BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay- per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected]

http://bmjopen.bmj.com/ on September 26, 2021 by guest. Protected copyright. BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open

Psychological distress and quality of life: Rationale and protocol of a prospective cohort study in a rural district in

For peer review only Journal: BMJ Open

Manuscript ID bmjopen-2017-016745

Article Type: Protocol

Date Submitted by the Author: 16-Mar-2017

Complete List of Authors: Uddin, Mohammed ; Swinburne University of Technology, Department of Statistics, Data Science and Epidemiology Bhar, Sunil ; Swinburne University of Technology, Department of Psychological Sciences Al-Mahmud , Abdullah ; Swinburne University of Technology, School of Design; Faculty of Health, Arts and Design Islam, Fakir; Swinburne University of Technology, Deapartment of Statistics, Data Science and Epidemiology

Primary Subject Public health Heading: http://bmjopen.bmj.com/ Secondary Subject Heading: Mental health

Psychological distress, Quality of Life, Validation, K1O, WHOQOL-BREF, Keywords: Rasch Analysis

on September 26, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 1 of 21 BMJ Open

Page 1 of 17 1 2 3 4 Psychological distress and quality of life: Rationale and 5 6 7 protocol of a prospective cohort study in a rural district in 8 9 10 Bangladesh 11 12 13 14 15 M Nazim UddinFor1, Sunil Bharpeer2, Abduallh reviewAl-Mahmud3, Fakir Monly Amirul Islam1,4* 16 17 18 1. Department of Statistics, Data Science and Epidemiology; Faculty of Health, Arts and 19 20 Design; Swinburne University of Technology, Hawthorn VIC 3122, Australia 21 22 23 2. Department of Psychological Sciences; Faculty of Health, Arts and Design; Swinburne 24 25 University of Technology, Hawthorn VIC 3122, Australia 26 27 3. School of Design; Faculty of Health, Arts and Design; Swinburne University of 28 29 Technology, Hawthorn VIC 3122, Australia 30 31

32 4. Organisation for Rural Community Development (ORCD), Dariapur, Narail, Bangladesh http://bmjopen.bmj.com/ 33 34 35 36 37 Email: M Nazim Uddin – [email protected]; Sunil Bhar - [email protected]; 38 39 Abdullah Al-Mahmud – [email protected]; Fakir M Amirul Islam* -

40 on September 26, 2021 by guest. Protected copyright. 41 [email protected] 42 43 44 Corresponding author: 45 46 47 Email: [email protected] 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 2 of 21

Page 2 of 17 1 2 3 ABSTRACT 4 Introduction: A significant proportion of the global burden of disease has been attributed to 5 mental and behavioural disorders. People with mental health disorders (MHD) have lower 6 levels of health-related quality of life than those without MHD. Several studies have shown 7 that in low-resource countries, a range of social determinants including poor health literacy 8 9 are critical in the epidemiologic transition of disease outcome. There is a lack of evidence of 10 mental health literacy, the prevalence and risk factors of common mental health problems, or 11 any validated instruments to measure psychological distress, or to evaluate the quality of life 12 in rural areas of Bangladesh. 13 14 Methods and Analysis: Using an appropriate statistical power and sample size calculation, a 15 sample of 1500For adults agedpeer 18-59 years review and 1200 older adults only aged 60 to 90 years will be 16 interviewed from a multi-level cluster random sample. Each participant will go through a 17 face-to-face interview, consisting of a semi structured questionnaire that will assess 18 participant awareness and knowledge of MHD, as well as their current mental health status 19 and how they managed such disorders. Along with the K10 and WHOQOL-BREF 20 questionnaire, information about the participant’s demographic and socio-economic status 21 22 will be collected. Internal consistency, validity, reliability, and item discrimination of K10 23 and WHOQOL-BREF instruments will be determined by item response theory. Awareness 24 and KAP of MHD will be assessed using Rasch analysis, and regression techniques. 25 Psychometric properties of WHOQOL-BREF in four domains will be assessed using 26 confirmatory factor analysis. Psychological distress severity will be defined from combined 27 K10 score and factors associated with it’s prevalence will be assessed. 28 29 Ethics and Dissemination: Human Ethics Approval was received from the Swinburne 30 University of Technology Human Ethics Committee (SHR Project 2015/065). Results of the 31 main trial will be submitted for publication in peer-reviewed journals. 32 http://bmjopen.bmj.com/ 33 34 Strengths and limitations of this study 35 36 • This study will be conducted in a large and representative sample from a typical rural 37 district to report the awareness, knowledge, attitudes and practice about mental health 38 disorders. Therefore, the study will provide a unique opportunity to assess the gaps in 39 public awareness of mental health disorders in rural Bangladesh.

40 • Data will be collected through a face-to-face interview process which will eliminate on September 26, 2021 by guest. Protected copyright. 41 the possible response bias given the literacy level is different in males and females 42 43 and in different age groups. 44 • The study will be able to estimate the prevalence of and factors associated with 45 psychological distress to conduct appropriate intervention programs. 46 • The study will use a sophisticated Rasch analysis technique to validate the 47 psychometric properties of the WHOQOL-BREF and Kessler 10 items (K10) 48 questionnaire to use in Bangladesh. The study will provide the first validated tools to 49 measure psychological distress and quality life in rural Bangladesh, which can be 50 used in other developing countries with the similar socio-economic condition. 51 • The limitation of the study include, we are attempting to conduct in one area of rural 52 53 Bangladesh. The study would need to be repeated in a random sample of other remote 54 areas in order for the results to be truly representative of the national level. 55 56 Keywords: Psychological distress, quality of life, validation, Rasch analysis, WHOQOL- 57 BREF, K10, Rural Bangladesh 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 3 of 21 BMJ Open

Page 3 of 17 1 2 3 Background 4 Mental Health Disorders: 5 Mental health disorders (MHD) are one of the leading contributors to the global burden of 6 disease in both high and low income countries, accounting for 13% of disease measured as 7 1 8 disability adjusted life years . Common types of MHD include conditions such as major 9 depressive disorders, bipolar disorders, anxiety disorders, psychotic disorders, and substance 2 10 use disorders . Of these, major depressive and anxiety disorders contribute to approximately 11 50% of the international disease burden 3. Such conditions are frequently comorbid with other 12 health and psychosocial problems and are among the most serious causes of morbidity 4-6. 13 Despite the adverse impacts of such mental health conditions, they receive little attention in 14 most low and middle-income countries, and hence, treatment of such conditions are not 15 considered nationalFor health peer priorities 7 8. Mentalreview health complaints only create a major public health 16 4 9 10 17 concern in Bangladesh, especially in rural areas . The prevalence of mental health disorders in such areas varies between 6.5% and 31% possibly due to the use of different 18 11 19 protocol and definitions of MHD . 20 21 Factors associated with mental health disorders: 22 23 A number of factors have been identified as increasing risk for MHD. These factors include a 24 lack of access to, and utilisation of, mental health services, stigma about mental health 25 treatment, and poor physical health 12. MHD are present to a greater extent amongst low 26 socioeconomic status (SES) groups 13, and research shows that financially poor countries 27 14-18 28 with poor health literacy are critical in the epidemiologic transition of disease outcome . 29 A lot of evidence supported that knowledge, attitude and practice (KAP) studies are needed 30 for greater awareness of prevention, diagnosis, risk factor control, and disease management 31 of psychological distress 14-21.

32 http://bmjopen.bmj.com/ 33 34 Mental health literacy has been defined as “knowledge and attitudes about MHD which aid 35 their recognition, management, and/or prevention” 22. Knowledge and positive attitudes 36 towards treatment are fundamental for individuals seeking help for MHD in a timely and 37 appropriate manner. In India, recent studies have reported limited knowledge of mental 38 health issues and high rates of negative attitudes towards seeking help for mental health 39 problems 23 24, yet similar literature for the neighbouring region of Bangladesh is lacking.

40 on September 26, 2021 by guest. Protected copyright. 41

42 43 Quality of Life: 44 45 46 In recent years, beyond traditional health indicators such as mortality and morbidity, there 47 has been a greater focus on measuring health and quality of life (QOL) as an important 25 48 outcomes in clinical and interventional studies . QOL is used as an important outcome 49 measure for evaluating intervention effects, such as the effect of medicine 26. The World 50 Health Organisation (WHO) defines quality of life (QOL) as an “individuals’ perception of 51 their position in life in the context of the culture in which they live and the value systems they 52 have in relation to their goals, expectations, standards and concerns” 27. 53 54 Quality of life is affected by a range of factors such as physical health, psychological state, 55 level of independence, social relationships, personal beliefs, and their relationships to silent 56 features of their environment 28. However, most of the research on QOL is focused on the 57 58 deleterious impact of chronic physical health diseases such as cancer, stroke, diabetes and 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 4 of 21

Page 4 of 17 1 2 3 HIV/AIDS 29-32. Furthermore, over the past 20 years, although a quite a number of tools have 4 been developed to measure QOL 33, with a few exceptions 34-36, these measure of QOL as 5 impacted by specific diseases. As an exception, the QOL Scale (QOLS) first developed by 6 American psychologist, John Flanagan 37 38, fits a more generalised definition of QOL that 7 can be used to assess QOL in an everyday context. The WHOQOL-BREF is a shortened, 26- 8 item instrument, rated on five-point Likert-type scales, planned to be used for large 9 39 10 population studies . This tool has not been applied to a general population setting in any 11 developing countries, including Bangladesh. 12 13 A Bangla version of the WHOQOL-BREF was initially developed in 2005 for use in a study 40 41 14 of adolescents and adults, residing only in the capital city of Bangladesh, . Since 42 43 15 then a BengaliFor version ofpeer the scale has beenreview used in some rural only areas of the country . The 16 dimensionality of WHOQOL-BREF Bangla version was assessed using classical test theory 17 (CCT) 40 41. In CTT approach, the items and the person latent trait are measured separately 18 and provide descriptive measures. In the CTT method, true score estimates are typically 19 obtained by summing responses across items, with the assumption that each item within the 20 same construct is valued equally. A further assumption of CCT is that the score difference 21 between the two adjacent response scales is uniform; but this does not occur in all 22 44 23 circumstances . Therefore, the CTT cannot be meaningfully and methodically compared in 45 46 24 an item-person continuum . These restrictions can be resolved using item response theory 25 (IRT) modelling (Rasch Analysis) despite IRT’s own assumptions regarding 26 unidimensionality, invariability and local independence 45 46. 27 28 Psychological distress: 29 30 In recent years, epidemiological studies have attempted to use less items to proficiently 31 47

32 measure and monitor the extent of psychological distress in the widespread community . http://bmjopen.bmj.com/ 33 For the purpose of measuring depression and anxiety few scales have been used so far. These 48 34 include the Beck Depression Inventory (BDI) , the Hospital Anxiety and Depression Scale 49 50 51 35 (HADS) , the Depression Anxiety Stress Scales (DASS) and the Kessler K10 . The K10 36 is one of the most popular tools for screening psychological distress in the general population 37 52. K10 comprises 10 items, rated on five-point Likert-type scales, which indicate the degree 38 of psychological distress are present among persons 51. The K10 have been established 39 predominantly from Western population samples 51-56. Information on cross-cultural validity

40 on September 26, 2021 by guest. Protected copyright. of the K10 is not available in Bangladesh. Therefore, we have used an interview-administered 41 57 42 version of K10 in our study due to the low rate of literacy in Bangladesh . 43 44 Reliability and validity of K10 and WHOQOL questionnaires: 45 46 47 Reliability and validity are context-specific attributes, showing applicability to some 48 populations over others. Where quality of life and physiological distress assessment 49 instruments are used in a variety of cultural settings, it is important to establish whether the 50 same aspects of life are equally important for adults across the lifespan, males and females, 51 people with and without education and for people with various relationship and employment 52 status. The above factors report to impact differently on people’s lives, but how significant 53 these differences are, and the issue of whether a reasonably valid tool can be developed in 54 58 55 spite of these differences needs to be addressed in Bangladesh . Therefore, the validation of 56 WHOQOL questionnaire as well as K10 questionnaire in a general population, both in adults 57 and older adults can provide empirical development of literature on these issues in 58 Bangladesh. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 5 of 21 BMJ Open

Page 5 of 17 1 2 3 4 Aims of the study 5 6 7 The current prospective cohort study will be conducted in a rural district of Bangladesh to 8 address a number of research questions and the specific aims are as follows: 9 10 I. Report the awareness, knowledge, attitudes and practice (KAP) using seven items 11 from the mental health literacy questionnaire of MHD. This information would be 12 useful in the formulation of policy for the awareness, training, management and 13 service delivery in regards to mental health issues in rural areas of Bangladesh and 14 15 may Forbe used as a peermodel in low and review middle-income countries only around the world. 16 II. Estimate the prevalence of psychological distress in both adults and older adults 17 according to socio-demographic status. 18 III. Estimate the quality of life in adults and older adults, their risk factors and the 19 variation according to the socio-demographic factors. 20 IV. To test the internal validity of the K10 scale using Rasch analysis 21 V. To investigate the factor structure of K10 scale using CFA and EFA 22 VI. To test the internal validity of the four domain WHOQOL-BREF questionnaire using 23 Rasch analysis. 24 25 VII. To assess the factor structure of the WHOQOL-BREF in a Bangladeshi rural general 26 population sample CFA and EFA 27 28 Methods 29 30 Study Population: 31

32 Bangladesh is a country of 160 million people divided into 64 districts. Each district is http://bmjopen.bmj.com/ 33 divided into sub-districts named (493 in total), and each further divided into 34 several Unions and the city centre known as pourashava. Each Union consists of 15-25 35 villages and the pourashava consists of 5-15 wards, those are divided into para or localities 36 that are comprised of several households. Participants will be recruited from the Narail 37 district (Figure 1). Narail is located approximately 200 km southwest of Dhaka. The 38 population of Narail District is 272,872, with approximately 40% of residents aged between 39 18-59 years and 19,000 (about 7%) of residents aged between 60-90 years. The study

40 on September 26, 2021 by guest. Protected copyright. 41 location was selected as it was considered to be representative of a typical rural demographic 42 in Bangladesh. The Narail District, with an estimated population density of 722 people per 2 2 43 km is comparable to the national rural population density of 873 people per km . It is not at 44 the extremity of remote locations nor is it a catchment of a metropolis such as Dhaka 59. 45 46 Sample size and statistical power 47 48 Prior data indicate that the prevalence of severe depression was 21% in older adults aged 60 60 49 or above, and 6.5% in adults aged between 18 and 60 . We assumed a margin of error of 5% 50 in prevalence rates for older adults, and of 3% in adults when estimating the true prevalence 51 of severe depression for each cohort in this rural area. Using a significance level of 0.05 and 52 statistical power above 80%, a required sample size of 1128 is needed for the older adults and 53 1283 for the adults, respectively. We did not adjust for cluster effects because only a single 54 participant will be recruited from each household. However, assuming non-response and 55 56 other possible problems such as; irrelevant or duplicate data, erroneous or misinterpreted data 57 and too little data acquired from the client in data collection, the sample size was adjusted. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 6 of 21

Page 6 of 17 1 2 3 The recruitment target will be increased to 1200 older adults aged between 60-90 years and 4 1500 adults aged between 18-59 years. . 5 6 7 8 Sampling Frame 9 10 A multi-level cluster random sampling technique will be used for this prospective cohort 11 study. Three unions from a total of 13 and one ward from a total of 9 located within Narail 12 District were randomly selected at level 1. Two to three villages or mahalla from each 13 selected union or ward were randomly selected at the second level. The Unions and Ward are 14 shown in Figure 1. The selected villages from different Unions were as follows: Banshgram 15 union: Tabra,For Betvita and peer Ballertop, Bhadrabilla review union: Ramshidi, only Bogura and Palaidanga, 16 17 Tularampur Union: Betenga, Mitna and Bamanhat, Ward 1: Kurigram and Maijpara. The 18 projected recruitment target will be set at 120 older adults and 150 adults from each village. 19 Each village or ward comprised between 360 to 480 households; and one eligible adult or 20 older adult will be interviewed from each household. 21 22 23 24 Recruitment strategy 25 26 First, the chief investigator and the head of the Department of Statistics, Data Science and 27 Epidemiology at the Swinburne University of Technology, Australia have written a request 28 letter to the district commissioner, district civil surgeon, police super, and union parishad 29 chairman’s for their cooperation. Second, interviewers will be provided a general overview of 30 the research to village leaders and/or school teachers, and will request their cooperation. 31 Once consent by the village leaders will be provided, the interviewer approached prospective 32 participants to explain the purpose of the study and to invite them to participate in the study. http://bmjopen.bmj.com/ 33 34 The interviewers will be instructed to interview an older adult first. If none were available in 35 this subgroup, the interviewer will approach an adult person of that household. Again, if 36 there will be more than one male or female adult in the same household, one individual was 37 38 selected, based on who was born closer to the month of January. However, to maintain an 39 approximately equal number of males and female participants, one female will be interview immediately after an interview of a male participant. Exclusion criteria will be: age of less 40 on September 26, 2021 by guest. Protected copyright. 41 than 18 years or more than 90 years, and having an illness that prohibited participation. 42 43 Quality assurance 44 All team members will be participated in an intensive two-day training program in Narail 45 before the will commencement of the survey. The purpose of the training is to outline the 46 47 rationale of the study, and the procedures and potential difficulties associated with data 48 collection. 49 50 51 52 Data Collection 53 Questionnaires 54 55 An interviewer-administered semi-structured questionnaire is being developed to collect 56 relevant socio-demographic data and to assess participant’s awareness, knowledge, attitudes 57 58 and practice regarding MHD. Data will be collected through a face-to-face interview 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 7 of 21 BMJ Open

Page 7 of 17 1 2 3 process, which allowed a greater possibility of collection of more truthful answers. Questions 4 regarding income and financial support will be collected. Smoking and smokeless tobacco 5 consumption status data will be collected as well. The questionnaire is shown in Table 1. 6 7 8 9 Awareness, Knowledge, Attitude and Practice about mental health disorders 10 11 Given the relative lack of validated mental health KAP data, the specific items included in the 12 final questionnaire were derived from the National Survey of Mental Health Literacy and 13 Stigma by Reavley and Jorm (2011) 61. The questionnaire will be assessed awareness of 14 seven common MHD (depressive disorders, bipolar disorders, anxiety disorders, psychotic 15 disorders, andFor substance peer use disorders ) reviewby asking if they have only ever heard of these diseases 16 with a possible response of “yes” or “no”. If they will response “yes”, they will be requested 17 to list at least one symptom of the disease to provide insight into the level of knowledge they 18 possessed. Attitudes towards use of treatment will be assessed by asking “do they need 19 20 treatment?” with a possible response of “yes” or “no”. The practice about MHD will be 21 assessed by asking if the participant or any of his/her relatives had ever experienced any 22 mental health problem, and from whom they would seek treatment, with a possible response 23 of “medical doctor”, “psychologist” or others such as “spiritual persons”. The detail 24 questions about KAP is reported in different section after Table 1. 25 26 27 Kessler Psychological Distress scale (K10) and WHOQOL-BREF questionnaires: 28 29 The K10 51 was used to assess psychological distress and World Health Organization Quality 30 of Life (WHOQOL-BREF) 62 was used to assess the quality of life of the participants. The 31 K10 was translated into Bengali independently by two bilingual translators including a 32 medical practitioner with experience in public health. The K10 questionnaire was translated http://bmjopen.bmj.com/ 33 using back-translated techniques 63. Regarding the WHOQOL questionnaire, the chief 34 investigator (AI) contacted the original developers of the WHOQOL-BREF quality of life 35 64 36 assessment test team to seek permission to use the Bengali version of the WHOQOL-BREF 37 for research purposes in Bangladesh. This version was also translated to accommodate Narail 38 languages. Validity of such translations were checked through the back-translation procedure 39 63.

40 on September 26, 2021 by guest. Protected copyright. 41 42 Planned Data Collection: 43 44 Location: Narail Upazilla (Figure 1). Duration of data collection: April-July, 2017 45 46 47 48 Outcome Variables: The first outcome variable will be the prevalence of psychological 49 distress which will be defined based on the sum of a five value response option for each of 10 50 Kessler questionnaire items, with the maximum score of 50 and the minimum score of 10, 51 then categorised according to the score obtained; no psychological distress (10 to 19), mild 52 (20 to 24), moderate (25 to 29) and severe psychological distress (30 to 50). 53 54 The second outcome variables will be the prevalence of awareness of MHDs comprised of 55 seven common mental health disorders: depression, anxiety, psychosis, drug addiction, 56 dementia, bipolar disorder and Alzheimer’s disease, defined as if the participants have ever 57 heard about the disorders. Prevalence of knowledge will be defined if the participants are 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 8 of 21

Page 8 of 17 1 2 3 able to identify at least one symptoms of the MHDs of them who are aware of the disorders, 4 positive attitudes will be defined if the participants who are aware of the MHDs are in favour 5 of taking medical treatment or psychological counselling, and practice if participants with 6 MHDs already take medical treatment or psychological counselling. 7 8 Validated K10 Questionnaire: the K10 comprises 10 questions that will be answered using a 9 five -point scale. We will use Rasch analysis to achieve adequate model fit for K10 that will 10 give internal consistency, reliability, unidimensionality and freedom from differential item 11 12 function (DIF) for sex (male/ female) and age (adult and elderly). 13 Validated WHOQOL questionnaire: To assess the quality of life of rural people in 14 15 Bangladesh For we will use peer the WHOQOL-BREF review tool, which consistsonly of 26 questions with 16 structured responses on a Likert scale of five points. Of the 26 questions, two assess the 17 perception of quality of life and health of the patient, and the others (24 questions) comprise 18 the physical, psychological, social and environment domains. Therefore, we will use Rasch 19 analysis, CFA and EFA for each domain and that will provide an optimal evaluation of the 20 quality of life of rural people. 21 22 23 24 Exposure variables: Demographic details for age, categorised as either adult (18 to 59 years) 25 or older adult (60 to 90 years), gender, level of education – categorised as no schooling, 26 primary school education (grade 1 to 5), secondary school education (grade 6 to 10) and 27 school secondary certificate (SSC) or above and socio-economic status (SES) will be 28 collected. SES was assessed according to Cheng et al. 65 asking whether "over the last twelve 29 30 months, in terms of household food consumption, how would you classify your socio- 31 economic status?" The possible answers were: (i) insufficient funds for the whole year; (ii)

32 insufficient funds some of the time; (iii) neither deficit nor surplus (balance); and (iv) http://bmjopen.bmj.com/ 33 sufficient funds most of the time. Data on current health problems (yes or no), number of 34 health problems, medication use and smoking status into three categories: ‘Never smoking’, 35 ‘Ever smoking’ and ‘Smokeless tobacco use’ will also be collected during the interview. 36 37 38 39 Planned Statistical analysis:

40 on September 26, 2021 by guest. Protected copyright. 41 Primary objective 1: Awareness, knowledge, attitude and practice of mental health 42 disorders and prevalence of mental health in rural area 43 44 Rasch analysis will be performed to compute person measures based on the awareness of 45 seven common MHD in a logarithmic scale and termed as ‘‘awareness score”. Awareness, 46 knowledge, attitudes and practice will be reported by age, gender, level of education and 47 other socio-economic factors using Chi-square tests. Multiple regression analysis techniques 48 will be used to investigate the factors associated with the combined awareness score. 49 RUMM2030 and SPSS 23 Software will be used for analysis. 50 51 An estimate of the prevalence of psychological distress will be calculated (using standard cut- 52 off values of combined K10 score, according to Victorian Population Health Survey) 66. 53 54 Factors associated with psychological distress will be investigated using Chi-square test and 55 binary or multi-nominal logistic regression techniques. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 9 of 21 BMJ Open

Page 9 of 17 1 2 3 4 5 Primary objective 2: Evaluate and Validate the Kessler K10 6 7 Rasch analysis will be used to find a possible unidimensional structure of K10. Moreover, we 8 will use the Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) to 9 explore the factor structure of the K10 questionnaire. 10 11 12 13 Primary objective 3: Evaluate and Validate the WHOQOL-BREF 14 15 ConfirmatoryFor Factor Analysis peer (CFA) will review be applied to the original only four domains (physical, 16 psychological, environmental, and social), assuming the orthogonally between factors and 17 invariance of the structure across items. Rash analysis will be used to check uni- 18 dimensionality, local independence, and invariability of the original four domains. If needed, 19 we will take necessary adjustments to improve the model fit. Finally, the proposed model 20 will be tested to get possible psychometric structural model for the WHOQOL-BREF in the 21 rural area in Bangladesh. 22 23 Follow-up data collection: 24 25 Using sophisticated statistical model we would expect to get a unidimensional structure of 26 K10 and WHOQOL-BREF. We plan to collect additional 300 samples to justify our findings. 27 28 29 30 Discussion 31 32 A questionnaire that demonstrates reliability, validity, and responsiveness in a certain context http://bmjopen.bmj.com/ 33 may not necessarily be appropriate for other contexts. The validation of the Bengali version 34 35 of the K10 as well as the WHOQOL-BREF questionnaires will enable them to be used in a 36 broader context in which similar circumstances arise. The K10 has not been used previously 37 in Bangladesh. Previously, the Bengali version of WHOQOL-BREE was used in specific 40 41 42 43 38 contexts such as in slum areas and on disease prone individuals . which limited the 39 validity, reliability and the general utility of the questionnaire. However, they failed to

40 include people of middle income, general rural people, and elderly people. The properties of on September 26, 2021 by guest. Protected copyright. 41 the test may change when it is used to assess people that are more representative of the 42 national socio-economic condition. In the present study, the inclusion of rural people aged 43 between 18-59 and 60-90 enhances the usefulness of the questionnaire. The validation of the 44 45 Bengali versions of WHOQOL-BREF and K10 are expected to enable generalisability of the 46 scales and will allow comparisons to be made between developing countries as well as 47 contrast between classical test theory and item response theory. 48 49 The study will present a unique opportunity to assess the gaps in public awareness of mental 50 health disorders in a rural district of Bangladesh. The purpose of the project is to create an 51 opportunity to work with a local non-government organization (the Organization for Rural 52 Community Development) and capture information that would be of practical benefit to 53 health policy planners in the Narail district and local health authorities. This information 54 could ultimately benefit the general population. Moreover, another benefit of the study will 55 be use of traditional CCT, Rasch analysis and EFA to validate the psychometric properties of 56 the WHOQOL-BREF and K1O questionnaire. These will then enable the exploration of 57 58 different statistical methods in order to produce the best model for Bangladeshi rural 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 10 of 21

Page 10 of 17 1 2 3 population. The study will be conducted upon a large sample of both adults and older adults 4 and utilised face-to-face data collection. A targeted intervention might be needed to increase 5 awareness of MHD. Moreover, we will be able to determine the amount of rural populations 6 which experience physiological distress and will explore most common socio-economic 7 factors that drive psychological distress are. This will then help us identify which socio- 8 9 economic factors can be targeted initially. The study will explore how many items of the K10 10 scale and WHOQOL-BREF supports the measurement properties, internal consistency, 11 reliability and uni-dimensionality. This investigation will guide us using Rasch and 12 Confirmatory Factor Analysis to find an appropriate modified questionnaire in terms of 13 acceptability, validity, reliability and responsiveness. There have been no prior studies to 14 document such information regarding community knowledge of mental health in Bangladesh. 15 For peer review only 16 17 There are limitations of our study methodology which are acknowledged. The study will be 18 conducted in one area of rural Bangladesh. Whilst we will attempt to capture the situation in 19 Narail, the study would obviously need to be repeated in a random sample of other remote 20 areas in order for the results to be truly representative of a national perspective. 21 22 23 This will be the first study to prompt a critical analysis of the psychometric properties of the 24 25 WHOQOL-BREF and K10 using Rasch analysis and CFA. The outcomes will deliver 26 measurement properties, internal validity, reliability and unidimensionality of the scale. The 27 findings will guide us to recommend further studies using validated questionnaires in clinical 28 assessments in the appropriately targeted samples. The findings will discuss, in relation to the 29 and culture, social beliefs about mental disorders, and suggestions for 30 improvements in mental health literacy and stigma towards mental health. Furthermore, we 31 hope this study will enable the formation of an early intervention program, targeted at those 32 who are expected to diagnose with mental illness, such as mild or moderate levels of http://bmjopen.bmj.com/ 33 34 psychological distress. This program, when developed, has the potential to reduce the 35 likelihood of secondary effects such as loss of employment, school dropout, relationship 36 collapse, disability and drug and alcohol problems. The study will also be able to suggest an 37 appropriate program which rural areas require for early intervention, and this is of great 38 importance, as delay in getting help and support for mental health problems increases the 39 chances of experiencing a difficult recovery.

40 on September 26, 2021 by guest. Protected copyright. 41 Author Affiliations 42 43 1. Department of Statistics, Data Science and Epidemiology; Faculty of Health, Arts and 44 Design; Swinburne University of Technology, Hawthorn VIC 3122, Australia 45 2. Department of Psychological Sciences; Faculty of Health, Arts and Design; Swinburne 46 47 University of Technology, Hawthorn VIC 3122, Australia 48 3. School of Design; Faculty of Health, Arts and Design; Swinburne University of 49 Technology, Hawthorn VIC 3122, Australia 50 4. Organisation for Rural Community Development (ORCD), Dariapur, Narail, Bangladesh 51 52 53 List of abbreviations 54 55 MHD, Mental health disorders; WHOQOL, World Health Organization Quality of Life; IRT, 56 Item response theory; CTT, Classical test theory; CFA, Confirmatory factor analysis; KAP, 57 Knowledge, attitudes and practice. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 11 of 21 BMJ Open

Page 11 of 17 1 2 3 4 5 Acknowledgement: 6 7 We acknowledge Dr Jason Skues to provide us valuable feedback about the project. We 8 9 particularly acknowledge the contribution of Md Rafiqul Islam, Md Sajibul Islam, Saburan 10 Nesa and Arzan Hossain who will collect the data from the participants households. Finally, 11 we would like to express our gratitude to the study participants for their voluntary 12 participation. 13 14 15 Author’s contribution:For peer review only 16 17 MNU and FMAI jointly designed the study. MNU drafted the manuscripts. SB, AAM and 18 FMAI contributed in writing the manuscripts. FMAI supervised the overall preparation of the 19 manuscript. All authors read and approved the final manuscript. 20 21 Funding 22 23 Data collection for this research project was funded by the Faculty of Health, Arts and Design 24 (FHAD) of the Swinburne University Technology under the Research and Development 25 Grant Scheme (RDGS). The funders had no role in the design of the study, data collection, 26 analysis, interpretation of data or writing the manuscript. 27 28 Competing interests: 29 30 The authors declare that they have no competing interests. 31 32 http://bmjopen.bmj.com/ 33 Ethical aspects: 34 35 Human Ethics Approval was received from the Swinburne University of Technology Human 36 Ethics Committee (SHR Project 2015/065) in accordance with the tenets of the Declaration of 37 Helsinki. Study participants provided written consent in order to participate. 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 References: 42 43 1. Tomlinson M, Swartz L, Daniels K. "No health without mental health" The global effort to 44 improve population mental health. Rout Stud Pub Hlth 2011:174-91. 45 2. Association AP. Diagnostic and statistical manual of mental disorders (DSM-5®): 46 American Psychiatric Pub, 2013. 47 3. Whiteford HA, Degenhardt L, Rehm J, et al. Global burden of disease attributable to 48 mental and substance use disorders: findings from the Global Burden of Disease 49 Study 2010. Lancet 2013;382(9904):1575-86. 50 51 4. Moussavi S, Chatterji S, Verdes E, et al. Depression, chronic diseases, and decrements in 52 health: results from the World Health Surveys. Lancet 2007;370(9590):851-58. 53 5. Patel V, Kleinman A. Poverty and common mental disorders in developing countries. B 54 World Health Organ 2003;81(8):609-15. 55 6. Lamers F, van Oppen P, Comijs HC, et al. Comorbidity Patterns of Anxiety and 56 Depressive Disorders in a Large Cohort Study: the Netherlands Study of Depression 57 and Anxiety (NESDA). J Clin Psychiat 2011;72(3):341-48. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 12 of 21

Page 12 of 17 1 2 3 7. Bruckner TA, Scheffler RM, Shen G, et al. The mental health workforce gap in low- and 4 middle-income countries: a needs-based approach. B World Health Organ 5 2011;89(3):184-94. 6 8. Hock RS, Or F, Kolappa K, et al. A new resolution for global mental health. Lancet 7 2012;379(9824):1367-68. 8 9 9. Hosain GMM, Chatterjee N, Ara N, et al. Prevalence, pattern and determinants of mental 10 disorders in rural Bangladesh. Public Health 2007;121(1):18-24. 11 10. Islam MM, Ali M, Ferroni P, et al. Prevalence of psychiatric disorders in an urban 12 community in Bangladesh. Gen Hosp Psychiat 2003;25(5):353-57. 13 11. Hossain MD, Ahmed HU, Chowdhury WA, et al. Mental disorders in Bangladesh: a 14 systematic review. Bmc Psychiatry 2014;14. 15 12. KilkkinenFor A, Kao-Philpot peer A, O'Neil A,review et al. Prevalence of onlypsychological distress, anxiety 16 and depression in rural communities in Australia. Aust J Rural Health 17 2007;15(2):114-19. 18 19 13. World Health Organization. The world health report: 2001. Mental health: new 20 understanding, new hope. Geneva: WHO; 2001. 2001. 21 14. Demaio AR, Otgontuya D, de Courten M, et al. Exploring knowledge, attitudes and 22 practices related to diabetes in Mongolia: a national population-based survey. BMC 23 Public Health 2013;13:236. 24 15. Rani PK, Raman R, Subramani S, et al. Knowledge of diabetes and diabetic retinopathy 25 among rural populations in India, and the influence of knowledge of diabetic 26 retinopathy on attitude and practice. Rural Remote Health 2008;8(3):838. 27 16. Zaman MJ, Patel A, Jan S, et al. Socio-economic distribution of cardiovascular risk 28 29 factors and knowledge in rural India. International journal of epidemiology 30 2012;41(5):1302-14. 31 17. Islam FMA, Chakrabarti R, Islam SZ, et al. Factors Associated with Awareness, Attitudes

32 and Practices Regarding Common Eye Diseases in the General Population in a Rural http://bmjopen.bmj.com/ 33 District in Bangladesh: The Bangladesh Population-based Diabetes and Eye Study 34 (BPDES). Plos One 2015;10(7). 35 18. Islam FM, Chakrabarti R, Dirani M, et al. Knowledge, attitudes and practice of diabetes 36 in rural Bangladesh: the Bangladesh Population based Diabetes and Eye Study 37 (BPDES). PLoS One 2014;9(10):e110368. 38 39 19. Norris SL, Engelgau MM, Narayan KM. Effectiveness of self-management training in type 2 diabetes: a systematic review of randomized controlled trials. Diabetes Care 40 on September 26, 2021 by guest. Protected copyright. 41 2001;24(3):561-87. 42 20. Al-Maskari F, El-Sadig M, Al-Kaabi JM, et al. Knowledge, attitude and practices of 43 diabetic patients in the United Arab Emirates. PLoS One 2013;8(1):e52857. 44 21. Mazzuca SA, Moorman NH, Wheeler ML, et al. The diabetes education study: a 45 controlled trial of the effects of diabetes patient education. Diabetes Care 1986;9(1):1- 46 10. 47 22. Jorm AF, Korten AE, Jacomb PA, et al. ''Mental health literacy'': A survey of the public's 48 49 ability to recognise mental disorders and their beliefs about the effectiveness of 50 treatment. Med J Australia 1997;166(4):182-86. 51 23. Liu MC, Tirth S, Appasani R, et al. Knowledge and Attitudes Toward Depression Among 52 Community Members in Rural Gujarat, India. J Nerv Ment Dis 2014;202(11):813-21. 53 24. Almanzar S, Shah N, Vithalani S, et al. Knowledge of and Attitudes Toward Clinical 54 Depression Among Health Providers in Gujarat, India. Ann Glob Health 55 2014;80(2):89-95. 56 25. Kuyken W, Orley J. Development of the Whoqol - Rationale and Current Status. Int J 57 Ment Health 1994;23(3):24-56. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 13 of 21 BMJ Open

Page 13 of 17 1 2 3 26. Chang KC, Wang JD, Tang HP, et al. Psychometric evaluation, using Rasch analysis, of 4 the WHOQOL-BREF in heroin-dependent people undergoing methadone 5 maintenance treatment: further item validation. Health Qual Life Out 2014;12. 6 27. Kuyken W, Orley J, Power M, et al. The World-Health-Organization Quality-of-Life 7 Assessment (Whoqol) - Position Paper from the World-Health-Organization. Soc Sci 8 9 Med 1995;41(10):1403-09. 10 28. WHOQOL-Measuring quality of life. The World Health Orgnaization Quality of Life 11 Instruments 1997. 12 29. Silva SM, Correa FI, Faria CDCD, et al. Psychometric properties of the stroke specific 13 quality of life scale for the assessment of participation in stroke survivors using the 14 rasch model: a preliminary study. J Phys Ther Sci 2015;27(2):389-92. 15 30. Debiec J,For Goch A, Chmielewski peer H, etreview al. Effect of diabetes onlyon neurological condition and 16 quality of life of patients with ischaemic stroke. Atherosclerosis 1999;144:192-92. 17 31. Jia HM, Zack MM, Thompson WW. The Effects of Diabetes, Hypertension, Asthma, 18 19 Heart Disease, and Stroke on Quality-Adjusted Life Expectancy. Value Health 20 2013;16(1):140-47. 21 32. Richardson J, Sinha K, Iezzi A, et al. Modelling utility weights for the Assessment of 22 Quality of Life (AQoL)-8D. Qual Life Res 2014;23(8):2395-404. 23 33. Berzon RA, Donnelly MA, Simpson RL, et al. Quality of life bibliography and indexes: 24 1994 update. Qual Life Res 1995;4(6):547-69. 25 34. Wahl AK, Rustoen T, Hanestad BR, et al. Quality of life in the general Norwegian 26 population, measured by the Quality of Life Scale (QOLS-N). Qual Life Res 27 2004;13(5):1001-09. 28 29 35. Ohaeri JU, Awadalla AW, El-Abassi AHM, et al. Confirmatory factor analytical study of 30 the WHOQOL-Bref: experience with Sudanese general population and psychiatric 31 samples. Bmc Med Res Methodol 2007;7.

32 36. Redko C, Rogers N, Bule L, et al. Development and validation of the Somali WHOQOL- http://bmjopen.bmj.com/ 33 BREF among refugees living in the USA. Qual Life Res 2015;24(6):1503-13. 34 37. Flanagan JC. Measurement of Quality of Life - Current State of the Art. Arch Phys Med 35 Rehab 1982;63(2):56-59. 36 38. Flanagan JC. Research Approach to Improving Our Quality of Life. Am Psychol 37 1978;33(2):138-47. 38 39 39. Development of the World Health Organization WHOQOL-BREF quality of life assessment. The WHOQOL Group. Psychol Med 1998;28(3):551-8. 40 on September 26, 2021 by guest. Protected copyright. 41 40. Izutsu T, Tsutsumi A, Islam M, et al. Validity and reliability of the Bangla version of 42 WHOQOL-BREF on an adolescent population in Bangladesh. Qual Life Res 43 2005;14(7):1783-89. 44 41. Tsutsumi A, Izutsu T, Kato S, et al. Reliability and validity of the Bangla version of 45 WHOQOL-BREF in an adult population in Dhaka, Bangladesh. Psychiat Clin Neuros 46 2006;60(4):493-98. 47 42. Zeldenryk L, Gordon S, Gray M, et al. Cognitive testing of the WHOQOL-BREF 48 49 Bangladesh tool in a northern rural Bangladeshi population with lymphatic filariasis. 50 Qual Life Res 2013;22(8):1917-26. 51 43. Laskar MS, Rahaman MM, Akhter A, et al. Quality of Life of Arsenicosis Patients in an 52 Arsenic-Affected Rural Area in Bangladesh. Arch Environ Occup H 2010;65(2):70- 53 76. 54 44. Bradley KD. Applying the Rasch model: Fundamental measurement in the human 55 sciences. Organ Res Methods 2005;8(2):249-50. 56 45. Bartholomew D. Fundamentals of Item Response Theory - Hambleton,Rk, 57 Swaminathan,H, Rogers,Hj. Brit J Math Stat Psy 1993;46:184-85. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 14 of 21

Page 14 of 17 1 2 3 46. Raykov T, Marcoulides GA. Fundamentals and Models of Item Response Theory. 4 Introduction to Psychometric Theory 2011:269-304. 5 47. Sunderland M, Mahoney A, Andrews G. Investigating the Factor Structure of the Kessler 6 Psychological Distress Scale in Community and Clinical Samples of the Australian 7 Population. J Psychopathol Behav 2012;34(2):253-59. 8 9 48. Beck AT, Erbaugh J, Ward CH, et al. An Inventory for Measuring Depression. Arch Gen 10 Psychiat 1961;4(6):561-&. 11 49. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiat 12 Scand 1983;67(6):361-70. 13 50. Parkitny L, McAuley J. The Depression Anxiety Stress Scale (DASS). J Physiother 14 2010;56(3):204-04. 15 51. Kessler RC,For Andrews peer G, Colpe LJ, etreview al. Short screening scales only to monitor population 16 prevalences and trends in non-specific psychological distress. Psychol Med 17 2002;32(6):959-76. 18 19 52. Kessler RC, Barker PR, Colpe LJ, et al. Screening for serious mental illness in the general 20 population. Arch Gen Psychiat 2003;60(2):184-89. 21 53. Andrews G, Slade T. Interpreting scores on the Kessler Psychological Distress Scale 22 (K10). Aust Nz J Publ Heal 2001;25(6):494-97. 23 54. Cairney J, Veldhuizen S, Wade TJ, et al. Evaluation of 2 measures of psychological 24 distress as screeners for depression in the general population. Can J Psychiat 25 2007;52(2):111-20. 26 55. Browne MAO, Wells JE, Scott KM, et al. The Kessler Psychological Distress Scale in Te 27 Rau Hinengaro: the New Zealand Mental Health Survey. Aust Nz J Psychiat 28 29 2010;44(4):314-22. 30 56. Furukawa TA, Kawakami N, Saitoh M, et al. The performance of the Japanese version of 31 the K6 and K10 in the World Mental Health Survey Japan. Int J Meth Psych Res

32 2008;17(3):152-58. http://bmjopen.bmj.com/ 33 57. Adult literacy Rate in Bangladesh. 2015. 34 58. Saxena S, Carlson D, Billington R, et al. The WHO quality of life assessment instrument 35 (WHOQOL-Bref): The importance of its items for cross-cultural research. Qual Life 36 Res 2001;10(8):711-21. 37 59. Bangladesh Bureau of Statistics. Population and Housing Census., 2010. 38 39 60. Hossain MD, Ahmed HU, Chowdhury WA, et al. Mental disorders in Bangladesh: a systematic review. BMC Psychiatry 2014;14:216. 40 on September 26, 2021 by guest. Protected copyright. 41 61. Reavley NJ, Jorm AF. National Survey of Mental Health Literacy and Stigma. 42 Department of Health and Ageing, Canberra 2011. 43 62. Harper A, Power M, Grp W. Development of the World Health Organization WHOQOL- 44 BREF quality of life assessment. Psychol Med 1998;28(3):551-58. 45 63. Brislin RW. Back-Translation for Cross-Cultural Research. J Cross Cult Psychol 46 1970;1(3):185-216. 47 64. Permission to use WHOQOL-100 and/or WHOQOL-BREF questionnaires 2015. 48 49 65. Cheng YH, Chi I, Boey KW, et al. Self-rated economic condition and the health of 50 elderly persons in Hong Kong. Soc Sci Med 2002;55(8):1415-24. 51 66. Serraglio A, Carson N, Ansari Z. Comparison of health estimates between Victorian 52 population health surveys and National Health Surveys. Aust Nz J Publ Heal 53 2003;27(6):645-48. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 15 of 21 BMJ Open

Page 15 of 17 1 2 3 Table 1: Variables will be collected for the study. 4 5 6 Variables Measures 7 8 Identification 9 10 11 Date of Birth 12 13 Date of data collection 14 15 Serial numberFor peer review only 16 17 Participants name 18 19 20 Mobile number (own) 21 22 Mobile number (relative) Mobile number (next of kin whom he/she can be 23 24 contacted) 25 26 Father or Husbands name Fathers name for males or unmarried males or females; 27 28 29 husband’s name for married or widowed women 30 31 Household Household such as Fakir Bari

32 http://bmjopen.bmj.com/ 33 Village Village 34 35 Union/Pourashava Union/Pourashava 36 37 38 Data collector’s signature 39

40 2.Socio-demographic on September 26, 2021 by guest. Protected copyright. 41 42 variables 43 44 2.1 Age Age in years (18-90 years) 45 46 47 2.2 Sex Sex (Female or Male) 48 49 2.3 Educational Qualification Education (No schooling, 1-5 years (primary), 6-10 50 51 (below school secondary certificate (SSC) pass), SSC or 52 53 higher secondary certificate (HSC) pass, and Bachelor of 54 55 56 above 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 16 of 21

Page 16 of 17 1 2 3 2.4 Marital status Marital (married (living with husband/wife), widowed, 4 5 never married, divorced or separate) 6 7 8 2.5 Occupation Occupation (Student, housewives, land owner, cultivates 9 10 other’s land, business, Govt. or non-govt. job, daily 11 12 labour, unemployed, retired, unable to work for disability) 13 14 2.6 Socio-economic status SES (insufficient money all the time, insufficient funds 15 For peer review only 16 some of the time, balanced (not good or bad), Sufficient 17 18 19 funds most of the time) 20 21 2.7 whom do you live with Live with (with own family, with relatives, with non- 22 23 relatives such as neighbour, live alone) 24 25 2.8 Income source Income (own income, help from family, help from 26 27 28 children, other relatives, children and relatives, 29 30 government help, cannot work and do not receive help 31

32 from children or government) http://bmjopen.bmj.com/ 33 34 2.9 Smoking status Never smoker, past smoker, current smoker, smokeless 35 36 37 tobacco only’, smoking and SLT 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 KAP Questionnaire: 42 43 A questionnaire will be adopted to collect data on participant’s awareness and KAP 44 regarding mental health, and data relevant to socio-demographic characteristics. Questions 45 46 evaluating knowledge, attitude and practice of mental health were associated with categorical 47 responses such as ‘‘yes’’ or ‘‘no’’. There are seven items related to mental health disorders 48 (depression, anxiety, psychosis, drug addiction, dementia, bipolar disorder and Alzheimer’s 49 disease) which will be included in the questionnaire. Each of the mental health indicators 50 consists of seven questions. For example, 51 52 1. Have you ever heard of depression? Yes=1, No=0 53 54 2. If yes, please list at least one symptom of depression. Answer was reported as “at least 55 one=1” and “none=0”. 56 57 3. Attitude towards mental health: If you or any of your friends or relatives have had any 58 mental health problems, do they seek treatment? Possible responses were, Yes=1, No=0. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 17 of 21 BMJ Open

Page 17 of 17 1 2 3 3.1 If the response was “Yes”, the question was “where would they seek treatment?” 4 with a possible response of “Medical doctor=1”, “Counselling=2” or “others, such as seek 5 advice from spiritual person=3” 6 7 4. The participants were asked whether they had any mental health disorders by asking “Do 8 you suffer from depression?” with possible answers of “Yes=1” and “No=0”. 9 10 4.1 If the response was “Yes”, the question was “Do you take treatment?” with a 11 possible response of “Yes=1” or “No=0”. 12 13 4.1.1 If the response was “Yes”, the next question was “Where did you go for treatment?” 14 with a possible response of “Medical doctor=1”, “psychological counselling=2” or “others, 15 such as seek Foradvice from peer spiritual person=3” review only 16 17 The attitude will be consider positive towards the treatment of mental health if they had a 18 positive response. Questions will evaluated the practice of mental health management 19 20 particularly for those who will be diagnosed with mental health, do they take treatment? If 21 yes, where did they go for treatment? 22 51 23 K10 Questionnaire 24 62 25 WHOQOL-BREF Questionnaire 26 27 28 29 30 31

32 http://bmjopen.bmj.com/ 33 34 35 36 37 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 18 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

32 http://bmjopen.bmj.com/ 33 34 35 36 37 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 Location of the Study Area

48 168x233mm (300 x 300 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 19 of 21 BMJ Open

Consent Form: 1 2 3 4 Interviewer note: Should be completed from one of the eligible members of aged ≥ 18 years 5 6 7 8 9 10 11 12 Patient’s consent 13 14 We are screening to identify people with depression and its risk factors, and to measure their quality of 15 life in people ofFor aged ≥ 18 years.peer The survey review will increase your awarenessonly of the disease outcome and 16 17 it’s risk factors. The community will be benefited from this study through the intervention programs 18 those the Organisation for Rural Community Development intend to conduct in the future. We do not 19 expect any risk for you if you participate in this study. Upon the completion, the results will be 20 published but your individual information will be kept confidential and your identification will not be 21 disclosed. We expect to continue our study for a longer period for which we may invite you again to 22 participate in our study. However, you are free to change your mind and can withdraw from the study 23 anytime without any obligation if you want. 24 25 26 27 28 Please provide your signature or thumb imprint if you agree 29 30 31 32 http://bmjopen.bmj.com/ 33 34 Signature/thumb imprint 35 36 37 38 39

40 Signature by the interviewer if the participant cannot provide signature. on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 20 of 21

1 2 STROBE Statement—checklist of items that should be included in reports of observational studies 3 4 5 Item No Recommendation 6 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 8 [Within the title page 1 and method and analysis section of the abstract page 2 ] 9 (b) Provide in the abstract an informative and balanced summary of what was done 10 and what was found [NA ] 11 12 Introduction 13 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 14 [ pages 3-4 ] 15 Objectives For3 peer State specific objectives, review including any prespecified only hypotheses [page 5 ] 16 17 Methods 18 Study design 4 Present key elements of study design early in the paper [ Methods page 5 ] 19 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 20 21 exposure, follow-up, and data collection [ pages 5-7 and page 9; Figure 1 ] 22 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of 23 selection of participants. Describe methods of follow-up [ pages 5-7 and page 9 ] 24 Case-control study—Give the eligibility criteria, and the sources and methods of 25 26 case ascertainment and control selection. Give the rationale for the choice of cases 27 and controls 28 Cross-sectional study—Give the eligibility criteria, and the sources and methods of 29 selection of participants 30 (b) Cohort study—For matched studies, give matching criteria and number of 31

32 exposed and unexposed http://bmjopen.bmj.com/ 33 Case-control study—For matched studies, give matching criteria and the number of 34 controls per case 35 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect 36 37 modifiers. Give diagnostic criteria, if applicable [ pages 7-8] 38 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 39 measurement assessment (measurement). Describe comparability of assessment methods if there is

40 more than one group [pages 15-17: table 1] on September 26, 2021 by guest. Protected copyright. 41 Bias 9 Describe any efforts to address potential sources of bias [ pages 6; Recruitment 42 43 strategy ] 44 Study size 10 Explain how the study size was arrived at [ pages 5; Sample size and statistical 45 power ] 46 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, 47 48 describe which groupings were chosen and why [ NA ] 49 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding [ 50 pages 8-9; planned statistical analysis ] 51 (b) Describe any methods used to examine subgroups and interactions [ NA ] 52 53 (c) Explain how missing data were addressed [ NA ] 54 (d) Cohort study—If applicable, explain how loss to follow-up was addressed Case- 55 control study—If applicable, explain how matching of cases and controls was 56 addressed 57 Cross-sectional study—If applicable, describe analytical methods taking account of 58 59 sampling strategy 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 21 of 21 BMJ Open

1 2 (e) Describe any sensitivity analyses 3 Continued on next page 4 5 Results 6 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, 7 examined for eligibility, confirmed eligible, included in the study, completing follow-up, and 8 analysed [ pages 5-6 and page 9 ] 9 (b) Give reasons for non-participation at each stage [ N/A ] 10 11 (c) Consider use of a flow diagram [ N/A ] 12 Descriptive 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information 13 data on exposures and potential confounders [ N/A now, we will get the information after data 14 collection] 15 For peer review only 16 (b) Indicate number of participants with missing data for each variable of interest 17 (c) Cohort study—Summarise follow-up time (eg, average and total amount) 18 Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 19 Case-control study—Report numbers in each exposure category, or summary measures of 20 21 exposure 22 Cross-sectional study—Report numbers of outcome events or summary measures 23 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their 24 precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and 25 why they were included [ N/A now, we will get the information after data collection] ] 26 27 (b) Report category boundaries when continuous variables were categorized [ N/A ] 28 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful 29 time period [ N/A ] 30 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 31

32 analyses [ N/A ] http://bmjopen.bmj.com/ 33 Discussion 34 Key results 18 Summarise key results with reference to study objectives [ page 9 ] 35 36 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. 37 Discuss both direction and magnitude of any potential bias [ page 10; paragraph 2 ] 38 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity 39 of analyses, results from similar studies, and other relevant evidence [ page 9-10 ]

40 on September 26, 2021 by guest. Protected copyright. 41 Generalisability 21 Discuss the generalisability (external validity) of the study results [pages 9 ; follow up data 42 collection ] 43 Other information 44 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, 45 46 for the original study on which the present article is based [ page 11 ] 47 48 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and 49 unexposed groups in cohort and cross-sectional studies. 50

51 52 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 53 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 54 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 55 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 56 57 available at www.strobe-statement.org. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open

Psychological distress and quality of life: Rationale and protocol of a prospective cohort study in a rural district in Bangladesh

For peer review only Journal: BMJ Open

Manuscript ID bmjopen-2017-016745.R1

Article Type: Protocol

Date Submitted by the Author: 26-May-2017

Complete List of Authors: Uddin, Mohammed ; Swinburne University of Technology, Department of Statistics, Data Science and Epidemiology Bhar, Sunil ; Swinburne University of Technology, Department of Psychological Sciences Al-Mahmud , Abdullah ; Swinburne University of Technology, Centre for Design Innovation (CDI); School of Design; Faculty of Health, Arts and Design Islam, Fakir; Swinburne University of Technology, Deapartment of Statistics, Data Science and Epidemiology

Primary Subject

Public health http://bmjopen.bmj.com/ Heading:

Secondary Subject Heading: Mental health

Psychological distress, Quality of Life, K1O, WHOQOL-BREF, Rasch Keywords: Analysis, Mental disorder

on September 26, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 1 of 22 BMJ Open

Page 1 of 18 1 2 3 Psychological distress and quality of life: Rationale and 4 5 6 7 protocol of a prospective cohort study in a rural district in 8 9 10 Bangladesh 11 12 13 14 15 M Nazim UddinFor1, Sunil Bharpeer2, Abdullah reviewAl Mahmud3, Fakir M only Amirul Islam1,4* 16 17 18 1. Department of Statistics, Data Science and Epidemiology; Faculty of Health, Arts and 19 20 Design; Swinburne University of Technology, Hawthorn VIC 3122, Australia 21 22 23 2. Department of Psychological Sciences; Faculty of Health, Arts and Design; Swinburne 24 25 University of Technology, Hawthorn VIC 3122, Australia 26 27 3. Centre for Design Innovation (CDI);School of Design; Faculty of Health, Arts and 28 29 Design; Swinburne University of Technology, Hawthorn VIC 3122, Australia 30 31 4. Organisation for Rural Community Development (ORCD), Dariapur, Narail, Bangladesh 32 http://bmjopen.bmj.com/ 33 34 35 36 37 Email: M Nazim Uddin – [email protected]; Sunil Bhar - [email protected]; Abdullah 38 39 Al Mahmud – [email protected]; Fakir M Amirul Islam* - [email protected]

40 on September 26, 2021 by guest. Protected copyright. 41 42 Corresponding author: 43 44 Email: [email protected] 45 46 47 48 49

50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 2 of 22

Page 2 of 18 1 2 3 ABSTRACT 4 Introduction: A significant proportion of the global burden of disease has been attributed to mental 5 and behavioural disorders. People with mental disorders (MD) have lower levels of health-related 6 quality of life than those without MD. Several studies have shown that in low-resource countries, a 7 range of social determinants including poor health literacy are critical in the epidemiologic transition 8 9 of disease outcome. There is a lack of evidence of mental disorder literacy, the prevalence and risk 10 factors of common mental health conditions, or any validated instruments to measure psychological 11 distress or evaluate the quality of life in rural areas of Bangladesh. 12 13 Aims of the study are to (1) Report the awareness, knowledge, attitudes and practice (KAP) of MD, 14 (2) Estimate the prevalence of and risk factors for psychological distress, (3) Measuring association 15 of psychologicalFor distress andpeer other socio review demographic factors withonly quality of life, (4) Test the 16 feasibility to use of K10 and WHOQOL-BREF questionnaire in rural Bangladesh for measuring 17 psychological distress and quality of life. 18 19 Methods and Analysis: A sample of 1500 adults aged 18-59 years and 1200 older adults aged 60 to 20 90 years will be interviewed from a multi-stage cluster random sample. Each participant will go 21 through a face-to-face interview to assess their awareness and KAP of MD., Information about the 22 participant’s demographic and socio-economic status will be collected along with psychological 23 distress (K10) and quality of life (WHOQOL-BREF) questionnaire. Internal consistency, validity, 24 25 reliability, and item discrimination of K10 and WHOQOL-BREF instruments will be determined by 26 using Rasch analysis, and regression techniques. 27 28 Ethics and Dissemination: Human Ethics Approval was received from the Swinburne University of 29 Technology Human Ethics Committee. Results of this research will be disseminated via scientific 30 forums including peer-reviewed publications and presentations at national and international 31 conferences.

32 http://bmjopen.bmj.com/ 33 Strengths and limitations of this study 34 35 36 • This study will be conducted using a large and representative sample from a typical rural 37 district to report the awareness, knowledge, attitudes and practice about mental health 38 disorders. Therefore, the study will provide a unique opportunity to assess the gaps in public 39 awareness of mental health disorders in rural Bangladesh.

40 • Data will be collected through a face-to-face interview process which will reduce the possible on September 26, 2021 by guest. Protected copyright. 41 response bias given the literacy level is different in males and females and in different age 42 groups. 43 • The study will be able to estimate the prevalence of the factors associated with psychological 44 distress to conduct appropriate intervention programs. 45 46 • The study will use a sophisticated Rasch analysis technique to validate the psychometric 47 properties of the WHOQOL-BREF and Kessler 10 items (K10) questionnaire for their use in 48 Bangladesh. The study will provide the first validated tools to measure psychological distress 49 and quality life in the rural areas of Bangladesh, which can be used in other developing 50 countries with the similar socio-economic condition. 51 • This study may suffer from the external validity as it is based on one area of rural 52 Bangladesh. The study could be repeated using random sample of other remote areas to better 53 portrait the intensity of psychological distress and quality of life at the national level. 54 55 56 Keywords: Mental disorder, psychological distress, quality of life, validation, Rasch analysis, 57 WHOQOL-BREF, K10, Rural Bangladesh 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 3 of 22 BMJ Open

Page 3 of 18 1 2 3 Background 4 5 Mental Disorders: 6 Mental disorders (MD) are one of the leading contributors to the global burden of disease in 7 both high and low income countries, accounting for 13% of disease measured as disability 8 adjusted life years 1. Common types of MD include conditions such as major depressive 9 10 disorders, bipolar disorders, anxiety disorders, psychotic disorders, and substance use disorders 2. Of these, major depressive and anxiety disorders contribute to approximately 11 3 12 50% of the international disease burden . Such conditions are frequently comorbid with other 4-6 13 health and psychosocial problems and are among the most serious causes of morbidity . 14 Despite the adverse impacts of such mental health conditions, they receive little attention in 15 most of the lowFor and middle-income peer countries, review and hence, treatment only of such conditions is not 16 considered national health priority 7 8. Mental health complaints create a major public health 17 concern in Bangladesh, especially in rural areas 4 9 10. The prevalence of mental disorders in 18 such areas varies between 6.5% and 31% possibly due to the use of different protocol and 19 11 20 definitions of MD . 21 22 Factors associated with mental disorders: 23 24 A number of factors have been identified as increasing risk for MD. These factors include a 25 lack of access to, and utilisation of, mental health services, stigma about mental health 26 treatment, and poor physical health 12. MD are present to a greater extent amongst low 27 socioeconomic status (SES) groups 13, and research shows that financially poor countries 28 with poor health literacy are critical in the epidemiologic transition of disease outcome 14-18. 29 A lot of evidence supports that KAP studies are needed for greater awareness of prevention, 30 14-21 31 diagnosis, risk factor control, and disease management of psychological distress .

32 http://bmjopen.bmj.com/ 33 34 Mental health literacy has been defined as “knowledge and attitudes about MD which aid 22 35 their recognition, management, and/or prevention” . Knowledge and positive attitudes 36 towards treatment are fundamental for individuals seeking help for MD in a timely and 37 appropriate manner. In India, recent studies have reported limited knowledge of mental 38 health issues and high rates of negative attitudes towards seeking help responsible for mental 39 health problems 23 24, yet similar literature for the neighbouring region of Bangladesh is 40 lacking. on September 26, 2021 by guest. Protected copyright. 41 42 Quality of Life: 43 44 In recent years, beyond traditional health indicators such as mortality and morbidity, there 45 has been a greater focus on measuring health and quality of life (QOL) as an important 46 25 47 outcomes in clinical and interventional studies . QOL is used as an important outcome 26 48 measure for evaluating intervention effects, such as the effect of medicine . The World 49 Health Organisation (WHO) defines quality of life (QOL) as “individuals’ perception of their 50 position in life in the context of the culture in which they live and the value systems they 51 have in relation to their goals, expectations, standards and concerns” 27. 52 53 Quality of life is affected by a range of factors such as physical health, psychological state, 54 level of independence, social relationships, personal beliefs, and their relationships to silent 55 features of their environment 28. However, most of the research on QOL is focused on the 56 deleterious impact of chronic physical health diseases such as cancer, stroke, diabetes and 57 29-32 58 HIV/AIDS . Furthermore, over the past 20 years, although a quite a number of tools have 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 4 of 22

Page 4 of 18 1 2 3 been developed to measure QOL 33, with a few exceptions 34-36, these measure of QOL are 4 impacted by specific diseases. As an exception, the QOL Scale (QOLS) first developed by 5 American psychologist, John Flanagan 37 38, fits a more generalised definition of QOL that 6 can be used to assess QOL in an everyday context. The WHOQOL-BREF is a shortened 7 version of, 26-item instrument, rated on five-point Likert-type scales, is used for large 8 39 9 population studies . This tool has not been applied to a general population setting in any 10 developing countries, including Bangladesh. 11 A Bengali version of the WHOQOL-BREF was initially developed in 2005 for using in a 12 40 41 13 study of adolescents and adults, residing only in the capital city of Bangladesh, Dhaka . 14 The Bengali version of the scale has been used in some rural areas of the country since 2005 42 43 15 . The dimensionalityFor peer of the Bengali review version of WHOQOL-BREF only was assessed using 16 Classical Test Theory (CTT) 40 41. In CTT approach, the items and the person latent trait are 17 measured separately and descriptive measures are provided. In the CTT method, true score 18 estimates are typically obtained by summing the responses across items, with the assumption 19 that each item within the same construct is valued equally. A further assumption of CTT is 20 that the score difference between the two adjacent response scales is uniform although this 21 does not occur in all circumstances 44. Therefore, the CTT cannot be meaningfully and 22 45 46 23 methodically compared in an item-person continuum . These restrictions can be resolved 24 using Item Response Theory (IRT) modelling (Rasch Analysis) despite IRT’s own 45 46 25 assumptions regarding unidimensionality, invariability and local independence . 26 27 Psychological distress: 28 29 In recent years, epidemiological studies have attempted to use less items to proficiently 30 measure and monitor the extent of psychological distress in the widespread community 47. 31

For the purpose of measuring depression and anxiety few scales have been used so far. These http://bmjopen.bmj.com/ 32 48 33 include the Beck Depression Inventory (BDI) , the Hospital Anxiety and Depression Scale 49 50 51 34 (HADS) , the Depression Anxiety Stress Scales (DASS) and the Kessler K10 . The K10 35 is one of the most popular tools for screening psychological distress in the general population 36 52. K10 comprises 10 items, rated on five-point Likert-type scales, which indicate the degree 37 of psychological distress prevalent among persons 51. The K10 has been established 38 predominantly from Western population samples 51-56. Information on cross-cultural validity 39 of the K10 is not available in Bangladesh. Therefore, we have used an interview-administered

40 on September 26, 2021 by guest. Protected copyright. version of K10 in our study due to the low rate of literacy in Bangladesh 57. 41 42 43 Reliability and validity of K10 and WHOQOL questionnaires: 44 45 Reliability and validity are context-specific attributes, showing applicability to some 46 populations over others. As quality of life and physiological distress assessment instruments 47 are used in a variety of cultural settings, it is important to establish whether the same aspects 48 of life are equally important for adults across the lifespan, males and females, people with 49 and without education and for people with various relationship and employment status. The 50 above factors report to impact differently on people’s lives, but how significant these 51 differences are, and the issue of whether a reasonably valid tool can be developed in spite of 52 58 53 these differences needs to be addressed in Bangladesh . Therefore, the validation of 54 WHOQOL questionnaire as well as K10 questionnaire in a general population, both in adults 55 and older adults can provide empirical development of literature on these issues in 56 Bangladesh. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 5 of 22 BMJ Open

Page 5 of 18 1 2 3 4 5 Aims of the study 6 7 The current prospective cohort study will be conducted in a rural district of Bangladesh to 8 9 address a number of research questions and the specific aims are as follows: 10 I. Report the awareness, knowledge, attitudes and practice (KAP) using seven items 11 12 from the mental health literacy questionnaire of MD. 13 II. Estimate the prevalence of psychological distress in both adults and older adults 14 according to socio-demographic status. 15 III. EstimateFor the quality peer of life in review adults and older adults only, their risk factors and the 16 variation according to the socio-demographic factors. 17 IV. Measure the association between psychological distress with quality of life. 18 V. To test the internal validity of the K10 scale using Rasch analysis 19 VI. To investigate the factor structure of K10 scale using Confirmatory Factor Analysis 20 (CFA) and Exploratory Factor Analysis (EFA) 21 22 VII. To test the internal validity of the four domain WHOQOL-BREF questionnaire using 23 Rasch analysis. 24 VIII. To assess the factor structure of the WHOQOL-BREF in a Bangladeshi rural general 25 population sample CFA and EFA 26 27 28 Methods 29 30 Study Population: 31 Bangladesh is a country of 160 million people divided into 64 districts. Each district is

32 divided into sub-districts named Upazilas (493 in total), and each Upazila further divided into http://bmjopen.bmj.com/ 33 several Unions and the city centre known as pourashava. Each Union consists of 15-25 34 villages and the pourashava consists of 5-15 wards which are divided into localities (also 35 known as ‘para’ in Bengali) that are comprised of several households. Participants will be 36 recruited from the Narail district (Figure 1). Narail is located approximately 200 km 37 38 southwest of Dhaka the capital city of Bangladesh. The population of Narail Upazila is 39 272,872, with approximately 40% of residents aged between 18-59 years and 19,000 (about

40 7%) of residents aged between 60-90 years. The study location was selected as it was on September 26, 2021 by guest. Protected copyright. 41 considered to be representative of a typical rural demographic in Bangladesh. Area of Narial 42 Upazila is 381.76 square km, located in between 23°02' and 23°17' north latitudes and in 43 between 89°23' and 89°37' east longitudes. It is bounded by Lohagara and Salikha Upazilas 44 on the north, Kalia and Kbhaynagar Upazilas on the south, Lohagara Upazila on the east, 45 Bagherpara and Sadar Upazilas on the west. The district consists of 13 unions, 231 46 villages, 1 pourashava, 9 wards. The total population of this Upazial is 273000 with 49% 47 48 male. Literacy of this Upazila for male is 63.3% and female is 59.3%. 73% people are 49 Muslim and approximately 27% are . Currently there is one 100-bed government 50 hospital and 11 private clinics consist of 115-bed facilities. Only 36 medical doctors are 51 working for the whole Upazilla and no specialized hospitals or clinic 59. The Narail Upazila 52 an estimated population density of 722 people per km2 is comparable to the national rural 53 population density of 873 people per km2. It is not at the extremity of remote locations nor is 54 it a catchment of a metropolis such as Dhaka 60. 55 56 Sample size and statistical power 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 6 of 22

Page 6 of 18 1 2 3 Prior data indicate that the prevalence of severe depression was 21% in older adults aged 60 4 or above, and 6.5% in adults aged between 18 and 60 61. We assumed a margin of error of 5% 5 in prevalence rates for older adults, and of 3% in adults when estimating the true prevalence 6 of severe depression for each cohort in this rural area. Using a significance level of 0.05 and 7 statistical power above 80%, a required sample size of 1128 is needed for the older adults and 8 9 1283 for the adults, respectively. We did not adjust for cluster effects because only a single 10 participant will be recruited from each household. However, assuming non-response and 11 other possible problems such as; irrelevant or duplicate data, erroneous or misinterpreted data 12 and too little data acquired from the client in data collection, the sample size was adjusted. 13 Hence, The recruitment target will be increased to 1200 older adults aged between 60-90 14 years and 1500 adults aged between 18-59 years. 15 For peer review only 16 Sampling Frame 17 18 A multi-stage cluster random sampling technique was used for this prospective cohort study. 19 Three unions from a total of 13 and one ward from a total of 9 located within Narail District 20 were randomly selected at level 1. Two to three villages or mahalla from each selected union 21 or ward were randomly selected at the second level. The Unions and Ward are shown in 22 Figure 1. The selected villages from different Unions were as follows: Banshgram union: 23 Tabra, Betvita and Ballertop, Bhadrabilla union: Ramshidi, Bogura and Palaidanga, 24 25 Tularampur Union: Betenga, Mitna and Bamanhat, Ward 1: Kurigram and Maijpara. The 26 projected recruitment target was set at 120 older adults and 150 adults from each village. 27 Each village or ward comprised between 360 to 480 households; and one eligible adult or 28 older adult will be interviewed from each household. 29 30 Recruitment strategy 31

32 First, the chief investigator and the head of the Department of Statistics, Data Science and http://bmjopen.bmj.com/ 33 Epidemiology at the Swinburne University of Technology, Australia have sent a letter to the 34 district commissioner, district civil surgeon, police super, and chairman of the union parishad 35 seeking their cooperation. Second, interviewers will be provide the village leaders and/or 36 school teachers with a general overview of the research and will request their cooperation. 37 Once received the consent by the village leaders, the interviewer will approach the 38 prospective participants to explain the purpose of the study and to invite them to participate 39 in the study.

40 on September 26, 2021 by guest. Protected copyright. 41 The interviewers will be instructed to interview an older adult first. If none are available in 42 43 this subgroup, the interviewer will approach an adult person of that household. Again, if there 44 will be more than one male or female adult in the same household, one individual will be 45 selected, based on who was born closer to the month of January. However, to maintain an 46 approximately equal number of males and female participants, one female will be interview 47 immediately after an interview of a male participant. Selected participants of age of less than 48 18 years or more than 90 years, and having an illness will be excluded from the study. 49 50 Quality assurance 51 52 All team members will be participated in an intensive two-day training program in Narail 53 before the will commencement of the survey. The purpose of the training is to outline the 54 rationale of the study, and the procedures and potential difficulties associated with data 55 collection. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 7 of 22 BMJ Open

Page 7 of 18 1 2 3 Data Collection 4 5 Questionnaires 6 7 An interviewer-administered semi-structured questionnaire is being developed to collect 8 relevant socio-demographic data and to assess participant’s awareness, knowledge, attitudes 9 and practice regarding MD. Data will be collected through a face-to-face interview process, 10 which will allow a greater possibility of collection of more truthful answers. Questions 11 regarding income, financial support and consumption of tobacco will also be collected. The 12 questionnaire is shown in Table 1. 13 14 Awareness, Knowledge, Attitude and Practice about mental disorders 15 For peer review only 16 Given the relative lack of validated mental health KAP data, the specific items included in the 17 final questionnaire were derived from the National Survey of Mental Health Literacy and 18 Stigma by Reavley and Jorm (2011) 62. The questionnaire will be used to assess the 19 awareness of seven common MD (depressive disorders, bipolar disorders, anxiety disorders, 20 psychotic disorders, and substance use disorders) by asking if they have ever heard of these 21 diseases with a possible response of “yes” or “no”. Based on their responses, they will be 22 further requested to list at least one symptom of the disease to provide insight into the level of 23 24 knowledge they possessed. Attitudes towards the use of treatment will be assessed by asking 25 the question “do they need treatment?” with a possible response of “yes” or “no”. The 26 practice about MD will be evaluated based on whether the participant or any of his/her 27 relatives had ever experienced any mental health problem, and from whom they would seek 28 treatment, with a possible response of “medical doctor”, “psychologist” or others such as 29 “spiritual persons”. The detail questions about KAP are reported in different section after 30 Table 1. 31 32 http://bmjopen.bmj.com/ 33 Kessler Psychological Distress scale (K10) and WHOQOL-BREF questionnaires: 34 35 The K10 51 was used to measure psychological distress and World Health Organization 36 Quality of Life (WHOQOL-BREF) 63 to assess the quality of life of the participants. The K10 37 was translated into Bengali independently by two bilingual translators including a medical 38 39 practitioner with experience in public health. The K10 questionnaire was translated using back-translated techniques 64. Regarding the WHOQOL questionnaire, the chief investigator 40 on September 26, 2021 by guest. Protected copyright. 41 (AI) contacted the original developers of the WHOQOL-BREF quality of life assessment test 65 42 team to seek permission to use the Bengali version of the WHOQOL-BREF for research 43 purposes in Bangladesh. This version was also translated to accommodate Narail languages. 44 Validity of such translations were checked through the back-translation procedure 64. 45 46 47 Planned Data Collection: 48 49 Location: Narail Upazilla (Figure 1). Duration of data collection: May-July, 2017 50 51 Outcome Variables: The first outcome variable will be the prevalence of psychological 52 distress which will be defined based on the sum of a five value response option for each of 10 53 Kessler questionnaire items, with the maximum score of 50 and the minimum score of 10, 54 then categorised according to the score obtained; no psychological distress (10 to 19), mild 55 (20 to 24), moderate (25 to 29) and severe psychological distress (30 to 50). 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 8 of 22

Page 8 of 18 1 2 3 The second outcome variables will be the prevalence of awareness of MDs comprising of 4 seven common mental disorders: depression, anxiety, psychosis, drug addiction, dementia, 5 bipolar disorder and Alzheimer’s disease, provided the participants have ever heard about the 6 disorders. Prevalence of knowledge will be measured based on whether the participants are 7 able to identify at least one symptom of the MDs and are aware of the disorders. Attitudes 8 9 will be deemed positive if the participants who are aware of the MDs are in favour of taking 10 medical treatment or psychological counselling, and practice if participants with MDs already 11 take medical treatment or psychological counselling. 12 13 Validated K10 Questionnaire: the K10 comprises 10 questions that will be answered using a 14 five -point scale. We will use Rasch analysis to achieve adequate model fit for K10 that will 15 give internalFor consistency, peer reliability, unidimensionality review and freedomonly from differential item 16 function (DIF) for sex (male/ female) and age (adult and elderly). 17 18 Validated WHOQOL questionnaire: To assess the quality of life of rural people in 19 Bangladesh we will use the WHOQOL-BREF tool, which consists of 26 questions with 20 structured responses on a Likert scale of five points. Of the 26 questions, two assess the 21 perception of quality of life and health of the patient, and the others (24 questions) comprise 22 the physical, psychological, social and environment domains. Therefore, we will use Rasch 23 analysis, CFA and EFA for each domain to provide an optimal evaluation of the quality of 24 25 life of rural people. 26 27 Exposure variables: Demographic details for age, categorised as either adult (18 to 59 years) 28 or older adult (60 to 90 years), gender, level of education – categorised as no schooling, 29 primary school education (grade 1 to 5), secondary school education (grade 6 to 10) and 30 school secondary certificate (SSC) or above and socio-economic status (SES) will be 31 collected. SES will be assessed according to Cheng et al. 66 asking whether "over the last

32 twelve months, in terms of household food consumption, how would you classify your socio- http://bmjopen.bmj.com/ 33 economic status?" The possible answers are: (i) insufficient funds for the whole year; (ii) 34 insufficient funds some of the time; (iii) neither deficit nor surplus (balance); and (iv) 35 sufficient funds most of the time. Data on ongoing health conditions (yes or no), number of 36 37 health problems, use of medication and smoking status classified into three categories: 38 ‘Never smoking’, ‘Ever smoking’ and ‘Smokeless tobacco use’ will also be collected during 39 the interview.

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 Planned Statistical analysis: 44 45 Primary objective 1: Awareness, knowledge, attitude and practice of mental health 46 disorders and prevalence of psychological distress in rural area 47 48 Rasch analysis will be performed to compute the person measures based on the awareness of 49 seven common MHD in a logarithmic scale and termed as ‘‘awareness score”. Differentials 50 of the measures of awareness, knowledge, attitudes and practice towards MD will be 51 evaluated across different major sub-domains of age, gender, level of education and other 52 socio-economic factors using Pearson Chi-square tests of independence. Multiple regression 53 analysis techniques will be used to investigate the factors contributing to the combined score 54 of awareness about the underlying MD. Data pertaining to the awareness about MD will be 55 collected using KAP questionnaire and analysed using RUMM2030 and SPSS 23 software. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 9 of 22 BMJ Open

Page 9 of 18 1 2 3 An estimate of the prevalence of psychological distress will also be calculated (using 4 standard cut-off values of combined K10 score, according to Victorian Population Health 5 Survey) 67. Factors associated with psychological distress will be investigated using Chi- 6 square test and binary or multi-nominal logistic regression techniques. 7 8 Finally, the association between the quality of life and potential contributing factors including 9 awareness, knowledge, and attitudes towards mental disorder and psychological distress will 10 also be assessed using regression techniques. 11 12 13 14 Primary objective 2: Evaluate and Validate the Kessler K10 15 For peer review only 16 Rasch analysis will be used to find a possible unidimensional structure of K10. Moreover, we 17 will use the Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) to 18 explore the factor structure of the K10 questionnaire. 19 20 Primary objective 3: Evaluate and Validate the WHOQOL-BREF 21 22 Confirmatory Factor Analysis (CFA) will be applied to the original four domains (physical, 23 psychological, environmental, and social), assuming the orthogonally between factors and 24 invariance of the structure across items. Rasch analysis will be used to check uni- 25 26 dimensionality, local independence, and invariability of the original four domains. If needed, 27 we will take necessary adjustments to improve the model fit. Finally, the proposed model 28 will be tested to get possible psychometric structural model for the WHOQOL-BREF in the 29 rural area in Bangladesh. 30 31

32 http://bmjopen.bmj.com/ 33 Follow-up data collection: 34 35 Subjects between 18 to 90 years of age are deemed to be the appropriate target population. In 36 Bangladesh, there is no prior longitudinal follow-up study on physiological distress and 37 quality of life. Therefore, we have used alternative approach to measure physiological 38 distress using K10 questionnaire as a proxy of depression or anxiety. Prior data indicate that 39 the prevalence of severe depression was 21% in older adults of age 60 or above, and 6.5% in

40 61 on September 26, 2021 by guest. Protected copyright. 41 adults aged between 18 and 59 . Based on the prior data, we expect 2250 participants will 42 be free from severe psychological distress at the baseline. There is no prior information to 43 predict the incidence of psychological distress per year. However, we would expect the 44 incidence rate of 3-4% per year, indicating about 100 new cases per year. The follow-up data 45 collection will be conducted in every two years from which the expected number of incidence 46 will be sufficient to study the factors associated with incidence of psychological distress. 47 48 Main baseline and follow-up studies are as follows: Baseline: May to July 2017 (2500 49 samples), second follow-up: May- July 2019 (2125 participants, 85% participation) and third 50 51 follow-up: May to July 2021 (1800 participants, 85% participation). The eligible criterion 52 will be followed from baseline. Although we acknowledge the importance of aiming for 53 maximum follow-up in any study. In practice, it is inevitable that losses to follow-up will 54 occur. 55 56 At the baseline, some procedure will follow to minimise the loss of follow-up. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 10 of 22

Page 10 of 18 1 2 3 1. Collect baseline information, e.g., addresses, phone numbers, but also for possible contacts 4 such as next of kin or close friends. 5 6 2. When feasible, we try to involve village leaders or Union Chairmans because they are 7 more likely to remain in that area, which makes it easier to track them down if they relocate. 8 9 10 3. Send participants newsletters periodically to keep them updated on the study' progress. 11 12 13 14 15 For peer review only 16 Discussion 17 18 A questionnaire that demonstrates reliability, validity, and responsiveness in a certain context 19 may not necessarily be appropriate for other contexts. The validation of the Bengali version 20 of the K10 as well as the WHOQOL-BREF questionnaires will enable them to be used in a 21 broader context in which similar circumstances arise. The K10 has not been used previously 22 in Bangladesh. Previously, the Bengali version of WHOQOL-BREE was used in specific 23 40 41 42 43 24 population such as those in slum areas and on disease prone individuals , which 25 limited the validity, reliability and the general utility of the questionnaire. However, they 26 failed to include people of middle income, general rural people, and elderly people. The 27 properties of the test may have changed since it was used to assess people that were more 28 representative of the national socio-economic condition during a specific period. In the 29 present study, the inclusion of rural people aged between 18-59 and 60-90 enhances the 30 usefulness of the questionnaire. The validation of the Bengali versions of WHOQOL-BREF 31 and K10 are expected to enable generalisability of the scales and will allow comparisons to 32 http://bmjopen.bmj.com/ be made between developing countries as well as contrast between classical test theory and 33 34 item response theory. 35 36 The study will present a unique opportunity to assess the gaps in public awareness of mental 37 disorders in a rural district of Bangladesh. This information would be useful in the 38 formulation of policy for the awareness, training, management and service delivery in regards 39 to mental disorder issues in rural areas of Bangladesh and may be used as a model in low and

40 middle-income countries around the world. The purpose of the project is to create an on September 26, 2021 by guest. Protected copyright. 41 opportunity to work with a local non-government organization (the Organization for Rural 42 Community Development) and capture information that would be of practical benefit to 43 health policy planners in the Narail district and local health authorities. This information 44 could ultimately benefit the general population. Moreover, another benefit of the study will 45 46 be validating the psychometric properties of the WHOQOL-BREF and K1O questionnaire 47 using the traditional CTT, Rasch analysis and EFA. These will then enable the exploration of 48 different statistical methods in order to produce the best model for Bangladeshi rural 49 population. The study will be conducted upon a large sample of both adults and older adults 50 and utilised face-to-face data collection. A targeted intervention might be needed to increase 51 awareness of MD. Moreover, we will be able to determine the proportions’ of rural 52 populations who experience physiological distress and to explore most common socio- 53 economic factors that drive psychological distress. This will help us identify the potential 54 socio-economic factors that can be targeted initially. The study will explore how many items 55 56 of the K10 scale and WHOQOL-BREF supports the measurement properties, internal 57 consistency, reliability and uni-dimensionality. This investigation will guide us using Rasch 58 and Confirmatory Factor Analysis to find an appropriate modified questionnaire in terms of 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 11 of 22 BMJ Open

Page 11 of 18 1 2 3 acceptability, validity, reliability and responsiveness. No prior studies have been undertaken 4 using similar approach to assess the community knowledge of mental disorders in 5 Bangladesh. 6 7 8 Despite several benefits of this study, the methodology used suffers from some limitations. 9 The study will be conducted in one rural area of Bangladesh, which may not delineate the 10 portrait of the awareness of MDs and the level of psychological distress for all rural areas in 11 12 Bangladesh. Whilst we will attempt to capture the situation in Narail, the study would 13 obviously need to be repeated in a random sample of other remote areas in order to be truly 14 reflective of a national perspective. 15 For peer review only 16 17 This will be the first study to prompt a critical analysis of the psychometric properties of the 18 WHOQOL-BREF and K10 using Rasch analysis and CFA. The outcomes will deliver 19 measurement properties, internal validity, reliability and unidimensionality of the scale. The 20 findings will guide us to recommend further studies using validated questionnaires in clinical 21 assessments using the appropriately targeted samples. The findings will discuss, in relation to 22 the Bengali language and culture, social beliefs about mental disorders, and suggestions for 23 improvements in mental health literacy and stigma towards mental health. Furthermore, we 24 25 hope this study will enable the formation of an early intervention program, targeted at those 26 who are expected to diagnose with mental illness, such as mild or moderate levels of 27 psychological distress. This program, when developed, has the potential to reduce the 28 likelihood of secondary effects such as loss of employment, school dropout, relationship 29 collapse, disability and drug and alcohol problems. The study will also be able to suggest an 30 appropriate program for rural areas that require early intervention, which is important, as 31 delay in getting help and support for mental health problems increases the likelihood of 32 experiencing a difficult recovery. http://bmjopen.bmj.com/ 33 34 Author Affiliations 35 36 37 1. Department of Statistics, Data Science and Epidemiology; Faculty of Health, Arts and 38 Design; Swinburne University of Technology, Hawthorn VIC 3122, Australia 39 2. Department of Psychological Sciences; Faculty of Health, Arts and Design; Swinburne

40 University of Technology, Hawthorn VIC 3122, Australia on September 26, 2021 by guest. Protected copyright. 41 3. School of Design; Faculty of Health, Arts and Design; Swinburne University of 42 Technology, Hawthorn VIC 3122, Australia 43 4. Organisation for Rural Community Development (ORCD), Dariapur, Narail, Bangladesh 44 45 46 List of abbreviations 47 48 MD, Mental disorders; WHOQOL, World Health Organization Quality of Life; IRT, Item 49 response theory; CTT, Classical test theory; CFA, Confirmatory factor analysis; KAP, 50 Knowledge, attitudes and practice. 51 52 53 Acknowledgement: 54 55 56 We acknowledge Dr Jason Skues to provide us valuable feedback about the project. We 57 particularly acknowledge the contribution of Md Rafiqul Islam, Md Sajibul Islam, Saburan 58 Nesa and Arzan Hossain who will collect the data from the participant’s households. Finally, 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 12 of 22

Page 12 of 18 1 2 3 we would like to express our gratitude to the study participants for their voluntary 4 participation. 5 6 7 8 9 Author’s contribution: 10 11 MNU and FMAI jointly designed the study. MNU drafted the manuscripts. SB, AAM and 12 FMAI contributed in writing the manuscripts. FMAI supervised the overall preparation of the 13 manuscript. All authors read and approved the final manuscript. 14 15 Funding For peer review only 16 17 Data collection for this research project was funded by the Faculty of Health, Arts and Design 18 (FHAD) of the Swinburne University Technology under the Research and Development 19 Grant Scheme (RDGS). The funders had no role in the design of the study, data collection, 20 analysis, interpretation of data or writing the manuscript. 21 22 Competing interests: 23 24 The authors declare that they have no competing interests. 25 26 27 Ethical and Dissemination: 28 29 Human Ethics Approval was received from the Swinburne University of Technology Human 30 Ethics Committee in accordance with the tenets of the Declaration of Helsinki. Study 31 participants provided written consent in order to participate. We anticipate that the findings 32 http://bmjopen.bmj.com/ will be of interest Ministry of Health Bangladesh and other similar international organisations 33 34 who work in the rural area in Bangladesh. We also expect that the study will generate 35 recommendations in relation to practice. Results of this research will be disseminated via 36 scientific forums including peer-reviewed publications, community briefs and presentations 37 at national and international conferences. 38 39

40 References: on September 26, 2021 by guest. Protected copyright. 41 42 1. Tomlinson M, Swartz L, Daniels K. "No health without mental health" The global effort to 43 44 improve population mental health. Rout Stud Pub Hlth 2011:174-91. 45 2. Association AP. Diagnostic and statistical manual of mental disorders (DSM-5®): 46 American Psychiatric Pub, 2013. 47 3. Whiteford HA, Degenhardt L, Rehm J, et al. Global burden of disease attributable to 48 mental and substance use disorders: findings from the Global Burden of Disease 49 Study 2010. Lancet 2013;382(9904):1575-86. 50 4. Moussavi S, Chatterji S, Verdes E, et al. Depression, chronic diseases, and decrements in 51 health: results from the World Health Surveys. Lancet 2007;370(9590):851-58. 52 5. Patel V, Kleinman A. Poverty and common mental disorders in developing countries. B 53 54 World Health Organ 2003;81(8):609-15. 55 6. Lamers F, van Oppen P, Comijs HC, et al. Comorbidity Patterns of Anxiety and 56 Depressive Disorders in a Large Cohort Study: the Netherlands Study of Depression 57 and Anxiety (NESDA). J Clin Psychiat 2011;72(3):341-48. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 13 of 22 BMJ Open

Page 13 of 18 1 2 3 7. Bruckner TA, Scheffler RM, Shen G, et al. The mental health workforce gap in low- and 4 middle-income countries: a needs-based approach. B World Health Organ 5 2011;89(3):184-94. 6 8. Hock RS, Or F, Kolappa K, et al. A new resolution for global mental health. Lancet 7 2012;379(9824):1367-68. 8 9 9. Hosain GMM, Chatterjee N, Ara N, et al. Prevalence, pattern and determinants of mental 10 disorders in rural Bangladesh. Public Health 2007;121(1):18-24. 11 10. Islam MM, Ali M, Ferroni P, et al. Prevalence of psychiatric disorders in an urban 12 community in Bangladesh. Gen Hosp Psychiat 2003;25(5):353-57. 13 11. Hossain MD, Ahmed HU, Chowdhury WA, et al. Mental disorders in Bangladesh: a 14 systematic review. Bmc Psychiatry 2014;14. 15 12. KilkkinenFor A, Kao-Philpot peer A, O'Neil A,review et al. Prevalence of onlypsychological distress, anxiety 16 and depression in rural communities in Australia. Aust J Rural Health 17 2007;15(2):114-19. 18 19 13. World Health Organization. The world health report: 2001. Mental health: new 20 understanding, new hope. Geneva: WHO; 2001. 2001. 21 14. Demaio AR, Otgontuya D, de Courten M, et al. Exploring knowledge, attitudes and 22 practices related to diabetes in Mongolia: a national population-based survey. BMC 23 Public Health 2013;13:236. 24 15. Rani PK, Raman R, Subramani S, et al. Knowledge of diabetes and diabetic retinopathy 25 among rural populations in India, and the influence of knowledge of diabetic 26 retinopathy on attitude and practice. Rural Remote Health 2008;8(3):838. 27 16. Zaman MJ, Patel A, Jan S, et al. Socio-economic distribution of cardiovascular risk 28 29 factors and knowledge in rural India. International journal of epidemiology 30 2012;41(5):1302-14. 31 17. Islam FMA, Chakrabarti R, Islam SZ, et al. Factors Associated with Awareness, Attitudes

32 and Practices Regarding Common Eye Diseases in the General Population in a Rural http://bmjopen.bmj.com/ 33 District in Bangladesh: The Bangladesh Population-based Diabetes and Eye Study 34 (BPDES). Plos One 2015;10(7). 35 18. Islam FM, Chakrabarti R, Dirani M, et al. Knowledge, attitudes and practice of diabetes 36 in rural Bangladesh: the Bangladesh Population based Diabetes and Eye Study 37 (BPDES). PLoS One 2014;9(10):e110368. 38 39 19. Norris SL, Engelgau MM, Narayan KM. Effectiveness of self-management training in type 2 diabetes: a systematic review of randomized controlled trials. Diabetes Care 40 on September 26, 2021 by guest. Protected copyright. 41 2001;24(3):561-87. 42 20. Al-Maskari F, El-Sadig M, Al-Kaabi JM, et al. Knowledge, attitude and practices of 43 diabetic patients in the United Arab Emirates. PLoS One 2013;8(1):e52857. 44 21. Mazzuca SA, Moorman NH, Wheeler ML, et al. The diabetes education study: a 45 controlled trial of the effects of diabetes patient education. Diabetes Care 1986;9(1):1- 46 10. 47 22. Jorm AF, Korten AE, Jacomb PA, et al. ''Mental health literacy'': A survey of the public's 48 49 ability to recognise mental disorders and their beliefs about the effectiveness of 50 treatment. Med J Australia 1997;166(4):182-86. 51 23. Liu MC, Tirth S, Appasani R, et al. Knowledge and Attitudes Toward Depression Among 52 Community Members in Rural Gujarat, India. J Nerv Ment Dis 2014;202(11):813-21. 53 24. Almanzar S, Shah N, Vithalani S, et al. Knowledge of and Attitudes Toward Clinical 54 Depression Among Health Providers in Gujarat, India. Ann Glob Health 55 2014;80(2):89-95. 56 25. Kuyken W, Orley J. Development of the Whoqol - Rationale and Current Status. Int J 57 Ment Health 1994;23(3):24-56. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 14 of 22

Page 14 of 18 1 2 3 26. Chang KC, Wang JD, Tang HP, et al. Psychometric evaluation, using Rasch analysis, of 4 the WHOQOL-BREF in heroin-dependent people undergoing methadone 5 maintenance treatment: further item validation. Health Qual Life Out 2014;12. 6 27. Kuyken W, Orley J, Power M, et al. The World-Health-Organization Quality-of-Life 7 Assessment (Whoqol) - Position Paper from the World-Health-Organization. Soc Sci 8 9 Med 1995;41(10):1403-09. 10 28. WHOQOL-Measuring quality of life. The World Health Orgnaization Quality of Life 11 Instruments 1997. 12 29. Silva SM, Correa FI, Faria CDCD, et al. Psychometric properties of the stroke specific 13 quality of life scale for the assessment of participation in stroke survivors using the 14 rasch model: a preliminary study. J Phys Ther Sci 2015;27(2):389-92. 15 30. Debiec J,For Goch A, Chmielewski peer H, etreview al. Effect of diabetes onlyon neurological condition and 16 quality of life of patients with ischaemic stroke. Atherosclerosis 1999;144:192-92. 17 31. Jia HM, Zack MM, Thompson WW. The Effects of Diabetes, Hypertension, Asthma, 18 19 Heart Disease, and Stroke on Quality-Adjusted Life Expectancy. Value Health 20 2013;16(1):140-47. 21 32. Richardson J, Sinha K, Iezzi A, et al. Modelling utility weights for the Assessment of 22 Quality of Life (AQoL)-8D. Qual Life Res 2014;23(8):2395-404. 23 33. Berzon RA, Donnelly MA, Simpson RL, et al. Quality of life bibliography and indexes: 24 1994 update. Qual Life Res 1995;4(6):547-69. 25 34. Wahl AK, Rustoen T, Hanestad BR, et al. Quality of life in the general Norwegian 26 population, measured by the Quality of Life Scale (QOLS-N). Qual Life Res 27 2004;13(5):1001-09. 28 29 35. Ohaeri JU, Awadalla AW, El-Abassi AHM, et al. Confirmatory factor analytical study of 30 the WHOQOL-Bref: experience with Sudanese general population and psychiatric 31 samples. Bmc Med Res Methodol 2007;7.

32 36. Redko C, Rogers N, Bule L, et al. Development and validation of the Somali WHOQOL- http://bmjopen.bmj.com/ 33 BREF among refugees living in the USA. Qual Life Res 2015;24(6):1503-13. 34 37. Flanagan JC. Measurement of Quality of Life - Current State of the Art. Arch Phys Med 35 Rehab 1982;63(2):56-59. 36 38. Flanagan JC. Research Approach to Improving Our Quality of Life. Am Psychol 37 1978;33(2):138-47. 38 39 39. Development of the World Health Organization WHOQOL-BREF quality of life assessment. The WHOQOL Group. Psychol Med 1998;28(3):551-8. 40 on September 26, 2021 by guest. Protected copyright. 41 40. Izutsu T, Tsutsumi A, Islam M, et al. Validity and reliability of the Bangla version of 42 WHOQOL-BREF on an adolescent population in Bangladesh. Qual Life Res 43 2005;14(7):1783-89. 44 41. Tsutsumi A, Izutsu T, Kato S, et al. Reliability and validity of the Bangla version of 45 WHOQOL-BREF in an adult population in Dhaka, Bangladesh. Psychiat Clin Neuros 46 2006;60(4):493-98. 47 42. Zeldenryk L, Gordon S, Gray M, et al. Cognitive testing of the WHOQOL-BREF 48 49 Bangladesh tool in a northern rural Bangladeshi population with lymphatic filariasis. 50 Qual Life Res 2013;22(8):1917-26. 51 43. Laskar MS, Rahaman MM, Akhter A, et al. Quality of Life of Arsenicosis Patients in an 52 Arsenic-Affected Rural Area in Bangladesh. Arch Environ Occup H 2010;65(2):70- 53 76. 54 44. Bradley KD. Applying the Rasch model: Fundamental measurement in the human 55 sciences. Organ Res Methods 2005;8(2):249-50. 56 45. Bartholomew D. Fundamentals of Item Response Theory - Hambleton,Rk, 57 Swaminathan,H, Rogers,Hj. Brit J Math Stat Psy 1993;46:184-85. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 15 of 22 BMJ Open

Page 15 of 18 1 2 3 46. Raykov T, Marcoulides GA. Fundamentals and Models of Item Response Theory. 4 Introduction to Psychometric Theory 2011:269-304. 5 47. Sunderland M, Mahoney A, Andrews G. Investigating the Factor Structure of the Kessler 6 Psychological Distress Scale in Community and Clinical Samples of the Australian 7 Population. J Psychopathol Behav 2012;34(2):253-59. 8 9 48. Beck AT, Erbaugh J, Ward CH, et al. An Inventory for Measuring Depression. Arch Gen 10 Psychiat 1961;4(6):561-&. 11 49. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiat 12 Scand 1983;67(6):361-70. 13 50. Parkitny L, McAuley J. The Depression Anxiety Stress Scale (DASS). J Physiother 14 2010;56(3):204-04. 15 51. Kessler RC,For Andrews peer G, Colpe LJ, etreview al. Short screening scales only to monitor population 16 prevalences and trends in non-specific psychological distress. Psychol Med 17 2002;32(6):959-76. 18 19 52. Kessler RC, Barker PR, Colpe LJ, et al. Screening for serious mental illness in the general 20 population. Arch Gen Psychiat 2003;60(2):184-89. 21 53. Andrews G, Slade T. Interpreting scores on the Kessler Psychological Distress Scale 22 (K10). Aust Nz J Publ Heal 2001;25(6):494-97. 23 54. Cairney J, Veldhuizen S, Wade TJ, et al. Evaluation of 2 measures of psychological 24 distress as screeners for depression in the general population. Can J Psychiat 25 2007;52(2):111-20. 26 55. Browne MAO, Wells JE, Scott KM, et al. The Kessler Psychological Distress Scale in Te 27 Rau Hinengaro: the New Zealand Mental Health Survey. Aust Nz J Psychiat 28 29 2010;44(4):314-22. 30 56. Furukawa TA, Kawakami N, Saitoh M, et al. The performance of the Japanese version of 31 the K6 and K10 in the World Mental Health Survey Japan. Int J Meth Psych Res

32 2008;17(3):152-58. http://bmjopen.bmj.com/ 33 57. Adult literacy Rate in Bangladesh. 2015. 34 58. Saxena S, Carlson D, Billington R, et al. The WHO quality of life assessment instrument 35 (WHOQOL-Bref): The importance of its items for cross-cultural research. Qual Life 36 Res 2001;10(8):711-21. 37 59. District Statistics. Secondary District Statistics 2017. 38 39 http://www.bbs.gov.bd/WebTestApplication/userfiles/Image/District%20Statistics/Na rail.pdf. 40 on September 26, 2021 by guest. Protected copyright. 41 60. Bangladesh Bureau of Statistics. Population and Housing Census., 2010. 42 61. Hossain MD, Ahmed HU, Chowdhury WA, et al. Mental disorders in Bangladesh: a 43 systematic review. BMC Psychiatry 2014;14:216. 44 62. Reavley NJ, Jorm AF. National Survey of Mental Health Literacy and Stigma. 45 Department of Health and Ageing, Canberra 2011. 46 63. Harper A, Power M, Grp W. Development of the World Health Organization WHOQOL- 47 BREF quality of life assessment. Psychol Med 1998;28(3):551-58. 48 49 64. Brislin RW. Back-Translation for Cross-Cultural Research. J Cross Cult Psychol 50 1970;1(3):185-216. 51 65. Permission to use WHOQOL-100 and/or WHOQOL-BREF questionnaires 2015. 52 66. Cheng YH, Chi I, Boey KW, et al. Self-rated economic condition and the health of 53 elderly persons in Hong Kong. Soc Sci Med 2002;55(8):1415-24. 54 67. Serraglio A, Carson N, Ansari Z. Comparison of health estimates between Victorian 55 population health surveys and National Health Surveys. Aust Nz J Publ Heal 56 2003;27(6):645-48. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 16 of 22

Page 16 of 18 1 2 3 4 5 6 Table 1: Variables will be collected for the study. 7 8 9 Variables Measures 10 11 Identification 12 13 14 Date of Birth 15 For peer review only 16 Date of data collection 17 18 Serial number 19 20 Participants name 21 22 23 Mobile number (own) 24 25 Mobile number (relative) Mobile number (next of kin whom he/she can be 26 27 contacted) 28 29 Father or Husbands name Fathers name for males or unmarried males or females; 30 31

32 husband’s name for married or widowed women http://bmjopen.bmj.com/ 33 34 Household Household such as Fakir Bari 35 36 Village Village 37 38 Union/Pourashava Union/Pourashava 39

40 on September 26, 2021 by guest. Protected copyright. 41 Data collector’s signature 42 43 2.Socio-demographic 44 45 variables 46 47 2.1 Age Age in years (18-90 years) 48 49 50 2.2 Sex Sex (Female or Male) 51 52 2.3 Educational Qualification Education (No schooling, 1-5 years (primary), 6-10 53 54 (below school secondary certificate (SSC) pass), SSC or 55 56 higher secondary certificate (HSC) pass, and Bachelor of 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 17 of 22 BMJ Open

Page 17 of 18 1 2 3 above 4 5 2.4 Marital status Marital (married (living with husband/wife), widowed, 6 7 8 never married, divorced or separate) 9 10 2.5 Occupation Occupation (Student, housewives, land owner, cultivates 11 12 other’s land, business, Govt. or non-govt. job, daily 13 14 labour, unemployed, retired, unable to work for disability) 15 For peer review only 16 2.6 Socio-economic status SES (insufficient money all the time, insufficient funds 17 18 19 some of the time, balanced (not good or bad), Sufficient 20 21 funds most of the time) 22 23 2.7 whom do you live with Live with (with own family, with relatives, with non- 24 25 relatives such as neighbour, live alone) 26 27 28 2.8 Income source Income (own income, help from family, help from 29 30 children, other relatives, children and relatives, 31

32 government help, cannot work and do not receive help http://bmjopen.bmj.com/ 33 34 from children or government) 35 36 37 2.9 Smoking status Never smoker, past smoker, current smoker, smokeless 38 39 tobacco only’, smoking and SLT

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 KAP Questionnaire: 44 45 46 A questionnaire will be adopted to collect data on participant’s awareness and KAP 47 regarding mental health, and data relevant to socio-demographic characteristics. Questions 48 evaluating knowledge, attitude and practice of mental health were associated with categorical 49 responses such as ‘‘yes’’ or ‘‘no’’. There are seven items related to mental health disorders 50 (depression, anxiety, psychosis, drug addiction, dementia, bipolar disorder and Alzheimer’s 51 disease) which will be included in the questionnaire. Each of the mental health indicators 52 consists of seven questions. For example, 53 54 1. Have you ever heard of depression? Yes=1, No=0 55 56 2. If yes, please list at least one symptom of depression. Answer was reported as “at least 57 one=1” and “none=0”. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 18 of 22

Page 18 of 18 1 2 3 3. Attitude towards mental health: If you or any of your friends or relatives have had any 4 mental health problems, do they seek treatment? Possible responses were, Yes=1, No=0. 5 6 3.1 If the response was “Yes”, the question was “where would they seek treatment?” 7 with a possible response of “Medical doctor=1”, “Counselling=2” or “others, such as seek 8 advice from spiritual person=3” 9 10 4. The participants were asked whether they had any mental health disorders by asking “Do 11 you suffer from depression?” with possible answers of “Yes=1” and “No=0”. 12 13 4.1 If the response was “Yes”, the question was “Do you take treatment?” with a 14 possible response of “Yes=1” or “No=0”. 15 For peer review only 16 4.1.1 If the response was “Yes”, the next question was “Where did you go for treatment?” 17 with a possible response of “Medical doctor=1”, “psychological counselling=2” or “others, 18 such as seek advice from spiritual person=3” 19 20 The attitude will be consider positive towards the treatment of mental health if they had a 21 positive response. Questions will evaluated the practice of mental health management 22 23 particularly for those who will be diagnosed with mental health, do they take treatment? If 24 yes, where did they go for treatment? 25 51 26 K10 Questionnaire 27 63 28 WHOQOL-BREF Questionnaire 29 30 31

32 http://bmjopen.bmj.com/ 33 34 35 36 37 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 19 of 22 BMJ Open

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

32 http://bmjopen.bmj.com/ 33 34 35 36 37 38 39

40 on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 Location of the Study Area

48 168x233mm (300 x 300 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 20 of 22

Consent Form: 1 2 3 4 Interviewer note: Should be completed from one of the eligible members of aged ≥ 18 years 5 6 7 8 9 10 11 12 Patient’s consent 13 14 We are screening to identify people with depression and its risk factors, and to measure their quality of 15 life in people ofFor aged ≥ 18 years.peer The survey review will increase your awarenessonly of the disease outcome and 16 17 it’s risk factors. The community will be benefited from this study through the intervention programs 18 those the Organisation for Rural Community Development intend to conduct in the future. We do not 19 expect any risk for you if you participate in this study. Upon the completion, the results will be 20 published but your individual information will be kept confidential and your identification will not be 21 disclosed. We expect to continue our study for a longer period for which we may invite you again to 22 participate in our study. However, you are free to change your mind and can withdraw from the study 23 anytime without any obligation if you want. 24 25 26 27 28 Please provide your signature or thumb imprint if you agree 29 30 31 32 http://bmjopen.bmj.com/ 33 34 Signature/thumb imprint 35 36 37 38 39

40 Signature by the interviewer if the participant cannot provide signature. on September 26, 2021 by guest. Protected copyright. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from Page 21 of 22 BMJ Open

1 2 STROBE Statement—checklist of items that should be included in reports of observational studies 3 4 5 Item No Recommendation 6 7 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 8 [Within the title page 1 and method and analysis section of the abstract page 2 ] 9 (b) Provide in the abstract an informative and balanced summary of what was done 10 and what was found [NA ] 11 12 Introduction 13 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 14 [ pages 3-4 ] 15 Objectives For3 peer State specific objectives, review including any prespecified only hypotheses [page 5 ] 16 17 Methods 18 Study design 4 Present key elements of study design early in the paper [ Methods page 5 ] 19 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 20 21 exposure, follow-up, and data collection [ pages 5-7 and page 9; Figure 1 ] 22 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of 23 selection of participants. Describe methods of follow-up [ pages 5-7 and page 9 ] 24 Case-control study—Give the eligibility criteria, and the sources and methods of 25 26 case ascertainment and control selection. Give the rationale for the choice of cases 27 and controls 28 Cross-sectional study—Give the eligibility criteria, and the sources and methods of 29 selection of participants 30 (b) Cohort study—For matched studies, give matching criteria and number of 31

32 exposed and unexposed http://bmjopen.bmj.com/ 33 Case-control study—For matched studies, give matching criteria and the number of 34 controls per case 35 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect 36 37 modifiers. Give diagnostic criteria, if applicable [ pages 7-8] 38 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 39 measurement assessment (measurement). Describe comparability of assessment methods if there is

40 more than one group [pages 15-17: table 1] on September 26, 2021 by guest. Protected copyright. 41 Bias 9 Describe any efforts to address potential sources of bias [ pages 6; Recruitment 42 43 strategy ] 44 Study size 10 Explain how the study size was arrived at [ pages 5; Sample size and statistical 45 power ] 46 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, 47 48 describe which groupings were chosen and why [ NA ] 49 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding [ 50 pages 8-9; planned statistical analysis ] 51 (b) Describe any methods used to examine subgroups and interactions [ NA ] 52 53 (c) Explain how missing data were addressed [ NA ] 54 (d) Cohort study—If applicable, explain how loss to follow-up was addressed Case- 55 control study—If applicable, explain how matching of cases and controls was 56 addressed 57 Cross-sectional study—If applicable, describe analytical methods taking account of 58 59 sampling strategy 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open: first published as 10.1136/bmjopen-2017-016745 on 1 September 2017. Downloaded from BMJ Open Page 22 of 22

1 2 (e) Describe any sensitivity analyses 3 Continued on next page 4 5 Results 6 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, 7 examined for eligibility, confirmed eligible, included in the study, completing follow-up, and 8 analysed [ pages 5-6 and page 9 ] 9 (b) Give reasons for non-participation at each stage [ N/A ] 10 11 (c) Consider use of a flow diagram [ N/A ] 12 Descriptive 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information 13 data on exposures and potential confounders [ N/A now, we will get the information after data 14 collection] 15 For peer review only 16 (b) Indicate number of participants with missing data for each variable of interest 17 (c) Cohort study—Summarise follow-up time (eg, average and total amount) 18 Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 19 Case-control study—Report numbers in each exposure category, or summary measures of 20 21 exposure 22 Cross-sectional study—Report numbers of outcome events or summary measures 23 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their 24 precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and 25 why they were included [ N/A now, we will get the information after data collection] ] 26 27 (b) Report category boundaries when continuous variables were categorized [ N/A ] 28 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful 29 time period [ N/A ] 30 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 31

32 analyses [ N/A ] http://bmjopen.bmj.com/ 33 Discussion 34 Key results 18 Summarise key results with reference to study objectives [ page 9 ] 35 36 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. 37 Discuss both direction and magnitude of any potential bias [ page 10; paragraph 2 ] 38 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity 39 of analyses, results from similar studies, and other relevant evidence [ page 9-10 ]

40 on September 26, 2021 by guest. Protected copyright. 41 Generalisability 21 Discuss the generalisability (external validity) of the study results [pages 9 ; follow up data 42 collection ] 43 Other information 44 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, 45 46 for the original study on which the present article is based [ page 11 ] 47 48 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and 49 unexposed groups in cohort and cross-sectional studies. 50

51 52 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 53 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 54 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 55 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 56 57 available at www.strobe-statement.org. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2