A TWO-LEVEL APPROACH TO INCREASE HAND HYGIENE COMPLIANCE AMONG
HEALTHCARE WORKERS IN A COMMUNITY HOSPITAL
By
Jay Jordan Liggins

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Epidemiology—Master of Science
2017

ABSTRACT
A TWO-LEVEL APPROACH TO INCREASE HAND HYGIENE COMPLIANCE AMONG
HEALTHCARE WORKERS IN A COMMUNITY HOSPITAL
By
Jay Jordan Liggins
Hand hygiene is a vital tool in the prevention of nosocomial infections such as central
line associated bloodstream infections (CLABSIs) and catheter associated urinary tract infections
(CAUTIs). This study implemented a two-level intervention centered on feedback and reminders
of the importance of hand hygiene. The objective of this study was to increase hand hygiene
compliance of healthcare workers to 70%, and to decrease CLABSI and CAUTI rates among inpatient units within a community hospital. Baseline data was collected from June 2016 to March
2017 and consisted of 1,991 hand hygiene observations. The two-level intervention was
implemented over six weeks when 3,438 hand hygiene observations were recorded. After the
first five weeks of the original intervention, and additional feedback component added to the
sixth week, there was a significant increase in hand hygiene compliance, from an average rate of
53% (November 2016 to March 2017) to 65% after six weeks (p-value < 0.0001). Based on the
results of this study, the odds of compliance are 1.17 (95% CI: 1.10-1.25) times higher for every
two weeks of additional the intervention time, controlling for healthcare worker type and for
hospital unit. There were ultimately too few CLABSI and CAUTI infection cases to analyze.
However, the significant increase in hand hygiene compliance in just over a month’s time,
concomitant with weekly feedback and daily reminders for healthcare workers who are in regular
and frequent contact with patients could potentially provide infection prevention units with
knowledge to help them reach their infection prevention targets.

ACKNOWLEDGMENTS

This thesis could not have been completed without the guidance from the great minds that
are a part of my committee: Drs. Lixin Zhang, David Todem, and Dorothy R. Pathak. Thank you
for offering your wisdom through the study design and analysis stages of this thesis. A special
thank you to Dr. Todem for providing assistance with the SAS coding. I would also like to thank
the professors who taught me in my undergraduate studies and caused my interest in
epidemiology to grow into what it is today: Drs. Elizabeth Wasilevich, Julie Wirth, David
Barondess, and Dorothy R. Pathak. I am also thankful to the infection prevention team at the
hospital that agreed to collaborate with me for this study. Everyone was extremely helpful and I
learned life and career lessons that I will never forget. Lastly, I want to thank my family and
loved ones for being there for me through this entire journey.

iii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. vi
LIST OF FIGURES .......................................................................................................... vii
KEY TO ABBREVIATIONS .......................................................................................... viii
CHAPTER 1: Aims and Objectives .....................................................................................1
CHAPTER 2: Background and Significance .......................................................................2
2.1 Central Line Associated Bloodstream Infections ..............................................2
2.1.1 What makes a CLABSI Case a Case?.................................................3
2.1.2 CLABSI Prevention ............................................................................4
2.2 Catheter Associated Urinary Tract Infections....................................................5
2.2.1 What Makes a CAUTI Case a Case? ..................................................5
2.2.2 CAUTI Prevention ..............................................................................6
2.3 Cost of CAUTIs and CLABSIs .........................................................................8
2.4 Hand Hygiene ....................................................................................................9
2.4.1 How Hand Hygiene Compliance is Captured .....................................9
2.4.2 The Current State of Hand Hygiene Compliance Interventions .......10
2.5 Significance of this Intervention Study............................................................11
CHAPTER 3: Methods ......................................................................................................12
3.1 Study Area .......................................................................................................12
3.2 The Two-Level Hand Hygiene Intervention ....................................................12
3.2.1 Weekly Posting of Hand Hygiene Compliance ................................12
3.2.2 Daily Reminder of the Importance of Hand Hygiene .......................13
3.3 Recording Hand Hygiene Observations...........................................................14
3.4 Joint Intervention .............................................................................................15
3.5 Testing Intervention Effect/Statistical Analysis ..............................................15
CHAPTER 4: Results ........................................................................................................17
4.1 Descriptive Characteristics of Baseline and Intervention Data .......................17
4.1.1 Baseline Data Characteristics ...........................................................17
4.1.2 Intervention Data Characteristics ......................................................17
4.2 Intervention Effect ...........................................................................................18
4.2.1 Trends in Baseline Data ....................................................................18
4.2.2 Intervention Effect on Hand Hygiene Compliance ...........................18
CHAPTER 5: Discussion/Conclusion ...............................................................................20
APPENDICES ...................................................................................................................24
APPENDIX A Tables ............................................................................................25

iv

APPENDIX B Figures ...........................................................................................31
APPENDIX C Hand Hygiene Observer Test ........................................................39
APPENDIX D IRB Approval Letter......................................................................43
REFERENCES ..................................................................................................................45

v

LIST OF TABLES

Table 1: Hand Hygiene Compliance and HCW Distribution for Baseline (June 2016 March 2017). Monthly hand hygiene compliance and distribution of healthcare workers (HCW)
during baseline period (June 2016 – March 2017). ...........................................................26
Table 2: Hand Hygiene and HCW Distribution for Study Intervention (Apr 2017 - May 7,
2017) and Joint Intervention (May 8, 2017 - May 12, 2017). Monthly hand hygiene
compliance and distribution of healthcare workers (HCW) during study intervention period (Apr
2017 – May 7, 2017) and joint intervention period (May 8, 2017 – May 12, 2017). ........27
Table 3: Hand Hygiene Compliance and Unstructured Model Results. Hand hygiene
compliance and full logistic regression model: hand hygiene compliance = month + HCW type +
Unit (random effects). P-values are significant at a value <0.05. SAS contrast statements were
used to see if there was a temporal (month) effect on the baseline data.
HCW=healthcare worker ..................................................................................................28
Table 4: SAS Contrasts. SAS Contrasts used to see if there was a temporal (month) effect on
baseline data within full logistic regression model: hand hygiene compliance = month + HCW
type + Unit (random effects). P-values are significant at a value <0.05............................28
Table 5: Hand Hygiene Compliance and Reduced Model. Results of segmented regression
model adjusting for healthcare workers and unit with a continuity condition imposed on the
regression parameters to constrain the fitted proportions to be the same on both sides of the
month of March (regarded as a knot). P-values are significant at <0.05. Jun 2016- Jul 2016 is
reference category. HCW= healthcare worker...................................................................29
Table 6: List of Studies that Include Feedback of HH Performance and Frequency of
Feedback. List of Hand Hygiene studies from Schweizer et al. 2014 and Kingston et al 2016 that
included feedback of hand hygiene performance. Studies with an ‘*’ have an immediate or
continuous feedback frequency with HCWs being directly observed ...............................30

vi

LIST OF FIGURES

Figure 1: Hand Hygiene Compliance Baseline (June 2016 - March 2017). Hand hygiene
compliance rates of 12 in-patient units during baseline period
(June 2016 - March 2017) ..................................................................................................32
Figure 2: Hand Hygiene Compliance Baseline + Intervention (June 2016 - May 2017). Hand
hygiene compliance for baseline and intervention time periods for full model. Blue diamonds =
baseline period; red squares = two-level intervention; green triangle = joint intervention with
partnered hospital. May 17a = May 1 – 7, 2017;
May 17b = May 8- May 12, 2017. ....................................................................................33
Figure 3: Central Line Insertion Site. Illustration of central line insertion site. Source: Pictures
of Central Venous Catheters. Digital image. The Joint Commission. The Joint Commission.
Web.
<https://www.jointcommission.org/assets/1/6/CLABSI_Toolkit_Tool_13_Pictures_of_Central_
Venous_Catheters_F.pdf>. ................................................................................................34
Figure 4: Indwelling Urinary Catheter System. Illustration of indwelling urinary catheter
system. Source: Indwelling Foley Catheter for Women. Digital image. Metro Health University
of Michigan Health. Metro Health University of Michigan Health, Web.
<https://metrohealth.net/healthwise/indwelling-foley-catheter-for-women/>. ..................35
Figure 5: Weekly Hand Hygiene Huddle Helper. Example of weekly hand hygiene huddle
helper that was distributed to units every other week. .......................................................36
Figure 6: Positive Bi-weekly Huddle Helper. Example of positive biweekly assessment huddle
helper that was distributed to units that had increased hand hygiene compliance after 2 weeks.
............................................................................................................................................37
Figure 7: Negative Bi-weekly Huddle Helper. Example of negative biweekly assessment
huddle helper that was distributed to units that had decreased
hand hygiene compliance after 2 weeks. ...........................................................................38

vii

KEY TO ABBREVIATIONS

CLABSI

Central Line Associated Bloodstream Infection

CAUTI

Catheter Associated Urinary Tract Infection

MRSA

Methicillin-Resistant Staphylococcus aureus

SSI

Surgical Site Infection

CDC

Centers for Disease Control and Prevention

NHSN

National Healthcare Safety Network

UTI

Urinary Tract Infection

MDRO

Multi-Drug Resistant Organism

VAP

Ventilator associate Pneumonia

Joint Commission

Joint Commission on Accreditation of Healthcare Organizations

HH Compliance

Hand Hygiene

Medsurg Unit

Medical Surgical Unit

viii

CHAPTER 1: Aims and Objectives
This thesis is an intervention study with an overall goal to increase hand hygiene
compliance. Hand hygiene is defined as the act of washing one’s hands with soap and water,
antiseptic hand wash or using antiseptic hand rub (14). The intervention that is being conducted
is defined as weekly feedback of hand hygiene compliance rates on a unit’s quality board, and
daily reminders of the importance of hand hygiene. This intervention is being conducted to see if
a two-level approach, feedback of hand hygiene compliance rates and reminders of its
importance, is able to increase hand hygiene compliance. This intervention consists of two aims
accompanied by two hypotheses.
Aim 1: To determine the impact on hand hygiene compliance using a two-level
intervention focused on hand hygiene feedback and reminders of the importance of hand
hygiene.
Hypothesis 1: A two-level intervention focused on hand hygiene compliance feedback
and reminders of the importance of hand hygiene, will increase hand hygiene compliance from
the 53% average rate of the five months prior to the intervention to 70%.
Aim 2: To see if hand hygiene compliance has an effect on Cather Associated Urinary
Tract Infection (CAUTI) and Central Line Associated Bloodstream Infection (CLABSI) rates.
Hypothesis 2: As hand hygiene compliance increases to 70%, a reduction in CAUTI and
CLABSI rates will be seen.

1

CHAPTER 2: Background and Significance
When a patient is admitted to a hospital, it is to receive treatment for a disease, infection
or condition, not to become infected. Infections acquired during a patient’s hospital stay are
called, “nosocomial infections.” According to the Center for Disease Control (CDC), “on any
given day, 1 in 25 hospital patients have at least one nosocomial infection,” (10). Nosocomial
infections can be acquired through cross contamination between patients or through invasive
procedures that might be used to treat patients. Some nosocomial infections include, but are not
limited to, Clostridium difficile (C. difficile), Methicillin-Resistant Staphylococcus Aureus
(MRSA), central line associated bloodstream infections (CLABSIs), catheter associated urinary
tract infections (CAUTIs), and Surgical Site Infections (SSIs). This study focuses on CLABSIs
and CAUTIs due to the cost of a single CLABSI infection and the rate at which CAUTIs occur,
being the highest among nosocomial infections. This chapter will go into detail on the infections
focused on in this study, and how hand hygiene can affect the nosocomial infection rate.
2.1 Central Line Associated Blood Stream Infections
Central Line Associated Bloodstream Infections (CLABSIs) are the result of bacteria
entering the body through a central line placement. A central line is a type of intravascular
device used in hospitals. Central lines are inserted into the body through one of the major blood
vessels and usually end at, or close to, a patient’s heart. Central lines are used for the withdrawal
of blood, hemodynamic monitoring, and the infusion of medicine into a patient’s body (10).
Illustration of a Central Line can be seen in Figure 3 (Appendix B).
In order for a CLABSI to occur, bacteria must be able to attach to the extraluminal or
intraluminal surface of a central line. After bacteria attaches to the central line, the bacteria then
begin to colonize, and could lead to an infection and potentially hematogenous dissemination (6).

2

There are multiple opportunities for bacteria to enter the body through a central line system.
These opportunities for bacterial contamination present themselves through a varying number of
causes: device defects, contamination from outsides sources (i.e. healthcare workers or visitors)
or it could be due to the severity of the disease that the patient is being treated for (6). Risks
associated with CLABSIs include factors such as the central line insertion point, number of times
the central line is drawn from, and central line insertion attempts. There is a common factor
among most, if not all, CLABSIs: “Alteration of the patient’s skin flora, either as a result of
antimicrobial therapy or by colonization with an epidemic strain carried on the hands of hospital
personnel,” (6). Staphylococcus aureus and coagulase-negative staphylococci are bacteria that
are mostly associated with CLABSIs (6).
2.1.1 What makes a CLABSI Case a Case?
According to the CDC, a CLABSI becomes a case when there is “a laboratory-confirmed
bloodstream infection where the central line was in place for greater than two calendar days on
the date of the event.” The ‘date of event’ is the date that a criterion for meeting a laboratoryconfirmed bloodstream infection first occurs. The criteria that need to be met consist of two
levels 1) patient must have either a fever above 38℃, chills, or hypotension and 2) a blood
specimen drawn that is not related to an infection at another site in the patient. The blood
specimen must be drawn twice on separate occasions. All criteria must be met within the sevenday infection window period as defined by the National Healthcare Safety Network (NHSN).
The infection window period begins on the date of either the first positive blood culture or the
first observed sign or symptom of disease and consists of the three days before and the three days
after. The central line also still needs to be inserted in the patient the day of the laboratory
confirmation, or the day before. The only exception to this rule is if the laboratory confirmation

3

comes on the same day, or the day after, the central, line was removed (11). In cases where a
patient is moved to another facility, and they have an implanted central line, the day that central
line is first accessed is considered the first day for CLABSI surveillance. According to the CDC
“Access is defined as line placement, insertion of needle into the port, infusion or withdrawal
through the line,” (11).
In the United States, the NHSN has been keeping track of CLABSI rates in hospitals.
Hospitals across the nation have been reporting their CLABSI rates to the CDC since 1970 (6).
According to the most recent report from the CDC, 3,655 acute care hospitals were able to report
CLABSI data and reported that there were 17,758 infections in 2014 (11). CLABSI rates are sent
to the NHSN, by month, as the number of infections over the number of days that a central line
has been inserted in a patient; the resulting fraction is then multiplied by 1000 (11).
=

#

1000

#

2.1.2 CLABSI Prevention
There are guidelines in place to help prevent CLABSI cases. For healthcare workers
certain steps are needed to maximize patient safety before and during central line placement 1)
perform hand hygiene before central line insertion, 2) use aseptic technique (e.g. wear a mask,
cap, gown, sterile gloves, and a sterile full body drape), 3) place the line where it minimizes the
risk of infection, and 4) prepare the insertion site with >0.5% chlorhexidine with alcohol, and
place a sterile dressing over the line placement site (12).
Maintaining a central line is just as important as the insertion of a central line. The CDC
has come up with CLABSI prevention strategies during the maintenance of a central line as well:
be sure to perform hand hygiene, bathe ICU patients over two months of age with chlorhexidine
on a daily basis, scrub the access port with antiseptic prior to use, use only sterile devices while

4

accessing lines, immediately replace dirty, wet, or out of normal position dressing, and be sure to
routinely change the dressings around the central line using aseptic technique (12). An important
step in CLABSI prevention is to remove unnecessary central lines (12).
2.2 Catheter Associated Urinary Tract Infections
Catheter Associated Urinary Tract Infections (CAUTIs) are Urinary Tract Infections
(UTIs) that are associated with urinary catheter placement. Urinary catheters (Foley catheters)
are devices that are used to watch urine levels during an ill patient’s hospital stay or after a
patient’s surgery. Illustration of a Foley catheter system can be seen in Figure 4 (Appendix B).
The most prevalent risk factor for a CAUTI is the length of time that a catheter is
inserted in a patient. The longer a catheter is inserted, the greater the chance for infection (21).
CAUTIs make up for 40% of nosocomial infections in hospitals in the United States, and
approximately 15-25% of patients have a urinary catheter inserted during their hospital stay (21).
According to the CDC, “of all the UTIs acquired in the hospital, approximately 75% are
associated with a urinary catheter,” (13). Unlike normal UTIs, there are a wide range of
pathogens that are associated with CAUTIs and many of them are multi-drug resistant organisms
(MDROs). Pathogens that have been commonly associated with CAUTIs are enterococci, gramnegative bacilli, and funguria (21).
2.2.1 What Makes a CAUTI Case a Case?
What makes CAUTIs different from normal UTIs is the placement of the urinary
catheter. This urinary catheter disrupts the normal environment of the patient and allows easy
access of pathogens into the bladder. Pathogens that gain access to the bladder can come from
the patient’s own flora, healthcare workers, or other outside objects (20).

5

UTIs are diagnosed using Symptomatic Urinary Tract Infection criteria, Asymptomatic
Bacteremic UTI or Urinary System Infection criteria. The diagnosis of a CAUTI is given to a
UTI when three criteria are met during the seven day infection window period: 1) A patient must
have either a fever greater than 38℃, suprapubic tenderness with no cause, costovertebral angle
pain/tenderness with no cause, or issues with urinary urgency, frequency, or dysuria. Urinary
urgency, frequency or dysuria cannot be counted as symptoms when catheter is in place. 2) A
urinary catheter must be inserted for more than two calendar days on the date of UTI diagnosis.
The catheter also needs to be in place the day of UTI diagnosis or the day before. If the catheter
was removed, then the date of UTI diagnosis must be the day of removal, or the day after
catheter removal. 3) A patient must also have a urine culture with no more than two species, one
of which is a bacterium of ≥105 CFU/ml (colony forming units per milliliter) (13).
Hospitals are also required to report monthly CAUTI rates to the NHSN similar to
CLABSI rate reporting. CAUTIs are calculated by the number of infections over the number of
days that a patient has a Foley catheter inserted; the resulting fraction is then multiplied by 1000
(13).
=

#

1000

#

2.2.2 CAUTI Prevention
CAUTI prevention is similar to CLABSI prevention. Guidelines have been implemented,
by the CDC, to help minimize the risk of infection during Foley catheter use, insertion, and
maintenance. In regards to catheter use, guidelines state that catheters should only be used in the
appropriate situations and in place for the necessary amount of time. Foley catheter use should be
minimized in all patients as much as possible as to lower the risk for CAUTI. Foley catheters
should be avoided in nursing home settings, and used only as necessary in operative patients.

6

The guidelines also state that if operative patients need a Foley catheter, that it should be
removed as soon as possible after their operation. Removal of a urinary catheter within 24 hours
is preferred, unless there is a need for the catheter to remain inserted for a longer duration of time
(15).
If a Foley catheter needs to be inserted, as with central lines, there are steps that can be
taken to help minimize a patient’s risk for a CAUTI during insertion. Guidelines for catheter
insertion state the following: hand hygiene should be performed before and after the Foley
catheter is inserted or handled, only properly trained healthcare workers who are aware of aseptic
technique and the use of sterile equipment (i.e. gloves, drapes, sponges, sterile solution for
cleaning, and single-packet use of lubricant jelly for insertion) for catheter insertion and
maintenance should be handling Foley catheters (15). Healthcare workers should also properly
secure Foley catheters to prevent them from moving around, and use the smallest catheter
possible to minimize damage to the urethra (15).
Maintenance of Foley catheters is equally as important as the insertion of the Foley
catheter. CDC guidelines for maintenance of Foley catheters help to minimize the risk for a
CAUTI while maintaining a Foley catheter: follow aseptic technique and keep a closed drainage
system, replace the catheter using sterile equipment and aseptic technique if the catheter
disconnects or begins to leak, consider using urinary catheters that are already connected and
have sealed junctions, be sure to have unobstructed urine flow (no kinks in the tubing), keep the
urine collection bag below the bladder and not on the floor, be sure to empty the urine collection
bag regularly and use a separate container for each patient, be sure to wear gloves and gowns
appropriately during urinary catheter maintenance, only change indwelling urinary catheters if
there is an infection, obstruction or if there is an issue with the closed system, do not use routine

7

antibiotic treatments in patients to prevent a CAUTI unless clinically appropriate, and do not use
antiseptics to clean periurethral area to prevent a CAUTI while the catheter is inserted (15).
2.3 Cost of CAUTIs and CLABSIs
CAUTIs and CLABSIs are not just negative outcomes for the patient’s health; they are
also negative outcomes in terms of cost for the hospital. A meta-analysis looked at the cost, and
the increased length of stay due to nosocomial infections (50).
In terms of cost, a CLABSI is the most expensive, per case, nosocomial infection
($45,814) when compared to other common nosocomial infections such as ventilator-associated
pneumonia ($40,144), SSIs ($20,785), C. difficile infections ($11,285), and CAUTIs ($896) (50).
MRSA was added as an additional category under the nosocomial infections, if studies included
in the meta-analysis recorded presence of MRSA (50). MRSA was only added as an additional
category under CLABSIs and SSIs. When MRSA was present in CLABSIs and SSIs the cost
attributed to that infection increased from $45,814 to $58,614 and from $20,785 to $42,300
respectively (50).
According to the most recent report filed by the CDC on the health status of the United
States, the average number of days spent in the hospital utilizing inpatient services is 6.1 days
(34). According to the meta-analysis, SSIs, ventilator-associated pneumonia, and CLABSIs
increase the length of a patient’s stay beyond the average number of hospital days: 11.2 days
(95% CI: 10.5-11.9), 13.1 days (95% CI: 11.9-14.3), and 10.4 days (95% CI: 6.9-15.2)
respectively (50). C. difficile infections did not increase the length of stay past the 6.1-day
average length of stay. According to the meta-analysis the length of stay recorded for patients
with C. difficile infections was 3.3 days (95% CI: 2.7-3.8) (50). As for length of stay, in regards

8

to CAUTIs, there was not enough data recorded in the studies gathered by the meta-analysis to
provide length of stay information (50).
2.4 Hand Hygiene
Even with the aforementioned guidelines on prevention of CLABSIs and CAUTIs, during
line/Foley placement and management, hospitals continue to look for more interventions in order
to cut down on CAUTI and CLABSI rates. An intervention that has been continually worked on
is the increase of hand hygiene compliance among healthcare workers. Hand hygiene is a vital
tool in stopping the spread of disease, and is the first step in the insertion and management of
both central lines and urinary catheters (6, 21). According to the CDC, hand hygiene is defined
as the washing of hands using soap and water, antiseptic hand wash, or the use of antiseptic hand
rub (14).
An intervention study published in the year 2000, found that increasing the hand hygiene
compliance from 47.6% to 66.2% showed a significant decrease in nosocomial infection rates
(38). Despite the importance of hand hygiene compliance and display of effect moving from just
a low to medium compliance level (38), hand hygiene compliance among healthcare workers
continues to be low. According to the Joint Commission, hand hygiene compliance among
healthcare workers is below 50% in most hospitals (25).
2.4.1 How Hand Hygiene Compliance is Captured
In most studies, in order to capture hand hygiene compliance, hospitals use covert
observers to monitor healthcare workers’ hand hygiene compliance. A study done in 2016
reported a 40.7% difference in hand hygiene observations of nurses who were directly observed
(85.8% compliance) versus hand hygiene observations of nurses who were covertly observed
(45.1% compliance) (27). The differences in covert and direct observations brought to light an

9

issue that makes hand hygiene observation difficult: the Hawthorne effect. The Hawthorne effect
is an idea that a person will behave differently if they are aware that they are being watched.
Another aspect to consider when gathering hand hygiene observations and implementing
interventions, is the burden of compliance. The uneven burden of hand hygiene compliance
addresses the issue that not all healthcare worker types have equal opportunities for patient
interaction that require hand hygiene. A recent study, found a statistically significant difference
in hand hygiene opportunities at a patient’s bedside, recorded in a seven-day period, for nurses
and physicians (4). During the seven-day period of hand hygiene observation, it was found that
there were 13,989 hand hygiene observations for nurses compared to the 2,516 observations for
physicians (4). This is important information so that hand hygiene interventions can be aimed at
healthcare workers that have more patient interaction, and thus have a greater impact on hand
hygiene compliance
2.4.2 The Current State of Hand Hygiene Compliance Interventions
Multiple intervention studies have been done trying to increase and sustain hand hygiene
compliance. Hand hygiene compliance interventions range from a single intervention to a multiintervention approach. According to a meta-analysis performed in 2013, 78% of the 45 studies
gathered consisted of a multi-level intervention approach (42). Despite the large number of
studies that consist of multi-level interventions it seems that interventions that consist of one to
two levels are the most effective. It was reported that studies that had between one to two levels
of intervention showed the largest increase in hand hygiene compliance with a statistically
significant, pooled odds ratio of 3.44 (95% CI: 1.11-10.68) (42). The discovery of a hand
hygiene intervention with as few levels as possible is key as to not add too much of a burden to
infection prevention teams.

10

2.5 Significance of this Intervention Study
This intervention falls under the two-level intervention approach that is seen to have the
greatest effect on increasing hand hygiene compliance (42). The first level of this study’s
intervention is the implementation of weekly reporting of hand hygiene compliance for each unit,
and the second level of this study’s intervention is the reminders of the importance of hand
hygiene.
Among the hand hygiene intervention studies from current literature, included in the
studies done by Schweizer et al. and Kingston et al., feedback of hand hygiene compliance is
included as part of the intervention in 33 of the combined 57 studies (26, 42). In the 33 studies,
none of them have weekly feedback, and only a few studies have immediate or continuous
feedback as displayed in Table 6 (Appendix A). The observations for the studies with immediate
or continuous feedback are not covert observations. The intervention of this current study is
significant because the hand hygiene observations are done covertly and are accompanied with
weekly feedback.
This study is also significant because the intervention is targeting nurses. According to
current literature, nurses hold the majority of hand hygiene observations (4). The intervention in
this study is rolled out to all healthcare workers, but nurses get the most reminders of the
importance of hand hygiene and of their unit’s hand hygiene compliance rate.

11

CHAPTER 3: Methods
3.1 Study Area
Collaborating with an Infection Prevention team in a community hospital, an intervention
was implemented that took aim to increase hand hygiene compliance among healthcare workers.
This intervention targeted 12 inpatient medical surgical units and excluded emergency units,
intensive care units, critical care units, and maternity wards.
3.2 The Two-Level Hand Hygiene Intervention
The hand hygiene intervention implemented in this study was composed of two parts:
weekly posting of hand hygiene compliance and the daily reminder of the importance of hand
hygiene.
3.2.1 Weekly Posting of Hand Hygiene Compliance
Healthcare workers in the hospital were notified by the Infection Prevention team that the
Infection Prevention team will be monitoring hand hygiene compliance rates, and that the hand
hygiene compliance rates will be posted on the unit’s quality board. The unit’s quality board can
be seen by all healthcare workers, but is mainly seen and utilized by the unit’s nurses and nurse
assistants. The quality board is where all of the unit’s goals and statistics are placed, so it is
frequently visited by the unit staff. The nursing managers of each unit were also notified that
they were responsible for posting the hand hygiene compliance rates on their unit’s quality
board. The hand hygiene compliance rates for each unit were displayed in graph format that
showed two graphs: a graph with their unit’s compliance alone (unit graph) and a graph that
showed all of the observed units’ hand hygiene compliance rates next to each other (unit-wide
graph). Each week both graphs would change between two different versions. For weeks one,
three, and five, the unit graph would be of the unit’s hand hygiene compliance for the previous

12

week and the unit-wide graph would show all of the observed units’ hand hygiene compliance
for the previous week (see Figure 5 in Appendix B). For weeks two, four and six, the graphs
would be a bi-weekly assessment: the unit graph would show the unit’s hand hygiene compliance
rate for the past two weeks and the unit-wide graphs would show the observed units’ average
hand hygiene compliance rate individually for the past two weeks (see Figures 6 and 7 in
Appendix B for examples).
The graphs that displayed the unit’s hand hygiene compliance were a part of a group of
documents that helped nursing managers talk about important topics during daily morning and
night shift meetings with their unit’s staff. The daily morning and night shift meetings were
called ‘daily huddles’ and the documents that help the nursing mangers were known as huddle
helpers; a huddle helper was created to talk about hand hygiene (see Figures 5, 6, and 7 in
Appendix B for examples of the huddle helpers).
3.2.2 Daily Reminder of the Importance of Hand Hygiene
The second part of this intervention involves the daily reminders of the hand hygiene
compliance rates and the importance of hand hygiene by the nurse managers in the daily huddles.
On the hand hygiene huddle helper, they were given talking points to help stress the importance
of hand hygiene in stopping the spread of disease. These documents also had talking points for
the bi-weekly assessment. If a unit’s hand hygiene compliance rate for the previous two weeks
was on a decline, the hand hygiene huddle helper continued to stress the importance of hand
hygiene compliance, in accordance with current methods at the partnered hospital. The huddle
helper reporting a declining hand hygiene compliance rate also stressed that the healthcare
workers talk amongst themselves on what the potential barriers are to performing hand hygiene,
Figure 8 (Appendix B). If a unit’s compliance rate for the previous two weeks was on an

13

increasing trend, the hand hygiene huddle helper was altered to tell the unit to keep up the good
work with the increasing hand hygiene compliance rates. The huddle helper also continued to
stress the importance of high hand hygiene compliance, Figure 7 (Appendix B).
3.3 Recording Hand Hygiene Observations
Hand hygiene observations were recorded via stealth observation, the current hand
hygiene observation method of the partnered hospital. The partnered hospital used the hand
hygiene data collection and observer validation technique that is given out by the Joint
Commission on Accreditation of Healthcare Organizations (the Joint Commission). The
partnered hospital used volunteer data collectors from different departments in the hospital. Each
potential data collector must go through an hour-long hand hygiene observation education
seminar. The education seminar taught the volunteers how to be data collectors and taught them
what constituted as proper documentation of a hand hygiene observation. After the seminar, the
potential data collectors took a sixteen-question test that consisted of different scenarios they
might see while out collecting data (Appendix C). The potential data collectors are only allowed
to get one question wrong on this test. If the potential data collector passes the test, then he/she is
considered an official stealth, hand hygiene data collector and may start collecting hand hygiene
observations.
While out collecting data, the data collectors are looking for observations that fall under
the partnered hospital’s hand hygiene policy. Healthcare workers at the partnered hospital are
taught, at new employee orientation, that any time they enter or exit a patient’s room they are to
follow proper hand hygiene. According to the partnered hospital’s policy, proper hand hygiene
includes the action of hand washing, antiseptic hand washing, or antiseptic product washing.
Entry and exit to a patient’s room are counted as two separate hand hygiene opportunities. Hand

14

hygiene observations were not documented if the data collector was witnessing an emergent or
urgent situation, where the patient needed immediate medical attention. The data collectors were
also told that if they were uncertain if the healthcare worker did or did not wash their hands, then
do not record the observation. Hand hygiene observations were gathered at random times over a
24-hour time period each day of the week during the six-week intervention period.
3.4 Joint Intervention
During the sixth week of the intervention, May 8th to May 12th The partnered hospital
instituted “coaches.” The coaches were instructed to talk with healthcare workers who were not
compliant with hand hygiene, and to remind them to wash their hands upon entry or exit of a
patient’s room. The coaches were also required to document any reason given by the noncompliant healthcare worker as to why that healthcare worker did not wash their hands.
3.5 Testing Intervention Effect/Statistical Analysis
In order to evaluate the effect of the intervention, a segmented logistic regression model
with random effects was used. The cutoff points for the segmented regression were obtained
using the results from preliminary analyses that suggested that the rates of hand hygiene
compliance over the entire sampling period 1) remained constant between June 2016 and July
2016 (Period 1), between August 2016 to October 2016 (Period 2) and between November 2016
to March 2017 (Period 3); and 2) increased linearly between April 2017 to the second week of
May 2017 (Intervention Period). A continuity condition was imposed on the regression
parameters to constrain the fitted proportions to be the same on both sides of the month of March
(regarded as a knot). Because the rate between November 2016 and March 2017 was constant,
the test of the slope of the regression line for the intervention period serves as a test of the
intervention effect. Specifically, this was assessed by looking at the interaction term in the

15

segmented regression model between the constant line fitted for November 2016-March 2017
and April 2017 to the second week of May 2017. The parameter estimate for this interaction
term provides an estimate of the linear slope for the intervention. These analyses were adjusted
for the effect of the type of healthcare worker, and the unit random effects term. The random
effects were imposed to account for the within-unit association, resulting from the same
healthcare workers contributing multiple hand hygiene observations pre and during intervention
period. All statistical analyses were conducted using SAS 9.4, and statistical significance was
established for p-values less than 0.05.
We were unable to evaluate the effect of hand hygiene compliance on CLABSI and
CAUTI rates because the number of observed infections during the intervention period (6 weeks)
was insufficient to perform a meaningful statistical analysis.
The data collection and intervention methods were approved and determined exempt via
the institutional review board at Michigan State University and the partnered hospital.

16

CHAPTER 4: Results
4.1 Descriptive Characteristics of Baseline and Intervention Data
This section contains hand hygiene compliance distribution information during baseline
and intervention time periods.
4.1.1 Baseline Data Characteristics
Pre-intervention data consisted of hand hygiene observations gathered by the partnered
hospital between the months of June 2016 and March 2017. Table 1 (Appendix A) and Figure 1
(Appendix B) display the hand hygiene compliance rates, based on 1991 observations, from
June 2016 to March 2017. Table 1 (Appendix A) shows, that during the baseline time period,
nurses had the majority of hand hygiene observations with 900 observations, followed by nurse
assistants with 368 observations, doctors with 356 observations, and finally the ‘other’ category
with 367 observations.
4.1.2 Intervention Data Characteristics
Intervention data consisted of hand hygiene observations that were gathered during April
2017 and the first two weeks of May 2017. Figure 2 (Appendix B) displays the hand hygiene
compliance rates of both the baseline and intervention time periods. In Figure 2 (Appendix B)
the baseline period is marked with blue diamonds, the original study intervention period is
marked with red squares, and the joint intervention is marked with a green triangle.
According to Table 2 (Appendix A), there were 3,438 hand hygiene observations during
the intervention time period. During the intervention time period, nurses had the most
observations with 1,200 observations, followed by nurse assistants with 614 observations,
doctors with 424 observations and finally the ‘other’ category with 1200 observations.

17

4.2 Intervention Effect
This section contains the results of the segmented regression model used to test the effect
of the intervention.
4.2.1 Trends in Baseline Data
Before a segmented regression could be fitted to the data, preliminary analyses were
conducted. An unstructured logistic regression model was fit to the data with the month of march
as the reference category adjusting for healthcare worker type, and the hospital units serving as
random effects (Table 2, Appendix A). SAS contrast statements were used to test whether there
was a temporal (month) effect on the hand hygiene compliance rates, and whether any temporal
structural effect could be imposed on these pre-intervention rates. These analyses suggested that
three segments could be used to describe the rates of hand hygiene compliance before the
intervention. The first segment represents the period June 2016-July 2016, the second segment
represents August 2016-October 2016, and the third segment represents November 2016-March
2017. The results of this analysis can be seen in Table 4 (Appendix A). The temporal trend
(month effect), tested by contrast of June 2016 - March 2017, was statistically significant at a 5%
level (p-value = 0.0181) for the pre-intervention period. The months within the three plateaus, in
the baseline data, were not statistically significant from one another: June 2016 - July 2016 (pvalue = 0.7388), August 2016 - October 2016 (p-value = 0.7458), and November 2016 - March
2017 (p-value = 0.9782).
4.2.2 Intervention Effect on Hand Hygiene Compliance
The full logistic regression model and SAS contrast statements suggested that the hand
hygiene compliance rates were found to be linearly related to the time lag following the initiation
of the intervention. The results of the segmented regression model can be seen in Table 5

18

(Appendix A). Under the segmented regression model the intervention effect is captured through
the slope of the regression line for the six week period (Apr 2017 – 2nd week of May 2017)
compared to the five months preceding the intervention (Nov 2016 – Mar 2017). The estimate of
that slope and the associated standard error are 0.3181 and 0.0657, and a simple t-test showed
that the slope was statistically different from zero with a p-value below 0.0001. The odds of hand
hygiene compliance will be exp(0.5*0.3181) = 1.17 (95% CI: 1.10-1.25) times higher for every
two additional weeks during the intervention, controlling for healthcare worker type and unit as
random effects. The estimate of the slope was multiplied by 0.5 since the unit of time in the
model for estimating the hand hygiene compliance was one month. The variance measuring the
heterogeneity of units on the hand hygiene compliance was estimated to be 0.0956 with a
standard error 0.0481 (data not shown in Table 5) (p-value = 0.0117, based on mixture
distribution).

19

CHAPTER 5: Discussion/Conclusion
Based on the results of this study, the hand hygiene compliance was significantly
increased over the full six weeks of intervention but did not reach the 70% goal. The two-level
intervention did increase hand hygiene compliance from 53% to 59% during the five weeks it
was implemented. Once the intervention was combined with the coaches from the paired
hospital, hand hygiene compliance significantly increased to 65%. There were ultimately too
few CLABSI and CAUTI infection cases to analyze.
Previous literature has stated that observed hand hygiene opportunities are unevenly
distributed among healthcare workers (4). This study’s observed hand hygiene opportunities
were also unevenly distributed among healthcare workers. As presented in Table 1 (Appendix
A), in the baseline and intervention data, nurses had the most hand hygiene observations. In the
intervention data, the ‘other’ category had the second highest number of observations, which is
different from the baseline data. The difference between the number of observations in the
baseline and intervention data (Tables 1 and 2, Appendix A) could be due to the fact that the
partnered hospital cast a wider net for healthcare worker types during the intervention time
period. The ‘other’ category, during the intervention time period, contained more healthcare
worker types in the intervention data than it did in the baseline data.
When observing the baseline data, there was a dip in hand hygiene compliance
percentage between the months of August 2016 and October 2016. After further investigation, it
was made known that the partnered hospital had changed observation techniques during the
aforementioned time period. Specifically, in the months of June 2016 and July 2016, the
Infection Prevention team collected hand hygiene observations themselves. During the August
2016 to October 2016 time frame, the partnered hospital enlisted college volunteers to collect

20

hand hygiene observations; this action was to alleviate the burden on the Infection Prevention
team . The Infection Prevention team believed the observations during this time period (August
2016 to October 2016) to be of low quality, and returned to collecting observations themselves
during the months of November 2016 to March 2017. That is why the rate of hand hygiene
compliance for this time period was used as a reference for testing the effect of the intervention.
The use of a segmented regression model is one of the strengths of this study. While
adjusting for healthcare worker type and for heterogeneity across units, this model demonstrates
a statistically significant increasing slope during the intervention period when compared to the
five months prior to the intervention. Even though the goal of 70% hand hygiene compliance was
not met, the increasing slope shows that it could have been met given a longer time period of
intervention and assuming that the rate of hand hygiene compliance increase remained the same
for the additional time of intervention.
The data collection process implemented by the partnered hospital is a key to this study’s
successful gathering of hand hygiene observations, and a perceived strength of this study. The
partnered hospital was able to gather over 50 validated hand hygiene observers, who were able to
maintain their covert status, and gather observations 24 hours a day. With the amount of hand
hygiene observers, this study was able to have a large sample size for the intervention period (N
= 3438).
The posting of the units’ compliance rate graphs are also a possible key to the increase in
hand hygiene compliance. A case could be made that the units seeing compliance rates as a
weekly or bi-weekly goal, instead of a monthly goal, could be a reason for the significant
increase in the compliance rates. It could be argued that narrowing the window of focus could
have made healthcare workers feel like the goal was more attainable. It is also worth noting that

21

the compliance rate graphs allowed units to see how they compared to other units. The
aforementioned comparison could add an element of competition among units to have better
hand hygiene compliance rates than other units in the study.
Despite the strengths of this study, there are several limitations. Some of the units that
were excluded from this study (i.e. intensive and critical care units) are units that usually hold the
bulk of the CAUTI and CLABSI infection numbers for any hospital (4). The intensive care,
critical care, emergency department, and maternity ward were excluded during study design due
to lack of hand hygiene observers at the partnered hospital. It was not until a several weeks
before the study began that the partnered hospital recruited over 50 hand hygiene observers. The
exclusion of the intensive and critical care units is a limitation that could be the reason for the
lack of CAUTI and CLABSI data. The difficulty of the units to comply with the intervention at
the beginning of the study is another limitation. The lack of intervention compliance during the
first week of the study could have held back the initial rise in hand hygiene compliance.
Another limitation has to do with the intervention time frame. A five-week time frame,
plus the week 6th week of additional feedback added to the original intervention via ‘coaching’,
was insufficient to reach the 70% compliance goal. A short time frame, paired with difficulty of
intervention compliance during the first week did not help in reaching the 70% compliance goal.
Given the observable trend that can be seen before and after the joint intervention, it is possible
that the goal of a 70% compliance rate can be reached given a longer intervention period, and
assuming that the estimated increase in hand hygiene compliance per every two additional weeks
of intervention will continue. While this study did demonstrate a significant increase in hand
hygiene compliance, the question of sustainability deserves further exploration.

22

The lack of control group is also a potential limitation as it may render uncertain whether
a true effect is being seen. However, despite the lack of a control group, it is highly likely that
the observed increase in hand hygiene compliance is due to the intervention for two reasons.
First, there were no other interventions running concurrently until the second week of May 2017.
Second, based on the segmented regression model, the increase in compliance during the
intervention period was statistically significant compared to the five months prior to the
intervention.
Hand hygiene compliance is, and will continue to be, a pressing topic in hospital
environments. The two-level intervention implemented in this study seems to be effective, but
needs to be tested in all units over a longer period of time. This will allow the sustainability of
the intervention to be tested.

23

APPENDICES

24

APPENDIX A

Tables

25

Hand Hygiene Compliance Rates
Month
Compliance
Jun 2016
57%
Jul 2016
57%
Aug 2016
44%
Sep 2016
44%
Oct 2016
45%
Nov 2016
55%
Dec 2016
57%
Jan 2017
54%
Feb 2017
53%
Mar 2017
48%
HCW Distribution
HCW Type
Compliant
Observations
Doctor
201
356
Nurse
469
900
Nurse Asst.
150
368
Other
177
367
Total
997
1991
Table 1: Hand Hygiene Compliance and HCW Distribution for Baseline (June 2016 March 2017). Monthly hand hygiene compliance and distribution of healthcare workers (HCW)
during baseline period (June 2016 – March 2017).

26

Hand Hygiene Compliance Rates
Month
Compliance
Apr 2017
54%
May 1-7, 2017
59%
May 8 - 12, 2017
65%
HCW Distribution
HCW Type
Compliant
Observations
Doctor
260
424
Nurse
722
1200
Nurse Asst.
271
614
Other
722
1200
Total
1975
3438
Table 2: Hand Hygiene and HCW Distribution for Study Intervention (Apr 2017 - May 7,
2017) and Joint Intervention (May 8, 2017 - May 12, 2017). Monthly hand hygiene
compliance and distribution of healthcare workers (HCW) during study intervention period (Apr
2017 – May 7, 2017) and joint intervention period (May 8, 2017 – May 12, 2017).

27

HCW
Type

Standard
Effect
Month
Estimate
Error
DF t Value p-value
Intercept
0.1151
0.1293
11
0.89
0.3924
Month
Jun2016
0.279
0.1918
724
1.46
0.1461
Month
Jul 2016
0.364
0.2047
724
1.78
0.0758
Month
Aug 2016
-0.2229
0.1815
724
-1.23
0.2198
Month
Sep 2016
-0.2871
0.1873
724
-1.53
0.1257
Month
Oct 2016
-0.1257
0.1594
724
-0.79
0.4306
Month
Nov 2016
0.2704
0.1931
724
1.4
0.1619
Month
Dec 216
0.2852
0.2103
724
1.36
0.1756
Month
Jan 2017
0.2234
0.1983
724
1.13
0.2603
Month
Feb 2017
0.1716
0.1729
724
0.99
0.3214
Month
Apr 2017
0.1913
0.09251
724
2.07
0.039
Month
May 2017a
0.4077
0.1176
724
3.47
0.0006
Month
May 2017b
0.6987
0.1358
724
5.14
<0.0001
Month
Mar 2017
0
.
.
.
.
HCW Type
Doctor
0.03257
0.08389
724
0.39
0.6979
HCW Type
Other
-0.2431
0.07248
724
-3.35
0.0008
HCW Type
PCT
-0.5788
0.07756
724
-7.46
<0.0001
HCW Type
Nurse
0
.
.
.
.
Table 3: Hand Hygiene Compliance and Unstructured Model Results. Hand hygiene
compliance and full logistic regression model: hand hygiene compliance = month + HCW type +
Unit (random effects). P-values are significant at a value <0.05. SAS contrast statements were
used to see if there was a temporal (month) effect on the baseline data. HCW=healthcare worker
Month
DF DF F Value p-value
Jun 2016-Jul 2016
1 724
0.11
0.7388
Aug 2016 - Oct 2016 2 724
0.29
0.7458
Nov 2016 - Mar 2017 4 724
0.11
0.9782
Jun 2016 - Mar 2017 9 724
2.24
0.0181
Table 4: SAS Contrasts. SAS Contrasts used to see if there was a temporal (month) effect on
baseline data within full logistic regression model: hand hygiene compliance = month + HCW
type + Unit (random effects). P-values are significant at a value <0.05.

28

Estimate
0.4007

Standard
Error
0.1606

Aug 2016 - Oct 2016

-0.5199

0.1541

Nov 2016 - May 2017b

-0.2302

0.1399

DF
11
61
7
61
7

61
7

Effect
Intercept

HCW
Type

Nov 2016-Mar 2017*Apr
2017- May 2017b
(Linear Time Effect)

0.3181

0.0657

Jun 2016 – Jul 2016

0

.

HCW Type

Doctor

0.06093

0.1072

HCW Type

Other

-0.1483

0.102

HCW Type

PCT

-0.4915

0.1035

.
61
7
61
7
61
7

t
Value
2.5

p-value
0.0298

-3.37

0.0008

-1.65

0.1004

4.84

<.0001

.

.

0.57

0.5699

-1.45

0.1465

-4.75

<0.0001

HCW Type
Nurse
0
.
.
.
.
Table 5: Hand Hygiene Compliance and Reduced Model. Results of segmented regression
model adjusting for healthcare workers and unit with a continuity condition imposed on the
regression parameters to constrain the fitted proportions to be the same on both sides of the
month of March (regarded as a knot). P-values are significant at <0.05. Jun 2016- Jul 2016 is
reference category. HCW= healthcare worker.

29

Publication

Feedback

Frequency of Feedback

Huis et al. 2013
bar charts of hand hygiene rates sent to ward manager
van den Hoogen et al. 2011*
feedback after missed hand hygiene opportunity
Martin-Madrazo et al. 2012
compliance rate for each health care worker
Henderson et al. 2012
compliance rate by job type
after missed hand hygiene opportunity. Bottle of
Linam et al. 2011*
alcohol based hand rub carried with them.
Health care workers had access to view and review
Marra et al. 2010
observations recorded
was not included in orginal design. Added due to
Ho et al. 2012
dropping compliance rate
Pittet et al. 2000

Twice over 6 months
Immediate
after 6 month study period
quarterly
immediate
Unknown how many times feedback was
reviewed.
Unknown

Helder et al. 2010

March and September every year for 4 years
2 months after each of the 7 biannual surveys
performance feedback in hospital newsletter
for 4 years
HH dashboard all HCWs could access; reports sent HCWs had unlimited access to dashboard
to institution leaders and mid-level managers. Reward (frequency of use not measured); reports sent
to managers on biweekly basis
for highest unit compliance
On completeness of hand washing technique not
compliance rate. Used flourescent hand rub solution
N/A

Bedat et al. 2010
Bischoff et al. 2000

HH compliance rates given during education sessions
HH compliance feedback with education sessions

Brown et al. 2003

feedback not on HH compliance but on noscomial
colonization in the unit.

Hugonnet et al. 2002

Aboumatar et al. 2012*

Grayson et al. 2011
Trick et al. 2007

van de Mortel et al. 2000
Mathai et al. 2011*

Mertz et al. 2010*
Monistrol et al. 2012
Pessoa-Silva et al. 2007
Picheansathian et al. 2008
Rosenthal et al. 2003

Rosenthal et al. 2005
Scheithauer et al. 2012*
Sharek et al. 2002
Swoboda et al. 2007*
Allegranzi et al. 2010
Tromp et al. 2012
Harbarth et al. 2002
Zerr et al. 2005

Armellino et al. 2012*

performance feedback in hospital newsletter

created audit system
performance detailed in handout/discussed with
HCWs

every two weeks for a month.
6 times over the span of 6 months
reports sent out biweekly. HH compliance
was shown once and then was decided
against.
did not specify if HCWs had access or how
often compliance rates were reported.

Annual education meeting for 3 years.
2 weeks after baseline prevalence gathered,
6 months and 12 months after start of study
bar charts displaying performance placed above sinks.
as well.
feedback after missed hand hygiene opportunity
immediate
HH compliance rates shown on whiteboard both
graphically and numerically. Meetings with HCWs
held biweekly for unit-specific feedback
every two weeks for a year?
performance feedback mailed to ward nurse
managers
after each of the 2 observations periods
performance feed back during follow up period after
interventions
months 28, 32 and 36 of 36 month study
perfmance feedback
frequency not specifically reported
bar charts of HH rates were displayed at meetings
Monthly from July 1998 to July 2002
feedback on Hand hygiene compliance rates through
graphic presentations at meetings and posted in ICUs
monthly posting
feedback on HH compliance after noncompliant
immediate
compliance data displayed
frequency not specified
voice prompt reminders if noncompliant
immediate
HH performance feedback
frequency not specified
Performance feedback
at baseline and at post intervention
graphs of HH compliance rates
8 week intervals
HCWs received educational material including data on
hospital-wide HH compliance
frequency not specified
feedback displayed on electronic boards mounted in
hallways and sumary reports delivered to unit
immediate and continuous feedback (door
supervisors via email
chimed everytime HCW entered/exited)

Table 6: List of Studies that Include Feedback of HH Performance and Frequency of
Feedback. List of Hand Hygiene studies from Schweizer et al. 2014 and Kingston et al 2016 that
included feedback of hand hygiene performance. Studies with an ‘*’ have an immediate or
continuous feedback frequency with HCWs being directly observed
30

APPENDIX B

Figures

31

Hand Hygiene Compliance Baseline
(June 2016 - March 2017)
100%
90%

Compliance Percentage

80%
70%
60%

57%

55%

57%

57%

50%
40%

44%

54%

53%
48%

45%

44%

30%
20%
10%
0%
Jun-16

Jul-16

Aug-16

Sep-16

Oct-16

Nov-16

Dec-16

Jan-17

Feb-17

Mar-17

Month

Figure 1: Hand Hygiene Compliance Baseline (June 2016 - March 2017). Hand hygiene
compliance rates of 12 in-patient units during baseline period (June 2016 - March 2017)

32

Hand Hygiene Compliance
Baseline + Intervention (June 2016 - May 2017)
100%
90%

Compliance percentage

80%
70%
60%

65%

57%

55%

57%

59%

53%

50%
40%

57%

44%
44%

54%
45%

48%

54%

30%
20%
10%
0%

Month

Figure 2: Hand Hygiene Compliance Baseline + Intervention (June 2016 - May 2017). Hand
hygiene compliance for baseline and intervention time periods for full model. Blue diamonds =
baseline period; red squares = two-level intervention; green triangle = joint intervention with
partnered hospital. May 17a = May 1 – 7, 2017; May 17b = May 8- May 12, 2017.

33

Figure 3: Central Line Insertion Site. Illustration of central line insertion site. Source:
Pictures of Central Venous Catheters. Digital image. The Joint Commission. The Joint
Commission. Web.
<https://www.jointcommission.org/assets/1/6/CLABSI_Toolkit_Tool_13_Pictures_of_C
entral_Venous_Catheters_F.pdf>.

34

Figure 4: Indwelling Urinary Catheter System. Illustration of indwelling urinary
catheter system. Source: Indwelling Foley Catheter for Women. Digital image. Metro
Health University of Michigan Health. Metro Health University of Michigan Health,
Web. <https://metrohealth.net/healthwise/indwelling-foley-catheter-for-women/>.

35

Figure 5: Weekly Hand Hygiene Huddle Helper. Example of weekly hand hygiene huddle
helper that was distributed to units every other week.

36

Figure 6: Positive Bi-weekly Huddle Helper. Example of positive biweekly assessment huddle
helper that was distributed to units that had increased hand hygiene compliance after 2 weeks.

37

Figure 7: Negative Bi-weekly Huddle Helper. Example of negative biweekly assessment
huddle helper that was distributed to units that had decreased hand hygiene compliance after 2
weeks.

38

APPENDIX C

Hand Hygiene Observer Test

39

Competency test for observers __________________________________
Name _________________________ Date _________________________
Review the scenarios below and record them on a hand hygiene data collection form just as you would if
you were watching the situation on a nursing unit. Do not read into the situation; what is written is all you
are able to observe.
1. A physician cleans her hands with hand sanitizer and heads toward a patient room. Her cell phone
rings and she answers, looks up something on the computer, and then enters the patient room 10
minutes later.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
2. A nurse enters a patient room and begins talking with a patient as she heads to the sink to wash her
hands. After drying her hands, she takes the patient’s blood pressure.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
3. A nursing assistant walks out of a patient room rubbing his hands together and shaking them dry.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Would not monitor in an emergency situation
4. A physician enters a patient room holding coffee and a chart, bypassing the hand sanitizer dispenser.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
5. A phlebotomist arrives at a patient’s room. The phlebotomist puts on gloves and enters the patient’s
room to draw a blood sample.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine

40

6. A staff member from Food and Nutrition applies hand sanitizer to his hands, enters the patient’s room
to deliver a tray, then applies hand sanitizer again on his way out of the room.
a. Wash on entry and exit
b. No wash on entry or exit
c. No wash on entry but wash on exit
d. Wash on entry but no wash on exit
e. Unable to determine
7. A nurse prepares to enter an isolation room by putting on a gown and gloves. She then enters the
patient room.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
8. An nurse applies hand sanitizer before entering a patient room. She then enters the patient room and
turn off the IV alarm. She realizes that she forgot a pair of gloves she will need to assess the patient’s
wound. She exits the patient room, grabs gloves, and re-enters the room to perform patient care.
a. Wash on entry, wash on entry (re-entry), and wash on exit
b. Wash on entry, no wash on exit, and no wash on entry
c. No wash on entry, no wash on exit, and no wash on entry
d. No wash on entry, no wash on exit, and wash on entry
e. Unable to determine
9. A code blue is called. Three health care workers run into the patient room without washing their hands
to stabilize the patient.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Would not monitor in an emergency situation
10. A physical therapist enters a patient room without washing her hands. She asks the patient several
questions while resting her hands on the bedrail. She exits the room without washing her hands.
a. Wash on entry and exit
b. No wash on entry or exit
c. No wash on entry but wash on exit
d. Wash on entry but no wash on exit
e. Unable to determine
11. A physician and a fellow use the hand sanitizer dispenser at the nurse’s station. They both go directly
into a patient’s room.
a. Wash on entry
b. Wash on exit
c. No wash on entry

41

d. No wash on exit
e. Unable to determine
12. A housekeeper uses the hand sanitizer dispenser, puts on gloves, then enters a patient room to begin
daily cleaning.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
13. A phlebotomist arrives on the unit already wearing gloves. She enters a patient room to draw a blood
sample.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Unable to determine
14. During physician patient rounds, the attending physician applies hand sanitizer when leaving the
patient room then immediately enters the next patient room while continuing to rub hand sanitizer into her
skin.
a. Wash on entry and exit
b. No wash on entry or exit
c. No wash on entry but wash on exit
d. Wash on entry but no wash on exit
e. Unable to determine
15. A nurse runs into a patient room as the patient begins falling from his bed.
a. Wash on entry
b. Wash on exit
c. No wash on entry
d. No wash on exit
e. Would not monitor in an emergency situation
16. A dietary tech picks up a tray, enters a patient room, places the tray on the patient’s table, leaves the
room and picks up another tray. The dietary tech is not wearing gloves per the hospital’s infection control
policy.
a. Wash on entry but no wash on exit
b. Wash on exit
c. No wash on entry or exit
d. Would not monitor because dietary are not considered health care workers needing to wash
e. Would not monitor in an emergency situation
Courtesy of Cedars-Sinai Health System and The Johns Hopkins Hospital and Health System

42

APPENDIX D

IRB Approval Letter

43

44

REFERENCES

45

REFERENCES

1)

Aboumatar, Hanan, et al. "Infection prevention promotion program based on the
PRECEDE model: improving hand hygiene behaviors among healthcare”

2)

Allegranzi, Benedetta, et al. "Successful implementation of the World Health
Organization hand hygiene improvement strategy in a referral hospital in Mali,
Africa." Infection Control and Hospital Epidemiology 31.2 (2010): 133-141.

3)

Armellino, Donna, et al. "Using high-technology to enforce low-technology safety
measures: the use of third-party remote video auditing and real-time feedback in
healthcare." Clinical Infectious Diseases (2011): cir773.

4)

Azim, Syed, Craig Juergens, and Mary-Louise McLaws. "An average hand hygiene day
for nurses and physicians: The burden is not equal." American Journal of Infection
Control (2016).

5)

Bédat, B., et al. "Successful hand hygiene improvement strategy in a referral children’s
hospital in Armenia." Journal of Hospital Infection 76.4 (2010): 362-363.

6)

Beekman, E. S., and K. D. Henderson. "Infections caused by percutaneous intravascular
devices." Mandell, Douglas, and Bennett's Principles and Practice of Infectious
Diseases 8 (2014).

7)

Bischoff, Werner E., et al. "Handwashing compliance by health care workers: the impact
of introducing an accessible, alcohol-based hand antiseptic." Archives of Internal
Medicine 160.7 (2000): 1017-1021.

8)

Brown, Samuel M., et al. "Use of an alcohol‐based hand rub and quality improvement
interventions to improve hand hygiene in a Russian neonatal intensive care
unit." Infection Control and Hospital Epidemiology 24.3 (2003): 172-179.

9)

Centers for Disease Control and Prevention "Catheter-associated Urinary Tract
Infections." Centers for Disease Control and Prevention.16 Oct. 2015. Web. 25 June
2017.

10)

Centers for Disease Control and Prevention "Healthcare-associated Infections." Centers
for Disease Control and Prevention, 16 Nov. 2015. Web.

11)

Centers for Disease Control and Prevention. "Bloodstream infection event (central lineassociated bloodstream infection and non-central line-associated bloodstream
infection)." Atlanta, GA: Centers for Disease Control and Prevention 4 (2016): 1-32.

46

12)

Centers for Disease Control and Prevention. "Checklist for prevention of central line
associated blood stream infections." (2014)

13)

Centers for Disease Control and Prevention. "Urinary tract infection (catheter-associated
urinary tract infection [CAUTI] and non-catheter-associated urinary tract infection [UTI])
and other urinary system infection [USI]) events." Centers for Disease Control and
Prevention, Atlanta, GA: http://www. cdc.gov/nhsn/PDFs/pscManual/7pscCAUTIcurrent.
pdf (2015).

14)

Centers for Disease Control and Prevention. "Hand Hygiene in Healthcare
Settings." Centers for Disease Control and Prevention, 24 Mar. 2017. Web.

15)

Gould, Carolyn V., et al. "Guideline for prevention of catheter-associated urinary tract
infections 2009." Infection Control & Hospital Epidemiology 31.4 (2010): 319-326.

16)

Grayson, M. Lindsay, et al. "Outcomes from the first 2 years of the Australian National
Hand Hygiene Initiative." Med J Aust 195.10 (2011): 615-9.

17)

Harbarth, Stephan, et al. "Interventional study to evaluate the impact of an alcohol-based
hand gel in improving hand hygiene compliance." The Pediatric Infectious Disease
Journal 21.6 (2002): 489-495.

18)

Helder, Onno K., et al. "The impact of an education program on hand hygiene
compliance and nosocomial infection incidence in an urban neonatal intensive care unit:
an intervention study with before and after comparison." International Journal of
Nursing Studies 47.10 (2010): 1245-1252.

19)

Henderson, Donna M., et al. "A Collaborative, Systems‐Level Approach to Eliminating
Healthcare‐Associated MRSA, Central‐Line–Associated Bloodstream Infections,
Ventilator‐Associated Pneumonia, and Respiratory Virus Infections." Journal for
Healthcare Quality 34.5 (2012): 39-49.

20)

Ho, Mei-lin, et al. "Effectiveness of multifaceted hand hygiene interventions in long-term
care facilities in Hong Kong: a cluster-randomized controlled trial." Infection Control &
Hospital Epidemiology 33.08 (2012): 761-767.

21)

Hooton, Thomas M. "Nosocomial urinary tract infections." Mandell, Douglas, and
Bennett's Principles and Practice of Infectious Diseases 8 (2014).

22)

Hugonnet, Stéphane, Thomas V. Perneger, and Didier Pittet. "Alcohol-based handrub
improves compliance with hand hygiene in intensive care units." Archives of Internal
Medicine 162.9 (2002): 1037-1043.

23)

Huis, Anita, et al. "Impact of a team and leaders-directed strategy to improve nurses’
adherence to hand hygiene guidelines: a cluster randomized trial." International Journal
of Nursing Studies 50.4 (2013): 464-474.

47

24)

Indwelling Foley Catheter for Women. Digital image. Metro Health University of
Michigan Health. Metro Health University of Michigan Health, Web.
<https://metrohealth.net/healthwise/indwelling-foley-catheter-for-women/>.

25)

Joint Commission. "Sustaining and spreading improvement in hand hygiene
compliance." The Joint Commission Journal on Quality and Patient Safety 41.1 (2015).

26)

Kingston, L., N. H. O'Connell, and C. P. Dunne. "Hand hygiene-related clinical trials
reported since 2010: a systematic review." Journal of Hospital Infection 92.4 (2016):
309-320.

27)

Kovacs‐Litman, Adam, et al. "Do physicians clean their hands? Insights from a covert
observational study." Journal of Hospital Medicine 11.12 (2016): 862-864.

28)

Linam, W. Matthew, et al. "Quality-improvement initiative sustains improvement in
pediatric health care worker hand hygiene." Pediatrics 128.3 (2011): e689-e698

29)

Marra, Alexandre R., et al. "Positive Deviance A New Strategy for Improving Hand
Hygiene Compliance." Infection Control & Hospital Epidemiology 31.01 (2010): 12-20.

30)

Martín-Madrazo, Carmen, et al. "Cluster randomized trial to evaluate the effect of a
multimodal hand hygiene improvement strategy in primary care." Infection Control and
Hospital Epidemiology 33.7 (2012): 681-688.

31)

Mathai, Ashu S., Smitha E. George, and John Abraham. "Efficacy of a multimodal
intervention strategy in improving hand hygiene compliance in a tertiary level intensive
care unit." Indian Journal of Critical Care Medicine 15.1 (2011): 6.

32)

Mertz, Dominik, et al. "Effect of a multifaceted intervention on adherence to hand
hygiene among healthcare workers: a cluster-randomized trial." Infection Control &
Hospital Epidemiology 31.11 (2010): 1170-1176.

33)

Monistrol, O., et al. "Impact of a hand hygiene educational programme on hospital‐
acquired infections in medical wards." Clinical Microbiology and Infection 18.12 (2012):
1212-1218.

34)

National Center for Health Statistics (US. "Health, United States, 2015: with special
feature on racial and ethnic health disparities." (2016).

35)

Pessoa-Silva, Carmem Lucia, et al. "Reduction of health care–associated infection risk in
neonates by successful hand hygiene promotion." Pediatrics 120.2 (2007): e382-e390.

36)

Picheansathian, Wilawan, Alan Pearson, and Prakin Suchaxaya. "The effectiveness of a
promotion programme on hand hygiene compliance and nosocomial infections in a

48

neonatal intensive care unit." International Journal of Nursing Practice 14.4 (2008): 315321.
37)

Pictures of Central Venous Catheters. Digital image. The Joint Commission. The Joint
Commission. Web.
<https://www.jointcommission.org/assets/1/6/CLABSI_Toolkit_Tool_13_Pictures_of_C
entral_Venous_Catheters_F.pdf>.

38)

Pittet, Didier, et al. "Effectiveness of a hospital-wide programme to improve compliance
with hand hygiene." The Lancet 356.9238 (2000): 1307-1312.

39)

Rosenthal, Victor D., Sandra Guzman, and Nasia Safdar. "Reduction in nosocomial
infection with improved hand hygiene in intensive care units of a tertiary care hospital in
Argentina." American Journal of Infection Control 33.7 (2005): 392-397.

40)

Rosenthal, Victor Daniel, et al. "Effect of education and performance feedback on
handwashing: the benefit of administrative support in Argentinean hospitals." American
Journal of Infection Control 31.2 (2003): 85-92.

41)

Scheithauer, Simone, et al. "Improving hand hygiene compliance rates in the
haemodialysis setting: more than just more hand rubs." Nephrology Dialysis
Transplantation 27.2 (2012): 766-770.

42)

Schweizer, Marin L., et al. "Searching for an optimal hand hygiene bundle: a metaanalysis." Clinical Infectious Diseases (2013): 670.

43)

Sharek, Paul J., et al. "Effect of an evidence-based hand washing policy on hand washing
rates and false-positive coagulase negative staphylococcus blood and cerebrospinal fluid
culture rates in a level III NICU." Journal of Perinatology 22.2 (2002).

44)

Swoboda, Sandra M., et al. "Isolation status and voice prompts improve hand
hygiene." American Journal of Infection Control 35.7 (2007): 470-476.

45)

Trick, William E., et al. "Multicenter intervention program to increase adherence to hand
hygiene recommendations and glove use and to reduce the incidence of antimicrobial
resistance." Infection Control & Hospital Epidemiology 28.01 (2007): 42-49.

46)

Tromp, Mirjam, et al. "The short-term and long-term effectiveness of a multidisciplinary
hand hygiene improvement program." American Journal of Infection Control 40.8
(2012): 732-736.

47)

van de Mortel, Thea, et al. "Maximizing handwashing rates in the critical care unit
through yearly performance feedback." Australian Critical Care 13.3 (2000): 91-95.

49

48)

van den Hoogen, Agnes, et al. "Improvement of adherence to hand hygiene practice using
a multimodal intervention program in a neonatal intensive care." Journal of Nursing Care
Quality 26.1 (2011): 22-29.

49)

Zerr, Danielle M., et al. "Decreasing hospital-associated rotavirus infection: a
multidisciplinary hand hygiene campaign in a children's hospital." The Pediatric
Infectious Disease Journal 24.5 (2005): 397-403.

50)

Zimlichman, Eyal, et al. "Health care–associated infections: a meta-analysis of costs and
financial impact on the US health care system." JAMA Internal Medicine 173.22 (2013):
2039-2046.

50