Psychosocial adjustment in patients after a first acute myocardial infarction: The contribution of salutogenic and pathogenic variables

Psychosocial adjustment in patients after a first acute myocardial infarction: The contribution of salutogenic and pathogenic variables

811 Psychosocial Adjustment in Patients After a First Acute Myocardial Infarction: The Contribution of Salutogenic and Pathogenic Variables Yaacov Dr...

1MB Sizes 0 Downloads 8 Views

811

Psychosocial Adjustment in Patients After a First Acute Myocardial Infarction: The Contribution of Salutogenic and Pathogenic Variables Yaacov Drory, MD, Shlomo Kravetz, PhD, Victor Florian, PhD, Israel Study Group on First Acute Myocardial Infarction * ABSTRACT. Drory Y, Kravetz S, Florian V, Israel Study Group on First Acute Myocardial Infarction. Psychosocial adjustment in patients after a first acute myocardial infarction: the contribution of salutogenic and pathogenic variables. Arch Phys Med Rehabil 1999;80:81 l-8. Objective: To ascertain the differential and independent impact of sociodemographic, medical, and psychologic variables assessedat patients’ hospital discharge on these patients’ psychosocial adjustment in several domains of life 3 to 6 months later. Design: Two-hundred ninety Israeli male patients, aged 30 to 65 years, with a documented first acute myocardial infarction (AMI) were interviewed once before discharge and again 3 to 6 months postinfarct. Sociodemographic, medical, and psychologic data were elicited at the first interview and completed from medical information in the hospital files. Psychosocial adjustment in seven significant life domains was evaluated by the Psychosocial Adjustment to Illness Scale-Self-Report Version (PAIS-SR) at the second interview. Hierarchical regression analysis was used to examine the relation between the sociodemographic, medical, and psychologic variables at discharge to psychosocial adjustment in the different life domains 3 to 6 months later. Results: Psychologic variables, such as depression, sense of coherence, and social support, and the sociodemographic variable of educational level at discharge predicted a relatively substantial amount of variance in psychosocial adjustment in most PAIS-SR-measured life domains. Low to moderate relations were found between such medical variables as Killip class, heart disease before AMI, other medical conditions, and perceived health before first AM1 and psychosocial adjustment in specific life domains. The results also raised the possibility

*Investigators who participated in the Israel Study Group on First Acute Myocardial Infarction, and their institutional affiliations: Yaacov Drory, MD, Uri Goldbourt, PhD, Hanoch Hod, MD, Elieser Kaplinsky, MD, Emique Z. Fisman, MD, Michael Motm, MD. Chaim Sheba Medical Center. Tel Hashomer: Yeheskiel Kishon. MD. Michael Kriwisky, MD, Wolfson Medical Center, Holon; Daniel David, MD, Hana Pauzner, MD, Meir Hospital, Kfar-Sava; Itzhak Shapira, MD, Amos Pines, MD, Margalit Drory, MSW, Arie Roth, MD, Shlomo Laniado; MD, Ichilov Hospital, Sourasky Medical Center, Tel Aviv: Zvi Schlesinger, MD, Assaf Harofeh, Zerifin; Alejandro Solodky, MD. Samuel Sclarovskv. MD Beilinson Medical Center. Petach-Tikvah: Izhar Zahavi. MD: Menachem Cane;;, MD, Hasharon Hospital, Peiach-Tikvah; l& Lear, MD: Laniado Hospital, Netanya; Victor Florian, WD, Bar Ilan University, Ramat Gan. Submitted-for publication June 16, 1998. Accepted in revised f&m December 17, 1998. Supported in part by grants from the Israel Health Ministry; the National Insurance Institution; the Administrator General, Israel Justice Ministry; and the Tel-Aviv University Research Fund. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any oreanization with which the authors are associated. ieprint requests to Yaacov Drory, MD, Cardiac Rehabilitation Institute, Chaim Sheba Medical Center, Tel-Hashomer, 52621, Israel 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003.9993/99/8007-5078$3.00/O

that part of the impact of the medical variables at discharge on psychosocial adjustment 3 to 6 months later may have been mediated by the psychologic variables. The centrality of the psychologic and domestic life domains to psychosocial adjustment in post-AM1 patients was also suggested by the results. Conclusions: Both external and internal pathogenic (depression) and health proneness variables (sense of coherence and social support) at discharge predict psychosocial adjustment in most life domains 3 to 6 months after AMI. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation ORONARY ARTERY DISEASE is a chronic physical illness that can curtail both physical and psychosocial C well-being and thus may adversely affect psychosocial adjustment in important life domains.1-7 Criteria for adjustment to coronary artery disease should therefore be multidimensiona1,3,7-10 involving interactions between the individual and his or her physical and social environments as well as intrapsychic processes.3Because of the multidimensional nature of its impact on quality of life, medical, sociodemographic, and psychologic variables have been shown to influence psychosocial adjustment to coronary artery disease.2,9s11 Extensive research has focused on the contribution of variables selected from these specific sets of characteristics to a single criterion of adjustment to coronary artery disease.6J2-16Furthermore, studies have been carried out to examine the relations between a variety of predictor variables and a variety of outcome variables.7-10J7-24 Few investigations, however, have been designed to disclose the differential relation of specified sets of independent variables to psychosocial adjustment in a variety of significant life domains.9,10,20,22 This study attempted to clarify the differential contribution of representative medical, sociodemographic, and psychologic variables to psychosocial adjustment in important domains of life. It was, in part, motivated by Drory and Florian’slo multivariate investigation of the psychosocial adjustment of male participants in a cardiac rehabilitation program. In keeping with the former study’s focus on the impact of psychologic variables on this adjustment, we were particularly concerned, in the current study, with the independent contribution of systematically selected psychologic variables to this adjustment. Drory and Florian’sr” investigation was based on the assumption that individuals can be characterized by health proneness traits that insulate them against such stressful conditions as coronary artery disease. They examined the relation between a variety of variables (medical, sociodemographic, and psychologic) and their research participants’ responses to Derogatis’ Psychosocial Adjustment to Illness Scale-Self-Report Version (PAIS-SR).25However, they were particularly interested in the contribution of the personality trait of hardiness to psychosocial adjustment in the life domains tapped by their multidimensional measure. In their study, hardiness represented health proneness, Arch

Phys

Med

Rehabil

Vol 80, July

1999

812

PSYCHOSOCIAL

ADJUSTMENT

and individuals high on hardiness were assumed to be capable of coping successfully with both illness-related and environmental sources of stress. A major conclusion of Drory and Florian’slo study was that hardiness was the most salient correlate of psychosocial adjustment in six of the seven life domains measured by PAIS-SR. This study also applied the PAIS-SRz5 to evaluate the differential and independent influence of sociodemographic, medical, and psychologic variables on psychosocial adjustment in coronary artery disease patients across seven separate life domains. As its theoretical framework, however, it used Antonovsky’s26s21theory of salutogenesis, which is a comprehensive variant of the health proneness paradigm. This theory furnishes an integrated conceptualization of the processes that protect an individual’s health in stressful situations and help prevent the development of illness. It also provides a theoretical account of effective coping after the occurrence of negative health events such as the outbreak of illness.26,27 Antonovsky26 developed the theory of salutogenesis as a complement to the traditional pathogenic model of illness. He replaced the “dichotomous classification of people as healthy or diseased in favor of their location on a multidimensional health ease/disease continuum.” Instead of focusing on the conditions that cause and maintain illness, the salutogenic model emphasizes the factors that help a person maintain or attain a position in the more healthy areas of the multidimensional health/illness space. Antonovsky26,27chose the term “sense of coherence” to refer to these salutogenic dispositions. He used Lazarus and Folkman’s28theory of coping to explicate the manner in which sense of coherence positively influences states of health. Accordingly, individuals with a strong sense of coherence perceive life circumstances as lessthreatening, believe in their ability to cope with threatening situations, and actually cope effectively with such situations. Sense of coherence is a perceived internal resource that helps the individual confront stressful situations and promotes adjustment to illness. Antonovsky26 defines sense of coherence as “a ‘global’ orientation; a way of viewing the world; a dispositional orientation; rather than a response to a specific situation.” According to Antonovsky’s theoretical approach, sense of coherence is made up of the following three interrelated components: (1) comprehensibility: the extent to which an individual views the internal and external stimuli that confront him or her as cognitively meaningful and as either predictable or explicable; (2) manageability: the extent to which an individual perceives available resources as sufficient to meet the demands posed by internal and external stimuli; and (3) meaningfulness: the extent to which the individual feels that life makes sense emotionally and that at least some of the everyday problems and demands of life are more of a challenge than a burden.26,27Antonovsky repeatedly emphasized that these three components of sense of coherence are entwined in a knot that cannot be untied and, therefore, recommended that scientific use should only be made of the global construct. Recently, Antonovsky’s claim regarding the influence of sense of coherence on physical and mental health has received empirical support.2g-32Furthermore, these studies confirmed, to a great extent, his theoretical argument that persons with a strong sense of coherence will perceive illness as less distressful. To compare the salutogenic approach to coping with stress and chronic illness with the pathogenic approach, and because of evidence that sense of coherence negatively overlaps depression30 depression was included among the psychologic variArch

Phys

Med

Rehabil

Vol 80, July

1999

AFTER

AMI,

Drory

ables that we investigated. Lately, researchers and clinicians in the area of cardiac disease have begun to pay increasing attention to depression.33 Symptoms of depression have frequently been found to be correlates of myocardial infarction, and all levels of clinical depression are thought to characterize approximately 50% of patients a short time after hospitalization for a myocardial infarction. 33,34Clinically significant depression predicts such health outcomes as mortality,17,35,36occurrence of angina,36 recurrent hospitalizations,15,37arrhythmia,38 and physical functioning. 3g Mental health during hospitalization and early convalescence is related to recovery and return to appropriate functioning in such significant life domains as sexual and vocational behavior. l2 Most models of coping with stress and the psychosocial consequences of illness consider sense of coherence and depression as inner resources or deficits; social support, however, is defined as an external resource.12,40To tap both internal and external resources, this study included perceived social support among its psychologic variables. Parameters of social networks and social support have been shown to influence the long-term consequences of myocardial infarction.39e41Social support has been found to positively affect both the short-term and long-term adjustment and psychosocial well-being of persons after a myocardial infarction. Furthermore, there is evidence that social support moderates the negative impact of stressduring convalesence.6,12,40,42,43 A weakness of most studies of the contribution of sense of coherence and other psychologic variables to coping successfully with stressful situations is that they are usually crosssectional. Therefore, they leave open the question of the direction of the influence between these variables and the reduction of stress.Drory and Florian’O have recommended that longitudinal research, especially with samples from populations whose medical condition puts them at risk for a great deal of stress,be carried out to answer this question. This study, which simultaneously investigated the impact of theoretically and clinically meaningful psychologic, sociodemographic, and medical variables at discharge and specific outcome variables 3 to 6 months later, followed this recommendation. METHODS Subjects Two hundred ninety Israeli men with documented first acute myocardial infarction (AMI) who were admitted into any of eight medical centers in central Israel participated in this study. Their mean age was 52 years (range 30 to 65), SD = 8 years. In years of education the mean was 12 years, SD = 4 years. Of these men, 29% were born in Asia or Africa; the others were born in Europe, America, or Israel. In 75% of the patients, this AM1 was the first clinical manifestation of coronary artery disease. Angina1 symptoms were reported by 12% of the subjects less than 4 months before AMI. A history of coronary artery disease ~4 months before AM1 was disclosed in 13% of subjects. Those men with a previous AM1 were excluded. One additional medical problem was reported by 33%, two by 23%, and three or more by 22%. Sixty-three percent reported very good or excellent health before AMI. On admission, 83% were classified Killip class 1; 16%, Killip class 2; and l%, Killip classes 3 and 4. The diagnosis of AM1 was established by the presence of at least two of the following three criteria: (1) typical chest pain (of no less than 20 minutes’ duration); (2) creatine kinase (CK) elevation 21.5 times the upper limit of normal or CK-MB fraction 15% when simultaneous reference CK exceeded the upper limit of normal; and (3) electrocardio-

PSYCHOSOCIAL

ADJUSTMENT

graphic changes compatible with Q wave or a non-Q wave AMI. Sociodemographic interview questions. The time 1 structured interview (which took place before discharge, about 1 week after AMI) incorporated questions on age, education, employment status, marital status, and country of origin, which were read to the patients by the interviewers who also recorded the verbal responses. Medical variables. Information on the medical variables was collected from two sources: patient records and the structured interview. Infarction severity, as indicated by Killip class, was taken from patient records. The presence and duration of coronary artery disease symptoms before first AM1 and the presence of other medical conditions were assessedby structured interview questions. The latter variable was assessed by a checklist of 40 medical problems (eg, hypertension and chronic obstructive pulmonary disease). Perceived health was determined by a l-item question in the structured interview asking the patient to rate his overall health in the past year by checking one of five response categories, ranging from poor (1) to excellent (5). This l-item question has been found to be strongly related to patient mortality and morbidity.44 Psychologic variables. Psychologic variables were measured by three standard questionnaires incorporated in the time 1 structured interview-ie, the Beck Depression Inventory,45 the Sense of Coherence Scale,26 and the Multidimensional Scale of Perceived Social Support.46 The Beck Depression Inventory-a 21-item questionnaire-is widely used to assess subclinical and clinical depression. High scores indicate high levels of depression; low scores indicate low levels of depression. The reliability of the Beck Depression Inventory, as indicated by the Cronbach alpha estimate of internal consistency, was 84. The Sense of Coherence Scale-a 29-item questionnaire-measures personality resources that are assumed to indicate the individual’s ability to maintain physical and mental health.26 Higher scoresindicate higher levels of this ability; lower scores indicate lower levels. Reliability of the Sense of Coherence Scale, as estimated by Cronbach’s alpha, was 87. The Multidimensional Scale of Perceived Social Support-a 12-item scale-assesses the individual’s present perception of the extent to which social support is available from family, friends, and a significant other.46 Higher scores indicate higher levels of perceived availability; lower scores indicate lower levels. Reliability of the Multidimensional Scale of Perceived Social Support, as estimated by Cronbach’s alpha, was .90. Psychosocial adjustment to illness. The self-report version of the Psychosocial Adjustment to Illness Scale-Self-Report Version (PAIS-SR)2swas used to evaluate the patients’ psychosocial adjustment at time 2. This 46-item multidimensional scale was designed to assess medical patients’ psychosocial adjustment by asking them to report both the degree to which their illness interferes with activities related to psychosocial adjustment and the degree to which they are able to carry on with these activities despite their illness. PAIS-SR measures adjustment in the following seven psychosocial domains: (1) health care orientation (current health posture and the degree to which it contributes to positive adjustment to illness and its treatment); (2) vocational environment (interference with job performance and adjustment and satisfaction with job, which is related to present illness); (3) domestic environment (difficulties produced in the home or usual family environment by present illness); (4) sexual relationships (any changes in sexual behavior or relationship because of the present illness or its consequences); (5) extended family relationships (impairment

AFTER

AMI,

813

Drory

by present illness of contact and other forms of socialization with family living outside the household); (6) social environment (degree to which illness interferes with social and leisure activities); and (7) psychologic distress (the degree to which psychologic difficulties are associated with present illness). Response categories for each item range from 0 (no problems) to 3 (serious problems). Thus, a high score in a specific domain indicates poor psychosocial adjustment in that domain. On the basis of a cardiac patient sample, Derogatis and Lopezz5 reported Cronbach alpha coefficients from .76 to .85 for all domains except for the extended family and health care orientation domains. Alpha coefficients for the latter were .62 and .47, respectively. The alpha coefficients for the Hebrew version of PAIS-SR applied in this study ranged from .60 to .89. PAIS-SR has been validated as a measure of psychosocial adjustment both in terms of clinician-patient agreement47 and in terms of long-term postsurgery recovery.48Thus, statistically significant positive correlations were uncovered between physicians’ and physiotherapists’ global ratings of patient psychosocial adjustment and the patients’ total PAIS-SR scores.47 Furthermore, such personality variables as preoperative assessedperfectionist body image ideals and frustration tolerance predicted PAIS-SR-measured psychosocial adjustment in the expected direction in several life domains after pelvic pouch surgery for ulcerative colitis.48 Stady design. This research is part of a larger, follow-up longitudinal study that focused on patients admitted to the hospital with a first AMI. For the purpose of this larger study, 531 male patients with a first AM1 who were between the ages of 30 and 65 years and who were hospitalized in one of eight medical centers located in central Israel were identified as research participants. Our research relates to participants who were interviewed twice: once before discharge-ie, about 1 week after AM1 (time 1); and again, 3 to 6 months after AM1 (time 2). Two hundred forty-one patients were not interviewed at time 2 because of such medical and technical difficulties as mortality and lack of a correct address or telephone number. Twenty-live of the research participants who were contacted at time 2 refused to be interviewed. Statistical analyses of the differences between the 290 patients who were interviewed at time 2 and the 241 patients who were interviewed only at time 1 did not uncover statistically significant differences for any of this study’s independent variables. Data collected at both times included sociodemographic, psychologic, and medical variables. Structured interviews, including self-report questionnaires, were conducted in the hospital at time 1 and took place in the outpatient clinics or the patient’s home at time 2. Additional medical information was obtained from the medical charts at time 1. Statistical analysis. Pearson product moment correlation analysis and hierarchical stepwise regression analysis were used to examine the relation of the sociodemographic, medical, and psychologic variables to psychosocial adjustment postAMI. Because of the large sample of participants in this study, the level of statistical significance was defined asp < .Ol. RESULTS Table 1 presents the means and standard deviations of the PAIS-SR measures of the research participants’ psychosocial adjustment to the AM1 in each of the seven life domains. Becausemost of the PAIS-SR items askrespondents to compare their present condition with their pre-illness condition, normative data for the PAIS-SR based on samples of persons without a known illness are not available. However, the values in table 1 are quite similar to those obtained from a sample of 128 men with coronary artery disease who were taking part in a Arch

Phys

Med

Rehabil

Vol 80, July

1999

814

PSYCHOSOCIAL Table Pais-SR

1: Mean Scores and Standard of the PAIS-SR Domains

Deviations

Mean

SD

Health care Vocational

.92 .93

.44 .74

Domestic Sexual

.49 .48

.44

Extended Social

Domain

ADJUSTMENT

family

.59 .41

.23 .47

Psychologic

.67 .53

.65

long-term cardiac rehabilitation program.‘O The largest discrepancy between the samples appeared in the vocational environment domain with the 128 long-term rehabilitants reporting more positive vocational adjustment (ie, mean for this sample in the vocational environment domain is 56). Given the relatively short period that has passed since our study’s sample first AMI, this finding is understandable. For both the sample of our research participants 3 to 6 months after a first AMI, and the sample of long-term rehabilitants, the means in all of the PALS-SRdomains were lower than 1. Because on the PAIS-SR 0 to 3 response scale, a score of 1 indicates mild problems, the participants in both samples can be considered to be reporting a reasonable degree of psychosocial adjustment to their cardiac illness. Hierarchical regression analysis was used to investigate the independent contribution of three categories of research variables-sociodemographic, medical, and psychologic-to the patients’ subsequent psychosocial adjustment in the seven PAIS-SR domains. The variables were entered in the analysis according to the following sequence: (1) sociodemographic variables; (2) medical; and (3) psychologic. Tables 2, 3, and 4 present the results of each step of this analysis, respectively. Psychologic and sociodemographic variables predicted the largest amount of variance in psychosocial adjustment across a variety of significant life domains 3 to 6 months after a first AM1 (tables 2, 3, 4). Medical variables appear to have contributed far less to psychosocial adjustment. Thus, the introduction of the medical variables to the regression analysis at step 2 added relatively little to the amount of variance accounted for at step 1 (table 3). Of the four medical variables added at step 2, Killip class, other medical conditions, and perceived health before first AM1 contributed statistically significant amounts of variance in psychosocial adjustment in a number of the PAIS-SR domains. When the psychologic variables were added at step 3, however, the contribution of the medical variables to the patients’ psychosocial adjustment was further reduced. Whereas, at step 2, the relation of perceived health before first AM1 to psychosocial adjustment in three domains (domestic, social, and psychologic) and the relation of other medical conditions to psychosocial adjustment in five domains (vocational, domestic, extended family, social, and psychologic) were statistically significant, at step 3, only the Table

2: Standardized

Regression

Health

Variables

Coefficients

of Hierarchical

*pi + p<

.Ol. .OOl.

Arch

Phys

Med

Rehabil

Vol 80, July

Drory

relation of perceived health before first AMI to psychosocial adjustment in the domestic and social domains and the relation of other medical conditions to psychosocial adjustment in the vocational domain remained statistically significant. At step 3, perceived health before first AM1 was negatively related and other medical conditions was positively related to psychosocial adjustment, indicating that low perceived health before first AM1 and the existence of other medical conditions at discharge were associated with low psychosocial adjustment in the above specified life domains (ie, lower scores on the PAIS-SR signify higher degrees of psychosocial adjustment). The positive relation of Killip class to psychosocial adjustment in the vocational, extended family, and social life domains remained statistically significant, although slightly reduced, even after the psychologic variables were introduced. This relation signifies that higher Killip classes (ie, infarction severity) at discharge predicted lower psychosocial adjustment in the abovementioned life domains. In addition, step 3 of the hierarchical regression revealed that coronary artery disease symptoms before the first AM1 were predictive only of psychosocial adjustment in the sexual relationships domain. Table 5 presents the correlations between the independent variables. As is evident from this table, the two medical variables whose relation with psychosocial adjustment was considerably reduced at step 3 of the hierarchical regression analysis (perceived health before first AM1 and other medical conditions) were correlated with the three psychologic variables at a statistically significant level. This table also shows that the correlations between severity of infarction as measured by Killip class and the psychologic variables were not statistically significant. Table 4 reveals that the three psychologic independent variables added at step 3 considerably increased the percentage of explained variance in psychosocial adjustment in the psychologic, extended family, and domestic domains. Of these three psychologic variables, depression explained the largest amount of variance in the patients’ psychosocial adjustment, both in terms of the number of life domains to which it was related and in terms of the amount of variance it explained in each of these domains. Depression thus predicted significant percentages of variance in psychosocial adjustment in all life domains except health care orientation. In two of these-psychologic distress and sexual relationships-depression explained comparatively large amounts of variance. The positive relation of depression to psychosocial adjustment in the above domains indicates that patients who reported higher levels of depression at discharge reported lower levels of psychosocial adjustment in these domains 3 to 6 months later. Sense of coherence was also shown to be positively related at a statistically significant level to four of the seven PAIS-SR life domains (health care orientation, extended family, social, and psychologic). Thus, patients who were characterized by higher levels of sense of coherence at discharge reported more positive psychosocial adjustment 3 to 6 months later. The relation of perceived social Regression

Analysis

for the 7 PAIS-SR

Domains: Social

Domestic

Sexual

Extended Family

-.33+

-.26+

.28+ -.22+

-.15*

.I1

.07

-

R2

AMI,

Vocational

Age Education Origin

AFTER

1999

.I4

.02

-.27’ .07

First Step Psychologic

-.32+ .I0

PSYCHOSOCIAL Table

3: Standardized

Regression

Variables

Coefficients Health

ADJUSTMENT

of Hierarchical

AFTER

Regression

Analysis

Vocational

Domestic

Sexual

-.33+

-.26+

-.24+

815

Drory for the 7 PAIS-SR

Domains:

Extended Family

Second

Social

Step Psychologic

.25+

Age Education Origin Killip

AMI,

class

.22+

-.16*

-.27+

.18+

.14*

.22+

.15* -.22+

-.33+

.23+

Heart disease before AMI Other medical conditions Perceived health before AMI R2

-

R2 change

-

.16*

.18+ -.21+

.I8 .07

.I7 .I0

.I9 .05

.I1 .09

.16* -.18+

.I9

.I8

.I2

.08

"p4.01. + p-c ,001.

support to psychosocial adjustment was negative and statistically significant in three life domains (domestic, extended family, and social). Persons with a greater extent of social support at discharge evinced more psychosocial adjustment in these domains 3 to 6 months later. The relation of two sociodemographic variables-educational level and age-to psychosocial adjustment was found to be statistically significant at step 1. At step 3, these variables continued to show a statistically significant relation to psychosocial adjustment. Whereas persons with high levels of education reported high levels of psychosocial adjustment in five life domains (vocational, domestic, extended family, social, and psychologic), older persons reported low levels of adjustment only in the sexual domain. Table 4 also reveals that the amount of variance explained by the research variables differed considerably acrossthe PAIS-SRmeasured life domains. Whereas a large percentage of variance was accounted for in the psychologic and domestic domains, a relatively small percentage of variance was explained in the health care orientation domain. Research variables accounted for intermediate amounts of variance in psychosocial adjustment in the extended family and social domains. To provide a basis by which to clarify the above findings, Pearson product moment correlations were computed among the seven PAIS-SR-measured life domains. Table 6 presents these correlations. This table shows that life domains varied in the extent of their interrelations. Furthermore, the extent to which each life domain was related to other life domains appears to parallel the amount of variance explained by the research variables for each. Thus, the psychologic and domestic Table

4: Standardized

Regression

Variables

Health

Coefficients

of Hierarchical

domains, in which the highest amounts of variance in psychosocial adjustment were accounted for by the research variables, were not only highly related to each other but also strongly associated with the five remaining domains. Health care orientation, in which the percentage of variance accounted for by the research variables was extremely low, also had relatively low correlations with the other domains. DISCUSSION Psychosocial adjustment to coronary artery disease is a complex multivariate process. In our study, several independent variables at discharge explained a relatively substantial amount of the variance in this process in most life domains measured by PAIS-SR 3 to 6 months later. In accordance with other research,4~9J0J7J9-22 although coronary artery disease is primarily a medical condition, these variables were in the psychologic and sociodemographic categories. Depression was thus found to be predictive of psychosocial adjustment in all life domains except for health care orientation. Patients who expressed a high level of depression at discharge reported low levels of psychosocial adjustment in each of these domains 3 to 6 months later. Although most of these relations were low, the relation of depression to psychosocial adjustment in the psychologic distress domain, as could be anticipated, was moderately high. This finding is in keeping with the prevalent view of depression as a major negative affect“9 and with the traditional pathogenic approach to the influence of psychologic variables on physical, mental, and psychologic well-being. This finding is also consistent with much recent research showing depression to be Regression

Vocational

Domestic

Sexual

-.31+

-.15*

-.19+

Age

for the 7 PAIS-SR

Domains:

Extended Family

Final Step

Social

Psychologic

-.18+

-.20+

.24+

Education Origin Killip Heart

Analysis

class disease

.22+ before

Other medical conditions Perceived health before Depression Perceived social support Sense of coherence R’ R* Change

.11*

AMI

.12*

.22+ .15* AMI

-.12* .17+ -.23+ .06 .06

.20 .02

.20+ -.17+ -.16+

.27+

.37 .20

.26 .07

-.16* .16*

.20+ -.21+ .32 .21

.30+

-.16* -.13* .31

-.33+ .44

.I2

.26

*p< .Ol. +p<.001.

Arch

Phys

Med

Rehabil

Vol 80, July

1999

816

PSYCHOSOCIAL Table

5: Pearson

Product

Moment

Correlations 1

Between 2

AFTER

AMI,

the Sociodemographic, 3

Drory

Medical,

4

5

and Psychologic 6

Variables

7

8

at Discharge 9

10

-

Age Education Origin Killip Heart

ADJUSTMENT

class disease

Other

before

medical

AMI

conditions

Perceived Depression

health

Perceived

social

Sense

.I1 .12x .13*

-.37+ -.Ol .09

-.03 -.04

.21+

.05

-.03

-.I0

before

AMI

-.17+ .07

support

.Ol -.20+

-.14*

.18+

of coherence

-.07

.02 .15*

.06

.09

-.21+ .04

-.31+ .24+

-.08 -.Ol

-.17+

.20+

-.39+

-

-.19+

.15*

-.53+

.51+

-.ll

-.I0 -.13*

-

.04

-.07 -.04

-.24+

-

"pi.01. +p<.001.

a predictor of poor psychosocial adjustment in coronary artery diseasepatients.8~12~15~17 The psychologic variable of sense of coherence26,27also contributed to psychosocial adjustment in four of the seven domains and was found to be the only independent variable related to the health care orientation domain. Antonovsky’s theory of salutogenesis provides a broad basis for empirical investigations and clinical applications of the understanding of the coping of persons who are experiencing stress for any number of reasons. Although, in the last few years, studies have been carried out to test Antonovsky’s theory,50 few of these studies have investigated samples of persons whose distress stems from such acute and chronic medical problems as AM1 and its consequences. The ambiguity and uncertainty associated with a first AM1 can be assumed to require the comprehensibility, manageability, and meaningfulness components that theoretically comprise senseof coherence. Thus, this study’s results regarding the impact of sense coherence on the psychosocial adjustment of a relatively large sample of men after a first AM1 provide particularly convincing support for the salutogenic approach to health proneness. The broad-spectrum impact of this variable confirms the theoretical conceptualization of sense of coherence as a personal resource that provides the individual with the means to maintain physical and psychologic wellbeing in the face of stress.26,27 In light of evidence that senseof coherence is negatively related to depression,30this finding that sense of coherence at discharge independently contributes significantly to psychosocial adjustment 3 to 6 months later is especially enlightening. In addition, educational level, a sociodemographic variable, was a predictor of psychosocial adjustment in five of the life domains, This finding accords with studies in which education was positively related to psychosocial adjustment to coronary artery disease.21 In general, education has been found to be positively associated with happiness and has even buffered the impact of such negative conditions as low pay.49 Table Variables

Health

Domestic Sexual

Vocational

family

.05

-

.27* .19* .24*

.58* .29* .33* .50*

.I0 .24*

.42*

*p<.oo1

Arch

Phys

Med

Product

Moment

Correlations

Domestic

Between SeXUal

PAIS-SR

Domains

Extended Family

Social

Psychologic

-

Health Vocational

Extended Social Psychologic

6: Pearson

Although medical variables, in general, did not explain a great deal of the variance across the seven life domains, differences were uncovered in the extent to which and the manner in which specific medical variables were related to psychosocial adjustment in these life domains. On the one hand, infarction severity, which is indexed by the relatively objective measure of Killip class, seemed to directly reduce psychosocial adjustment in the vocational, extended family, and social life domains 3 to 6 months postinfarction. On the other hand, although perceived health before first AM1 and the presence of other medical conditions at the time of a first AMI, both measured by self-report, were related to psychosocial adjustment in various domains, these relations appeared to be mediated by the three major psychologic variables examined by this study. Thus, the medical variable, other medical conditions, which in step 2 of the hierarchical regression was shown to be negatively related to psychosocial adjustment significantly in the extended family domain (table 3), may have attained this effect by increasing depression and decreasing social support and senseof coherence. Consequently, when depression, social support, and sense of coherence were entered in the regression at step 3, the impact of other medical conditions on psychosocial adjustment in the extended family domain was no longer found to be statistically significant (table 4). Evidence for this possibility is provided in table 5. As pointed out in the results, this table indicates that whereas Killip classis not related to any of the psychologic variables, other medical conditions and perceived health were significantly related to the two measures of positive health proneness, social support, and sense of coherence, and to the one measure of negative well-being, depression. In addition to showing the relative explanatory power of the different categories of independent variables, this study provides some insight into the significance of the life domains indexed by PAIS-SR. The psychologic and domestic domains were the two domains in which the percentage of variance in

Rehabil

Vol 60, July

1999

.48* .58* .59* .62*

.34" .41* ,401

.54* .45"

.54*

-

PSYCHOSOCIAL

ADJUSTMENT

psychosocial adjustment accounted for by the independent variables was highest (table 4). These particular domains were also found to be highly related to each other and to the other domains (table 6), indicating their potential as core domains of psychosocial adjustment for post-AM1 patients. Very little of the variance in psychosocial adjustment in the health care orientation domain was accounted for by the independent variables. As stated above, the relation between this domain and sense of coherence was the only one that reached statistical significance. Additionally, the correlations between this and the other domains were relatively low (table 6). Health care orientation may be more of a measure of a positive attitude toward the use of medical service than of psychosocial adjustment. As a relatively specific form of problem-focused coping (see Lazarus and Folkman2* for the distinction between problem-focused and emotion-focused coping), it could be a specific consequence of sense of coherence. Further evidence of the specificity of this domain is its relatively low reliability (Cronbach alpha = .60). Additional research is necessary to clarify the processes underlying the relations disclosed by this study. Our findings indicate that both pathogenic and salutogenic psychologic internal and external resources contribute to the psychosocial adjustment of persons with cardiovascular disease in a broad range of life domains. These findings, together with the theoretical rationale that motivated the selection of the research variables, also raise interesting questions and point in corresponding research directions regarding the possible processes that may underlie the impact of the research variables on psychosocial adjustment. Antonovsky26,27 suggests that the salutogenic influence of sense of coherence might be accomplished by means of the relation of this health proneness disposition to the choice of particular coping styles. Research, possibly using linear structural equation models, into the possibility that the relation of sense of coherence and depression to psychosocial adjustment might be mediated by specific coping styles associated differentially with the comprehensibility, manageability, and meaningfulness components of senseof coherence could provide further theoretical insight into the process of adjustment to chronic life-threatening illness. Such research would also constitute testsof the specific implications of Antonovky’s theory. Problemfocused versus emotion-focused coping2* and active versus avoidant coping29 are prominent conceptual candidates for such research because they seem to be related to the abovementioned components of senseof coherence. Furthermore, our findings should direct the clinician to assess internal and external psychologic resources when designing individual and group programs for cardiovascular rehabilitation. CONCLUSION A central finding of this study is that for patients with a first AMI, nonmedical variables-especially such internal and external positive resources as sense of coherence and social support, and such experiences of emotional distress as depression-at the time of discharge from the hospital are related to psychosocial adjustment in significant domains of life 3 to 6 months later. The findings also suggest that certain medical conditions may limit psychosocial adjustment by reducing positive resources and increasing emotional distress and that the psychologic and domestic life domains may be central to adjustment in other life domains. Acknowledgment: We gratefully acknowledge the assistance of Dov Har-Even in statistical analysis.

AFTER

AMI,

817

Drory

References 1. Croog SH, Levine S. Life after a heart attack: social and psychological factors eight years later. New York: Human Sciences Press; 1982. 2. Moos RH, Schaefer JA. The crisis of physical illness: an overview and conceptual approach. In: Moos RH, editor. Coping with physical illness. New perspectives. Vol 2. New York: Plenum Press; 1984. p. 3-25. 3. Derogatis LR, Fleming MP, Sudler NC, Della Pietra L. Psychological assessment. In: Nicassio PA, Smith TW, editors. Managing chronic illness: a biopsychosocial perspective. Washington (DC): American Psychological Association; 1995. p. 59-115. 4. Neil1 WA, Branch LG, DeJong G, Smith NE, Hogan CA, Corcoran PJ, et al. Cardiac disability: the impact of coronary heart disease on patients’ daily activities. Arch Intern Med 1985;145: 1642-7. 5. Krantz DS. Cognitive processes and recovery from heart attack: a review and theoretical analysis. J Hum Stress 1980;6:27-38. 6. Waltz M, Badura B, Pfaff H, Schott T. Marriage and the psychological consequences of a heart attack: a longitudinal study of adaptation to chronic illness after 3 years, Sot Sci Med 1988;27: 149-58. I. Mayou R, Foster A, Williamson B. Psychosocial adjustment in patients one year after myocardial infarction. J Psychosom Res 1978;22:447-53. 8. Stem MJ, Pascale L, Ackerman A. Life adjustment postmyocardial infarction: determining predictive variables. Arch Intern Med 1977;137:1680-5. 9. Trelawny-Ross C, Russell 0. Social and psychological responses to myocardial infarction: multiple determinants of outcome at six months. J Psychosom Res 1987;31:125-30. 10. Drory Y, Florian V! Long-term psychosocial adjustment to coronary artery disease. Arch Phys Med Rehabil 1991;72:326-31. 11. Blumenthal JA, Emerv CF. Rehabilitation of uatients followine myocardial infarction. J Consult Clin Psycho1 1988;56:374-81. 12. Holahan CJ, Moos RH, Holahan CK, Brennan PL. Social support, coping, and depressive symptoms in a late-middle-aged sample of patients reporting cardiac illness. Health Psycho1 1995;14:152-63. 13. Papadopoulos C. A survey of sexual activity after myocardial infarction. Cardiovasc Med 1978;3:821-6. 14. Maeland JG, Havik OE. Return to work after a myocardial infarction: the influence of background factors, work characteristics and illness severity. Stand J Sot Med 1986;14:183-95. 15. Havik OE, Maeland JG. Patterns of emotional reactions after a myocardial infarction. J Psychosom Res 1990;34:271-85. 16. Garrity TE Vocational adjustment after the first myocardial infarction: comparative assessment of several variables suggested in the literature. Sot Sci Med 1973:7:705-7. 17. Ladwig KH, Roll G, Breithardt D, Budde T, Borggrefe M. Post-infarction depression and incomplete recovery 6 months after acute myocardial infarction. Lancet 1994;343:20-23. 18. Kottke TE, Young DT, McCall MM. Effect of social class on recovery from myocardial infarction: a follow-up study of 197 consecutive patients discharged from hospital. Minn Med 1980;63: 590-7. 19. Moser DK, Dracup K. Psychosocial recovery from a cardiac event: the influence of perceived control. Heart Lung 1995;24:273-80. 20. Terry DJ. Stress, coping and coping resources as correlates of adaptation in myocardial infarction. Br J Clin Psycho1 1992;31: 215-25. 21. Helgeson V. Moderators of the relation between perceived control and adjustment to chronic illness. J Pers Sot Psycho1 1992;63: 656-66. 22. Wiklund I, Sanne H, Vedin A, Wilhelmsson C. Psychosocial outcome one year after a first myocardial infarction. J Psychosom Res 1984;28:309-21. 23. Rose GL, Suls J, Green PJ, Lounsbury P, Gordon E. Comparison of adjustment, activity and tangible social support in men and women patients and their spouses during the six months post-myocardial infarction. Ann Behav Med 1996;18:264-72. 24. Lloyd GG, Cawley RH. Distress or illness? A study of psychological symptoms after myocardial infarction. Br J Psychiatry 1983;142: 120-5. Arch

Phys

Med

Rehabil

Vol 80, July

1999

818

PSYCHOSOCIAL

ADJUSTMENT

25. Derogatis LR, Lopez MC. Psychosocial Adjustment to Illness Scale (PAIS & PAIS-SR) scoring, procedures & administration manual-I. Baltimore (MD): Clinical Psychometric Research; 1983. 26. Antonovsky A. Unraveling the mystery of health. San Francisco: Jossey-Bass; 1987. 27. Antonovsky A. Personality and health: testing the sense of coherence model. In: Friedman HS, editor. Personality and disease. New York Wiley; 1990. p. 155-77. 28. Lazarus RS, Folkman S. Stress, appraisal and coping. New York Springer; 1984. 29. Dangoor N, Florian V. Women with chronic physical disabilities: correlates of their long-term psychosocial adaptation. Int J Rehabil Res 1994;17:159-68. 30. Kravetz S, Drory Y, Florian V. Hardiness and sense of coherence and their relation to negative affect. Eur J Personality 1993;7: 233-44. 31. Petrie K, Azariah R. Health-promoting variables as predictors of response to a brief pain management program. Clin J Pain 1990;6:43-6. 32. Motzer SU, Stewart BJ. Sense of coherence as a predictor of quality of life in persons with coronary heart disease surviving cardiac arrest. Res Nursing Health 1996;19:287-98. 33. Carney RM, Freedland KE, Rich MW, Jaffe AS. Depression as a risk factor for cardiac events in established coronary heart disease: a review of possible mechanisms. Ann Behav Med 1995;17: 142-9. 34. Schleifer SJ, Macari-Hinson MM, Coyle DA, Slater WR, Kahn M, Gorlin R, et al. The nature and course of depression following myocardial infarction. Arch Intern Med 1989;149: 1785-9. 35. Frasure-Smith N, Lesperance F, Talajic M. Depression following mvocardial infarction, Imoact on 6-month survival. JAMA 1993; 2?0:1819-25. 36. Ladwig KH, Kieser M, Konig J, Breithardt G, Borggrefe M. Affect&e disorders and survival after acute myocardial infarction. Results from the Post-Infarction Late Potential Studv. Eur Heart J 1991;12:959-64. 37. Levine JB, Covino NA, Slack WV, Safran C, Safran BB, Boro JE, et al. Psychological predictors of subsequent medical care among

AFTER

38. 39.

40. 41. 42. 43. 44. 45. 46. 47.

I

Arch

Phys

Med

Rehabil

Vol 80, July

1999

48. 49. 50.

AMI,

Drory

patients hospitalized with cardiac disease. J Cardiopulm Rehabil 1996;16:109-16. Camey RM, Freedland KE, Rich MW, Smith LJ, Jaffe AS. Ventricular tachycardia and psychiatric depression in patients with coronary artery disease. Am J Med 1993;95:23-8. Ryan TJ, Anderson LJ, Antman EM, Braniff BA, Brooks NH, Califf RM, et al. ACC/AHA guidelines for the management of patients with acute myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Acute Myocardial Infarction). J Am Co11 Cardiol 1996;28:1328-428. Shumaker SA, Czajkowski SM, editors. Social support and cardiovascular disease. New York Plenum Press; 1994. Berkman LF. The role of social relations in health promotion. Psychosom Med 1995;57:245-54. El1 K, Haywood LJ. Social support and recovery from myocardial infarction: a panel study. J Sot Service Res 1984;7:1-19. Moser DK. Social support and cardiac recovery. J Cardiovasc Nursing 1994;9:27-36.- Gold M. Franks P. Erickson P. Assessincr the health of the nation. The predictive validity of a preferencelbased measure and selfrated health. Med Care 1996;34: 163-77. Beck AT, Ward CH, Mendelson M, Mock JE, Erbaugh JK. An inventory for measuring depression. Arch Gen Psychiatry 1961;4: 561-71. Zimet GD, Dahlem MW, Zimet SG, Farley GK. The Multidimensional Scale for Perceived Social Support. J Pers Assess 1988;52: 30-41. Stubbing DG, Haalboom P, Barr P. Comparison of the Psychosocial Adjustment to Illness Scale-Self Report and clinical judgment in patients with chronic lung disease. J Cardiopulm Rehabil 1998;18:32-36. Weinryb RM, Gustavsson JP, Barber JP. Personality predictors of dimensions of psychosocial adjustment after surgery. Psychosom Med 1997;59:626-31. Argyle M. The psychology of happiness. London: Methuen; 1987. Antonovsky A. The structure and properties of the Sense of Coherence Scale. Sot Sci Med 1993;36:725-33.