An assessment of awareness of mental health conditions and its association with socio-demographic characteristics: a cross-sectional study in a rural district in Bangladesh

Bangladesh is a country of 163 million people divided into 64 districts. Participants were recruited from the Narail district, which is located approximately 200 km south-west of Dhaka, the capital city of Bangladesh. The population of Narail district is 272,872, with approximately 40% of residents aged between 18 and 59 years and 19,000 (about 7%) of residents aged between 60 and 90 years. The study location was selected as it was reflective of typical rural demography in Bangladesh. The Narail district with an estimated population density of 722 people was comparable to the national rural population density of 873 people per square kilometre. Narail is not at the extremity of remote locations nor is it a catchment of a metropolis such as Dhaka [33]. The area of Narail is 381.76 km2, located in between 23°02′ and 23°17′ north latitudes and in between 89°23’and 89°37′ east longitudes. The district is surrounded by Lohagara and Salikha upazilas on the north, Kalia and Abhaynagar upazilas on the south, Lohagara upazila on the east, and Bagherpara and Jessore Sadar upazilas on the west. The ratio of male to female (48.5 to 51.5) resembled quite well with that of Bangladesh (48.9 to 51.1) [34]. 72.3% of sample data attained primary education or above as opposed to 72.9% [35] of the national population, while 27.7% of it had no education, which was comparable to the national 27.1% [35]. In the make-up of the population with respect to marital status, the Married group of Narial sample data (79.9%) was fairly proximate to the national level (80.01%) [36]. With respect to the availability of funds, a determinant for socio-economic condition, the sample population having insufficient funds some / most of the time accounted for 32.2 and 23.2% [37] in Bangladesh. In summary, the socio-demographic make-up of the Narail district delineated well the typical characteristics of a rural district of Bangladesh. Moreover, researchers carried out studies [32, 38, 39] in Narail district earlier and motived to advance further studies based on the population of Narail district.

The sample comprised 2425 participants, aged between 18 to 90 years from the Narail district. The sample including 1147 older adults and 1278 adults. Prior data indicate that the prevalence of severe depression was 21% in older adults aged 60 years or above, and 6.5% in adults aged between 18 and 60 years [40]. We assumed a margin of error of 5% in prevalence rates for older adults, and of 3% for adults when estimating the true prevalence of severe depression for each cohort in this rural area [40]. Using a significance level of 0.05 and statistical power above 80%, a required sample size of 1128 was needed for the older adult cohort and 1283 for the adult group.

A multi-stage cluster random sampling technique was used for this study. Three unions from a total of 13 and one ward from a total of 9 wards of Narail upazilla were randomly selected at level 1. Two to three villages or mohalla from each selected union or ward were randomly selected at the level 2. About 120 older adults and 150 adults were interviewed from each of the villages. Recruitment strategy and quality assurance in data collection were described in details previously [38]. In brief, all team members participated in an intensive 2-day training programme in Narail before the commencement of the survey. The purpose of the training was to outline the rationale of the study, and the procedures and potential difficulties associated with data collection. The interviewers were instructed to visit every household within the randomly selected villages and to interview one household member of an older adult first. If none were available in this subgroup, the interviewers were approached an adult person of that household. If there was more than one male or female adult in the same household, one individual was selected, based on who was born closer to January. However, to maintain an approximately equal number of males and female participants, one female was interviewed immediately after a male participant. The recruitment started from a corner of a village and continued until the recruitment of a maximum of 250 participants was reached for a large village where the number of eligible participants were greater than 250. In case of fewer than 250 households in a village, the requirement continued to the adjacent village to reach the number to 250.

Given the relative lack of validated MHL data, the specific items measuring MHL were sourced from the National Survey of Mental Health Literacy and Stigma in Australia [41]. A questionnaire comprising these items assessed participants’ awareness of six MHCs (depression, bipolar disorder, anxiety, psychosis, dementia, and drug addiction) by asking if they had ever heard of these conditions with a possible response of “yes” or “no”. Based on their responses, they were asked to list at least one symptom of these conditions, this provided insight into the level of knowledge they possessed. Attitudes towards the use of treatment was assessed by asking the question “do they or their relatives need treatment?” with a possible response of “yes” or “no”. Participants were also asked if they or their relatives had ever experienced any MHCs, and if so, if they had undertaken treatment from a “medical doctor”, “psychologist” or others such as “spiritual persons”.

The outcome variables were determined by the level of awareness of MHCs divided into two groups, common (depression, anxiety and drug addiction) and severe (psychosis, dementia, and bipolar disorder) MHCs. Knowledge was measured based on whether the participants could identify at least one symptoms of the MHCs given that they were aware of the conditions. Attitudes were defined as positive if the participants who were aware of the MHCs were in favour of taking medical treatment or psychological counselling. Practice was defined based on whether participants’ themselves or their relatives with MHCs had undertaken medical treatment or psychological counselling.

Demographic details for age was categorised into adult (18 to 59 years) and older adult (60 to 90 years) groups. Level of education was categorised into five categories: no schooling, primary school level of education (grade 1 to 5), secondary school level of education (grade 6 to 10), school secondary certificate (SSC) or higher secondary certificate (HSC), bachelor’s degree or above. SES was assessed according to Cheng et al. [42] asking whether “over the last twelve months, concerning household food consumption, how would you classify your socioeconomic status?” The possible answers were: (i) insufficient funds for the whole year; (ii) insufficient funds some of the time; (iii) neither a deficit nor surplus (balance); and (iv) sufficient funds most of the time. Occupation was categorised as students, housewives, landowners, labourers, business men or women, government or non-government employees and retired persons.

Participant’s sociodemographic characteristics including age, gender, level of education, occupation and SES were reported using descriptive statistics. Association of MHCs with gender, levels of education and other categorical factors were evaluated using Chi-square tests. The linear-by-linear association option in Chi-square test was used to report trends for ordinal categorical variables. Rasch Analysis is a unique approach of mathematical modelling based upon a latent trait and accomplishes stochastic (probabilistic) conjoint additivity (conjoint means measurement of persons and items on the same scale and additivity is the equal-interval property of the scale) [43]. Rasch models are used when a set of questionnaire items are intended to be summed together to provide a total score. In this study, Rasch analysis [44] was performed to compute the person measures based on the awareness for common and severe MHCs on a logarithmic scale and termed as “awareness score”. The negative value of an awareness score indicates the person has below average awareness, zero value indicates average and positive value indicates above average level of awareness of the items. Differentials of the measures of awareness and help-seeking attitudes and behaviour in relation to MHCs were evaluated across different major categories of age, gender, the level of education and other socioeconomic factors using Pearson Chi-square tests of independence. Multiple regression analysis techniques were used to evaluate the direction and strength of the effects of socio-economic factors gender, age, SES and education in predicting awareness of underlying MHCs. Data were analysed using RUMM2030 [45] and SPSS 24.0 [46].