Prevalence and predictors of family accommodation in obsessive-compulsive disorder in an Indian setting.

BACKGROUND: Family accommodation (FA) is a phenomenon whereby caregivers assist/facilitate rituals or behaviors related to obsessive-compulsive disorder (OCD). FA, however, has been explored primarily in the Western population, and it is unclear to what extent it might be present in diverse cultural settings. At present, little is known about the extent and predictors of FA among caregivers of adult OCD patients in India. AIMS: The study aims to assess the extent, clinical correlates, and predictors of FA in the caregivers of adults with OCD. SETTINGS AND
DESIGN: Cross-sectional study conducted in an outpatient setting in a tertiary-care hospital.
MATERIALS AND METHODS: Hundred and one adult patients of either gender with Diagnostic and Statistical Manual of Mental Disorders-5 diagnosis of OCD and 101 caregivers were included. The patients were assessed using Yale-Brown Obsessive Compulsive Scale (YBOCS), Hamilton Rating Scale for Depression (HAM-D), World Health Organization Disability Assessment Schedule Version 2.0 12-item version (WHO-DAS 2.0.12), Clinical Global Impressions Scale for Severity (CGI-S), and Clinical Global Impressions Scale for Improvement. The FA Scale-Self Rated Version (FAS-SR) was applied on caregivers after Hindi translation. STATISTICAL ANALYSIS: Descriptive statistics, group comparisons, and Pearson's product moment correlations were carried out. Multiple linear regression modeling was performed with the total FAS-SR score as the dependent variable.
RESULTS: About 92% of caregivers displayed at least some form of FA. Higher scores on HAM-D, YBOCS, WHODAS, and CGI-S were associated with higher scores on FAS-SR scale, which reached statistical significance (P < 0.01).
CONCLUSIONS: FA in OCD appears to be a frequent phenomenon. Higher FA is associated with higher symptom severity and disability, emphasizing its clinical and research relevance for future studies. Copyright:

INTRODUCTION

Obsessive–compulsive disorder (OCD) is a psychiatric disorder characterized by obsessions and compulsions[1] with an estimated prevalence of approximately 1%–3% in adults.[2] Although the symptoms of OCD are considered to be an individual phenomenon, its severity, course, and outcome need to be viewed from an interpersonal context. Existing literature suggests that family involvement in OCD possibly exerts a bidirectional influence, i.e., familial involvement may determine the course of the OCD, and conversely, the symptomatology of OCD may determine the psychological burden and stress perceived by families.[3] A construct related to family involvement in OCD, which has garnered research interest over the years, is that of family accommodation (FA). FA refers to how family members assist in compulsive rituals, provide reassurance, or modify their routines to alleviate or avoid the distress experienced by a patient suffering from OCD.[4] The concept of accommodation and the instruments developed for its quantitative assessment have provided a systematic framework for exploring how family members can be influenced by the patient's disorder.

The initial studies on FA were conducted primarily in pediatric OCD. Allsopp and Verduyn found that 70% of parents reported being involved with their child's symptoms in some way.[5] Some level of FA occurs in the families of virtually all children and youth with OCD.[678910]

Different methods have been employed to assess accommodating behaviors in relatives of OCD patients. For example, Shafran et al. applied a self-administered questionnaire in family members of individuals with OCD and found that 60% of the family members were involved to some extent in rituals performed by the patient.[11] Another group developed a clinician-administered instrument, the FA Scale (FAS) which assesses the nature and frequency of accommodating behaviors of family members of persons with OCD.[1213] Quantitative measurement of FA with the help of the newly developed tools confirmed that this phenomenon is highly prevalent in families of patients with OCD, with rates ranging as high as 89%[13] to 96.9%.[141516] However, the results have to be viewed with caution owing to methodological limitations of previous studies such as low sample size, cross-sectional study design, heterogeneous sample (not considering type of OCD/definition of chief caregiver or family/family history of OCD/anxiety disorders), and measures used for assessment of accommodation, to name a few.

A strong relationship between FA and obsessive–compulsive symptom severity has been reported and replicated.[1213161718] FA has also been associated with worse treatment outcomes in both adult and pediatric OCD patients.[4]

It is worth noting that FA emerged as one of only five variables that predicted treatment outcomes in the largest and most systematic trial of OCD treatment in childhood OCD.[19] FA is also related to poorer patient and family functioning,[812] greater symptom severity,[810121720212223] and increased distress.[918] A recent meta-analysis of the association between levels of FA and OCD symptom severity reported that FA was significantly associated with OCD symptom severity with medium effect size.[24] Based on these findings, there is growing interest by clinical researchers in developing a family-based intervention for OCD directed at reducing FA.[25]

Till date, there is a paucity of literature on FA in OCD from the Indian settings. To the best of our knowledge, only three prior studies[262728] from India have investigated the role of FA in OCD, in either a relatively smaller sample[26] or pediatric age group.[27] Since recent literature suggests that FA appears to be an important factor associated with the severity and outcome of OCD not only in the pediatric population but also adult patients, there is a need for research on FA in adult OCD patients from India. Furthermore, even though the prevalence and burden of OCD had been demonstrated across cultures, there is a need for further cross-cultural research to delineate the nature of accommodation-related behaviors across diverse populations, such as that of lower and middle-income countries like India. Patterns of caregiving and beliefs about illness vary widely in different cultures and can lead to substantially different responses to obsessive–compulsive symptoms in a patient. FA has been explored primarily in American and European western societies, and it is unclear to what extent cultural factors might impact accommodation in our cultural setting.

At present, little is known about the extent of FA among caregivers of adult OCD patients in India. FA as a construct, as well as its clinical correlates, deserve further exploration to delineate its role in mediating disease severity, functional impairment as well as treatment outcomes. Ours is the first study in India to explore the self-report of FA by caregivers of adult patients of OCD. The aim of the present exploratory study was two fold: (1) to assess the prevalence of FA in a sample of healthy family members of adults with OCD and (2) to assess the clinical correlates and predictors of FA in OCD. It was hypothesized that FA would demonstrate a significant positive association with symptom severity as well as functional impairment.

MATERIALS AND METHODS

Study design

A cross-sectional, clinic-based outpatient study conducted in the Psychiatry OPD, All India Institute of Medical Sciences (AIIMS), New Delhi, India.

Study subjects

A convenience sample of 101 individuals diagnosed with OCD according to Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5), fulfilling the selection criteria, attending the outpatient clinic constituted the patient group (Group 1). Similarly, 101 nonill first-degree relatives or spouse/partners of the patients of OCD fulfilling the selection criteria formed the caregiver group (Group 2). All individuals were recruited via nonprobability sampling.

Selection criteria for participants

Inclusion criteria for Group 1 (obsessive–compulsive disorder patients)

Meets DSM-5 criteria of OCD, with Yale–Brown Obsessive Compulsive Scale (YBOCS) score >8

The total duration of illness more than a year

Age 18–60 years

Hamilton Rating Scale for Depression (HAM-D) score ≤14

No other Axis I psychiatric disorder ascertained by Mini International Neuropsychiatric Interview Version 6 (MINI 6.0)[29]

Willing to give informed consent to participate in the study.

Inclusion criteria for Group 2 (caregivers)

For each patient, we selected one family member based on the following criteria:

First-degree relatives or spouse/partner of the patients fulfilling three out of the following four criteria:

Has the most frequent contact with the patient

Supports the patient and his care financially

Is the most frequent participant in the treatment

Is to be contacted by the clinical staff in emergency.

Taking care of the patient for last 1 year or more

Not involved in the care of any other family member suffering from severe physical or mental illness

No lifetime diagnosis of Axis-I psychiatric disorder (as ascertained by MINI 6.0)[29]

Age 18–60 years

Willing to give informed consent to participate in the study.

Exclusion criteria for both the groups:

Following were the exclusion criteria:

Any lifetime diagnosis of Axis-I psychiatric disorder (other than OCD and Mild depressive episode in Group 1) or mental retardation (as per clinical history)

Clinical history of neurological disorder, cerebrovascular disorder, head injury with loss of consciousness >5 min, or epilepsy

Clinical history of any other major medical disorder interfering with assessment (excluding medically controlled diabetes mellitus, hypertension, and hypothyroidism)

Clinical history of substance use disorder (except nicotine dependence).

Measures

Obsessive–compulsive disorder patients

A systematic face-to-face interview that consisted of structured and semistructured components was used to collect data from patients and family members. All sociodemographic and illness characteristics were obtained through the administration of a semistructured pro forma. Diagnostic evaluation and screening for other Axis I Psychiatric disorders was done with the help of Mini International Neuropsychiatric Interview Version 6.0 as well as clinical interviewing.

Sociodemographic data: age, gender, religion, occupation, monthly income, marital status, education, and contact details

Semistructured clinical pro forma: patient's presenting complaint, age at onset, the total duration of illness, course of illness, presence of other comorbid disorders, past psychiatric history, family history of psychiatric disorders, treatment details, and mental status examination including insight.

In addition, the following rating scales were included in the assessment of OCD patients: YBOCS,[30] 17-item HAM-D,[31] World Health Organization Disability Assessment Schedule Version 2.0 12-item version (WHO-DAS 2.0.12),[32] Clinical Global Impressions Scale for Severity (CGI-S), and Clinical Global Impressions Scale for Improvement (CGI-I).[33]

Caregivers/family members

The sociodemographic details of the caregivers were recorded. The FA was measured using the Hindi translated version of 19-item FAS-Self Rate (FAS-SR).[34] The first section of the FAS-SR includes an OCD symptom checklist: the relative is asked to identify the types of obsessions and compulsions experienced by the patient in the past week. The second section of the FAS-SR asks the relative to draw on the patient's OCD symptoms endorsed in Section 1 to identify accommodating behaviors in which he/she (the relative) has engaged in the preceding week. The following FA behaviors are assessed by items in the FAS-SR: (1) providing reassurance regarding OCD symptoms, (2) refraining from saying/doing things because of the patient's OCD, (3) facilitating compulsions, (4) facilitating avoidance, (5) tolerating odd behaviors/household disruption, and (6) modifying personal routines because of patient symptoms. The FAS-SR demonstrated excellent internal consistency and reliability (Cronbach's alpha = 0.90).[34]

Translation of Family Accommodation Scale-Self Rated

Before translation, the authors of the original scale[34] were contacted, and permission was obtained for Hindi translation. The WHO's translation–backtranslation methodology was used to translate the scale.[35] The final version after backtranslation was sent to the original authors of the instrument. The original authors approved the backtranslated English version with a few minor corrections. These suggestions were incorporated to prepare the final Hindi version.

Procedure

Patients attending the psychiatry outpatient clinic at AIIMS were screened for the presence of OCD by a clinical interview based on DSM-5 criteria. Those who screened positive for the presence of OCD were then assessed for suitability for inclusion into the study. A key caregiver was identified based on the selection criteria mentioned above. Both the patient and their key caregivers were included in the study, after informing them of the details of the study and obtaining written informed consent from them. Patients were then administered a socio-demographic and semi-structured clinical pro forma, YBOCS, CGI-S, CGI-I, and WHO-DAS 2.0.12 scales. The primary caregivers were administered the sociodemographic pro forma and FAS-SR (Hindi version).

Statistical analysis

Statistical analysis was performed using SPSS v20 (released 2011, IBM Corp, Armonk, NY, USA). Descriptive statistics were used for various sociodemographic and clinical variables. Measures of central tendency and distribution were calculated for the continuous variables. Frequency tables were constructed for the categorical variables. Group comparisons of FAS-SR scores (continuous variable) were made using t-test for mean differences between the two subgroups. To explore the relationship between clinical and psychosocial variables among themselves and with the continuous variables, Pearson correlation was used to explore the association between clinical variables and FA. Finally, the variables found to be significant at the conventional alpha level of 0.05 in the above analyses were then entered as predictor variables into multiple linear regression models with total FA score as the dependent variable (P to enter = 0.05; P to leave = 0.10).

Ethical aspects

Ethical clearance was obtained from the Institute Ethics Committee. Informed consent was obtained from the individuals before inclusion into the study. The individuals had the right to withdraw their consent from participation at any time after inclusion into the study. Individual's nonparticipation in the study did not have any bearing on their treatment. The identity of the individuals in the study was kept confidential. All the scales included in the study, except the FA scale-self rated version, are freely available in the public domain for research and academic purpose. Formal permission was sought from the authors of the FAS-SR version for translating in in Hindi and standardizing it for the study.

RESULTS

Sociodemographic profile

The sociodemographic profile of the patients, as well as caregivers, is shown in Table 1. The mean age of the patients was 34.12 (standard deviation [SD]: 11.39) years. There was a slight preponderance of females (55.4%), and the majority of the sample was married (64.4%). The mean age of the caregivers was 44.09 (SD: 10.09) years. Majority of the caregivers were spouses (59.4%) of the patients, followed by parents (31.7%).

Table 1

Descriptive statistics of sociodemographic variables of patients/caregivers

Variables	Mean (SD) or frequency (%)
Patients (n=101)
Age (years)	34.12 (11.39)
Gender
Male	45 (44.6%)
Female	56 (55.4%)
Marital status
Never married	35 (34.7%)
Married	64 (63.4%)
Separated/divorced	2 (1.9%)
Caregivers (n=101)
Age (years)	44.09 (10.9)
Gender
Male	76 (75.2%)
Female	25 (24.8%)
Marital status
Never married	5 (4.8%)
Married	96 (95.2%)
Separated/divorced	-
Relationship to patient
Spouse/partner	60 (59.4%)
Parent	32 (31.7%)
Sibling	6 (5.9%)
Adult child	3 (2.9%)

SD – Standard deviation

Clinical profile of obsessive–compulsive disorder patients

The clinical profile of the OCD patients is shown in Table 2. The mean duration of illness was 6.12 (SD: 5.44) years, and the mean age of onset was 27.7 (SD: 9.98) years. The mean duration of treatment was 2.46 (SD: 2.14) years. The mean YBOCS score was 27.54 (SD: 6.631). Five patients (5%) had mild illness (YBOCS 8–15), 22 (21.8%) patients had moderately severe illness (YBOCS 16–23), 39 (38.6%) had severe illness (YBOCS 24–31), and 35 (34.7%) patients had extremely severe illness (YBOCS 32–40). The course of illness was continuous for all except one patient. There was a positive family history of OCD in 5.9% of patients, and 17.8% of patients demonstrated symptoms (not amounting to a disorder) of other OC spectrum disorders. 14.9% of patients had been previously hospitalized for their illness. Insight (assessed on a clinical basis) for OC symptoms was present in 84.2%, partially present in 13.9%, and absent in 2% of the patients.

Table 2

Descriptive statistics of the clinical variables (n=101)

Variables (units)	Mean (SD) or frequency (%)
Age at onset (years)	27.703 (9.981)
Total duration of illness (years)	6.12 (5.44)
Duration of treatment (years)	2.46 (2.14)
HAM-D score	8.46 (2.304)
Y-BOCS score	27.54 (6.631)
WHO-DAS 2.0.12 score	21.37 (8.179)
CGI-S score	4.49 (0.832)
CGI-I score	3.64 (0.976)
FAS-SR score	37.28 (15.774)
Insight
Complete	85 (84.2%)
Partial	14 (13.9%)
Absent	2 (2%)
Obsessive-compulsive spectrum symptoms (not amounting to disorder)
Present	18 (17.8%)
Family history of OCD
Positive	6 (5.9%)
Previous hospitalization
Present	15 (14.9%)

HAM-D – Hamilton Rating Scale for Depression; YBOCS – Yale-Brown Obsessive Compulsive Scale; WHO – World Health Organization; WHO-DAS 2.0.12 – WHO Disability Assessment Schedule Version 2.0 12-item version; CGI-S – Clinical Global Impressions Scale for Severity; CGI-I – Clinical Global Impressions Scale for Improvement; FAS-SR – Family Accommodation Scale-Self Rated; OCD – Obsessive–compulsive disorder; SD – Standard deviation

Family accommodation in caregivers

The frequency of OCD symptoms reported by the caregivers during the past 1 week is shown in Table 3. Contamination obsessions (75.2%) were most commonly reported, followed by harming (32.7%) and sexual obsessions (18.8%). Cleaning/washing compulsions (72.3%) constituted the commonest of compulsions, followed by checking (45.5%), repeating (33.7%), and counting (22.8%) compulsions.

Table 3

Frequency of obsessive-compulsive disorder symptoms as reported during the past one week in Family Accommodation Scale-Self Rated (n=101)

OCD symptom	n (%)
Obsessions
Harming obsessions	32 (32.7)
Contamination obsessions	76 (75.2)
Sexual obsessions	19 (18.8)
Hoarding/saving obsessions	3 (2.9)
Religious obsessions	10 (9.9)
Obsessions with need for symmetry or exactness	9 (8.9)
Somatic obsessions	12 (11.9)
Miscellaneous obsessions	27 (26.7)
Compulsions
Cleaning/washing compulsions	73 (72.3)
Checking compulsions	46 (45.5)
Repeating rituals	34 (33.7)
Counting compulsions	23 (22.8)
Ordering/arranging compulsions	10 (9.9)
Hoarding/saving/collecting compulsions	2 (1.9)
Miscellaneous compulsions	61 (60.4)

OCD – Obsessive-compulsive disorder

The mean score on the FAS-SR scale was found to be 27.28 (SD: 15.7). The response frequency for each of the 19 items in Part 2 of the FAS-SR is shown in Table 4. Approximately 92.1% of caregivers demonstrated some form of accommodation behavior for at least 1 day in the past week. Thirty-nine percent of caregivers reported having provided reassurance for OCD-related worries every day for the past week. Up to 26.7% of caregivers reported directly participating in the patient's compulsions, and 29.7% reported facilitating rituals/compulsion by providing items, all for 4–6 days in a week. About 15.9% of caregivers resorted to at least one of the accommodation-related behaviors for 4–6 days in a week, and 10.9% of caregivers resorted to such behaviors every day, demonstrating high levels of FA.

Table 4

Response frequency for each item of the Family Accommodation Scale-Self Rate

Item	Number of days this past week
	Never, n (%)	1 day, n (%)	2-3 days, n (%)	4-6 days, n (%)	Every day, n (%)
1. I reassured my relative that there were no grounds for his/her OCD-related worries	8 (7.9)	6 (5.9)	16 (15.8)	31 (30.7)	40 (39.6)
2. I reassured my relative that the rituals he/she already performed took care of the OCD-related concern	11 (10.9)	6 (5.9)	18 (17.8)	27 (26.7)	39 (38.6)
3. I waited for my relative while he/she completed compulsive behaviors	16 (15.8)	14 (13.9)	23 (22.8)	25 (24.8)	23 (22.8)
4. I directly participated in my relative’s compulsions	42 (41.6)	16 (15.8)	9 (8.9)	27 (26.7)	7 (6.9)
5. I did things that made it possible for my relative to complete compulsions	33 (32.7)	13 (12.9)	15 (14.9)	22 (21.8)	18 (17.8)
6. I provided my relative with OCD with items he/she needs to perform rituals or compulsions	27 (26.7)	20 (19.8)	21 (20.8)	30 (29.7)	3 (2.9)
7. I did things that allowed my relative to avoid situations that might trigger obsessions or compulsions	18 (17.8)	22 (21.8)	28 (27.7)	22 (21.8)	11 (10.9)
8. I helped my relative make simple decisions when he/she couldn’t do so because of OCD	36 (35.6)	30 (29.7)	32 (31.7)	2 (1.9)	1 (1.0)
9. I helped my relative with personal tasks when his/her ability to function was impaired by OCD	54 (53.5)	18 (17.8)	14 (13.9)	11 (10.9)	4 (3.9)
10. I helped my relative prepare food when he/she couldn’t do so because of OCD	65 (64.4)	13 (12.9)	11 (10.9)	10 (9.9)	2 (1.9)
11. I took on family or household responsibilities that my relative couldn’t adequately perform due to OCD	20 (19.8)	22 (21.8)	27 (26.7)	19 (18.8)	13 (12.9)
12. I avoided talking about things that might trigger my relative’s obsessions or compulsions	42 (41.6)	24 (23.8)	20 (19.8)	7 (6.9)	8 (7.9)
13. I stopped myself from doing things that could have led my relative to have obsessions or compulsions	48 (47.5)	14 (13.9)	19 (18.8)	12 (11.9)	8 (7.9)
14. I made excuses or lied for my relative when he/she missed work or a social activity because of his/her OCD	86 (85.1)	11 (10.9)	2 (1.9)	2 (1.9)	0
15. I didn’t do anything to stop unusual OCD-related behaviors by my relative	31 (30.7)	13 (12.9)	24 (23.8)	26 (25.7)	7 (6.9)
16. I put up with unusual conditions in my home because of my relative’s OCD	49 (48.5)	13 (12.9)	15 (14.9)	14 (13.9)	10 (9.9)
17. I cut back on leisure activities because of my relative’s OCD	28 (27.7)	25 (24.8)	25 (24.8)	13 (12.9)	10 (9.9)
18. I changed my work or school schedule because of my relative’s OCD	97 (96.1)	4 (3.9)	0	0	0
19. I put off some of my family responsibilities because of my relative’s OCD	33 (32.7)	25 (24.8)	37 (36.6)	6 (5.9)	0

OCD – Obsessive–compulsive disorder

Association between family accommodation and clinical characteristics of the obsessive–compulsive disorder patients

The correlation matrix demonstrating a univariate association between the extent of FA and clinical characteristics of OCD patients is shown in Table 5. FAS-SR scores showed a positive correlation (P < 0.01) with HAM-D scores, YBOCS score, 12-item WHO-DAS, and CGI-S scores. Hence, higher scores of FA were associated with higher scores for depression, more severe obsessive–compulsive psychopathology, and a greater amount of disability.

Table 5

Correlation matrix for the study variables (n=101)

S.No.	Variable	1	2	3	4	5	6	7	8	9
1	Patient’s age (years)	1
2	Total duration of illness (years)	0.475**	1
3	Duration of treatment (years)	0.319**	0.601**	1
4	HAM-D score	0.185	0.113	0.373**	1
5	YBOCS score	0.043	0.043	0.071	0.512**	1
6	WHO DAS 2.0.12 score	0.045	0.006	0.187	0.559**	0.828**	1
7	CGI-S score	−0.023	0.003	0.074	0.410**	0.791**	0.723**	1
8	CGI-I score	−0.015	−0.37	0.075	0.291**	0.499	0.537	0.449	1
9	FAS-SR score	0.096	−0.009	0.172	0.589**	0.773**	0.766**	0.750**	0.482	1

Statistical analysis by Pearson’s correlation. **Correlation is significant at the 0.01 level (two-tailed). HAM-D – Hamilton Rating Scale for Depression; YBOCS – Yale-Brown Obsessive Compulsive Scale; WHO-DAS 2.0.12 – WHO Disability Assessment Schedule Version 2.0 12-item version; CGI-S – Clinical Global Impressions Scale for Severity; CGI-I – Clinical Global Impressions Scale for Improvement; FAS-SR – Family Accommodation Scale-Self Rated

Multiple linear regression modeling

Multiple linear regression analysis was performed with the FAS-SR scale score as dependent variable, and the variables initially found to be significantly correlated with accommodation in the univariate analyses as independent variables [Table 6]. Thus, the following independent variables were entered in the models: HAM-D scores, YBOCS score, WHO-DAS 2.0.12 scores, and CGI-S scores. Subsequently, the other independent variables also entered were the age of the patient, total duration of illness, duration of treatment, and CGI-I scores. The four variables, namely HAM-D score, Y-BOCS score, WHO-DAS 2.0.12 score, and CGI-S score together explained 69.9% of the variance (adjusted R2 = 0.699, F (8,92) =29.960, P < 0.001). Higher scores on HAM-D, YBOCS, WHO-DAS 2.0.12, and CGI-S were associated with higher scores on FAS-SR scale. There was no problem of collinearity among the variables. The variance inflation factor (VIF) and tolerance for the independent variables are as follows: Age: tolerance = 0.742, VIF = 1.348; total duration of illness: tolerance = 0.522, VIF = 1.915; duration of treatment: tolerance = 0.526, VIF = 1.9; HAM-D score: tolerance = 0.579, VIF = 1.728; YBOCS Score: tolerance = 0.223, VIF = 4.482; 12-WHO DAS: tolerance = 0.256, VIF = 3.9; CGI-S score: tolerance = 0.351, VIF = 2.848; and CGI-I score: tolerance = 0.698, VIF = 1.432.

Table 6

Stepwise multiple linear regression predicting family accommodation in obsessive-compulsive disorder

Family members (n=101)	FAS-SR total scorea
	Betab	t	P
Constant		−6.266	<0.001
Age (years)	0.077	1.205	0.231
Total duration of illness (years)	−0.081	−1.072	0.286
Duration of treatment (years)	0.040	0.527	0.600
HAM-D score*	0.198	2.740	0.007
Y-BOCS score*	0.301	3.008	0.003
WHODAS-12 2.0*	0.213	2.161	0.033
CGI-S score*	0.337	3.632	<0.001
CGI-I score	0.059	0.893	0.374

aR2=0.723, Adjusted R2=0.699, F (7,93)=29.960, P<0.001; bStandard coefficients; *Statistically significant at P < 0.05. HAM-D – Hamilton Rating Scale for Depression; YBOCS – Yale-Brown Obsessive Compulsive Scale; CGI-S – Clinical Global Impressions Scale for Severity; CGI-I – Clinical Global Impressions Scale for Improvement; FAS-SR – Family Accommodation Scale-Self Rated; WHO – World Health Organization; WHO-DAS 2.0.12 – WHO Disability Assessment Schedule Version 2.0 12-item version

DISCUSSION

This is the first study to examine the self-report account of FA among caregivers of OCD patients in an Indian setting. The sociodemographic profile of the patients was in keeping with previous studies.[161826] Majority of the caregivers were spouses (59.4%), followed by parents (31.7%), which again was similar to the pattern observed in other family studies on adult OCD patients.[11121334] The mean age at onset of OCD was found to be 27.7 years which was higher than some of the other studies in adult patients.[1826] This could be attributed to the higher female preponderance in the patient group. OCD has been shown to typically occur in women at a later age, compared to men, who are known to have an earlier onset.[3637] The mean duration of illness was 6.12 years and mean YBOCS score 27.54, signifying that the majority of the patients suffered from a severe illness. This could be because of the characteristics specific to the sample. Often, the patients referred to a tertiary clinic are the most severe ones, which is reflected in the high symptom scores as well as longer duration of illness. Similar clinical profile of patients has been described in previous studies from other tertiary clinic settings from India.[28]

Accommodating behaviors were found to be highly prevalent among family members of our patients, and the results were similar to the rates mentioned in previous studies.[162038] The frequency of directly participating in compulsions/rituals (every day) was 6.9%, which is much lower than the rates reported in previous studies from the Western population.[18] However, direct participation in compulsions for 4–6 days in a week yielded a comparable figure (26.7%).

The most important finding of the current study suggested that higher levels of FA are associated with greater symptom severity as well as functional impairment in individuals with OCD. This finding is consistent with previous studies primarily conducted in the Western population,[89121316] as well as the Indian population.[28] This particularly highlights an important finding that FA is consistently demonstrated across different populations from diverse cultural settings. However, the scale to measure FA in this study (FAS-SR Hindi version) was adapted from the FAS-SR scale, which was originally developed in a Western setting. Therefore, it is not equipped to pick any culture-specific manifestations of accommodation behaviors which might be specific to our population. Qualitative studies in the future exploring the nuances of this construct might be able to shed better light on cross-cultural issues. This will also pave the way for developing family-based interventions targeting accommodation behaviors among family members of patients with OCD, in keeping with the sociocultural and value systems of our study population.

Second, although this study clearly demonstrates that FA is associated with more severe psychopathology and higher disability scores, the cause–effect relationship between accommodation behavior and illness-related parameters cannot be adequately gleaned from these findings due to the cross-sectional nature of the study. Whether higher rates of accommodation in family members was a consequence of higher levels of psychopathology in the patient or vice-versa can only be determined in the future through prospective studies. However, the current study clearly highlights the potential role of FA in adversely affecting the trajectories of the outcome of OCD in patients. Longitudinal studies in the future will be able to further delineate the role of FA in the clinical and functional outcome of OCD.

Strengths

The primary strengths of our study were the recruitment of a large sample of psychopathology-free relatives of patients who were well characterized with respect to the clinical presentation of the disorder. We used the self-report version of the FA Scale to explore the caregiver's first-hand account of involvement in accommodation-related behaviors. The FAS-SR scale was translated and standardized in accordance to the rigorous WHO's translation–backtranslation methodology. Data of all the 101 caregivers were also used for the assessment of the internal consistency of the scale. The Cronbach's alpha was 0.93, indicating a good interitem correlation.[39] The use of multiple linear regression analyses allowed us to identify predictors of FA.

Limitations

One of the limitations was the cross-sectional design and purposive sampling of the study. The association between symptom severity and accommodation, as observed in this study is, however, clinically pertinent and instrumental in guiding future research in this area. Patients were recruited via purposive sampling in a tertiary clinic setting. The sample size was determined as per convenience. Possibly because of this, our sample had a higher representation of female patients as compared to community samples. The inclusion of patients with more severe psychopathology could attribute to the high levels of FA obtained in the caregivers, which might not be truly representative of patients in the community. These factors limit the generalisability of the results and highlight the need for further studies in this area. Second, patients were not assessed for comorbid personality disorders, which might be an important factor mediating the level of FA. Furthermore, a family history of anxiety disorders was not evaluated. Previous research demonstrated that a positive family history of another anxiety disorder emerged as a predictor of greater accommodation and correlated with all the dimensions of accommodation.[18] Furthermore, personality traits as well as subsyndromal depressive and anxiety symptoms in caregivers were not assessed, which can influence the phenomena of FA. In this study, we assessed FA in only one key caregiver. A more comprehensive analysis in other family members may reflect on the patterns of accommodation in different family members of the patient.

CONCLUSIONS

The present study adds useful information to the existing Indian research, revealing that FA is a common phenomenon in caregivers of patients with OCD. FA is particularly frequent when the patient has prominent contamination/washing symptoms. Clinicians should be aware of this high probability, and future research should be geared toward the development of family-based psychotherapeutic interventions, which might be particularly beneficial for these families. Whether FA is a determinant or a consequence of OCD severity remains to be elucidated. There is a need for prospective studies to delineate the clinical salience of FA as a predictor of disease severity as well as treatment outcomes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.