Explanatory Models and Medication Adherence in Patients with Depression in South India
Dushad Ram1, Adarsh Lakkur Siddappa2, Rajesh Raman3, Basavana Gowdappa Hattur4
1 Associate Professor, Department of Psychiatry, JSS Medical College and Hospital, Mysore, Karnataka, India.
2 Assistant Professor, Department of Medicine, JSS Medical College and Hospital, Mysore, Karnataka, India.
3 Professor, Department of Psychiatry, JSS Medical College and Hospital, Mysore, Karnataka, India.
4 Professor, Department of Medicine, JSS Medical College and Hospital, Mysore, Karnataka, India.
NAME, ADDRESS, E-MAIL ID OF THE CORRESPONDING AUTHOR: Dr. Dushad Ram, Associate Professor, Department of Psychiatry, JSS Hospital, MG Road Mysore-570004, Karnataka, India.
E-mail: dushadram@hotmail.com
Introduction
Conceptualization of depression may have bearing on treatment seeking. It may affect adherence behaviour of the patients.
Aim
To find out the explanatory models and their relationship with socio-demographic variables and medication adherence in patients with depression.
Materials and Methods
Fifty-eight condivutive patients with depression in remission were recruited as per selection criteria. Socio-demographic details were collected. Patients were assessed using Mental Distress Explanatory Model Questionnaire (MDEMQ) and Morisky Medication Adherence Scale (MMAS).
Results
Significant scores were observed in all dimensions of explanatory models. In the Mann-Whitney U test the patient’s marital status (MU=113.500, p=0.05, sig≤0.05, 2-tailed), and family history of mental illness (MU=165.5, p=0.03, sig≤0.05, 2-tailed) had a statistically significant group difference in the score of MDEMQ. In linear regression analysis, four predictors (MDEMQ subscales Stress, Western physiology, Non-Western physiology and Supernatural) had significantly predicted the value of MMAS (R2=0.937, f=153.558, p<0.001).
Conclusion
Findings of this study suggested that patients with depression harbor multidimensional explanatory model. The levels of explanatory models are inversely associated with levels of medication adherence.
Introduction
Major Depressive Disorder (MDD) is one of the most common psychiatric disease; with a lifetime prevalence estimates upto 17% [1]. Conceptualization (explanatory model) of depression varies with culture, ethnics and comorbid psychiatric disorder [2–4]. In India, common model of explanation about depression are supernatural and psychosocial [5,6]. Explanatory Model (EM) is the notion about an episode of sickness and its treatment that is employed by all those engaged in the clinical process [7]. EM influences patient’s help seeking behaviour [8]. Those with a supernatural model of explanation seek help from a traditional healer [9–11], while those with a social model seek more from social networks [12].
Only 60% of patients with depression seek treatment from physician, and non-adherence ranges from 20-50% [13–15]. Supernatural explanatory model of mental illness may hamper the medication adherence [10]. Few studies in India have addressed the EM in depression. Grover at al., reported Karma-deed-heredity category as commonest explanatory model (77.4%), followed by psychological model (62.2%) and social model (40.2%) and opined that this might have treatment implications [5]. Other studies from different parts of the world have examined the EM and treatment seeking behaviour in patients with depression and found that non-medical models were associated with less preference for medical treatment [16,17].
There is a paucity of study that have examined the relationship of EM and medication adherence among patients with depression. Buus N et al. had done a qualitative study in 16 patients with depression and found that patients with predominant psychosocial explanatory model were ambiguous about treatment with antidepressants [18]. Thus, this study was conducted to explore the explanatory model and its relationship with medication adherence in patients with depression. We hypothesized that the levels of explanatory models are negatively associated with the levels of medication adherence.
Materials and Methods
This was a cross-sectional hospital based study. A sample size of 60 was recommended by institutional reviewers, which was based on number of patients visiting for depression in psychiatry outpatient department every day, Patient’s selection criteria, and duration of this study.
A total of 58 consecutive patients with depression, who were in remission and living in the community, were recruited when they attended the outpatient department of psychiatry at a tertiary care hospital (in South India) for follow up over a period of three months (February 2014-April 2014). The inclusion criteria were both genders with an International Classification of Disease, Tenth Revision (ICD-10) diagnosis of depressive episodes, currently in remission as per treating psychiatrist and aged 16-65 years. Patients with any chronic physical illness were excluded as it may interfere with the conceptualization of mental illness due to its physical nature and induced discomfort or distress. An ICD-10 diagnosis of mental retardation or dementia was also excluded due to reliability issues. Patients who satisfied the selection criteria were first assessed with a socio-demographic proforma designed for this study. Then conceptualization of mental illness was assessed using MDEMQ that has 45 items; 5 points (1-5) item rating with a possible minimum score of 45 and the maximum score of 225 [19]. The items can be clustered into four explanatory categories (with a possible score range); Western Physiology (9-45), Non-Western Physiology (4-20), Supernatural (19-95) and Stress (13-65). At last medication adherence was assessed with eight items MMAS [20]. Each item can be rated in two points (0-1). Adherence is low when the score is >2, medium when the score is 1-2, and high when score is 0. The data were analyzed using SPSS version 16.0. Analysis of demographic variables and explanatory models was done with descriptive statistics. Since normal distribution could not be established in the analysis of data distribution; group difference of demographic variables on the score of MDEMQ were obtained with Mann-Whitney U (two groups) and Kruskal Wallis test (three or more groups). Linear regression analysis was conducted to know if score on MDEMQ could significantly predict patients score on MMAS. The level of statistical significance was kept at p<0.05 for all tests.
Results
In this study [Table/Fig-1] majority of the patients were educated, married, unemployed and belonged to the Hindu nuclear family. [Table/Fig-2] reveals mean and standard deviation of scores on MDEMQ (Mean 97.05, SD±22.76) and its subscales: Stress (Mean 39.03±9.42), Western physiology (Mean 19.72±5.53), Supernatural (Mean 32.67±13.13) and Non-Western physiology (Mean 5.62±2.15).
Socio-demographic characteristics.
| Variable | n | % |
|---|
| Gender | Male | 24 | 41.4 |
| Female | 34 | 58.6 |
| Occupation | Unemployed | 39 | 67.2 |
| Employed | 19 | 32.8 |
| Education | Uneducated | 11 | 19.0 |
| Primary | 4 | 6.9 |
| Middle | 5 | 8.6 |
| High school | 17 | 29.3 |
| Higher secondary | 14 | 24.1 |
| Graduate | 6 | 10.3 |
| Postgraduate | 1 | 1.7 |
| Marital status | Married | 49 | 84.5 |
| Single | 9 | 15.5 |
| Family type | Nuclear | 47 | 81.0 |
| Joint | 11 | 19.0 |
| Residence | Rural | 26 | 44.8 |
| Urban | 32 | 55.2 |
| Religion | Hindu | 52 | 89.7 |
| Muslim | 5 | 8.6 |
| Christian | 1 | 1.7 |
Explanatory models and scores.
| Variables | Minimum | Maximum | Mean | Std. Deviation (±) |
|---|
| Total score on MDEMQ | 60.00 | 191.00 | 97.05 | 22.76 |
| Stress | 20.00 | 72.00 | 39.03 | 9.42 |
| Western physiology | 11.00 | 35.00 | 19.72 | 5.53 |
| Non-Western physiology | 4.00 | 13.00 | 5.62 | 2.15 |
| Supernatural | 19.00 | 73.00 | 32.67 | 13.13 |
In the Mann-Whitney U, test a statistically significant group difference on the score of MDEMQ was observed for patient’s marital status (Mann-Whitney U=113.500, p=0.05, sig≤0.05, 2-tailed) and family history of mental illness (Mann-Whitney U=165.5, p=0.03, sig≤0.05, 2-tailed). No statistically significant result was observed in Kruskal Wallis Test [Table/Fig-3].
Group difference of demographic variables on score of MDEMQ.
| Variables | N | Mean Rank | Sum of Ranks | Mann-Whitney U | Z | Asymp. Sig. (Two-tailed) |
|---|
| Gender | Male | 24 | 28.16 | 704.0 | | | |
| Female | 34 | 30.52 | 1007.0 | 379.0 | -0.527 | 0.59 |
| Occupation | Unemployed | 39 | 31.25 | 1187.5 | | | |
| Employed | 19 | 26.18 | 523.5 | 313.5 | -1.091 | 0.27 |
| Religion* | Hindu | 52 | 29.92 | 1526.0 | | | |
| Muslim | 5 | 26.43 | 185.0 | 157.0 | -0.514 | 0.60 |
| Education | Uneducated | 11 | 27.71 | 332.5 | | | |
| Educated | 47 | 29.97 | 1378.5 | 254.5 | -0.414 | 0.67 |
| Marital status | Married | 49 | 30.68 | 1503.5 | | | |
| Single | 9 | 18.69 | 149.50 | 113.5 | -1.900 | 0.05 |
| Family type | Nuclear | 47 | 28.68 | 1319.5 | | | |
| Joint | 11 | 32.62 | 391.5 | 238.5 | -0.722 | 0.47 |
| Domicile | Rural | 26 | 28.61 | 772.5 | | | |
| Urban | 32 | 30.27 | 938.5 | 394.5 | -0.375 | 0.70 |
| Family history | Without family history | 46 | 31.90 | 1467.5 | | | |
| With family history | 12 | 20.29 | 243.5 | 165.5 | -2.126 | 0.03 |
| Knowledge of treatment | Magico- religious | 11 | 33.82 | 372.0 | | | |
| Allopathic | 47 | 28.49 | 1339.0 | 211.0 | -0.944 | 0.34 |
| Source of information | Family and society | 19 | 25.47 | 484.0 | | | |
| Health professionals | 39 | 31.46 | 1227.0 | 294.0 | -1.271 | 0.20 |
* one patient was Christian
A linear regression was conducted using the enter method to see if MDEMQ score can predict the value of MMAS [Table/Fig-4]. Five predictors explained 93.7% of the variance (R2=0.937, F= 153.558, p<0.001). These predictors were total score of MDEMQ, (Beta=-2.064, t=-13.675, p<0.001) and its subscales Stress (Beta=0.651, t= 6.265, p<0.001), Western physiology (Beta=0.437, t= -7.339, p <0.001), Non-Western physiology (Beta=0.244, t= 3.693, p<0.001) and Supernatural (Beta=1.094, t=9.701, p<0.001) [Table/Fig-5].
Group difference of demographic variables on score of MDEMQ in Kruskal Wallis Test.
| Variables | N | MeanRank | Chi-square | df Asymp.Sig |
|---|
| Knowledgeof courseof illness | Continuous | 8 | 36.12 | 2.318 | 2.31 |
| Recurrent | 42 | 29.42 |
| Other | 8 | 23.31 |
| Preferredmethod | Pharmacologicaltreatment only | 49 | 30.50 |
| Pharmacologicaland psychological | 8 | 25.62 | 1.738 | 2.41 |
| Magico -religious | 1 | 11.50 |
| Referredby | Self | 5 | 43.60 |
| Family members | 43 | 29.62 | 5.514 | 2.06 |
| Health professionals | 10 | 21.95 |
| Treatmenttypesoughtbefore | Magico-religious | 27 | 30.30 |
| Allopathic | 29 | 28.00 | 1.143 | 2.56 |
| Ayurvedic | 2 | 40.50 |
Regression of explanatory models.
| ModelPredictors | UnstandardizedCoefficients | StandardizedCoefficients | t | Sig. |
|---|
| Beta | Std. Error | Beta |
|---|
| (Constant) | 30.051 | 2.241 | | 13.407 | 0.000 |
| Stress | 0.829 | 0.132 | 0.651 | 6.265 | 0.000 |
| Western physiology | 1.037 | 0.141 | 0.437 | 7.339 | 0.000 |
| Non-Western physiology | 1.251 | 0.339 | 0.244 | 3.693 | 0.001 |
| Supernatural | 1.116 | 0.115 | 1.094 | 9.701 | 0.000 |
Dependent variable: Total score MMAS
R2=0.937, F=153.558, p<0.001
Discussion
Demographic characteristics
Demographic characteristics can be explained on the basis of a population characteristic around the study centre. Hindus are among the majority, literacy is on the rise, there is a trend of the nuclear family and unemployment is a major problem in that community.
Explanatory models
Scores on MDEMQ (both total and subscales) indicate patients had multiple explanatory models of depression, similar to a report by Grover S et al., (2012) [5]. Such observation was also made for other mental illnesses in tertiary care setup [9,21,22]. Prevailed explanatory models are usually symptom based and depression as a syndrome is likely to have multiple explanatory models. Medical literacy is on the rise and patients are now likely to have medical model apart from traditional non-medical explanatory model. Probably, non-scientific and scientific +++belief may coexist in the same individual at the same time as multiple facets of a broader worldview [23,24].
Explanatory models and demographic variables
Statistically significant group difference was observed for patients marital status and family history of mental illness on the score of MDEMQ. In India, unmarried status or inability to get married in time (within 35 years of age) is often considered as a cause of mental illness, while after marriage inability to adjust with their spouse or in-laws is considered as the cause [25,26]. After marriage due to cultural practice the patient belonging to Hindu family are more likely to harbor non-Western explanations such as ayurvedic concepts of mental illnesses (that mental illness is a result of excited gas, vata dosha or imbalanced food intake) and other traditional believe [27–29]. Patients with a family history of mental illness are more likely to be subjected for validation of supernatural explanation of illness due to their family background of mental illness [30,31].
Explanatory models and treatment adherence
Consistent with our hypothesis, a linear regression analysis subscales score of MDEMQ was inversely associated with level of medication adherence. As per the MMAS scoring instruction more score actually indicates less adherences. Thus, a statistically positive association between MDEMQ and MMAS is clinically an inverse relation. Interestingly models did not have significant differential effect on medication adherence. Models of illness are often determined by demographic and cultural characteristics of patients that may directly or indirectly influence the medication adherence [32]. In a qualitative study, Buus N et al., found that psychosocial or biomedical models are non- prominent in patients with depression [18]. If medicine is ineffective, the explanatory models legitimised alternative strategies towards recovery, including non-adherence [18]. In another report non-adherence was more due to apprehension of addictive or harmful effect of antidepressant, and stigma associated with psychotropic medication [33]. Thus, health care service provider must be aware of and sensitive to the patient’s model of illness, because beliefs and perceptions influence individual, family and community acceptability to treatment [34].
Limitation
Limitations of this includes small sample size, no socio-cultural evaluation, and knowledge about the illness was not assessed (especially biomedical model). Further studies are needed addressing limitation of this study.
Conclusion
Findings of this study suggest that patients with depression harbor multidimensional explanatory model. The levels of explanatory models are inversely associated with levels of medication adherence.
* one patient was ChristianDependent variable: Total score MMASR2=0.937, F=153.558, p<0.001
[1]. Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, National Comorbidity Survey Replication. The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R) JAMA 2003 289(23):3095-105. [Google Scholar]
[2]. Fung K, Wong YLR, Factors influencing attitudes towards seeking professional help among East and Southeast Asian immigrant and refugee women Int J Soc Psychiatry 2007 53(3):216-31. [Google Scholar]
[3]. McCabe R, Priebe S, Explanatory models of illness in schizophrenia: Comparison of four ethnic groups Br J Psychiatry 2004 185:25-30. [Google Scholar]
[4]. Fritzsche K, Anselm K, Fritz M, Wirsching M, Xudong Z, Schaefert R, Illness attribution of patients with medically unexplained physical symptoms in China Trans cult Psychiatry 2013 50(1):68-91. [Google Scholar]
[5]. Grover S, Kumar V, Chakrabarti S, Hollikatti P, Singh P, Tyagi S, Explanatory models in patients with first episode depression: A study from North India Asian J Psychiatry 2012 5(3):251-57. [Google Scholar]
[6]. Patra BN, Grover S, Aggarwal M, Avasthi A, Chakrabarti S, Malhotra S, Explanatory models in patients with obsessive compulsive disorder: An exploratory study Indian J Soc Psychiatry 2013 29(1-2):38-45. [Google Scholar]
[7]. Kleinman A, Patients and healers in the context of culture: An exploration of the borderland between anthropology, medicine, psychiatry 1980 BerkeleyUniversity of California Press [Google Scholar]
[8]. Karasz A, Cultural differences in conceptual models of depression Soc Sci Med 2005 60(7):1625-35. [Google Scholar]
[9]. Chakraborty K, Das G, Dan A, Bandyopadhyay G, Chatterjee M, Perceptions about the cause of psychiatric disorders and subsequent help seeking patterns among psychiatric outpatients in a tertiary care centre in Eastern India German J Psychiatry 2013 16(1):7-14. [Google Scholar]
[10]. Ram D, Patil S, Basavana Gowdappa H, Level of paranormal beliefs and its relationship with explanatory models, treatment adherence and satisfaction Arch Clin Psychiatry 2016 43(3):51-55. [Google Scholar]
[11]. Callan A, Littlewood R, Patient satisfaction: Ethnic origin or explanatory model? Int J Soc Psychiatry 1998 44(1):1-11. [Google Scholar]
[12]. Dejman M, Ekblad S, Forouzan AS, Baradaran-Eftekhari M, Malekafzali H, Explanatory model of help-seeking and coping mechanisms among depressed women in three ethnic groups of Fars, Kurdish and Turkish in Iran Arch Iran Med 2008 11(4):397-406. [Google Scholar]
[13]. Hasin DS, Goodwin RD, Stinson FS, Grant BF, Epidemiology of major depressive disorder: Results from the National Epidemiologic Survey on Alcoholism and Related Conditions Arch Gen Psychiatry 2005 62(10):1097-106. [Google Scholar]
[14]. Kripalani S, Yao X, Haynes RB, Interventions to enhance medication adherence in chronic medical conditions: A systematic review Arch Intern Med 2007 167(6):540-50. [Google Scholar]
[15]. Pampallona S, Bollini P, Tibaldi G, Kupelnick B, Munizza C, Patient adherence in the treatment of depression Br J Psychiatry 2002 180:104-09. [Google Scholar]
[16]. Okello ES, Neema S, Explanatory models and help-seeking behaviour: Pathways to psychiatric care among patients admitted for depression in Mulago hospital, Kampala, Uganda Qual Health Res 2007 17(1):14-25. [Google Scholar]
[17]. Hwang WC, Myers HF, The explanatory model of illness catalogue: Ethnic Differences in women’s illness beliefs and help-seeking for depression J Cult Diversity 2013 20(2):57 [Google Scholar]
[18]. Buus N, Johannessen H, Stage KB, Explanatory models of depression and treatment adherence to antidepressant medication: A qualitative interview study Int J Nurs Stud 2012 49(10):1220-29. [Google Scholar]
[19]. Eisenbruch M, Classification of natural and supernatural causes of mental distress. Development of a mental distress explanatory model questionnaire J Nerv Ment Dis 1990 178(11):712-19. [Google Scholar]
[20]. Morisky DE, Green LW, Levine DM, Concurrent and predictive validity of a self-reported measure of medication adherence Med Care 1986 24(1):67-74. [Google Scholar]
[21]. Kermode M, Bowen K, Arole S, Joag K, Jorm AF, Community beliefs about treatments and outcomes of mental disorders: A mental health literacy survey in a rural area of Maharashtra, India Public Health 2009 123(7):476-83. [Google Scholar]
[22]. Andrew G, Cohen A, Salgaonkar S, Patel V, The explanatory models of depression and anxiety in primary care: A qualitative study from India BMC Research Notes 2012 5:499 [Google Scholar]
[23]. Nemeroff C, Rozin P, The makings of magical mind. In: Rosengren KS, Johnson CN, Harris PL eds Imagining the impossible: Magical, scientific, and religious thinking in children 2000 New York, NYCambridge University Press [Google Scholar]
[24]. Irwin HJ, An introduction to parapsychology 2004 4th edJefferson, NCMcFarland:227 [Google Scholar]
[25]. Kotar AB, Mental health among married and unmarried women Int J Indian Psycho 2014 01(03):161-64. [Google Scholar]
[26]. Sharma I, Pandit B, Pathak A, Sharma R, Hinduism, marriage and mental illness Indian J Psychiatry 2013 55:S243-49. [Google Scholar]
[27]. Tirodkar MA, Baker DW, Makoul GT, Khurana N, Paracha MW, Kandula NR, Explanatory models of health and disease among South Asian immigrants in Chicago J Immigr Minor Health 2011 13(2):385-94. [Google Scholar]
[28]. Ramu MG, Venkataram BS, Manovikara (mental disorders) in ayurveda Ancient Sci Life 1985 4(3):165-73. [Google Scholar]
[29]. Acharya YT, Charakasamhita 1941 Bombay, IndiaNirnaya Sagar Press [Google Scholar]
[30]. Grover S, Nebhinani N, Chakrabarti S, Shah R, Avasthi A, Relationship between first treatment contact and supernatural beliefs in caregivers of patients with schizophrenia East Asian Arch Psychiatry 2014 24(2):58-67. [Google Scholar]
[31]. Srinivasan TN, Thara R, Beliefs about causation of schizophrenia: Do Indian families believe in supernatural causes? Soc Psychiatry Psychiatr Epidemiol 2001 36(3):134-40. [Google Scholar]
[32]. Kucukarslan SN, A review of published studies of patients’ illness perceptions and medication adherence: Lessons learned and future directions Res Social Adm Pharm 2012 8(5):371-82. [Google Scholar]
[33]. Cabassa LJ, Hansen MC, Palinkas LA, Ell K, Azúcar Y, Nervios: Explanatory models and treatment experiences of hispanics with diabetes and depression Social Science and Medicine (1982) 2008 66(12):2413-24. [Google Scholar]
[34]. Dahab M, Charalambous S, Hamilton R, Fielding K, Kielmann K, Churchyard GL, That is why I stopped the ART: Patients and providers? Perspectives on barriers to and enablers of HIV treatment adherence in a South Africa workplace programme BMC Public Health 2008 8:63 [Google Scholar]