Health-Related Quality of Life Among First-Degree Relatives of Patients With Obsessive-Compulsive Disorder in Italy Umberto Albert, M.D., Ph.D. Virginio Salvi, M.D. Paola Saracco, Psy.D. Filippo Bogetto, M.D. Giuseppe Maina, M.D.

Objective: The aim of this study was, first, to examine health-related quality of life among relatives of Italian patients with obsessive-compulsive disorder and, second, to search for potential predictors of quality of life among these relatives. Methods: Health-related quality of life was assessed among 64 nonpsychiatrically ill family members of 48 patients with obsessive-compulsive disorder by using the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36). The YaleBrown Obsessive-Compulsive Scale (Y-BOCS) and the Structured Clinical Interview for DSM-IV Axis II Disorders were used to assess obsessive-compulsive disorder among the patients. Mean SF-36 scores of participants were compared with expected scores for 2,031 persons from the Italian general population (Italian norms) by using the one-sample t test. Correlates of health-related quality of life were examined by means of independent-sample t tests and Pearson correlations; variables significantly associated with SF-36 subscales were entered into a stepwise multiple linear regression analysis with the physical and mental components as dependent variables. Results: Relatives of patients with obsessive-compulsive disorder showed a greater impairment in health-related quality of life in the SF-36 subscales of role limitations due to physical health, vitality, social functioning, role limitations due to emotional problems, and mental health. When a stepwise multiple linear regression analysis was performed with the physical component and the mental component of the SF-36 as dependent variables, female gender, older age, and the total score on the Family Accommodation Scale predicted a poorer score on the physical component, whereas the only predictor of a poorer score on the mental component was the patient’s Y-BOCS total score. Conclusions: The study provides evidence that obsessive-compulsive disorder impairs health-related quality of life among family members of patients with obsessive-compulsive disorder, even among healthy family members. Involving family members in the treatment of obsessive-compulsive disorder could improve their perceived quality of life. (Psychiatric Services 58:970–976, 2007)

The authors are affiliated with the Department of Neuroscience, University of Turin, Italy, Via Cherasco 11, Torino 10126, Italy (e-mail: [email protected]). 970

PSYCHIATRIC SERVICES

O

bsessive-compulsive disorder has a major impact on the lives of patients and anyone who lives with them. Several self-report surveys found that the disease substantially interferes with patients’ daily activities; disrupts family, social, and working life; and disturbs emotional well-being—all of which reduce quality of life (1–3). In recent years, health-related quality of life has been specifically investigated among individuals with obsessivecompulsive disorder with tools such as the Medical Outcomes Study 36Item Short-Form Health Survey (SF-36), which allows comparisons with general population norms. The first study that examined health-related quality of life in this population suggested that patients with obsessive-compulsive disorder appear similar to persons in the U.S. general population in the domains of physical health but are impaired in the domains of mental health, including social functioning (4). Similar findings emerged from a Spanish study (5), which found that compared with persons in the general population, patients with obsessive-compulsive disorder were more impaired in all domains except for physical functioning. Other authors, using the same instrument, confirmed the poorer health-related quality of life of patients with obsessive-compulsive disorder in all domains except for physical functioning, as compared with community norms (6,7).

♦ ps.psychiatryonline.org ♦ July 2007 Vol. 58 No. 7

Moreover, quality of life among patients with obsessive-compulsive disorder is equally compromised (8) or even more compromised in some dimensions, such as psychological wellbeing and social relationships, than the quality of life among patients with other chronic psychiatric disturbances, such as schizophrenia (9). All of these findings suggest that obsessive-compulsive disorder is a severe condition, with a high impact on patients’ health-related quality of life. Obsessive-compulsive disorder, moreover, appears to impair family functioning in several ways. Families of patients with obsessive-compulsive disorder are often more engaged in the illness than families of other psychiatric patients—that is, compulsions usually involve family members and the home itself (10,11). Van Noppen and colleagues (12) hypothesized that family members of patients with obsessive-compulsive disorder move along a continuum that ranges from participating and assisting in the rituals to resisting and overtly opposing the rituals. Accommodation is the term used to refer to the familial responses that are specifically related to obsessive-compulsive symptomatology: it includes behaviors such as feeling obliged to assist a relative with obsessive-compulsive disorder when he or she is performing a ritual or respecting the rules that obsessive-compulsive disorder imposes on the patient and, at the opposite side, interfering with the rituals or actively opposing them (10). Such accommodating behaviors have been investigated by using different methods. Shafran and colleagues (13) administered a questionnaire to 88 family members of individuals with obsessive-compulsive symptoms: 60% of the family members were involved to some extent in rituals performed by the affected family member. Other researchers developed a specific instrument, the Family Accommodation Questionnaire (FAQ) (14), which assesses the nature and frequency of accommodating behaviors of family members of persons with obsessive-compulsive disorder. The investigators found a high degree of accommodation; moreover, accommodation of the family members was PSYCHIATRIC SERVICES

related to distress among patients and relatives. Another study from the same group, found that 88% of families accommodated the symptoms of 36 adult patients with obsessive-compulsive disorder and that there was a correlation between high accommodation scores (as measured with the Family Accommodation Scale [FAS]), dysfunctional family interactions (as measured with the Patient Rejection Scale), and family distress and burden (15). In another study, 73 relatives of patients with obsessive-compulsive disorder were administered the FAQ and other rating scales measuring anxiety and depression; families that displayed a high degree of accommodation showed high levels of anxiety and depression (16). Besides being directly involved in patients’ symptoms, family members of patients with obsessive-compulsive disorder frequently suffer from the consequences of living with and caring for people with a chronic and disabling disease (17). Various authors have defined this kind of nonspecific stress as family burden; it includes feelings of frustration, anger, and guilt, as well as loss of social activities. Magliano and colleagues (18) were the first to investigate the burden perceived by 32 key relatives of patients with obsessive-compulsive disorder compared with that perceived by 26 key relatives of patients with depression. Moderate to severe family burden was detected in both groups, with half to three-quarters of relatives of patients with obsessivecompulsive disorder experiencing poor social relationships, difficulty in taking holidays, and neglect of hobbies. Furthermore, relatives of patients with obsessive-compulsive disorder displayed depressive feelings more frequently than relatives of patients with depression. Cooper (10) administered to 181 relatives of patients with obsessivecompulsive disorder a ten-item questionnaire that addressed the disturbances caused to family members by behaviors related to obsessive-compulsive disorder. The authors found that 82% of relatives displayed at least some disruption in personal and social life, whereas over 60% reported problems like marital discord, loss of

♦ ps.psychiatryonline.org ♦ July 2007 Vol. 58 No. 7

leisure, and financial problems as a result of their relatives’ disorder. Also, Black and colleagues (19) investigated the functioning of 15 spouses of patients with obsessive-compulsive disorder: 60% reported that caring for the relative with obsessive-compulsive disorder was burdensome, and nearly the two-thirds directly participated in the patients’ rituals. Although obsessive-compulsive disorder has been shown to exert a great influence on family functioning, health-related quality of life of relatives living with patients with obsessive-compulsive disorder has never been systematically investigated. The understanding of which relatives are more prone to experience a worsening in quality of life, moreover, could lead to the creation of specific strategies aimed at improving the family environment and consequently improving the family members’ quality of life. Previous reports have shown that higher levels of family accommodation and caregiver burden predicted poorer family functioning and poorer response to treatment, including exposure and response prevention and pharmacotherapy. Moreover, reducing accommodation of obsessivecompulsive disorder symptoms or reducing caregivers’ burden has proved useful in previous studies (20,21). Therefore, the aim of the study presented here was, first, to examine health-related quality of life among relatives of patients with obsessivecompulsive disorder and, second, to search for potential predictors of quality of life among these relatives.

Methods Sample We enrolled family members of patients with a principal diagnosis of obsessive-compulsive disorder. The presence of obsessive-compulsive disorder was determined with a score of 16 or greater on the YaleBrown Obsessive-Compulsive Scale (Y-BOCS). Patients were enrolled from the Mood and Anxiety Disorders Unit of the University of Turin, Italy. This is a tertiary referral center located within the university hospital and specialized in the treatment of patients with obsessive-compulsive disorder. 971

All patients gave their informed consent before enrollment of a family member in the study. The protocol was reviewed and approved by the ethical committee of the Regione Piemonte. A systematic face-to-face interview that consisted of structured and semistructured components was used to collect data from patients. Diagnostic evaluation and axis I comorbidities were recorded with the Structured Clinical Interview for the DSM-IV Axis I Disorders (SCID-I) (22). For patients, all sociodemographic and illness characteristics were obtained through the administration of a semistructured interview, developed and used in previous studies (23–26). Information was gathered in four areas. First, sociodemographic data were obtained—that is, age, sex, marital status (single, married, divorced, or widowed), and years of education. Second, information was gathered on the onset of obsessive-compulsive disorder—that is, disease onset was coded as the first month of occurrence of obsessive and compulsive symptoms, when at least one of the symptoms caused marked distress, was time-consuming (more than one hour a day), or interfered with the person’s normal daily functioning (normal routine and occupational and social activities). Third, information was gathered on obsessive-compulsive symptomatology. For each patient up to three primary obsessions and compulsions were listed by using the symptom checklist found in the YBOCS (27,28). The Y-BOCS measures the severity of obsessive-compulsive disorder. Five items on the YBOCS inquire about obsessions and five items inquire about compulsions. Each of the ten items is rated on a 5point Likert scale ranging from 0, none, to 4, extreme. Total possible scores range from 0 to 40. Fourth, information was gathered on personality disorders. Personality status was assessed by using the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II) (29). In addition, the following rating scales were included in the assessment of patients with obsessive-compulsive disorder: the Hamilton Rating Scale for Anxiety (HAM-A) (30) and 972

the 17-item Hamilton Rating Scale for Depression (HAM-D) (31). For both rating scales the total score is derived from summation of the items. The interview and all the ratings were completed by psychiatrists with at least four years of experience in anxiety and mood disorders. High reliability and diagnostic concordance have been documented in previous reports (25,32). For each patient, we selected one or more family members on the basis of the following criteria: family member was older than 18 years, was living with the patient for at least two years, did not have any history of mental disorder, was not involved in the care of any other family member with severe physical or mental illness, and gave informed consent to participate in the study. Family members’ general information was collected by means of informal interviews; then, in order to ensure the absence of any history of mental disorder, all potentially enrolled family members were directly interviewed with the SCID-I, NonPatient version (33). Accommodation for obsessive-compulsive disorder and health-related quality of life of family members who satisfied the above-mentioned inclusion criteria were then evaluated. Family accommodation was measured by using a semistructured interview developed by Calvocoressi and colleagues (15), the FAS. Thirteen items were used. The first nine core items assessed two types of family accommodation in obsessive-compulsive disorder: participation in symptom-related behaviors (items 1–5) and modifications of functioning (items 6–9). The scores on these nine items are summed to yield a family accommodation score. Item 10 assesses relatives’ distress when engaged in accommodating behaviors; items 11–13 assess the consequences of not accommodating the patient. Higher scores on items 1–5 indicate a greater involvement in obsessivecompulsive symptoms, higher scores on items 6–9 indicate a greater modification of functioning, and higher scores on items 10–13 indicate a greater personal distress and heavier consequences of not participating in patient’s symptom-related behaviors. PSYCHIATRIC SERVICES

Health-related quality of life was assessed by using the SF-36 (34–36). The SF-36 contains eight scales for assessing physical functioning, role limitations due to physical health, bodily pain, general health, vitality, social functioning, role limitations due to emotional problems, and mental health. For each scale, possible scores range from 0, worst possible health, to 100, best possible health. Summary scales include a physical composite and a mental composite that are expressed as t scores (mean±SD= 50±10). The SF-36 has been validated for its use in Italian, and Italian norms are available for it (37). Data analysis One-sample t tests were used to compare scores on SF-36 subscales of the family members of patients with obsessive-compulsive disorder with expected scores, which were based on 2,031 persons from the Italian general population (Italian norms) (37). We then examined the demographic and clinical correlates of quality of life first by comparing SF-36 scores with variables of the participants (sex, age, kinship, and FAS scores) and second by comparing SF-36 scores with variables of their relative with obsessive-compulsive disorder (sex, age, age at onset of obsessive-compulsive disorder, DSM-IV axis I and II comorbidity, YBOCS total score, scores on the obsession and compulsion subscales of YBOCS, and HAM-A and HAM-D scores). Comparisons were made by way of independent-samples t tests in the case of dichotomous variables and by way of Pearson product-moment correlations in the case of continuous variables. Finally, the variables found to be significant at the conventional p level of .05 were entered into a stepwise multiple linear regression analysis with, as dependent variables, the physical and mental components of the SF-36.

Results Seventy-seven family members were initially interviewed. All relatives who were approached to participate gave their consent, as did the patients they lived with. Seven family members had a history of mental disorder, three were caring for other severely

♦ ps.psychiatryonline.org ♦ July 2007 Vol. 58 No. 7

ill relatives, and three had been living with their relative with obsessivecompulsive disorder for less than two years. These family members failed to meet the inclusion criteria and were excluded from the subsequent assessments, thus 64 family members participated in the study. As shown in Table 1, among the participants 26 (41%) were men, and the mean age was 49.9 years. Fiftytwo percent were parents, 36% were spouses, 6% were offspring, and 6% were siblings. Table 1 also reports characteristics of the 48 patients with obsessive-compulsive disorder whose family members participated into this study. Mean SF-36 scores for family members of patients with obsessivecompulsive disorder along with population-based means (Italian norms) are listed in Table 2. Because our sample differed from the sample of the Italian population in gender ratio, we performed a different analysis considering males and females separately. Compared with the Italian population, relatives of patients with obsessive-compulsive disorder showed a greater impairment in health-related quality of life in the subscales of role limitations due to physical health, vitality, social functioning, role limitations due to emotional problems, and mental health. These differences, except for role limitations due to physical health, remained significant when men and women were examined separately. Although the mean age of our sample (49.9±12.8 years) did not differ from the mean age of the Italian population sample (47.73 years), we performed a one-sample t test comparing mean scores of participants in the age ranges 35–44 (N=11), 45–54 (N=17), and 55–64 (N=25) with the mean scores of the general Italian population for the corresponding age ranges (in the other age ranges there were too few persons). Although the statistical power was reduced, some differences between participants in our study and the Italian population remained statistically significant (vitality, social functioning, and mental health for those in the age range 45–54, and role limitations due to physical health for those in the age range 45–54, general health, PSYCHIATRIC SERVICES

vitality, social functioning, role limitations due to emotional problems, and mental health for those in the age range 55–64). All significant correlations among continuous variables and one or more SF-36 subscales are shown in Table 3. Concerning dichotomous variables, female family members scored significantly lower than male family members on subscales of the SF-36: physical functioning (80.00 versus 90.77; t=2.34, df=62, p=.023), bodily pain (59.13 versus 80.69; t=3.52, df=62, p=.001), vitality (45.92 versus 57.69; t=2.62, df=62, p=.011), and the global physical component (47.78 versus 53.13; t=2.22, df=62, p=.030). Moreover, being a parent was associated with a greater impairment in social functioning than being a spouse (55.30 versus 72.26; t=3.16, df=3, p=.040). When patients’ variables were taken into account, relatives of patients with personality disorders had significantly lower scores in the domains of general health (57.08 versus 68.44; t=2.00, df=62, p=.050), role limitations due to emotional problems (46.14 versus 67.96; t= 2.119, df=62, p=.038), and in the global mental component (35.93 versus 42.12; t=2.075, df=62, p=.042). Other patients’ characteristics, such as gender, age, age at obsessive-compulsive disorder onset, axis I comorbidity (actual or lifetime), and HAMA total score, did not predict their relatives’ quality of life. When a stepwise multiple linear regression analysis was performed with the physical and mental components of the SF-36 as dependent variables, female gender, older age, and the FAS total score predicted a poorer score on the physical component, whereas the only predictor of a poorer score on the mental component was the Y-BOCS total score (Table 4).

Table 1

Sociodemographic and clinical characteristics of 64 family members and 48 patients with obsessivecompulsive disorder Variable Family member Age (M±SD) Gender Male Female Kinship Parent Spouse Offspring Sibling Patients with obsessivecompulsive disorder Age (M±SD) Gender Male Female Age at onset of disorder (M±SD) Axis I comorbidity Current Yes No Lifetime Yes No Axis II comorbidity Yes No Y-BOCS (M±SD)a Total score Obsession subscore Compulsion subscore HAM-D (M±SD score)b HAM-A (M±SD score)c a

b

c

N

%

49.9±12.8 26 38

41 59

33 23 4 4

52 36 6 6

35.3±11.3 24 24

50 50

22.1±8.4 25 23

52 48

35 13

73 27

30 18

63 38

25.9±6.0 13.3±3.1 12.6±3.5 8.8±4.7 10.9±6.3

Yale-Brown Obsessive-Compulsive Scale. Possible scores on the Y-BOCS range from 0 to 40, with higher scores indicating more severe symptoms. Possible scores on the obsession subscale range from 0 to 20, with higher scores indicating more severe symptoms. Possible scores on the compulsion subscale range from 0 to 20, with higher scores indicating more severe symptoms. 17-item Hamilton Rating Scale for Depression. Possible scores range from 0 to 52, with higher scores indicating more severe symptoms. Hamilton Rating Scale for Anxiety. Possible scores range from 0 to 56, with higher scores indicating more severe symptoms.

Discussion This study was a first attempt to examine health-related quality of life among relatives of patients with obsessivecompulsive disorder and to search for potential predictors of poorer quality of life by using a validated and highly used instrument, the SF-36. Our data show that family members of patients with obsessive-compulsive

♦ ps.psychiatryonline.org ♦ July 2007 Vol. 58 No. 7

disorder are significantly more impaired in quality-of-life measures than the Italian general population. This is true especially in the mental health components of health-related quality of life measured by the SF-36. Scores from our study population on the SF-36 subscales—vitality, social 973

Table 2

Scores on the 36-Item Short-Form Health Survey (SF-36) for family members of patients with obsessive-compulsive disorder, compared with Italian norms Family members Italian norms (M)

One-sample t test

Total (N=64 )

Males (N=26)

Females (N=38)

SF-36 subscalea

M

M

M

Physical functioning Role limitations due to physical health Bodily pain General health Vitality Social functioning Role limitations due to emotional problems Mental health

84.38 18.72 90.77 12.78 80.00 20.93 84.46

88.15 81.02

65.23 67.89 61.52 50.70 61.91

82.15 78.81 68.75 66.05 80.51

a b c d

SD

39.50 26.18 22.71 18.43 23.29

68.27 80.69 67.00 57.69 67.29

SD

42.16 22.33 17.88 18.29 27.16

Total Males Females Total (N= (N= (N= 2,031) 999) 1,032) tb

SD

65.16 59.13 57.76 45.92 58.22

37.99 25.22 25.03 17.16 19.77

78.21 73.67 65.22 61.89 77.43

54.66 41.30 57.65 41.69 52.61 41.46 76.16 55.19 20.03 59.54 19.31 52.21 20.22 66.59

–.04

Males p

tc

.971

Females p

td

1.05

.306

74.13 69.01 63.19 57.98 73.89

–2.63 .011 –1.68 –1.77 .082 .43 –1.31 .197 –.50 –4.86