- Email: [email protected]
Prevalence and correlates of family dysfunction and poor adjustment to chronic illness in specialty clinics
J CUE-ial
Vol. 43, No. 4, pp. 373-383, 1990 Printed in Chat Britain. All rights reserved
Copyright 0
0895-4356/90 $3.00 + 0.00 1990 Pergamon Prcaa pk
PREVALENCE AND CORRELATES OF FAMILY DYSFUNCTION AND POOR AI)JUSTMENT TO CHRONIC ILLNESS IN SPECIALTY CLINICS* KATHLEEN &PIN,‘?
MARGARET FITCH,‘** GINA BoI-& BROWNE””
and PAUL CORJZY~
‘Graduate Department of Nursing, Faculty of Nursing, University of Toronto, Toronto, Ontario, Toronto General Division, Toronto Hospital Corporation, Toronto, Ontario, ?khool of Nursing, Faculty of Health Sciences, McMaster University, Hamilton. Ontario, ‘Department of Clinical Epidemiology and Biostatistics, Faculty of Health Sciences, McMastcr University, Hamilton, Ontario and ‘Department of Preventive Medicine and Biostatistics, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada (Received in revised form 19 July 1989)
Abstract-This study estimated the prevalence of poor adjustment and family dysfunction among three chronically ill clinic patient populations and ass&d the biological, situation& social and psychological variables which most explained poor adjustment. Recently referred subjects were approached and 216 chronically ill subjects (from oncology, rheumatology and gastroenterology clinics) completed a Meaning of Illness Questionnaire, the M&aster Family Assessment Device, and the Psychosocial Adjustment to Illness Self-report Scale. In addition, information describing their biological, disease and socioeconomic stattis was obtained from the clinic record. Respondents were generally representative of other new referrals to the clinics (ineligible for the subsequent trial) in disease characteristics but uniformly.came from a more advantaged socioeconomic situation and were better adjusted to illness. Subjects from the three clinics were comparable on meaning, family function and adjustment variables. The proportion of subjects with family dysfunction was 30% and with poor adjustment to illness was 36%, high by community standards. Nevertheless, healthy family functioning and high levels of positive adjustment to chronic illness prevailed and were remarkably similar across clinic settings. Severity or type of disease was not related to adjustment outcomes nor to the level of observed disability. Rather, as hypothesized, meaning given the illness, followed by family function, and disability variables combined to explain 57% of the variance in adjustment outcomes. An intervention designed to improve family function and the meaning given illness was judged suitable. Family function
Psychosocial adjustment
INTRODUCTION
Most people with chronic illness adjust to their conditions without change in their mood *Collaborative project supported by the Ontario Ministry of Health, No. 01724, Ontario, Canada. ?A11 correspondence should be addressed to: Kathleen Arpin, Professor, Chair of Graduate Prograqmw, Faculty of Nursing, 50 St George Street, Tsronto, Ontario, Canada MS1 IA1 (Tel: 1-416-978-2848).
Chronic illness
Specialty clinics
or social roles [l]. However, a small but economically significant group of less adjusted chronically ill are concentrated in outpatient specialty clinics [2]. The majority of patients referred to these clinics suffer from co-existing chronic physical and psychosocial maladies [3]. The substantial body of information regarding adjkts$ment to chronic conditions is specific to cert.&n diseases. There is little comparative inform&ion about adjustment patterns across 373
I
374
KATHLEEN ARPINet
diseases [l]. Stress and social support literature suggest that adjustment to a chronic illness stressor is mediated by the attitude or meaning given the event and the quality of family support as a coping resource [4]. Less is known about how the meaning given illness and family function variables may combine with the situational disease factors to explain adjustment outcomes. These findings raised two questions: (1) Is adjustment to chronic illness related to the type and severity of the illness situation, or is it more closely related to cognitive appraisal (the meaning of illness) and family functioning; and (2) Are correlates of psychosocial adjustment similar across different chronic illnesses? Lazarus and Folkman [4] have proposed that the evaluation (of illness as a stress) consists of a primary appraisal of the relevance and seriousness of the event and a secondary appraisal of one’s coping resources to deal with it. An individual’s “commitments” regarding what is important [4, pp. 56621 and “beliefs regarding one’s ability to control the outcome” [4, pp. 65-661 were also considered important variables in adjustment. These authors’ central proposition that cognitive appraisal (which may be affected by the situation) leads to coping which “assists in maintaining social function, morale and somatic health” [4, p. 1821suggested the importance of the meaning ascribed to illness rather than its nature. There is accumulating research evidence that adjustment to chronic illness is independent of its type or severity. Psychological adaptation among non-cerebrally damaged chronically ill outpatients [5] is remarkably effective and fundamentally independent of specific disease such as cancer [l, 61, renal failure [7], or stroke [8]. For patients with lung disease [9, lo], renal disease [I 11,or burn trauma [12] adjustment has not been shown to be related to the severity of disease. The relationship between treatment status among the chronically ill and their adjustment outcomes is equivocal, however [13]. Cassileth et al. [6] found that psychosocial status of seven types of cancer patients closely correlated with treatment status. The psychological state was best for those on follow-up treatment, adequate for those in active treatment and worse for those receiving palliative treatment for cancer. Yet, in other studies that controlled for type of treatment, social support [14] and, in particular, sexual/affectional patterns in the marriage, were more closely related to adjustment among
al.
cancer patients than were prognosis or disability [ 151.These findings suggested that future studies of adaptation to chronic illness should continue to weigh the importance of the biological variables of disease severity, function level, and treatment status against other psychosocial variables. Psychosocial variables correlating with adjustment outcomes among patients with a variety of chronic diseases are life stress and social isolation [ 161, unfavourable self concepts, social relationships or attitudes toward illness [17,‘18]. A summary of the best literature on psychosocial correlates of adjustment to illlness revealed that social suport [13, 19-211 and cognitive variables [4,22,23] (which include the meaning given life events such as illness or treatment [2429]), combine to explain between 27 and 57% of the variation in a person’s eventual adjustment to a chronic illness [18]. Literature on social support linked family support to good adjustment. Among patients receiving chronic home dialysis [30], family or household members’ support [19] was more important than support provided by professionals in developing good vocational adaptation. Similar relationships were observed in cancer patients [6,31], in stroke patients [20], and in patients with chronic pain [32-341. Among cancer patients, family expressiveness with less conflict and more moral-religious emphasis was positively related to the patient’s adjustment [35]. The spouse’s cognitive interpretation of the patient’s chronic pain syndrome was closely related to the outcome of both partners [23]. The family variable was crucial to the rehabilitation outcome in adult arthritics, aphasics and other chronic disabling conditions [36]. No study could be located which simultaneously weighed the relative importance of a variety of disease variables against social and psychological variables of family function and meaning given illness in explaining adjustment to illness across a variety of diseases. In light of Lazarus and Folkmans’ [4] conceptualization of stress-coping it was hypothesized that neither the type nor severity of the disease stressor would be related to adjustment of the chronically ill. Rather, adjustment to the illness would be related to family functioning and the cognitive appraisal or meaning variables, and these relationships would be similar across different types of diseases. The baseline data from a larger study designed to test a clinical intervention were used to test these two hypotheses.
Family Function and Adjustment in Chronic Illness
315
Table 1. Procedure for obtaining sample ALL NEW REFERRALS Mixed cancer 241
I
Gastroenterology 409
(N = 1288) Rheumatology 638
ELIGIBLE SUBJECTS (A’= 520) Mixed cancer 118
Gastroenterology 171
I
INELIGIBLE: (N = 768) -not a chronic condition -unable to commute -retarded/cerebral damage --lives alone with no regular contact with a single person -non-English speaking patient or family member -already receiving counselling --25yr old or >80yr old -disallowed by attending clinicians
Rheumatology 231 DECLINED: (N = 302) -refused meetings (not needed) = 250 -refused meetings (needed) = 41 = 5 -refused quest
CONSENTING Mixed cancer 84
SUBJECTS (N = 218)
Gastroenterology 89
Rheumatology 45 DEATHS (N = 2)
BASELINE MEASURES (N = 216)
METHODS
Subjects were recruited from the Ambulatory Service Clinics of the Toronto General Hospital Division of the Toronto Hospital Corporation. These clinics/services are regional referral centres for a number of cancer, gastroenterological, and rheumatological chronic diseases. Table 1 outlines the procedure for obtaining the sample over a 9 month period from each of three clinics. From a total of 1288 new referrals, 768 subjects were ineligible for the subsequent trial; 286 of these either did not have a chronic condition or were not returning to the clinic, 213 others lived out of town, 89 lacked familiarity with English, 65 lived alone with no regular contact with others, 56 were disallowed eligibility by the attending physician or clinic nurse, 3 1 were already receiving counselling, 15 were ineligible due to age, 11 were either retarded or had cerebral damage, 1 was not a new referral, and data on 1 subject was missing. Of 520 eligible subjects, 302 declined participation, 2 died prior to the baseline measures, and 216 agreed to participate. 80% of those who declined participation were rated by the research nurses as well adjusted based on a global clinical index of adjustment, a judgement later corroborated by the baseline measures concerned with family function and psychosocial adjustment.
The project coordinator and research nurses administered the baseline bio-social-culturaldemographic inventory and the meaning, family function, adjustment questionnaires listed in Table 2 to consenting subjects at the time of their first appointment. The extensive methods of classifying the biological data and assessing their reliability and validity is reported elsewhere [37]. Chart data were abstracted as the measure of some of the biological variables. Other disability or treatment variables, such as severity of illness and/or functional ability using standardized classification systems, required the clinical judgement of the research nurses. The reliability of the nurses’ clinical judgement was compared with that of the physician on 18 subjects, and the degree of agreement is reported in Table 2 next to the variable and the classification system. The social and psychological explanatory variables were measured using instrumentation reported in Table 2. The Family Assessment Device [38] was chosen to assess family function after a careful review of nine other measures {39]. The reliability and validity of this measure have been documented [38,40] and are reported to classify clinicians’ judgements correctly regarding healthy and unhealthy families, with a sensitivity of 64% and a specificity of 67%. The reasons for developing the Meaning of Illness Questionnaire and the extensive analyses
376
KATHLEENARPINet
al.
Table 2. Biological, social, psychological variables, measures and reliability checks Construct Biological/disease variables
Variable -Clinician’s primary diagnosis -Months of time since diagnosis -Stage or severity of illness in cancer in rheumatoid disease in gastroenterological disease -Prognosis -Disfigurement -Functional
ability
-Additional illnesses before and since diagnosis -Disability related to combined diseases -Treatment status
Measure
Reliability validity
Chart abstraction Chart abstraction Chart abstraction using: [47, 48, 49, SO] -stage of cancer -curability 151, 521 -functional ability in rheumatoid disease [53, 54, 551 -modified severitv and leneth of gastroenterolbgical diseases Clinician’s judgement 1 = terminal to 7 = positive outcome and 5 year survival rate [47, 48, 49) in cancer Clinician’s judgement 1 = no disfigurement to 7 = severe (shocking) disfigurement Clinical judgement using Kamoskv Scale 1561 ECOG Scale [57j _ Chart abstraction [58]
-
Clinical judgement using Karnosky scale [56] Clinical judgement using 1 = active, 2 = follow up
Social variables
-Date of birth -Sex -Educational level -Marital status -Working status -Education of head of household -Combined family income -Country or culture -Occupation
Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Blishen (1976) socioeconomic status [59]
Social support variables
Family function
Psychological variables
Coping, cognitive appraisal of illness
McMaster Family Assessment Device (FAD) [38, 401 53 items on 4ooint scale -Communicaiion -Problem solving -Roles -Affective involvement -Affective responsiveness -Behaviour control -General functioning Meaning of Illness Questionnaire [41]
Outcome variable
Psychosocial adjustment
K = 0.96
K =
0.56
P,, = 83%
ICC = 0.68 ICC = 0.46 ICC = 0.68 K =0&t
Test/retest r = 0.66 to 0.76 Cronbach’s alpha = 0.74 0.75 0.72 0.78 0.83 0.72 0.92 K = 0.484.99
Factors: 33 items on 3- and ‘I-ooint scales I. Impact of Illness on Daily Living R2 = 0.46060 II. Type of Stress: (negative attitude of harm, loss, threat) and Functional Context III. Degree of Stress; Change in Commitments; Secondary Appraisal of Coping Resources IV. Positive of Illness as __ -_ .. Attitude Challenge; Hope, Motivation, Control V. Expectancy and Reoccurrence of Illness Psychosocial Adjustment to Convergent validity Illness Scale (PAIS) [42] Cronbach’s alpha = 45 items on 4-point scale -Health care orientation 0.63 0.81 -Vocational environment -Domestic environment 0.67 -Sexual relationships 0.80 -Extended familv relationshios 0.66 -Social environment _ 0.78 -Psychological distress 0.80 -Total adjustment score
K = Kappa; ICC = intraclass correlation coefficient; PO= observed agreement.
Family Function and Adjustment in Chronic Illness
to assess its reliability, factor structure, and validity have been previously reported [41]. In a sample of 320 subjects approached for participation in the present study, three meaning items explained (R2) 60% of the variance in adjustment. Three meaning factors explained 46% of the variance in adjustment. The “outcome” variable of interest was psychosocial adjustment to illness (PAIS-SR) [42], selected after a careful review of other measures of adjustment [43]. The PAIS-SR is a measure of change in seven dimensions of a person’s life as a result of the illness. The reliability and convergent and predictive validity of this measure are well documented [l 1,42,44], and the degree of agreement between the PAIS total score and other valid measures of attitude, mood, or symptomatology generally ranges between r = 0.60 and 0.81. A PAIS raw score of O-34 corresponds to clinical judgements of good adjustment; 35-53 indicates fair adjustment; and 54 or greater indicates very poor adjustment to illness [ll, 12,42,45]. RESULTS
In the analysis of results the representatiueness of participating subjects from each clinic is described in terms of sociodemographic and disease severity variables. Eligible subjects were compared with 68 ineligible subjects (largely from the gastroenterology and rheumatology clinics) who agreed to do part or all of the questionnaire. Participating subjects were compared with 36 persons who declined participation but agreed to complete some or all of the questionnaire. The impact of any biis due to those events on study results is highlighted. The comparability of subjects between each clinic was assessed and differences between the chronically ill groups are highlighted. Thereafter, the questions of prevalence and correlates of adjustment are addressed. Representativeness
Socioeconomic data describing the nature and severity of persons’ diseases, as in Table 2, were gathered from all new referrals. These data allowed comparisons between those eligible and ineligible and those who consented to and declined participation in the later trial. The details of this extensive comparison are reported elsewhere [37J. In summary, eligible consenting chronically ill subjects in all three clinics were uniformly a more socioeconomically advanCE43/4-E
311
taged group, slightly better adjusted with somewhat more favourable meanings given the illness. The effect of these differences would be to lower the prevalence of dysfunction in this sample. Comparability of participating clinics
subjects across
Subjects in all three clinics differed in their prognosis, age, education, marital status, occupational prestige, and family income. However, as shown in Table 3, subjects with different diseases did not differ in their baseline adjustment to illness, family function and majority of types of meaning given illness. Rheumatology subjects viewed their illness more negatively than others. Gastroenterology subjects had more expectation their illness would reoccur. Prevalence of dysfunction among the chronically ill
Estimates of prevalence of a disorder based on samples from tertiary centres are inflated compared to community samples due to the concentrating factor which accompanies the process of referral [2,3]. This fact should be kept in mind when interpreting the results. (i) Family dysfunction. The families from the three clinics in the present study were classified as healthy or unhealthy, using the cutoff points described [40] on each dimension of family functioning. This comparative prevalence of family dysfunction among the subjects in three specialty clinics is displayed in Table 4. In general, the prevalence of family dysfunction (in every dimension) across three chronically ill groups was remarkably similar (x2 = 1.18, p = 0.55) and was in general 30% of the sample (inflated compared to the community). Noteworthy among the dimensions of family function is the high prevalence of unhealthy communication (44%), poor involvement (44%), and problems in behaviour control (44%). (ii) Adjustment. Kruskal-Wallis was used to test differences in the levels of good, fair, and poor adjustment across the three clinics. No important differences were found in the number of subjects with various levels of adjustment to illness across the three clinics. As in Table 5, the vast majority (approximately two-thirds) of the chronically ill adjusted to their illness with little change in their life. Only a small proportion of the chronically ill exhibited poor adjustment to illness or major changes due to illness. There
378
KATHLEEN ARPIN et
al.
was a middle group of subjects who reported a fair to moderate degree of change in their life (adjustment) attributable to the illness. In all three clinic samples, these same patterns were observed. The combined prevalence of fair to poor adjustment of 36% again reflected the concentration of these disorders in outpatient specialty clinics. Nevertheless, was there any relationship between the prevalence of poor adjustment and that of family dysfunction? The correlations between the dimensions of family function and adjustment or psychological distress ranged between 0.18 and 0.38. The weak correlations indicate that something other than family dysfunction must explain poor adjustment to illness or psychological distress. Consequently, all baseline data including type of disease (clinic) were used to assess the combination of biological, disease, socioeconomic, family function, or meaning given illness variables which most explained the variation in subjects’ adjustment to illness. Baseline illness
correlates
of adjustment
to chronic
From stepwise regression analyses performed on the adjustment to illness variable within each clinic it was consistently found that the two meaning given illness factors, namely, the “negative attitude” (Factor I) and the “adverse impact” (Factor II), as well as some aspect of family dysfunction always entered the model before any of the biological and disease variables. Although no overall formal test of interaction of all of these effects across clinics was performed, their partial regression coefficients were similar. For example, the coefficients associated with the meaning of illness of Factor I were not significantly different across the clinics. The remaining analyses were performed on the combined dataset with the clinic factor in the model. A formal test of the effect that the meaning given illness and family function variables had in excess of the contribution of disease severity and disability was carried out. The partial correlation coefficient associated with the cognitive and family function variables adjusted for the disease and severity variables was 0.57 (p < 0.0000001) whereas the partial correlation coefficient associated with the disease severity and family function variables adjusted for the cognitive and family function variables was 0.30 (p < 0.0001). That is, the meaning of illness and
319
Family Function and Adjustment in Chronic Illness Table 4. The percentage of subje&s with di&ent types of fatily dysfunction in three samples of chronicallv ill
Oncology Rheumatology Gastroenterology p-Value N=45 N = 88 N = 80
Dimensions of
familyfunction
15% 42% 17% 41% 31% 51% 34%
Problem solving Communication Roles Affective involvement Affective responsiveness Behaviour control General family functioning
18% 40% 27% 40% 29% 36% 24%
0.85 0.14 0.28 0.60 0.96 0.19 0.55
18% 47% 21% 48% 31% 41% 31%
p = probability associated with the chi-square test of three independent proportions. family function variables exceeded the importance of disease severity and disability in explaining adjustment to illness. Severity of disease alone explained only 7% of the variation in adjustment to illness. Thereafter, forward stepwise multiple regression analyses using free entry and listwise deletion were conducted, as in Table 6, in search of the most important combination of variables listed in Table 2 which explained adjustment to chronic illness. The socioeconomic variables of educational level, marital and working status, occupational prestige and education of the head of household, and ethnic origins were entered as dummy variables, along with clinic, to represent the type of disease. Although each of the socioeconomic variables were interrelated, none were important in explaining a person’s adjustment. The discrete biological variables entered along with clinic into a second equation were: months of time since diagnosis, age, stage (severity) of illness, the Karnosky and ECOG approaches to measuring disability, the clinician’s assessment of disfigurement, prognosis, and treatment status. Only disability and little disfigurement together explained 32% of the variance in patients’ poor adjustment to illness [F: (3,212) = 44.47, p < O.OOOOl]. Disability or function level, however, was only
weakly related to the biological severity or stage of illness (r = 0.31) and unrelated to disfigurement (I = -0.08). Further, the more definitive measure of severity of illness such as “stage” of illness tias weakly related to adjustment (r = 0.25). Consequently, disability may serve as a marker for some other psychosocial variable (associated with disability) which would explain adjustment to illness. The seven dimensions of family functioning were entered, along with clinic, into a third equation, and only the dimension of poor role enactment in the family entered to explain 15% of the variance in adjustment [F: (1,211) = 36.15, p < O.OOOOl].As before, none of the dimensions of family function correlated very strongly with adjustment. The five meaning of illness factors, three statistically significant coping behaviour items, eight categories of previous and current commitments (1 =present; 0 = absent) and realignment in commitments (0 = no change; 1 = change) were entered into a fourth regression equation along with clinic. The negative views of illness as stress (harm, loss, threat), poor view of functional status (disability, deterioration, disfigurement) and concept of illness as “adversely affecting daily living” together explained 42% of the variance in a person’s poor adjustment to illness. Other
Table 5. The percentage of subjects with problems in adjusting to illness in oncology, rheumatology and gastroenterology tertiary clinics Clinic: PASSR Raw-score Adjustment: Good o-34 Fair 35-53 Poor 54+ Total
Oncology N = 83
Rheumatology N=45
Gastroenterology N = 88
Total 216
10%
62%
60%
64%
24%
29%
21%
26%
6%
9%
13%
10%
(100%) (100%) (100%) (Kruskal-Wallis y* = 2.26, p = 0.32)
(100%)
KATHLEEN fiRPIN et al.
380
Table 6. Biological, social, psychological and clinic variables which most explain poor adjustment among the chronically III (multiple regression with listwise deletion) Construct Biological and socioeconomic N =216
Social support Psychological (meaning given illness) N=189
Summary equation of biological, social psychological variables N = 188
Variable
/l*
(a) Socioeconomic (b) Biological/disease -Disability (ECOG) -Disability (Karnosky) -Disfigurement Family functioning -Roles Type of stress/illness (negative attitude) of illness as harm, loss, threat and functional contextdisability, deterioration, disfigurement Impact of illness on daily living Feel you need to know more before you can act Realigned commitments Optimistic attitude of illness as challenge; hope, motivation, control Meaning: Factor II Type of stressfulness (negative attitude) of harm, loss, threat, and functional context Meaning: Factor I Impact of illness on living Genera1 family functioning Disability (ECOG) Disfigurement Meaning: Realigned commitments Treatment status Meaning: Feel you need to know more
R2
p-Value?
None entered 5.80 -0.22 - 1.50
0.27 0.30 0.32
0.00 0.00 0.01
8.52
0.15
0.00
0.56 0.42 0.70 2.56
0.32 0.42 0.45 0.46
0.00 0.00 0.00 0.01
0.48
0.02
0.53
0.32
0.00
0.33 3.96 4.18 -1.38
0.43 0.47 0.51 0.53
0.00 0.00 0.00 0.01
2.49 2.11
0.55 0.56
0.01 0.03
-0.29
0.45 0.57 0.05 [F: (8,179) = 29.66, p < O.OOOOl]
*Regression coefficients in the final model. tp-Values associated with B in the final model.
meaning of illness items and factors (realigned little optimism) entered to commitments, explain 48% of the variance in a person’s poor adjustment to illness [F: (5,183) = 33.52, p < O.OOOOl]. The negative view of the illness as stressful and associated with poor function (Factor II) was moderately correlated with the clinician’s rating of disability (ECOG) where (r = 0.44). However, as with the disability assessment, there was only a weak relationship between the view of the illness as disabling and the objective reality regarding the stage or severity of illness (r = 0.23). The clinician may have observed illness behaviour more than disability related to disease. All of the important biological, social, and psychological (meaning) variables were entered, along with clinic as a dummy variable, into a summary regression equation (Table 6) in search of the combination of factors which most explained a person’s adjustment to chronic illness. The meaning variables of “the view of illness as a harm, loss, threat associated with disability, deterioration, and disfigurement” and the view of the “illness (adversely) affecting
their daily living” together with poor family function in general, entered to explain 47% of the variance in a person’s poor adjustment to chronic illness. These variables, coupled with the clinician’s rating of disability, little disfigurement, and additional meaning and treatment variables, collectively explained 57% of the variance in a person’s adjustment [I;: (8,179) = 29.66, p c O.OOOOl].This pattern of the meaning of illness variables exceeding the importance of disease severity or disability variables was observed in every clinic. In general, the cognitive appraisal or meaning variables, followed by family function variables, had the most explanatory power. The biological variables of observed disability and treatment status together although statistically significant, add little (6%) in explaining adjustment of the chronically ill when weighed against the other psychosocial variables. The weak correlations between the meaning of illness factors and dimensions of family function (r ranges between 0.02 and 0.24) indicate that these two variables are relatively independent and make a separate contribution to living with chronic illness.
Family Function and Adjustment in Chronic Illness
One measurement bias in the measure of adjustment to illness is in the direction of the scoring. In this PAIS measure, it is only possible to have no change = 0 or change (deterioration) = l-3. Consequently, it fails to detect improvements in various dimensions .of people’s lives since their illness which have been observed. A one-question item regarding a person’s improvement since the illness from 0 = none to 4 = a great deal in each of the vocational, domestic, sexual, social, extended family, psychological distress components of the adjustment scale were added at the end of that respective section of the PAIS. A total improvement in adjustment score was derived as the sum of each source of improvement. Again, the chronically ill from the three clinics were remarkably similar in their report of life improvements since the illness. More than 60% of subjects reported fair to substantial improvement in areas of their life since the illness. Similar separate and summary regression analyses were conducted as in the previous analyses in search of factors associated with this improvement. The cognitive appraisal or meaning variables again exceeded the importance of all others and explained 14% of a person’s life improvement since the illness [F: (2,186) = 14.88, p < O.OOOOl]. When weighed with all other variables in Table 2, “the degree of stress, change in commitments and appraisal of (sufficient) coping resources” coupled with an “optimistic attitude of illness as a challenge: hope, motivation and control” entered first to explain 14% of the improvements in adjustment to illness. These, together with lower levels of education and greater periods of time since diagnosis, explained 18% of the variance in patient’s improvement in adjustment to illness [F: (4,183) = 10.09, p < O.OOOOl]. Summary
In these specialty clinics, the chronically ill are remarkably similar in their levels of adjustment, family function. and meaning given illness. Thirty percent of respondents had unhealthy patterns of family function; 36% had fair to poor levels of adjustment to illness. The respondents represented the more advantaged socioeconomic group of all new referrals to three specialty outpatient clinics; however, neither the socioeconomic nor biological disease severity situation variables were important in explaining a person’s adjustment to illness. Rather, the
381
unfavourable meanings of “harm, loss, threat” and “disabling, deteriorating, disfiguring”, “(adverse) impact of illness on total life”, “realigned commitments”, and “poor family function” collectively explained 57% of the variance in poor adjustment to illness. Similar negative meaning or unhealthy family communication patterns explained 31% of the variance in a person’s level of psychological distress. By contrast, approximately 60% of subjects reported moderate to sign&ant improvements in their lives since their diagnosis. The more favourable meaning given the stressfulness of illness of “change in commitments, availability of coping resources”, “view of (illness as) a challenge, coupled with a sense of hope, motivation, control” among the less educated and less recently diagnosed subjects explained 18% of the variance in their improvement since the illness (with little contribution from the biological, disease, or family function variables). Discussion and implications
Although a cross-sectional design cannot determine the direction of causality, the results of this study pointed to the importance of the cognitive appraisal variable in explaining adjustment to the stress of illness. They support the central propositions of Lazarus and Folkman that the cognitive appraisal aspect of “coping assists in maintaining social function, morale and somatic health” [4, p. 1821. This finding, coupled with the high prevalence of poor family functioning and adjustment among the chronically ill in specialty clinics, served as sufficient justification for the subsequent trial to assess the “Efficacy of Family Clinic Appointments in Promoting Adjustment to Chronic Illness” [371. These findings also support the clinical efforts to assist patients and/or their families who have unfavourable views of their illness, with a cognitive reappraisal of their situation so that they may achieve a more optimistic, hopeful and satisfied outlook. Cognitive reappraisal could be emphasized in individual, family or group orientations to treatment. Since there was little relationship between disease severity and meaning given the illness, negative or unfavourable attitudes toward illness should be used rather than disease severity for identifying subjects in need of referral to psychosocial services. Conversely, subjects with a favourable and hopeful outlook, regardless of the incongruence with the
382
KATHLEEN
severity of their disease, should be encouraged to maintain their bases of hope. In the family approaches to treatment, the effective coping behaviours of problem-solving and communication should be emphasized. Cognitive behavioural orientations to family treatment would also emphasize assisting families to achieve a more optimistic interpretation of members’ behaviour. Treatment frequently appears to facilitate changed attitudes or more optimistic meanings attributed to familiar patterns of behaviour in families rather than real differences in the pattern of behaving per se. Future research could test this proposition. Since a relationship was found between both the meaning of illness and family variables and adjustment to illness, our future research will test the efficacy of family meetings in promoting more favourable meaning given illness, family function and, ultimately, adjustment to illness. If efficacious, the cost and benefit of this service, designed to augment conventional medical outpatient care, will be assessed [37].
REFERENCES 1.
2.
3.
4. 5.
6. 7.
8.
9.
10.
Cassileth BR, Lusk EJ, Strouse TB, Miller DS, Brown LL, Cross PA, Tenaglia AN. Psychosocial status in chronic illness. N Engi J Med 1984; 311: 5066511. Spaulding WB. The psychosomatic approach to the practice of medicine. In: Lipowski ZJ et al., Eds. Psychosomatic Medicine. New York: Oxford University Press; 1977: 457-469. Eastwood MR. Epidemiological studies in psychosomatic medicine. In: Lipowski Z J et al., Eds. Psychosomatic Medicine. New York: Oxford University Press; 1977: 41 I-420. Lazarus RS, Folkman S. Stress, Appraisal and Coping. New York: Springer; 1984. Davies RK, Quinlan DM, McKegney FP, Kimball CP. Organic factors and psychological adjustment in advanced cancer patients. Psychosom Med 1973; 35: 464-47 1. Cassileth BR, Lusk EJ, Strouse TB, Miller DS, Brown LL, Cross PA. A psychological analysis of cancer patients and their next of kin. Cancer 1985; 55: 72-76. Livesley WJ. Symptoms of anxiety and depression in patients undergoing chronic haemodialysis. J Psychosom Rcs 1982; 26: 581-584. Bishop DS, Epstein NB, Keitner GI, Miller IW, Srinivasan SV. Stroke: Morale family functioning, health status and functional capacity. Arch Phys Med RehabB 1986; 67: 84-87. Dudley DL, Glaser EM, Jorgenson BN, Logan DL. Psychosocial concommitants to rehabilitation in chronic obstructive pulmonary disease: Part 2. Psychosocial treatment. Chest 1980: 77: 554-561. Judd L, Browne G, Craig H, Norman G, Guyatt G. The prevalence and correlates of poor psychosocial adjustment in patients with chronic airflow limitation. In: King K, Ed. NursIag Research: Science for Quality Care. Proc 10th National Nursing Research Conference. University of Toronto; 1986: 109-I 12.
ARPIN et
al.
11. DeNour AK. Psychosocial Adjustment to Illness Scale (PAIS): A study of chronic hemodialysis patients. J Psychosom Res 1928; 26(l): 11-12. 12. Browne G, Byrne C, Brown B. Psychosocial prognosis of burn survivors. Burns 1985; 12: 28-35. 13. Goldberg RJ, Cullen LO. Factors important to the psychosocial adjustment to cancer: A review of the evidence. Sot Sci Med 1985; 20: 803-807. 14. Revenson TA, Wollman CA, Felton BJ. Social supports as stress buffers for adult cancer patients. Psvchosom Med 1983: 45: 321-331. 15. Taylor SE, Lichman’ RR, Wood JV, Bluming HZ, Dosik GM, Leibowitz RL. Illness related and treatment related factors in psychological adjustment to breast cancer. Cancer 1985; 55: 2506-2513. 16. Lewis FM, Bloom JR. Psychosocial adiustment to breast cancer: A review of selected literature. Int J Psvchiat Med 1978: 9(l): 1-17. 17. Hyman MD. Social psychological factors affecting disability among ambulatory patients. J Cbron Dis 1975; 28: 199-216. 18. Counte MA, Bieliauskas LA, Pavlou M. Personal stress and personal attitudes in chronic illness. Arch Phys Med Rehabil 1983; 64: 272-275. 19. Broadhead WE, Kaplan BH, James SA, Wagner EH, Schoenbach VJ, Grimson R, Neyden S, Tibblin G, Gehlbach S. The epidemiologic evidence for a relationship between social support and health. Am J Epidemiol 1983; 117: 525-541. 20. Evans RL, Northwood LK. Social support needs in adjustment to stroke. Arch Phys Med Rehabil 1983;64: 61-64. 21. McQueen DV, Siegnist J. Social factors in the etiology of chronic disease: An overview. SIX Sci Med 1982; 16: 353-367. 22. Viney LL, Westbrook MT. Coping with chronic illness: Strategy preferences, changes in preferences and associated emotional reactions. J Chron Dis 1984; 37: 489-502. 23. Block AR, Bayer SL. The spouse’s adjustment to chronic pain: cognitive and emotional factors. Sot Sci Med 1984; 19: 1313-1317. 24. Streiner DL. Norman GR. McFarlane AH. Rov RG. Quality of life events and their relationship to itrain. Schizophrenic Bull 1981; 7(l): 34-42. 25. Schroeder DH, Costa PT. Influence of life event stress on physical illness: Substantial effects or methodological flaws? J Pers Sot Psycho1 1984; 46: 853-863. 26. McFarlane AH, Norman GR, Streiner DL, Roy RG. The process of social stress: Stable, reciprocal and mediating relationships. J Health Sot Behav 1983; 24: 160-173. 27. McFarlane A, Norman G, Streiner D, Roy R, Scott D. Longitudinal study of the influence of the psychosocial environment on health states: A preliminary report. J Health Sot Behav 1980; 21: 124-133. 28. Moos RH, Billings AG. Conceptualizing and measuring coping resources and processes. In: Goldberger L, Breznitz S, Eds. Handbook of Stress: Theoretical and Clinical Aspects. New York: The Free Press; 1982: 2 12-230. 29. Moos RH, Cronkite RC, Billings AG, Finney JW. Index of Life Change Events. In: HeaItb and Daily Living Form Manual. Palo Alto, California: Social Ecology Laboratory, Veterans Administration and Standard University Medical Center; 1984. 30. Palmer S, Canzona L. Conley J, Wells G. Vocational adaptation of patients on home dialysis: Its relationship to personality, activities and support received. J Psychosom Res 1983; 27(3): 201-207.
Family Function and Adjustment in Chronic Illness 31.
32.
33. 34. 35. 36.
31.
38. 39. 40.
41.
42.
43. 44.
Baider L, DeNour AK. Couples’ reactions and adjustment to mastectomy. Int J Psychiat Med 1984; M(3): 265-270. Shanfield SB, Heiman EM, Cape DN, Jones JR. Pain and the marital relationship: Psychiatric distress. Pain 1979; 7: 343-351. Block AR. An investigation of the response of the spouse to chronic pain behaviour. Psychosem Med 1981; 43: 415-423. Flor H, Turk DC, Scholz OB. Impact of chronic pain on the spouse: Marital, emotional and physical consequences. J Psychosom Ree 1987; 31(l): 63-71. Spiegel D, Bloom JR, Gollheil E. Family environment as a predictor of adjustment to metastatic cancer. J Psycheaee Oneel 1983; l(1): 33-44. Bishop DS, Epstein NB. Family problems and disability. In: Behioural Problems and the Disabled. Baltimore: Williams and Williams; 1980: 387-364. Arpin K, Fitch M, Browne G, Corey P. Efficacy of family clinic appointments in promoting adjustment to chronic illness: Methodological manual and appendices. Unpublished manual, Ontario Ministry of Health Project No. 01724, University of Toronto, Faculty of Nursing, Toronto, 1988. Epstein N, Baldwin L, Bishop D. The McMaster Familv Assessment Device. J Marital Fam Ther 1983: 9(2): i71-180. Speer J, Sachs B. Selecting the appropriate family assessment tool. Pediatric Nursing 1985; 11: 349-355. Miller I, Bishop D, Epstein N, Keitner G. The McMaster Family Assessment Device: Reliability and validity. J Marital Fam Ther 1985; 11: 345-356. Browne G, Bvme C. Roberts J. Streiner D. Fitch M. Corey P, Arpm K. The Meaning of Illness’Question: naire: Reliability and validity. Nursing Res 1988; 37(6): 368-373. Derogatis LR, Lopez MC. PAIS and PAISSR Scorhrg Manual I. Baltimore: Johns Hopkins University, School of Medicine; 1983. Weisman M. The assessment of social adjustment: An update. Arch Gen Psycbiat 1981; 38: 250-1258. Morrow GP, Chiarello RJ, Derogatis L. A new scale of assessing patients’ psychosocial adjustment to medical illness. Psychol Med 1978; 9: 605-610.
45.
46.
47. 48. 49. 50. 51.
52.
53. 54. 55.
56.
57.
58.
59.
383
Salata P. Variables which distinguish good and poor function&t outcomes in CGLD natients. Unnublished Master’s ‘project, McMaster University Hamilton, Ontario; 1981. Ruberman W, Weinblatt E, Goldberg JD, Chaudhary BS. Psychosocial influences on mortality after myocardial infarction. N hgl J Med 1984; 311: 552-559. JSeahrs0, Myers MH. Marmal for Smghrg of Cancer, 2nd edn. Philadelphia, Penn.: J. B. Lippincott; 1983. DeVita V, Helman S, Rosenberg S. Cancer, Principles and Prmetke of Oncology, 2nd edn. Philadelphia, Penn.: J. B. Lippincott; 1985. Nealan, T. Managementof the Patient with Cancer. Toronto: W. B. Saunders; 1986. Haskell C. Cancer Treatment, 2nd edn. Toronto: W. B. Saunders, 1985. Rodnan GP, Schumacher RH, Zuaifier N. Criteria for determination of progression of rheumatoid arthritis and of functional capacity of patients with the disease. In: Primer on the Rheumatoid Diseases, 8th edn. Atlanta, Ga: The Arthritis Foundation; 1983: 209. Steinbrocker 0, Traeger CH, Batterman RC. Therapeutic criteria in rheumatoid arthritis. JAMA 1949; 140: 659-662. Farmer RG. Clinical features and natural history of inflammatory bowel disease. Med Clin North Am 1980; 64(6): 1103-l 116. Harvey RF, Bradshaw JM. A simple index of chron’s disease activity. Lmcet 1980; March 8: 514. Best WR, Becktel JM, Singleton JW, Kern F. Development of a chron’s disease activity index. Gastroenterology 1976; 70: 439-444. Kamosky DA, Abelman WH, Craier LF, Burchenal JH. The use of nitrogen mustards in the palliative treatment of cancer. Cancer 1948; l(4): 634-656. Skeel RT. Systematic assessment of the patient with cancer. In: Skeel RT, Ed. Manual of Cancer Therapy. Boston: Little, Brown Co.; 1982: 13-21. Royal College of General Practitioners. CIasslBcation of Diseaws, Problems nod Procedures (Occasional Paper 26, ISBN 850 84094 5). May 1984. Blishen BR, McRoberts HA. A revised socioeconomic index for occupations in Canada. Can Rev Se&d An&r@ 1976; 13(l): 65-73.
Vol. 43, No. 4, pp. 373-383, 1990 Printed in Chat Britain. All rights reserved
Copyright 0
0895-4356/90 $3.00 + 0.00 1990 Pergamon Prcaa pk
PREVALENCE AND CORRELATES OF FAMILY DYSFUNCTION AND POOR AI)JUSTMENT TO CHRONIC ILLNESS IN SPECIALTY CLINICS* KATHLEEN &PIN,‘?
MARGARET FITCH,‘** GINA BoI-& BROWNE””
and PAUL CORJZY~
‘Graduate Department of Nursing, Faculty of Nursing, University of Toronto, Toronto, Ontario, Toronto General Division, Toronto Hospital Corporation, Toronto, Ontario, ?khool of Nursing, Faculty of Health Sciences, McMaster University, Hamilton. Ontario, ‘Department of Clinical Epidemiology and Biostatistics, Faculty of Health Sciences, McMastcr University, Hamilton, Ontario and ‘Department of Preventive Medicine and Biostatistics, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada (Received in revised form 19 July 1989)
Abstract-This study estimated the prevalence of poor adjustment and family dysfunction among three chronically ill clinic patient populations and ass&d the biological, situation& social and psychological variables which most explained poor adjustment. Recently referred subjects were approached and 216 chronically ill subjects (from oncology, rheumatology and gastroenterology clinics) completed a Meaning of Illness Questionnaire, the M&aster Family Assessment Device, and the Psychosocial Adjustment to Illness Self-report Scale. In addition, information describing their biological, disease and socioeconomic stattis was obtained from the clinic record. Respondents were generally representative of other new referrals to the clinics (ineligible for the subsequent trial) in disease characteristics but uniformly.came from a more advantaged socioeconomic situation and were better adjusted to illness. Subjects from the three clinics were comparable on meaning, family function and adjustment variables. The proportion of subjects with family dysfunction was 30% and with poor adjustment to illness was 36%, high by community standards. Nevertheless, healthy family functioning and high levels of positive adjustment to chronic illness prevailed and were remarkably similar across clinic settings. Severity or type of disease was not related to adjustment outcomes nor to the level of observed disability. Rather, as hypothesized, meaning given the illness, followed by family function, and disability variables combined to explain 57% of the variance in adjustment outcomes. An intervention designed to improve family function and the meaning given illness was judged suitable. Family function
Psychosocial adjustment
INTRODUCTION
Most people with chronic illness adjust to their conditions without change in their mood *Collaborative project supported by the Ontario Ministry of Health, No. 01724, Ontario, Canada. ?A11 correspondence should be addressed to: Kathleen Arpin, Professor, Chair of Graduate Prograqmw, Faculty of Nursing, 50 St George Street, Tsronto, Ontario, Canada MS1 IA1 (Tel: 1-416-978-2848).
Chronic illness
Specialty clinics
or social roles [l]. However, a small but economically significant group of less adjusted chronically ill are concentrated in outpatient specialty clinics [2]. The majority of patients referred to these clinics suffer from co-existing chronic physical and psychosocial maladies [3]. The substantial body of information regarding adjkts$ment to chronic conditions is specific to cert.&n diseases. There is little comparative inform&ion about adjustment patterns across 373
I
374
KATHLEEN ARPINet
diseases [l]. Stress and social support literature suggest that adjustment to a chronic illness stressor is mediated by the attitude or meaning given the event and the quality of family support as a coping resource [4]. Less is known about how the meaning given illness and family function variables may combine with the situational disease factors to explain adjustment outcomes. These findings raised two questions: (1) Is adjustment to chronic illness related to the type and severity of the illness situation, or is it more closely related to cognitive appraisal (the meaning of illness) and family functioning; and (2) Are correlates of psychosocial adjustment similar across different chronic illnesses? Lazarus and Folkman [4] have proposed that the evaluation (of illness as a stress) consists of a primary appraisal of the relevance and seriousness of the event and a secondary appraisal of one’s coping resources to deal with it. An individual’s “commitments” regarding what is important [4, pp. 56621 and “beliefs regarding one’s ability to control the outcome” [4, pp. 65-661 were also considered important variables in adjustment. These authors’ central proposition that cognitive appraisal (which may be affected by the situation) leads to coping which “assists in maintaining social function, morale and somatic health” [4, p. 1821suggested the importance of the meaning ascribed to illness rather than its nature. There is accumulating research evidence that adjustment to chronic illness is independent of its type or severity. Psychological adaptation among non-cerebrally damaged chronically ill outpatients [5] is remarkably effective and fundamentally independent of specific disease such as cancer [l, 61, renal failure [7], or stroke [8]. For patients with lung disease [9, lo], renal disease [I 11,or burn trauma [12] adjustment has not been shown to be related to the severity of disease. The relationship between treatment status among the chronically ill and their adjustment outcomes is equivocal, however [13]. Cassileth et al. [6] found that psychosocial status of seven types of cancer patients closely correlated with treatment status. The psychological state was best for those on follow-up treatment, adequate for those in active treatment and worse for those receiving palliative treatment for cancer. Yet, in other studies that controlled for type of treatment, social support [14] and, in particular, sexual/affectional patterns in the marriage, were more closely related to adjustment among
al.
cancer patients than were prognosis or disability [ 151.These findings suggested that future studies of adaptation to chronic illness should continue to weigh the importance of the biological variables of disease severity, function level, and treatment status against other psychosocial variables. Psychosocial variables correlating with adjustment outcomes among patients with a variety of chronic diseases are life stress and social isolation [ 161, unfavourable self concepts, social relationships or attitudes toward illness [17,‘18]. A summary of the best literature on psychosocial correlates of adjustment to illlness revealed that social suport [13, 19-211 and cognitive variables [4,22,23] (which include the meaning given life events such as illness or treatment [2429]), combine to explain between 27 and 57% of the variation in a person’s eventual adjustment to a chronic illness [18]. Literature on social support linked family support to good adjustment. Among patients receiving chronic home dialysis [30], family or household members’ support [19] was more important than support provided by professionals in developing good vocational adaptation. Similar relationships were observed in cancer patients [6,31], in stroke patients [20], and in patients with chronic pain [32-341. Among cancer patients, family expressiveness with less conflict and more moral-religious emphasis was positively related to the patient’s adjustment [35]. The spouse’s cognitive interpretation of the patient’s chronic pain syndrome was closely related to the outcome of both partners [23]. The family variable was crucial to the rehabilitation outcome in adult arthritics, aphasics and other chronic disabling conditions [36]. No study could be located which simultaneously weighed the relative importance of a variety of disease variables against social and psychological variables of family function and meaning given illness in explaining adjustment to illness across a variety of diseases. In light of Lazarus and Folkmans’ [4] conceptualization of stress-coping it was hypothesized that neither the type nor severity of the disease stressor would be related to adjustment of the chronically ill. Rather, adjustment to the illness would be related to family functioning and the cognitive appraisal or meaning variables, and these relationships would be similar across different types of diseases. The baseline data from a larger study designed to test a clinical intervention were used to test these two hypotheses.
Family Function and Adjustment in Chronic Illness
315
Table 1. Procedure for obtaining sample ALL NEW REFERRALS Mixed cancer 241
I
Gastroenterology 409
(N = 1288) Rheumatology 638
ELIGIBLE SUBJECTS (A’= 520) Mixed cancer 118
Gastroenterology 171
I
INELIGIBLE: (N = 768) -not a chronic condition -unable to commute -retarded/cerebral damage --lives alone with no regular contact with a single person -non-English speaking patient or family member -already receiving counselling --25yr old or >80yr old -disallowed by attending clinicians
Rheumatology 231 DECLINED: (N = 302) -refused meetings (not needed) = 250 -refused meetings (needed) = 41 = 5 -refused quest
CONSENTING Mixed cancer 84
SUBJECTS (N = 218)
Gastroenterology 89
Rheumatology 45 DEATHS (N = 2)
BASELINE MEASURES (N = 216)
METHODS
Subjects were recruited from the Ambulatory Service Clinics of the Toronto General Hospital Division of the Toronto Hospital Corporation. These clinics/services are regional referral centres for a number of cancer, gastroenterological, and rheumatological chronic diseases. Table 1 outlines the procedure for obtaining the sample over a 9 month period from each of three clinics. From a total of 1288 new referrals, 768 subjects were ineligible for the subsequent trial; 286 of these either did not have a chronic condition or were not returning to the clinic, 213 others lived out of town, 89 lacked familiarity with English, 65 lived alone with no regular contact with others, 56 were disallowed eligibility by the attending physician or clinic nurse, 3 1 were already receiving counselling, 15 were ineligible due to age, 11 were either retarded or had cerebral damage, 1 was not a new referral, and data on 1 subject was missing. Of 520 eligible subjects, 302 declined participation, 2 died prior to the baseline measures, and 216 agreed to participate. 80% of those who declined participation were rated by the research nurses as well adjusted based on a global clinical index of adjustment, a judgement later corroborated by the baseline measures concerned with family function and psychosocial adjustment.
The project coordinator and research nurses administered the baseline bio-social-culturaldemographic inventory and the meaning, family function, adjustment questionnaires listed in Table 2 to consenting subjects at the time of their first appointment. The extensive methods of classifying the biological data and assessing their reliability and validity is reported elsewhere [37]. Chart data were abstracted as the measure of some of the biological variables. Other disability or treatment variables, such as severity of illness and/or functional ability using standardized classification systems, required the clinical judgement of the research nurses. The reliability of the nurses’ clinical judgement was compared with that of the physician on 18 subjects, and the degree of agreement is reported in Table 2 next to the variable and the classification system. The social and psychological explanatory variables were measured using instrumentation reported in Table 2. The Family Assessment Device [38] was chosen to assess family function after a careful review of nine other measures {39]. The reliability and validity of this measure have been documented [38,40] and are reported to classify clinicians’ judgements correctly regarding healthy and unhealthy families, with a sensitivity of 64% and a specificity of 67%. The reasons for developing the Meaning of Illness Questionnaire and the extensive analyses
376
KATHLEENARPINet
al.
Table 2. Biological, social, psychological variables, measures and reliability checks Construct Biological/disease variables
Variable -Clinician’s primary diagnosis -Months of time since diagnosis -Stage or severity of illness in cancer in rheumatoid disease in gastroenterological disease -Prognosis -Disfigurement -Functional
ability
-Additional illnesses before and since diagnosis -Disability related to combined diseases -Treatment status
Measure
Reliability validity
Chart abstraction Chart abstraction Chart abstraction using: [47, 48, 49, SO] -stage of cancer -curability 151, 521 -functional ability in rheumatoid disease [53, 54, 551 -modified severitv and leneth of gastroenterolbgical diseases Clinician’s judgement 1 = terminal to 7 = positive outcome and 5 year survival rate [47, 48, 49) in cancer Clinician’s judgement 1 = no disfigurement to 7 = severe (shocking) disfigurement Clinical judgement using Kamoskv Scale 1561 ECOG Scale [57j _ Chart abstraction [58]
-
Clinical judgement using Karnosky scale [56] Clinical judgement using 1 = active, 2 = follow up
Social variables
-Date of birth -Sex -Educational level -Marital status -Working status -Education of head of household -Combined family income -Country or culture -Occupation
Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Clinician’s inventory Blishen (1976) socioeconomic status [59]
Social support variables
Family function
Psychological variables
Coping, cognitive appraisal of illness
McMaster Family Assessment Device (FAD) [38, 401 53 items on 4ooint scale -Communicaiion -Problem solving -Roles -Affective involvement -Affective responsiveness -Behaviour control -General functioning Meaning of Illness Questionnaire [41]
Outcome variable
Psychosocial adjustment
K = 0.96
K =
0.56
P,, = 83%
ICC = 0.68 ICC = 0.46 ICC = 0.68 K =0&t
Test/retest r = 0.66 to 0.76 Cronbach’s alpha = 0.74 0.75 0.72 0.78 0.83 0.72 0.92 K = 0.484.99
Factors: 33 items on 3- and ‘I-ooint scales I. Impact of Illness on Daily Living R2 = 0.46060 II. Type of Stress: (negative attitude of harm, loss, threat) and Functional Context III. Degree of Stress; Change in Commitments; Secondary Appraisal of Coping Resources IV. Positive of Illness as __ -_ .. Attitude Challenge; Hope, Motivation, Control V. Expectancy and Reoccurrence of Illness Psychosocial Adjustment to Convergent validity Illness Scale (PAIS) [42] Cronbach’s alpha = 45 items on 4-point scale -Health care orientation 0.63 0.81 -Vocational environment -Domestic environment 0.67 -Sexual relationships 0.80 -Extended familv relationshios 0.66 -Social environment _ 0.78 -Psychological distress 0.80 -Total adjustment score
K = Kappa; ICC = intraclass correlation coefficient; PO= observed agreement.
Family Function and Adjustment in Chronic Illness
to assess its reliability, factor structure, and validity have been previously reported [41]. In a sample of 320 subjects approached for participation in the present study, three meaning items explained (R2) 60% of the variance in adjustment. Three meaning factors explained 46% of the variance in adjustment. The “outcome” variable of interest was psychosocial adjustment to illness (PAIS-SR) [42], selected after a careful review of other measures of adjustment [43]. The PAIS-SR is a measure of change in seven dimensions of a person’s life as a result of the illness. The reliability and convergent and predictive validity of this measure are well documented [l 1,42,44], and the degree of agreement between the PAIS total score and other valid measures of attitude, mood, or symptomatology generally ranges between r = 0.60 and 0.81. A PAIS raw score of O-34 corresponds to clinical judgements of good adjustment; 35-53 indicates fair adjustment; and 54 or greater indicates very poor adjustment to illness [ll, 12,42,45]. RESULTS
In the analysis of results the representatiueness of participating subjects from each clinic is described in terms of sociodemographic and disease severity variables. Eligible subjects were compared with 68 ineligible subjects (largely from the gastroenterology and rheumatology clinics) who agreed to do part or all of the questionnaire. Participating subjects were compared with 36 persons who declined participation but agreed to complete some or all of the questionnaire. The impact of any biis due to those events on study results is highlighted. The comparability of subjects between each clinic was assessed and differences between the chronically ill groups are highlighted. Thereafter, the questions of prevalence and correlates of adjustment are addressed. Representativeness
Socioeconomic data describing the nature and severity of persons’ diseases, as in Table 2, were gathered from all new referrals. These data allowed comparisons between those eligible and ineligible and those who consented to and declined participation in the later trial. The details of this extensive comparison are reported elsewhere [37J. In summary, eligible consenting chronically ill subjects in all three clinics were uniformly a more socioeconomically advanCE43/4-E
311
taged group, slightly better adjusted with somewhat more favourable meanings given the illness. The effect of these differences would be to lower the prevalence of dysfunction in this sample. Comparability of participating clinics
subjects across
Subjects in all three clinics differed in their prognosis, age, education, marital status, occupational prestige, and family income. However, as shown in Table 3, subjects with different diseases did not differ in their baseline adjustment to illness, family function and majority of types of meaning given illness. Rheumatology subjects viewed their illness more negatively than others. Gastroenterology subjects had more expectation their illness would reoccur. Prevalence of dysfunction among the chronically ill
Estimates of prevalence of a disorder based on samples from tertiary centres are inflated compared to community samples due to the concentrating factor which accompanies the process of referral [2,3]. This fact should be kept in mind when interpreting the results. (i) Family dysfunction. The families from the three clinics in the present study were classified as healthy or unhealthy, using the cutoff points described [40] on each dimension of family functioning. This comparative prevalence of family dysfunction among the subjects in three specialty clinics is displayed in Table 4. In general, the prevalence of family dysfunction (in every dimension) across three chronically ill groups was remarkably similar (x2 = 1.18, p = 0.55) and was in general 30% of the sample (inflated compared to the community). Noteworthy among the dimensions of family function is the high prevalence of unhealthy communication (44%), poor involvement (44%), and problems in behaviour control (44%). (ii) Adjustment. Kruskal-Wallis was used to test differences in the levels of good, fair, and poor adjustment across the three clinics. No important differences were found in the number of subjects with various levels of adjustment to illness across the three clinics. As in Table 5, the vast majority (approximately two-thirds) of the chronically ill adjusted to their illness with little change in their life. Only a small proportion of the chronically ill exhibited poor adjustment to illness or major changes due to illness. There
378
KATHLEEN ARPIN et
al.
was a middle group of subjects who reported a fair to moderate degree of change in their life (adjustment) attributable to the illness. In all three clinic samples, these same patterns were observed. The combined prevalence of fair to poor adjustment of 36% again reflected the concentration of these disorders in outpatient specialty clinics. Nevertheless, was there any relationship between the prevalence of poor adjustment and that of family dysfunction? The correlations between the dimensions of family function and adjustment or psychological distress ranged between 0.18 and 0.38. The weak correlations indicate that something other than family dysfunction must explain poor adjustment to illness or psychological distress. Consequently, all baseline data including type of disease (clinic) were used to assess the combination of biological, disease, socioeconomic, family function, or meaning given illness variables which most explained the variation in subjects’ adjustment to illness. Baseline illness
correlates
of adjustment
to chronic
From stepwise regression analyses performed on the adjustment to illness variable within each clinic it was consistently found that the two meaning given illness factors, namely, the “negative attitude” (Factor I) and the “adverse impact” (Factor II), as well as some aspect of family dysfunction always entered the model before any of the biological and disease variables. Although no overall formal test of interaction of all of these effects across clinics was performed, their partial regression coefficients were similar. For example, the coefficients associated with the meaning of illness of Factor I were not significantly different across the clinics. The remaining analyses were performed on the combined dataset with the clinic factor in the model. A formal test of the effect that the meaning given illness and family function variables had in excess of the contribution of disease severity and disability was carried out. The partial correlation coefficient associated with the cognitive and family function variables adjusted for the disease and severity variables was 0.57 (p < 0.0000001) whereas the partial correlation coefficient associated with the disease severity and family function variables adjusted for the cognitive and family function variables was 0.30 (p < 0.0001). That is, the meaning of illness and
319
Family Function and Adjustment in Chronic Illness Table 4. The percentage of subje&s with di&ent types of fatily dysfunction in three samples of chronicallv ill
Oncology Rheumatology Gastroenterology p-Value N=45 N = 88 N = 80
Dimensions of
familyfunction
15% 42% 17% 41% 31% 51% 34%
Problem solving Communication Roles Affective involvement Affective responsiveness Behaviour control General family functioning
18% 40% 27% 40% 29% 36% 24%
0.85 0.14 0.28 0.60 0.96 0.19 0.55
18% 47% 21% 48% 31% 41% 31%
p = probability associated with the chi-square test of three independent proportions. family function variables exceeded the importance of disease severity and disability in explaining adjustment to illness. Severity of disease alone explained only 7% of the variation in adjustment to illness. Thereafter, forward stepwise multiple regression analyses using free entry and listwise deletion were conducted, as in Table 6, in search of the most important combination of variables listed in Table 2 which explained adjustment to chronic illness. The socioeconomic variables of educational level, marital and working status, occupational prestige and education of the head of household, and ethnic origins were entered as dummy variables, along with clinic, to represent the type of disease. Although each of the socioeconomic variables were interrelated, none were important in explaining a person’s adjustment. The discrete biological variables entered along with clinic into a second equation were: months of time since diagnosis, age, stage (severity) of illness, the Karnosky and ECOG approaches to measuring disability, the clinician’s assessment of disfigurement, prognosis, and treatment status. Only disability and little disfigurement together explained 32% of the variance in patients’ poor adjustment to illness [F: (3,212) = 44.47, p < O.OOOOl]. Disability or function level, however, was only
weakly related to the biological severity or stage of illness (r = 0.31) and unrelated to disfigurement (I = -0.08). Further, the more definitive measure of severity of illness such as “stage” of illness tias weakly related to adjustment (r = 0.25). Consequently, disability may serve as a marker for some other psychosocial variable (associated with disability) which would explain adjustment to illness. The seven dimensions of family functioning were entered, along with clinic, into a third equation, and only the dimension of poor role enactment in the family entered to explain 15% of the variance in adjustment [F: (1,211) = 36.15, p < O.OOOOl].As before, none of the dimensions of family function correlated very strongly with adjustment. The five meaning of illness factors, three statistically significant coping behaviour items, eight categories of previous and current commitments (1 =present; 0 = absent) and realignment in commitments (0 = no change; 1 = change) were entered into a fourth regression equation along with clinic. The negative views of illness as stress (harm, loss, threat), poor view of functional status (disability, deterioration, disfigurement) and concept of illness as “adversely affecting daily living” together explained 42% of the variance in a person’s poor adjustment to illness. Other
Table 5. The percentage of subjects with problems in adjusting to illness in oncology, rheumatology and gastroenterology tertiary clinics Clinic: PASSR Raw-score Adjustment: Good o-34 Fair 35-53 Poor 54+ Total
Oncology N = 83
Rheumatology N=45
Gastroenterology N = 88
Total 216
10%
62%
60%
64%
24%
29%
21%
26%
6%
9%
13%
10%
(100%) (100%) (100%) (Kruskal-Wallis y* = 2.26, p = 0.32)
(100%)
KATHLEEN fiRPIN et al.
380
Table 6. Biological, social, psychological and clinic variables which most explain poor adjustment among the chronically III (multiple regression with listwise deletion) Construct Biological and socioeconomic N =216
Social support Psychological (meaning given illness) N=189
Summary equation of biological, social psychological variables N = 188
Variable
/l*
(a) Socioeconomic (b) Biological/disease -Disability (ECOG) -Disability (Karnosky) -Disfigurement Family functioning -Roles Type of stress/illness (negative attitude) of illness as harm, loss, threat and functional contextdisability, deterioration, disfigurement Impact of illness on daily living Feel you need to know more before you can act Realigned commitments Optimistic attitude of illness as challenge; hope, motivation, control Meaning: Factor II Type of stressfulness (negative attitude) of harm, loss, threat, and functional context Meaning: Factor I Impact of illness on living Genera1 family functioning Disability (ECOG) Disfigurement Meaning: Realigned commitments Treatment status Meaning: Feel you need to know more
R2
p-Value?
None entered 5.80 -0.22 - 1.50
0.27 0.30 0.32
0.00 0.00 0.01
8.52
0.15
0.00
0.56 0.42 0.70 2.56
0.32 0.42 0.45 0.46
0.00 0.00 0.00 0.01
0.48
0.02
0.53
0.32
0.00
0.33 3.96 4.18 -1.38
0.43 0.47 0.51 0.53
0.00 0.00 0.00 0.01
2.49 2.11
0.55 0.56
0.01 0.03
-0.29
0.45 0.57 0.05 [F: (8,179) = 29.66, p < O.OOOOl]
*Regression coefficients in the final model. tp-Values associated with B in the final model.
meaning of illness items and factors (realigned little optimism) entered to commitments, explain 48% of the variance in a person’s poor adjustment to illness [F: (5,183) = 33.52, p < O.OOOOl]. The negative view of the illness as stressful and associated with poor function (Factor II) was moderately correlated with the clinician’s rating of disability (ECOG) where (r = 0.44). However, as with the disability assessment, there was only a weak relationship between the view of the illness as disabling and the objective reality regarding the stage or severity of illness (r = 0.23). The clinician may have observed illness behaviour more than disability related to disease. All of the important biological, social, and psychological (meaning) variables were entered, along with clinic as a dummy variable, into a summary regression equation (Table 6) in search of the combination of factors which most explained a person’s adjustment to chronic illness. The meaning variables of “the view of illness as a harm, loss, threat associated with disability, deterioration, and disfigurement” and the view of the “illness (adversely) affecting
their daily living” together with poor family function in general, entered to explain 47% of the variance in a person’s poor adjustment to chronic illness. These variables, coupled with the clinician’s rating of disability, little disfigurement, and additional meaning and treatment variables, collectively explained 57% of the variance in a person’s adjustment [I;: (8,179) = 29.66, p c O.OOOOl].This pattern of the meaning of illness variables exceeding the importance of disease severity or disability variables was observed in every clinic. In general, the cognitive appraisal or meaning variables, followed by family function variables, had the most explanatory power. The biological variables of observed disability and treatment status together although statistically significant, add little (6%) in explaining adjustment of the chronically ill when weighed against the other psychosocial variables. The weak correlations between the meaning of illness factors and dimensions of family function (r ranges between 0.02 and 0.24) indicate that these two variables are relatively independent and make a separate contribution to living with chronic illness.
Family Function and Adjustment in Chronic Illness
One measurement bias in the measure of adjustment to illness is in the direction of the scoring. In this PAIS measure, it is only possible to have no change = 0 or change (deterioration) = l-3. Consequently, it fails to detect improvements in various dimensions .of people’s lives since their illness which have been observed. A one-question item regarding a person’s improvement since the illness from 0 = none to 4 = a great deal in each of the vocational, domestic, sexual, social, extended family, psychological distress components of the adjustment scale were added at the end of that respective section of the PAIS. A total improvement in adjustment score was derived as the sum of each source of improvement. Again, the chronically ill from the three clinics were remarkably similar in their report of life improvements since the illness. More than 60% of subjects reported fair to substantial improvement in areas of their life since the illness. Similar separate and summary regression analyses were conducted as in the previous analyses in search of factors associated with this improvement. The cognitive appraisal or meaning variables again exceeded the importance of all others and explained 14% of a person’s life improvement since the illness [F: (2,186) = 14.88, p < O.OOOOl]. When weighed with all other variables in Table 2, “the degree of stress, change in commitments and appraisal of (sufficient) coping resources” coupled with an “optimistic attitude of illness as a challenge: hope, motivation and control” entered first to explain 14% of the improvements in adjustment to illness. These, together with lower levels of education and greater periods of time since diagnosis, explained 18% of the variance in patient’s improvement in adjustment to illness [F: (4,183) = 10.09, p < O.OOOOl]. Summary
In these specialty clinics, the chronically ill are remarkably similar in their levels of adjustment, family function. and meaning given illness. Thirty percent of respondents had unhealthy patterns of family function; 36% had fair to poor levels of adjustment to illness. The respondents represented the more advantaged socioeconomic group of all new referrals to three specialty outpatient clinics; however, neither the socioeconomic nor biological disease severity situation variables were important in explaining a person’s adjustment to illness. Rather, the
381
unfavourable meanings of “harm, loss, threat” and “disabling, deteriorating, disfiguring”, “(adverse) impact of illness on total life”, “realigned commitments”, and “poor family function” collectively explained 57% of the variance in poor adjustment to illness. Similar negative meaning or unhealthy family communication patterns explained 31% of the variance in a person’s level of psychological distress. By contrast, approximately 60% of subjects reported moderate to sign&ant improvements in their lives since their diagnosis. The more favourable meaning given the stressfulness of illness of “change in commitments, availability of coping resources”, “view of (illness as) a challenge, coupled with a sense of hope, motivation, control” among the less educated and less recently diagnosed subjects explained 18% of the variance in their improvement since the illness (with little contribution from the biological, disease, or family function variables). Discussion and implications
Although a cross-sectional design cannot determine the direction of causality, the results of this study pointed to the importance of the cognitive appraisal variable in explaining adjustment to the stress of illness. They support the central propositions of Lazarus and Folkman that the cognitive appraisal aspect of “coping assists in maintaining social function, morale and somatic health” [4, p. 1821. This finding, coupled with the high prevalence of poor family functioning and adjustment among the chronically ill in specialty clinics, served as sufficient justification for the subsequent trial to assess the “Efficacy of Family Clinic Appointments in Promoting Adjustment to Chronic Illness” [371. These findings also support the clinical efforts to assist patients and/or their families who have unfavourable views of their illness, with a cognitive reappraisal of their situation so that they may achieve a more optimistic, hopeful and satisfied outlook. Cognitive reappraisal could be emphasized in individual, family or group orientations to treatment. Since there was little relationship between disease severity and meaning given the illness, negative or unfavourable attitudes toward illness should be used rather than disease severity for identifying subjects in need of referral to psychosocial services. Conversely, subjects with a favourable and hopeful outlook, regardless of the incongruence with the
382
KATHLEEN
severity of their disease, should be encouraged to maintain their bases of hope. In the family approaches to treatment, the effective coping behaviours of problem-solving and communication should be emphasized. Cognitive behavioural orientations to family treatment would also emphasize assisting families to achieve a more optimistic interpretation of members’ behaviour. Treatment frequently appears to facilitate changed attitudes or more optimistic meanings attributed to familiar patterns of behaviour in families rather than real differences in the pattern of behaving per se. Future research could test this proposition. Since a relationship was found between both the meaning of illness and family variables and adjustment to illness, our future research will test the efficacy of family meetings in promoting more favourable meaning given illness, family function and, ultimately, adjustment to illness. If efficacious, the cost and benefit of this service, designed to augment conventional medical outpatient care, will be assessed [37].
REFERENCES 1.
2.
3.
4. 5.
6. 7.
8.
9.
10.
Cassileth BR, Lusk EJ, Strouse TB, Miller DS, Brown LL, Cross PA, Tenaglia AN. Psychosocial status in chronic illness. N Engi J Med 1984; 311: 5066511. Spaulding WB. The psychosomatic approach to the practice of medicine. In: Lipowski ZJ et al., Eds. Psychosomatic Medicine. New York: Oxford University Press; 1977: 457-469. Eastwood MR. Epidemiological studies in psychosomatic medicine. In: Lipowski Z J et al., Eds. Psychosomatic Medicine. New York: Oxford University Press; 1977: 41 I-420. Lazarus RS, Folkman S. Stress, Appraisal and Coping. New York: Springer; 1984. Davies RK, Quinlan DM, McKegney FP, Kimball CP. Organic factors and psychological adjustment in advanced cancer patients. Psychosom Med 1973; 35: 464-47 1. Cassileth BR, Lusk EJ, Strouse TB, Miller DS, Brown LL, Cross PA. A psychological analysis of cancer patients and their next of kin. Cancer 1985; 55: 72-76. Livesley WJ. Symptoms of anxiety and depression in patients undergoing chronic haemodialysis. J Psychosom Rcs 1982; 26: 581-584. Bishop DS, Epstein NB, Keitner GI, Miller IW, Srinivasan SV. Stroke: Morale family functioning, health status and functional capacity. Arch Phys Med RehabB 1986; 67: 84-87. Dudley DL, Glaser EM, Jorgenson BN, Logan DL. Psychosocial concommitants to rehabilitation in chronic obstructive pulmonary disease: Part 2. Psychosocial treatment. Chest 1980: 77: 554-561. Judd L, Browne G, Craig H, Norman G, Guyatt G. The prevalence and correlates of poor psychosocial adjustment in patients with chronic airflow limitation. In: King K, Ed. NursIag Research: Science for Quality Care. Proc 10th National Nursing Research Conference. University of Toronto; 1986: 109-I 12.
ARPIN et
al.
11. DeNour AK. Psychosocial Adjustment to Illness Scale (PAIS): A study of chronic hemodialysis patients. J Psychosom Res 1928; 26(l): 11-12. 12. Browne G, Byrne C, Brown B. Psychosocial prognosis of burn survivors. Burns 1985; 12: 28-35. 13. Goldberg RJ, Cullen LO. Factors important to the psychosocial adjustment to cancer: A review of the evidence. Sot Sci Med 1985; 20: 803-807. 14. Revenson TA, Wollman CA, Felton BJ. Social supports as stress buffers for adult cancer patients. Psvchosom Med 1983: 45: 321-331. 15. Taylor SE, Lichman’ RR, Wood JV, Bluming HZ, Dosik GM, Leibowitz RL. Illness related and treatment related factors in psychological adjustment to breast cancer. Cancer 1985; 55: 2506-2513. 16. Lewis FM, Bloom JR. Psychosocial adiustment to breast cancer: A review of selected literature. Int J Psvchiat Med 1978: 9(l): 1-17. 17. Hyman MD. Social psychological factors affecting disability among ambulatory patients. J Cbron Dis 1975; 28: 199-216. 18. Counte MA, Bieliauskas LA, Pavlou M. Personal stress and personal attitudes in chronic illness. Arch Phys Med Rehabil 1983; 64: 272-275. 19. Broadhead WE, Kaplan BH, James SA, Wagner EH, Schoenbach VJ, Grimson R, Neyden S, Tibblin G, Gehlbach S. The epidemiologic evidence for a relationship between social support and health. Am J Epidemiol 1983; 117: 525-541. 20. Evans RL, Northwood LK. Social support needs in adjustment to stroke. Arch Phys Med Rehabil 1983;64: 61-64. 21. McQueen DV, Siegnist J. Social factors in the etiology of chronic disease: An overview. SIX Sci Med 1982; 16: 353-367. 22. Viney LL, Westbrook MT. Coping with chronic illness: Strategy preferences, changes in preferences and associated emotional reactions. J Chron Dis 1984; 37: 489-502. 23. Block AR, Bayer SL. The spouse’s adjustment to chronic pain: cognitive and emotional factors. Sot Sci Med 1984; 19: 1313-1317. 24. Streiner DL. Norman GR. McFarlane AH. Rov RG. Quality of life events and their relationship to itrain. Schizophrenic Bull 1981; 7(l): 34-42. 25. Schroeder DH, Costa PT. Influence of life event stress on physical illness: Substantial effects or methodological flaws? J Pers Sot Psycho1 1984; 46: 853-863. 26. McFarlane AH, Norman GR, Streiner DL, Roy RG. The process of social stress: Stable, reciprocal and mediating relationships. J Health Sot Behav 1983; 24: 160-173. 27. McFarlane A, Norman G, Streiner D, Roy R, Scott D. Longitudinal study of the influence of the psychosocial environment on health states: A preliminary report. J Health Sot Behav 1980; 21: 124-133. 28. Moos RH, Billings AG. Conceptualizing and measuring coping resources and processes. In: Goldberger L, Breznitz S, Eds. Handbook of Stress: Theoretical and Clinical Aspects. New York: The Free Press; 1982: 2 12-230. 29. Moos RH, Cronkite RC, Billings AG, Finney JW. Index of Life Change Events. In: HeaItb and Daily Living Form Manual. Palo Alto, California: Social Ecology Laboratory, Veterans Administration and Standard University Medical Center; 1984. 30. Palmer S, Canzona L. Conley J, Wells G. Vocational adaptation of patients on home dialysis: Its relationship to personality, activities and support received. J Psychosom Res 1983; 27(3): 201-207.
Family Function and Adjustment in Chronic Illness 31.
32.
33. 34. 35. 36.
31.
38. 39. 40.
41.
42.
43. 44.
Baider L, DeNour AK. Couples’ reactions and adjustment to mastectomy. Int J Psychiat Med 1984; M(3): 265-270. Shanfield SB, Heiman EM, Cape DN, Jones JR. Pain and the marital relationship: Psychiatric distress. Pain 1979; 7: 343-351. Block AR. An investigation of the response of the spouse to chronic pain behaviour. Psychosem Med 1981; 43: 415-423. Flor H, Turk DC, Scholz OB. Impact of chronic pain on the spouse: Marital, emotional and physical consequences. J Psychosom Ree 1987; 31(l): 63-71. Spiegel D, Bloom JR, Gollheil E. Family environment as a predictor of adjustment to metastatic cancer. J Psycheaee Oneel 1983; l(1): 33-44. Bishop DS, Epstein NB. Family problems and disability. In: Behioural Problems and the Disabled. Baltimore: Williams and Williams; 1980: 387-364. Arpin K, Fitch M, Browne G, Corey P. Efficacy of family clinic appointments in promoting adjustment to chronic illness: Methodological manual and appendices. Unpublished manual, Ontario Ministry of Health Project No. 01724, University of Toronto, Faculty of Nursing, Toronto, 1988. Epstein N, Baldwin L, Bishop D. The McMaster Familv Assessment Device. J Marital Fam Ther 1983: 9(2): i71-180. Speer J, Sachs B. Selecting the appropriate family assessment tool. Pediatric Nursing 1985; 11: 349-355. Miller I, Bishop D, Epstein N, Keitner G. The McMaster Family Assessment Device: Reliability and validity. J Marital Fam Ther 1985; 11: 345-356. Browne G, Bvme C. Roberts J. Streiner D. Fitch M. Corey P, Arpm K. The Meaning of Illness’Question: naire: Reliability and validity. Nursing Res 1988; 37(6): 368-373. Derogatis LR, Lopez MC. PAIS and PAISSR Scorhrg Manual I. Baltimore: Johns Hopkins University, School of Medicine; 1983. Weisman M. The assessment of social adjustment: An update. Arch Gen Psycbiat 1981; 38: 250-1258. Morrow GP, Chiarello RJ, Derogatis L. A new scale of assessing patients’ psychosocial adjustment to medical illness. Psychol Med 1978; 9: 605-610.
45.
46.
47. 48. 49. 50. 51.
52.
53. 54. 55.
56.
57.
58.
59.
383
Salata P. Variables which distinguish good and poor function&t outcomes in CGLD natients. Unnublished Master’s ‘project, McMaster University Hamilton, Ontario; 1981. Ruberman W, Weinblatt E, Goldberg JD, Chaudhary BS. Psychosocial influences on mortality after myocardial infarction. N hgl J Med 1984; 311: 552-559. JSeahrs0, Myers MH. Marmal for Smghrg of Cancer, 2nd edn. Philadelphia, Penn.: J. B. Lippincott; 1983. DeVita V, Helman S, Rosenberg S. Cancer, Principles and Prmetke of Oncology, 2nd edn. Philadelphia, Penn.: J. B. Lippincott; 1985. Nealan, T. Managementof the Patient with Cancer. Toronto: W. B. Saunders; 1986. Haskell C. Cancer Treatment, 2nd edn. Toronto: W. B. Saunders, 1985. Rodnan GP, Schumacher RH, Zuaifier N. Criteria for determination of progression of rheumatoid arthritis and of functional capacity of patients with the disease. In: Primer on the Rheumatoid Diseases, 8th edn. Atlanta, Ga: The Arthritis Foundation; 1983: 209. Steinbrocker 0, Traeger CH, Batterman RC. Therapeutic criteria in rheumatoid arthritis. JAMA 1949; 140: 659-662. Farmer RG. Clinical features and natural history of inflammatory bowel disease. Med Clin North Am 1980; 64(6): 1103-l 116. Harvey RF, Bradshaw JM. A simple index of chron’s disease activity. Lmcet 1980; March 8: 514. Best WR, Becktel JM, Singleton JW, Kern F. Development of a chron’s disease activity index. Gastroenterology 1976; 70: 439-444. Kamosky DA, Abelman WH, Craier LF, Burchenal JH. The use of nitrogen mustards in the palliative treatment of cancer. Cancer 1948; l(4): 634-656. Skeel RT. Systematic assessment of the patient with cancer. In: Skeel RT, Ed. Manual of Cancer Therapy. Boston: Little, Brown Co.; 1982: 13-21. Royal College of General Practitioners. CIasslBcation of Diseaws, Problems nod Procedures (Occasional Paper 26, ISBN 850 84094 5). May 1984. Blishen BR, McRoberts HA. A revised socioeconomic index for occupations in Canada. Can Rev Se&d An&r@ 1976; 13(l): 65-73.