DESIGN FACTORS FOR EVALUATING CHILD RESISTANT PACKAGING By Cory Jay Wilson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Packaging – Doctor of Philosophy 2021 DESIGN FACTORS FOR EVALUATING CHILD RESISTANT PACKAGING ABSTRACT By Cory Jay Wilson Child resistant packaging (CRP) is required for a wide range of hazardous household products including over-the-counter (OTC) and prescription medication. This type of packaging is intended to serve as a last line of defense that provides a physical barrier to prevent children under the age of 5 years from accessing substances that can be harmful to them. Studies indicate that child-resistant packaging is difficult for adults to open and potentially leading to inadequate reclosure (intentionally or unintentionally), or the transfer of contents to non-CRP containers and pill boxes, entire circumvention of the child resistant mechanism, all of which can facilitate unintentional medication ingestion by young children. The overarching objective of this research was to design and develop a novel child-resistant mechanism for a reclosable container / closure packaging system (RCCPS) for medication that is both adult-friendly and child-resistant. This research postulated that differences in anthropometric hand data (children vs adults) could be leveraged to develop a child resistant mechanism that distinguishes between the capabilities of adults and young children to apply torque to a RCCPS for medication. Young children (3-5 years) were found to utilize fewer grip types (3: cylindrical, spherical, and pronated spherical) when interacting with cylindrical push and turn child resistant packaging when compared with grip types employed by adults over 18 (6: cylindrical, spherical, and pronated spherical, box, pulp, lateral). Utilizing this insight, a two-piece, inner cap / overcap packaging system (patent pending, U.S. application no. 16/421,631) was developed which shields specific portions of the hand from gripping the inner cap by restricting the exposed surface area of the same. Our hypothesis was that the functional surface area (area where the inner cap is exposed) could be “tuned”. One-hundred fifty participants were tested in this study including: 50 children (4- 12 years), 50 adult males (18-67) and 50 adult females (18-78). The average peak torque achieved by children aged 4 (n= 9) was significantly (p<0.0001) lower (4.0177 lbs-inch +/- 1.17 lbs-inch) with our experimental design compared to the control (continuous thread medication bottle with screw cap, 7.40818 lbs-inch+/-1.17 lbs-inch). There was no evidence of a significant difference (p=0.9581) among senior adults (65+ years, n=11) when using the control (18.5375 lbs-inch+/- 1.17 lbs-inch) compared to senior adults (65+ years) using our design (13.5047 lbs-inch +/- 1.17 lbs-inch). Additionally, the amount of average peak torque 4-year-olds generated with the experimental design (4.0177 lbs-inch +/- 1.17 lbs-inch) was significantly lower (p<0.0001) than senior adults’ average peak torque (13.5047 lbs-inch +/- 1.17 lbs-inch) with the experimental design. As such, findings may have implications for the design of effective child-resistant and adult-friendly closures. TABLE OF CONTENTS LIST OF TABLES..………………………………………………………………………….….vi LIST OF FIGURES...………………………………………………………………………….viii CHAPTER 1: PROBLEM STATEMENT ......................................................................... 1 1.1 Background ........................................................................................................... 2 1.2 Poison exposure case reporting ............................................................................ 6 CHAPTER 2: LITERATURE REVIEW .......................................................................... 15 2.1 Background - Child-resistant Packaging for Medications ..................................... 15 2.2 Literature review methodology ............................................................................. 16 2.3 Stage 1 – Identification of performance metrics ................................................... 16 2.3 Stage 2 - User interaction studies ........................................................................ 23 2.3.1 User interaction studies approach ................................................................. 23 2.3.2 Rationale for approach .................................................................................. 23 2.3.2 Grip strategy .................................................................................................. 25 2.3.3 User studies of dissimilar and simultaneous motions with CRP .................... 32 2.4 Summary ............................................................................................................. 36 CHAPTER 3: PILOT STUDY ........................................................................................ 38 3.1 Pilot Study ........................................................................................................... 38 3.1.1 Pilot Study Aims ............................................................................................ 39 3.1.2 Materials and Methods .................................................................................. 39 3.1.3 Archived Study S/S/CR Methodology - subject of analysis ........................... 39 3.1.4 Methodology for analyzing the S/S/CR study ................................................ 41 3.2 Pilot Study Results ............................................................................................... 44 3.2.1 Children who were unable to open CRP ....................................................... 44 3.2.2 Children who opened CRP ............................................................................ 48 3.3 Pilot Study Discussion ......................................................................................... 50 3.4 Pilot Study Conclusions ....................................................................................... 53 3.5 Pilot Study Limitations ......................................................................................... 54 CHAPTER 4: DESIGN APPROACH ............................................................................. 57 4.1 Design considerations ......................................................................................... 57 4.2 Design hypothesis rationale ................................................................................. 58 4.2.1 Hand regions used for grip contact ............................................................... 58 4.2.2 Thumb and fingers grip contact ..................................................................... 60 4.3 Design Hypothesis ............................................................................................... 62 4.4 Design development rationale ............................................................................. 64 4.5 Design Theory ..................................................................................................... 67 iv 4.6 Design Approach ................................................................................................. 69 CHAPTER 5: DESIGN EVALUATION STUDY ............................................................. 74 5.1 Design hypothesis ............................................................................................... 74 5.2 Materials and Methods ........................................................................................ 75 5.2.1 Procedure ......................................................................................................... 83 5.3 Demographic Statistics ........................................................................................ 86 5.3.1 Participant age, gender, and ethnicity ........................................................... 86 5.3.2 Thumb breadth measurement ........................................................................... 88 5.4 Average peak torque results ................................................................................ 90 5.4.1 Average peak torque by thumb breadth ........................................................ 90 5.4.2 Average peak torque by package .................................................................. 92 5.4.3 Average peak torque by age ......................................................................... 94 5.5 Discussion ......................................................................................................... 100 5.6 Conclusions ....................................................................................................... 104 5.7 Limitations ......................................................................................................... 105 APPENDICES ............................................................................................................. 107 Appendix A – IRB APPROVED DOCUMENTS ....................................................... 108 Recruitment Flyer ................................................................................................. 108 Consent Form A ................................................................................................... 109 Consent Form B ................................................................................................... 113 Assent Form (8-11 years) .................................................................................... 117 Assent Script ........................................................................................................ 119 Data Collection Sheet .......................................................................................... 120 Appendix B – FLAT FILE ......................................................................................... 122 Sight, Sound, and Child Resistance ..................................................................... 122 BIBLIOGRAPHY ......................................................................................................... 125 v LIST OF TABLES Table 1: Outline of research methodology and goals. ..................................................... 1 Table 2: Poisoning exposure case reporting entities. ...................................................... 7 Table 3: Unintentional Poisoning. .................................................................................... 9 Table 4: Healthcare costs. ............................................................................................. 13 Table 5: Problem statement summary of goals. ............................................................ 14 Table 6: Literature review goals. ................................................................................... 15 Table 7: Types of Child-Resistant Packaging. ............................................................... 17 Table 8: Testing protocols for Child-Resistant Packaging. ............................................ 19 Table 9: ISO 13127 Mechanical testing of reclosable packaging for squeeze and turn and push and turn packaging. ....................................................................................... 22 Table 10: Frequency of adults presenting a grip type by treatment. .............................. 28 Table 11: Adult grip choices with push/squeeze and turn closures. .............................. 30 Table 12: Literature review summary of goals. .............................................................. 37 Table 13: Pilot study goals. ........................................................................................... 38 Table 14: Children who were unable to open CRP (closure and container grips identified). ...................................................................................................................... 47 Table 15: Children who successfully opened CRP. ....................................................... 50 Table 16: Frequency of adult user grip choices when opening 28 mm cylindrical squeeze and turn (SaT), and push and turn (PaT) closures. ......................................... 53 Table 17: Pilot study summary of goals. ........................................................................ 56 Table 18: Design approach goals. ................................................................................. 57 Table 19: Design approach summary of goals. ............................................................. 73 vi Table 20: Design evaluation study goals. ...................................................................... 74 Table 21: Nomenclature for package surface area exposure levels. ............................. 78 Table 22: Number of adult participants by self-identified gender and age, and number of child participants by parental/guardian-identified age and gender. ............................... 87 Table 23: Number of adult participants by self-identified ethnicity, and number of child participants by parental/guardian-identified ethnicity. .................................................... 87 Table 24: Participant thumb breadth (proximal interphalangeal joint). .......................... 89 Table 25: Age groups, number of participants, and thumb breadth (proximal interphalangeal joint) for analysis. ................................................................................. 94 Table 26: Design evaluation summary of goals. .......................................................... 106 Table 27: S/S/CR Study Flat File opened packages. .................................................. 122 Table 28: S/S/CR Study Flat File unopened packages. .............................................. 123 vii LIST OF FIGURES Figure 1: Unintentional poisonings 1972-2013. ............................................................... 6 Figure 2: Poison Statistics National Data (2016). .......................................................... 10 Figure 3: Top substance categories of unintentional poisoning exposures. .................. 11 Figure 4: National Electronic Injury Surveillance System (NEISS) query results (2000- 2017). ............................................................................................................................ 12 Figure 5: Dissimilar and simultaneous motions for push and turn (a), and squeeze and turn (b) packaging systems. .......................................................................................... 18 Figure 6: General force exertions involved in applying torque to a cylinder closure. ..... 24 Figure 7: Grip types. ...................................................................................................... 26 Figure 8: Torque measurement device example. .......................................................... 31 Figure 9: Torquimeter device example. ......................................................................... 33 Figure 10: Squeeze and turn closures measured with tridigit and bidigit grips. ............. 34 Figure 11: Opening sequence diagram for a squeeze and turn closure. ....................... 36 Figure 12: Grip types related to grasping cylinder closure packaging. .......................... 43 Figure 13: Frequency of children who were unable to open CRP (closure grips identified). ...................................................................................................................... 45 Figure 14: Frequency of children who were unable to open CRP (closure and container grips identified). ............................................................................................................. 46 Figure 15: Frequency of children who successfully opened CRP (closure grips identified). ...................................................................................................................... 48 Figure 16: Frequency of children who successfully opened CRP (container and closure grips identified). ............................................................................................................. 49 Figure 17: Regions of the hand. .................................................................................... 59 Figure 18: Example of a cylindrical grip on a closure. ................................................... 59 viii Figure 19: Example of thumb or finger contact with a flat surface. ................................ 61 Figure 20: Grip types related to gripping cylindrical closure packaging. ........................ 63 Figure 21: One-piece closure with pared cap and container mechanisms. ................... 65 Figure 22: Two-piece closure with paired cap and container mechanisms. .................. 65 Figure 23: Two-piece closure with paired cap mechanisms. ......................................... 66 Figure 24: Conceptual two-piece closure design. .......................................................... 68 Figure 25: Grips shielding of conceptual two-piece closure design. .............................. 69 Figure 26: Control design. ............................................................................................. 70 Figure 27: Design concept bridge. ................................................................................ 70 Figure 28: Design concept variations. ........................................................................... 71 Figure 29: Design surface area exposure. .................................................................... 76 Figure 30: 3D printed design variations. ........................................................................ 77 Figure 31: Single exposure 0, Package 5 and Control Package 1 examples. ............... 79 Figure 32: Design concept variations surface area details. ........................................... 81 Figure 33: Secure Pak Digital Torque Tester with control package clamped for testing. ...................................................................................................................................... 83 Figure 34: Measurement of participant thumb breath at the proximal interphalangeal joint. ............................................................................................................................... 88 Figure 35: Example of participant thumb contact orientation relative to the inner blue cap of each treatment package. .................................................................................... 91 Figure 36: Thumb breadth and average peak torque. ................................................... 91 Figure 37: Design concepts’ imposed surface area restriction. ..................................... 92 Figure 38: Average peak torque generated across all package treatments. ................. 93 Figure 39: Average peak torque generated across all package treatments by age groups. .......................................................................................................................... 95 ix Figure 40: Senior adult average peak torque across all package treatments. ............... 96 Figure 41: Average peak torque for children across all package treatments. ................ 97 Figure 42: Average peak torque for children and senior adults by prototype package. . 98 Figure 43: Control package max/min and average peak torque for adults and children. ...................................................................................................................................... 99 Figure 44: Average peak torque between control and prototype 1 with children and adults. .......................................................................................................................... 100 Figure 45: Range of average peak torque for all packages with senior adults and children. ....................................................................................................................... 102 x CHAPTER 1: PROBLEM STATEMENT Table 1: Outline of research methodology and goals. Research Goals(s) Problem Statement Literature Review Pilot Study Design Approach • Describe the scope of the problem • Narrow the problem focus • Describe how the focused problem will be addressed • Identify performance metrics and user interaction patterns related to the problem • Identify relevant gaps in knowledge • Gather data to address relevant identified knowledge gaps • Analyze results • Develop a design approach that addresses gaps in knowledge Design Evaluation Study • Evaluate design based on identified performance metrics, user interactions patterns, pilot study data analysis, and industry standards • Report significant findings • Describe implications of results • Summarize findings Results Discussion Conclusions 1 1.1 Background The U.S. Centers for Disease Control (CDC) defines a poison as “any substance, including medications, that is harmful to your body if too much is eaten, inhaled, injected, or absorbed through the skin” and classifies an unintentional poisoning as “when a person taking or giving too much of a substance did not mean to cause harm” (CDC, 2015). Unintentional poisoning by ingestion of household chemicals and medications was considered by physicians to be the leading cause of injury to young children under the age of 5 years in the mid-century 1900s (AAP, 1950). World War II (1939-1945) and the subsequent second industrial revolution were phases of rapid industrialization which led to mass proliferation of household chemicals being sold and stored in the homes with tenuous regulation. Without foresight, state death certificates from 1940 to 1949 reported over 400 (Bain, 1954) unintentional poisoning fatalities in each year for young children under the age of 5 years. This prompted the American Academy of Pediatrics’ Committee on Accident Prevention to recommend the creation of poison control centers that would provide specialized diagnoses and treatment assistance over the telephone for poisonings incidents. The first poison control center in the United States opened in Chicago in 1953, and by 1957 there were 17 poison control centers across the nation. The National Clearinghouse for Poison Control Centers was later established in 1957 as a hub to provide therapeutic and diagnostic information, and also coordinate the collection of data among centers. In 1958, the American Association of Poison Control Centers (AAPCC) was formed as a professional membership society that produced and distributed poisoning prevention materials and offered guidance to its members. 2 In response to data reports from the AAPCC, the American Medical Association (AMA) and US the Food and Drug Administration (FDA) successfully urged Congress to pass the Hazardous Substances Labeling Act of 1960 (Scriba, 1961). This Act banned substances that were too hazardous for common use in households and also required other, less hazardous, substances to be labeled with cautionary statements that warned consumers of potential harm. Additionally, a national education program (Public Law 87- 319) that designated the third week in March as National Poison Prevention Week (NPPW) was passed in an effort to raise public awareness of the issue. However, unintentional poisonings and deaths of young children continued to occur, and by 1961 the number of reported poison-related deaths of children under the age of 5 had reached approximately 450 incidents. Educating consumers and raising public awareness were not enough; new approaches were sought. At this point, manufacturers, advocates, and the government began to heavily explore functional packaging design as a potential way to physically restrict children from gaining access to harmful products. In 1966, two independent studies were conducted utilizing child-resistant packaging (CRP). Both studies demonstrated a greater than 75 percent reduction in unintentional child ingestions associated with the use of child-resistant packaging (Breault, 1974; Scherzo, 1970). Largely based on the effectiveness of the strategy documented in the aforementioned studies (CPSC, 2005), the US Congress drafted and adopted the Poison Prevention Packaging Act (PPPA) of 1970 (PPPA, 1970). The PPPA requires the manufacturers of hazardous household products to use “special packaging”. Special packaging (generally referred to as child-resistant packaging, or CRP) is defined by the 3 PPPA as “packaging that is designed or constructed to be significantly difficult for children under 5 years of age to open or obtain a toxic or harmful amount of the substance contained therein within a reasonable time and not difficult for normal adults to use properly, but does not mean packaging which all such children cannot open or obtain a toxic or harmful amount within a reasonable time”. Nearly one year after the PPPA mandate, aspirin ingestion related deaths in children declined 50 percent (CPSC, 2004). The enactment of the Consumer Product Safety Act (CPSA, 1972), transferred the enforcement of the PPPA from the FDA to the newly formed Consumer Products and Safety Commission (CPSC). In partnership with the US Environmental Protection Agency (EPA), who regulates CRP for rodenticides, fungicides and pesticides, the CPSC remains the enforcement Agency affiliated with the Act to this day. The core data collection system related to poisoning data at the CPSC is the National Electronic Injury Surveillance System (NEISS). NIESS collectively functions among the CPSC, US Centers for Disease Control and Prevention (CDC) and National Center for Injury Prevention and Control (NCIPC) to collect data on consumer-product related injuries from a nationally representative sample of U.S. hospital emergency departments. An expansion of that system: The National Electronic Injury Surveillance System - Cooperative Adverse Drug Event Surveillance Project (NEISS-CADES) collects data on adverse drug events (ADE) treated in hospital emergency departments. ADE refers to “an injury related to the outpatient use of a drug resulting from allergic reaction, side effect, unintentional overdose or secondary effect. Intentional drug injuries (e.g., suicide attempts) and injuries resulting from alcohol, tobacco, or illicit drug use are excluded.” (ODPHP, 2018). The data collected consists of outpatient demographics, 4 diagnoses, medication name and dose, tests and treatments preformed in the hospital emergency department. Over the many years since the PPPA’s inception and adoption of CRP, fatalities of young children associated with unintentional poisoning by ingestion of household chemicals and drugs has drastically declined. Analysis of NEISS-CADES data from 1964 to 1992 by the CPSC showed that CRP for prescription medicine reduced the death rate of children under 5 by 1.4 deaths per million, a 45 percent reduction (Rodgers, 1996). Another study which analyzed aspirin-related fatalities from 1973 to 1990 indicated a 34 percent reduction during the specified timeframe (Rodgers, 2002). Overall, CRP has helped reduce unintentional poisoning fatalities by 88 percent (CPSC, 2016) for children under age 5. Fatalities have decreased from 216 per year in 1972 (see Figure 1) to an average of 33 per year (January 2008 – December 2013). While this data represents a drastic decline in fatalities, preventable unintentional poisoning exposures that result in hospitalizations and deaths of young children continue to occur (Gummin et al., 2017). Work remains to be done. 5 Annual rate of child (< 5 years) fatalities from unintentional poisonings 1972 to 2013 250 200 150 100 50 s e i t i l a t a F 0 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 Year Figure 1: Unintentional poisonings 1972-2013. Annual rate (1972 - 2013) of fatalities from unintentional poisonings for young children under the age of 5 years (CPSC, 2016). 1.2 Poison exposure case reporting Today, poison control centers and hospitals are two main nodes for data collection related to poisoning (see Table 2). Injury victims and caregivers often seek immediate treatment when harm occurs and hospital emergency departments and poison control call centers can offer a timely care response. The volume of injury cases reported by these entities generates the data necessary to analyze public health concerns on a national scale. This data is used to monitor and evaluate public safety and welfare, identify outbreak developments, understand rates of exposure, and identify illness and death before patterns of injury become a persistent and widespread public health epidemic. 6 Table 2: Poisoning exposure case reporting entities. Data reporting entity Data collection system Collection method Reported age grouping category for young children Data inclusion criteria for reported cases American Association of Poison Control Centers (AAPCC) National Poison Data System (NPDS) 55 Poison control centers that collectively serve the U.S. and its territories Children under the age of 6 years (data for individual ages may be available for some statistics) “Human exposure cases” reported by telephone (“cases that are not duplicates and classified by the PC [poison center] as CLOSED. A case is closed when the PC has determined that no further follow-up/recommendations are required or no further information is available”). Exposures are listed by generic categories “a robust generic coding system categorizes product data into 1092 generic codes” (Gummin et al., 2017). Consumer Products Safety Commission (CPSC) National Electronic Surveillance System (NEISS) Nationally representative probability sample of hospitals in the U.S. and its territories Children under the age of 5 years “Poisonings (ingestions) and chemical burns to children under age 5 associated with drugs, medications or any other substances” (NEISS, 2017). “Patient information is collected from each NEISS hospital for every emergency department visit associated with a consumer product or a poisoning to a child younger than five years of age. The total number of product- related hospital emergency department visits nationwide can be estimated from the sample of cases reported in the NEISS.” (CPSC, 2018b). The American Association of Poison Control Centers (AAPCC) is now comprised of 55 poison control centers (2017) across the United States. These control centers are staffed by toxicology specialists; including pharmacists, physicians and nurses who provide telephone guidance to the public as well as health care facilities. Poison exposures are categorized by product or substance and coded into generic product categories. In 2016, the top five poison exposure product categories for children under 7 the age of 6 years were cosmetics/personal care products (13.3%), household cleaning substances (11.1%), analgesics, a single category of medication (9.21%), foreign bodies/toys/miscellaneous (6.48%), and topical preparations (5.07%). Reported case data of poisoning exposures from poison control centers are regularly uploaded every 8 minutes to the National Poison Data System (NPDS) and published in an annual report by the AAPCC. The NPDS can track poisoning exposures in near real-time, enabling the detection of outbreak events that pose a public health concern with rapidity. In 2016 the AAPCC received over 2.1 million calls of human poison exposure events - nearly one call every 15 seconds (Mowry, Spyker, Brooks, Zimmerman, & Schauben, 2016). Every year in the United States over 950,000 unintentional poisoning incidents involving children under the age of 5 years are reported to poisoning control centers (Mowry, Spyker, Brooks, Zimmerman, & Schauben, 2000-2016). This is the equivalent of five percent of all children under the age of 5 in the United States experiencing an unintentional exposure to hazardous substances (see Table 3). 8 Table 3: Unintentional Poisoning. Poison (99.4% unintentional) exposures (2000 through 2016) for young children under 5 years of age. aPopulation Data (Population Division, U.S. Census Bureau) (Bureau). bExposure Data (Annual Report of the American Association of Poison Control Centers’ National Poison Data System) (Mowry et al., 2000-2016). Year Total Populationb Percentage exposed 5.71% 5.81% 6.06% 6.07% 6.06% 5.93% 5.88% 6.05% 6.11% 6.11% 5.74% 5.44% 5.27% 5.05% 4.97% 4.89% 4.81% 5.64% 0.0047 Total Exposuresa (<5 years old) (<5 years old) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Average Std.dev 1,095,166 1,120,275 1,176,737 1,189,213 1,198,262 1,180,842 1,171,912 1,217,626 1,239,134 1,237,011 1,158,006 1,095,669 1,053,240 1,001,945 986,926 973,330 958,633 1,120,819 94,948 19,178,293 19,298,217 19,429,192 19,592,446 19,785,885 19,917,400 19,938,883 20,125,962 20,271,127 20,244,518 20,189,589 20,125,988 19,980,509 19,854,133 19,877,544 19,912,499 19,927,037 19,861,719 320,309 Young children under the age of 6 years accounted for 46.4% of the 2.1 million reported poison exposures in 2016; that is, exposures to any type of toxic substance. Across all age groups, for every 100,000 people in the United States 660 poison exposures were reported (Gummin et al., 2017). For every 100,000 children in the United States age 5 and under; an average of 4,964 poison exposures was reported. The highest incidence of poisoning reported was among the ages of 1 (8,083 per 100,000 children) and 2-year old children (7,675 per 100,000 children) (see Figure 2). Data from 2016 indicates that 77.8% of all poison exposures across all age groups reported were unintentional, 18.1% were intentional, and 2.5% were classified as 9 adverse reactions ; contrast this with data comprised solely of children less than 5, where 99.4% of exposures were unintentional (Gummin et al., 2017). Figure 2: Poison Statistics National Data (2016). Poison exposures per 100,000 population across all age groups (Gummin et al., 2017). Cosmetics and personal care products, analgesics (pain medication), and household cleaning products were consistently in the top 3 categories of exposure substances for children under six during the 16-year period from 2000 through 2016 (see Figure 3) (Mowry et al., 2000-2016). 10 Figure 3: Top substance categories of unintentional poisoning exposures. Poisoning (99.4% unintentional) exposures (2000 through 2016) for young children under 6 years of age (Mowry et al., 2000-2016). Hospital emergency departments also serve as a valuable source for reporting poisoning exposures. The National Electronic Surveillance System (NEISS) functions under the CPSC to collect common patient data related to any injury involving a consumer product (see Table 1). Data collected includes; date when injury occurred, product involved, number of patients injured, patient sex and age, diagnosis, location where injury happened, and body part(s) injured. Emergency department patient data aligns with the common data elements NEISS collects in an attempt to not place an undue burden on emergency department staff. NEISS collects injury data from a probability sample of 100 hospitals with 24-hour emergency departments (that have more than six beds) and then provides national estimates for the number and type of consumer product-related injuries. Unintentional poisoning exposures for young children continue to remain a common occurrence (see Figures 3 and 4). An average of over 84,000 children under 11 the age of 5 were taken to the emergency department for unintentional poisoning exposures each year from 2000 to 2017 (CPSC, 2018a). During the same period, an annual average of 12,500 young children were admitted to the hospital and treated or observed for more serious exposures (see Figure 4). Available associated healthcare costs from emergency department visits, and poison control center service calls are estimated for 2005 and depicted in Table 4. Despite best efforts from industry, government agencies, and academia the occurrence of these incidents persist (Lovegrove et al., 2014; Spiller et al., 2013). Figure 4: National Electronic Injury Surveillance System (NEISS) query results (2000-2017). Poisoning (99.4% unintentional) exposures for young children under 5 years of age. (5 Year Age Groups: 00 - 04 years; Diagnosis Selection: Poisoning (68); Disposition Selection: 0=no Injury, 1=treated & Released, Or Examined & Released Without Treatmnt, 6=left Without Being Seen, 2=treated & Transferred, 4=treated & Admitted For Hospitalization, Hospitalized, 5=held For Observation, 8=fatalities, Including Doa, Died In Er, Expired In Hosp.). 12 Number of Exposures Cost per patient (USD) $1,840c Total Cost (USD) $161,496,800 Table 4: Healthcare costs. Estimated healthcare cost of pediatric (age < 5 years) unintentional poisoning exposures (2005). a National Electronic Injury Surveillance System (NEISS) query results 2005 (CPSC, 2018a). b National Poison Data System (NPDS): Annual Report 2005 (Mowry et al., 2000-2016). cTadros et al., 2016 (Tadros, Layman, Davis, Bozeman, & Davidov, 2016). dZaloshnja et al., 2008 (Zaloshnja et al., 2008). Treatment 1 Treated and Released from Hospital 87,770a Emergency Department (NEISS) 2 Admitted to Hospital Through the 11,283a $14,235c $160,613,505 Emergency Department(NEISS) 3 Poison control center service call 1,180,842b $34.94d $47,162,829 (AAPCC) 4 Total Medically-Attended Injuries Cost $369,273,134 Available data indicates 83.6% of all reported unintentional poisoning cases (some cases consist of multiple routes of exposure) in 2016 were ingestion related followed by dermal absorption (6.96%), inhalation (6.32%) and ocular absorption (4.23%). Additionally, the CPSC has identified and indicates known contributors (CPSC, 2005) for unintentional pediatric ingestions of medications as: • “Availability of non-special packaging, on request, for prescription medication • Availability of one non-special packaging size of over-the-counter medications • Inadequate quality control by manufacturers leading to defective closures • Misuse of special packaging in the home (leaving the cap off or unsecured, transferring the contents to a non-special package) • Violations of the law by the pharmacist and/or the dispensing physician” As mentioned, special packaging “is designed or constructed to be significantly difficult for children under 5 years of age to open or obtain a toxic or harmful amount of the 13 substance contained therein within a reasonable time and not difficult for normal adults to use properly, but does not mean packaging which all such children cannot open or obtain a toxic or harmful amount within a reasonable time”. Thus, this research investigates ‘special packaging’ for medication with a focus on ergonomic packaging design factors that may influence “misuse of special packaging in the home (leaving the cap off or unsecured, transferring the contents to a non-special package)”. Table 5: Problem statement summary of goals. Research Goal(s) Summary Problem Statement • Describe the scope • Unintentional pediatric of the problem poisoning • Narrow the problem • Continuous thread cylinder focus • Describe how the • focused problem will be addressed type child-resistant packaging (medication bottle/cap), that is difficult for adults to open, may be inadequately closed and compromise child- resistance (Literature review) Identify gaps in knowledge, patterns of user interaction, and performance metrics related to opening child- resistant packaging in order to develop a design methodology that addresses both adult ease of use and child-resistance 14 CHAPTER 2: LITERATURE REVIEW Table 6: Literature review goals. Research Literature Review Goal(s) • Identify performance metrics and user interaction patterns related to the problem • Identify relevant gaps in knowledge 2.1 Background - Child-resistant Packaging for Medications Child-resistant packaging for medications differ from other potentially hazardous products because unique considerations exist due to the nature of the product. Medication for chronic conditions must be taken as prescribed to manage aliments and provide benefits that can be lifesaving. The functional utility of the package is paramount for these types of packages because anything that hinders convenient access is likely to result in noncompliance and unintended behaviors (Conn et al., 2015). Studies indicate older adults have difficulties opening CRP and, as a consequence, may engage in unintended behaviors that circumvent child‐resistant features (Blok, Ruiter, & Wever, 2016; Jacobson, Rock, Cohn, & Litovitz, 1989; Yoxall, Rodriguez-Falcon, & Luxmoore, 2013). Child-resistant packaging that is difficult to open may be poorly reclosed (intentionally or unintentionally), may result in the product being transferred to non-CRP secondary containers and pill boxes, or even left open entirely, all of which can facilitate poisoning exposure incidents (CPSC, 2005). In a review of intoxication causes for children under the age of 6, 86% of emergency department visits for medicine poisoning were found to be the result of the child gaining access to an adult medicine (SafeKids, 2013). In 38% of cases it was found that the medication the child ingested was normally used by a grandparent, 31% 15 by the mother, 12% by a sibling, 8% by a father, 5% by an aunt/uncle and 6% was unknown. Results suggest the need for CRP features which strike a balance between facilitating ease of use for adults (particularly older adults) while providing a restrictive barrier for young children (Beckman, Bernsten, Parker, Thorslund, & Fastbom, 2005; Yoxall et al., 2006). 2.2 Literature review methodology Herein, we leverage the two-stage model (Peebles & Norris, 2003) for our literature review. Specifically, we first (1) identify performance metrics related to ergonomic design, and then (2) gather insights from published user-interaction studies that are relevant to child-resistant packaging design for medication to inform the design of a novel CRP closure. The benefits of using ergonomics data in the early stages of the design process are widely recognized because they are integral to the design of safe and usable products (Ramsey, 1985; Rennie, 1981). Stage one of this model identifies performance metrics that may influence ergonomic design in the context of CRP. Stage two reviews existing user-interaction studies that measure performance metrics identified as critical during stage one. The objective of this literature review is to synthesize meaningful patterns of insights from user interactions studies with performance metrics relevant to ergonomic CRP design. The ultimate goal of this work is the creation of a CRP design methodology that considers both adult ease of use and child-resistance in its design process. 2.3 Stage 1 – Identification of performance metrics Various types of packaging are recognized as special packaging (commonly referred to as child-resistant packaging or CRP) based on CPSC performance testing; 16 these broad types have been categorized in a standard published by ASTM International. The standard, ASTM D3475, lists the standard categories of design classification of child-resistant packages. Thirteen types are identified (see Table 7), six types are categorized as reclosable, three as non-reclosable, two are considered dispensers, and the remaining two types are boxes/trays and aerosols (ASTM, 2018). Work herein is concentrated on a reclosable system for solid oral dosage forms. Reclosable packaging is defined as “package which, after it has been initially opened, is capable of being reclosed with a similar degree of security and is capable of being used a sufficient number of times to dispense the total contents without loss of security”. Table 7: Types of Child-Resistant Packaging. ASTM D3475 (2018) Standard Classification of Child-Resistant Packaging types (ASTM, 2018). ASTM D3475-18 Standard Classification of Child-Resistant Packages TYPE Classification RECLOSABLE PACKAGING—CONTINUOUS THREAD CLOSURE RECLOSABLE PACKAGING LUG FINISH CLOSURE RECLOSABLE PACKAGING SNAP CLOSURE UNIT NON-RECLOSABLE PACKAGING FLEXIBLE (STRIP/POUCH) UNIT NON-RECLOSABLE PACKAGING RIGID UNIT RECLOSABLE PACKAGE I II III IV V VI VII AEROSOL PACKAGES VIII NON-RECLOSABLE PACKAGING SEMI-RIGID (BLISTER) IX X XI XII DISPENSER (MAY BE REMOVED) XIII RECLOSABLE PACKAGING—SEMI-RIGID (BLISTER) DISPENSERS (NOT INTENDED TO BE REMOVED) BOX OR TRAY PACKAGE RECLOSABLE PACKAGING FLEXIBLE The two most common types of reclosable child-resistant packaging systems are ASTM type I and II “push and turn” and ASTM type I “squeeze and turn” (Blok et al., 2016; Rodriguez-Falcon & Yoxall, 2010; Yam, 2010). A push and turn packaging system “requires the user to apply a downward force while simultaneously applying a 17 torque to unscrew the closure from the container” (ISO, 2012). A squeeze and turn packaging system “requires the user to squeeze the closure at designated points while simultaneously applying a torque to unscrew the closure from the container (ISO, 2012)” (see Figure 5). These types of closures employ the concept of dissimilar and simultaneous motions as a child-resistant mechanism. The force exertions required to operate a push and turn system are compression and torque, while squeeze and turn systems utilize pinch/grip strength and torque. Figure 5: Dissimilar and simultaneous motions for push and turn (a), and squeeze and turn (b) packaging systems. ISO (International Organization for Standardization) 8317:2015 (ISO, 2015) and ISO 13127:2012 (ISO, 2012) are the most relevant reclosable CRP test standards 18 because they comprehensively specify performance requirements, user panel testing, and mechanical assessments for reclosable packages (see Table 8). By comparison, the testing protocol dictated by the regulations authorized under the PPPA (PPPA, 1970) (found in 16 CFR 1700) encompasses both reclosable and non-reclosable packaging but does not specify mechanical assessments for the packages. ISO 8317 is an adoption of the PPPA’s testing protocol but specifically pertains to reclosable packaging, and ISO 13127 is a supplemental extension to 8317 that specifies test methods for mechanical assessments for reclosable CRP (see Table 8). Table 8: Testing protocols for Child-Resistant Packaging. Comparable features of PPPA and ISO 8317 / 13127 testing protocols for reclosable child-resistant packaging. Testing protocol PPPA 16 CFR 1700 (PPPA, 1970) ISO 8317 (ISO, 2015) ISO 13127 (ISO, 2012) [supplemental extension of ISO 8317] Package type All-inclusive, child- resistant packaging Reclosable, child- resistant packaging Reclosable, child- resistant packaging Panel testing: Children (42 – 51 months in age) Younger-adult (18 – 45 years in age) Senior adult (50-70 years in age) Panel testing: Children (42 – 51 months in age) Senior adult (50-70 years in age) Panel testing: Children (42 – 51 months in age) Senior adult (50-70 years in age) Test subjects 19 Table 8 (cont’d) Task Children: two 5- minute test periods. First period (try) opening container without demonstration. Second period (try) opening package after a visual demonstration. Adults: One 5–minute period and one 1- minute period. 5- minute period to open package, 1-minute period to open and close another identical package (screening test if necessary) Children: two 5- minute test periods. First period (try) opening container without demonstration. Second period (try) opening package after a visual demonstration. Adults: One 5– minute period and one 1-minute period. 5-minute period to open package, 1-minute period to open and close another identical package (screening test if necessary) Testing dependent on package type: • Torque release test • Squeeze test • Non-squeeze torque test • Press down and turn engagement test • Push and turn test • Reverse ratchet torque test • Disassembly test • Rotational torque test • Push-off force • Application force Data interface Observational Observational Mechanical Success/failure Quantitative: Mean opening times and standard deviation, ratio of total packages opened to tested, number of adult re- securing failures Qualitative: “Opening method, (e.g., normal opening, teeth, etc.).” Success/failure Quantitative: Mean opening times and standard deviation, ratio of total packages opened to tested. number of adult re-securing failures Qualitative: Opening method: “teeth (or any other means)” T-test to determine whether there is a statistical difference between the test data and reference data Reference data: rotational torque and applied forces (a normal force oriented horizontal or vertical relative to the closure). Assessment of results (minimal) 20 The reference data categories specified in ISO 13127 are data metrics relevant to reclosable CRP design. This standard establishes mechanical test methods where “data generated by these mechanical test methods are suitable for comparing child resistant characteristics of related reclosable packaging systems” (ISO, 2012). ISO 13127 identifies and describes application forces “unidirectional (vertical and lateral) compression and tension forces”, and rotational torque as reference data in relation to push/squeeze and turn systems (see Table 9). Reference data is defined as “data generated from type-approved child resistant packages using the testing methods specific to the packaging as given in this International Standard”(ISO, 2012). In other words, a child-resistant package that has passed ISO 8317 panel testing (see Table 8) is considered type-approved and can then be mechanically tested under ISO 13127 to generate reference data. These reference data are used as performance metrics intended to measure the efficacy of relevant child-resistant features of a reclosable packaging system. Application forces and torque are not comprehensive a list of data that can be used to inform the ergonomic design of CRP. Instead, application forces and torque can be considered performance metrics by which CRP features can be evaluated. This research will utilize these performance metrics as criteria to direct a focused review of user interaction studies. 21 Table 9: ISO 13127 Mechanical testing of reclosable packaging for squeeze and turn and push and turn packaging. Reference data Task description Reclosable packaging system ISO 13127 (ISO, 2012) mechanical testing specified Squeeze and turn Squeeze test Application force Non-squeeze torque test Rotational torque Progressive application of a load to opposite sides of the closure’s squeeze points Anticlockwise rotation of the closure without squeezing Push and turn Press down and turn engagement test Application force and rotational torque A progressively increasing normal force to the closure while rotating anticlockwise Push and turn test Application force and rotational torque Reverse ratchet toque test Rotational torque Disassembly test Application force A normal force of sufficient magnitude to engage the mechanism while rotating in an anticlockwise direction Anticlockwise rotation without application of downward force Progressively increasing tensile load 22 2.3 Stage 2 - User interaction studies 2.3.1 User interaction studies approach In order for a user to open or close a continuous-thread, reclosable child- resistant container they must first choose an effective grip type to exert suitable application forces on the closure. They must generate enough rotational torque while maintaining their grip type to successfully turn the closure in any given direction (conventionally, counterclockwise for removal). This second stage of the review identifies relevant grip strategies by reviewing ethnographic studies where participants are video-recorded while interacting with packaging. Next, studies that evaluate the design concept of dissimilar and simultaneous motions (application force and rotational torque) as a child-resistant mechanism are reviewed. 2.3.2 Rationale for approach In relation to a reclosable container /closure system, torque (commonly in newton meters [Nm], or pounds-inch units) is described as either application or removal torque. In other words, the total radial force applied to the closure in either a clockwise (application) or counterclockwise (removal) direction. The general force exertions involved in the opening of a closure are diagramed in Figure 6 based on Pheasant and O’Neill’s (1975) general torque model for cylindrical objects. 23 G (grip force) T (torque) T = µ G D T: torque (Nm) µ: coefficient of friction between the hand and object G: Grip force (N) D: cylindrical diameter (m); moment arm of friction forces Figure 6: General force exertions involved in applying torque to a cylinder closure. Adapted from (Pheasant & Oneill, 1975) and (Seo, Armstrong, Ashton-Miller, & Chaffin, 2007). The model depicts that the resultant torque (T) exerted on a cylinder depends on grip force (G), the coefficient of friction between the hand and the object (µ), and the object diameter (D); where (T = G x µ x D). The torque (T) that a user can apply to a closure is limited by the user’s ability to functionally apply a grip force (G) (Rowson & Yoxall, 2011). A user may be strong enough to rotate a closure open but will fail if they cannot maintain a grip on the closure while attempting to generate sufficient torque to turn the closure. A frictional force between a user’s hand and the material of the closure develops when a grip force is applied to the closure. This frictional force resists contact from slipping between the closure and the user’s hand. The limit of this friction force is the product of the maximum grip force the user applies and the maximum coefficient of friction between the skin of the user’s hand and the material of the closure (Rowson & Yoxall, 2011). There have been numerous studies that measure human grip force (G) using grip dynamometers (Giampaoli et al., 1999) and pinch gauges (Mathiowetz et al., 1985). These grip strength measurements are not directly applicable to cylinder package 24 closures because other studies have shown strength measurements to be dependent on the geometry of the test mechanism (Crawford, Wanibe, & Nayak, 2002; Yoxall et al., 2006). Force data from users applying grip force across parallel bars of a grip dynamometer or across a rectangular pinch gauge have shown weak correlation with the forces used to operate container closures (Rice, Leonard, & Carter, 1998). Thus, concepts for user grip strategies, application forces, and torque are evaluated by reviewing user studies in either the broad context of cylindrical closure systems or the specific context of cylindrical closure CRP if data is available. 2.3.2 Grip strategy In order to open or close a reclosable child-resistant package, a user must first grip the packaging to exert forces on it. Certain gripping types of the hand are inherently stronger than others when performing specific tasks (Rowson & Yoxall, 2011); thus, the ability to succeed in opening or closing a container is partially related to grip choice. Yoxall et al. identified seven types of grips that could be employed when trying to open relevant, cylinder-shaped containers (see Figure 7). 25 (A) Spherical (B) Cylindrical (C) Pronated Cylindrical (D) Pulp (E) Box (F) Lateral (G) Tip Figure 7: Grip types. Grip types related to grasping cylinder closure packaging. Identified by (Rowson & Yoxall, 2011). In an ethnographic study (Yoxall et al., 2007), grip types that participants used while manipulating packaging was recorded. User interactions with 6 different packages: three (60 mm, 75 mm, and 85 mm closure diameters) jars, one 28 mm closure diameter water bottle, one yoghurt pot, and one 25 g flexible crisp packet were video-recorded, observed, and then analyzed. Fifty users between the ages of 18 and 26 63 years were asked to open each package without any explicit instructions on how to open the packages given. Trends for user preferences of grip types were attributed to the inherent flexion range (a function hand size) of certain grip types being easier to achieve with larger or smaller closure diameters. For example lateral grips, with limited grip span between the thumb and index finger, occurred more often with smaller closure diameters while wider spherical and box grips were observed more often with larger closure diameters (see Table 10). The ability to apply grip force (G) with the hand is strongest when the hand is moderately flexed (Seo & Armstrong, 2008). When the hand has to narrow or contract (shortening of the muscles) below moderate flex to grip small objects, grip force is decreased due to inefficient contact with the entire grasp of the hand (Barry, 2009). The hand is also weaker when the fingers are near full extension, grip force decreases when widening or stretching the hand beyond moderate flex to grasp large objects (Barry, 2009). 27 Table 10: Frequency of adults presenting a grip type by treatment. Summary of adult user grip choices when applying torque to cylinder closure packaging Adapted from (Yoxall et al., 2007). Grip type (n = 50 participants) 85 mm (jar) 75 mm (jar) 60 mm (jar) 28 mm (water bottle) Spherical Box Lateral Pulp Cylindrical Cylindrical (with ring finger) Cylindrical (pronated) 24 19 13 18 19 8 - 1 5 - 3 - - 7 20 27 - 9 - - 9 9 2 2 - 2 2 1 28 In another video-recorded ethnographic study (Yoxall et al., 2013), under the more specific context of CRP, gripping choices for two plastic bleach bottles with 28 mm cylinder closures were evaluated. One bottle featured a push and turn closure and the other bottle used a squeeze and turn closure. A random sample of 57 individuals (age and gender were not specified) were tasked with opening each package. No additional grip types identified in this study (see Table 11) that were not identified in the previous study (see Figure 7). Frequency of grip type used to initially open the closure was recorded (see Table 11) and use frequency by grip type was reported. It can be inferred that the operational context of the closure may influence the choice of grip type employed to open the closure, given that the closures in this study were similar in appearance and geometry. A lateral grip type for the squeeze and turn closure was the most frequent grip type and shows agreement with the 28 mm water bottle closure data from the previous study (Yoxall et al., 2007). In contrast, both lateral and box grip types were frequently used for the push and turn closure (see Table 11). This raises a few interesting questions. Does a child’s lack of knowledge to operate CRP impact their grip choices when interacting with a CRP? Are there implications for CRP design? Searches for similar video-recorded ethnographic studies that reported grip types of children under the age of 5 when interacting with cylinder closure packaging systems did not yield any results. Further research in this area is needed. 29 Table 11: Adult grip choices with push/squeeze and turn closures. Frequency of adult user grip choices when apply torque to cylinder squeeze and turn, and push and turn closures (Yoxall et al., 2013). Grip type (n = 57 participants) Squeeze and turn bleach bottle closure Push and turn bleach bottle closure Spherical Cylindrical Tip Pulp Lateral Box Cylindrical (pronated) - 6 - 7 38 3 3 30 6 7 - 2 19 22 1 In another study (Rowson & Yoxall, 2011), 19 females and 15 males (ages were not reported) were tasked with trying to open a torque measuring device constructed in such a way that it resembled a cylinder jar with a closure. A torque sensor was mounted inside of a glass jar with a jar lid attached to the device to measure the amount of torque a user could generate when gripping the closure. The torque sensor was removable and could be placed in other jars and attached to other removable lids (see Figure 8). Figure 8: Torque measurement device example. Example of a torque measurement device constructed to resemble a cylinder jar packing systemReprinted from Applied Ergonomics, Vol 42, Issue 5, J. Rowson, A. Yoxall, Hold, grasp, clutch or grab: Consumer grip choices during food container opening, Pages 627-633, Copyright (2011), with permission from Elsevier. Users were shown seven grip types (see Figure 7) and asked to attempt to open the torque jar using the pictured grips. The construction of the device precluded actual removal of the “lid” but enabled the research team to measure the amount of torque users exerted trying to twist the closure. Three jar closures of varied diameters (55 mm, 75 mm and 110mm closures) were used for testing with the torque device. Each user 31 conducted 21 trials (7 grips [see Figure 7] for each of the 3 lid sizes) in a randomized order with suitable rest periods. The torque that would normally be required to open jars of 55, 75 and 110 mm closure diameter were referenced from a previous study (Yoxall et al., 2010) and compared to the torque values users generated in this study (Rowson & Yoxall, 2011). Men were able to produce higher maximum torque values than women for each type of grip. Other studies have shown that hand exertions of force by adult females is generally around two-thirds of that of males (Mathiowetzet al., 1986; Fothergill et al., 1992; Steenbekkers, 1993) and males generally have larger hands (breadth, width, and length) (DTI, 2002). Key findings from this study were that “Women from the survey seemed to have a tacit knowledge of the implication of grip choice, whereby they chose the grip type that was able to apply sufficient strength to overcome the required jar openability torque, regardless of comfort. Men however, were seen to have no such limitations, allowing them to employ a wide choice of grips, and, as such, being able to choose their preferred type.” (Rowson & Yoxall, 2011). Are these results extendable to young children that have comparably smaller hands and less strength than adults? Are young children similarly limited in their functional grip choices? What are the implications for CRP design? Further research in this area is needed. 2.3.3 User studies of dissimilar and simultaneous motions with CRP In a study by Bonifim et al. torque measurements (maximum grip force was not measured) were taken with the three different squeeze and turn mouthwash packages using a bi-digit (pulp) and a tri-digit (box) gripping technique to squeeze the closures 32 (Bonfim, Medola, & Paschoarelli, 2016). Each package was drained of its contents and a torquimeter was mounted inside (see Figure 9). Figure 9: Torquimeter device example. Torquimeter (left) embedded inside a squeeze and turn mouthwash package to measure torque. Tri-digit (box) and Bi-digit (pulp) grips used to generate torque in this study. Reprinted from International Journal of Industrial Ergonomics, Vol 54, Gabriel H.C. Bonfim, Fausto O. Medola, Luis C. Paschoarelli, Correlation among cap design, gripping technique and age in the opening of squeeze-and-turn packages: A biomechanical study, Pages 178-183, Copyright (2016), with permission from Elsevier. The maximum torque generated by 100 adults and children (ranging in age from 3 to 60+ years old) was measured and recorded. Results were characterized by age group: (A) 3-5 years, (B) 8-12 years, (C) 13-17 years, (D) 30-59 years, and (E) 60+ years. The key finding of this study was that children under the age of 5 (group A) when using a tri-digit grip were able to produce torques greater than all other age groups that used a bi-digit griping method (see Figure 10) (Bonfim et al., 2016). Conversely, the bi- digit grip in the 3-5 age group produced the lowest torque of all age groups and grip types. Thus, it can be inferred that a CRP that could limit the functional grip strategies of young children to certain grip types while allowing adults to retain certain grip types 33 could potentially distinguish the ability to apply torque between adults and young children. Figure 10: Squeeze and turn closures measured with tridigit and bidigit grips. Torque measurements in (newton meters) for tri-digit (box) and bi-digit (pulp) grips types by age group. Reprinted from International Journal of Industrial Ergonomics, Vol 54, Gabriel H.C. Bonfim, Fausto O. Medola, Luis C. Paschoarelli, Correlation among cap design, gripping technique and age in the opening of squeeze-and-turn packages: A biomechanical study, Pages 178-183, Copyright (2016), with permission from Elsevier. 34 A different analytical approach was used in another study (Yoxall et al., 2013) that utilized computer simulations, a numerical hand model, and task modeling for the opening motions employed when manipulating a squeeze and turn CRP. The CRP closure model was a 3D laser-scanned 28 mm bleach bottle squeeze and turn closure. The hand model was arranged in a bi-digit (pulp grip) configuration (see Figure 9, right). Forces were applied to the model in squeeze and turn sequence. During the squeeze sequence two opposing forces were applied to each fingertip contacting the closure. The force was ramped up linearly until the closure deformed enough to allow the closure to rotate (see Figure 11). Once the closure was sufficiently deformed the linear increase in squeeze force remained constant. While the closure was deformed, two additional forces were applied to the fingertips in opposite directional tangential to the closure to induce the turning motion. The turn motion was continued until the closure lugs cleared the stop lugs on the bottle. Data from the sequence carried out in the study indicated a “the application of the ‘turn’ force causes the thumb and finger to accelerate rapidly meaning there is no longer a stable structure to support the ‘squeeze’ force; the index finger quickly tends towards hyperextension at its joints, the thumb towards flexion.” (Yoxall et al., 2013). A significant increase in joint stress was observed when the turn sequence is engaged and may explain why users find the dissimilar and simultaneous motion of squeeze and turn difficult (Yoxall et al., 2013). A study that analyzed a press and turn sequence using the modeling approach was not found during the course of our literature review, but isolating these motions and understanding the impact dissimilar and simultaneous motions have on each other likely has critical implications for CRP design. 35 Container’s lug Cap’s lug Container’s lug Cap’s lug Figure 11: Opening sequence diagram for a squeeze and turn closure. Squeeze and turn closure operating sequence for opening. Reprinted from International Journal of Industrial Ergonomics, Vol 54, Gabriel H.C. Bonfim, Fausto O. Medola, Luis C. Paschoarelli, Correlation among cap design, gripping technique and age in the opening of squeeze-and-turn packages: A biomechanical study, Pages 178-183, Copyright (2016), with permission from Elsevier. 2.4 Summary ISO 8317 panel testing and ISO 13127 mechanical testing were identified as standard test methodologies that provide reference data that can be used as performance metrics relevant to ergonomic design characteristics for reclosable child resistant containers. Torque was identified as the dependent outcome variable influenced by application forces (unidirectional [vertical and lateral] compression and tension forces) that users generate employing gripping strategies. Testing methods utilized in relevant research studies include video-recorded ethnographic studies and user studies (Bonfim et al., 2016; Rowson & Yoxall, 2011; Yoxall et al., 2007; Yoxall et al., 2013). A gap in knowledge that was identified during the course of our literature review was that no publications identify grip choices for young children under the age of 5 years when interacting with cylinder child-resistant 36 packaging systems. Additionally, studies that evaluate how the application forces involved in dissimilar and simultaneous motions function as a child-resistant mechanism and their impact on ease of use are needed. Table 12: Literature review summary of goals. Research Goal(s) Summary Literature Review • Identify performance metrics and user interaction patterns related to the problem • Performance metrics: application of force (unidirectional [vertical and lateral] compression and tension forces) and torque. • User interaction patterns: grip strategies when applying torque • Identify relevant gaps in knowledge • (Pilot Study) What grip types do young children use when interacting with reclosable cylinder closure child-resistant packaging? • Hypothesis: Young children under the age of 5 years utilize fewer grip types when interacting with cylinder type child- resistant packaging than adults use 37 CHAPTER 3: PILOT STUDY Table 13: Pilot study goals. Research Pilot Study Goal(s) • Gather data to address identified knowledge gaps: o What grip types do young children use when interacting with reclosable cylinder closure child- resistant packaging? o Hypothesis: Young children under the age of 5 years utilize fewer grip types when interacting with cylinder type child-resistant packaging than adults use • Analyze results • Draw conclusions 3.1 Pilot Study To fill the identified gap in knowledge related to children’s use of varied grip styles with CRP containers, we reviewed video previously collected by our lab of children interacting with CRP. This post-hoc review was conducted to test the hypothesis that children under the age of 5 years do not utilize all seven styles of grips identified by Yoxall et al. 2013 when attempting to apply torque to a CRP continuous thread cylinder closure. Identification and comparison of grip types utilized by young children was used to aid in the development of a design methodology that considers both child-resistance and adult ease of use by understanding the differences in grip strategies between children and adults. Data originally collected during the summers of 2004 and 2005 (Sight, Sound and Child Resistance-S/S/CR IRB i019713) comprised of 38 100 children between the ages of 3.4 to 5 years interacting with push down and turn CRP was reviewed; grips were recorded and utilized as a pilot study for this work. 3.1.1 Pilot Study Aims Our broad goal was to develop a pilot understanding of the grip styles employed by young children (aged 3 to 5 years) when interacting with child-resistant, push and turn packaging. With this in mind, we had the following proximal aims: 1. Conduct a post-hoc review of archived video footage from our “Sight, Sound and Child Resistance Study” (S/S/CR) (IRB i019713) to identify grip types young children utilize when applying torque with push down and turn child-resistant packaging (CRP) and 2. Compare the identified child grip types with adult grip types previously identified while using push down and turn CRP in another study (Yoxall et al., 2013). 3.1.2 Materials and Methods The S/S/CR study (IRB i019713) adapted the Testing Procedure for Special Packaging dictated by 16 CFR 1700.20 (IRB i019713). The two-part study was conducted and video-recorded in 2004-2005 utilizing children aged 3 to 5 years who interacted with push down and turn, child-resistant packaging (closure/container diameter 42mm, closure height 16mm). Video footage of part two was reviewed and analyzed for the purpose of our present study. 3.1.3 Archived Study S/S/CR Methodology - subject of analysis The first part of the S/S/CR study was conducted in 2004 and utilized child- resistant push down and turn closures that were glued to the container which prevented the packaging from being opened. The second part of the S/S/CR study was conducted in 2005 where the closures were not glued to the container and could be removed by test participants. 39 Part two of the S/S/CR study recruited 100 children between the ages of 3-5 years that had no physical or mental disabilities that would have impacted their ability to open packages. Participants were recruited and scheduled by Great Lakes Marketing (GLM) and research was conducted at various daycares in Toledo, Ohio. Two hypotheses were investigated in the original S/S/CR study 1. an opaque (continuous thread child-resistant pharmacy vial with a push down and turn closure) package will provide better child-resistance than a transparent package and 2. a package that muffles the sound of the product will provide better child- resistance than a package that allows the product to rattle. Four treatments were tested: Treatment (A) a transparent container that rattles, treatment (B) a transparent container that doesn’t rattle, treatment (C) an opaque container that rattles, and treatment (D) an opaque container that doesn’t rattle. The types of closures and containers were numbered by treatment so they were identifiable on the video. Daycare teachers brought two children into a testing room at the test site. Each child was assigned a subject number, and their sex and date of birth was recorded prior to testing. Children were tested in pairs and asked to sit down in front of a researcher who instructed the children throughout testing, while two other researchers videotaped the test. Each child sat on their own individual rectangular piece of carpet that was placed on the floor. Each of the children was provided with one CRP and both children were given the same treatment. Treatments were rotated (first pair of children received treatment A, and the next pair received treatment B, and so on). The children were told that the test is not a contest or a race and that there are no rewards. They were also told that there 40 is not a right way to open these packages and that they should feel free to open the packages in any way they like. The children were then instructed by a researcher to “Please try to open the package”. The researcher did not continually encourage the children to interact with the package. The researcher working with the children used a stopwatch was used to keep track of the total time the children interacted with the treatments. Another researcher (working with the camera equipment behind a one-way mirror) used a camcorder to record the two children being tested. The children were seated so that they were facing the camera with their backs to the wall. The camcorder screen had a counter which showed and recorded the time that each child spends trying to open a given treatment so that the precise time that each child spends with each package was accurately documented. If the child opened the package or indicated they wanted to stop testing, testing was ended at that point. Once the testing was ended the children were debriefed with the following message, “Never open packages like this when you are by yourself. This kind of package might have something in it that would make you sick”. The children were then escorted back to their classrooms. 3.1.4 Methodology for analyzing the S/S/CR study Available study data included archived video files (.avi format, 640x480 resolution), IRB documentation (IRB i019713) and a flat file capturing demographic information, and information regarding whether or not the child successfully opened the package, opening times, time of total interaction with packaging, and treatment information. Other information such as camcorder model, package container height, and container/closure material composition were not available. 41 Archived video footage of children tested in pairs from the Sight, Sound and Child Resistance study, where closures were not glued (part II), was reviewed and analyzed for grip types. Of the fifty video files comprised of 100 children attempting to open push and turn vials (closure/container diameter 42mm, closure height 16mm), fifteen video clips contained 19 children (aged 41-67 months) who were able to open their packages. (Some clips contained a successful pair, other videos contained only a single successful child). Each video was approximately 5 minutes in length unless both children opened the package before 5 minutes expired. Video footage was reviewed on an ASUS model N43J laptop with a 14” display resolution of 1366x768 using Windows Media Player Classic software version 1.7.1. The goal of the pilot study was to identify the grip types young children utilize when interacting with a cylindrical type, child-resistant package, however the action of opening the package terminates exploration of other grips. Therefore, an additional 25 videos containing footage of 50 children (aged 41-65 months) that were unable to open their vials were randomly selected for review to identify grip types used throughout the 5-minute test period. A flat file (see Appendix B) capturing demographic information, whether or not the child successfully opened the package, opening times, time of total interaction with packaging, and treatment information created by researchers who conducted the S/S/CR study also informed the present study. Children reviewed in our study ranged from 41 to 67 months (3 to 5 years) in age with 38 children identifying as male and 31 children as female. 42 Spherical Cylindrical Pronated Cylindrical Pulp Box Lateral Tip Figure 12: Grip types related to grasping cylinder closure packaging. Seven categories of grip types (see Figure 12) identified and utilized in previous studies (Bonfim et al., 2016; Rowson & Yoxall, 2011; Yoxall et al., 2007; Yoxall et al., 2013) were used to identify the grips employed by children in the development of this data set. Each video file contained audio and video playback and was played until one of the seven identifiable grip type was observed being used by a child to apply torque; video was reviewed a minimum of two times (focusing on a single participant in the pair during a single review). The video playback was paused and advanced frame by frame to identify and confirm grip types. Every grip type observed to apply torque to the closure employed by each child during the course of the 5-minute video was compared to Figure 12 and coded in binary fashion (occurred yes/no). If the package was opened, video playback was rewound and resumed to positively identify which grip was (G employed to successfully defeat the push and turn mechanism. Although, the grip type ) used on the container, itself, was not identified in previous studies (Bonfim et al., 2016; Rowson & Yoxall, 2011; Yoxall et al., 2007; Yoxall et al., 2013), herein we attempted to characterize it (as well as what was utilized on the closure) to garner potential, additional insights. 43 3.2 Pilot Study Results 3.2.1 Children who were unable to open CRP Video footage of 50 children (aged 3 to 5 years) who were unable to open their packages over the course of a full 5 minutes was reviewed. Three grip types (spherical, pronated cylindrical, and cylindrical) were identified for the closure using the established taxonomy previously discussed (see Figure 12). When multiple grip types were used by a single child, each was coded as being utilized by that participant. Spherical grips were used by all 50 children on the push and turn closure while attempting to open their packages (see Figure 13). Additionally, 39 of the 50 children also used a cylindrical grip and 18 children also used a pronated cylindrical grip. 44 n e r d l i h c f o r e b m u N 60 50 40 30 20 10 0 Grip type Closure grips attempted by children (3-5 years) when unable to open CRP/CRC 22 28 19 20 11 9 Spherical Cylindrical Pronated cylindrical Female Male Figure 13: Frequency of children who were unable to open CRP (closure grips identified). Number of children (aged 3 to 5 years) who utilized each of the seven specific cylindrical grip types (Rowson & Yoxall, 2011) on the closure when interacting with a 42 mm diameter push down and turn child-resistant package. All subjects were unable to open the closure over a 5-minute time period and some children used multiple grip types (n = 50, 22 females and 28 males). Grip types in opening attempts when video from 50 children (n=22 females and n=28 males) was reviewed (see Figure 14). Grip types were coded first by the grip type used on the container to apply torque, then by the grip type used on the closure to apply torque (Spherical = S, Cylindrical = C, and Pronated Cylindrical = PC). All 50 children who were unable to open their packages attempted using a C/S (cylindrical grip on the container and spherical grip on the closure) grip type to try to open their package. Multiple grips were attempted by children such that: 37 children also used a C/C grip, 35 children used a S/S, 10 used a C/PC grip, and 6 used a S/C grip to try to open the package (see Table 14). 45 Grips attempted on container/closure by children (3-5 years) when unable to open CRP/CRC C/S C/C S/S C/PC S/C n e r d l i h C 60 50 40 30 20 10 0 Grip types Figure 14: Frequency of children who were unable to open CRP (closure and container grips identified). Number of children (aged 3 to 5 years) who attempted specific cylinder grip types (Rowson & Yoxall, 2011) on the container and closure when interacting with a 42 mm diameter push down and turn child- resistant package. All subjects were unable to open the closure over a 5-minute time period. Grip types were coded first by the grip type used on the container, then by the grip type used on the closure (Spherical = S, Cylindrical = C, and Pronated Cylindrical = PC). 46 Table 14: Children who were unable to open CRP (closure and container grips identified). Number of children (aged 3 to 5 years) attempted specific cylinder grip types (Rowson & Yoxall, 2011) on the container and closure when interacting with a 42 mm diameter push down and turn child-resistant package. All subjects were unable to open the closure over a 5-minute time period. Grip types were coded first by the grip type used on the container, then by the grip type used on the closure (Spherical = S, Cylindrical = C, and Pronated Cylindrical = PC). Female children ranged in age from 42 to 65 months and male children were aged 41 to 60 months (n=50). Container Cylindrical Spherical Cylindrical Spherical Cylindrical (grip) Closure Spherical Spherical (C/S) (S/S) (C/PC) (S/C) (grip) Males Females Total Female age (months) Average age (months) Standard deviation Males age (months) Average Age (months) Standard deviation Cylindrical Cylindrical Pronated Cylindrical (C/C) 27 23 50 20 15 35 42-65 42-65 48.86 48.64 6.57 6.68 9 11 18 42-58 46.72 4.78 18 19 37 42-65 49.28 7.19 5 2 7 65 65 0 41-60 41-60 41-57 42-52 41-60 48.81 48.15 46.89 47.00 48.39 5.42 5.00 4.51 3.81 5.59 47 3.2.2 Children who opened CRP Only the grip type used to successfully open the package was recorded. Two grip types, pronated cylindrical and spherical, were identified as the grips used to defeat the push and turn mechanism when reviewing video footage of the 19 children who were able to open their packages (see Figure 15). Spherical grips comprised the vast majority of successful openings with 17 of the 19 children who opened their package utilizing this grip to open the closure, while 2 children (tested at the same time) used a pronated cylindrical grip to open their packages. Closure grips used by children (3-5 years) when opening CRP/CRC n e r d l i h c f o r e b m u N 18 16 14 12 10 8 6 4 2 0 6 11 Spherical Pronated cylindrical 2 Grip type Female Male Figure 15: Frequency of children who successfully opened CRP (closure grips identified). Number of children (aged 3 to 5 years) who successfully opened a 42 mm diameter push down and turn child-resistant package and the specific grip type they utilized to open the closure (n = 19, male= 11, female = 8). Grip types used on the container of the package during opening were also identified. Grip types were coded first by the grip type used on the container, followed by the grip type used on the closure (Spherical = S, Cylindrical = C, and Pronated 48 Cylindrical = PC). Fourteen of the 19 children who opened their packages used a C/S (cylindrical grip on the container and spherical grip on the closure) grip type, 3 used a S/S grip, 2 used a C/PC grip, and none of the 19 children used a S/C or C/C grip to open their package (see Figure 16). Additional demographic information, opening time, grip types, age and gender data are provided in (Table 15). Grips used on container/closure by children (3-5 years) when opening CRP/CRC n e r d l i h c f o r e b m u N 16 14 12 10 8 6 4 2 0 Grip type 9 5 C/S 2 1 S/S 2 C/PC Male Female Figure 16: Frequency of children who successfully opened CRP (container and closure grips identified). Number of children (aged 3 to 5 years) who successfully opened a 42 mm diameter push down and turn child-resistant package and the specific grip types they utilized on the container and closure to open the package (n = 19). Grip types were coded first by the grip type used on the container, then by the grip type used on the closure (Spherical = S, Cylindrical = C, and Pronated Cylindrical = PC). 49 Table 15: Children who successfully opened CRP. Data for 42 mm closure diameter child-resistant push and turn package: opening time, grip types, age and gender. Grip types were coded first by the grip type used on the container, then by the grip type used on the closure (Spherical = S, Cylindrical = C, and Pronated Cylindrical = PC). Push and turn package (42 mm closure diameter) Opening time (min/sec) (container, closure) Gender Age (months) Grip type 0:53 2:13 1:59 3:20 0:08 0:24 1:12 2:18 4:18 2:05 1:52 3:34 3:18 0:09 1:06 2:30 0:31 1:31 4:41 C/PC C/PC C/S C/S C/S C/S C/S C/S C/S C/S C/S C/S C/S C/S C/S C/S S/S S/S S/S F F M M F F M F F M M M M F M M M F M 67 66 47 46 48 60 43 44 50 56 64 60 65 41 48 50 50 60 50 3.3 Pilot Study Discussion Children between the ages of 41 to 67 months utilized 3 grip types (spherical, cylindrical, and pronated cylindrical grips, see Figure 12) when gripping the closure and applying torque to a push and turn 42 mm in diameter. Of the three closure grip types utilized by children, only two of those grip types, spherical and pronated cylindrical, resulted in successful openings (see Figure 15). In the Rowson and Yoxall (2013) study, adults utilized 6 of the 7 identified grip types when interacting with child-resistant push and turn closures applied to bleach bottles. Results support the hypothesis that children 50 aged 3 to 5 years utilize fewer grips when attempting to open these cylinder-type closures. Unlike previous studies which focus solely on the closure hand and utilize adults, we report grip types on both the container and the closure with children in this study. The most frequently occurring grip type per participant was a spherical grip on the closure and a cylindrical grip on the container (see Figure 14, C/S). Children had the most success opening cylindrical type child-resistant packaging with this grip type (see Figure 16). However, children were also successful in opening their packages using a pronated cylindrical grip on the closure and a cylindrical grip on the container (see Figure 16, C/PC), and spherical grips on both the container and closure (see Figure 16, S/S). This seems to suggest there are multiple biomechanical actions based on grip choices that children can employ to overcome the push and turn child-resistant closure. Child-resistant packaging is a balance of providing a barrier for young child and facilitating adult ease of use. Literature (Yoxall et al., 2013) indicates adults may commonly utilize 6 of the 7 identified (see Table 16) grip types: spherical, cylinder, pronated cylinder, pulp, box, and lateral grips when interacting with cylindrical type child-resistant packaging. Not all adults can utilize all these grip types and some may be more comfortable or effective than others (Rowson & Yoxall, 2011). Children aged 3 to 5 years that we observed only commonly used three (cylinder, pronated cylinder, and spherical for both the closure and container) of the 7 identified (see Figure 3) grip types when interacting with cylindrical type child-resistant (push and turn continuous thread) packaging in this study. 51 As such, cylindrical closure type CRP that restricts young children from effectively using spherical, cylinder, and pronated cylindrical grips may be beneficial in limiting access to young children. For example, children in the 3-5 age group using a bi-digit grip (see Figure 10) produced the lowest torque of all age groups and grip types children in the Bonfim study (Bonfim et al., 2016). However, it is still important for any potential child-resistant mechanism to facilitate adult ease of use by retaining the ability for adults to utilize grip types that adults find comfortable and effective. Design of cylindrical type child-resistant packaging mechanisms should consider the biomechanical interactions of spherical, cylindrical, and pronated cylindrical grips (see Figure 12) of young children and the forces they can impart to potentially defeat a child- resistant mechanism. Further identification and comparison of grip types utilized by young children and adults may aid in developing design methodologies that facilitate simultaneous child-resistance and adult ease of use by understanding the differences in contextual grip strategies between children and adults. 52 Table 16: Frequency of adult user grip choices when opening 28 mm cylindrical squeeze and turn (SaT), and push and turn (PaT) closures. Data referenced from (Yoxall et al., 2013). Child (aged 41-67 months) grip choices successfully opening PaT closures (S/S/CR). Child (aged 41-60 months) grip choices explored when failing to open and successfully opening PaT closures (S/S/CR). Cylindrical Shperical Lateral Pronated Cylindrical Pulp (D) Box (E) (F) Tip (G) Grip types Adults (PaT) Adults (SaT) (A) (B) (C) n = 57 adults (unspecified ages) Grips utilized to open (28mm closure) 6 - 7 6 1 3 2 7 22 3 19 38 n = 19 children, aged 41-67 months. Grips utilized to open (42mm closure) - - - - Children (PaT) 17 - 2 - - n = 50 children, aged 41-65 months. Grips explored when unable to open (42mm closure) 50 39 18 - - Children (PaT) - - 3.4 Pilot Study Conclusions The aims of this study were to identify the type of grips young children aged 3 to 5 years utilize (1) when interacting with push and turn CRP and (2) compared to (see Figure 12 and Table 11) the grip types adults used in a another related, published study (Yoxall et al., 2013). Children in this study only used three grips, spherical, cylindrical and pronated cylindrical, on both the closure and container (a 42 mm push and turn closure) over the course of 5 minutes of interaction. When grip combinations (container and closure) were examined, 5 types of grips strategies emerged: C/S (cylindrical grip 53 on the container and spherical grip on the closure), C/C, S/S, C/PC, S/C and C/S grips (see Table 14). Children were able to open push and turn closures using three combinations S/C, S/S, and C/PC grips (see Figure 16). Adults in the Yoxall et al. (2013) study used 6 of the 7 grip types (only the tip grip was not utilized, see Table 10) when attempting to open a 28 mm diameter child- resistant push and turn bleach container package. However, in our study, it was determined that young children only utilized 3 of the 7 grip types (spherical, cylindrical, and pronated cylindrical grips, see Figure 13) when interacting with a 42 mm diameter push and turn closure for a period of 5 minutes. Further, only two of those grip types employed on the closure, spherical and pronated cylindrical, resulted successful openings. Design of child-resistant mechanisms that restrict young children from effectively using spherical, cylindrical and pronated cylindrical grip types may be beneficial for cylindrical type child-resistant packaging provided adults retain the ability to utilize grip types they find comfortable and effective. 3.5 Pilot Study Limitations A major limitation of our study is the post-hoc nature of the work. Specifically, the review of the article by Yoxall et al. providing a taxonomy of different grips used by adults interacting with CRP/CRC systems led us to postulate that perhaps children were more limited in their approach (and that this could be leveraged to the benefit of CRP/CRC system design). As a preliminary exploration of this, we (post-hoc) reviewed a large data set comprised of videos of young children interacting with CRP/CRC to characterize the frequency of use of the grip styles identified by Yoxall (using adults as participants). The benefit of the post-hoc review is that it provided a relatively large 54 sample size of young children interacting with CRP/CRC (a protected class of participants that can be difficulty to reach); that said, its use does present limitations. Given the post-hoc, exploratory nature of the work, there is a confound regarding the container/closure systems employed in the two studies. Specifically, the child data that we reviewed utilized a larger (42 mm) closure, compared to the 28 mm closure used in the adult study (Yoxall et al., 2013). Research has shown that smaller diameter closures may afford more grip choices (Barry, 2009; Lin, Chen, Lin, & Hung, 2015). Grip choices can also be influenced by the size of the individual’s hand and the size of the closure. This could not be further explored due to a lack of data such as the container/closure height, texture, material composition, the shape and spacing of the closures’ gripping ridges, and a lack of hand anthropometric data on the hand size of the children and adults in both of these studies. Additional research related to grip choices children utilize when interacting with cylindrical type child-resistant packaging across a range of closure diameters is needed. 55 Table 17: Pilot study summary of goals. Research Goal(s) Summary Pilot Study • Gather data to • 69 children (19 that address relevant identified knowledge gaps • Analyze results opened CRP and 50 that could not) under the age of 5 years were video- recorded interacting with CRP and their grip types were identified • Hypothesis: Young children under the age of 5 years utilize fewer grip types when attempting to apply torque on cylinder type child-resistant packaging than adults • Young children under the age of 5 years use fewer grip types (3) on the closure when interacting with push and turn CRP than adults (6) use • Draw conclusions • CRP that restricts children from effectively using spherical, cylindrical and pronated cylindrical grips may be beneficial 56 CHAPTER 4: DESIGN APPROACH Table 18: Design approach goals. Research Goal(s) Design Approach 4.1 Design considerations Develop a design approach that addresses gaps in knowledge: • CRP that restricts children from effectively using spherical, cylindrical and pronated cylindrical grips may be beneficial o not significantly impact adult gripping strategies An ideal CRP paradigm would restrict access for unintended users (i.e. children) without significantly impacting accessibility with intended users (i.e. adult users of medication) (Blok et al., 2016; Imrhan, 1994). This can be achieved if a CRP mechanism can functionally separate the opening capabilities of young children (unintended users) from adults (intended users). Designing a solution may seem feasible from an engineering perspective, however understanding the mechanisms required to operate the CRP has the potential to influence the outcomes of both types of users. Adults may get confused with complicated directions and children may figure out simplistic instructions, designs that are overly intuitive, or can be easily defeated with trial and error. As a result, the solution needs to employ a cognitive strategy that is easy to explain and operate for adults without being readily apparent to young children. Other design considerations include the need for a solution that can be economically produced with current manufacturing processes and machinery (including at high speeds), that is capable being integrated into a variety of pharmacy systems (e.g. central fill pharmacy, online pharmacy, retail pharmacy, institutional pharmacy). For this 57 initial conceptual design phase: (1) the design’s dimensions were kept comparable to currently available products on the market and (2) operational characteristics (continuous thread closure) of the design were selected so that existing capping capabilities could be utilized. It is worth noting injection and compression molding manufacturers were engaged early (consultation is ongoing) in the design process to ensure the design can be adapted to at least one of these processes, however the design needs to be further developed to specifically tailor it to these manufacturing processes. 4.2 Design hypothesis rationale Based on our literature review and pilot study, we speculated that (1) the hand torque (Replogle, 1983) strength of 4 year-old boys and girls may be comparable to male and female senior (62-92 years) adults. Additionally, we postulated that (2) both grip choice and hand size (related to closure diameter) could be explored as possible factors to build a child-resistant mechanism that can distinguish the hand torque generation capabilities of intended users from unintended users. 4.2.1 Hand regions used for grip contact The major regions of the hand (see Figures 17 and 18) that typically makes contact with a closure system in order to grip and apply torque includes: the fingers, thumb, the abductor pollicis muscles. 58 Figure 17: Regions of the hand. Identified major regions of the hand involved in gripping objects. Palm, Fingers, Thumb, abductor pollicis muscles between the index finger and thumb (Kozin, Porter, Clark, & Thoder, 1999; Kroemer, 1986). Specifically, the thumb, index finger and APM are typically the contact regions of the hand involved in using a cylindrical grip on a closure (Kozin et al., 1999; Kroemer, 1986). Figure 18: Example of a cylindrical grip on a closure. Illustration of the abductor pollicis muscles, index finger, and thumb involved in cylindrical grip when contacting a container closure. 59 It’s worth noting there are many variations of the seven grips (see Figure 12) identified in literature (Rowson & Yoxall, 2011), including a cylinder grip where the middle finger and palm make closure contact when gripping rather than the index finger and APM. For this reason, the design developed herein is focused on the generalized regions of the hand that are commonly involved in all grip types and variations. Two of those regions, the fingers and thumbs, require greater fine motor control than the palm and APM because each finger and thumb are controlled independently and are composed of multiple joints (Hume, Gellman, McKellop, & Brumfield, 1990; Maas, Veeger, & Stegeman, 2018). 4.2.2 Thumb and fingers grip contact When contacting an object with the thumbs and fingers, the phalanges (bones) of the fingers and thumbs are the support structures that compress the skin (Maceo, 2009; Pawluk & Howe, 1999). Since these bones are approximately cylindrical in shape, the critical contact area is near the midline of the bone (Ikeda, Kurita, Ueda, Matsumoto, & Ogasawara, 2004; Pawluk & Howe, 1999). The further away from the midline of the bone, the weaker the contact forces are (Maceo, 2009) because the bone (i.e. the support structure) curves away from the object surface and the skin is less compressed (see Figure 19). 60 Figure 19: Example of thumb or finger contact with a flat surface. Conceptual illustration of the bones of the fingers and thumbs as the support structures that compress the skin when contacting an object. The further away from the midline of the bone, the weaker the contact forces are (Maceo, 2009) because the bone curves away from the object surface and the skin is less compressed. Due to the smaller size of childrens’ fingers it can be conjectured that a child’s hand to object-surface contact area is significantly less than an adult-sized hand. As a frame of reference, the average middle finger length and breadth of four-year old (male and female) children is approximately 40% less than adult females’ (aged 18-56 years) (Garrett, 1970; Snyder, 1975). As a result, a child gripping with only the thumb and fingers is likely to have a significantly smaller surface area to compress skin and transfer useful forces for gripping. As such, we postulate their grip will be less effective compared to the larger, adult-sized hands using the same type of grip (e.g. finger and thumb). In light of this, we theorized that children may instead primarily rely on the large area of the palm and/or the elastic APM to grip because these regions require less 61 specific fine motor control (CPSC, 2002). In this scenario the thumb and fingers would serve as ancillary structures that help support the APM and palm when gripping. In general, it can be conjectured that larger fingers (supported by larger phalanges) will more efficiently generate torque than smaller fingers because larger fingers will have more bone-supported skin contact with the gripping surface of a cylindrical object. 4.3 Design Hypothesis Our literature review shows that adults commonly use 6 (A-F) grip types (see Figure 16) when interacting with push and turn-type cylindrical medication containers, and our pilot data suggests that young children typically only utilize three grips (A-C). It has been suggested that children are limited in approach due to underdeveloped motor skills (CPSC, 2002) and the small size of their hands (Garrett, 1970; Snyder, 1975). We conjecture that the 3 grip types children were observed to use are likely due to the fact that each primarily utilizes the palm (spherical grip, Figure 20A) and the APM (cylindrical and pronated cylindrical grips, Figure 20B and 20C), not the fingers and thumb. The use of these three grips would enable children (who have smaller hands as previously discussed) to generate their maximum torque. 62 (A) Spherical (B) Cylindrical (C) Pronated Cylindrical (D) Pulp (E) Box (F) Lateral (G) Tip Figure 20: Grip types related to gripping cylindrical closure packaging. Grip types identified by (Rowson & Yoxall, 2011). Considering this, we postulated that it would be beneficial to design a cylindrical closure that targets the palm and the APM regions of the hand by preventing these grip styles/regions of the hand from effectively contacting the closure while allowing the larger, adult-sized fingers and thumbs to efficiently contact the closure surface area (see Figure 19). In other words, without effective palm and APM contact smaller child- 63 sized hands may struggle to generate enough thumb and finger object surface area contact to generate sufficient torque. Design hypothesis: Limiting the ability of participants to use grips which rely on the abductor pollicis muscles (APM) or those that utilize palm contact with the working features of a closure system will significantly reduce the torque children are capable of generating (unintended users) without significantly impacting torque adults are capable of generating (intended users). 4.4 Design development rationale A two-piece continuous thread screw cap closure was selected as the basis for design development with the following rationale: Most cylindrical child-resistant packaging systems either use (1) a one-piece closure mechanism that also requires a paired mechanism on the container (see Figure 21) or, (2) use a two-piece closure mechanism with a paired mechanism on the container (see Figure 22) or (3) a two-piece closure with a paired mechanism on both caps of the closure (see Figure 23). 64 Figure 21: One-piece closure with pared cap and container mechanisms. (Left) Child-resistant “Squeeze and turn” packaging system with lug style mechanisms paired between the one-piece closure and container. (Right) Child-resistant “Line up arrows” packaging system with arrow mechanisms paired between the one-piece closure and container. Figure 22: Two-piece closure with paired cap and container mechanisms. (Left) Child-resistant “Push and turn” packaging system with lug style mechanisms paired between the two-piece closure (red cap is nested inside the white overcap) and container. 65 Figure 23: Two-piece closure with paired cap mechanisms. (Left) Child-resistant “Push and turn” packaging system with mating mechanisms paired between the caps of the two-piece closure (a translucent cap is nested inside a white overcap). The ridges of the translucent cap interlock with the ridges of the white cap when the overcap is pushed down, allowing both caps to rotate simultaneously. A fourth, theoretical, mechanism would consist of a one-piece closure without a paired mechanism on the container. It would be an ideal solution from material cost (only one- piece closure), manufacturing, and industry integration perspectives. A one-piece closure would (theoretically) be comprised of less material relative to a two-piece closure, less complex to manufacturer, and could adapt to currently existing threaded containers because a paired mechanism would not need to be added to the container. It is, however, unlikely a cap and container packaging system design solution can be engineered that is functionally child-resistant and simultaneously senior adult friendly without utilizing paired mechanisms (either mechanisms paired between the cap and container or mechanisms paired between caps on a two-piece closure). A child-resistant mechanism seemingly needs to have a conditional change in how it operates when an 66 intended user interacts with it versus when an unintended user interacts with it. This conditional change is typically executed by movement of the mechanism. For example, a continuous thread, two-piece push and turn mechanism involves the overcap being pushed down (conditional change) to mate with the inner cap (see Figure 23). The mating of the paired mechanisms allows the inner cap to be unscrewed by rotating the overcap. If the overcap is not pushed down (conditional change) then the inner cap will not rotate. Another example is how the squeeze and turn mechanism works. It is typically a one-piece lug cap with a paired lug-style mechanism on the container (see Figure 21, left). The mechanism operates by squeezing opposing sides of the cap until it deforms (conditional change) and the paired lugs on the cap and the container either disengage or cannot engage as the cap is rotated. Of the four approaches mentioned above, a two-piece closure with a paired mechanism between the inner cap and overcap is the most practical to develop. This is because a paired mechanism is not required for the container. A closure solution of this type will have the capability to adapt with currently available threaded containers and therefore likely have the greatest potential for large scale integration. 4.5 Design Theory Our design goal was to develop a system that precluded the palm and APM regions of the hand from effectively contacting the functional mechanisms of the closure design. Our approach to this goal utilized a two-piece closure consisting of an inner cap nested inside an overcap. Specific areas of the overcap were “cut away” exposing areas of the inner cap; the cut-aways, or functional exposure area, enabled a deliberate portion of the inner cap to be gripped using the fingers and thumbs (see Figure 24). 67 Figure 24: Conceptual two-piece closure design. Areas of the overcap are cut away and expose corresponding areas of the nested inner cap. The areas of the overcap that are not “cut away” shield the inner cap so that there is not enough exposed surface area of the inner cap to be effectively gripped using the APM and palm regions of the hand (see Figure 25). The overcap freely rotates while the threaded inner cap is stationary once engaged with the container. The rigidity of the material used to form the overcap structure does not allow a grip applied to the overcap to aid closure removal. Specifically, the grip types and hand regions we want to restrict are the APM region of the hand with cylinder and pronated cylinder grips, and the palm with a spherical grip (see Figure 25). 68 Figure 25: Grips shielding of conceptual two-piece closure design. The overcap shields the palm from contacting the inner (white) cap and creates a bridged gap between the abductor pollicis muscles (APM) region of the hand and the inner cap. In theory, these restrictions will make it necessary for an individual interacting with the closure to rely on their thumb and fingers to effectively and efficiently grip the inner cap. Therefore, we postulated smaller child-sized hands will have limited thumb and finger to (inner) cap surface area contact, due to the smaller phalanx’s (as discussed previously). In turn, this will limit the amount of torque they can generate, while larger adult thumbs and fingers will be able to more efficiently contact the inner cap and, subsequently, apply a greater amount of torque. 4.6 Design Approach To better understand and characterize the performance of the concept, several versions of the design allowed us to test across a spectrum that ranges from adult friendly (least restrictive functional area) to child-resistant (most restrictive functional 69 area). A one-piece continuous thread screw cap served as the control (no restriction, 100% cap surface area exposure) (see Figure 26). Figure 26: Control design. One-piece continuous thread closure without surface area gripping restriction. Figure 27: Design concept bridge. The overcap’s lateral bridge runs across the inner white cap to interrupt efficient midline gripping of the thumbs and fingers. Our concept design was modified to further limit the amount of functional surface area available by interrupting the midline of the thumbs and fingers during grippping. This was accomplished by creating a lateral bridge across the “cut away” section (cut away exposed the functional gripping area of the innercap) of the overcap (see Figure 27). Several variations of the design were created that imposed restrictions on the 70 amount of inner cap surface area, the functional grip area that was exposed and available for gripping. These variations ranged from the least restrictive control (100% inner cap surface are exposure) to the most restrictive design (approximately 13.3% inner cap surface area exposure) (see Figure 28). Figure 28: Design concept variations. Overcap design variations each allow access to a different amount of surface area in which to grip the inner white cap. 71 Two treatments of interest were developed (exposure- single and double sided) were crossed with bridge (present and absent) with two levels of bridge width (1mm and 2mm). That said, the functional exposure area (the white inner cap areas above and below the bridge) was held constant for the treatments which had bridges. This was done to isolate and evaluate the efficacy of the varying the bridge width so that the effect of a 2 mm bridge versus a 1 mm bridge could be assessed. Both intended (adults) and unintended users (children) should be able to demonstrate their maximum hand torque capabilities with the control design. We hypothesized a progressive degradation in the ability to generate torque across designs as a function of limiting functional surface area. If we observe a point at which unintended users’ hand torque drops significantly without intended users’ hand torque also dropping significantly, we will have developed evidence that supports our design hypothesis. The design can then be further developed and optimized. 72 Table 19: Design approach summary of goals. Research Goal(s) Summary Design Approach Develop a design approach • A design concept was that addresses gaps in knowledge: • CRP that restricts children from effectively using spherical, cylindrical and pronated cylindrical grips may be beneficial developed that restricts the APM region of the hand from making functional closure contact when using cylinder and pronated cylinder grips and restricts the palm with a spherical grip. • Children may be limited to primarily using their thumbs and fingers to grip and therefore may be less effective at generating torque • not significantly • Compared to young impact adult gripping strategies children (< 5 years), adults may be able to generate torque more efficiently with their thumbs and fingers due to their larger hand size 73 CHAPTER 5: DESIGN EVALUATION STUDY Table 20: Design evaluation study goals. Research Aim Goals(s) Design Evaluation Study • Evaluate design based on identified performance metrics (torque), user interactions patterns (grip type), pilot study data analysis: o Children may be limited to primarily using their thumbs and fingers to grip and therefore may be less effective at generating torque o Compared to young children (< 5 years), adults may be able to generate torque more efficiently with their thumbs and fingers due to their larger hand size • Report significant findings • Describe implications of results • Summarize findings 5.1 Design hypothesis Our hypothesis is that if the overcap shields specific portions of the inner cap from being gripped, then the maximum torque that can be applied to the inner cap will be limited. If this postulate holds, the functional surface area (area where the inner cap is exposed) could be “tuned” to allow some users entry, while precluding others. The overarching aim of this design evaluation study was to identify how limiting the functional, exposed surface area imposed by our design (a two-piece, inner cap and overcap packaging system; patent pending: U.S. application no. 16/421,631) impacts 74 the ability of users with varied hand sizes to exert a rotational hand torque enabling the removal of a cap from a continuous thread bottle. Explicit objectives presented in this chapter were to develop data for 4-year-old children and senior adults (65+) of various hand sizes (related by thumb breadth measurement) that: 1. examines the effect of limiting the functional surface area of the top and lateral sides of the inner cap on rotational torque that can be generated, and 2. examines the efficacy of a bridge to limit functional surface area 5.2 Materials and Methods Seven, white HDPE (Alpha Packaging) medication bottles (120 cc capacity, 38- 400; 38-millimeter neck diameter and 400 style neck finish) were labeled ‘1’ through ‘7’ and paired with seven cap designs (a one-piece cap that served as a control and 6 two- piece caps that represented the varied experimental treatments) (see Figure 28). The one-piece cap (control) and the 6 two piece caps (experimental designs) were each tightened (30 lbs-inch +/- 0.4) with a Secure Pak Digital Torque Tester and glued (JB Weld ClearWeld 5-minute epoxy, tensile strength 4400psi) to the neck of bottles 1 through 7, such that only the overcap of the two-piece caps was capable of rotation. The one-piece (38-400, 40 mm outside diameter [O.D.], continuous thread) cap was used as a control because it did not have a secondary overcap that would restrict the surface area of grip. The other 6 two-piece (six 38-400, 40 mm O.D., 12.5 mm height, nylon continuous thread inner caps, and six 44 mm O.D., nylon overcaps) caps restricted the amount of inner cap surface area exposed (limiting the functional grip area); in short, they utilized the overcap to shield the area of the inner cap available for participants to grip. The amount of inner cap surface area exposed by the overcap 75 (functional grip area) is based on a 7 mm reference line that extends from the outer circumference of the inner cap towards center of mass of the inner cap. The bounds of overcap surface area exposure was based on a finger compressing a maximum of 7 mm (Tomlinson, Lewis, Carre, & Franklin, 2013). Specifically, a reference line that extends approximately 7 mm from the outer circumference of the inner cap towards center of mass of the inner cap was constructed. A horizontal cut line perpendicular to the reference line represents the bounds of the exposure (see Figure 29). Figure 29: Design surface area exposure. The amount of inner cap surface area exposed by the overcap is based on a 7 mm reference line that extends from the outer circumference of the inner cap towards center of mass of the inner cap. A horizontal cut line perpendicular to the reference line shows the bounds of the exposure. Overcaps were either double-sided, with open lateral faces (2-4) on opposite sides of the cap, or a single (5-7), open, lateral face. Additionally, to provide for more variation in the gripping surface area, four of the overcaps (Packages 3, 4, 6 and 7) also had a “bridge” that horizontally bisected the open face(s) of the overcap, further limiting the available grip area (see Figure 30). Packages 3 and 6 had a 1 mm width bridge and Packages 4 and 7 had a 2 mm bridge; the functional surface area for a given exposure treatment (single or double) was held the same so that the efficacy of the bridge could be examined independently of functional surface area (e.g. 6 vs 7). 76 Figure 30: 3D printed design variations. Design variations consist of a two-piece (inner cap and overcap) packaging system for use with a continuous thread bottle. Variations of the design each allow access to a different amount of surface area in which to grip the inner cap (pictured white). Variations 5 through 7 have a single-sided opening and variations 2 through 4 are double-sided. The bridge width was varied by treatment; larger bridges obstructed more of the midline of the inner cap from being exposed, while smaller bridges exposed more of the inner cap midline. Treatments 2-7 were categorized by exposure level. Double-sided packages (2-4) were coded as double exposure level 0-2, respectively. Exposure level 0 indicated the absence of a bridge, while levels 1 and 2 refer to the bridge width in millimeters. Single-sided packages 5-7 were similarly coded (see Table 21). 77 Table 21: Nomenclature for package surface area exposure levels. Exposure level Characteristic Double exposure 0 No bridge Double exposure 1 1 mm bridge Double exposure 2 2 mm bridge Single exposure 0 No bridge Single exposure 1 1 mm bridge Single exposure 2 2 mm bridge Package 2 3 4 5 6 7 We hypothesized that the larger bridge would interfere with participants ability to exert rotational forces, resulting in lower rotational forces than those that were generated in designs with smaller bridges (which had a larger functional area of the inner cap exposed). All bridge thicknesses were 0.6 mm; potential bridge widths were bounded by the structural limitations of the 3D printing process utilized to create the prototypes (1 mm width). The overcap of Package 5 limited the total amount of surface area of the inner cap (available to grip) to 14.4% by vertically shielding off 100% of the top cap and 74% of the lateral sides of the cap (see Figure 31). By comparison, the one-piece cap used as a control (Pkg 1) had 100% of its surface area available for gripping. 78 Figure 31: Single exposure 0, Package 5 and Control Package 1 examples. (Left) Package 5: Two-piece cap design in which the overcap shields off portions of the inner cap’s surface area from being gripped. (Right) Control Package 1: One-piece cap design in which the cap is unshielded and does not restrict any portion of the surface area from being gripped. The outer cap of Package 6 limits the available surface area of the inner cap to 13.3% by shielding off 100% of the top cap portion, as well as 74% of the lateral side of the cap, and by utilizing an additional 1 mm x 0.6 mm (width x thickness) bridge that bisects the midline height of the inner cap (see Figure 32). This bridge obstructs efficient gripping by the thumb or fingers. Package 7 has a wider bridge (2mm as compared to the 1 mm height in Package 6); it utilizes a 2 mm x 0.6 mm bridge. The scaling of Package 7 is done so that the proportion of inner cap available for gripping is kept the same (13.3% of total surface area) as Package 6, and only the midline bridge width is increased to further obstruct efficient gripping by the thumb or fingers. Comparing Package 6 or 7 to Package 5 will allow for inference on the effect of adding a bridge (on torque) compared with merely shielding the top and lateral portions of the 79 cap. Double exposure Packages (2-4) mirrors single exposure Packages 5-7 by translating the lateral opening and bridges to the opposite lateral side (see Figure 32). It is expected that the increase in the available surface area will for Packages 2 through 4 will be less restrictive than their counterparts (higher torque will be generated comparatively). 80 Figure 32: Design concept variations surface area details. Overcap design variations each allow access to a different amount of surface area in which to grip the inner cap. All caps were 3D printed using a Creality3D CR-10S 3D printer (0.4 mm nozzle, 0.2 mm layer height, print temperature of 280°C) using nylon (tensile strength at yield/break: 7,582/6072 psi, flexural modulus 152,000 psi, max elongation at break 27%) filament. MatterHackers PRO Series White Nylon 1.75 mm filament was used for 81 the overcaps, and MatterHackers PRO Series Blue Nylon 1.75 mm filament was used for the inner caps (6) and the control cap (1). Each inner cap, overcap, and control cap were 3d printed in batches of five (13 batches; 5 one-piece caps, 30 inner caps and 30 overcaps). Each individual cap was measured across cap diameter/height and bridge width/thickness (if applicable) using a Digital Vernier Caliper to the nearest 0.1 mm. Caps were weighed using a Mettler Toledo (30216549 Model ME203TE Precision Balance, 220 g Maximum Load Capacity x 1.0 mg Readability) balance to the nearest 0.1 gram to ensure consistent quality in the prototyping process. Each participant required a single set of caps that consisted of, one-piece cap (Package 1), 6 inner caps, and 6 overcaps (Packages 2-7). Two sets within +/- 0.1 (gram and mm) comparative tolerance were drawn from the batches for testing in this study. The maximum rotational torque (nearest 0.1 lbs-inch, ±1% of full scale or reading, range 0-100 lbs-inch) that people could exert on the inner cap was tested for each treatment (1-7) using a Secure Pak Digital Torque Tester (see Figure 33). Treatment order for Packages 1-7 was randomized for each subject with Microsoft Excel Version 1809 randomization order. Independent variables of interest were gender, age, thumb breadth, and bridge width, and the dependent variables were average peak torque generated by thumb size, by package, and by age. 82 Figure 33: Secure Pak Digital Torque Tester with control package clamped for testing. 5.2.1 Procedure One hundred-fifty people (50 women aged 18+ years, 50 men 18+, and 50 children aged 4 through 12 years) were recruited for this study (MSU Study ID STUDY00001724). The three groups of people in this study were selected as being likely to vary in ability to generate torque and hand size. Power calculations utilized published data from (Rowson & Yoxall, 2011) to suggest a sample size of 9 for an effect size of 0.59 assuming 7 treatments being compared (α =0.05, power = 0.93). Since we do not know how many people of a specific age or thumb size will participate within each group, we estimated 50 people in each group is sufficient to give us 3-5 subgroups for comparison within each group (I.E 4-year-olds can be compared with senior adults 65+ years). Subjects were screened and excluded if they had self-indicated history of physical or mental impairments that impacted their ability to open packages (see Consent forms A and B, Appendix A). Two groups were targeted for inclusion in the study; a group of adults, Subject Group A (50 women aged 18+ years and 50 men 18+) (see Flyer, Appendix A) and a group of elementary aged children, Subject Group B (50 children aged 4 through 12 years), were recruited through the Michigan State University 83 Family Listserv and the SONA System Community Paid Pool online recruitment platforms. Additionally, participants were also provided IRB approved fliers to take to family and friends. Each participant received a cash incentive of $15 for participating. Testing took place at the School of Packaging’s Packaging HUB (Healthcare, Universal Design and Biomechanics) laboratory. Researchers provided all subjects with a verbal explanation detailing all test procedures. Subject Group A was provided with an IRB approved consent (see Consent form A, Appendix A) which they were asked to review and sign. Parents and Guardians were also provided an IRB approved consent form (see attached Consent form B, Appendix A) for child participants comprising Group B. Group B participants were also provided with an IRB approved assent information form (see attached Assent form, Appendix A) if they were between the ages of 8-12 years. A verbal assent script (see attached Assent script, Appendix A) was read to subjects under the age of 8 years to ascertain their willingness to participate (through verbal or nonverbal cues such as the nod of the head). Participants that consented to testing were assigned a subject number and either self-identified or parents/guardians identified the following demographic information: Gender, age, and ethnicity. Data collection sheets (see Data Collection Sheet, Appendix A) from participants were identified by subject number. No reference to a participant’s subject number was made on their consent forms. During the informed consent process, a member of the research team verbally reiterated the screening criteria (see Consent form, Appendix A) to verify eligibility. Participants were informed that testing would take no more than 15 minutes of their time and that they could quit or skip any portion of the testing and still receive the $15 84 incentive. The researcher also provided a brief explanation of what was to be done. Specifically, participants were asked to "twist a series of 7 caps to the best of the best of their ability by exerting the most force that they can without hurting themselves." Subjects were tested one at a time. The subject was seated in a height adjustable chair at a table in the testing room. The height of the chair was adjusted to a level that the participant indicated was comfortable for them while seated at the table. The breadth of their thumb was then measured to the nearest 0.1 millimeters using a FstDgte 510-160 digital caliper. The breadth of the proximal interphalangeal joint of the thumb (the widest digit of the human hand) was measured and used as an upper limit estimation for phalangeal surface area interaction. Subjects were then tasked with applying torque to seven different packages (treatments). A researcher sequentially mounted each treatment (see Treatment randomization order sheet, Appendix A) inside the clamps of a Secure Pak Digital Torque Tester (see Figure 3) to measure peak torque to the nearest 0.1 pound-inch and in an order randomized for each participant (using Microsoft Excel Version 1809 randomization). Participants were instructed to attempt to rotate the “blue cap” (inner cap is blue and overcap is white) in a clockwise direction for a period of 10 seconds. A demonstration was performed by a researcher and then the subject was allowed to practice the motion on a demonstration sample. The following instructions were verbally communicated as the demonstration was conducted and verbally communicated again prior to the first trial: “Please place your thumb horizontally across the side of the blue cap and squeeze with your hand tightly as if you were preparing to turn the cap. 85 The blue cap is glued and will not rotate. Try to twist the blue cap clockwise applying the highest force you can without hurting yourself. Please stay seated and use only one hand on the cap, do not use the other hand to help. Begin trying when I say ‘START’ and you can stop when I say ‘STOP’”. The researcher used a stopwatch timer and verbally indicated when the subject should “START” and “STOP”. Peak torque was recorded for each trial for each participant. Additionally, tested packages were removed from the torque tester which was reset, and the next test package was mounted. This not only prepped the next trial but also allowed for a brief rest period for the participants; the rest period could be extended if the participant elected to do so. Notes: Some packages began to fail from observed fatigue due to repeated use. Packages failed by layer delamination, glue bond failure between cap and container or a combination of both. These packages were replaced from the set of backup packages when they failed and the trial that failed was repeated. 5.3 Demographic Statistics Consenting study participants were assigned a subject number and the following demographic information was collected: Gender, age, and ethnicity. All data collected (see Data Collection Sheet, Appendix A) from participants was identified by their subject number and no reference to their subject number was made on the consent forms. 5.3.1 Participant age, gender, and ethnicity A total of 150 participants were tested in this study. Fifty participants were adult males between the ages of 18-67 years, 50 participants were adult females between the ages of 19-78 years and 50 participants were children (26 males, 4-11 years and 24 females 4-12 years) between the ages of 4-12 years (see Table 22). One hundred and 86 fourteen participants self-identified as Caucasian, two as African American, one as American Indian, seven as Mixed, seventeen as Asian, eight as Hispanic, and one as Indian (see Table 23). Table 22: Number of adult participants by self-identified gender and age, and number of child participants by parental/guardian-identified age and gender. Number of child participants Age (years) Female child Male child Total 4 5 4 9 5 4 5 9 6 4 3 7 7 2 6 8 8 5 1 6 9 - 4 4 10 11 12 Total 1 2 3 2 1 3 1 - 1 24 26 50 Number of adult female participants Age (years) 18-29 30-39 40-49 50-59 Adult Females 11 18 8 3 Number of adult male participants Age (years) 18-29 30-39 40-49 50-59 Adult Males 16 16 6 7 60+ (60-78) 10 60+ (60-67) 5 Total 50 Total 50 Table 23: Number of adult participants by self-identified ethnicity, and number of child participants by parental/guardian-identified ethnicity. Ethnicity Children (4-12 years) Adult Males (18-67 years) Adult Females (18-78 years) White / Caucasian Black / African American American Indian Mixed Asian Hispanic Indian Total 40 - - 3 6 1 - 50 36 2 1 1 7 3 - 50 38 - - 3 4 4 1 50 Total 114 2 1 7 17 8 1 150 87 5.3.2 Thumb breadth measurement The breadth of the proximal interphalangeal joint of the thumb was measured to the nearest 0.1 millimeters (see Figure 34 and Table 24). This measurement was used as a proportional estimation of hand size among individuals. In general, a larger thumb was considered indicative of a larger hand and therefore likely to have greater contact compared to a hand with a smaller thumb breadth. Figure 34: Measurement of participant thumb breath at the proximal interphalangeal joint. 88 Table 24: Participant thumb breadth (proximal interphalangeal joint). Participant thumb breadth (mm) Number of participants 50 50 50 Age (years) Children (4-12) Adult females (18-78) Adult males (18-67) Average Min Max 15.1 12.2 18.4 18.4 14.8 21.7 20.6 17.7 22.7 Child thumb breadth Number of participants Age (years) Thumb breadth (mm) Max 13.8 14.3 14.3 15.6 16.3 16.0 18.4 17.4 17.6 Average Min 12.3 12.7 12.2 13.7 14.4 14.2 15.3 16.3 17.6 13.1 13.5 13.5 14.7 15.0 15.2 16.7 16.7 17.6 4 5 6 7 8 9 10 11 12 9 9 7 8 6 4 3 3 1 Number of participants 11 18 8 3 10 Number of participants 16 16 6 7 5 Adult female thumb breadth Age (years) 18-29 30-39 40-49 50-59 60-78 Thumb breadth (mm) Max 20.3 20.2 20.4 19.2 21.7 Average Min 14.8 16.7 16.8 16.5 16.8 17.2 18.6 18.4 18.0 19.6 Adult male thumb breadth Thumb breadth (mm) Max 22.1 22.0 22.0 22.6 22.7 Average Min 18.1 17.7 20.2 20.1 20.6 19.9 20.3 21.2 21.2 21.7 Age (years) 18-29 30-39 40-49 50-59 60-67 89 5.4 Average peak torque results Statistical analysis was conducted using SAS Version 9.4 (Statistical Analysis Software). Peak torque was not normally distributed and, as such, was log transformed. Average peak torque is reported as back-transformed mean estimates with standard error unless otherwise noted throughout the remainder of this chapter. The mixed procedure was used to make statistical inferences about the average peak torque. The model included fixed effects: Thumb*Package, Package, AgeGroup, Package*AgeGroup. Package*Order, Sex*AgeGroup, Sex were included as random effects, as there was no evidence of significance. Due to an unequal distribution of sex and age group, Tukey's adjustment was used for comparisons involving sex and age group while Bonferroni was used for the multiple comparisons with the packages. 5.4.1 Average peak torque by thumb breadth Participants were instructed to place their thumb horizontally along the circumference of the cap while gripping the opposite side in a manner they found comfortable (see Figure 35). Each treatment package was sequentially mounted in a digital torque tester in a randomized order and the peak torque was recorded. Peak torque (log transformed due to non-normality) was graphed versus thumb breadth (see Figure 36). Across all seven packages the average peak torque was shown to increase as thumb breadth increased (Pearson Correlation were calculated for all treatments r(148), p<.0001: package1=0.74; pkg2=0.79; pkg3=0.65; pkg4=0.74; pkg5=0.76; pkg6=.062; pkg7=0.54). 90 Figure 35: Example of participant thumb contact orientation relative to the inner blue cap of each treatment package. Figure 36: Thumb breadth and average peak torque. Slope relationship of participant thumb breadth (mm) and average peak torque (log-transformed) for each of the seven treatment packages (labeled 1-7). Participants generated greater torque with Package 1 (control), and as thumb breadth increases torque increases across all packages. (Pearson Correlation r(148), p<.0001: package1=.74; pkg2=.79; pkg3=.65; pkg4=.74; pkg5=.76; pkg6=.62; pkg7=.54). 91 5.4.2 Average peak torque by package The average peak torque generated by all participants across each package treatment was compared across packages. The control package was absent any surface area restriction allowing the cap to be gripped freely. Packages 2-7 were comprised of overcaps that increasingly shielded specific areas of their inner caps from being gripped (see Figure 37). Figure 37: Design concepts’ imposed surface area restriction. Overcap design variations each allow access to a different amount of surface area in which to grip the inner cap. As expected, participants were able to generate larger rotational forces as the functional surface area, the exposed area of the inner cap, increased. The control package (Package 1) resulted in the highest reported average for peak torque (reported 92 as back-transformed means with 95% confidence intervals) while the average peak torque progressively declined with each subsequent treatment package as surface area exposure decreased (see Figure 38). No significant difference was found (α=0.05, p<0.0001) when the varied bridge sizes were compared within an exposure level (i.e. between double-sided packages (C 1 mm bridge and D 2 mm bridge) or between single- sided packages F (1 mm bridge) and G (2 mm bridge). Figure 38: Average peak torque generated across all package treatments. Control package A, absent of any surface area restriction, generated the highest average peak torque while each subsequent package demonstrated declining torque as package surface area restriction increased (back-transformed; different letters indicate statistical significance at α=0.05. No significant difference was found when the varied bridge sizes were compared within an exposure level (i.e. between double exposure 1 and 2 (Package 3, 1 mm bridge; Pkg 4, 2 mm bridge; respectively) (α=0.05, p=1.0000) or between single exposure 1 and 2 (Package 6, 1 mm bridge; Pkg 7, 2 mm bridge; respectively) (α=0.05, p=1.0000). 93 5.4.3 Average peak torque by age The average peak torque across all package treatments was compared with participant age as a covariate (see Figure 39). Tukey's adjustment was used for comparisons (different letters indicate statistical significance at α=0.05). Age groups were structured for analysis as follows (see Table 25): Table 25: Age groups, number of participants, and thumb breadth (proximal interphalangeal joint) for analysis. Age Group (years) 4-year-olds 5-year-olds 6-year-olds Children 7 to 12 Adults 18 to 64 Senior adults 65+ Thumb breadth (mm) Average 13.1 13.5 13.5 15.4 19.4 20.1 Number of participants 9 9 7 25 89 11 Min 12.3 12.7 12.2 13.7 14.8 18.1 Max 13.8 14.3 14.3 18.4 22.7 21.7 Because control package 1 demonstrated a strong linear correlation (r(148) =.74 , p<.0001) between increasing thumb size and average peak torque, age groups informed on the data related to thumb size. Specifically, children aged 4, 5, and 6 years had average thumb breadths that were 13.1, 13.5 and 13.5 mm, respectively (see Table 25). Individual age groups for these children were created to if see their average peak torque was significantly different despite similar thumb sizes. An age group for older children between 7 to 12 years old (average thumb size 15.4mm) was also created for comparison with younger children if there was no difference in torque was detected among 4, 5, and 6-year-olds. Age groups for younger adults (18-64 years) and senior adults (65+ years) were created because we also want to know how treatment packages will perform specifically with senior adults (senior adult accessibility) and 4- year-old children (child-resistance). 94 Figure 39: Average peak torque generated across all package treatments by age groups. Four-year-old children generated the least amount of torque when interacting with all seven package treatments while senior adults 65+ years generated less torque compared to younger adults (different letters indicate statistical significance at α=0.05). Out of all the age groups, 4-year-old children generated the least amount of average peak torque (see Figure 34). Five (4.67 lbs-inch+/-1.13 lbs-inch) and 6-year-olds (4.69 lbs-inch +/-1.13 lbs-inch) generated greater average peak torque (p < 0.0001 for both comparisons) than 4-year-olds (3.90 lbs-inch +/-1.13 lbs-inch). Senior adults generated less average peak torque (p < 0.0001) compared to younger adults (10.09 lbs-inch +/- 1.13 lbs-inch VS 10.90 lbs-inch +/-1.13 lbs-inch, respectively). Since senior adults generated the lowest average peak torque of the adult age groups, the average peak torque across each package was compared. Bonferroni was used for the multiple comparisons with the packages. Average peak torque declined across each subsequent package (1-7) as surface area restriction increased (see Figure 40). The average peak torque that the oldest adult group generated when testing 95 with Package 2 did not yield evidence of statistical significance when the average peak torque that they generated with Package 1 (control) was compared (p=0.9581). No evidence of a significant difference was found (α=0.05, p=1.0000) when torque generated from treatments of double exposure – bridge size 1mm or 2mm (Packages 3 and 4) were compared. The same was true when single exposure level with bridges 1mm and 2mm were compared (Packages 6 and 7) with senior adults. Figure 40: Senior adult average peak torque across all package treatments. Average peak torque generated across all package treatments by senior adults (65-78 years) (back- transformed; different letters indicate statistical significance at α=0.05). Average peak torque declined across each subsequent package as surface area restriction increased. The youngest group, four-year-olds, not surprisingly, generated the least amount of average peak torque across all age groups (see Figure 41). Bonferroni was used for the multiple comparisons with the packages. Unlike the results for senior adults (see Figure 40), average peak torque did not decline across each subsequent package (1-7) as the amount of exposed surface area available for rotation decreased (see Figure 41). 96 When the average peak torques generated during the control trials (Package 1) were compared to those generated during trials with the experimental packages (2-7), the amount the torque was reduced by experimental treatments ranged between 45% ((7.4818-4.1275) / 7.4818 *100) with Package 6 to 65% ((7.4818-2.649) / 7.4818 *100) with Package 5 for 4-year-olds. No significant difference in peak torque readings were evidenced (α=0.05, p=1.0000) when treatments with double exposure (bridges in 1mm and 2mm- Packages 3 and 4 were compared. The same held true for the torque generated from packages with single exposures – bridges 1mm and 2mm (Packages 6 and 7) with 4-year-olds. Figure 41: Average peak torque for children across all package treatments. Average peak torque generated across all package treatments by 4-year-olds (back-transformed; different letters indicate statistical significance at α=0.05). Average peak torque did not decline as surface area restriction increased across Packages 2-7. Compared to Control Package 1, the range of average peak torque generated was reduced between 45% to 65% across prototype Packages 2 through 7. Average peak torques for senior adults and 4-year-olds were compared across prototype packages 2-7 (see Figure 42). The percentage of average peak torque 97 reduction between senior adult torque and child torque ranged from 70% ((13.5047- 4.0177) / 13.5047 *100) with Package 1 to 43% ((6.2449-3.5395) / 6.2449 *100) with Package 7. In other words, Package 2 demonstrated the largest gap in average peak torque (70% reduction) between senior adults and 4-year-olds while Package 7 demonstrated the smallest gap at 43%. The amount of average peak torque 4-year-olds generated with Package 2 (4.0177 lbs-inch +/- 1.17 lbs-inch) was significantly lower (p<0.0001) than senior adults average peak torque with Package 2 (13.5047 lbs-inch +/- 1.17 lbs-inch). Package 2’s results suggest it has the greatest ratio of adult accessibility to child resistance and was selected for further analysis (see Figure 42). Figure 42: Average peak torque for children and senior adults by prototype package. Average peak torque generated across all prototype package treatments by senior adults (65-78) and children (4 years) (back-transformed with standard error bars). Package B demonstrated the largest reduction in average peak torque between senior adults and 4-years-olds. To further characterize adult accessibility and child resistance, it is critical to identify the range of average peak torque that both adults and young children can 98 similarly generate. Using Control Package 1 as a baseline, the average peak torque of 4-year-olds ranged from approximately 4 to 12 lbs-inch and with adults (18-78 years) it ranged from approximately 8 to 45 lbs-inch (see Figure 43). The lower adult limit of 8 lbs-inch and the upper 4-year-olds limit of 12 lbs were selected as the range for analysis. Nine adults (22-73) who generated average peak torque in the range of 8-12 lbs-inch were compared to eleven children (4 and 5-year-olds) who similarly generated torque in that same 8-12 lbs-inch range. Five-year-olds that generated between 8-12 lbs-inch average peak torque were also included with 4-year-olds as a more stringent test group. Senior adult (65-78) vs child (4 years) min/max & average peak torque by package ) h c n i - s b l ( e u q r o t k a e p e g a r e v A 30 25 20 15 10 5 0 11.6 8.9 Senior Adults 4-year-olds 1 Package 1 Figure 43: Control package max/min and average peak torque for adults and children. Average and max/min peak torque generated with Control Package 1 (control) by senior adults and 4- year-old children. An approximate range of 8-12 lbs-inch of torque was demonstrated between senior adults and 4-year-old-children using the average minimum peak torque adults (8.9 lbs-inch) generated and the average maximum peak torque children (11.6 lbs-inch) generated. There was no evidence of a significant difference (p=1.0000) among the average peak torques generated by children (4 and 5 years) (9.4097 lbs-inch+/-1.10 lbs-inch) 99 using Package 1, adults (22-73 years) using Package 1 (11.0355 lbs-inch+/- 1.05 lbs- inch), or adults (22-73 years) using Package 2 (10.7228 lbs-inch +/- 1.05 lbs-inch) (see Figure 44). However, the average peak torque for children (4 and 5 years) was significantly (p<0.0001) lower (4.8908 lbs-inch +/- 1.10 lbs-inch) with Package 2 when compared to Control Package 1 (children), Control Package 1 (adults), and Prototype Package 2 (adults). Figure 44: Average peak torque between control and prototype 1 with children and adults. Average peak torque generated between package treatments 1 (control) and 2 (prototype) (different letters indicate statistical significance at α=0.05). Nine adults (22-73 years) who generated average peak torque in the range of 8-12 lbs-inch were compared to eleven children (4 and 5-year-olds) who similarly generated torque in that same 8-12 lbs-inch range. Average peak torque for children was significantly (p<0.0001) lower with Prototype Package 2 when compared to Control Package 1 (children), Control Package 1 (adults), and Prototype Package 2 (adults). 5.5 Discussion Thumb breadth at the proximal interphalangeal joint was measured and used as a proportional estimation of hand size among individuals. In general, a larger subject thumb breadth was considered to be indicative of a larger hand and therefore likely to 100 have greater hand to object contact area compared to subjects with smaller thumb breadths. Pheasant and O’Neill’s (1975) cylindrical torque model suggests the resultant torque (T) exerted on a cylinder depends on grip strength (G), the coefficient of friction between the hand and the object (µ), and the object diameter (D) where (T = G x µ x D). Applying this model, we would expect as hand size increases the resultant torque would also increase: First, as thumb breadth (or contact area) increases the coefficient of friction between the hand and the object (µ) would also increase because greater contact area between the skin and an object is associated with a higher friction coefficient (Bullinger et al., 1979; Comaish & Bottoms, 1971; Seo, 2008). Second, grip strength (G) tends to increase as hand size increases because people with larger hands tend to exhibit greater muscular strength (Aghazadeh, Lee, & Waikar, 1993; Crawford et al., 2002). Our results are supported by this previous work; all subjects in our study the average peak torque increased as thumb breadth (hand size) increased (see Figure 36). However, a general association among hand size, grip strength and contact area does not explain the variation in torque generated across age groups. Seo and Armstrong (Seo & Armstrong, 2008) found that the ratio of cylinder diameter to hand length with a cylindrical grip explained 62% of the variance in grip force, 57% of the variance in normal force, and 71% of the variance in contact area. When comparing young children (<5 years) with adults (18+ years) in our study, we speculate that differences in fine motor skills, cognitive development (Davis, Pitchford, & Limback, 2011), and haptic perception and exploration (Kalagher & Jones, 2011; Lederman & Klatzky, 2009) may also play a role in how much hand torque is ultimately generated. 101 For example, senior adult (65+ years) average peak torque consistently declined across each treatment package 1-7 as surface area exposure available for gripping declined. However, when 4-year-olds interacted with the same packages no apparent pattern was observed across prototype Packages B though G (see Figure 40). This data suggests Figure 45: Range of average peak torque for all packages with senior adults and children. Average peak torque (with min and max range) generated across all package treatments by senior adults (65-78) and children (4 years). Treatment packages reduced surface area exposure available for gripping from Control Package 1 (100%) to Package 7 (13.3%). Senior adult torque consistently declined from Package 1 to Package 7 as surface area exposure declined. Child torque declined from Package 1 to Package 2 but no clear pattern of torque decline was observed relative to surface area exposure among Packages 2-7. that, compared to senior adults, the 4-year-olds in this study may not have been able to effectively maximize their grip using only their thumbs and fingers to apply torque to the prototypes’ specific exposed surface area regions. Of the prototype packages, Prototype Package B demonstrated the highest ratio (3.35) of senior adult average peak torque to child average peak torque (see Figure 45). This method of limiting hand to 102 object contact area may be a suitable method for executing child resistance without significantly impacting senior adult accessibility. Lastly, the thumb breadth range of 4-year-olds (12.3-13.8 mm) did not overlap with adult thumb breadth range (14.8-22.7 mm), meaning 4-year-olds likely had smaller hands and less object contact area than adults. Despite having smaller hands, the amount of peak torque 4-year-olds generated with Control Package A coincided with an 8-12 lbs-inch range that some adults (22-73 years) also generated. Within this 8-12 lbs- inch range the average peak torque for adults (22-73 years; 11.0355 lbs-inch+/- 1.05 lbs-inch) and young children (aged 4-5 years; 9.4097 lbs-inch+/-1.10 lbs-inch) did not show evidence of statistical differences (p<0.4669) when interacting with Control Package 1. In other words, the weakest adults and strongest 4 and 5-year old children we tested demonstrated similar peak torque generation capabilities with Package 1 (see Figure 44). When these children and adults’ average peak torque for Package 1 was compared with Prototype Package 2 (see Figure 44) two main observations were evident. First, the amount of average peak torque adults generated with Control Package 1 (11.0355 lbs-inch+/- 1.05 lbs-inch) and Prototype Package 2 (10.7228 lbs- inch +/- 1.05 lbs-inch) were not statistically significantly different (p<0.0001) and varied by less than 3 percent. Package 1 and 2 performing similarly with adults suggests they may be approximately equal in terms of adult (22-73 years) accessibility (see Figure 44). By contrast, when the average peak torque that children could generate on treatment 1 versus 2 were compared the amount of average peak torque children generated with Control Package 1 (9.4097 lbs-inch+/-1.10 lbs-inch) and Prototype Package 2 (4.8908 lbs-inch +/- 1.10 lbs-inch) were statistically significantly different 103 (p<0.0001). The average peak torque generated with Package 2 was 48 percent less ((9.4097-4.8908) / 9.4097 x 100) than Package 1. In sum, our data suggests Prototype Package 2’s design may be suitable as a child resistant mechanism that is also capable of similar levels of senior adult accessibility when compared to a continuous thread screw cap (Control Package 1). 5.6 Conclusions • Average peak torque increased across all packages as thumb size increased. • When comparing the experimental designs to the screw cap Control Package 1, limiting the exposed surface area of the top and lateral sides of the prototypes’ inner cap reduced the average peak torque across all subjects. • Among the Double and Single exposure levels 1 and 2 (limiting the exposed surface area of the inner cap using of overcap bridges of varying width), a statistically significant difference in average peak torque was not found. • Prototype Package 2 did not demonstrate a statistically significant difference in senior adult average peak torque when compared to a continuous thread screw cap (Control Package 1). However, 4 and 5-year-old children demonstrated 48% less average peak torque generation when comparing Control Package 1 with Prototype Package 2. Prototype Package 2’s design may be suitable as a child resistant mechanism that is also capable of similar levels of senior adult accessibility when compared to a continuous thread screw cap (Control Package 1). 104 5.7 Limitations The findings in this study are limited to a single closure diameter of 40 millimeters. Other studies have shown that, for cylindrical closures approximately 20-80 mm in diameter, the amount of torque individuals can generate increases as the diameter of the closure increases (Crawford et al., 2002; Ray & Biswas, 2012; Rohles, Laviana, & Moldrup, 1984; Yoxall & Janson, 2008; Yoxall et al., 2010). It is unknown if the effectiveness of our surface area limiting method for cylindrical closures has a linear relationship below and above closure diameters of 40 mm with young children and adults. Further research is needed. Additionally, prototype packages were 3D-printed and there may be performance differences compared to production quality products due to differences in texture, physical tolerances and robustness of the prototypes. 105 Table 26: Design evaluation summary of goals. Research Goal(s) Summary Design Approach • Evaluate design based on • No evidence of a identified performance metrics (torque), user interactions patterns (grip type), pilot study data analysis: o Children may be limited to primarily using their thumbs and fingers to grip and therefore may be less effective at generating torque o Compared to young children (< 5 years), adults may be able to generate torque more efficiently with their thumbs and fingers due to their larger hand size • Report significant findings • Describe implications of results • Summarize findings statistically significant difference was found in the average peak torque generated by senior adults using a continuous thread screw cap medication bottle and Prototype Package 2 • There was evidence of a statistically significant difference between the average peak torque generated by children (aged 4 and 5) using a continuous thread screw cap medication bottle and Prototype Package 2 • Prototype Package 2’s design may be suitable as a child resistant mechanism that is also capable of senior adult accessibility performance similar to a continuous thread screw cap 106 APPENDICES 107 Appendix A – IRB APPROVED DOCUMENTS Recruitment Flyer Opportunity to participate in a research project on packaging design An Analysis of the Impact of Hand Anthropometry and Surface Area Restriction on Removal Torque Michigan State University’s School of Packaging is conducting a study to investigate how the application of torque varies based on (1) hand size and (2) surface area available for gripping using seven different cap designs. You are invited to participate in this study. During this study, we will test the amount of torque each participant is able to apply to each of seven different packages. Research will take place at Michigan State University’s School of Packaging in room 159 and will take approximately 10 minutes (but no more than 15 minutes). If a child is participating parents must give permission for their child to participate. Children will also be asked if they want to participate. To participate in this study, you must: • Be an adult 18 years or older OR be a child between the ages of 4 to 11 years old • Have no physical or mental impairments that impact your ability to open packages • Verbally agree to participate after a brief explanation of the study In exchange for participation you will receive $15. If at any time the participant is uncomfortable with the testing or wishes to discontinue the process, they may discontinue participation without penalty (i.e. they will still receive the $15). If you have questions or comments regarding this study, please contact Laura Bix at bixlaura@msu.edu or phone 517-355-4556. If you have questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this study, you may contact, anonymously if you wish, the Michigan State University's Human Research Protection Program at 517-355-2180 or e-mail irb@msu.edu or regular mail at 4000 Collins Road, Suite 136, Lansing, MI 48910. To schedule participation in this torque study, please contact Cory Wilson at 517-410-2766. 108 Consent Form A PARTICIPANT INFORMATION STATEMENT AND CONSENT FORM An Analysis of the Impact of Hand Anthropometry and Surface Area Restriction on Removal Torque PURPOSE OF RESEARCH You are invited to participate in a research study of packaging designs. We are investigating if the amount of surface area an individual can grip, and the size of their hand affects how much torque (or rotational force) they can apply to a cap and container package. If the idea is successful it can contribute to designing more effective packaging. It has the potential to inform design from an ease of use perspective, but also can be utilized for special cases (e.g. child resistance). Michigan State University’s School of Packaging is conducting a study to investigate how the application of torque varies based on hand size and surface area available for gripping using seven different cap designs. You are being asked to participate in this study. During this study, we will test the amount of torque each participant is able to apply to each of the seven different packages that have the caps glued onto them; caps cannot be removed. To evaluate torque generation, we will conduct testing which will take approximately 10 minutes (but no more than 15 minutes maximum): • We will measure the width of your thumb by using a digital caliper that gently touches two points on the side of your thumb to measure the distance between those points. • A researcher will give you verbal instructions and a brief demonstration of the task to complete. We will then ask you to twist a series of seven caps applying the highest force you can (without hurting yourself) for a period of 10 seconds. These caps are glued and will not twist off. • You will be given a rest period between each 10-second cap trial. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 109 To participate, you must: • Be an adult aged 65+ years OR an adult male 18+ years old at the time of testing • Have no physical or mental impairments that impact your ability to open packages • Have this form (signed by you) WHAT YOU WILL DO Research will take place in the Packaging Building Room 159 and will take approximately 10 minutes (but no more than 15 minutes). As part of this research a participant number will be assigned to you. We will record your gender, ethnicity, age, and thumb size. Data will only be tracked by participant number and will not be tied to your name. You will be seated in a chair with a table in front of you; we will adjust the table height to where you feel the task will be most comfortable for you. A brief verbal explanation of what we will be doing will be provided. Each package will be subsequently mounted in a Secure Pak Digital Torque Testing Machine that will sit on the table in front of you. You will be instructed to use one hand to grip the cap and twist it. The Torque Testing machine will display the torque value generated to the researcher and will then be recorded. A rest period will be provided as the researcher mounts the next package into the torque tester. When you are ready another randomly assigned package will be tested. This process will repeat until seven total packages are tested. POTENTIAL BENEFITS You will not benefit personally from being in this study. However, we hope that, in the future, other people might benefit from this study because it will contribute to designing more effective packaging. YOUR RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 110 You have the right to say no to participate in the research. You can stop at any time after it has already started. There will be no consequences if you stop and you will not be criticized. You will not lose any benefits that you are scheduled to receive. PRIVACY AND CONFIDENTIALITY Data will only be tracked by subject number and will not be tied to your name. Collected data (stored by subject number) will be held for a minimum of three years on password protected computers in the packaging HUB lab (currently room 159). Only the MSU appointed researchers and the Institutional Review Board will have access to the research records at MSU. COSTS AND COMPENSATION FOR BEING IN THE STUDY In exchange for your participation you will receive $15. If at any time you are uncomfortable with the testing or wish to discontinue the process, you may discontinue participation without penalty (i.e. you will still receive $15). POTENTIAL RISKS There is little immediate risk to you in participating in this research. You will be tasked with an ordinary activity normally encountered in daily life that consists of applying a rotational twisting force to a bottle container and cap package. Rest periods will be provided between each cap trial. The probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological activities. If you are injured as a result of this study to a point that requires medical attention, and you have insurance for medical care, your insurance carrier will be billed in the ordinary manner. As with any medical insurance, any costs that are not covered or in excess of what are paid by your insurance, including deductibles, will be your responsibility. The University’s policy is not to provide financial compensation for lost wages, disability, pain or discomfort unless required by law to do so. This does not mean that you are giving up any legal rights you may have. CONFLICT OF INTEREST The student investigating the design presented herein (Cory Wilson), is the sole inventor and owner of the patent (pending) for the paradigm being tested. CONTACT INFORMATION If you have concerns or questions about this study, such as scientific issues, or to report an injury, please contact Dr. Laura Bix, Associate Professor of Packaging at 448 Wilson Road # 114 Michigan State University 48823 at 517-355-4556 or bixlaura@msu.edu. If you have questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 111 this study, you may contact, anonymously if you wish, the Michigan State University's Human Research Protection Program at 517-355-2180 or e-mail irb@msu.edu or regular mail at 4000 Collins Road, Suite 136, Lansing, MI 48910. DOCUMENTATION OF INFORMED CONSENT Your signature below means that you voluntarily agree to participate in this research study. Participant name: ________________________________________ (Please Print) Participant signature: __________________________________ ______________________________ You will be given a copy of this form to keep. Date: 112 Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. Consent Form B PARTICIPANT INFORMATION STATEMENT AND CONSENT FORM An Analysis of the Impact of Hand Anthropometry and Surface Area Restriction on Removal Torque PURPOSE OF RESEARCH Your child is invited to participate in a research study of packaging designs. We are investigating how the amount of surface area an individual can grip, and the size of their hand affects how much torque (or rotational force) they can apply to a cap and container package. If the idea is successful it can contribute to designing more effective packaging. It has the potential to inform design from an ease of use perspective but also can be utilized for special cases (e.g. child resistance) to make packaging safer. Michigan State University’s School of Packaging is conducting this research to investigate how the application of torque varies based on hand size and surface area available for gripping using seven different cap designs. Your child is being asked to participate in this study. During this study, we will test the amount of torque each participant is able to apply to each of the seven different packages. To evaluate torque generation, we will conduct testing which will take approximately 10 minutes (but no more than 15 minutes maximum): • We will measure the width of your child’s thumb by using a digital caliper that gently touches two points on the side of their thumb to measure the distance between those points. • A researcher will give your child verbal instructions and a brief demonstration of the task to complete (i.e. twisting each of the seven caps). Specifically, we will instruct them to twist the cap by applying the highest force they can (without hurting themselves) for 10 seconds each. These caps are glued and will not twist off. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 113 • Your child will be asked to repeat this process for seven different caps and will be given a rest period between each 10-second cap trial. To participate, your child must: • Be 4 to 7 years old at the time of testing • Have no physical or mental impairments that impact their ability to open packages • Have this form (signed by ALL guardians) • Verbally agree to participate after a brief explanation of what we want your child to do WHAT YOUR CHILD WILL DO Research will take place in the Packaging Building Room 159 and will take approximately 10 minutes (but no more than 15 minutes). As part of this research a participant number will be assigned to your child. We will record the participant’s gender, ethnicity, age, and thumb size. Data will only be tracked by participant number and will not be tied to their name. Each participant will be seated in a chair with a table in front of them. We will adjust the table height to a position that they deem comfortable. A brief verbal explanation of what we will be doing will be provided. Each of seven packages will be subsequently mounted in a Secure Pak Digital Torque Testing Machine that will sit on the table in front of the participant. Participants will be told that caps cannot be removed, and the will be instructed to use only one hand to grip the cap and twist it. The Torque Testing machine will display the torque value generated; this value will be recorded. A rest period will be provided as the researcher mounts the next package into the torque tester. When the participant is ready another randomly assigned package will be tested. This process will repeat until seven total packages are tested. POTENTIAL BENEFITS Your child might not benefit personally from being in this study. However, we hope that, in the future, other people might benefit from this study because it will contribute to designing more effective packaging. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 114 YOUR CHILD’S RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW Your child has the right to say no to participate in the research. They can stop at any time after it has already started. There will be no consequences if they stop and they will not be criticized. They will not lose any benefits that they normally receive. PRIVACY AND CONFIDENTIALITY Data will only be tracked by subject number and will not be tied to your child’s name. Collected data (stored by subject number) will be held for a minimum of three years on password protected computers in the packaging HUB lab (currently room 159). Only the MSU appointed researchers and the Institutional Review Board will have access to the research records at MSU. COSTS AND COMPENSATION FOR BEING IN THE STUDY In exchange for your child’s participation your child will receive $15. If at any time your child is uncomfortable with the testing or wish to discontinue the process, they may discontinue participation without penalty (i.e. they will still receive $15). POTENTIAL RISKS There is little immediate risk to your child in participating in this research. Your child will be tasked with an ordinary activity normally encountered in daily life that consists of applying a rotational twisting force to a bottle container and cap package. Rest periods will be provided between each cap trial. The probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological activities. If your child is injured as a result of this study to a point that requires medical attention, and you have insurance for medical care, your insurance carrier will be billed in the ordinary manner. As with any medical insurance, any costs that are not covered or in excess of what are paid by your insurance, including deductibles, will be your responsibility. The University’s policy is not to provide financial compensation for lost wages, disability, pain or discomfort unless required by law to do so. This does not mean that you or your child are giving up any legal rights you or your child may have. CONFLICT OF INTEREST The student investigating the design presented herein (Cory Wilson), is the sole inventor and owner of the patent (pending) for the paradigm being tested. CONTACT INFORMATION If you or your child has concerns or questions about this study, such as scientific issues, or to report an injury, please contact Dr. Laura Bix, Associate Professor of Packaging at 448 Wilson Road # 114 Michigan State University 48823 at 517-355-4556 or bixlaura@msu.edu. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 115 If you or your child has questions or concerns about your child’s role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this study, you may contact, anonymously if you wish, the Michigan State University's Human Research Protection Program at 517-355-2180 or email irb@msu.edu or regular mail at 4000 Collins Road, Suite 136, Lansing, MI 48910. DOCUMENTATION OF INFORMED CONSENT Your signature below means that you voluntarily agree to participate in this research study. Please have ALL guardians read, sign and date this form and bring it to the research study. Your child’s name: ________________________________________ (Please Print) Parent or Legal Guardian’s name: ________________________________________ (Please Print) Parent or Legal Guardian’s signature: Date: _____________________________________ _____________________________ Parent or Legal Guardian’s name: ________________________________________ (Please Print) Parent or Legal Guardian’s signature: Date: ___________________________________ ______________________________ You will be given a copy of this form to keep. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 116 Assent Form (8-11 years) PARTICIPANT INFORMATION STATEMENT A Package Opening Study You are invited to participate in a research study of packaging designs. We are studying to see if how big your hand is affects how well you can twist a package. If this idea is successful, it will help make better packaging. During this study, we will test how well you can twist seven different packages (like in the picture above). Testing will take about 10 minutes: • You will sit down in a chair at a table and we will measure the size of your thumb using a plastic tool that will gently touch the sides of your thumb • There will be seven packages placed in front of you • We will tell you to twist the caps using the most force that you can without hurting yourself, keeping in mind that the caps cannot come off. We will tell you to use one hand and show you a specific way to grab the package and twist the cap package • You will be asked to twist seven different packages • You will be able to rest your hand until you are ready to try again • After you twist seven packages you are done To participate, you must: • Be between the ages of 8 to 11 years old • Agree to participate after we tell you what to do You can only participate if both you and your parents agree for you to be in the study. Nobody will be upset if you do not want to be in the study. It is your decision. You will be given $15 for participating in this study. If you decide to be in the study, and later Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 117 change your mind that is okay too. You can stop being in the study anytime you like. You will still be given $15 if you decide you are uncomfortable and want to stop. We will write down your gender, ethnicity, age, and thumb size. Your name will not be in any report of the results of this study. Only the MSU researchers and the Institutional Review Board will have access to the research records at MSU. If you have any questions about the study, you can either tell your parents and have them talk to me, or you can talk to me. Here is the phone number and address of a person who can help if you have questions: Laura Bix, 448 Wilson Road # 114 Michigan State University 48823 at 517-355-4556. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 118 Assent Script Hi, I am a student from Michigan State University. My name is ________ and I am trying to learn about how children open packages. I would like you to help me with this research. If you do, I will ask you to twist the caps of seven different packages like you were trying to take them off, but you should know that they won’t come off. You do not have to do it, if you do not want to. Nobody will be upset with you if you decide you would rather do something else. Does this sound like it would be something you would like to do? Thank you. Approved by a Michigan State University Institutional Review Board effective 12/11/2019. This version supersedes all previous versions. MSU Study ID STUDY00001724. 119 Data Collection Sheet Data Collection Form Subject # _______ Demographic Information 1. Gender ______________________________ 2. Age (years) _______________________________ 3. Ethnicity _________________________________ 120 FOR RESEARCHER USE ONLY Peak Torque (inch • pounds) Treatment A B C D E F G Researcher Notes: 121 Appendix B – FLAT FILE Sight, Sound, and Child Resistance Table 27: S/S/CR Study Flat File opened packages. Children that opened child resistant packages. 122 Table 28: S/S/CR Study Flat File unopened packages. Children that did not open child resistant packages. 123 BIBLIOGRAPHY 124 BIBLIOGRAPHY AAP. (1950). American Academy of Pediatrics. Unpublished data. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/00000698.htm. Aghazadeh, F., Lee, K., & Waikar, A. (1993). Impact of anthropometric and personal variables on grip strength. J Hum Ergol (Tokyo), 22(2), 75-81. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/7963483 ASTM. (2018). ASTM D3475-18 Standard Classification of Child-Resistant Packages. In: ASTM International, West Conshohocken, PA. Bain, K. (1954). Death Due to Accidental Poisoning in Young Children. Journal of Pediatrics, 44(6), 616-623. doi:Doi 10.1016/S0022-3476(54)80002-3 Barry, F. (2009). A study of ease of use issues and potential design solutions for designers and consumers. W. Bradley Fain, Ph. D. Keith A. Kline, MS May 2009. Beckman, A., Bernsten, C., Parker, M., Thorslund, M., & Fastbom, J. (2005). The difficulty of opening medicine containers in old age: A population-based study. Pharmacy World & Science, 27(5), 393-398. doi:10.1007/s11096-005-7903-z Blok, D., Ruiter, I., & Wever, R. (2016). Operability of child-resistant caps: review of current solutions and alternative ideas. Bonfim, G. H. C., Medola, F. O., & Paschoarelli, L. C. (2016). Correlation among cap design, gripping technique and age in the opening of squeeze-and-turn packages: A biomechanical study. International Journal of Industrial Ergonomics, 54, 178-183. doi:10.1016/j.ergon.2016.06.004 Breault, H. J. (1974). Five years with 5 million child-resistant containers. Clin Toxicol, 7(1), 91-95. doi:10.3109/15563657408987980 Bureau, C. National KIDS COUNT. Population Division, U.S. Census Bureau. Retrieved from https://datacenter.kidscount.org/data/tables/101-child-population-by-age- group 125 CDC. (2015). Home and Recreational Safety - Poisoning. Retrieved from https://www.cdc.gov/homeandrecreationalsafety/poisoning/index.html Conn, V. S., Ruppar, T. M., Chan, K. C., Dunbar-Jacob, J., Pepper, G. A., & De Geest, S. (2015). Packaging interventions to increase medication adherence: systematic review and meta-analysis. Curr Med Res Opin, 31(1), 145-160. doi:10.1185/03007995.2014.978939 CPSA. (1972). Consumer Product Safety Act, Public Law 92-573; 86 Stat. 1207, October 27, 1972, Sec. 30; 15 U.S.C. CPSC. (2002). Age determination guidelines: relating children's ages to toy characteristics and play behavior: U.S. Consumer Product Safety Commission (CPSC). CPSC. (2004). Consumer Product Safety Review. https://www.cpsc.gov/s3fs- public/pdfs/blk_media_cpsr_nws33.pdf CPSC. (2005). Poison Prevention Packaging: A Guide for Healthcare Professionals. U.S. Consumer Product Safety Commission Retrieved from https://www.cpsc.gov/s3fs-public/384.pdf CPSC. (2016). Pediatric Poisoning Fatalities from 1972 through 2013. https://www.cpsc.gov/s3fs-public/PPPAMortality2013.pdf CPSC. (2018a). Consumer Product Safety Commission. National Electronic Injury Surveillance System 1998-2018 on NEISS Online Database, released April, 2018. Generated at https://www.cpsc.gov/cgibin/NEISSQuery/home.aspx. on: July 2, 2018 at 21:52:25. In. CPSC. (2018b). National Electronic Injury Surveillance System (NEISS). NEISS Probability Samples Retrieved from https://www.cpsc.gov/Research-- Statistics/NEISS-Injury-Data Crawford, J. O., Wanibe, E., & Nayak, L. (2002). The interaction between lid diameter, height and shape on wrist torque exertion in younger and older adults. Ergonomics, 45(13), 922-933. doi:10.1080/00140130210162243 126 Davis, E. E., Pitchford, N. J., & Limback, E. (2011). The interrelation between cognitive and motor development in typically developing children aged 4-11 years is underpinned by visual processing and fine manual control. British Journal of Psychology, 102, 569-584. doi:10.1111/j.2044-8295.2011.02018.x DTI. (2002). Consumer and Competition Policy Directorate: Specific Anthropometric and strength data for people with dexterity disability. Department of Trade and Industry, London., (URN 02/743) Garrett, J. (1970). ANTHROPOMETRY OF THE AIR FORCE FEMALE HAND. 87. Giampaoli, S., Ferrucci, L., Cecchi, F., Lo Noce, C., Poce, A., Dima, F., . . . Menotti, A. (1999). Hand-grip strength predicts incident disability in non-disabled older men. Age and Ageing, 28(3), 283-288. doi:DOI 10.1093/ageing/28.3.283 Gummin, D. D., Mowry, J. B., Spyker, D. A., Brooks, D. E., Fraser, M. O., & Banner, W. (2017). 2016 Annual Report of the American Association of Poison Control Centers' National Poison Data System (NPDS): 34th Annual Report. Clin Toxicol (Phila), 55(10), 1072-1252. doi:10.1080/15563650.2017.1388087 Hume, M. C., Gellman, H., McKellop, H., & Brumfield, R. H., Jr. (1990). Functional range of motion of the joints of the hand. Journal of Hand Surgery, 15(2), 240- 243. doi:10.1016/0363-5023(90)90102-W Ikeda, A., Kurita, Y., Ueda, J., Matsumoto, Y., & Ogasawara, T. (2004, 28 Sept.-2 Oct. 2004). Grip force control for an elastic finger using vision-based incipient slip feedback. Paper presented at the 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566). Imrhan, S. N. (1994). Muscular Strength in the Elderly - Implications for Ergonomic Design. International Journal of Industrial Ergonomics, 13(2), 125-138. doi:Doi 10.1016/0169-8141(94)90079-5 ISO. (2012). Packaging -- Child resistant packaging -- Mechanical test methods for reclosable child resistant packaging systems. International Standards Organization(BS EN ISO 13127:2012), 28. 127 ISO. (2015). Child-resistant Packaging - Requirements and Testing Procedures for Reclosable Packages. International Standards Organization(IS EN ISO 8317:2015), 31. Jacobson, B. J., Rock, A. R., Cohn, M. S., & Litovitz, T. (1989). Accidental Ingestions of Oral Prescription Drugs - a Multicenter Survey. American Journal of Public Health, 79(7), 853-856. doi:Doi 10.2105/Ajph.79.7.853 Kalagher, H., & Jones, S. S. (2011). Developmental change in young children's use of haptic information in a visual task: The role of hand movements. Journal of Experimental Child Psychology, 108(2), 293-307. doi:10.1016/j.jecp.2010.09.004 Kozin, S. H., Porter, S., Clark, P., & Thoder, J. J. (1999). The contribution of the intrinsic muscles to grip and pinch strength. Journal of Hand Surgery-American Volume, 24a(1), 64-72. Retrieved from ://WOS:000078426900010 Kroemer, K. H. E. (1986). Coupling the Hand with the Handle - an Improved Notation of Touch, Grip, and Grasp. Human Factors, 28(3), 337-339. Retrieved from ://WOS:A1986D868700009 Lederman, S. J., & Klatzky, R. L. (2009). Haptic perception: A tutorial. Attention Perception & Psychophysics, 71(7), 1439-1459. doi:10.3758/App.71.7.1439 Lin, M. C., Chen, M. S., Lin, Y. H., & Hung, Y. C. (2015). A User-Centric Evaluation Procedure for the Design of Child-Resistant Medicine Bottles. Human Factors and Ergonomics in Manufacturing & Service Industries, 25(2), 211-227. doi:10.1002/hfm.20595 Lovegrove, M. C., Mathew, J., Hampp, C., Governale, L., Wysowski, D. K., & Budnitz, D. S. (2014). Emergency Hospitalizations for Unsupervised Prescription Medication Ingestions by Young Children. Pediatrics, 134(4), E1009-E1016. doi:10.1542/peds.2014-0840 Maas, H., Veeger, H. E. J., & Stegeman, D. F. (2018). Understanding the constraints of finger motor control. Journal of Electromyography and Kinesiology, 38, 182-186. doi:10.1016/j.jelekin.2017.10.004 Maceo, A. V. (2009). Qualitative assessment of skin deformation: a pilot study. Journal of Forensic Identification, 59(4), 390-440. 128 Mathiowetz, V., Kashman, N., Volland, G., Weber, K., Dowe, M., & Rogers, S. (1985). Grip and Pinch Strength - Normative Data for Adults. Archives of Physical Medicine and Rehabilitation, 66(2), 69-74. Retrieved from ://WOS:A1985ABW1100001 Mowry, J. B., Spyker, D. A., Brooks, D. E., Zimmerman, A., & Schauben, J. L. (2000- 2016). Annual Reports of the American Association of Poison Control Centers' National Poison Data System (NPDS). 2000-2016. Clin Toxicol (Phila). doi:10.1080/15563650.2016.1245421 Mowry, J. B., Spyker, D. A., Brooks, D. E., Zimmerman, A., & Schauben, J. L. (2016). 2015 Annual Report of the American Association of Poison Control Centers' National Poison Data System (NPDS): 33rd Annual Report. Clin Toxicol (Phila), 54(10), 924-1109. doi:10.1080/15563650.2016.1245421 NEISS. (2017). National Electronic Injury Surveillance System - NEISS Coding Manual. U.S. Consumer Product Safety Commission. ODPHP. (2018). National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance Project. Office of Disease Prevention and Health Promotion. Retrieved from https://www.healthypeople.gov/2020/data- source/national-electronic-injury-surveillance-system-cooperative-adverse-drug- event Pawluk, D. T. V., & Howe, R. D. (1999). Dynamic Contact of the Human Fingerpad Against a Flat Surface. Journal of Biomechanical Engineering, 121(6), 605-611. doi:10.1115/1.2800860 Peebles, L., & Norris, B. (2003). Filling 'gaps' in strength data for design. Appl Ergon, 34(1), 73-88. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/12523807 Pheasant, S., & Oneill, D. (1975). Performance in Gripping and Turning - Study in Hand-Handle Effectiveness. Applied Ergonomics, 6(4), 205-208. doi:Doi 10.1016/0003-6870(75)90111-8 PPPA. (1970). Poison Prevention Packaging Act of 1970, (PPPA) Public Law 91-601, 84 Stat.1670, December 30, 1970, Sec. 3; 15 U.S.C. 1472. 129 Ramsey, J. D. (1985). Ergonomic Factors in Task-Analysis for Consumer Product- Safety. Journal of Occupational Accidents, 7(2), 113-123. doi:Doi 10.1016/0376- 6349(85)90022-7 Ray, G. G., & Biswas, S. P. (2012, 9-12 July 2012). Investigation on hand twisting force developed by the Indian elderly for opening different kinds of bottle packaging. Paper presented at the 2012 Southeast Asian Network of Ergonomics Societies Conference (SEANES). Rennie, A. M. (1981). The Application of Ergonomics to Consumer Product Evaluation. Applied Ergonomics, 12(3), 163-168. doi:Doi 10.1016/0003-6870(81)90005-3 Replogle, J. O. (1983). Hand Torque Strength with Cylindrical Handles. Proceedings of the Human Factors Society Annual Meeting, 27(5), 412-416. doi:10.1177/154193128302700519 Rice, M. S., Leonard, C., & Carter, M. (1998). Grip strengths and required forces in accessing everyday containers in a normal population. American Journal of Occupational Therapy, 52(8), 621-626. doi:DOI 10.5014/ajot.52.8.621 Rodgers, G. B. (1996). The safety effects of child-resistant packaging for oral prescription drugs. Two decades of experience. JAMA, 275(21), 1661-1665. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8637140 Rodgers, G. B. (2002). The effectiveness of child-resistant packaging for aspirin. Arch Pediatr Adolesc Med, 156(9), 929-933. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12197802 Rodriguez-Falcon, E. M., & Yoxall, A. (2010). Accessibility is in the Palm of Your Hand. Designing Inclusive Interactions, 45-+. doi:Doi 10.1007/978-1-84996-166-0_5 Rohles, F. H., Laviana, J. E., & Moldrup, K. (1984). Wrist Twisting Strength of four Year Olds: Product Packaging Implications. Proceedings of the Human Factors Society Annual Meeting, 28(1), 90-94. doi:10.1177/154193128402800126 Rowson, J., & Yoxall, A. (2011). Hold, grasp, clutch or grab: Consumer grip choices during food container opening. Applied Ergonomics, 42(5), 627-633. doi:10.1016/j.apergo.2010.12.001 130 SafeKids. (2013). An In-Depth Look at Keeping Young Children Safe Around Medicine. Retrieved from https://www.safekids.org/sites/default/files/documents/ResearchReports/An%20I n- Depth%20Look%20at%20Keeping%20Young%20Children%20Safe%20Around %20Medicine%20-%20March%202013.pdf Scherzo, R., Strainer, MAC.,. (1970). "We Can Prevent Accidental Poisonings By Using Child-Resistant Containers". Unpublished paper presented at the Congress De Pediatria, Bogota, Columbia. Scriba, G. T. (1961). The Federal Hazardous Substances Labeling Act. Business Lawyer, 17(1), 137-156. Retrieved from ://WOS:A1961CAJ6500007 Seo, N. J., & Armstrong, T. J. (2008). Investigation of Grip Force, Normal Force, Contact Area, Hand Size, and Handle Size for Cylindrical Handles. Human Factors, 50(5), 734-744. doi:10.1518/001872008x354192 Seo, N. J., Armstrong, T. J., Ashton-Miller, J. A., & Chaffin, D. B. (2007). The effect of torque direction and cylindrical handle diameter on the coupling between the hand and a cylindrical handle. Journal of Biomechanics, 40(14), 3236-3243. doi:10.1016/j.jbiomech.2007.04.023 Snyder, R. G. (1975). Physical Characteristics of Children As Related to Death and Injury for Consumer Product Design and Use. UM-HSRI-BI-75-5 Final Report Contract FDA-72-70 May 1975 Retrieved from https://math.nist.gov/~SRessler/anthrokids/child.html#Description%20of%20Data %20Presentation Spiller, H. A., Beuhler, M. C., Ryan, M. L., Borys, D. J., Aleguas, A., & Bosse, G. M. (2013). Evaluation of Changes in Poisoning in Young Children 2000 to 2010. Pediatric Emergency Care, 29(5), 635-640. doi:10.1097/PEC.0b013e31828e9d00 Tadros, A., Layman, S. M., Davis, S. M., Bozeman, R., & Davidov, D. M. (2016). Emergency Department Visits by Pediatric Patients for Poisoning by Prescription Opioids. The American journal of drug and alcohol abuse, 42(5), 550-555. doi:10.1080/00952990.2016.1194851 131 Tomlinson, S. E., Lewis, R., Carre, M. J., & Franklin, S. E. (2013). Human finger friction in contacts with ridged surfaces. Wear, 301(1-2), 330-337. doi:10.1016/j.wear.2012.12.039 Yam, K. L. (2010). The Wiley encyclopedia of packaging technology. John Wiley & Sons. Yoxall, A., & Janson, R. (2008). Fact or friction: a model for understanding the openability of wide mouth closures. Packaging Technology and Science, 21(3), 137-147. doi:10.1002/pts.785 Yoxall, A., Janson, R., Bradbury, S. R., Langley, J., Wearn, J., & Hayes, S. (2006). Openability: Producing design limits for consumer packaging. Packaging Technology and Science, 19(4), 219-225. doi:10.1002/pts.725 Yoxall, A., Langley, J., Janson, R., Lewis, R., Wearn, J., Hayes, S. A., & Bix, L. (2010). How Wide Do You Want the Jar?: The Effect on Diameter for Ease of Opening for Wide-mouth Closures. Packaging Technology and Science, 23(1), 11-18. doi:10.1002/pts.874 Yoxall, A., Luxmoore, J., Austin, M., Canty, L., Margrave, K. J., Richardson, C. J., . . . Lewis, R. (2007). Getting to grips with packaging: Using ethnography and computer simulation to understand hand-pack interaction. Packaging Technology and Science, 20(3), 217-229. doi:10.1002/pts.755 Yoxall, A., Rodriguez-Falcon, E. M., & Luxmoore, J. (2013). Carpe diem, Carpe ampulla: a numerical model as an aid to the design of child-resistant closures. Appl Ergon, 44(1), 18-26. doi:10.1016/j.apergo.2012.04.006 Zaloshnja, E., Miller, T., Jones, P., Litovitz, T., Coben, J., Steiner, C., & Sheppard, M. (2008). The impact of poison control centers on poisoning-related visits to EDs - United States, 2003. American Journal of Emergency Medicine, 26(3), 310-315. doi:10.1016/j.ajem.2007.10.014 132