| Article Access Statistics|
| Viewed||708 |
| Printed||10 |
| Emailed||0 |
| PDF Downloaded||51 |
| Comments ||[Add] |
Click on image for details.
|Year : 2018
: 60 | Issue : 3 | Page
|How valid is obsessive-compulsive inventory-revised scale among Sri Lankan adults?
Buddhika Senanayake1, Mahesh Rajasuriya2, Chathurie Suraweera2, Carukshi Arambepola3
1 Postgraduate Institute of Medicine, University of Colombo, Colombo, Sri Lanka
2 Department of Psychological Medicine, Faculty of Medicine, University of Colombo, Colombo, Sri Lanka
3 Department of Community Medicine, Faculty of Medicine, University of Colombo, Colombo, Sri Lanka
Click here for correspondence address and
|Date of Web Publication||16-Oct-2018|
| Abstract|| |
Background: Obsessive-compulsive disorder (OCD) is a neuropsychiatric disorder with a lifetime prevalence of 1.9%–3.0% in the general population. It is an easily missed diagnosis. Although several paper-based tools have been developed, none are culturally validated for Sri Lankans to screen for OCD at field level.
Aims: This study aimed to translate and assess the validity and reliability of obsessive-compulsive inventory-revised scale (OCI-R) for Sri Lankan adults.
Setting and Design: This study was a case–control study.
Materials and Methods: This was conducted among 89 OCD patients and 89 controls recruited from the National Hospital of Sri Lanka to assess the criterion validity of OCI-R (an 18-item tool on common OCD symptoms using six subscales), by applying it and the gold standard (clinical diagnosis made by two independent consultant psychiatrists) to the same patient simultaneously. Before this, the tool was translated into local language (Sinhala) by bilingual experts and two psychiatrists using the forward-backward translation method. Receiver operating characteristics (ROC) curve was drawn to determine the cutoff value to identify OCD in Sri Lanka.
Results: The translated tool demonstrated the following: sensitivity 84.4%, specificity 85.6%, positive predictive value 85.4%, negative predictive value 84.6%, and positive and negative likelihood ratios of 5.86 and 0.18, respectively. The cutoff value for diagnosing OCD was 21 according to the ROC curve. Internal consistency (Cronbach's alpha reliability coefficient) of all six domains and the total scale showed values exceeding Nunnally's criteria of 0.7.
Conclusions: Sinhala version of the OCI-R scale was identified as a valid and reliable screening instrument to be applied in Sri Lankan adults.
Keywords: Adults, obsessive-compulsive disorder, Sri Lanka, validity
|How to cite this article:|
Senanayake B, Rajasuriya M, Suraweera C, Arambepola C. How valid is obsessive-compulsive inventory-revised scale among Sri Lankan adults?. Indian J Psychiatry 2018;60:318-23
|How to cite this URL:|
Senanayake B, Rajasuriya M, Suraweera C, Arambepola C. How valid is obsessive-compulsive inventory-revised scale among Sri Lankan adults?. Indian J Psychiatry [serial online] 2018 [cited 2019 Nov 17];60:318-23. Available from: http://www.indianjpsychiatry.org/text.asp?2018/60/3/318/243372
| Introduction|| |
Obsessions are defined as repeated and persistent thoughts, or impulses that are experienced, at some time during the disturbance, as intrusive, unwanted, and that in most individuals cause marked anxiety or distress. Compulsions are repetitive actions that the individual feels driven to perform in response to an obsession or according to rules that must be applied rigidly. Although it has been underrecognized and undertreated, it has emerged in recent times as a neuropsychiatric disorder across the globe as one with disabling outcomes. The lifetime prevalence of obsessive-compulsive disorder (OCD) is around 1.9%–3.0% in the general population, and is closely linked with comorbidities such as depression, anxiety, and tendency to suicide. OCD patients show poor psychological, social, and physical quality of life, and also affect that of their loved ones., According to a recent hospital-based study, 76.5% of OCD patients showed an increasing tendency for renal damage and hyperlipidemia due to overt fluid restriction patterns and irrational eating, and severe self-neglect with difficulty in performing basic self-care. The affected individuals taking longer time to complete a task also seemed to influence their interpersonal relationships. In the United States of America, OCD affects one in 40 adults leading to an estimated economic impact from direct medical costs of $2.1 billion and loss of productivity of $5.9 billion.
Several instruments, such as the semi-structured Yale–Brown obsessive-compulsive scale (YB-OCS), obsessive-compulsive inventory (OCI), interpretation of intrusions inventory, and OCI-Revised (OCI-R), are commonly used to assess OCD. Of them, OCI-R is derived from OCI and has several improvements over others such as reduced overlapping across subscales, simplified scoring of the subscales, and elimination of the redundant frequency scale. OCI-R developed by Foa et al. is an 18-item tool that measures the full range of OCD symptoms in different settings within six subscales (washing, checking, ordering, obsessing, hoarding, and neutralizing) based on a five-point Likert scale ranging from 0 (“not at all”) to 4 (“extremely”). OCI-R has been validated in different regions, including Western and Asian countries, with demonstrably good validity and reliability.,, However, its authors have emphasized on modifying this tool when applied in other countries to further enhance its validity. Compulsions and obsessions mainly assessed in OCD need to be studied in relation to a particular country with specific knowledge on its traditions, culture, beliefs, social norms, and values, thus requiring cultural adaptation of the translated tools.
Sri Lanka differs from the other South Asian countries by its unique free healthcare system and the sociocultural background. It has a well-established primary healthcare system, where family health workers deliver health services at domiciliary level. Nevertheless, owing to social stigma on psychological illnesses leading to poor health seeking in the curative sector, there is a hidden burden of disease related to OCD at the community level. In this backdrop, a brief paper-based simple tool is helpful for effectively utilizing the grass root level healthcare workers to screen the families under their care for OCD and thereby do timely referral of the probable cases for the treatment to hospitals. Results will be useful for healthcare providers of the country to implement low-cost interventions for disease detection, and also for other countries with no community psychiatrists to make a field diagnosis of OCD. The objective of this study was to translate and assess the validity and reliability of OCI-R to screen adults for OCD in the field setting in Sri Lanka.
| Materials and Methods|| |
A validation study was conducted to assess the judgmental and criterion validity, and reliability of OCI-R-Sinhala version (OCI-R-S) when applied among Sri Lankan adults.
Translation and cultural adaptation of Obsessive-Compulsive Inventory-Revised
OCI-R was translated into the Sinhala (the main official language in Sri Lanka) using the forward-backward translation method. The forward translation was done by a bilingual expert fluent in both Sinhala and English together with a psychiatrist and by another bilingual expert working alone, with special attention paid to the simplicity of wording, preservation of original meanings, and cultural suitability. By comparing with each other and through further discussions with a group of third bilingual expert, community physician, and a psychiatrist, a provisional translation was yielded. The back translation into English was done by two new bilingual experts independently with no reference to the English original. The experts and principal investigator (PI) then compared it with the original to ascertain whether original meanings of the items have been retained or not.
Judgmental validity of Obsessive-Compulsive Inventory-Revised-Sinhala
The judgmental validity of the translated OCI-R (OCI-R-S) was assessed in the form of face, consensual, and content validity to assess whether the conceptual definition of OCD was appropriately converted into operational terms or not. Face validity was assessed by appraising the relevance of the tool to the domains under investigation by experts in the fields of psychiatry, psychology, and community health. Content validity was assessed by a multidisciplinary panel of experts in community medicine, psychiatry, psychology, sociology, and languages, and a lay person representing the community for the appropriateness of wording used, acceptability in the local context, and relevance of each item in assessing OCD. Consensual validity was assessed for the agreement of a number of experts that a measure is valid by carrying out cultural adaptation and relevant modifications of the translated tool using the modified Delphi process. The same panel of experts was asked to rate their agreement on each item on a scale of 1 (total disagreement)–5 (total agreement). If 70% or more of the ratings for an item were in the range of 4–5, those items were retained in the final version. If not, those items were rerated and modified in the consensus of all experts.
The OCI–R-S was pretested among 10 patients attending the outpatient department and 10 diagnosed OCD patients attending the psychiatry clinic at the Colombo North Teaching Hospital, Ragama, for assessing its timing and comprehensibility. In the same sample, the reliability of OCI-R-S was assessed between the PI and data collectors by simultaneous administration of the tool.
Criterion validity of Obsessive-Compulsive Inventory-Revised-Sinhala
A case–control study was conducted to assess the criterion validity of OCI-R-S against its gold standard (i.e., clinical diagnosis). Following informed written consent, the participants were recruited for the study from the National Hospital of Sri Lanka, Colombo, which is the premier tertiary referral hospital for adults in Sri Lanka. The study participants were within the age range of 18–75 years. Those with conditions, such as schizophrenia, schizotypal and delusional disorders, critical illness, and those not conversant in the official local language (Sinhala), were excluded from the study. Cases were patients attending the psychiatry clinic with a clinical diagnosis of OCD newly made according to the diagnostic and statistical manual of mental disorders-4th edition (DSM-IV) criteria. Controls were patients attending the outpatient's department for minor general medical symptoms and excluded from OCD using the same criteria. They were neither psychiatrically ill nor hospital-admitted patients. Diagnoses in both groups were made independently by two consultant psychiatrists. Thereafter, the OCI-R-S was administered simultaneously to those who were consecutively sampled from each relevant clinic by medical graduates, who were not involved in the care, and blinded to the disease status of participants. Each participant was given an overall score by converting their responses in the OCI-R-S into scores according to the OCI-R manual.
The minimum sample required for the study was calculated using the equation for testing criterion validity of a tool. Based on 95% confidence level (Z) of 1.96 and confidence interval width of 0.2, the number of cases required was 89 OCD patients for an expected sensitivity of 0.64 (range: 0.64–0.74),, and was 89 non-OCD patients for an expected specificity of 0.64.,,
A cutoff value for categorizing a person as having OCD based on OCI-R-S was derived using the receiver operating characteristics (ROC) curve analysis. In this analysis, the clinical diagnosis (presence or absence of OCD) was taken as the gold standard and overall scores of OCI-R-S as the test variable. The shortest length from the area under the curve (AUC) to the diagonal line in the ROC curve indicated the optimal OCI-R-S score cutoff value, which was determined by the following formula; (1-SN)2+ (1-Sp)2 in which, SN= sensitivity; Sp= specificity. As for its validity, an AUC of 0.5 was assumed to indicate that the scoring system does no better than chance in discriminating between diseased and nondiseased whereas a scoring system with perfect discrimination would have an AUC of 1.0. Finally, the criterion validity was assessed by comparing the results obtained in the presence or absence of OCD by the OCI-R-S tool with that of the gold standard, using validity indicators (sensitivity, specificity, predictive values, and likelihood ratio). The reliability of OCI-R-S was assessed by internal consistency – the extent to which items within a dimension are correlated with each other to give the same result, for which, Cronbach's alpha coefficients were calculated for individual domains and the total instrument. The magnitude of 0.7 or above was considered satisfactory internal consistency. The inter-rater reliability of the OCI-R-S (between the PI and data collectors) was also assessed by calculating the Kappa coefficient.
Ethics approval was obtained from the Ethics Review Committee, Faculty of Medicine, University of Colombo (EC–13-180). An information sheet containing a brief but comprehensive description about the purpose of the study, its importance and benefits were provided to each participant and were given adequate time to read and clarify any doubts regarding the study. If the participant agreed to answer the questions, the informed written consent was taken before the interview. Participation in the research was entirely voluntary while the participants had full freedom to quit from the study at any point and refuse to answer any question that she/he did not like to answer within the interview.
| Results|| |
The response rate of cases (n = 90) and controls (n = 90) was 100%. The clinical diagnosis of OCD made by the two psychiatrists did not show any discrepancy. As shown in [Table 1], compared to the controls, cases consisted of a significantly higher proportion of patients aged <40 years (63.3%), of Sinhalese ethnicity (90%), with secondary education or higher (61.1%), currently unemployed (72.2%), and having <four family members (63.3%).
|Table 1: Characteristics of the cases (obsessive-compulsive disorder patients) and controls (nonobsessive compulsive disorder patients)|
Click here to view
OCI-R-S was comprehensively evaluated for its conceptual equivalence to the original scale at the translation and cultural adaptation phases. During the Delphi process, 16 questions were selected directly to the final tool. Of others, the English phrase “that get in the way” (question number one in the subscale of hoarding) was completely reworded. Question number 10 of the neutralizing subscale was also reworded according to the experience of experts. Pretesting confirmed that all the questions and response categories were understandable requiring only minor modifications. The average time taken per questionnaire was 8–10 min. Results demonstrate that the judgmental validity established of OCI-R-S in the Sri Lankan setting.
Confirmatory factor analysis (CFA) was conducted to assess the factor structure of OCI-R-S using LISREL 8.8 software (Scientific Software International, Skokie, Illinois, USA). The originally proposed OCI-R factor structure exhibited good to excellent fit in the current sample, supporting the originally postulated six-factor structure (χ2 = 179.2, P < 0.001; goodness of fit index = 0.952, comparative fit index = 0.91; and root mean square error of approximation = 0.059). The overall OCI-R-S scores of cases ranged from 9 to 53, with a mean of 28.52 (standard deviation [SD] = 8.63) and median of 27.5 (interquartile range: 23–33.25), while it ranged from 0 to 33, with a mean of 12.44 (SD = 7.37) for controls. [Figure 1] shows the ROC curve plotted on sensitivity against (1-specificity) for different cutoff values of overall OCI-R-S scores. AUC of this curve was 0.915 (95% confidence interval: 0.88–0.96) and significant (P < 0.01), indicating good performance of the model. Cutoff values of total OCI-R-S ranged from 0 to 53. The shortest distance (d2) in the ROC curve was 0.0451, which corresponded with an overall score of 20.5, sensitivity of 0.844, and specificity of 0.856. The validity indicators (sensitivity, specificity, and predictive values) of this cutoff level, as well as inter-rater reliability between the PI and data collectors (range: 0.85–0.89), are shown in [Table 2].
|Figure 1: Receiver operating characteristics curve for overall scores of Obsessive-Compulsive Inventory-Revised-Sinhala in the study population|
Click here to view
|Table 2: Validity and reliability of the Obsessive-Compulsive Inventory-Revised-Sinhala as indicators to detect obsessive-compulsive disorder|
Click here to view
[Table 3] summarizes the reliability findings on internal consistency. All six domains and the total scale showed high internal consistency with Cronbach's alpha values exceeding the Nunnally's criteria of 0.7.
|Table 3: Internal consistency of the total scale and domains of Obsessive-Compulsive Inventory-Revised-Sinhala|
Click here to view
| Discussion|| |
OCI-R-S demonstrated that satisfactory criterion validity and reliability when applied among Sri Lankan adults. The main aim of having a scientific approach in the translation process of a tool is to retain the original meaning within the context of cultural appropriateness. Accordingly, the modifications following translation were based on achieving the conceptual equivalence of the original item rather than on achieving word-to-word equivalence. The psychometric properties of this instrument were evaluated for the relevance in determining OCD, appropriateness of wording used, and acceptability in the local context through the judgmental validity of the translated OCI-R. The multidisciplinary panel selected for this purpose was appropriate. The use of the modified Delphi technique was also appropriate to achieve the consensus of the experts. CFA showed that the current data fit well to the six-factor model originally suggested by Foa et al.
Criterion validity of OCI-R-S was assessed against the clinical diagnosis made independently by two psychiatrists based on DSM-IV criteria. It was in concurrence with the accepted practice of appraising validity against a criterion that is known or believed to be close to the truth. Our controls were not matched for age and sex with cases. However, the cases and controls attending the same hospital and being similar in their health seeking behaviour minimised selection bias, and thereby ensured comparability. Furthermore, OCI-R overall scores on a continuous scale provided us the opportunity to derive the best cutoff value appropriate for our cultural setting. Accordingly, our findings demonstrate OCI-R-S as a valid instrument to screen for OCD in Sri Lanka, which enables 84.4% of patients with OCD to be properly classified as having the disease, (sensitivity) and 86.6% individuals without OCD to be classified as not having the disease (specificity) at an optimal cutoff value of 21. In comparison, the psychometric properties of the original OCI-R developed by Foa et al. were 65.6% sensitivity, and 63.9% specificity at the same optimal cutoff value, and of the Korean version by Woo et al. were 74% sensitivity and 69% specificity at 22 optimal cutoff value. The findings suggest that the wide application of this tool across both developed and developing countries. On the contrary, an optimal cutoff value of 36 was seen among African Americans for OCD with a sensitivity of 64.7% and specificity of 88.6%. The cutoff value derived for the 42-item OCI was 42.
Achieving the best trade-off between false negatives and false positives (i.e., sensitivity and specificity) is of utmost importance in a screening tool. OCD is a disease which could create issues for the screened people in two ways. Failure to identify OCD (false negatives) may affect patient's physical, psychosocial, and occupational activities of life, eventually leading to poor quality of life. On the other hand, erroneously classifying healthy persons as having OCD (false positives) and referring them for health services would lead to a waste of resources and create mental trauma for them and their family members as well. This trade-off is crucial in the case of OCI-R-S, where the OCD status is determined by applying a cutoff on the overall scores. The ideal setting to derive this cutoff value would have been in a community study, where a continuous distribution of the score is expected. However, owing to feasibility issues, the cutoff value was derived from the same clinic-based case–control study.
Since the OCI-R-S is intended to be used to screen individuals in the community, the best setting for validating this instrument would have been in the same. However, it was not possible in our study due to logistics: The need to screen a large number to identify the necessary number of cases (community prevalence of OCD is very low), time-consuming, cost, and undue time and resources of psychiatrist. By recruiting hospital cases, validity could be overestimated, owing to the psychiatric clinic constituting more severe OCD cases than in the community. Compared to European Americans, the main reason for African Americans producing relatively higher optimal cutoff values during OCI-R validation was confined to the more severe symptoms of OCD patients.
Validity would also be overestimated by recruiting hospital controls constituting more under stress owing to their illness than in the community. According to Foa et al., 65.6% sensitivity and 63.9% specificity shown at an optimal cutoff value of 21 with “nonanxious controls” (similar to community controls) shifted to an optimal cutoff value of 18 with sensitivity increasing to 74% and specificity to 75.2% with “individuals with anxiety” (anxious controls). Another similar type of study using patients having other anxiety disorders as controls also demonstrated a much lower optimal cutoff score of 14 with a sensitivity of 64% and specificity of 82.5%. This shows the potential value of OCI-R-S in detecting OCD even among anxiety disorder patients.
The positive predictive value (PPV) is the most useful measure in appraising a screening tool since it gives an indication of what cost and effort the screening program would require. The OCI-R-S demonstrated a PPV of 85.4%, which confirmed the acceptability of this tool for detection of OCD. It was compatible with a study by Ferdinando et al. that validated a tool among adults in the district of Kalutara, Sri Lanka, to detect short-term depression with a PPV of 90.1%. Satisfactory internal consistency and inter-rater reliability also demonstrate the usefulness of this tool in field settings. Its low cost, less time consumption, the ability to be administered by persons with less professional training such as public health midwife, and noninvasiveness are also among the factors that indicate the feasibility and usefulness of this instrument. Any structured or semi-structured instruments were not used by the clinicians to make the diagnosis of OCD. This was identified as a limitation. Failure to conduct convergent and discriminant validity of the OCI-R-S using other measures such as YBCS, and OCI was another limitation of this study.
| Conclusions|| |
The OCI-R-S was found to be a valid and reliable instrument when tested against consultant psychiatrist's diagnosis (gold standard) to detect OCD among Sri Lankan adults. With an optimal cutoff value of 21 derived from ROC analysis, the tool demonstrated: sensitivity 84.4%, specificity 85.6%, PPV 85.4%, negative predictive value 84.6%, and positive and negative likelihood ratios of 5.86 and 0.18, respectively. Satisfactory reliability was shown with relatively high internal consistency in all six domains and the total scale.
The authors would like to thank all of the patients who participated in the study. The authors would also like to acknowledge the director of the National Hospital of Sri Lanka.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
American Psychiartric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th
ed. Washigton, DC: American Psychiatric Association; 2013.
Murray CJ, Lopez AD. Global Burden of Disease: A Comprehensive Assessment of Mortality and Morbidity from Diseases, Injuries and Risk Factors in 1990 and Projected to 2020. Harvard: World Health Organization; 1996.
Karno M, Golding JM, Sorenson SB, Burnam MA. The epidemiology of obsessive-compulsive disorder in five US communities. Arch Gen Psychiatry 1988;45:1094-9.
The New Jersey Medical School. Center for Obsessive Compulsive Disorder (OCD) and Related Disorders. Psychiatry. New Jersey: New Jersey Medical School; 2013.
Albert U, Salvi V, Saracco P, Bogetto F, Maina G. Health-related quality of life among first-degree relatives of patients with obsessive-compulsive disorder in Italy. Psychiatr Serv 2007;58:970-6.
Cicek E, Cicek IE, Kayhan F, Uguz F, Kaya N. Quality of life, family burden and associated factors in relatives with obsessive-compulsive disorder. Gen Hosp Psychiatry 2013;35:253-8.
Hertenstein E, Thiel N, Herbst N, Freyer T, Nissen C, Külz AK, et al.
Quality of life changes following inpatient and outpatient treatment in obsessive-compulsive disorder: A study with 12 months follow-up. Ann Gen Psychiatry 2013;12:4.
World Health Organization. Global burden of obsessive-compulsive disorder in the year 2000. In: Ayuso-Mateos JL, editor. Global Program on Evidence for Health Policy (GPE). Geneva: World Health Organization; 2006.
Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL, et al.
The yale-brown obsessive compulsive scale. I. Development, use, and reliability. Arch Gen Psychiatry 1989;46:1006-11.
Foa EB, Kozak MJ, Salkovskis PM, Coles ME, Amir N. The Validation of a New Obsessive-Compulsive Disorder Scale: The Obsessive-Compulsive Inventory. Washington, DC: American Psychological Association; 1998.
Obsessive Compulsive Cognitions Working Group. Psychometric validation of the obsessive beliefs questionnaire and the interpretation of intrusions inventory: Part I. Behav Res Ther 2003;41:863-78.
Foa EB, Huppert JD, Leiberg S, Langner R, Kichic R, Hajcak G, et al.
The obsessive-compulsive inventory: Development and validation of a short version. Psychol Assess 2002;14:485-96.
Abramowitz JS, Deacon BJ. Psychometric properties and construct validity of the obsessive-compulsive inventory – Revised: Replication and extension with a clinical sample. J Anxiety Disord 2006;20:1016-35.
Peng ZW, Yang WH, Miao GD, Jing J, Chan RC. The Chinese version of the obsessive-compulsive inventory-revised scale: Replication and extension to non-clinical and clinical individuals with OCD symptoms. BMC Psychiatry 2011;11:129.
Zermatten A, Van der Linden M, Jermann F, Ceschi G. Validation of a French version of the obsessive-compulsive inventory-revise d in a non-clinical sample. Rev Eur Psychol Appl 2006;56:151-5.
Huppert JD, Walther MR, Hajcak G, Yadin E, Foa EB, Simpson HB, et al.
The OCI-R: Validation of the subscales in a clinical sample. J Anxiety Disord 2007;21:394-406.
Abramson JH, Abramson ZH. Survey Methods in Community Medicine. 5th
ed. London: Churchil Livingston; 1999.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th
ed. Washington DC: American Psychiatric Association; 1994.
Hulley SB, Cummings SR, Browner WS, Grady D, Hearst N, Newman TB. Designing Clinical Research: An Epidemiologic Approach. 2nd
ed. Philadelphia: Lippincot Williams & Wilkins; 2001.
Williams M, Davis DM, Thibodeau MA, Bach N. Psychometric properties of the obsessive-compulsive inventory revised in African Americans with and without obsessive-compulsive disorder. J Obsessive Compuls Relat Disord 2013;2:399-405.
Woo CW, Kwon SM, Lim YJ, Shin MS. The obsessive-compulsive inventory-revised (OCI-R): Psychometric properties of the korean version and the order, gender, and cultural effects. J Behav Ther Exp Psychiatry 2010;41:220-7.
Pudrovska T, Anikputa B. The role of early-life socioeconomic status in breast cancer incidence and mortality: Unraveling life course mechanisms. J Aging Health 2012;24:323-44.
Kumar R, Indrayan A. Receiver operating characteristic (ROC) curve for medical researchers. Indian Pediatr 2011;48:277-87.
Nunnally JC, Bernstein IH. Psychometric Theory. USA: McGraw-Hill Inc.; 1994.
Sumathipala A, Murray J. New approach to translating instruments for cross-cultural research: A combined qualitative and quantitative approach for translation and consensus generation. Int J Methods Psychiatr Res 2000;9:87-95.
Ferdinando KDAR. Prevalence and Correlates of Depression in Adults of 25-45 Years in the Kalutara DDHS area. Colombo: Post Graduate Institute of Medicine, University of Colombo (D 1618); 2006.
Dr. Buddhika Senanayake
No: 65/A, Ananda Mawatha, Colombo 10
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3]