145%; 1 £51 ”1‘3“?“ ' 1,. , 1. ‘.' $31.3: ‘1 '1 3.3-4 :1 3 >1} 4 ‘ :1 J: ,2" “‘7 ‘L'L‘XL ‘33:} 1‘3“. P2131. ql‘g .nh‘ 3’ ‘ ' “"‘u "‘34. 1}." "413%. 14131“ W 1&1 -~‘§- “L."Vmu“ SHW 3 1,31 .usp’ 1: ‘ N “4341551 .m‘w“ 9; 45‘ ”m“. . .5”, . “1 ”I“ ulna k.) 4 c \. "L 153431??? . 1‘ at“: 2.1411,. 53.5554?“ L5 '1 ‘35‘1 ‘H‘fizflfit‘fic 9:4 “513‘. 1“ .41 a” 4 4133".“ “$317" 1‘2151‘13113‘51‘1 ”do“ . L? “1“ “'qitfl‘bg‘ P‘wadmlkt L‘HL 1. ‘4 'N’ I ‘5‘ ' ‘3 ”.121? “L' L ‘4 K 3 ‘1"11‘1.‘ . #1:. .1“. . 5;.- 2‘ 51.03“ ’L 331‘ .5 A». $525 233:; .3" flag“ 2“ ‘14.: N31?“ fifii‘i‘w i“ #12:” .1 1 45.5 ”‘1‘ ("a ”41% ’ 13:36! - f1“-\.‘-{'1““..’1‘R 4 3mm“. 35%:1511“ @‘skstz , , 6.151511%: #11? 112%.! 1312‘ ‘nn. H 1- 1243:: “$131.4: M41411:1 53; M. 11.15% 4553., . 555,455 5 2 ~ 1341“» 1 1,1,, x. 54%? “ 1.3% 3““ :5 W .. .1 31‘.- 1 “ I“ ”’ .. {c31‘ "1-3“ 4. "1M“ Lag}; (Lt ‘ “"w 323,313. RAP" fig: ‘fir " in," ‘9 . ‘ ‘3‘!“ Nfiik‘fi “lg? 1:31: 1;“‘1312' 12‘1“: 3.3%:(25‘12‘133 - ‘31“! _‘ 1 1 3.4115311. 315.. .. gififg “4.1 xxx-:3. ‘L‘Ef‘uw 5:15 ‘1 $5} ~ ii“‘“.}‘ 1. - ‘ 1 1“ ‘3 ‘3‘}? 113% . qi‘ V“$‘*‘““‘ "‘ 3? x’q‘fv‘ ‘Ir‘éfl 5:315 ' 51 :13“ 5'11 1' '2‘ L > M" ‘ " ‘ 4‘5 , - ” ‘L‘S‘Tw‘x‘h. t in: ' ‘44:“..5124 3“ ‘Wk‘i‘uw ‘ M {‘4 'L .. . _ . L , ‘L . , 3&1 W324; 14 - , ‘.~ NV ~4| :' "gm-2 “ % 53:41:1th 81:41 ‘34. ‘— 13..” ‘1, ii , _ _ . .‘ .3. ,5“. L‘ ‘R‘E‘hf if}: d, $130“: .1?“ “3‘5? ‘ 1; . , 1 ‘ . x . "L 4. x ‘L ' .‘1 L -4 \. CE‘ 24. ‘ 4~4 j . $313.1 g5“??-.“§§“" s.» - .1435?“ £135 “‘13:”? Y 5&2 5.3:}: ‘s‘x ‘44"x . w} baa -§iw:#‘l , Wt! _ . , . u . “,2” mm 33“» ‘52“. 35:“ j‘1‘53133‘331‘ ,‘1 , 4333933 mifizmg.“ 1) .g;‘ V a‘.‘ ' 1' V . ‘1 ‘L, .. - 4 .. , , was... .4 543...“. W‘fi“ ' . 53g? 41% 455*: we 5 '- 12h 3; ' ‘41:: Qinjéi'xfn‘v . ' 13%“ #1331: ‘34. ff??? 1.}? “42.41% ““31“ :13” 1‘ Ln‘tE““'~‘-I~ ' ‘ a 4 wwfim‘é‘tkig“; 1y. Wi‘fiam .'- \‘fig‘iw “a“ 3243.1. b41132: Th 7 m 1%. .‘4 "4 ' ~§' “‘5'. M v _ , . . I ‘ I‘- . u k . ‘35?”451517‘5‘525845 51.3.24 R 5753.1 ‘3“ WW3 45% ‘ . 15:15:25.?“ “3%; "3&553“ 41:11:35.... ‘ E z: + 5.1. . . wk; ‘- ~~m% .5, .. 3145;25:- “4&15‘3‘, n. 1.12% ‘ $1 ‘1 N 13.5,? “335111. MESS-L1“.txttn‘1‘}da:.:§‘~ 2&3“: ' ‘lg‘h w.‘ 4.33%; ‘V 'pi“ 42:213. wfi’g ‘1". 1‘13? A. ' ‘3 l "A A- ~ h“ ‘L nfi‘g‘ffitw n . “r_' 4 “$53: . . :P‘Komfi %§k‘}ée hawk-x y. up’L ‘~ 1.52%; “fig: ~ ”m "1413:... 55313. ‘1 "7 .‘LL R1113 ‘5. an. . 4 .fr‘sq ‘1‘m‘zfivm $331: $3 AL "1 :‘L‘ft‘t 13““: 'y ‘32.! “‘ fink“ I f .L I. Kl it 1‘ ‘1‘, , b it. 4:111: 14‘“? 5 fi‘flwafih ‘ “11.. 36-- . ‘43" f.\‘,, ‘ i‘lffim 4“ . 4.1 ‘ . Vljlgawu . $223?“ ‘ . - . .- :"1‘ ‘ " BL , I ""1 ‘ 3 L “ hvfihggg ‘ 41:151.. , , ,1 . 55$}: '4 it 1! 4:5: 1‘ haw-1 P . 1‘." 'r ,- {7’} 4:; 1': “tiff“. 5 1‘1. x _f ""3 “5. 5.. 24%.» ‘*§1‘1\*‘ a :w ‘KWIt 7‘3“" ‘I . 1 .4 #3; ‘ . M “3’ .. 1' ‘ 4 fig“ ~~L .- ‘L ‘9: 1141;53! ‘fig‘ifi‘a‘ {"£“"L§1§%Q¢ fig. “1?ng . 1:31;???” ":31: k ’17:. w%:‘ 5L“\.» \‘SL‘V‘ ‘2 5:} , 455%.» 1301:.“ £31? x131“??? - .L 14%.?“ 4;... E; 1.53,}... RX hwfiwfi mag: 1.1:? : Ev?» 3‘ “W“ 4. .' 1‘ ' ‘ 1 3 4‘1. 351‘ ._ ‘1'". . "‘- 1 3111.39 c 3-5" 31m}; .. .11, -. ' u... ‘21:; ”.wéfiaxgv‘ 35%,; 4 1 .14- $514,} Elf-«1.2“ 1135.4. 1 . ‘13.». \L:“ .thfigg; ’3‘ “3;... 1‘53?) 119%.» fi', ‘ 1.. . 1; 1, .5. *7“ y . . . ,4» . WEE“: -. =1L14:1‘.~.. L :c-3 v. ‘- 72:35 » a» . _ “‘4. "vi t‘&\%3: 5 ; . ,1: . n. :‘z‘ 3: >MAL‘L “,9“ ‘ 3“» 129.34.,- '33?“ ‘tfid 5%fi§.fifi‘gfiw ‘535‘ Q‘Qfifi“ ‘ it? "‘ “$3” - M3131. . ‘r «\f- ‘ . .. f ‘ I“! 1 3: 4171713153“ xi '31. o-L. I . .,- gas : . ». -‘cgf’f ‘ J. ‘ v‘. 4 ‘5? 1‘3 Y.‘ '5. ".1 Jpn; figfigfifiw 4:12;}? 0’.-.“ 7: ., s4.»i‘1\ai»a'; %& 3* “~35? - ‘35 ‘5}33- " ' i; . 1 ’5“. . x-‘W" “961'“‘gm 52m , brig. .... tut; “13313.15: . . ‘k we . 4 t‘" “39‘“ 4:31,: 33““). "h“ 11‘ ‘5 F57“? 7325‘ 11.1 x. - 2"?” $5: W 5 ¢ - , “$14: '1; -1“ 4‘ -.“.1'/’- “E' 1415;119:511 ' $31445. “gay“: “31‘2"“ ,g THESlS ’ ’1 ill? 51A. “Y s Michigan fitate ‘: v '- fl vaersity 53 This is to certify that the dissertation entitled THE RELATIONSHIP BETWEEN RESIDENTIAL PROGRAM CHARACTERISTICS AND PATIENT'S INTEGRATION INTO THE COMMUNITY AND SATISFACTION WITH THEIR LIVING ENVIRONMENT presented by Robert C. Davis has been accepted towards fulfillment of the requirements for Ph.D. degreein Education M/Maz Major professor mm 8 Date “WWW,” , .. F m -,p”- dun ” mill 3 lililililil” MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE RELATIONSHIP BETWEEN RESIDENTIAL PROGRAM CHARACTERISTICS AND PATIENT'S INTEGRATION INTO THE COMMUNITY AND SATISFACTION WITH THEIR LIVING ENVIRONMENT BY Robert C. Davis A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling and Educational PsychOIOgy 1984 © 1985 ROBERT CHRISTIAN DAVIS All Rights Reserved ABSTRACT THE RELATIONSHIP BETWEEN RESIDENTIAL PROGRAM CHARACTERISTICS AND PATIENT'S INTEGRATION INTO THE COMMUNITY AND SATISFACTION WITH THEIR LIVING ENVIRONMENT BY Robert C. Davis The purpose of this study was to determine if resi— dential prOgrams (group homes and supervised apartments) for the chronically mentally ill differixitheir program charac- teristics and social climates and if program characteristics and social climate are related to residents' integration into the community and satisfaction with their living environment. In addition, the relationship between resident characteristics and their integration into the community and satisfaction with their living environment was examined. Finally, residents' overall level of community integration and satisfactiOn with the environment was discussed. A review of the literature identified very few studies which examined program characteristics of residential settings. Seventy-eight residents in eight group homes and ten SuPervised apartments participated in the study. FOrty-three hOuse staff and apartment supervisors also participated. The research design was divided into two sections. The first section, or two formal hypotheses, examined g '.;x" Robert C. Davis whether the two independent variables, "program characteristics" and "SOCial climate" differed between residential settings. Analysis by univariate analysis of variance was statistically significant (p (.05) for both variables. The second section of the design, or the remaining four formal hypotheses, asked whether program characteristics and social climate are re- lated to the dependent variables: residents' integration into the community and satisfaction with the living environ- ment. Four multiple regression analysis equations, matching each of the independent and dependent variables were conducted. The major findings were that three social climate charac- teristics were significantly related to residents' integration into the community. Two program characteristics were also significantly related to residents' integration into the community. Residents' satisfaction with their living environ- ment was significantly related to all ten social climate characteristics, however, no program characteristics were significantly related to residents' satisfaction. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . LIST OF APPENDICES . . . . . . . . . . . . . . . Chapter I. II. IV. STATEMENT OF THE PROBLEM. . . . . . . . . Introduction. . . . . . . . . . . . . . Need for the Study. . . . . . . . . . . . Theory. . . . . . . . . . . . . Purpose of the Study. . . . . . . . . . . Hypothesis. . . . . . . . . . . . . . . Overview. . . . . . . . . . . . . . . . . REVIEW OF THE LITERATURE. . . . . . . . . Research in Both Residents' Characteristics and Program Factors . . . Summary. . . . . . . . . . . . . . . . Research on Program Factors . . . . . . . Summary. . . . . . . . . . . . Research on Residents' Characteristics. . Summary. . . . . . . . . . . . . . . . DESIGN OF THE STUDY . . . . . . . . . . . Selection and Description of the Subjects Description of Residential Settings . . . Measures. . . . . . . . . . . . . Procedures for Collecting the Data. Statistical Hypothesis. . . . . . . . -,- Design. . . . . . . . . . . . . . . . . Analysis. . . . . . . . . . . . . . . . . Summary. . . . . . . . . . . . . . . . ANALYSIS OF RESULTS . . . . . . . . . . . Hypothesis I: Differences Between Residential Settings' Social Environments . . . Hypothesis II: Differences Between Residential Settings' Program Characteristics . . . . Hypothsis III: The Relationship Between Residents' Community Integration and Residential Settings' Social Climate. . . ii o Page iv vi 80 84 85 Chapter Page Hypothesis IV: Relationship Between Residents' Satisfaction and Residential Settings' Social Climate. . . . . . . . . . . . 87 Hypothesis V: Relationship Between Residents' Community Integration and Residential PrOgram Characteristics . . . . . . 89 Hypothesis VI: Relationship Between Residents' Satisfaction and Residential Program Characteristics . . . . . . . . . . 91 The Relationship Between Residents' Characteristics and Their Integration into the COmmunity. . . . . . . . . . . . . 93 The Relationship Between Residents' Characteristics and Their Consumer Response Scores . . . . . . . . . . . . . 93 Overall Results on External Integration and COnsumer Response Scales. . . . . . . . . . 96 Summary. . . . . . . . . . . . . . . . . . . 96 V. SUMMARY . . . . . . . . . . . . . . . . . . . . 101 Results . . . . . . . . . . . . . . . . . . . . 103 Discussion. . . . . . . . . . . . . . . . . . . 105 Limitations . . . . . . . . . . . . . . . . . . 121 Recommendations for Further Research. . . . . . 122 APPENDICES. . . . . . . . . . . . . . . . . . . 124 REFERENCES. . . . . . . . . . . . . . . . . . . 152 iii LIST OF TABLES Table Page 2.1 Summary of Studies Examining Characteristics of Residential Programs Associated with Com- munity Adjustment and/or Tenure . . . . . . . . . 36 2.2 Summary of Studies Finding Residential Programs Superior to Other Placements on Measures of Adjustment or Community Tenure. . . . . . . . . . 37 2.3 Summary of Studies Examining Characteristics of Individuals Served in Residential Facilities and Their Effect on Community Adjustment and/or Tenure. . . . . . . . . . . . . . . . . . . . . 46 3.1 Demographic Data for Population. . . . . z . . . . 51 4.1 Summary Data for Analysis of Variance of Residential Settings' Social Environments (COPES Scores). . . . . . . . . . . . . . . . . . 81 4.2 Summary Data for Analysis of Variance of Residential Settings‘ Program Characteristics (RPCS Scores)_. . . . . . . . . . . . . . . . . . 84 4.3 Equltiple Regression Summary Table for Relation- ship Between External Integration and COPES Scores. . . . . . . . . . . . . . . . . . . . . 86 4.4 Pdultiple Regression Summary Table for the Rela- tionship Between Consumer Response Scores and COPES Scores. . . . . . . . . . . . . . . . . . . 88 4-5 Multiple Regression Summary Table for the Rela- tionship Between External Integration and Residential Program Characteristics . . . . . . . 90 4.6 Multiple Regression Summary Table for the Rela- tionship Between Consumer Response Scores and Residential Program Characteristics . . . . . . . 92 4'7 Multiple Regression Summary Table for the Rela- tionship Between Residents' Characteristics and Integration into the Community. . . . . . . . 94 iv Table Page 4.8 Multiple Regression Summary Table for the Relationship Between Residents' Characteristics and Consumer Response Scores. . . . . . . . . . . 95 4.9 Overall Results on External Integration and Consumer Response Scales. . . . . . . . . . . . . 97 LIST OF APPENDICES Consent for Participation . . . . . . . . PerSOnal Data Sheet . . . . . . . . . . . Residential Program Characteristics Scale Community-Oriented Programs Environmental Scale . . . . . . . . . . . . . . . . . External Integration Scale. . . . . . . . Consumer Response Scale . . . . . . . . . COPES Subscale Definitions. . . . . . . . Community-Oriented PrOgram Environmental Scale Scoring Key . . . . . . . . . . . . Summary Data for COPES Scores by Residential Setting . . . . . . . . . Stunmary Data for RPCS Scores by Residential Setting . . .. . . . . . .. Sununary Data for CR8 and EI Scores by Reijiential Setting . . . . . . . . . . . Sununary Data for Residents' Characteristics by Residential Setting. . . .. . . . . . vi Page 124 125 126 132 138 141 147 148 149 CHAPTER 1 STATEMENT OF THE PROBLEM Introduction The introduction of neuroleptic agents and the devel— opment of community mental health centers have provided the underpinnings for a reform movement in the treatment of hospitalized chronically mentally disabled patients. 11115 reform movement has been directed at deinstitution— alizing patients and offering outpatient care in local conuminities (Williams, Bellis and Wellington, 1980). In the :zeal of social reform, deinstitutionalization efforts have (often confused locus of care and quality of care. Channgrg the location of care does not, in itself, ensure the hinnanizing of mental health care. Changed 10cation must be accompanied by carefully designed programs (Bachrach, 1978). Adequate housing is considered by most proponents of deinstitutionalization as one of, if not, the basic service in a system of c0mmunity support services for formerly hospitalized patients (Carling, 1978). One leading proponent, Paul J. Carling (1978), sug- gests the primary goal of c0mmunity residential housing revolves around the principle of normalization. In this COntext, he says, adequate housing allows expatients to 1 hh__i_.--____ iiiiiiiiE-IIII'I reenter the community and maximize their participation in "normal" community activities through providing them with enough supportive structure. In residential terms, this means a continuum of housing options ranging frOm minimal programmatic and environmental supports to greater and greater levels of these supports, as needed. The sheltered living arrangements in this continuum of housing options offer different levels of social and psychological support to residents. It is this variance in housing programs and individual needs which requires us to look at the question of what characteristis of residen- tial housing programs are correlated with normalization or irytegration into the community. Need for the Study There are several reasons why it is important to study what cfllaracteristics of residential housing programs are cor- relatedi with residents' integration into the community. Prob- ably, the most important reason is the size of the problem. Between ‘1955 and 1975 there was a 65 percent decrease in the census of patients in state mental hospitals, from 559,000 to 193,000. While annual census was decreasing, admissions to state hospitals increased from 178,000 in 1955 to a peak of 390,000 in 1972, and had only declined to 375,000 by 1974. This trend to lower census and higher admissions reflects a trend toward short-term hospitalization and a growing pro- pOrtion of readmissions. These statistics surely reflect 2 L‘ A , i e:>:‘>“fr‘; gnaw-AW the lack of community-based support systems (Bassuk and Gerson, 1978). The last several years have witnessed growing concern over the nation's commitment to deinstitutionalization. Thousands of chronically mentally ill have been discharged to inadequate housing and nonexistent services in the com- munity (Carling, 1978; Lamb, 1979). This continuing failure to bring about a more humane existence for these individuals is a source of guilt, shame and frustration for c0untless communities (Report to Secretary of Health, 1980). The Task Force for the Development of Community Resi— (iential and Rehabilitative Programs of the New York State Department of Mental Hygiene (1976) concluded that "If the honue, or rather a supportive home environment, is so cen— tral. to our well-being and development, then the concept of the tuome as a supportive living environment, whether tran- sitional or permanent, must be a concern of the first mag- nitude in the development of COmmunity programs for the mentally disabled (p. 26)J' I If community residential services are to promote the humane, normalization process of community reentry, resi- dential program characteristics must be examined to deter- mine their impact on community reentry and quality of life. The literature examining residential placement for chronic psychiatric patients abounds with methodological difficulties. Controlled research is almost entirely lacking (R09 and Rausch, 1975), but descriptive reports have much to say in favor of such programs (Test and Stein, 1978). 3 As is true of most studies of public service programs, these studies usually lack randOm assignment or control groups (Suchman, 1967). Another problem is the reliance on dependent measures like rehospitalization and employment (Anthony, et al, 1972), even though they have been termed unreliable by others (Erickson and Paige, 1973). Such mea- sures of outcome do not provide an understanding of the process of returning patients to the COmmunity (Carpenter, 1978). There remains a great need for descriptive research which focuses on the quality of patient life and patient satisfaction with this environment. The present research attempts to address some of these issues. Theory This research is based on several assumptions regarding community treatment of the chronically mentally disabled. These theories and the backgrOund and rationale for community treatment will be discussed in order to provide a concep- tual framework for the present study. Deinstitutionalization has been defined as a process involving two elements: Shunning or avoidance of tradi- tional settings (particularly state hospitals) for the care Of the mentally ill, and the COncurrent expansion of com- munity-based facilities for the care of these individuals (Bachrach, 1978). The philosophy of deinstitutionalization emphasizes 4 the rights of individuals in a strong civil libertarian sense and holds that the primary avenue to change is through modification of the environment. The goal of deinstitu- tionalization is to humanize mental health care and reverse what is seen as the dehumanizing influences which are be- lieved to be part and parcel of the institutional approach to treatment (Bachrach, 1978). I The philosophy and goals of deinstitutionalization proceed from some fundamental assumptions about mental health care. First, it is assumed that c0mmunity mental health is a good thing and that COmmunity-based care is preferable to institutional care for most, if not all, mental patients. Community care is seen as the more ther- apeutic and represents the treatment of choice. A second assumption is that local communities can and are willing to assume responsibilities in the care of the mentally ill. Third, deinstitutionalizatiOn assumes that community-based programs can perform the functions of the mental hospital, equally or better (Bachrach, 1978). Williams, Bellis and Wellington (1980) have described 'the various historical forces contributing to the deinsti- tutionalization movement. They point out that the Great Depression and World War II left hospitals for the mentally ill without financial resources to hire needed staff or build additional facilities made necessary to care for increased admissions. By the end of World War II, state hospitals could not provide even minimal custodial care and no state hospital met the minimal standards of operations 5 of the American Psychiatric Association. Hospitals began turning to neuroleptics and physical procedures such as shOCk treatments and psychosurgery as admissions increased and staffing decreased. The twenty-year period after World War II was, accord- ing to Williams, et al, a period of social unrest, politi— cal liberalism and economic expansion. America attempted to bring its previously excluded minorities into the main- stream. The federal government initiated prOgressive pro- grams and the courts eliminated legal supports to discrimi- nation. There was also, at this time, an increased interest in psychiatry and mental health issues. This interest was, in part, due to the large number of men turned down for military service due to psychiatric impairment. Psychiatry also developed quick and effective methods of treating war Oneuroses and returning its victims to active duty. In addi- tion, Williams, et al, say that psychoanalytically-based psychiatry seemed to offer some people some understanding Of the causes and atrOCities of World War II. These factors, and others, led to the federal govern— Inent taking a major role in the provision of mental health Services. Congress created the National Institute of Mental Health and the Community Mental Health Center Act Of 1963. Community Mental Health Centers were seen as the replacements of the state hospitals. Their focus was on providing mental health services to previously under- served, poor urban and rural areas. Unfortunately, the 6 Community Mental Health Centers were not required to de- velop programs for the chronically mentally ill. Another problem was that the federal government by-passed state governments to work with local communities. (Williams, et al, point out that the state governments were often by- passed as the federal government created and funded serviCe programs because they were often seen as conservative and racist.) As state hospitals began to discharge chronically disabled patients back to the community, state hospital funds did not accompany the patients in most states. State funds were saved as the federal programs bore the cost of community treatment. In addition, federal judges affirmed the rights of hospital patients to adequate care and treat- ment, resulting in some upgrading of hospital programs and increased discharges back to the community. Other factors leading state hospitals to discharge the lchronically disabled back to the community included the introduction of neuroleptics which were able to control bizarre symptoms and many clinicians' beliefs that it would aid in recovery and minimize the effects of institutionalization. Test and Stein cite Barton (1966), Goffman (1961), Gruenberg (1967) and Hansell and Benson (1971), as having described the "institutional syndrome,‘ which may be charac- terized as apathy, lack of initiative, loss of interest, apparent inability to plan for the future, and lack of individuality. These characteristics are believed to de- velop through an interaction of the premorbid inadequate life-style, the disease process, and the institutional 7 environment which limits contact with the Outside world and assumes management of an individual's life to such an extent that an individual's own perSOnality functions atrophy. The work of others (Ludwig and Farrelly, 1966; Ludwig and Farrelly, 1967; and Towbin, 1969) has led to the theory that patients become active participants in this process and that they develop techniques to maintain their dependent, non-responsible patient status. Empirical evidence for the existence of an insti- tutional syndrome comes from several sources, according to Test and Stein. Wing (1962) surveyed the attitudes and behavior of male schizophrenics with over two years stay in the hospital and found that the longer the stay, the more unfavorable their attitude toward discharge. Honigfeld and Gillis (1967) found that time in the hospital is linearly related to the development of a "social break- down syndrome." Paul (1969), in a review of research On the chronic patient, cites studies showing that the longer the patient remains in the hospital, the less the chances Of his/her discharge. Evidence that any psychiatric hospitalization, regard— less of length of stay, may have negative effects is also discussed in Test and Stein. Langsley and Kaplan (1968), Surveyed studies demonstrating the adverse effects of hospitalization and the label "mentally ill" in the atti- tudes and expectations of the patient him/herself, the patient's family and COmmunity members to him/her. In addition, Mendel and Rapport (1969) found that a history 8 of previous psychiatric hospitalizations heavily influenced the decision to rehospitalize, independent of severity of the individual's current illness. The attempts to make hospitalization more humane and effective, reviewed by Test and Stein, show that while in-hospital adjustment improves, it is unrelated to release rates or post-hospital adjustment. Wing and Brown (1961) surveyed three mental hospitals in Britain differing in the degree of social and humane treatment. Measures of patient's symptomatology demonstrated that there was less critical disturbance in the hospitals with more advanced social treatment. Linn (1970) followed this line of re- search, however, and found variables related to humane treatment (e.g., hospital atmosphere, good facilities, humanistic policies toward patients) unrelated to treat- ment outcome (defined as rapid release). Efforts to improve in-hospital treatment have led to programs ranging from highly psychodynamic approaches to programs emphasizing teaching of coping skills for community living. After surveying the literature in this area, Paul (1969)_concluded that intensive treatment programs fre- quently improve within-hospital adjustment and they increase and speed up release rates; however, they are only slightly related to post-hospital adjustment and are unrelated to length of community stay. One study (Fairweather, 1964) described by Test and Stein for example, c0mbined milieu and learning theory approaches in an experimental program aimed at resocialization and instrumental role performance. 9 "Zia.” M Problem-solving patient grOups were formed and a step- system with responsibilities, passes and funds contingent upon appropriate behavior was initiated. A traditional ward program served as a control. The results showed sig- nificant differences in favor of the within-hospital per- formance on the experimental prOgram, and that the experi- mental program led to earlier release rates. However, a six month post-release follow up showed rehospitalization rates demonstrated no difference between the experimental and control group. There have also been attempts made to shorten hospi- tal stays in the hope of eliminating the negative effects of institutions while moving patients away from the hospi- tal and into the community. Among those reported by Test and Stein is a study by Caffey, et al, (1968). Newly admitted male schizophrenics in a Veterans Administration Hospital were randomly assigned to one of three treatment conditions. Condition A consisted of "normal hospital care" with the usual aftercare; condition B consisted of larief intensive care with special aftercare; and condition C3 consisted of normal hospital care with special aftercare. rl‘he mean times in the hospital were 80, 29 and 86 days, respectively. The study fOund no statistically signifi- cant difference between readmission rates or time out of the hospital before readmission for the three conditions. Another study of brief hospitalization by Herz, et al, (1977)_c0mpared three conditions including brief hos- pitalization (average of 11 days) with transitional day 10 care; brief hospitalizatiOn without transitional day care; and standard hospitalization (average of 60 days); all patients were offered outpatient aftercare. Two-year follow up on these 175 patients found no significant dif— ferences between readmission rates for the three groups. Herz did find that differences between groups in levels of psychopathology and inadequacy of role functioning favored the brief hospitalization group and that the use of day care reduced the number of inpatient days. Gove and Lubach (1969) used an experimental group which received three days of intensive treatment to alleviate severe anxiety and disorganization, followed by several weeks in a "readjustment area" intended to prepare for com— munity reentry. Patients who had been admitted from the same geographic area the previous year served as controls. Results indicated no significant difference in readmission rates, but when readmission did occur, the length of treat— ment for the experimental group was significantly shorter. However, differences could be due to other variables dif- fering from one year to the next. Mendel (1968) and Rhine and Mayerson (1971) report favorable results in their studies on short-term hospitali— zation, but neither program utilized a control group. In general, Test and Stein conclude that short-term hOSpitalization studies report readmission rates that are no higher than those for longer term hospitalizatiOn. Short-term hospitalization research, therefore, suggests that long-term hospitalization is not necessary for the 11 treatment of the severely disturbed. Unfortunately, re— admission rates for short-term hospitalization remain high and there still remains a “revolving door" problem. In recent years, many attempts have been made to de- velop programs in the community which w0uld both solve the revolving door problem with readmissions and offer a humane alternative to hospitalization. As stated earlier, these programs have centered around the principal of normalization by maximizing participation in COmmunity activities and pro— viding supportive structures. Segal (1976) points out that emphasis on community care has created increased variance in the types of shel- tered care programs available. These programs range from halfway houses, foster care homes, lodges, apartments and large dormitories to emergency respite centers. Segal says, "we come to the question of determining the characteristics which make a difference with respect to the type of shel- tered care an individual may best profit from." When speak- ing of profit, Segal is referring to "the extent to which “the level of social involvement or social integration of the individual is enhanced by his or her SOCial environ- ment." In summary, this section on theory has covered the background and rationale for community treatment. Included Were discussions of the philosophy, gOals and assumptions of deinstitutionalization. Also discussed were the histori- cal forces contributing to this movement. The clinical rationale and empirical evidence for deinstitutionalization 12 has been reviewed, along with attempts to correct the pro- blems of institutionalization. Finally, current problems in community treatment and deinstitutionalization were introduced. Purpose of the Study The primary purpose of this study is to determine what characteristics of residential hOusing programs for the chronically mentally ill are correlated with residents' in- creased integration into the community and satisfactiOn with their environment. In order to accomplish this pur- pose, residential programs are assessed in relationship to modifiable program rules and structure, i.e., presence or absence of a curfew. Next, the social environment of resi- dential programs are assessed by measuring the "climate" or "personality" of the residential programs. Finally, indi— vidual characteristics of residents are measured in order to describe the population served. Hypotheses Six main hypotheses are investigated in this study. They are stated generally in this section and in statistical form in Chapter Three. The main hypotheses to be investi- gated are: (1). The residential settings studied will differ in their social environments. 13 (2) The residential settings studied will differ in their structure and program characteristics. (3)_ Residential settings' social environments are im- portant factors in resident's integration into the community. (4)’ Residential settings' social environments are im- portant factors in the resident's satisfaction with their residential placement. (5) Residential settings' structure and program charac- teristics are important factors in resident's inte— gration into the community. (6). Residential settings' structure and program charac- teristics are important factors in resident's satis- faction with their residential placement. Overview In Chapter II the literature on residential placement of the mentally ill will be reviewed. In Chapter III the design and analysis of the study will be presented, in- cluding a description of the subjects and methodology and a description of the analysis used. Chapter IV will pre- sent the results of the hypothesis testing, as well as the results of the supplementary analysis. The study will be concluded in Chapter V with the summary and conclusions, along with the implications for future research. ;.. 1 i.e.,, J... CHAPTER II REVIEW OF THE LITERATURE Research conducted on psychiatric patients has focused on many factors involved in the process of rehabilitation. For the purpose of this study, only research related to the area of COmmunity residential placement of the chronically mentally disabled will be reported. This literature review will be divided into three broad areas. The first section will include studies that have examined the characteristics of both the individuals served in residential placement and the prOgrams which serve them. The second sectiOn will deal with studies which examined just program characteristics of residential place- ment. The third section will examine those studies which examined just the characteristics of the individuals served in residential placement. A few studies examining both the characteristics of the individuals served in residential placement as well as the programs which serve them will be described and dis- cussed in sections two and three. The remaining studies examining both individuals served and programs serving them are discussed in their entirety in section one. Few studies employed adequate control groups, if they were used at all. In addition, the Outcome measures often 15 appeared to be selected out of the need to support a par— ticular program rather than to test hypotheses. Research in Both Residents' Characteristics and Program Factors This section will focus on studies that have examined the characteristics of both the individuals served in resi- dential placement and the programs which serve them. While other studies addressing both patient and program charac- teristics have been divided between sections two and three, the studies in this section are presented here in an attempt to preserve their unity. Segal and Aviram (1978) studied the influence of indi- vidual characteristics and the social environment on the internal and external integration of formerly hospitalized mental patients living in community-based sheltered-care facilities, such as halfway houses, family-care houses, and board-and-care homes. Internal integration (II) was defined as social integration within the hOme and the ex- tent to which access to community life was actively sup- ported by the home. External integratiOn (EI)_was defined as the extent of access to and participation in community life whether supported by the home or not. To obtain the sample, the state of California was divided into three master strata: Los Angeles County, the Bay Area, and all other counties in the state. In the Los Angeles and Bay area strata, a two-stage cluster sample 16 was designed with sheltered-care facilities as the pri- mary sampling unit and individuals within facilities as the second stage. In the third stratum, comprising "all other counties," a three-stage cluster sample was designed using counties as primary selection units, facilities as ‘ the second stage, and individuals as the third stage. There were 499 resident interviews attempted with a loss (due to refusal and inaccessibility) of 12%; 10% of the 234 operators refused to participate. Scales were developed to measure external and internal social integration. The 650 possible variables used to predict SOCial integration were reduced to 26 significant predictors (p (.10). These 26 predictors were modifiable by policy action or represented individual characteristics that had to be controlled in the analysis to interpret effects on the more modifiable variables on the criterion measures. The selected predictors also added a reasonable amount of additional, independently explained variance to the model. Segal and Aviram found that community characteristics, resident characteristics, and sheltered-care facility char- acteristics, in that order, influenced E1. The most im- portant community characteristic across all ages and types of psychopathology was the response of neighbors to the residents. Positive responses, such as inviting residents into their homes (on an individual basis, not as a group), and having more than casual c0nversations with them, led to higher EI. Another important community characteristic 17 l x was that the closer the home was to community resources and services, the higher the El scores of residents. Among resident characteristics, the strongest associate of EI was sufficient spending money for residents to do things they wanted. Being an involuntary resident, that is not choosing the place they lived, was also associated with lower levels of El. This was true for those who wanted to leave the facility and those who were indifferent to leaving. The most important facility characteristic of El was its psychi- atric environment, as measured by the Community Oriented Program Environmental Scale (Moos, 1974b). Homes with pro- grams emphasizing resident involvement, support from staff and other residents, open expression of feelings, and a structured program with clear expectations for residents showed higher EI. Homes in which residents, as a group, were isolated from their families and neighbors showed de- creased EI. II was influenced most by COmmunity characteristics followed by facility characteristics and resident charac- teristics, in that order. As with ET, the most important community characteristic effecting II was positive response from neighbors. In addition, rural locations were asso- ciated with higher II than urban ones. Again, as for BI, a good psychiatric environment in the home was associated with higher II and a bad psychiatric environment with lower II. A positive attitude toward use of COmmunity psy- chological services was associated with higher II and homes with female operators tended to promote higher II in 18 .. NLQ-fl residents. For resident characteristics, level of psycho- logical distress was negatively correlated with II and sufficient spending money was positively correlated to II. Segal and Aviram (1978) conclude that the people they studied have never been integrated into the mainstream of society. They suggest that outcomes of community care should be assessed by comparing the number of persons main— taining an adequate level of functioning who, given past experience, might deteriorate to less adequate social func— tioning if confined to an institution. Ware (1979) studied the relationship between demographic and treatment variables in community-based rehabilitation programs. Subjects were 167 psychiatric patients residing in a 73-bed, board and care home over a 24 month period. Residents were contacted six mOnths following discharge from the home and divided into two groups. The first grOup, or success group, included those former residents who had lived in the community independent of any inpatient psychiatric facility for six months following discharge frOm the home. The second group, or failure group, consisted of these for- mer residents that were admitted to an inpatient facility within the six-month period following their discharge from the home. The two groups were compared for age, sex, race, diagnosis, educational level completed, prior work history, length and number of times hospitalized, and number of years since first hOSpitalization. The groups were also compared according to employer of their psychiatrist (private or public), and current enrollment in a job, school, or outside rehabilitation program. 19 Analysis of the data through stepwise regression pro— cedure indicated three variables that were significant (p (.001) in predicting treatment outcome. Number of times hospitalized during treatment, participation in work therapy and number of years spent in the mental.health system were significant in predicting ability to remain independent from inpatient care. Additionally, age at discharge was found to have a curvilinear relationship with outcome. Ware concluded that the efficacy of psychiatric treat- ment methods, as they relate to success in the community, is questionable. Smith and Smith (1979) studied 130 mental patients discharged from two divisions of a large state hospital in Michigan. Patients diagnosed as mentally retarded or suffering from organic brain damage, and those whose primary diagnOSis was related to substance abuse were excluded from the study. Immediately following discharge, each patient was assessed on seven different scales. Scales used assessed (1) plans and abilities; (2) desire to leave; (3) institutionalization; (4)chrrent hospital experience; (5) family and living situation; (6) after- care; and (7) employment and mobility. The seven dimen- sions were analyzed for independence from one another and internal reliability was assessed by calculating their alpha scores. The authors attempted to relate scores on the seven dimension to recidivism and COmmunity adjustment of the subjects. Recidivism was defined as a return to 20 the hospital at least once within a nine-month period after discharge. COmmunity adjustment was measured by a follow-up scale covering patient's adjustment in the areas of employment, family relationships, interpersonal relationships, SOCial and recreational activities, and overall involvement with the community. Follow-up data was collected at three months post-discharge. Any sub- jects rehospitalized before three months time completed follow-up questiOnnaires and were assessed immediately after readmission. For purposes of analysis, the patients were divided into two groups: those with scores on COmmunity adjust- ment above the mean, and those with scores below the mean. The patients were also divided into groups con- taining those with scores above the mean and those with scores below the mean for each of the seven diminsions. The dimension scores were compared in a series of 2x2 Tables with the adjustment scores and the incidence of recidivism and nonrecidivism. Chi-square tests on the 2x2 Tables indicated that scores on four of the dimensions were significantly related to the incidence of recidivism or to the com- munity adjustment scores or to both. Those dimensions were (1) Family and Living Situation; (2) Aftercare; (3)_Plans and Abilities; and (4) Desire to Leave. Scores on Family and Living Situation discriminated significantly between high and low community adjustment (p (.01) and between recidivism and nonrecidiviSm (p 4.01). 21 The authors also noted that although a large number of patients (29.9 percent) adjusted poorly in supportive family situations, few of them returned to the hospital. Scores on Aftercare were related to both adjustment (p <.01) and recidivism (P (.05) with those patients re- ceiving aftercare scoring higher on community adjustment and lower on recidivism. Scores on Plans and Abilities were significantly related to only recidivism (p (.05). The results showed that while patients who were considered capable of making plans and living in the community were likely to stay out of the hospital, many of those who were not regarded in this way were still able to remain out- side the hospital. The Desire to Leave dimension was sig- nificantly related to the patient's adjustment in the com- munity (p (.01), but not to their eventual recidivism. The authors concluded that wanting to stay in the COm- munity was necessary for success in the community, but that simply wanting to go back to the hospital is not al- ways sufficient to guarantee one's return in these days of tighter hospital admission policies. No significant differences were observed in the ad- justment and recidivism scores for the dimensiOn of Insti- tutionalization, Current Hospital Experience, and Employ- ment and Mobility. The authors concluded that measures of chronicity and lack of social COmpetence, both of which were thought to characterize patients who had spent long periods of time inside mental hospitals, did not appear to 'be related to outcomes in the COmmunity. 22 ‘- —u. _ ——r w_ _L...~ 2'? Finally, Smith and Smith say the results of their study demonstrate that a patient's staying out of the hos- pital does not necessarily imply that he or she has made a successful return to community life. Smith and Smith suggest that more humane and client-centered outcome measures of community adjustment should be used with mea- sures such as recidivism. The last study to be discussed in this section ex- amining both the characteristics of the individuals served in residential placement as well as the prOgrams which serve them was by Hull and Thompson (1981). The authors studied 157 community residential facili- ties for the mentally ill in Manitoba, Canada,examining: (1) Individual characteristics of residents; (2) Social structural characteristics of the residences; (3) Staff attitudes; and (4) Community characteristics. These variables were studied for their influence on "normalization." Normalization was defined as "the utili- zation of means which are culturally normative as possible, in order to establish and/or maintain personal behaviors and characteristics which are culturally normative as possible." The level of normalization was assessed by use of a 30 rating environmental normalization scale designed to correspond to the Program Analysis of Ser- vices System (PASS) Field Manual (Wolfensburger and Glenn, 1975). There were 296 residents studied with a median age 23 of 57 and a median number of 2.25 psychiatric hospi- talizations. Residents had spent a median of 7.5 years in the hospital; 49.2 percent were male and 50.7 percent were female, 75% were diagnosed schizophrenic. A step wise regression procedure was undertaken with the following results: The largest proportiOn of the variance, 41 percent, was accounted for by four characteristics of the home itself plus one community measure. The more residents in a home, other things being equal, the lower the normalization score. Similarly, the greater the number of disability groups in the home and the higher the proportion of males, the lower the level of normalization achieved by that home. If the home was an independent living facility, it was likely to achieve a higher normalization score than if it was a Board and Care Home, even when other variables were con- trolled. The higher the average family income in the com- munity, the higher the average normalization score of the residences in the community. Three individual characteristics accounted for 30% of the variance. The longer the time the residents had spent in institutions, the lower the envirOnmental nor- malization score achieved by a resident. Two measures of Social Competence were positively related to environ- mental normalization scores. The first reflected skills in the utilization of community services and resources and the second measured interpersonal skills. A number of variables directly measuring behavior 24 problems or which could act as a measure for pathOIOgy were entered into the equation. Of these, however, only length of institutionalization was related to the environ- mental normalization score. The authors suggest the development of residential facilities which house people at a number of points on the distribution of these characteristics. Overall, they suggest that smaller residences which pro- vide more opportunities for independence, which serve only one disability group, e.g., the mentally ill, and which are located in a middle income community are more normalizing than thOSe with the opposite characteristics. Summary The literature reviewed in this section has examined both patient's characteristics and program factors effecting patient outcome. The literature reviewed has raised ques- tions regarding the use of outCOme measures such as rehos- pitalization or community tenure (Smith and Smith, 1979). These commonly used measures ignore such goals of deinsti- tutionalization as humane treatment, patient satisfaction with their environment and integration into the mainstream of society. Atkinson (1975)_has pointed out the weakness of such outcome measures as rehOSpitalization because return to, the h05pital may reflect a lack of alternative placements rather than actual need for rehospitalization. Some 25 studies examined here (Segal and Aviram, 1978) (Hull and Thompson, 1981) have attempted to evaluate programs with more meaningful outcome measures. In the present study, the original goals of deinsti— tutionalization, client satisfaction and integration into 'the community were used as the outcome criteria for success. Prtxgram factors relating to these outcome criteria were examined . 26 Research on PrOgram Factors This section will focus on how various program and treatment characteristics have been associated with patient outcomes in residential programs. Braun, Kochansky, Shapiro, Greenberg, Gudeman, JohnSOn and Shore (1981) have reviewed controlled studies of deinstitutionalization. They conducted an extensive search of published works for studies satisfying generally accepted scientific standards including: (1) randOm as- signment to experimental and control programs; (2) patients well characterized before randomization; (3)_outcomes mea- sured with validated instruments and criteria; (4)_follow- up covering a high proportion of the subjects for a reason- able period of time; and (5) large enough number of patients and observations for statistical analysis. Their purpose was to evaluate deinstitutionalization for the effect of specific programs on patients. Three types of studies were reviewed. Those studies dealing with residential placement will be discussed here. Braun and associates reviewed a study by Weinman, et al, (1978) studying outcomes for chronically mentally ill Patients released from Philadelphia State Hospital. The study randomly assigned 516 patients, 90% of whom were diagnosed as schizophrenic and who could not be placed with relatives, to community placement with the support of com- munity members called "enables." In one experimental group, PrOfessional staff assisted the enables, giving 27 L them advice and support; in the other, they gave most of their attention directly to the patients. One control group randOmly selected from the same population as the experimental groups received socio-environmental treat- ment. A second, nonrandomized control group of 109 patients was discharged from the hospital through tradi- tional ward programs. Findings indicated that community treatment generated fewer readmissions over a 24 month post-treatment period than SOCio-environmental treatment. However, this difference was not statistically significant. The patients released from the traditional wards had the highest return rate. This difference was significant from both the community and socio—environmental treatment pro- grams. Community treatment patients showed a significantly greater improvement in self-esteem during treatment than both control groups. There was no difference in readmission rates or self-esteem for the two experimental COnditions. However, the patients in the enabler-centered condition mani- fest significantly less psychiatric disability at the com— Pletion of treatment than their counterparts in the patient- centered condition. Weinman, et al, conclude "perhaps the most important contribution of the project was the role Created for indigenous community members as social change agents" p. 154. "Employing indigenous community members fOr a service role with patients is demonstrated by the comparable and even somewhat more favorable impact of enables over professional staff on treatment outcome“ p. 148. I Linn, Caffey and Klett (1977) studied 572 chronically 28 mentally ill patients from five different Veterans Ad- ministration Hospitals. The patients were randomly as- signed to foster care homes outside the hospital (experi- mental group) or to continued hOSpital care (control group). A large number of the foster care patients were excluded after random assignment because their condition had deteri- orated. This made for considerable differences between the experimental and COntrol groups. Patients in the experi- mental group had fewer hospitalizations and a lower preva- lence of alcoholism and chronic brain syndrome. They also differed significantly from the control patients on a mea- sure of potential for community living. Four months after placement, the experimental subjects showed significant improvement over controls in social functioning and overall adjustment. The authors conclude that foster care placement was associated with this improve- ment, but Braun, et al, suggest the deficiencies in the study may have distorted the results. Braun and associates also reviewed studies at Soteria House by Mosher, et al, (1975, 1978). In this study, the Experimental patients lived in a small homelike facility Staffed by paraprofessionals. The COntrol patients were admitted to the inpatient service of a community mental health center. The patients were unmarried, mostly young ““3 experiencing their first episode of schizophrenia. All corl'trol subjects received neuroleptic medication in an actlive-treatment facility with a staff-patient ratio of 1'5 to 1. Controls also received aftercare upon discharge. 29 ,, ' , » ,, * tax; ‘ L.».__.‘_v. The experimental subjects received significantly less neuroleptic medication before discharge (17% versus 100%) and after discharge, as well as less outpatient care after leaving the house. In a two-year follow-up, experimental subjects showed less global symptomatology and better com- munity psychosocial adjustment than centrols. The experi- mental group also showed higher occupational levels and more frequently lived independently. Finally, the experi- mental subjects tended to have fewer readmissions than the control subjects. A fourth study reviewed by Braun, et al, (Polak and Kirby, 1976) reduced the need for psychiatric beds in southwest Denver to 1/100,000 population. The experi- menters assigned clients randomly to a home (experimental group or to the hospital (control group). The same clini- cal team provided treatment to both groups and no criteria was set for exclusion of clients. The experimental group received treatment in small, diversified, community-based social environments. These environments included private homes housing no more than two clients. Each home was backed up by psychiatric nurses and a psychiatrist on 24— hOur call. An observation apartment staffed by a psychology Student and his wife provided 24-hour supervision for clients reQUiring more intensive care. A variety of socialization activities, such as craft groups, were also provided. Firlally, psychiatric hospital beds were used as a back-up to all of these environments. Major emphasis was placed on home visits, immediate crisis service, social systems 30 ""' ' 7 " « k MW, i7 intervention, and rapid tranquilization. Outcome measures completed by clients and community informants at a four month follow-up indicated that community placement was more effective than psychiatric hospitalization (p (.05). on four measures. Of the first 48 clients assigned to the original home group, ten could not be treated in the home because they became overly violent or suicidal. While the authors claim that the similarity between home and hospital groups actually increased after these ten clients were re- moved from the experimental group, these "broken design" clients raise questions as to the study's c0nclusions. The last study (Linn, Caffey and Klett, 1980) re- viewed by Braun, et al, to be discussed here, studied the characteristics of foster homes producing different Out- comes. Patients whose current hospitalization averaged 45 months were randomly assigned to foster care hOmes or con- tinued hospitalization. Results for foster care placed patients indicated that (a) the greater the thal number of people in the home, the more likely the patient was to deteriorate (p <.01); (b) the greater the number of patients in the home, the more likely the patient was to deteriorate (P <.05); (c) the presence of children in the home was more likely to lead to improved functioning (p (.05); (d) a higher degree of sponsor-initiated activities in the home ledi to improvement in nonschizophrenics, but deterioration in schizophrenics; and (e) greater supervision was also asSSOciated with nonschizophrenics' improvement and schizo- phrenics' deterioration. After one year of placement, the 31 g ii _ .- will” -... _. .. W. rehospitalization rate for foster care patients was 38%, compared to a national average of almost 50%. In a nonexperimental descriptive study of community care for the deinstitutionalized mentally ill, Datel, Murphy and Pollack (1978) found that rural residental placements produced nearly twice as many community tenure days as urban placements. It was also found that when readmitted to the hospital, rurally-placed clients spent about half as much time in the hospital as did urban-placed clients. An interesting study by Lamb and Goertzel (1971, 1972). studied high and low expectation environments. Hospitalized patients were randOmly assigned to a halfway house with day care and vocational rehabilitation (high expectation) or to a boarding home (low expectation). Patients assigned to the high expectation environment maintained a higher level of function and activity, but also returned to the hospital at a greater rate than the patients in the low expectation environment. The authors felt this was due to the greater demands placed on the high expecatation group. The authors, nevertheless, reCOmmended the high expectation environment over the low expectation envirOnment, which they believed closely resembled the hospital back ward. The authors concluded that patients placed in low expectation environments were not really in the community. 1 A study examining length of stay in sheltered care was conducted by Johnson, Glick and Young (1980). This study examined the relationship between size of a placement 32 .9911. ‘? setting and movement out of the facility. InformatiOn was collected on 15 facilities divided into small (capa- city of six or less). medium (seven to 15). and large (16 or over) settings. After 12 months, Chi—squared analysis showed that patient tenure was associated to a highly sig- nificant degree (p (.001) with the size of the facility. Patients in larger facilities stayed longer. Patients in smaller facilities m0ved more often. Unfortunately, the authors did not obtain information as to where or why these patients moved. The authors suggest that larger facilities may allow more distance and the intensity of interpersonal relationships may be less demanding. This interpretation is questionable, however, since there is no way to judge whether these moves were the result of a "success" or "failure." Several studies have examined the characteristics of family care sponsors in foster homes. ‘ Giovannoni and Ullman (1961) found the best predictor of successful placement was not the length of prior hos- pitalization, as found in other studies (Ullman and Berkman, 1959; Lee, 1963; Lyle and Trail, 1961). but the presence of a male in the foster care home. The male figure did not have to be a positive role model, but his existence was crucial. Bloom (1976). in a survey of 28 home operators in Canada, found that married operators were more effective than those who were unmarried. Mendelsohn (1964) found that the more successful home operators stressed money in their decision to use their home for patient placement. Successful 33 _..._' i imahmw homes also utilized group meetings and contained frOm one to six patients. Finally, Tcheng-Laroche, Murphy, and Engelsmann (1976)_found successful home operators main— tained clear and firm rules. Brown (1959), in a study also cited under the indi- vidual characteristics section, studied 156 chronic schizo- phrenics to determine which of a variety of environments contributed to their success a year following discharge. Success was determined by lack of rehospitalization and social adjustment in terms of employment, social inter- action and need for supervision. Failures were highest in patients who went to boarding homes, their parents or their spouses. The lowest failures were those who went to a lodge or to live with a sibling. The author concluded that pa- tients may do better in a less personal environment that is more socially demanding and not as protected as living with parents or spouses. Rog and Raush (1975) examined 26 studies of halfway houses and found measures of success varied COnsiderably, as did selection of residents. Houses also varied in average length of stay, from 1% months to 30 months, and in terms of the stage of rehabilitation at which the studies were conducted. Control groups were utilized in only two studies. Median figures showed that 79.5% of the residents ad- justed to community living, 58.3% were living independently in the community and 55.2% were employed or in school. Only 20.5% of the halfway house residents required 34 M 12;,- 1‘“ news?" 1 rehospitalization. The authors concluded that the re- sults suggest that an expatient's chance of being read- mitted to a mental hospital are less after living in a halfway house. One of the controlled studies included in the Rog and Raush (1975)_survey (Gumrukcu, 1968) matched 15 ex- residents of a halfway house with 15 patients who entered the community directly frOm the hospital. The groups were matched by sex, age, educational background, degree of "illness," and date of release from the hospital. After a year, no halfway house residents were rehospitalized, while three control group members required hospitali- zation. Ten former halfway house residents obtained sus- tained employment, as compared to four controls. The other controlled study in the Rog and Raush (1975). survey (Rutman, 1971) randomly assigned patients deter- mined to require hospitalization to hospitals or a half- way house. At the end of 18 months, the group showed no differences in terms of job adjustment, living arrange- ment, or financial dependency. Rehospitalization rates following discharge were about 20% for both groups within the 18 month period. The author implies in his conclusions that the halfway house can serve as a substitute for re— hospitalization. 35 TABLE 2.1.--Summary of Studies Examining Characteristics of Residential Programs Associated with Com- munity Adjustment and/or Tenure. Variable Study Number of Patients in Facility Total Number of People in Facility Sponsor Initiated Activities Degree of Supervision Group Meetings High Expectancy Environment Clear, Firm Rules Day Center Employment Vocational Rehabilitation Rural Setting Male Sponsor Married Operators Presence of Children Johnson, et al, (1980) Cunningham, et al, (1969) Mendelsohn (1964) Linn, et al, (1980) Linn, et al, (1980) Linn, et al, (1980) Linn, et al, (1980) Mendelsohn (1964) Lamb, et al, (1971, 1972) Tcheng-Laroche, et al, (1976) Cunningham, et al, (1969) Cunningham, et al, (1969) Cunningham, et al, (1969) Datel, et al, (1978), Giovannoni, et al, (1961)_ Bloom (1976)_ Linn, et al, (1980) 36 TABLE 2.2.--Summary of Studies Finding Residential Programs Superior to Other Placements on Measures of Adjustment or Community Tenure- Reskkmthfl.mxgrmm OUru'Plaxments suxues Lodges, Siblings FbsterIkmes HalfimnrHouse FOSUflrCarelkme Hmmflikefbcifity Hahfieyikmse Bomxfingikmes,8pmmes Parenhs Hospitalization Boamfihglkme Hospitalization Active Treatment Hospital Dnectlusdrmgetn-um Comnuuty Brown (1959) _ Linn, et al, (1980) Lamb, et al, (1971, 1972) Linn, et al, (1977) Mosher, et al, (1975, 1978) Gunmukcu (1968) Halfway House Hospitalization Rutman (1971) Small, Diversified Hospitalization Polak, et al, (1976) OImunitydkmed Program Summary The literature reviewed in this section has considered residential program characteristics relationship to patient outcome. The literature has shown that such factors as the number of patients residing in a facility may have a rela- tionship to community adjustment or tenure. Other factors Such as the amount of sponsor-initiated activities and Whether the program is a high or low expectancy environment may also influence patient outcome. Different residential facility models such as boarding homes, foster care homes and halfway houses have also been studied and compared. 37 Research designs and methodologies used have made generalizing results difficult. Differences in patient populations, geographical location, outcome criteria and length of follow~up also make COmparisons difficult. The studies reviewed dorufi:address the process of returning patients to the community nor do they address integration into the community or satisfaction with the environment. PrOgram characteristics are described in terms such as active treatment, socio-environmental treat- ment or community treatment without clearly enumerating the essential components of such programs. The general- ization or duplication of these poorly described prOgrams is tenuous, at best. There remains a need for descrip- tive research which attempts to identify significant treatment factors which are generalizable and allow for replication elsewhere. 38 Research on Residents' Characteristics This section will focus on patient characteristics and background factors previously associated with patient outcome in residential programs. The first questiOn pertaining to the residential placement of chronically mentally disabled patients might be whether or not mental patients will stay in the com- munity. This was the question asked in,a study by Drake and Wallach (1979). They studied 110 hospitalized, func— tionally psychotic patients who were judged by staff to be well enough to care for themselves and well enough for dis— charge to be realistically possible. The subjects were rated by ward technicians, social workers and an observer from outside the hospital as to their preference for living in the hospital or in the community. Analysis of variance showed that individual living preference ratings showed strong relations (p 4.001) to past community tenure and to prospective community tenure (9 month follow-up). even when past community tenure was held constant. The authors ruled out the possibility that living preference ratings were really based on past community tenure, rather than on the patient's behavior in the present, since the observer had no kn0wledge of past community tenure. The possibility that living preference ratings were really based on psycho- pathology was also unlikely, because all the subjects were functioning relatively well and capable of leaving the hospital. 39 Another study examining patient's living preference and its influence on outcome in residential placement was conducted by Nevid, Capurso and Morrison (1980). They studied 32 ex-hospital patients living in ten family care homes. The study was an attempt to determine whether real- ideal similarity in patient judgments about their foster or family care home was related to their satisfaction with the home and with their adjustment to community living. The Community-Oriented PrOgram Environmental Scale (COPES), a 100-item, true/false questionnaire (Moos, 1974b). was administered to each patient. Patients were asked to indicate their present or real conditions and also their preferred or ideal conditions. In addition, patients were asked to rate their family care homes on four-point scales, measuring (a) their general satisfaction with the home, and (b) the overall quality of the home. Family caretakers and primary therapists were asked to rate the patients on four- point scales, measuring (a) adjustment to the home, (b) level of social functioning, and (c) expectations of future com- munity adjustment. Finally, the Katz Adjustment Scale (Katz and Lyerly, 1963) was completed by family caretakers to assess patients' adjustment to community living and degree of overt behavioral pathology. The study supported previous findings (Moos, 1974b). that patient-judged real-ideal similarity of the treatment environment is positively and significantly related to their satisfaction with the treatment program (.44gr<.73, M = .59). 40 Overt psychotic behavior and socially obstreperous behavior were found to be generally lowest in treatment conditions in which the judged real environment came closest to the patient's View of ideal conditions. The authors concluded that "the degree to which treatment environments match patient expectations, may predict successful adjustment to the community program" (p. 119). In a related study, Lamb (1980) compared patients who did not remain in a residential placement with those who did remain. Lamb found, at a six month follow-up, 32 of 101 psychiatric patients in a board-and-care home had moved on. Ten were living independently; nine had moved to another board-and-care home; seven had returned to their families; two left to live with a boyfriend or girlfriend; one was in a drug rehabilitation program; two had moved to a half- way house; and one person could not be located. Lamb found that of those who left, (1)_69% had re- sided at the board-and-care home for 12 months or less, while only 22% of those who remained had resided there for 12 months or less; (2).34% of those who left were under 30 years of age, while only 12% of those who remained were under 30 years of age; (3) 69% of those who left had goals or a desire to change something, whether realistic or not, while 38% of those who remained had goals; (4) of those who left, 47% had been hospitalized in the previous year and 23% of those who remained had been hospitalized during the previous year. Lamb suggests that while there are 41 always legitimate reasons to move, many of those who moved may have been resisting the pull of dependency. Another possible explanation is that this group was less psychia- trically stable, and their move could have been an attempt to relieve symptoms and conflicts. Other studies have found that length of hospitali- zation or number of hospitalizations are associated with tenure or adjustment in community residential placements. Their findings have varied, perhaps in part, due to the fact that they have examined a variety of residential programs. Sandall, Hawley and Gordon (1975) studied 72 apart- ment graduates and found that apartment residents who left apartments for more structured and supervised envi- ronments had been in hospitals for a quarter the length of time. On the other hand, Baganz, Smith, Goldstein, and Pou (1971) found successful placements in a YMCA had less prior length of hospitalization (5.3 years versus 9.1 years). In another study examining still another type of residential program, Johnston (1974) used halfway house residents in an investigation of residents' personal characteristics related to success in independent living in the community. Extensive demoqraphic data was collected on all clients admitted (n=70) and discharged (n=60) from three halfway houses over a six month period. Data was also collected on clients who had already been discharged for six months (n=41). Only two variables were found to be significant in regard to length of stay in a program or 42 future success in the community. These two variables were the number of prior hospitalizations and the total number of days hospitalized with community success having fewer hospitalizations and fewer days in the hospital (p <.05). Similar findings were reported by Lee (1963) in a study that examined yet another type of residential pro- gram, family care homes. Lee found that patients who were not rehospitalized after placement had a quarter the length of prior hospitalizations than those who were hospitalized. This was the only significant variable found to effect rehospitalization. Studies have found a variety of variables such as age, prior history of hospitalization, diagnosis, and living preference to be significantly related to community tenure or adjustment, or both. These variables have not been found to be significant in all studies, and results have not always been consistent. Lyle and Trail (1961) found the patients making suc- cessful adjustment to foster homes were: older, schizo- phrenic, advocated recreational interests and were not as interested in leaving the hospital as those who did not adjust well to the foster home. Ullman and Berkman (1959). studied characteristics of foster home patients and found that greater length of total hospitalization was associated with patients who adjusted successfully to the home and remained or moved on to greater independence. Patients who stayed briefly and were rehospitalized were character- ized by smaller periods of hospitalization. 43 However, Simon, Heggestad and Hopkins' (1968) study of successful placements in foster homes did not support those of Lyle and Trail (1961) and Ullman and Berkman (1959). Simon, et al, found successful foster home placements were older at the time of their first hospi- talization, had a shorter period between hospitalization and placement, and had fewer prior admissions. Patients who were less chronic showed better post-hospital adjust- ment. Simon, et al, state that their failure to support the earlier findings could be due to a shortage of younger patients in their study. Another study finding a negative correlation between length of community tenure and prior hospitalization in a different setting, was conducted by Cunningham, Botwinik, Dolson and Weickert (1969). They conducted a five-year .follow-up study of halfway house residents and found that only 40% remained in the community for two years. The majority were rehospitalized within one year and tended to be the patients with less prior hospitalization. Suc- cessful placements were more involved in employment, vo- cational rehabilitation or day center, were single and were located in large halfway houses. Brown (1959) studied 156 chronic schizophrenics a year after discharge. Patients were placed in a variety of environments ranging from lodges to boarding homes to parents or spouses. Results showed that the length of prior hospitalization had no relationship to rehos- pitalization; however, those with more prior hospitalizations 44 were more socially adjusted in terms of employment, social interaction or need for supervision. There was also a tendency for older patients to have lower hospital reentry rates and greater social adjustment. Lamb and Goertzel (1977) followed 99 severely dis- abled individuals (determined by support from Supplemental Security Income). and, like Brown (1959). found older sub- jects were rehospitalized less than those who were younger. One study has taken a unique approach to the question of post-hospital adjustment. Steinberg, Yu, Brenner and Krieger (1974) used Rotter's Locus of Control Scale (Rotter, 1966) to predict independent functioning as measured by a scale containing items relating to such behavior as employ- ment, handling money, preparing food, and having friends. The subjects were 112 schizophrenics who constituted five groups: (1) 25 patients in their first week of hospitali- zation; (2) 25 chronic patients who had spent at least one year in continuous hospitalization; (3) 18 former chronic patients who had been residing in apartments for more than 12 months; (4) 30 former chronic patients who had been living in boarding homes more than 12 months; and (5).14 chronic patients who were tested prior to discharge from the hospital and again after living in apartments for from one to seven months. Locus of control scores (I-E scale)_failed to discrimi- nate among,any of the groups tested and scores did not change over time. There were, however, significant negative cor- relations between I-E scores (the higher the score, the 45 more external the locus of control; the lower the score, the more internal the locus of control) and independence scores for patients living in apartments for more than one year (r=-.67, p <.005). Furthermore, I-E scores predicted level of independent functioning for apartment patients out of the hospital less than one year (r=-.52, p (.025). The authors concluded that the I-E scale may correlate with and predict level of independent functioning for schizo— phrenics in situations where behavior is more free to vary and that it might be useful in determining appropriate placement for patients leaving a mental hospital. TABLE 2.3.-- Summary of Studies Examining Characteristics of Individuals Served in Residential Facilities and Their Effect on Community Adjustment and/or Tenure. Variable Study Number of Prior Hospitalizations Johnston (1974) Simon, Heggestad and Hopkins (1968) Cunningham, Botwinik, Dolson and Weickert (1969) Brown (1959), Length of Prior Hospitalization Sandall, Hawley and Gordon (1975) Baganz, Smith, Goldstein and Pou (1971) Johnston (1974) Lee (1963)‘ ' Ullman and Berkman (1959). Age Lamb (1980). Lyle and Trail (1961) Simon, Heggestad and ‘ Hopkins (1968) Brown (1959) ' Lamb and Goertzel (1977) 46 TABLE 2.3.--continued Variable Study Marital Status Cunningham, Botwinik, Dolson, Weickert (1969). Diagnosis Lyle and Trail (1961)_ Residential Stability Lamb (1980). W The literature reviewed has considered patient charac- teristics and background factors in relation to patient out- come in residential settings. As noted earlier, few studies employed control groups or random assignment. The litera- ture has shown that some variables have been more frequently involved in assessing residential outcome than others. Number of prior hospitalizations, length of prior hospitali- zations and age are the three variables which have been most often associated with outcome in the studies reviewed (See Table 2.3). The studies reviewed have not always shown the same variables to be significant and the effect of a variable has not always been in the same direction. These differ- ences may, in part, be due to the different follow-up periods, different outcome criteria, different pOpulations studied, and the different methodoloqies used. Differences may also be related to the interaction between individual and background factors with differeing residential programs 47 ranging from foster homes to halfway houses to lodges, etc. Clearly, there remains a need for descriptive re- search which provides an understanding of the process of returning patients to the community and which focuses on the quality of patient life and satisfaction with the environment. 48 CHAPTER III DESIGN OF THE STUDY The purpose of this section is to present the design of the study. The following sections are included: selection and description of the subjects, description of residential settings, measures, procedures for collecting the data, statistical hypothesis, research design and analysis. Selection and Description of the Subjects The subjects in this study were residents of super- vised homes or supervised apartments for the mentally ill connected with the Clinton, Eaton, Ingham Community Mental Health Board. Residents in these prOgrams were also clients of Community Mental Health and eligible for a variety of aftercare services. Nearly all residents were considered to be chronically mentally ill. Most residents were diag— nosed as schizophrenic and had a history of multiple psy- chiatric hospitalizations. Nearly all residents received some sort of government financial assistance due to their illness. All residents in supervised living programs during the week of May 21, 1984 were considered as potential subjects. Each resident was asked to participate in the study and 49 was given a consent form. Residents agreeing to partici- pate were asked to complete the questionnaires. Eighty-six beds were available in the eight homes and 31 beds were available in the ten apartments. The resi- dents occupying these 117 residential beds provided a maximum of 117 potential subjects. However, at the time the data was collected, only 96 beds were occupied reducing the actual number of possible subjects from 117 to 96. Seventy-eight (78) residents or 81% of the possible number of subjects participated in the study. No analysis for differences between residents who refused to participate and those who agreed to participate was conducted dUe to the large percentage (81%) of residents participating. Table 3.1 presents demOgraphic data for all 96 possible subjects. Appendix L, page 148 presents a summary of resident's characteristics by residential setting. 50 TABLE 3.1.--Demoqraphic Data for Population Variable Number of Subjects Percentage Sex Race Female Male Caucasian Black Hispanic Bi-racial Marital Status Single Divorced Married Widowed Diagnosis Schizophrenia Personality Disorder Affective Disorder Schizo Affective Retardation Organic Brain Syndrome Adjustment Disorder Current Day Program None Job Volunteer Work School Vocational Training Day Treatment Other More Than One of Above Mean = 33.9 S.D. = 12.0 Range = 18 - 70 Educational Level 11.8 2.1 7 - 18 Mean S.D. Range 49 arcane N N—A \JONQKDNUIUIQ no 0 o o o ‘vdLIJ-AOD-AMN—a 51 TABLE 3.1.--continued Variable Number of Psychiatric Hospitalizations Mean = 4.6 S.D. = 3.3 Range = 0 - 19 Year of First Hospitalization Mean = 1970 S.D. = 15.7 Range = 1948 - 1983 Year of Most Recent Hospitalization Mean = 1979 S.D. = 14.6 Range = 1967 - 1984 Total Months Hospitalized Mean = 33.4 S.D. = 69.8 Range = O - 404 Number of Months in Present Residential Setting Mean = 10.0 SOD. = 9.4 Range = 0 - 42 Amount of Weekly Spending Money (in dollars) Mean = 16.9 S.D. = 15.8 Range = 0 - 54 a Standard Deviation Description of Residential Settings The Clinton, Eaton, Ingham Community Mental Health Board maintains a comprehensive residential housing program. A continuum of living situations are provided, allowing the freedom to place clients in settings that, hopefully, match 52 their needs. These settings range from highly dependent and structured, with large amounts of staff intervention, to supervised apartment settings with minimal structure and less frequent staff contact. The different settings are examined in greater detail below, starting with the most dependent settings, and proceeding to less dependent settings. ABBOTT HOUSE DESCRIPTION Abbott House is an eighteen (18) bed "Room and Board" facility. The house is open and staffed 24 hours a day by CMH employees. The house is "double-staffed" at all times, except overnight. PHILOSOPHY The house is designed to provide a pleasant and sup- portive environment with a minimum of structure. Services provided include: two prepared meals a day (residents pre- pare their own breakfast), help holding and monitoring medi- cation, help budgeting spending money (if necessary), and six hours a day of structured activities. Residents are required to meet only minimal expectations. The few expectations enforced include: no violence, no pro- perty destruction, acceptable personal hygiene, and compliance with taking prescribed medications. Unlike other houses, the program does not usually require Day Treatment attendance, household chores, or peer group meetings. 53 ACTIVITY PROGRAM A full time activity aide offers residents 24 hours per week of structured day time activity. Activities include: crafts, discussion groups, group outings, cooking, games, and just about everything residents wish to help plan. These groups help give residents the positive feelings of construc- tive activities. ANNEX Three of Abbott House's 18 beds are actually located in the Abbott Annex, two blocks away from the house. Annex residents do their own housekeeping, purchasing and most of their own cooking. They may choose to purchase meals from the house or to make their own. Other house services offered include: socialization, support, medication monitoring, help with budgeting, and activity groups. House staff also help Annex residents run a house meeting each week. COOK PROGRAM The cooking and cleaning at Abbott House is done by Community Mental Health clients. This meets the physical needs of the house, while giving gainful employment to other clients. Cooks are supervised by the house business aide. This allows for regular supervision, evaluation, and the potential for placement in competitive employment. HYATT & EUREKA AFC HOMES Hyatt and Eureka Houses are both 12 bed licensed Adult Foster Care Homes, providing room and board, and 24 hour supervision. Both homes are located in rural areas. Program- ming is provided outside the homes, and clients who are unable to attend day programs may remain in the home during the day, as there are staff available at all times. Length of stay is dependent upon client needs. Staff prepare meals and hold and monitor medications. There are no regular resident meetings. WHISPERING PINES DESCRIPTION Whispering Pines is a twelve (12) bed, private Adult Foster Care Home, on contract with Community Mental Health Board. Staff consists of a full time manager, and 5 additional aide staff. The house is "double staffed" from 3:30 to 11:00 p.m. on weekedays, plus all weekend. PHILOSOPHY The house prOgram is structured to gently encourage inde- ,pendence, while allowing a resident to be more dependent, if needed. The prOgram teaches and encourages basic living skills, by asking residents to help with cooking, cleaning, laundry, and other household responsibilities. While in Whispering Phines, residents are expected to see a case manager/therapist weekly, attend day programming daily, and keep medication appointments. House staff help residents by reminding them of appointments and holding medication. JEROME AND OASIS AFC HOMES DESCRIPTION Both Jerome and Oasis homes are licensed Adult Foster Care Homes. Each house is open from 3:30 p.m. to 9:00 a.m., Monday 55 through Friday, and 24 hours on weekends and holidays. During these times, they provide room, board, and client supervisions. Jerome House is a 6 bed facility. Oasis provides 9 beds. Each house is operated by Community Support Services employees. Staff consists of a full time manager, two full time afternoon/ evening aides, and two full time over-night aides. In general, both houses are "double staffed" from about 3:30 to 8:30 p.m., and "single staffed" the rest of the evening, plus weekends. PHILOSOPHY House programing is structured to enable the greatest possible independence for each resident, and to help each resi- dent learn and deal with the responsibilities of community life. Residents learn skills and confidence in daily living skills by doing house chores, such as: cooking, vacuuming, and meal clean- up, etc. Residents are expected to attend day programing. Once familiar with public transporation, residents must take the bus to day programing, medication appointments, and therapist appointments. STEP SYSTEM It is difficult to expect this sort of independence frOm clients who function at many different levels. To allow for these differences, both houses have a Step Level Program which allows residents to move both up and down six step levels, as their needs and desires change. A low step level means lower responsibilities and also lower privileges. For example, a client on Step Level One gets reminded to do chores, reminded to get up in the mornings, and gets his/her mOney 56 on a daily basis. On the other hand, a high step level means large amounts of responsibility, and also greater privileges. A resident who gets to Step Level Six may c0me and go as he/ she pleases, will handle money on a monthly basis, and is never reminded by staff to do chores. The move from Step Level One to Level Six may take as little as a couple of months, or as long as years. GROUP SYSTEM One of the most important features of both Oasis and Jerome House is the fact that residents are in charge of making many of the decisions. The residents meet three times a week to handle problems, plan acitivites, change step levels, and give each other support. Problems are handled mainly through a problem note system. When problems occur, a note is written by residents or staff. At meetings a solution is arrived at (anywhere from a penalty job to a discussion and warning). Staff intervention in the group process is mainly limited to offering weekly feedback to the group and resubmitting problem notes inadequately dealt with by the group. TRANSITIONAL LIVING HOUSE (TLH) DESCRIPTION TLH is a six month time limited prOgram, with room for nine clients. The house is open for clients frOm 3:30 p.m. to 10:00 a.m., weekdays, and all weekend. The house is staffed by a full time manager, who works primarily days, and by two resident aides who work afternOOn/ evenings and weekends. The facility is not licensed and no regular staffing is provided overnight. 57 - ~11? . PHILOSOPHY TLH provides a setting for clients to develop living skills and a knowledge of community resources (housing, schooling, employment, etc.). Since time is limited a specific five step level system helps clients make the decisions and gather the resources to live independently. When clients move in, they must agree to: (1) participate in the group decision making process; (2) work on the step system; (3) follow hOuse rules; (4) hold and take their own medication; and (5) make financed room and board payments. STEP SYSTEM To aid the progresson to independent living, each resi— dent has his/her own Step Program handbook. The handbook out- lines five (5) step levels, and uses worksheets and check-offs to help the client achieve a variety of objectives. To move from Level 1 to Level 2, for instance, a resident has to do things like: apply for financial assistance, do a "heavy cleaning" job, and ride the buses to two different places. To move from Level 4 to Level 5, a resident must supervise at least 5 grocery trips, plan at least 1 meal per week, plan and organize a recreational activity, and develop an independent living monthly budget actually, these are Only some of the things needed to move from Step 4 to Step 5). GROUP SYSTEM Residents meet five times a week (once with staff present) to make decisions and solve problems. Problems discussed range anywhere from an undone chore, to perSOnal problems, to broken rules. Group members are expected to make a written "problem 58 note" when they see a problem, then submit it for later group discussion. When a client is clearly not meeting the requirements to live in TLH, the group may put the client on "contract." This usually gives the client a specific period of time to work on problem areas. When the time is up, and if the client made little effort to correct the problem, the group may ask him/ her to leave TLH. This decision is made with staff feedback, and is usually used as a last resort, as few residents want tO "kick out" a fellow resident who has a problem or problems. HIGH STREET DESCRIPTION High Street is a nine (9) bed group living situation. Residents are charged rent only. They shop for their own food and do their own cooking. They hold house meetings during the week to solve group problems and divide up the tasks of cleaning. Staff spend about 40 hours a week at the home, including weekends, and work basically as consul- tants. There are some group activities, outings, and assis- tance with shopping, etc. Requirements to participate in constructive day activities is decided on a case by case basis, with the resident working the details out with their case manager . SUPERVISED APARTMENTS DESCRIPTION This is a program of semi-independent living that is 59 tied into the Residential Case Management Unit. The apart- ments consist of hOuses, 2 to 5 bedrooms, scattered through the city. Each house has a primary contact person who holds a house meeting with the residents once a week. These meet- ings are designed to work on interpersonal problems that exist in any group living situation. Residents of the program are required to pay rent, participate in a constructive day activi— ty, be responsible house mates and good neighbors. There are regularly scheduled shopping trips, laundry trips, and evening social acitivites for members of this program. This program is designed to follow up On gains made in some of the more structured residential homes. This pro- gram is tied into the Case Management Unit with various staff responsible for at least one supervised apartment. Measures Five forms of instrumentation were utilized in the study; these included: (a) Personal Data Sheet, (b) Resi- dential Program Characteristics Scale; (c) Community- Oriented Program Environmental Scale (COPES) (MOOS, 1974b); (d) External Integration Scale (Segal and Aviram, 1978); and (e) Consumer Response Scale (Segal and Aviram, 1978). 60 Personal Data Sheet As the review of the literature has shown, various individual and background characteristics of residents in supervised settings have been associated with resident outCOme. In this study, a data sheet was used to elicit this information in order to describe the subjects. The data sheet was designed to gather the following information about each resident: resident's age, sex, race, educational level completed, marital status, diag- nOSis, number of psychiatric hospitalizations, total length of time hospitalized, number of years since first hospitali- zation, date Of most recent hospitalization, tenure in their present supervised living placement, amount of spending money available to client, and current enrollment in a job, school, or rehabilitation program (see Table 3.1). The data sheet was completed by the resident's Case Manager in the Community Mental Health (CMH) system. 1. Item Development Items for the Personal Data Sheet were selected on the basis of previous research findings. hose individual characteristics Of residents most Often associated with patient outcome in earlier studies were included. 2. Reliability and Validity The reliability and validity of the Pers0nal Data Sheet in this study was undetermined and beyond the scope Of the present study. The actual relia- bility Of the information Obtained by use of the 61 Personal Data Sheet was assumed to be comparable to information currently utilized in psychiatric facili- ties. Residential Program Characteristics Scale The literature reviewed has described residential programs in only the broadest terms, i.e., fOster care homes, halfway houses, supervised apartments. In this study, the Residential Program Characteristic Scale was one of the instruments used to describe programs and how they vary, in greater detail. The Residential Program Characteristic Scale consisted Of three scales designed to measure: (1) the degree Of Structure in a program; (2) the degree of Resident Respon- sibility/Independence permitted or encOuraged by the pro- gram; and (3) the degree of Staff Supervision. Items making up these scales focused on program charac- teristics which were fairly discrete and easily modifiable. They included program characteristics, such as curfews, residents controlling their own money, staff supervising medication, etc. These scales were completed by the staff Of each house or by the staff person supervising each apartment. 1. Item Development Items for the Residential Program Characteristic Scale (RPCS) were generated by the researcher and CMH staff knowledgeable in the area of residential 62 placement. Items judged to be behaviorally descrip- tive Of program factors, modifiable, and useful in differentiating between various programs were COn- sidered for use. Items were then grouped according-to conceptual compatability, arriving at the present three scales. 2. Reliability Coefficient alpha was used as a measure of reliability for the RPCS. Coefficient alpha is a measure of equivalence and internal consistency. The RPCS was administered to 43 house staff and apartment supervisors participating in the present study. Cronbach's alpha was used to determine the internal consistency of each subscale. Results - showed an alpha of .82 for Structure; .84 for Independence/Autonomy; and .93 for Staff Super- vision. These results suggest that the RPCS had strong reliability in the present study. 3. Validity The RPCS has face validity and content validity. Content validity is the degree to which scale items represent the content which the scale is designed to measure. Content validity for the RPCS has been provided by developing items around specific Objectives and sampling a large number of items suggested by knowledgeable persons. Construct validity is the extent to which the scale measures the hypothetical COnstructs involved. 63 This type of validity is difficult to attain. Some construct validity was achieved because the RPCS was able to differentiate between resi- dential programs (see Chapter IV, page 74, Hypo- thesis 2). Community-Oriented Program Environmental Scale The Community-Oriented Program Environmental Scale (COPES) was developed by Moos (1974b) to assess the social environments of community-based psychiatric treatment pro- grams, e.g., halfway houses. community care homes or day programs. In its present form, the scale is a 100-item, ten-subscale instrument. A short version with 40 questions is also available and was utilized in the present study. The first three subscales, Involvement, Support and Spon- taneity, are conceptualized as measuring Relationship dimen— sions. These three subscales assess the extent to which staff support members and members support and help each other, and the amount of spontaneity, or free and open ex- pression, existing within these relationships (Moos, 1974). The next four subscales, Autonomy, Practical Orient— ation, Personal Problem Orientation, and Anger and Aggres- sion, are conceptualized as Personal Development, or Treat- ment PrOgram, dimensions (Moos, 1974b). The last three subscales of Order and Organization, Program Clarity, and Staff COntrol are COnceptualized as 64 assessing System Maintenance dimensions (Moos, 1974b). These scales were completed by participating resi- dents of each program in the present study. 1. Item Development Most items on the COPES were adopted from the Ward Atmosphere Scale (WAS)(Moos, 1974a). The WAS was developed to measure social climates of psychi- atric treatment programs as perceived by patients and staff. The theories of environmental press developed by Murray (1938) and Stern (1970) provided the original basis for the WAS. Additional items for the COPES were formulated from program descrip- tions and interviews of patients and staff in various community programs. A resulting 130-item form of the COPES was administered to members and staff in 21 community-oriented treatment programs (day centers, residential centers, COmmunity care homes, etc.). Tested in the 21 programs were 373 members and 203 staff. The current ten-subscale form Of the COPES was derived using the following criteria: a. Each subscale should have acceptable internal consistency, and each item should correlate more highly with its own than any other subscale. Two of the original 12 scales were dropped because they did not meet this criteria. Internal COn- sistencies were calculated using Cronbach's 61 and average-within-program item variances. b. When possible, not more than 80%, nor less 65 than 20% of subjects shOuld answer in one direction. This criterion was set to avoid items characteristic only Of extreme programs. Ninety—five percent Of COPES items meet this criterion. c. There should be approximately the same number Of items scored true as scored false within each subscale, to COntrol for acquies- cence response set. d. Items should not correlate significantly with the Halo Response Set Scale, a scale de- veloped to assess both positive and negative halo in program perceptions and given to members and staff. Last, means and standard deviations on all subscale scores were calculated for each program, separately, for members and staff. The results Of one-way analysis of variance indicated all ten subscales differentiated among the original 21 programs at p (.01 for all subscales for members and for nine of ten subscales for staff. The short version (40 questions) of the COPES was utilized in the present study. Correlations between the short version and the full length (100 question) version of the COPES were above .75 for 14 of the original 21 pro- grams for both members and staff scores. The lowest cor- relation (.68) was for members. The ten subscales of the COPES measure distinct, although correlated characteristics of member and staff perceptions of community—based programs (Moos, 1974b). 66 2. Reliability Internal consistencies (Kuder-Richardson formula 20) for the initial group of 21 programs were cal— culated following Stern (1970) using average within- ' program item variances. The subscales have accept- able internal consistency, with a mean of .78 for staff and a mean of .79 for members. Item-to-sub- scale correlations are moderate to high average, with a mean of .47 for staff and a mean Of .41 for members. The intercorrelations of the ten subscale scores have been calculated for the same original 21 pro- grams. The highest intercorrelation is .50, and the only cluster of subscales showing even moderate intercorrelations in both member and staff samples was composed of the Relationship dimension of Involve- ment, Support, and Spontaneity. Test-retest reliability has not been calculated for the COPES; however, test-retest reliability analysis for the WAS has been satisfactory. Since the COntent and the structure of the ten COPES and the ten WAS sub- scales are directly parallel, these results may be generalized as applicable to COPES (Moos, 1974b). 3. Validity The COPES construct validity is strengthened by the fact that the results of one-way analysis of variance indicates that all ten subscales signifi- cantly differentiated among the original 21 programs 67 for both member and staff responses. The actual proportion Of subscale variance accounted for by differences amOng programs was ascertained by esti- mated Omega-Squared (Hays, 1963). The percentages varied from a low of five percent on the Practical Orientation subscale for staff to a high of over 50 percent on both the Autonomy and Order and Or- ganization subscales for staff. These results may, Of course, vary greatly, depending on the particular sample of programs studied. Further evidence for the validity of COPES is that patient and staff perceptions measured by the WAS are only minimally, if at all, related to their tendency to answer in socially-desirable directions. The Crowne-Marlowe Social Desirability Scale and the Social Desirability subscale of the Ward Initiative Scale (WIS) were used in a study of patients in four different state hospital wards. There was a slight positive relationship between the Crowne-Marlowe and the WAS Relationship Dimension (average r'= .12). The Crowne-Marlowe was not correlated with other WAS dimensions. Finally, staff who answered in a socially- desirable direction had a slight tendency to also answer the WAS items in somewhat more desirable directions. The correlations were generally low, although four out of 94 were above .20 (Moos, 1974b). 68 External Integration Scale The External-Integration Scale developed by Segal and Aviram (1978)_was administered to residents in the present study. External-Integration (EI) has been defined as the extent of access to, and participation in, community life, whether supported by the home or not. The seven-factor analytically derived EI subscales are: (1) Attending to oneself; (2) Access to community resources; (3) Access to basic and personal resources; (4) Familial access and participation; (5) Friendship access and participation; (6) Social Integration through community groups; and (7) Use of COmmunity facilities. 1. Item Development Segal and Aviram developed two separate social integration scales. The first, External Integration (EI). was developed to measure SOCial integration into the community. The seCOnd, Internal Integration (II). was developed to measure social integration into the residential facility. Only the El scale was utilized in the present study. During the development Of the BI scale, all items thOught to be part of the original conception of social integration that were skewed more than 90% were eliminated. All items originally thought to belong in El or the II scale, respectively, were separately intercorrelated to produce two matrices of approximately 80 items each. These matrices were 69 cluster-analyzed to determine what major clusters could be within the El and II content areas. Once the best clusters were derived, they were put into one large correlation matrix containing both EI and II clusters. This large correlation matrix was then factor-analyzed, using the principal factor solution with varimax rotation to simple structure. These pro- cedures produced the present 12 subscales, seven com- prising EI, and five comprising II. There was little overlap between items on the two scales. 2. Reliability The internal consistencies (as measured by Alpha) of the subscales making up EI and II, have been com- puted during the development of the scale. Also com- puted were the average item-to-subscale and the average item-to-other-subscale correlation for each subscale. The major criterion used for retaining a subscale was a high average item-to—subscale correla— tion versus a low average item-to-other-subscale cor- relation. All 12 subscales for both EI and II met this criterion. The EI scale has acceptable internal consistency with good item-to-subscale correlations. Internal consistencies (Alpha) scores for BI range from a high of .91 on the Access to community resources subscale to a low of .65 on the Use of community facilities subscale. The average alpha score for the seven scales was .78. The average item-to-subscale 7O correlation ranges from a low of .65 on Attending to oneself subscale to a high of .78 on the Friend- ship Access and Participation subscale. The average item-to-subscale correlation for the seven scales was .72. Average item—to-other-subscale correla- tion ranges from a high of .39 on the Friendship Access and Participation subscale to a low of .26 on Attending to Oneself subscale. The average item-to- other-subscale correlation score for all seven scales was .30. Finally, there was a positive significant rela- tionship of all the subscales to each other (average intercorrelation of .70). Given this relationship, Segal and Aviram decided to add the normalized scores from each subscale to generate the External Integra— tion scale score. This procedure gave equal weight to each subscale in the total score. There is no known test-retest reliability infor- mation available for the social integration scales. 3. Validity The External Integration Scale has face validity and content validity. The EI scale purports to meas- ure social integration into the community as measured by access and participation in a variety of activities outside Of the residential facility. There was little overlap during the scales development between items on the EI and II scales. This seems to be a practical validation of the original conceptual distinction 71 between the two scales and strengthens their content validity. Content validity was also strengthened, in that the items and subscales were developed around specific Objectives and a large number of items were sampled. The degree to which the El scale measures the con- struct of social integration into the community is difficult to demonstrate. Some evidence of construct validity can be assumed if the ET scale distinguishes between residents and different residential prOgrams. The scale has successfully distinguished between resi- dential programs and various predictors in past re- search (Segal and Aviram, 1978). Consumer Response Scale Residents in the present study were asked to complete the Consumer Response Scale (Segal and Aviram, 1978). This scale asked residents to rate their satisfaction with the home or apartment where they resided. Residents were asked if they find their living in the home/apartment helpful, if the rules are good, if they have enough pri-~ vacy, goes if they feel they have enough influence with what on in the house/apartment, if they have the spending money they need, etc. The original scale has been modified in this study to include a Liekert Type response scale. 1. Item Development The Consumer Response Scale was developed by Segal and Aviram (1978) to determine residents' satisfaction with their living facility. The scale was developed to cover many aspects of supervised living, including: the physical environment, oper-. ation of the hOme (rules), treatment, etc. 2. Reliability and Validity The reliability and validity of the Consumer Response Scale is undetermined. Procedures for Collecting the Data Data was collected during the week of May 21, 1984, from residents at a regularly scheduled house or apartment meeting. All residents were asked to participate. Resi- dents were told that the study was to help determine which components of the program were most helpful. Those resi- dents volunteering to participate were given consent forms, COPES, External Integration Scales, and Consumer Response Scales. CMH Case Managers met as a group with the researcher. They also received a description of the study and its pur- pose. Case Managers were asked to complete a personal data sheet for each of their clients residing in One of the supervised settings. Finally, each house staff member or apartment super- visor was contacted. A description Of the study and its 73 purpose was provided. House staff and apartment super- visors were asked tO complete the Residential Program Characteristic Scale at a weekly staff meeting. Statistical Hypotheses The following hypotheses were tested in the present research: H : O NO difference will be found between residential settings' social environments, as measured by residents' reports on the Community-Oriented Pro- gram Environmental Scale (COPES). Residential settings' social environments will vary, as measured by residents' reports on the COPES. No difference will be found between residential settings' structure and program characteristics, as measured by staff report on the Residential Program Characteristics Scale (RPCS). Residential settings' program characteristics will vary, as measured by staff report on the RPCS. There will be no relationship between programs' "social climate,‘ across residential settings, as measured by the COPES, and resident integration into the community, as measured by the External Integra- tion Scale (EI). There will be a relationship between programs' "social climate," across residental settings, as measured by the COPES, and resident integration into 74 the community as measured by the E1 scale. There will be no relationship between programs' "social climate," across residential settings, as measured by the COPES, and resident satisfaction with the residential setting, as measured by the Consumer Response Scale (CRS). There will be a relationship between programs' "social climate,‘ across residential settings, as measured by the COPES, and resident satisfactiOn with the residential setting, as measured by the CRS. There will be no relationship between program charac- teristics, across residential settings, as measured by the Residential Program Characteristics Scale (RPCS), and resident integration into the community, as measured by the El scale. There will be a relationship between program charac- teristics, across residential settings, as measured by the RPCS, and resident integration into the com- munity, as measured by the El scale. There will be no relationship between program charac- teristics, across residential settings, as measured by the RPCS, and resident satisfaction with the resi- dential setting as measured by the CRS. There will be a relationship between program charac- teristics, across residential settings, as measured by the RPCS, and resident satisfaction with the resi- dential setting as measured by the CRS. 75 Design The general design of this study was descriptive. The purpose was to determine what characteristics of resi— dential programs for the chronically mentally ill were cor— related with residents' social integration into the community and satisfaction with their placement. Review of the liter- ature clearly indicates the need for research in this area. The studies reviewed have contributed very little to our knowledge Of what actually takes place in residential set- tings to promote residents' reintegration into the community and satisfaction with their environment. The present des- criptive research is necessary to broaden this understanding and to develop a useful theory. The present study was also intended to generate further questions and stimulate new approaches to research in this area. Two simultaneous studies were conducted in the research. The first was primarily concerned with whether or not there are differences between residential programs. The second was concerned with how these differences, if they exist, effect the dependent variables. Design I The first design addressed the broad research question: Are there differences between residential programs? In order to answer this question, residents' scores on the COPES were compared. (Supervised apartments were counted as one residential setting). Next, house managers or apartment supervisors' scores on the RPCS were compared. 76 Both comparisons were nested designs with residential set- tings as the unit of analysis. Design II The second design addressed the broad research question: What residential program characteristics (social en- vironment, rules, etc.) are assOciated with residents' in- tegration into the COmmunity and satisfaction with their residential placement? There were 13 independent variables and two dependent variables in this design. The independent variables were the ten COPES subscales for residential settings and the three RPCS subscales for residential settings. The de- pendent variables were the residents‘ External Integration Scale scores and the residents' Consumer Response Scale scores . Analysis For Design I two univariate analysis of variance were used to determine if there were any differences between residential settings on (1) the Residential Program Charac- teristics Scale and (2) the Community-Oriented Program Environmental Scale. A univariate method of analysis was selected because these variables were expected to measure distinct program and social climate characteristics. For Design II, four multiple regression equations were used to determine whether any relationship existed between the subscales of the independent variables (Residential 77 Program Characteristics Scale and the Community-Oriented Program Environmental Scales) and the two dependent variables (External Integration Scale and the Consumer Response Scale). This form of analysis allows for examin— ation of the relationship between independent and dependet variables as well as the strength of this relationship. Multiple regression was especially well suited to this type Of analysis because it has more power and allows for finer detection and better prediction. Summary Seventy-eight subjects agreed to participate in the study. This number represented 81% of the residents in eight residential treatment homes and ten residential apartments (apartments were counted as one residential setting during statistical analysis). Nearly all subjects were considered to be chronically mentally ill and most were diagnosed as schizophrenic. NO attempt was made to randomize subjects in the present descriptive research. Subjects completed three questionnaires, including: the Community-Oriented Program Environmental Scale (COPES), the External Integration Scale (EI), and the Consumer Response Scale (CRS). Residential home staff and apart- ment supervisors completed the Residential Program Charac- teristics Scale (RPCS). Each subjects' case manager in the Community Mental Health system completed a Personal Data Sheet for their client/subject. 78 Statistical hypotheses were formulated to determine if there were differences between residential settings' prOgram characteristics and/or social environments. Uni- variate analysis of variance were used for this purpose. Hypotheses were also formulated to determine if residential settings' program characteristics and/or social environments were related to the subject's satisfaction with their residential setting and/or the subject's inte- gration into the community. Multiple regression analysis was used for this purpose. The develoPment Of the Residential PrOgram Charac- teristics Scale was also presented. Results of reliability tests showed a low alpha score of .82 and a high Of .93. The results of the hypotheses tested are reported in Chapter IV. 79 CHAPTER IV ANALYSIS OF RESULTS The statistical hypotheses, an analysis of the data and a summary of the results are presented in this chapter. The first and second hypotheses were tested by univariate analysis Of variance. The third, fourth, fifth, and sixth hypotheses were tested by pairwise multiple regression analysis. Hypothesis I: Differences Between Residential Settings' Social Environments Null Hypothesis: No difference will be found between resi- dential settings' social environments, as measured by residents' reports on the COm- munity-Oriented Program Environmental Scale (COPES). Alternative Hypothesis: Residential settings' social environ- ments will vary, as measured by resi- dents' reports on the COPES. Significant differences were found between residential set- tings' social environments for three of the ten subscales on the COPES (p <405). Mean squares, F ratios, and F prob- abilities are shown in Table 4.1. Means, standard deviations, and the number of Observations per residential setting are found in Appendix I. 80 TABLE 4.1.--Summary Data for Analysis of Variance of Resi- dential Settings' Social Environments (COPES scores). Variable Mean F F Prob- (COPES subscale) Squares Ratio ability COPES 1 (Involvement) Between group a 1.6443 1.043 .41 Within grOups 1.5766 COPES 2 (Support)? Between groups 3.9663 3.684 .01 Within groups 1.0766 COPES 3 (Spontaneity). Between groups 2.6440 2.017 .06 Within groups 1.3108 COPES 4 (Autonomy)? Between groups 4.3094 4.357 .01 Within groups .9891 » COPES 5 (Practical Orientation). Between groups 1.1143 .928 .50 Within groups 1.2006 COPES 6 (Personal Problem Orientation). Between groups .5850 .428 .90 Within groups 1.3678 COPES 7 (Anger and Aggression)‘ Between groups .6190 .659 .72 Within groups .9389 COPES 8 (Order and Organization). Between groups 1.4892 1.249 .29 Within groups 1.1923 81 TABLE 4.1.--continued Variable Mean F F Prob- (COPES Subscale)‘ Squares Ratio ability COPES 9 (Program Clarity) Between groups 1.0913 1.159 .34 Within groups .9415 COPES 10 (Staff Control) Between groups 1.6414 2.178 .04* Within groups .7537 Significant at the .05 level. a Degrees Of Freedom Between Groups, 8. b Degrees of Freedom Within GrOups, 64. Univariate analysis of variance were performed to examine differences between residential settings for each subscale of the COPES. The significance level was set at .05. Differences in mean scores on subscale 2 (Support), subscale 4 (Autonomy), and subscale 10 (Staff Control) were significant (see Table 4.1). The null hypothesis of no difference in mean scores was, therefore, rejected in favor of the alternative hypothesis. Residential settings' sOcial environments did vary, as measured by the COPES on sub- scales 2, 4, and 10. 82 Hypothesis II: Differences Between Residential Settings' Program Characteristics Null Hypothesis: No difference will be found between resi- dential settings' program characteristics, as measured by staff report on the Resi- dential PrOgram Characteristics Scale (RPCS). Alternative Hypothesis: Residential settings' program char- acteristics will vary, as measured by staff report on the RPCS. Significant differences were found between residential settings' program characteristics for each of the three subscales on the RPCS (p<<.05). Mean squares, F ratios, and F probabilities are shown in Table 4.2. Means, standard deviations, and the number of observations per individual residential setting are found in Appendix J. Univariate analysis Of variance were performed to examine differences between residential settings for each subscale of the RPCS. The significance level was set at .05. Differences in mean scores were significant on all subscales (see Table 4.2). The null hypothesis of no difference between mean scores was therefore rejected in favor of the alternative hypothesis. Residential settings' program characteristics did vary as measured by the RPCS On all subscales. 83 TABLE 4.2.-— Summary Data for Analysis of Variance of Resi- dential Settings' PrOgram Characteristics (RPCS scores). Variable Mean F F Prob- (RPCS subscale) Squares Ratio ability RPCS (Structure) Between groups: 611.5010 38.572 <.o1 * Within groups 15.8535 RPCS (Responsibility/ Independence)_ Between groups 618.2311 75.089 <_.01 * Within groups 8.2333 RPCS (Supervision) Between groups 1035.3634 74.685 < .01 * Within groups 13.8630 * Significant at the .05 level. a Degrees of Freedom Between groups, 8. b Degrees of Freedom Within groups 41, 44, 38, respectively. 84 Hypothesis III: The Relationship Between Residents' Community Integration and Residential Settings' Social Climate Null Hypothesis: There will be no relationship between programs' "social climate," across residential setttings, as measured by the COPES, and resident integration into the COmmunity, as measured by the External Integration Scale (EI). Alternative Hypothesis: There will be a relationship be- tween programs' "social climate," across residential settings, as measured by the COPES, and resident integration into the community, as measured by the External Integration Scale (EI). Hypothesis III was tested by using a pairwise multiple regression equation with the ten COPES subscales as inde- pendent variables and the total EI scale score as the dependent variable. The results Of the regression analysis were significant for COPES subscales 3 (Spontaneity), 2 (Support), and 8 (Order and Organization). A summary Of the results is presented in Table 4.3. The null hypothesis was rejected as there is some relationship between subtest scores on the COPES and scores on the BI scale. 85 .cofluopomeoo MOM ucofloflmmsmca Ho>041oocmuoHoe no Ho>o41m .Ho>oa me. be nsmoamecoam . x..." xx. 9;. .3. .1. ACOHHZMHCORHO anooum He:0mnooc e mmmoo Fm. mom._ ems. NNF. meo.1 Acohmmonmoa can “waste e mmmoo mm. mom._ ems. awe. one. Assocousmc e mmmoo he. Nmm.e mee. mme. meo. Aspanmao seasonal m mmooo Ne. eme.e mee. mar. oeo. AcoHomocofluo Hmoenomnoc m mmooo mo. seem. oer. mme. emo.ulaonbcoo mombmc op mmmoo mo. mme.m mme. me_. see. Aucoso>ao>ch P mmooo « mo. owe.m oer. mme. m_N.1 AcceneNHcmouo one noonoc m mmmoo « mo. ome.m ewe. owe. meo. Aunooo5mc N mmmoo « mo. aoa.m mmo. one. wee. Axoeocmucoomc m mmooo ooam> m mo osam> e m moon no whom loamomhom mmmooc cocoOHwflomHm Haouo>o N nonnm cumccmum oHomHHo> ucoocomoccH .monoom mmmoo coo coflumumoucH Hmcuouxm coospom mflnmcofluoaom MOM canoe xnoEEom coammonmom oamfluaozli.m.v mamme 86 Hypothesis IV: Relationship Between Residents' Satisfaction and Residential Settings' Social Climate Null Hypothesis: There will be no relationship between programs' "social climate," acrOss resi- dential settings, as meausred by the COPES, and resident satisfaction with the resi- dential setting, as measured by the Con- sumer Response Scale (CRS). Alternative Hypothesis: There will be a relationship between . programs' "s0cial climate," acrOss residential settings, as measured by the COPES, and resident satisfaction with the residential setting, as measured by the Consumer Response Scale (CRS). A pairwise multiple regression equation with the ten subscales of the COPES as independent variables and the total score on the Consumer Response Scale as the dependent variable was used to test Hypothesis IV. The results Of the regression analysis were significant at the .05 level for all ten COPES subscales entered into the equation. However, COPES subscale 2 (Support) accounted for 21% of the total 38% Of the variance accounted for by all ten subscales combined. A summary of the analysis is presented in Table 4.4. The null hypothesis was rejected in favor of the al- ternative hypothesis. There does appear to be a relation- ship between subscale scores on the COPES and scores on the Consumer Response Scale, across residential settings. 87 .Ho>oe me. one be bemonmaceam , , Po. mwm.m Nam. Nae. eeo. ianflocmncOch m mmooo 4 Pony emo.e Nmm. Nee. owe. Assoconemv e mmooo . eo.uv mme.e Fem. e_e. emo. Aonumwhcmeuo 6cm hoonoc w mmaoo . Po.uv sem.m omm. ewe. omo.1 AcoHumucloo Emanoum Hegemnwov e mmaoo , Fo.v mem.e Gem. mme. meo. Acceomucoano Hmoeoomnmc m mmmoo « eo.v ope.e Fem. are. one. Aaonncoo ommumv ow mmooo . Po.v mm_.m amm. mme. . Pom. incoEo>Ho>ch F mmmoo , eo.v _me.a oem. Roe. ome.1 Aconmonmme ocm nomcac e mmooo . Po.v mmm.__ eem. ewe. ear. isnahaao Emumoumv m mmmoo , eo.u, omm.ee Fem. Gee. Fem. Aonooosmc N mmmoo osao> m mo osao> m m ouom mo opom .Aoaoomodm mmmoov OOCMOHMHcmHm Hamuo>o N Houum pudendum oaomfino> pcoocomoccH .1ilnnu .mouoom mmmoo wcm mwuoom omcommom noeomcou is... fiestas 2: as was sees 53...... 39:91.... 5...... 88 Hypothesis V: Relationship Between Residents' Community Integration and Residential Program Characteristics Null Hypothesis: There will be no relationship between program characteristics, across resi- dential settings, as measured by the Residential Program Characteristics Scale (RPCS), and resident integration into the community as measured by the ET scale. Alternative Hypothesis: There will be a relationship between program characteristics, across residential settings, as measured by the RPCS, and resi- dent integration into the COm- munity as measured by the El scale. Hypothesis V was tested by using a pairwise multiple rwegression equation with the three Residential Program (Ikaaracteristic Scale (RPCS) subscales as independent vari- aloles and the total External Integration (EI) Scale score as; the dependent variable. The results of the regression anealysis were significant (p‘<305) for two RPCS subscales, Ressponsibility/Independence and Structure. Results were not: significant for the third RPCS subscale, Supervision. A Slanmmry Of the results is presented in Table 4.5. The nul.3. hypothesis was rejected as there is a relationship betheen subtest scores on the RPCS and scores on the ET scale. 89 .Ho>oa mo. be nemoaoecmnm . no. emm.m «or. omm. mew. Asoflmfl>uomsmv momm * vo. Nvm.m nmo. omr. mme.1 onsuosnpmv momm a No. vem.m Fmo. Fem. mow. AmococcomooaH \sonaanamcoomomc moem ooHo> m mo moam> m m doom mo doom .Amoamomodm mommv oocmOHMHcmHm Hamuo>o Monnm cuoccopm oaooauo> unoccomoocH N .moaomfluouoouono Eonmonm Hoepcocamom can cofloonmoch aocumoxm coosuom mflnmcoflpoaom onu Mom oaooe wuoEEDm coflmmoumom oamfiuaozll.m.v mqmoH me. one be hemofleacmnm . . Fe. _m_.e eve. ems. eem.1 Aouenosunmc momm eo. oem.m meo. eom. mmo.P Acoama>nooemc momm mm. New. oeo. sew. Pam. AmocoocmooocH \xnaeannmcoomomc momm osem> m mo osaa> m we whom no memo Aoamomosm mommc oOQMOHMHcmHm Haono>o Mouum chocGMpm oHQoHno> pcopcomoccH .moaumfluouomnoco Eonmoum Hoflucocflmom can mouoom uncommom Hoasmcoo cooBuom mflzmcofiooaom can now canoe anoEESm COammonmom oamflpaszll.w.¢ mumme 92 The Relationship Between Residents' Characteristics and Their Integration into the Community The relationship between residents' characteristics and their score on the External Integration Scale was tested by using a pairwise multiple regression equation. The re- sults of the regression analysis were significant at the .05 level for seven characteristics of residents (see Table 4.7): education, age, number of psychiatric hospitalizations, months in current residential setting, sex, amount of spending money, and date of most recent hospitalization. There is a significant relationship between the characteristics of residents and their scores on External Integration. The Relationship Between Residents' Characteristics and Their Consumer Response Scores A pairwise multiple regression equation was used to 'test the relationship between residents characteristics arni their score on the Consumer Response Scale (CRS). Tile results of the regression analysis showed education tc> be significant at the .05 level (see Table 4.8). TTiere appears to be very little relationship between rtssidents' characteristics and their scores on the CRS. 93 .Hm teas marmcoapoaou o>Hoomoc 8 mm: came mcaom m .Ho>oa me. he nemonunemnm . me. mee.e «em. Fee. eme.u oOAnmNHHmonmom umnam mo. mnm.e New. ewe. Fwo.1 ooNfiHonHm loom mnucoz Houoe « mo. eo_.N oem. vow. mom. CanoNHHmonmom umoq « mo. emm.m omm._ mme. emo.1 hoooz unaccomm mo bosoem a No. ewm.m omm. oer. mee.1 oxom : Fo. one.m vmm. Pee. owe. oofluuom Hoaucocemom ocounno ca mnucoz a so. mmv.v mom. mow. com. mooflomNHHoon Imom mo Monsoz « mo. meo.w mme. For. emm.1 om< 4 mo. ome.m ewe. NMF. mmm. cofluoosom osHo> m m0 ooao> m m opom mo doom mocMOHmacmHm Haono>o m nounm choocoum oaooano> pooccomoocH .wpflcoeeoo one once cofiumuooch can moapmfluouomumso .mpcmwflmwm coozoom maerOHumaom one How OHQMB humEEsm :onmmummm mamauaszit.n.v mamme 94 .Ho>ma me. one on nemonmaemam , Ne. com. mme. eke. omo.1 mcoaumNHHmhad Imom mo HooEDZ mm. sea. eme. oee. mm_.u epoeNHHmnaomom nmuam we. eeo.e mme. ome. Nmo. xom em. ee~.. one. «om. see. was mm. eve.e GNP. ewe. Fe_.1 emuaamoao Imom mnpcoz Hopoe we. eme._ are. eme. moo. meannom Hmaucooemom #COHHDU CH mSHQOZ me. mem.e Poe. see. mme.1 mono: mcaocomm mo pcsoem we. mme.m omo. mme. arm. coaumuaampaomom ammo , .mo. eem.m see. Pee. me~.1 coaumosom osHo> m mo ooao> m m ouom mo whom mocooamecmam Haono>o N Honnm cnmocoum maooano> unoccommccH a! 3 .mouoom uncommom HoESmcoo can moaumfluouoonono .mucowflmom coosuom mflnmcoHpmHom map How canoe SHMEEsm :Oflmmmnmmm mHmHuH=2:I.m.w mumme ——n—~- —. 7 Overall Results on External Integration and Consumer Response Scales The dependent measures (External Integration Scale and Consumer Response Scale) both contained Liekert type response scales with one being the low point, three repre- senting the midpoint, and five being the high point. The mean total score for External Integration across residential settings was 129.042 with a maximum possible score of 220. The average response per question on the five point Liekert type scale was 2.9, or near the midpoint on the scale (see Table 4.9). The mean total score for Consumer Response across residential settings was 67.205 with a maximum possible score of 100. The average response per question on the iLiekert type scale was 3.36, or slightly beyond the mid- point on the scale (see Table 4.9) . Means and standard deviations for individual residential settings are found in Appendix K. Summary 'The first two hypotheses were tested to determine if reSidenrtial settings differed in their prOgrams (Residential PrOgraHICharacteristics Scale) and social environments (COmmunity-Oriented Program Environmental Scale). A uni- ‘Variate analysis of variance found the residential settings (aid differ at the .05 level of significance for both variables. 96 oow mm.m www.me mom.hm mmcommom Hoeomcoo omm Nmm.~ eom.em ~eo.mme sceumnooueH Hocnooxm ouoom oaoflmmom :oflbmoso Mom cOHpoH>oo ouoom EDEonz ouoom coo: pudendum coo: Hopoe oanmano> ll l| 1' I' .moaoom oncommom amasmcoo can coaponmoch Hmcuopxm co muasmom Haono>OIl.m.v mamme 97 The remaining four hypotheses were tested using mul- tiple regression analysis. The relationship between resi- dential settings' programs (RPCS) and residents' integration into the community (External Integration Scale) was examined, as was the relationship between residential settings programs (RPCS) and residents' satisfaction with their living environ— ment (Consumer Response Scale). This process was repeated by examining the relationship between residential settings' social environments (COPES) and residents' integration into the community (EI) and satisfaction with their living envi- ronment (CRS). The following is a summary of the results for each hypothesis test: 1. Hypothesis I asked if there were differences be- tween residential settings' social environments, as meas- ured by the Community-Oriented Program Environmental Scale (COPES). The null hypothesis was rejected at the .05 level On three of the ten COPES subscales (Support, Autonomy, and Staff Control). Residential settings did differ in their social environments. 2. Hypothesis II asked if residential settings dif— fered in program characteristics, as measured by the Resi- dential Program Characteristics Scale (RPCS). Significant differences (p‘<.05) were found between residential settings' program characteristics for all three of the RPCS subscales (Structure, Responsibility/Independence, and Supervision). The null hypothesis was rejected at the .05 level of signifi- cance. 98 3. Hypothesis III examined the relationship between programs' "SOcial climate," across residential settings, as measured by the COPES, and residents' integration into the community, as measured by the External Integration Scale (ET). The COPES subscales for Spontaneity, Support, and Order and Organization were found to be significant at the .05 level of significance and the null hypothesis was rejected. 4. Hypothesis IV looked for a relationship between programs' "social climate" (COPES), acrOss residential settings, and resident satisfaction with their living environment, as measured by the Consumer Response Scale (CRS). All ten subscales of the COPES were significant at the .OSlevel and the null hypothesis was rejected. 5. Hypothesis V asked if there was a relationship between program characteristics (RPCS), across residential programs, and resident's integration into the community (ET). The results were significant (p<<.05) for two RPCS subscales: Responsibility/Independence and Structure and the null hypothesis was rejected. 6. Hypothesis VI examined the relationship between the program characteristics (RPCS) of the residential set- ting and residents' satisfactiOn with living in their resi- dential setting (CRS). The results were significant at the .05 level for the program characteristic, Structure. This finding, however, appears to be a statistical artifact and the null hypothesis was not rejected. 99 Additional analysis examined the relationship between residents' characteristics and their integratiOn into the community and satisfactiOn with their environment. Only education was significantly related to satisfactiOn with the environment at the .05 level. Education, age, number of psychiatric hospitalizations, months in current resi- dential setting, sex, amount of spending money and date Of last hospitalization were all significantly related to integration into the community at the .05 level. Last, analysis of the overall integration Of residents into the community and their overall satisfactiOn with their living environment were performed. The average response (2.9) on a Liekert type five point scale for integration into the community was very near the midpoint on the scale (3.0), indicating residents "SOmetimes" participate in activities in the community and that they have "not much trouble“ arranging these activities. The average response (3.36) on a Liekert type five point scale for satisfaction with the living environment was slightly greater than the midpoint (3.0) on the scale, indicating residents find their environment to be a little better than "Okay,' "Somewhat satisfied," or "Adequate." In Chapter V a summary of the study will be presented. The findings will be discussed and conclusions presented. Limitations of the study and implications for future re- search will be discussed. 100 CHAPTER V SUMMARY In this chapter the study is summarized and conclu- sions based on the data analysis are explored. A dis- cussion of the results as well as the limitations of the study are included, along with suggestions for future re- search in the area. The purpose of this study was twofold: first to determine if residential treatment settings differed in their prOgram factors and social climate. Secondly, tO examine possible relationships between program factors and SOCial climates, and residents' integration into the community and satisfaction with their environment. The impetus for the study grew out of the need to examine residential treatment on the basis of two of the original goals of deinstitutionalization: reintegration into the community and humane treatment. Impetus was also provided by the scarcity of previous research attempting to iden- tify significant treatment factors and their relationship to the original goals of deinstitutionalization. Previous research examining residential treatment was reviewed across the following areas: (1) the relationship between residen- tial treatment settings' program characteristics and patient outCOme, and (2) the relationship between characteristics 101 of individuals served in residential treatment settings and patient outcome. With few exceptions (Segal & Aviram, 1978), the studies reviewed tended to label programs with terms such as "community treatment" without attempting to iden- tify significant prOgram, environmental or treatment charac- teristics. Seventy-eight residents of eight residential group homes and ten supervised apartments participated in the study. These residents completed questionnaires pertaining to the social environment of their present residential placement, their integration into the community and their satisfaction with the residential setting. Forty-three house staff and apartment supervisors completed a questionnaire examining program characteristics of the residential setting where they worked. Case managers/therapists completed a personal data sheet on each of their clients in the residential place- ments studied. The research design for this study was divided into two sections. The first section examined whether the two lindependent variables, program characteristics and social climate, differed between residential settings. The second section examined the relationship between these two inde- jpendent variables and the two dependent variables, resident's integration into the community and their satisfaction with the residential placement. In addition to the results of the formal hypotheses testing, analysis was performed to determine if resident's characteristics were related to the dependent variables. Lastly, the overall degree of 102 residents' integration into the community and satisfaction with their residential placement was reviewed. Results A univariate analysis of variance was used to deter- Inine if residential settings differed in their prOgram <2haracteristics and social climate. The significance level vvas set at .05. Significant differences were found On all tflnree of the subscales measuring prOgram characteristics (EStructure, Responsibility/Independence, and Supervision). Iflnree of ten subscales measuring social climate differed ssignificantly (Support, Autonomy and Staff Control). Multiple regression analysis were performed to test fkor a relationship between prOgram characteristics and resi- <fl£ents' integration into the community, as well as satis- Iftaction with the residential setting. Two program charac- ‘tearistics (Responsibility/Independence and Structure) were fOund to be significantly related to residents' integration irrt01the community at the .05 level. Responsibility/Indepen- which staff use measures to keep residents under necessary bentrols, i.e., in the formulation of rules, the scheduling ndemned building were it not for COmmunity residential PlEiC2ement. There can be no doubt that community integration and: consumer satisfaction are humane goals for residential treeEitment. The present research has attempted to identify Scnrue of the SOcial, environmental and program factors related to these goals in the hope that their modification ‘"i¥LJ. lead to the enhancement of community living for the chI'Onically mentally ill. 120 Limitations The use of non-random sampling and the absence Of a control group has an impact on the external validity of the present study. In addition, all subjects were volun- teers and generalization is restricted to the type Of subject volunteering. The sample was described in detail and a large proportion (81%) of the population agreed to participate, facilitating some generalizing of results. In addition to sampling limitations, there are pos- sible limitations in the measures used and the method Of data collection. The self-report instruments in the pre- sent study are accurate to the extent that such percep- tiOns are accurate and to the extent that the individual is willing to honestly express them. The contents of the iaistruments used in this study were not expected to be enfloarrassing, threatening, or sensitive to social desir- akiility. In addition, the effect of social desirability OII one instrument (COPES) has been previously studied arldi only a slight correlation was fOund. Finally, this study examined a limited number of ‘Véixriables. Other variables not considered in the present rEisearch, such as type of medication and the level admini- S1leered, individual personality factors, etc., could have ex) impact on the results. 121 Recommendations for Further Research With the deinstitutionalization movement has come the need to develop aftercare systems for the chronically men- tally ill in the community (Meyerson and Herman, 1983) . Re- search aimed at developing, evaluating, and improving these aftercare systems and residential services in particular, must be conscious Of several realities. First, concepts of "cure" and "discharge" need to be traded in for concepts connoting long term disabilities requiring possibly life- long supports (Test, 1981) . Humane treatment, patient satis- faction and maximum participation in society are reasonable goals and need to' be given more importance in research. Second, the deinstitutionalization movement has come to express and serve cultural values in our society. These values include that: "autonomy, choice, and interdependence are preferable to confinement, incompetence, and dependence; individual contentment is a worthwhile goal; one is best treated in the most natural setting with genuinely caring People; we all share sooial responsibility to include in our daily lives those who have special and sometimes negatively Vaerd differences" p. 118 (Estroff, 1981). It is important that research continue to study community treatment for its S“Cr-tess or lack of success at achieving the goals of dein- Stitutionalization. Outcome measures such as those used in the present study, community integration and satis- f‘E'Qtion with the living environment, must play a role in future residential treatment research. 122 Psychiatric indicators such as symptom remission and discharge rates have proven to be very poor predictors of community tenure or performance (Meyerson & Herman, 1983). There is an on-going need for studies examining the way services are delivered (SOCial climate, amount of struc- ture, etc.) as well as looking at what services are deliv- ered (halfway houses, day treatment, etc.) and what the outcome is (community integration, acquisition of skills, etc.) . A specific research question raised by the present study is; what is the impact of program characteristics (RPCS) on the social climate (COPES) of a residential The present research treated program charac- The setting. ‘teristics and social climate as separate variables. prTDgram characteristics were seen as the basic frame work (Iilles, etc.) or skeleton to a residential setting while the social climate was seen as the life (how people are reaJhly treated) and flesh of the setting. There can be littile doubt that these two variables influence one an- OthEBI: and in turn, effect the residents living in the Set“Z‘—:Lng. Lastly,there remains a need for rigorously designed reSEBEirch which utilizes random assignment at all levels of r'EESearch in residential treatment. 123 APPENDICES APPENDI X A APPENDIX A CONSENT FOR PARTICIPATION I freely consent to participate in the Residential Program Characteristics Study. I understand that the study's purpose is to examine the relationship between residential program characteristics and resident's satisfaction with their living environment and their participation in activities outside the residential program. I understand that all results will be treated with strict confidence and that my individual results will remain anonymous. I also understand that I am free to disc0ntinue Iny participation at any time. Finally, I understand that the overall results of the study will be made available to Ine upon request. Reusident or Staff Member Signature Date APPENDIX B ‘12. ‘13. 14:. APPENDIX B Personal Data Sheet Case # Residence: Age Sex Race Educational level completed Marital status Diagnosis Number of psychiatric hospitalizations Total length of time hospitalized ‘ (months) Year of first hospitalization Most recent hospitalization Number of months in present supervised home or apartment Amount of spending money available to client weekly Current day prOgram: None Job Volunteer Work School Vocational prOgram Day Treatment 125 APPENDIX C mNMBHQ >HHm5wo mmEHumEom wamumm um>wz mEoon mo mzflcmmao meacum>om mmasm mEHu p50 mucmaa no .mmEau omm mEHu camuumo a pm as pom umsE mpcmofimmm wmzm no mmzon pm mmHuH>Huom Hmaoom AhaxGGB ummma pmv omasomnom .Hmasmom ucmewmm mmum Ho mocm30aam cm m>Hmomu mucmoflmmm Emumwm muoc Emaboum monoco omcmflmmm “masmmm omumom mommmm ammuso HmmmHEmHU new mmasu unmao mmcflumme mmson.A%mem3 ummma Dav Umufisvmu .HMHzmmm mommafl>flum Umcumm buHB Emumoum mwum musuosuuw .Emumoum H50» muflm ummb 30H£3 mmcommmn may mcflaouao an xuoz 50> :oHQB CH mocmoflmmu map mbfluommp mmmmam mamom moaumaumuomumso Emumoum Hmflucmoammm U XHQZMQQAN .NP .FF .OF 126 mhm3a< awamsmo mmEHumEOm Namnmm H0>®Z U030HHM mum muouflma> >0:0§ :30 ma©:m: mu:mcflmmm cmuune:m: mu Honooa< Au:mE:HmuHmu:m .u:mu:mummu .umxumEMmQva poauumfip mmm:Hm:Q 0p mo:mumHU m:Hme3 AmHHE mo m:HH mun ummz A.oum .>#H>Huom amp .:0Hum0H©mEv mu:mEu:H0mmm :30 m:Hmmmx o:m m:Hm:muum How manflm:0mmmu mum mp:moammm m:0Ham0HUmE :30 mmfl>gmmsm mucmoflmmm m#:mowmmu 30: m0 m>0HQQM\u00Hmm mu:moflmmm mu:mcflmmu nonuo m0 Hmmmflsmflo m0 0>0HQQM\mm:0mHo mu:mpammm U030aam mumo mammE outdone ©:m >93 mucmowmmm om3oaam mumm 0m90: H0 m.u:maao mo:m©:mmmo:H\>#flHHQHm:0mmmm .vN .mm .NN .PN .om .me .mP .br .09 .me .ve .mw .HH Um::ap:oouuo xac:mmm4 127 m m w v m N m N mmm3H< NAHMSmD meHymEom mamumm AamflOOm .Hm:0apmmuomuv mmflua>fluom Umasomsom .umanmm: mua:mmuo mmmum mmuoso mmfl>ummsm mmmum mmm:fla:mmao How meoon mmfl>umm5m MMMDm mu:moflmmu 0p monoco :mflmmm mmmum 3mmuso mmfl>nmosm mmmum mommmm mma>ummsm mmmum mmauoooum How mmocm mmmum m::mE m:mHm mwmgm mu:moflmmu m0 mm:fluwme “madam: uo:U:00 mmmum >ua>fluom amp .mmmumnu .:0Hu00Home ..0.H .ma:mE nu:flommm ucmuuomfia muouH:OE mmmum moma mmfl>ummsm mmmum mcnmumum usosuvm :OHmH>uwmsm mmmum .mm .mm .vm .mm .Nm .em .om .mm .mm .mm .mm .mm .HHH U0::au:00nlo xflo:mmmd 128 APPENDIX D house or apartment you live in. APPENDIX D COMMUNITY-ORIENTED PROGRAMS ENVIRONMENTAL SCALE in the following way: (Form S) The remaining questions are True/False questions about the Please mark your answers True - Check the T if you think the statement is True or mostly True. False - Check the F if you think the statement is False or mostly False. Please be sure to answer every statement. Members put a lot of energy into what they do around here. The healthier members here help take care of the less healthy ones. Members tend to hide their feelings from one another. There is no membership government in this program. This program emphasizes training for new kinds of jobs. Members hardly ever discuss their sexual lives. It‘s hard to get people to argue around here. Members' activities are carefully planned. If a member breaks a rule, he knows what the consequences will be. Once a schedule is arranged for a member, the member must follow it. This is a lively place. Staff have relatively little time to encourage members. 129 True False Appendix D-—continued 13. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Members say anything they want to the staff. Members can leave here anytime with- out saying where they are going. There is relatively little emphasis on teaching members solutions to practical problems. Personal problems are openly talked about. Members often criticize or joke about the staff. This is a very well organized program. If a member's program is changed, staff always tell him why. The staff very rarely punish members by taking away their privileges. The members are proud of this program. Members seldom help each other. It is hard to tell how members are feeling here. Members are expected to take leader- ship here. Members are expected to make detailed, specific plans for the future. Members are rarely asked personal ques- tions by the staff. Members here rarely argue. The staff make sure that this place is always neat. Staff rarely give members a detailed explanation of what the program is about. Members who break the rules are punished for it. 130 True False Appendix D--continued 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. There is very little group spirit in this prOgram. Staff are very interested in following up members once they leave the program. Members are careful about what they say when staff are around. The staff tend to discourage criticism frOm members. There is relatively little discussion about exactly what members will be doing after they leave the program. Members are expected to share their personal problems with each other. Staff sometimes argue openly with each other. This place usually looks a little messy. The program rules are clearly understood by the members. If a member fights with another member, he will get into real trouble with the staff. True Reproduced by special permission of the Publisher, Consulting Psychologists Press, Inc., Palo Alto, CA from Community-Oriented Programs Environmental Scale Rudolf H. Moos copyright 1974. False 94306, bY Further reproduction is prohibited without the Publisher's consent . 131 APPENDIX E APPENDIX E EXTERNAL INTEGRATION SCALE This group of questions concerns activities outside the house or apartment where you live. Please circle the response which best describes your activities. Please answer every question. Very I. Often Often Sometimes Rarely Never 1. On a typical day do you go to a coffee shop or restaurant? 5 4 3 2 1 2. On a typical day do you go to the shopping center or local sh0pping area? 5 4 3 2 1 3. How often in a typical week do you order food from out- side or eat out at a local restaurant? 5 4 3 2 1 4. How often in a typ- ical week do you make a purchase at a local store? 5 4 3 2 1 A Half/ None Little Half Most All 5. On a typical day how much of your time between 8 a.m. and 5 p.m. is spent at the house? 5 4 3 2 1 6. On a typical day how much of your time between S p.m. and 11 p.m. do you spend at home? 5 4 3 2 1 132 Appendix E--continued Very Not Very Easy Easy Much Dif- Dif- II. Trouble ficult ficult If you have to arrange your own transportation, without the aid of (operator's name), or walk, how easy would it be to: 7. Go to a sh0pping center or a large shopping area: 5 4 3 2 1 8. Go to a park: 5 4 3 2 1 9. Go to a library: 5 4 3 2 1 10. Go to a movie: 5 4 3 2 1 11. Go to a community center: 5 4 3 2 1 12. G0 to a restaurant or coffee sh0p: 5 4 3 2 1 13. Go to a bar: 5 4 3 2 2 14. Go to a public_ transportation: 5 4 3 2 1 15. Go to the place of worship you prefer: 5 4 3 2 1 16. G0 to an organi- zation that offers individuals an Opportunity to do volunteer work: 5 4 3 2 1 17. Go to a barber shop or beauty parlor: 5 4 3 2 1 18- Take a walk in a pleasant area: 5 4 3 2 1 133 Appendix E--continued Not Very Very Much Dif- Dif- III. Easy Easy Trouble ficult ficult If you wanted, how easy would it be to obtain, outside this house or without the aid of (operator's name) the following things: 19. Meals 5 4 3 2 1 20. Medical care 5 4 3 2 1 21. Laundry services 5 4 3 2 1 22. Clothing 5 4 3 2 1 23. Toilet supplies and incidentals 5 4 3 2 1 24. A telephone 5 4 3 2 1 IV. How easy would it be, if you wanted to: 25. Telephone and just talk to a member of your immediate family: 5 4 3 2 1 26. Telephone and just talk to a more dis- tant relative: 5 4 3 2 1 27. Get together with a member of your imme- diate family: 5 4 3 2 1 28. Get together with a more distant rela- tive: 5 4 3 2 1 134 Appendix E--continued On a typical day how often do y0u visit with: 29. 30. v. Members of your immediate family More distant relatives How easy would it be, if you want to: 31. 32. 33. 34. Telephone and just talk to a close friend out- side the house Telephone and just talk to an acquaintance out- side the house Get together with a close friend not in this facility or another like it Get together with an acquaintance not in this facility or another like it Very Often Often Sometimes Rarely Never 5 4 3 2 1 5 4 3 2 1 Not Very Very Much Dif— Dif- Easy Easy Trouble ficult ficult 5 4 3 2 1 5 4 3 2 1 5 4 3 2 1 5 4 3 2 1 135 Appendix E--continued Very Often Often Sometimes Rarely Never On a typical day, how often do you: 35. Visit with close friends not in this house: 5 4 3 2 1 36. Visit with acquait- ances not in this house ' 5 4 3 2 1 VI. On a typical day, how often do you: - 37. Visit with close friends not in - this house 5 4 3 2 1 38. Visit with acquaint- ances not in this house 5 4 3 2 1 39. D0 volunteer work 5 4 3 2 1 40. Join in the activi- ties of social or political groups outside the house for people who are not considered for- mer patients 5 4 3 2 1 VII. On a typical day how often do you: 41 . Go to the park 5 4 3 2 1 42. G0 to the library 5 4 3 2 1 43 . Participate in some Outside sports ac- tivity 5 4 3 2 1 44 - Go to special sports or entertainment events 5 4 3 2 1 136 Appendix E--continued Reproduced by special permission of the Publisher, John Wiley and Sons, Inc., New York, NY 10158, from The Mentally Ill in Community—Based Sheltered Care: A Study of Community Care and Social IntegratiOn by Steven P. Segal and Uri Aviram copyright 1978. Further reproduction is prohibited without the Publisher's consent. 137 APPENDI X F APPENDIX F CONSUMER RESPONSE SCALEa The following questions concern the house or apartment in which you live. Please give your opinion. Please be sure to answer every question. 1. Do you find the living arrangements here to be: Very Good Good Adequate Poor Very Poor 5 4 3 2 1 For what you get, the amount you pay to live here is: A very good A bar- A fair Too Far too bargain gain amount Much Much 2 1 If you know of someone looking for a residential care home, would you recommend this place: Very With few With serious Not at Highly Highly Reservations Reservations All 5 4 3 2 1 Living here is: Comfor- Unc0m- Very Un- Very Comfor- table for- comfor- Comfortable table Enough table table 5 4 3 2 1 The food here is: Very Ade- Very Good Good quate Poor Poor 5 4 3 2 1 Are you bored here: N r Almost Occa- eve Never sionally Usually Always 5 4 2 1 The rules here are: Very Ade- Very Good Good quate Poor Poor 5 4 3 2 1 DO you feel that the appearance and cleanliness of the hOuse is: Very Ade- Very Good Good quate Poor Poor 5 4 3 2 1 138 Appendix F--continued 9. 10. 12. 14. The amount of privacy here is: Very Ade- Very Good Good quate Poor Poor 5 4 3 2 1 Do you feel that living here is: Very Somewhat Very Safe Safe Safe Unsafe Unsafe 5 4 3 2 1 Do you feel that living here is: Very helpful Help- Somewhat Not help- Very Un— to me ful helpful ful to me helpful to me 5 4 3 2 1 How satisfied are you with the amount of influence you have in what goes on in the house: Very Satis- Somewhat Dissat- Very Dis- Satisfied fied satisfied isfied satisfied 5 4 3 2 1 How satisfied are you with how much you are expected to participate in house activities and chores: Very Satis- Somewhat Dissat- Very Dis— Satisfied fied satisfied isfied satisfied 4 3 2 1 How satisfied are yOu with the amount of therapy or treatment you get: Very Satis- Somewhat Dissat- Very Dis- Satisfied fied satisfied isfied satisfied 4 3 2 1 Do you feel that the amount of recreational facilities and activities here are: Very Good Good Adequate Poor Very Poor 5 4 3 2 1 How satisfied are you with the number of close friends you have here: Very Satis- SOmewhat Dissat- Very Dis- Satisfied fied satisfied isfied satisfied 4 3 2 139 Appendix F--continued 17. About how often do you feel as though you want to move from here: Very Almost Often Often Occasionally Never Never 5 4 3 2 1 18. Do you feel that your needs are taken care of here: Very Very Well Well Okay Poorly Poorly 5 4 3 2 1 19. As far as doing what you want to do or say around here, do you feel that you are: Very Satis- Somewhat Dissat- Very Dis- Satisfied fied satisfied isfied satisfied 4 20. DO you feel safe on the street: Always Almost During the day Almost Always but not at night never Never 5 4 3 2 1 a Adapted for use with a five point Liekert type scale. Reproduced by special permission of the Publisher, John Wiley and Sons, Inc., New York, NY 10158 frOm The Mentally Ill in Community-Based Sheltered Care: A Study of Community Care and Social Integration by Steven P. Segal and Uri Aviram, copyright 1978. Further reproduction is prohibited without the Publisher's consent. 140 APPENDI X G APPENDIX G COPES SUBSCALE DEFINITIONS INVOLVEMENT measures h0w active members are in the day-to-day functioning of their program (spending time constructively, being enthusiastic, doing things on their own initiative). SUPPORT measures the extent to which members are en- couraged to be helpful and supportive toward other members and how supportive staff are toward members. SPONTANEITY measures the extent to which the program - encourages members to act openly and to express their feelings openly. AUTONOMY assesses how self-sufficient and independent members are encouraged to be in making decisiOns about their personal affairs (what they wear, where they go) and in their relationships with the staff. PRACTICAL ORIENTATION assesses the extent to which the member's environment orients him toward preparing him- selfforrelease from the program. Such things as training for new kinds of jobs, looking to the future, and setting and working toward goals are considered. PERSONAL PROBLEM ORIENTATION measures the extent to which members are encouraged to be concerned with their per- sonal problems and feelings and to seek to understand them. ANGER AND AGGRESSION measures the extent to which a member is allowed and encouraged to argue with members and staff, to become openly angry, and to display other aggressive behavior. ORDER AND ORGANIZATION measures the importance of order and organization in the program in terms of members (how do they look), staff (what they do to encourage order), and the house itself (how well it is kept). PROGRAM CLARITY measures the extent to which the member knows what to expect in the day-to-day routine of his program and the explicitness of the program rules and procedures. STAFF CONTROL assesses the extent to which the staff use measures to keep members under necessary controls (e.g., in the formulation of rules, the scheduling Of activi- ties, and in the relationships between members and staff). 141 APPENDIX G--continued Reproduced by special permission of the Publisher, Consulting Psychologists Press, Inc., Palo Alto, CA 94306, from COmmunity-Oriented Programs Environmental Scale by Rudolf H. Moos copyright 1974. Further reproduction is prohibited without the Publisher's consent. 142 APPENDIX H APPENDIX H Community-Oriented Program Environmental Scale Scoring Key Subscale Item Number Scoring Direction 1. Involvement 1 11 21 31 2. Support 2 12 22 32 HHJ'TJr-J "1861-3 3. Spontaneity 3 4. Autonomy 4 5. Practical Orientation 5 —h A "it-3W8 "JD-38"! "J'TJPJ'TJ 6. Personal Problem Orientation 6 7. Anger & Aggression 7 8. Order & Organization 8 N \l "JD-3'66 GEN-3'11 arm-3'11 143 Appendix H--continued Subscale Item Number Scoring DeviatiOn 9. Program Clarity 9 19 29 39 HWHi-l 10. Staff Control 10 to O HHWH Reproduced by special permission of the Publisher, Consulting Psychologists Press, Inc., Palo Alto, CA 94306, from Community-Oriented Programs Environmental Scale by Rudolf H. Moos copyright 1974. Further reproduction is prohibited without the Publisher's permission. 144 APPENDIX I Acoflumucoflno 3m.3 mm.3 43.3 mm. mm. mm. mm.3 mm. mo.3 .o.m Emanou: mm.3 me.3 oo.~ 06.3 oo.~ oo.3 33.3 04.3 oe.3 om.3 cams accomnmmc e mmmoo mm. Nm.3 00.3 mm. em.3 53.3 mo.3 43.3 33.3 .o.m lc03umucanuo me.3 om.3 ow.~ ee.3 mN.3 om.~ mm.3 oe.3 04.3 04.3 cams Hmonuomumc m mmmoo we. NN.3 mm. mm. mm. mo.3 me.3 wq.3 an. .o.m mo.m on.~ oo.m ev.m mm.3 ov.~ NN.3 oo.3 oe.3 om.m :mmz “>50:03:ao>ch 3 mmmoo m3 om m m 4 m m 03 m. 03 muowflazm mmmou mo quEsz mm mm m m G 3 33 33 3 m3 urcatwuam mo HODEOZ .muma .um mmCH: Hmnoe 6.:5m emu: mHm OCADDOm Hafiucwt3moz >b mouoom mmmoo How mumo NMOEESm H xHozmmm< 145 00.0 0k. 30.3 me. om. 03.3 03. 00. me. 00. 03.N 00.0 00.N 00.0 mk.N 00.N 33.0 00.0 00.N 00.3 00.0 00. mm. 03.3 mm. 00. 00.3 N0. 00. 0N.3 00.~ 03.0 00.N 00.N 0m.m 0N.m NN.0 00.~ 00.N 00.N 00.0 0N.3 03.3 40. 00.3 00.3 00. m~.3 mm. mm. Nm.~ mm.m 00.N 0N.m 00.3 00.3 03.N 00.~ 00.N 0N.N 00.0 mm. 00. 33. 30.3 03.3 00. 00.3 -.3 03.3 33.3 00.3 00.3 00.N 00.N 00.3 33.N 00.3 00.N 00.N .mne: .um mmcnm 30309 c.msm :03: mHmeo 650:60 .:.q.e meanmmmnsz “has: axons: uuonne :Hupmm Hmflu:mcammm .D.m :mmz .o.m :mmz .O.w :moz .D.m :mw: Aaonucoo uwmumv 03 mmmoo launumao Emnm0umv m mmmoo AcoHuMNfl:mmuo cc: umpuov m mmmOU Aconmmummm cc: uwm:mv n mmmoo waanum> COUCHUCOOIIH xHDzmmm< 146 APPENDIX J >0 topmHOOHMO :0:0 0:03 m:000©0>0p Onmncmum Ocm m:mwz .m:0000w Haeucopflmmh .m00m0>00m:m 0:0E0ummm p:m mwm0m omsoc M0 00 U000000:0Epm mm3 mamow Oeumauouomnmno Emumoum HmHucmpfimmm one x 00.0 00.0 03.0 00.3 30.0 03.0 00.0 00.0 00.0 00.00 00.03 00.00 00.00 00.00 30.00 33.00 00.00 00.00 00.30 00.3 00.3 00. 00.0 00.3 00.0 30.0 03.0 00.0 03.00 00.00 00.00 00.00 00.30 00.00 00.00 00.00 00.00 03.00 00.0 00.3 03.3 00.3 00.3 03.0 03.0 00.0 00.0 00.30 00.00 00.00 30.30 00.00 00.00 00.00 00.30 00.30 00.00 .0000 .00 0000: a0 mwuoom mumx 00% 00:0 xumEEom 000000 00000000000 h xHozmmm< .o.m :mmz :00m0>nmm:m mwmum .o.m :00: 00:0O:mmmO:H \3000000mcommmm .Q.m :moz 00:00:00m Ammamomndm mommo 0 mabmeum> 147 APPENDIX K .poosuonrmfl0: :0205050 0 :0 U000000 00 mHm mcfluuwm 0000:000000 >0 m0uoom Hm 0:0 mmo 00w mama >00EE:m M XHQmemm 148 APPENDIX L 0.3 0.0 0.0 0.0 0.0 0.0 0.0 3.0 0.0 0.0 .0030 00000 0000000 3.0 0.0 0.0 0.0 0.0 0.0 0.0 3.0 0.03 0.0 0O000000000 0.0 3.0 0.00 0.0 3.03 0.0 0.0 0.0 0.00 3.33 .000 0>000000< 0.03 3.00 0.00 0.00 0.0 0.03 0.03 0.0 0.03 0.0 00000000 .0000 0.00 0.00 0.00 0.00 3.00 3.30 0.00 0.30 0.00 0.33 0000000000000 30m0050ou0mv: mHmOCmMHQ 0.3 0.0 3.33 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0030002 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 000000: 0.00 3.Nm N.NN 0.0 3.03 0.0 0.0 m.m0 w.mm m.mm ©0000>0Q o.m3 n.33 3.00 o.oo3 m.mw 0.00 0.003 0.0m 0.33 3.00 00m00w Ammmuc0ou0mv 000000 0000002 0.3 0.0 3.33 0.0 0.0 0.0 0.0 0.0 0.0 0.0 00000000 3.~ 0.0 0.0 o.o~ 0.0 0.9 3.0 0.0 0.0 0.0 00:0000: m.3m m.om a.mm 0.00 o.oo3 3.mw 3.N3 0.003 3.0m a.mm anammosmo 0.0 3.0 0.0 0.00 0.0 0.03 0.03 0.0 0.03 3.33 00000 A0mmuc0000mv 0000 0.00 0.00 0.00 0.00 3.00 3.30 0.30 0.00 0.03 0.00 0002 0.30 0.00 3.00 0.00 0.00 0.00 0.03 0.00 3.00 3.00 000500 . A0mmuc0000mv x0m 0m 00 m 0 0 3 33 33 3 03 000000000 00 00QESZ .0004 .00 00000 00000 0.000 0000 00000 020000 .0.0.0 0000000002 0003: 000000 000000 Ucfluu0m Hmwucwcflm0m 0090000> 0000000 00000000000 30 00000000000005; 0.0c0wflm0m 00m 0002 >00EE:m q XHQmeL< 149 0:0000m 00000000000 0303 3303 0003 0003 3003 0003 0003 3003 0003 3303 000050 00000000000000: 0:0000 0002 00 000» 0303 0003 0303 0003 0003 0303 0303 0003 0003 0003 3:00EV 0000000000 Iammom 00000 00 000» 0.0 0.0 3.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 Acmwev 0:00000000000000 00000050>00 mo umnesz Acmmev. 0.33 0.03 0.03 0.33 0.03 0.03 0.03 0.03 3.33 0.33 00>0q 00000000000 0.00 3.mm 0.30 0.00 0.00 0.00 0.00 0.00 0.0m 0.30 00< 3 0 3 0 o 0 o 3 o o 0>on¢ 0:0 00:9 00oz 0 o o o o o o o o 0 00:00 00 0 0 m 0 3 0 0 0 0 . 0000000000 000 m3 0 o 0 0 0 0 3 o 3 00c00umoo> 0 o 3 o o o 3 0 o 0 000000 0 0 o 0 o 0 o o o 0 00000500> 0 0 0 o o 3 o o o 0 now 00 0 m o 0 0 o 3 m 30 0:02 30u0000000 no .020 E000000 >00 0000000 0.3 0.0 0.0 0.0 0.0 0.03 0.0 0.0 0.0 o.o .000 0005005004 0.0 0.0 0.0 0.00 0.03 0.03 0.00 0.0 0.0 0.0 0>0000mm¢ 000000 0.0.00000 0000:0000 . 00000 . 00 00000 0000.0 0.000 0000 00000 000003. 0.0.0 0000000003 0000: 0000000 000000 0030000> owneducoollq xHozm00< 150 m.o_ m.vm m.om 0.5 o.» m.op o.» o.m m.m~ F.n_ o.or m.oF _.h o.N m.m m.m m.mr m.FF o.oF m.m— «.mm F.F_ m.qP m.q m.n o.h n.wr F.mn o.wmp m.wm .mum< .um mmcflm Hmuoe u.m:m :mH: mHmmem3 mo HCSOEd AcmeV mCAuuwm ucmwmum :H mcucoz mo pmnESZ AcmmEv UwNHHmuHmmo: mnucoz Hmuoe wannaum> cwscflucoollq xHszmmd REFERENCES REFERENCES Anthony, W.A.; Buell, G.J.; Sharratt, 8.; Althoff, M.E. --Efficacy of psychiatric rehabilitation. Psychological Bulletin, 1972, Z§ (6), 447-456. Atkinson, R.M.--Current and emerging models of residential psychiatric treatment with special reference to the California situation. American Journal of Psychiatry, 1975, 123, 391-396. Bachrach, L.L.--A conceptual approach to deinstitution- alization. Hospital and Community Psychiatry, 1978 (29) 573-578. Baganz, P.C.; Smith, A.E.; Goldstein, R.; and Pou, N.K. --The YMCA as a halfway facility. Hospital and Com- munity Psychiatry, 1971, 3.__2_ (5), 156-159. Barton, R.--Institutional Neurosis. Bristol: John Wright and Sons, Ltd., 1966. Bassuk, E.L.; and Gerson, S.--Deinstitutionalization and mental health services. Scientific American, 238 (2): February, 1978. Bloom, J.--Whither approved homes? Analytical-descriptive study of the Parkland Region Approved Home Program. Canada's Mental Health, 1976, 23 (1), 37-40. Braun, P.; Kochansky, G.; Shapiro, R.; Grunburg, 5.; Gudeman, J.E.; Johnson, 8.; Shore, M.F.--Overview: Deinstitutionalization of psychiatric patients. American Journal of Psychiatry, 1981 (138), 736-749. Brown, G.W.--Experiences of discharged chronic schizo— phrenic patients in various types of living groups. Milbank Memorial Fund Quarterly, 1959, 21 105-131. Caffey, E.M.; Jones, R.D.; Diamond, L.S.; Burton, E.; and Bowen, W.T.--Brief hospital treatment of schizo- phrenia--Early results of a multiple hospital study. Hospital and Community Psychiatry, September, 1968, 282-287. 152 Carling, P.J.--Community residential services for the chronically disabled: An overview. HoriZOn House Institute, 555 Wissahickon Ave., Philadelphia, PA 19144. --Faculty Presentation at National Institute of Mental Health Staff College: "Beyond Deinstitutionalization: Developing Community-Based Mental Health Services." Rockville, MD (March, 1978) and Berkely, CA (June, 1978). Carpenter, M.D.--Residential placement for the chronic psychiatric patient: A review and evaluation of the literature. Schizophrenia Bulletin, 1978, 4 (3), 384-397. Cunningham, M.K.; Botwinik, W.; Dolson, J.; and Weickert, A.A.--Community placement of released mental patients: A five-year study. Social Work, 1969, 14 54-61. Datel, W.E.; Murphy, J.G.; and Pollack, P.L.--Outc0me in a deinstitutionalization program employing service inte- gration methodology. Journal of Operational Psychology, 1978, _9_' 6-24- Department of Mental Health and Mental Retardation Services, New York City. Reprot of the New York City Task Force on Community Residences. New York: Department of Mental Health and Mental Retardation Services, July, 1976. Drake, R.E.; Wallack, M.A.--Will mental patients stay in the community? A social psychological perspective. Journal of Consulting and Clinical Psychology, 1979 (47), 285-294. ' Erickson, R.C.; and Paige, A.B.--Fallacies in using length- of-stay and return rates as measure of success. Hospital and Commpnity Psychiatry, 1973 (24), 559-561. Estroff, S.E.--Psychiatric deinsitutionalization: A socio- cultural analysis. Journal of Social Issues, 1981, 37 (3), 116-131. Fairweather, G.W. (ed.)--Social psychology in treating mental illness. New York: Wiley and Sons, Inc., 1964. Giovannoni, J.M.; and Ullman, L.P.--Characteristics of family care homes. International Journal of SOcial Psychiatry, 1961, 1, 299-306. Goffman, E.--Asylums. Garden City, NY: Doubleday and Com- pany, Inc., 1961. Gove, J.; and Lubach, J.E.--An intensive treatment program for psychiatric inpatients: A descriptive evaluation. Journal of Health and Social Behavior, 1969, 12' 225-236. 153 Gruenberg, E.--The social breakdown syndrome--some origins. American Journal of Psychiatry, 1967, 123, 1481-1489. Gumrukcu, P.--The efficacy of a psychiatric halfway house: A 3-year study of a therapeutic residence. Sociological Quarterly, 1968, 9, 374-386. Hansell, N.; and Benson, M.L.--Interrupting long-term patient- hood: A cohort study. Archives of General Psychiatry, 1971, 24, 238-243. ‘ Hays, W.--Statistics for psychologists. New York: Holt, Rinehart and Winston, 1963. Herz, M.I., et al--Brief hospitalization: A two-year follow- up. American Journal of Psychiatry, 1977, 134, 502-507. Honigfeld, G.; and Gillis, R.--The role of institutionali— zation in the natural history of schizophrenia. Dis. Nerv. System, 1967, 38, 660-663. Hull, J.T. and Thompson, J.C.--Factors which contribute to normalization in residential facilities for the mentally ‘ ill. Community Mental Health Journal, 1981, ll (2), 107-113. Johnson, G.R.; Glick, I.D.; Young, R.--Length of stay of patients in sheltered care and physical space: A compara- tive study. Psychological Reports, 1980, 41, 703-708. Johnston, T.--Factors in the Rehabilitation of Ex-Mental Patients in Halfway Houses. (Doctoral Dissertation, California School of Professional PsycholOgY)( 1974. Lamb, H.R.-—Board-and-care home wanderers. Archives of General Psychiatry, 1980, 37, 135-137. Lamb, H.R.--The New Asylums in the Community. Archives of Generaly Psychiatr , 1979, 36. Lamb, H.R. and Goertzel, V.--Discharged mental patients-- are they really in the community? Archives of General Psychiatry, 1971, 24, 29-34. Lamb, H.R. and Goertzel, V.--High expectation of long-term ex-state hospital patients. American Journal of Psychiatry, 1972, 129 (4), 471-175. Lamb, H.R. and Goertzel, V.--The long-term patient in the area of COmmunity treatment. Archives of General Psychiatry, 1977, 34, 679-682. Langsley, D.G.; and Kaplan, D.M.--The treatment of families in crisis. New York: Greene and Stratton, 1968. Lee, D.T.--Family care: Selective and prediction. American Journal of Psychiatry, 1963, 120, 561-566. Linn, L.S.--State Hospital environment and rates of patient discharge. Archives of General Psychiatry, 1970, 2g, 345-351. Linn, M.W.; Caffey, E.M., Jr.; Klett, C.J.--Hospital vs. community (foster) care for psychiatric patients. Archives of General Psychiatry, 1977, ii, 78-83. Linn, M.W.; Klett, C.J.; and Caffey, E.M., Jr.--Foster home characteristics and psychiatric patient outcome. Archives of General Psychiatry, 1980, 31, 129-132. Ludwig, A.M.; and Farrelly, F.--The code of chronicity. Archives of Gemeral Psychiatry, 1966, 12, 562-563. Ludwig, A.M.; and Farrelly, F.--The weapons of insanity. American Journal of Psychotherapy, 1967, 21, 737-749. Lyle, C.M. and Trail, O.--A study of psychiatric patients in foster homes. Social Work, 1961, g, 82-88. Mendel, W.M.--On the abolition of the psychiatric hospital. Psychology Newsletter, 1968, 2. Mendel, W.M.; and Rapport, S.--Determinants of the decision for psychiatric hospitalization. Archives of General Psychiatry, 1969, 32, 321-328. Mendelsohn, I.--The family care program at Fort Logan Mental Health Center. Journal of the Fort Logan Mental Health Center, 1964, 3, 131-140. Meyerson, A.T. and Herman, G.S.--What's new in aftercare? A review of most recent literature. Hospital and Community Psychiatry, 1983, 34 (4), 333-342. Moos, R.--Community-Oriented Progrems Environment Scale Manual. Consulting Psychologists Press, Palo Alto, CA, 1974b. Moos, R.--Evaluating treatment environments: A social eco— logical approach. New York: Wiley and Sons, Inc., 1974c. Moos, R.--Ward Atmgsphere Scale Manual. Consulting Psychol- ogists Press, Palo Alto, CA, 1974a. Mosher, L.R. and Menn, A.Z.--Community residential treatment for schizophrenia: Two-year follow-up. Hospital and Com- munity Psychiatry, 1978, 32, 715-724. 155 Mosher, L.R.; Menn, A.; Matthews, S.M.--Soteria: Evaluation of a home-based treatment for schizophrenia. American Journal of Orthopsychiatry, 1975, £5 455-467. Murray, H.--Explorations in personality. New York: Oxford University Press, 1938. National Plan for Chronically Mentally Ill. Report to Secretary of Health. U.S. Department of Health and Human Services, Washington, D.C., 1980. Nevid, J.S.; Capurso, R.; Morrison, J.K.--Patient's adjust- ment to family--care as related to their perceptions of real-ideal differences in treatment environments. American Journal of Community Psychology, 1980, 8 (1), 117-120. Paul, G.L.--Chronic mental patient: Current status--future directions. Psychological Bulletin, 1969, 11, 81-94. Polak, P.R.; Kirby, M.W.--A model to replace psychiatric hospitals. Journal of Nervous and Mental Disorders, 1976, 162, 13-22. Rhine, M.W. and Mayerson, P.--Crisis hospitalization within a psychiatric emergency service. American Journal of Psychiatry, 1971, 127, 1386-1391. Rog, D.J.; Rausch, H.L.--The psychiatric halfway house: How is it measuring up? Community Mental Health Journal, 1975, ll (2), 155-161. ~Rotter, J.B.--Generalized expectations for internal versus external control of reinforcement. PsycholOgical Monographs, 1966, 82 (1, Whole Number 609). Rutman, I.D.--Preventing chrOnicity: A study of three alter- natives. Unpublished report, Horizon House, Philadelphia, October, 1971. Sandall, H.; Hawley, T.T.; and Gordon, G.C.--The St. Louis Community Homes Program: Graduate support for long-term care. American Journal of Psychiatry, 1975, 132 (6), 617-622. Segal, S.P.--Individual Characteristics Affecting the Shel- tered Care Needs of the Mentally Ill---paper presented at: Community Living Arrangements fOr the Mentally Ill and Dis- abled: Issues and options for public policy, a working conference sponsored by the National Institute of Mental Health, September 22-24, 1976. Published: Washington, D.C., USDHEW, 1976, in Conference Proceedings. Segal, S.P.; and Aviram, U.--The mentally ill in community- based care: A study of COmmunity care and social integra- tion. New York: Wiley and Sons, 1978. 156 Simon, S.H.; Heggestad, W.; and Hopkins, J.--Some factors relating to success and failure of male chronic schizo- phrenics On their first foster home placement. Community Meptal Health Journal, 1968, g (4), 314-318. Smith, C.J.; and Smith, C.A.--Evaluating outcome measures for deinstitutionalization programs. Social Work Research and Abstracts, 1979 (15), 23-30. Steinberg, T.; Yu, J.; Brenner, H.; and Kriegr, P.--Locus of control and community adjustment among schizophrenic patients. Journal of Clinical Psychology, 1974, 39 (1), 101. Stern, G.--People in context: Measuring person environment congruence in education and industry. New York: Wiley and Sons, 1970. Suchman, E.A.--Evaluative Research Principles and Practice in Public Service and Social Action Programs. New York: Russell Sage Foundation, 1967. Tsceng-Laroche, F.C.; Murphy, H.B.M.; and Engelsmann, F.-- The effect of foster home placement on psychiatric patients: Observed changes and contributory factors. Canada's Mental Health, 1976, 23 (1), 13-15. Test, M.S.--Effective community treatment of the chronically mentally ill: What is necessary? Journal of Social Issues, 1981, 37 (3), 71-86. Test, M.A. and Stein, L.I.--The clinical rationale for com- munity treatment: A review of the literature. In Alter- native to Mental Hospital Treatment. Edited by Stein, L.I. and Test, M.A. New York: Plenum, 1978. Towbin, A.P.--Self-care unit: Some lessons in institutional power. Journal of Consulting and Clinical Psychology, 1969, 33, 561-570. Ullman, L.P.; and Berkmen, U.S.--Efficacy of placement of neuropsychiatric patients in family care. Archives of General Psychiatry, 1959, 1, 273-274. Ware, B.L.--Demographic, bachground, and treatment factors influencing the rehabilitation of psychiatric patients. (Doctoral Dissertation, California School of Professional Psychology, 1979. Weinman, B. and Kleiner, R.J.--The impact of COmmunity living and community member intervention on the adjust- ment of the chronic psychotic patient, in Alternatives to Mental Hospital Treatment. Edited by Stein, L.I. and Test, M.A. New York: Plenum Press, 1978. Williams, D.H.; Bellis, E.C.; Wellington, S.W.--Deinstitu- tionalization and sooial policy: Historical perceptions and present dilemmas, American Journal of Orthopsychiatry, 1980, 22 (1), 54-64. Wing, J.K.--Institutionalism in mental hospitals. British Journal of Social and Clinical Psychology: 1962, 1, 38-51. Wing, J.K.; and Brown, G.W.--Social treatment of chronic schizophrenia: A comparative study of three hospitals. Journal of Mental Science, 1961, 107, 847-861. Wolfensburger, W.; and Glenn, L.--Prggram Analysis of Services System Field Manual. Toronto: National Institute on Mental Retardation, 1975. 158 MICHIGAN STATE UNIV. LIBRARIES WWIWIHWIHHIWIIWIH”NHHIIHHHIWI 31293106359692