|Ahead of print
Validation of the hospital anxiety depression scale - Marathi version in detecting anxiety and depression in cancer patients and caregivers
Jayita K Deodhar1, Savita S Goswami2, Lekhika N Sonkusare2
1 Department of Palliative Medicine; Psycho-oncology Unit, Tata Memorial Hospital; Homi Bhabha National Institute, Mumbai, Maharashtra, India
2 Psycho-oncology Unit, Tata Memorial Hospital, Mumbai, Maharashtra, India
|Date of Submission||08-Dec-2020|
|Date of Decision||12-May-2021|
|Date of Acceptance||14-May-2021|
|Date of Web Publication||29-Jun-2022|
Jayita K Deodhar,
Department of Palliative Medicine; Psycho-oncology Unit, Tata Memorial Hospital; Homi Bhabha National Institute, Mumbai, Maharashtra
Source of Support: None, Conflict of Interest: None
Background: The Hospital and Anxiety Depression Scale (HADS) is useful for screening depression and anxiety in cancer. It has not been validated in the Marathi language, which is the third most common language in India. We aimed to examine the reliability and validity of the Marathi-translated version of HADS in cancer patients and their caregivers.
Methods: In a cross-sectional study design, we administered the Hospital and Anxiety Depression Scale-Marathi version (HADS-Marathi) to 100 participants (50 patients and 50 caregivers) after obtaining their informed consent. The team Psychiatrist, who was blind to the HADS-Marathi scores, interviewed all participants and identified the presence of anxiety and depressive disorders using the diagnostic criteria of the International Classification of Diseases – 10th edition. We measured internal consistency using Cronbach's alpha, receiver operating characteristics, and factor structure. The study was registered with the Clinical Trials Registry-India (CTRI)
Results: The internal consistency of HADS-Marathi was good with 0.815, 0.797, and 0.887 for anxiety and depression subscales and total scale, respectively. The area under curve figures were 0.836 (95% Confidence Interval [CI]: 0.756 - 0.915), 0.835 (95% [CI]: 0.749–0.921), and 0.879 (95% [CI] 0.806–0.951) for anxiety and depression subscales, and total scale, respectively. The best cutoffs identified were 8 (anxiety), 7 (depression), and 15 (total). The scale displayed a three-factor structure, with two depression subscale and one anxiety subscales items loading on to the third factor.
Conclusion: We found that the HADS-Marathi version is a reliable and valid instrument for use in cancer patients. However, we found a three-factor structure, possibly reflecting a cross-cultural effect.
Keywords: Anxiety, cancer, depression, screening, validation
The Marathi version of the Hospital Anxiety and Depression Scale is a reliable and valid tool in screening for depression and anxiety in oncology setting. This tool can be useful in detecting psychological distress in cancer patients and their caregivers, whose first language is Marathi.
|How to cite this URL:|
Deodhar JK, Goswami SS, Sonkusare LN. Validation of the hospital anxiety depression scale - Marathi version in detecting anxiety and depression in cancer patients and caregivers. Indian J Cancer [Epub ahead of print] [cited 2022 Aug 7]. Available from: https://www.indianjcancer.com/preprintarticle.asp?id=348449
| » Introduction|| |
Approximately, 20% of cancer patients suffer from depression, and around 1 in 10 have anxiety.,, Depression and anxiety can have a negative impact on cancer patients' quality of life, adherence to treatment, and survival. It is, therefore, essential to screen and assess for depression and anxiety.
Among the frequently used screening tools, the Hospital and Anxiety Depression Scale (HADS) is a reliable and valid screening measure for anxiety and depression in patients with physical illnesses. It contains 14 items with an Anxiety subscale (HADS-A) and a Depression subscale (HADS-D), both containing seven intermingled statements. Several reviews have reported that the HADS is a good instrument for assessing caseness and severity of symptoms of anxiety and depression in patients in primary care, with psychiatric illnesses as well as physical illnesses. A two-factor structure is evident from these reviews. The cut-off scores of 8 for each subscale of HADS gives sensitivity and specificity ranging between 0.7 and 0.9, which is very good.
Researchers and clinicians have extensively used HADS in cancer patients. Mitchell et al. in their meta-analysis reported pooled sensitivities of 82%, 71%, and 80% and specificities of 77%, 82.6%, and 77.8% for total, depression, and anxiety subscales, respectively, for depressive disorders. The sensitivities and specificities for anxiety were 83.9% and 48.7%, respectively, and 69.9% and 78.7% for the total score and anxiety subscale, respectively. The cutoffs for screening for psychiatric disorders were 10 or 11, 5, and 7 or 8 on the HADS total, the HADS depression, and anxiety subscales, respectively.
The HADS has been used in cancer patients in various translated versions as a screening tool to detect psychological distress. Authors validated a Slovenian version in 202 female cancer patients and stated correlation coefficients of 0.81 and 0.91 for the depression and anxiety subscales, respectively. In a validation study in Iranian language in 167 breast cancer patients, the HADS was found to be reliable and valid. A study on the HADS Mandarin version reported sensitivities and specificities of 84 and 73%; 72 and 86%; and 84 and 68% for the HADS anxiety, depression, and total scores, respectively. A Greek version validated in a palliative care unit showed good reliability and validity for the tool. Various studies with the translated versions have confirmed a two-factor structure.,,, Overall, validation studies of the translated version of HADS have been done in both mixed as well as specific cancer populations, like breast cancer, head and neck cancer, and lung cancer and in different stages and settings in many Non-English speaking countries.
The HADS has been translated and validated in very few Indian languages for use in cancer patients. The Malayalam version showed good reliability, with a Cronbach's alpha of 0.81, 0.71, and 0.85 for the anxiety, depression, and total scores and good acceptability.
Marathi is the vernacular language of patients visiting our hospital in the state of Maharashtra. It is the third most common language spoken in India (Hindi and Bengali being the first two) and the 10th official language spoken in the world. It is crucial to identify psychosocial distress of cancer patients by using appropriate screening tools in the language understood by the patients. There are no validation studies of the Marathi-translated version of HADS in oncology settings. It is, therefore, important to evaluate whether the Marathi version of the scale is a reliable and valid measure in detecting anxiety and depression in cancer patients and their caregivers. The purpose of this study was, therefore, to examine the validity and reliability of the Hospital and Anxiety Depression Scale-Marathi version (HADS-Marathi) in screening for anxiety and depression in cancer patients.
| » Methodology|| |
We conducted this cross-sectional study in the Psycho-oncology department of a tertiary cancer care center. We included a convenient sample of cancer patients and their caregivers of the age of 18 years and above referred to the service, able to read and write Marathi and English. The exclusion criteria were participants with any unstable medical or psychiatric disorder requiring emergency attention and with pre-existing psychiatric disorders and/or taking prescribed medications for psychiatric disorders.
We screened 140 participants by undertaking a screening history of both patients and caregivers. A hundred participants (50 patients and 50 caregivers) after giving informed consent were recruited and their demographic details recorded [Appendix – Recruitment flowchart]. The Psycho-oncology staff administered the Marathi translation of the HADS (HADS-Marathi), which the participants completed [Appendix 1]. Subsequently, the team Psychiatrist, who was blind to the score on the HADS-Marathi, conducted a psychiatric interview, with a formal mental status examination for all participants. We used the diagnostic criteria in Chapter V for Mental and Behavioral Disorders of the International Classification of Diseases-10 for diagnosis of depressive and anxiety disorders (ICD-10) as “gold standard.” Additionally, the participants completed a brief questionnaire on the understandability and time required to complete the HADS-Marathi and items found difficult. All participants were interviewed. We had obtained the Marathi translation of HADS from GL assessment and Mapi research trust with permission for use, according to their protocol. The study was done with the approval of Institutional Ethics Committee (Final Approval number 1097) and registered with the Clinical Trials Registry-India (CTRI/2018/03/012568).
Descriptive measures were used for participant characteristics. We tested the reliability of the HADS-Marathi using internal consistency calculated by Cronbach's alpha. We assessed construct validity by factor analysis using principal component analysis with oblimin rotation. Convergent validity was measured using the Chi-squared test between scores on the subscales (anxiety and depression) and total score and diagnosis of anxiety and depressive disorders using the ICD-10) by psychiatric interview (identified as “cases”). We calculated sensitivity and specificity. We analyzed the receiver operating characteristic curve and measured the area under curve (AUC) for the subscales and total HADS-Marathi.
| » Results|| |
One hundred participants (55 men and 45 women) with the median age of 55.5 years (range 18–65) were enrolled and completed the HADS-Marathi. Of 50 patients and 50 caregivers, almost 70% had completed schooling. Sixty participants had some form of employment at the time of assessment. The mean age of patients was 37.3 years (Standard deviation (SD) 14.3) and that of caregivers was 41.1 years (SD 11.1). The most common cancers in the patients were brain, breast, and head and neck.
The characteristics of study participants are described in [Table 1].
Internal consistency measured by Cronbach's alpha for the anxiety and depression subscales of HADS-Marathi was 0.815 and 0.797, respectively. The Cronbach's alpha for the total scale was 0.887.
The interitem correlations are displayed in [Table 2]. The inter-domain correlation value for anxiety and depression subscale was 0.764.
Convergent validity was calculated by correlation of the score on HADS-Marathi with the presence of depression and anxiety on formal mental state examination carried out by the psychiatrist in the research team, using diagnostic criteria in Chapter V for Mental and Behavioural Disorders of the International Classification of Diseases-10 for diagnosis of depressive and anxiety disorders (ICD-10), using different cutoffs.
Area under curve, sensitivity, and specificity
For anxiety subscale on HADS-Marathi, AUC was 0.836 (95% Confidence Interval [CI]: 0.756-0.915). Utililzing 8 as the optimal cutoff, the sensitivity was 72% and specificity was 76%, whereas a cut off of 7 yielded a sensitivity of 79% and specificity of 73% [Figure 1]a.
|Figure 1: Receiver operating characteristics. (a) Receiver operating characteristics of Hospital Anxiety Depression Scale-Marathi – anxiety subscale;(b) Receiver operating characteristics of Hospital Anxiety Depression Scale-Marathi – depression subscale;(c) Receiver operating characteristics of Hospital Anxiety Depression Scale-Marathi – total scale|
Click here to view
For depression subscale on HADS-Marathi, AUC was and 0.835 (95% [CI]: 0.749 - 0.921). Using 8 as the optimal cutoff, the sensitivity was 64% and specificity was 86%, whereas a cut off of 7 yielded a sensitivity of 74% and specificity of 86% [Figure 1]b.
The AUC for total score on HADS-Marathi was 0.879 (95%[CI]: 0.806–0.951). With 15 as the optimal cutoff, the sensitivity was 83% and specificity was 89%, whereas a cutoff of 14 yielded a sensitivity of 84.5% and specificity of 74% [Figure 1]c.
The mean scores with SDs for patients were 8.9 (5.04), 8.3 (4.4), and 17.1 (8.6) for HADS anxiety subscale, depression subscale, and total scale, respectively. The mean scores with SD for caregivers were 8.8 (4.8), 7.6 (5.3), and 16.5 (9.6) for HADS anxiety subscale, depression subscale, and total scale, respectively.
We found that in factor analysis of the correlation matrix, all items of the questionnaire correlated with statistical significance (<0.01), the determinants was >0.00001, and no two items had a correlation of 0.8 (more than 0.5). The Kaiser–Meyer–Olkin test of sampling adequacy was 0.885 (greater than 0.5) and Bartlett's test was significant at X2 = 508.007, P <0.0001. Factor loading values ranged from 0.380 to 0.435 and 0.308 to 0.483, for anxiety and depression subscales, respectively.
The scale displayed a three-factor structure [Table 3]. The values for Factor 1 ranged between -0.002 and 0.822, corresponding to anxiety subscale. However, two items (8 and 12) from the depression subscale loaded on to Factor 1. Item numbers 4, 14, and 6 contributed to Factor 2, suggestive of depression. Factor 3 consisted of two items from the depression subscale (10 and 2) and one item from the anxiety subscale (item 7). Items 1 and 5 had cross-loading across two factors.
All participants found the HADS-Marathi acceptable and took an average of 5.9 minutes to complete.
| » Discussion|| |
The HADS is one of the standard tools to measure depression and anxiety in cancer patients and has been validated in multiple languages. Among Indian languages, HADS have been validated in Malayalam in oncology patients. In this paper, we report the validation of the Marathi-translated version in patients and caregivers undergoing treatment in the hospital and referred to psycho-oncology service.
The HADS-Marathi had an excellent internal consistency of Cronbach's alpha above 0.7 in both subscales of anxiety and depression and the total score. This result is similar to other translations.,,
The AUCs were 0.836, 0.835, and 0.879 for anxiety, depression, and total scores, respectively, indicating a good performance for HADS-Marathi, as reported in other reviews. The optimal cut-off score was 8 for the identification of cases on both subscales., In primary care patients, this was reported to be the valid cut-off point. For cancer patients, the threshold might be more applicable to avoid underdetection. We found that the sensitivity was better with a cutoff of 7 for the anxiety subscale (79%), but the specificity was better with a cut-off of 8. For the depression subscale, a cutoff of 7 yielded a sensitivity of 74% and specificity of 84%, which was more reliable.
In our study, the total score cutoff of 15 seemed the best trade-off with the sensitivity and specificity of 74% and 86% respectively. This finding is comparable to other studies in the Asian populations., Among Indian languages, there was a better sensitivity and specificity with the cutoff of 16 on the total score. Singer et al. in their review on HADS validation using a rigorous method of identifying the cut-off score by obtaining a balanced score, clinical score, and a specific score recommended a cut-off of 15 for the full scale. Other authors have also recommended the use of total score.,
We found a three-factor structure in the HADS-Marathi. This conclusion is in contrast to the two-factor structure according to Zigmond and Snaith's original study, studies done in general population, and those in cancer patients and their caregivers.,,,, A few authors have reported a three-factor structure in their validation of HADS English., Most of the studies on translated versions of HADS in cancer patients have also reported a two-factor structure., However, authors validating the Swedish version reported that a three-factor, as well as a two-factor structure gave a better fit than a single dimension.
In our factor analysis, one of the depression items (“I feel as if I am slowed down”) loaded on to the anxiety subscale. An explanation could be that the participants felt “slowed down” in concentration, which could be due to anxiety. The depression subscale had items related to the ability to enjoy (“I can laugh and see the funny side of things,” “I can enjoy a good book or radio or TV program,” “I feel cheerful”). The third factor had two items related to depression (“I have lost interest in my appearance” and “I still enjoy the things I still used to”) and one of anxiety (“I can sit at ease and feel relaxed”). Also, in the Malayalam translation, two items related to depression loaded on the anxiety subscale.
All our participants found the HADS-Marathi version acceptable items of the Marathi version and took about 5 minutes to complete the questionnaire. This finding is in contrast to the Malayalam version where the participants had difficulty in understanding the item “I get a sort of frightened feeling like butterflies in the stomach.” One explanation for this could be that the participants felt some hesitation or concern that it could affect the therapeutic alliance, although we did emphasize on anonymity and confidentiality while taking informed consent. Maters et al. in their review of HADS translations has pointed out that most authors have not reported on the difficulties in cross-cultural use in translated versions in Non-English speaking populations.
Our study is the first to validate the Marathi translation of HADS in cancer patients and their caregivers. It is reliable with a high internal consistency around 0.8 for the subscales and the full scale, which is higher than that found in other Indian language versions. Our report adds to the existing literature of better utility of the total score, especially among Indian languages. Our conclusion of a three-factor structure raises the possibility that our Marathi speaking population find that some features of depression align more with anxiety, signifying that challenges in cross-cultural use of tools developed in English-speaking countries.
Our study had certain limitations. We used the clinical psychiatric interview for diagnosis of depression and anxiety as per diagnostic criteria of Chapter V of ICD-10 as the gold standard. A structured research diagnostic tool like the Structured Clinical Interview for Diagnosis or Mini-International Neuropsychiatric Interview would have added more rigor., However, we wanted to test the scale in a routine clinical setting which could translate better to regular practice. Also, the psychiatrist assessing the participants was blind to the scores on the HADS-Marathi. Other authors have used clinical diagnostic criteria in different versions of Diagnostic and Statistical Manual of Mental disorders in their validation procedures. Our sample size was modest, similar to another study. We included caregivers in the study population to be able to identify both cases and noncases. However, it is possible that caregivers did not have any physical problems which interfered with their scoring on the depression items on the HADS, unlike the patients. Future studies with a larger sample size will help consolidate the validation procedure.
| » Conclusion|| |
We found that the HADS-Marathi version is a reliable and valid instrument for use in cancer patients. The total score was more useful for the identification of depression and anxiety disorders. We report a three-factor structure, possibly resulting from cultural bias. Screening for clinical depression and anxiety in oncology setting using this tool will help in identification, appropriate referral, and prompt management of cancer patients, which in turn will result in a better quality of life. Further research comparing HADS in different Indian languages would be useful to identify strengths and challenges in cross-cultural use.
We acknowledge our patients, hospital administration, and staff for their support.
Declaration of patient consent
The authors certify that they have obtained all appropriate participant consent forms. The study participants have given their consent and understand that their names and initials will not be published, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Linden W, Vodermaier A, MacKenzie R, Greig D. Anxiety and depression after cancer diagnosis: Prevalence rates by cancer type, gender, and age. J Affect Disord 2012;141:343-51.
Niedzwiedz CL, Knifton L, Robb KA, Katikireddi SV, Smith DJ. Depression and anxiety among people living with and beyond cancer: A growing clinical and research priority. BMC Cancer 2019;19:943.
Brown LF, Kroenke K, Theobald DE, Wu J, Tu W. The association of depression and anxiety with health-related quality of life in cancer patients with depression and/or pain. Psychooncology 2010;19:734-41.
Vodermaier A, Linden W, Siu C. Screening for emotional distress in cancer patients: A systematic review of assessment Instruments. J Natl Cancer Inst 2009;101:1464-88.
Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361-70.
Morse R, Kendell K, Barton S. Screening for depression in people with cancer: The accuracy of the hospital anxiety and depression scale. Clin Eff Nurs 2005;9:188-96.
Mitchell AJ, Chan M, Bhatti H, Halton M, Grassi L, Johansen C, et al
. Prevalence of depression, anxiety, and adjustment disorder in oncological, haematological, and palliative-care settings: A meta-analysis of 94 interview-based studies. Lancet Oncol 2011;12:160-74.
Vodermaier A, Millman RD. Accuracy of the hospital anxiety and depression scale as a screening tool in cancer patients: A systematic review and meta-analysis. Support Care Cancer 2011;19:1899-908.
Miklavčič IV, Snoj Z, Mlakar J, Pregelj P. Validation of the Slovenian version of hospital anxiety and depression scale in female cancer patients. Psychiatr Danub 2008;20:148-52.
Montazeri A, Vahdaninia M, Ebrahimi M, Jarvandi S. The hospital anxiety and depression scale (HADS): Translation and validation study of the Iranian version. Health Qual Life Outcomes [Internet] 2003;1:14. Available from: http://hqlo.biomedcentral.com/articles/100.1186/1477-7525-1-14
. [Last accessed on 2021 May 10].
Wang GL, Hsu SH, Feng AC, Chiu CY, Shen JF, Lin YJ, et al
. The HADS and the DT for screening psychosocial distress of cancer patients in Taiwan. Psychooncology 2011;20:639-46.
Mystakidou K, Tsilika E, Parpa E, Katsouda E, Galanos A, Vlahos L. The hospital anxiety and depression scale in Greek cancer patients: Psychometric analyses and applicability. Support Care Cancer 2004;12:821-5.
Saboonchi F, Wennman-Larsen A, Alexanderson K, Petersson LM. Examination of the construct validity of the Swedish version of Hospital Anxiety and Depression Scale in breast cancer patients. Qual Life Res 2013;22:2849-56.
Villoria E, Lara L. Assessment of the hospital anxiety and depression scale for cancer patients. Rev Med Chil 2018;146:300-7.
Smith AB, Selby PJ, Velikova G, Stark D, Wright EP, Gould A, et al
. Factor analysis of the hospital anxiety and depression scale from a large cancer population. Psychol Psychother 2002;75:165-76.
Djukanovic I, Carlsson J, Årestedt K. Is the hospital anxiety and depression scale (HADS) a valid measure in a general population 65-80 years old? A psychometric evaluation study. Health Qual Life Outcomes 2017;15:193.
Thomas BC, Devi N, Sarita GP, Rita K, Ramdas K, Hussain BM, et al
. Reliability and validity of the Malayalam hospital anxiety and depression scale (HADS) in cancer patients. Indian J Med Res 2005;122:395-9.
CENSUS OF INDIA 2011.
World Health Organization. The ICD-10 classification of mental and behavioural disorders. Int Classif 1992;10:1-267.
Li Q, Lin Y, Hu C, Xu Y, Zhou H, Yang L, et al
. The Chinese version of hospital anxiety and depression scale: Psychometric properties in Chinese cancer patients and their family caregivers. Eur J Oncol Nurs 2016;25:16-23.
Singer S, Kuhnt S, Götze H, Hauss J, Hinz A, Liebmann A, et al
. Hospital anxiety and depression scale cutoff scores for cancer patients in acute care. Br J Cancer 2009;100:908-12.
Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the hospital anxiety and depression scale. An updated literature review. J Psychosom Res 2002;52:69-77.
Akizuki N, Akechi T, Nakanishi T, Yoshikawa E, Okamura M, Nakano T, et al
. Development of a brief screening interview for adjustment disorders and major depression in patients with cancer. Cancer 2003;97:2605-13.
Razavi D, Delvaux N, Farvacques C, Robaye E. Screening for adjustment disorders and major depressive disorders in cancer in-patients. Br J Psychiatry 1990;156:79-83.
Cosco TD, Doyle F, Ward M, McGee H. Latent structure of the hospital anxiety and depression scale: A 10-year systematic review. J Psychosom Res 2012;72:180-4.
Iani L, Lauriola M, Costantini M. A confirmatory bifactor analysis of the hospital anxiety and depression scale in an Italian community sample. Health Qual Life Outcomes 2014;12:84.
Michopoulos I, Douzenis A, Kalkavoura C, Christodoulou C, Michalopoulou P, Kalemi G, et al
. Hospital anxiety and depression scale (HADS): Validation in a Greek general hospital sample. Ann Gen Psychiatry 2008;7:4.
Annunziata MA, Muzzatti B, Altoé G. Defining hospital anxiety and depression scale (HADS) structure by confirmatory factor analysis: A contribution to validation for oncological settings. Ann Oncol 2011;22:2330-3.
Rodgers J, Martin CR, Morse RC, Kendell K, Verrill M. An investigation into the psychometric properties of the hospital anxiety and depression scale in patients with breast cancer. Health Qual Life Outcomes 2005;3:41.
Brandberg Y, Bolund C, Sigurdardottir V, Sjödén P-O, Sullivan M. Anxiety and depressive symptoms at different stages of malignant melanoma. Psychooncology 1992;1:71-8.
Maters GA, Sanderman R, Kim AY, Coyne JC. Problems in cross-cultural use of the hospital anxiety and depression scale: “No butterflies in the desert.” PLoS One 2013;8:e70975.
Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al
. The mini-international neuropsychiatric interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatr [Internet] 1998;59:22-33. Available from: https://pubmed.ncbi.nlm.nih.gov/9881538/
[Last accessed on 2020 Sep 16].
[Table 1], [Table 2], [Table 3]