Perceptions and utilisation of complementary and alternative medicine practices among hospital patients in Bangladesh

Complementary and alternative medicine (CAM) has played an important role in providing universal access to essential health care services globally. Conventional medicine (CM) driven health care practices are well-developed in Bangladesh; however, millions of people utilise CAM-based healthcare services for specic health conditions or health benets due to high out-of-pocket payment (74%) in Bangladesh, while the global average is only 32%. Lack of evidence exists about the perception and utilisation of CAM in Bangladesh. This study aimed to estimate the prevalence correlates of the perception and utilisation of CAM among patients who received health care at a tertiary hospital, Bangladesh. This study comprised a cross-sectional study with 1,183 individuals from the cross-sectional survey among patients who received health care from Government Unani and Ayurvedic Medical College Hospital in Dhaka, Bangladesh. Logistic regression analyses were employed to estimate the adjusted effect of independent factors on CAM health care services utilisation.


Introduction
Complementary and alternative medicine (CAM) has been the traditional method of meeting people's basic healthcare requirements (1). This still holds for one-third of the world's population who lack access to CM. In these circumstances, millions of individuals rely on CAM and its practitioners for their primary healthcare. Often it is their only option due to a lack of access to conventional healthcare, geographic isolation, and high conventional healthcare costs (2). Although conventional medical access has improved rapidly in recent decades the use of CAM is increasing globally in illness prevention, control, and management (2,3) .
A large percentage of people in developed countries use CAM (4)(5)(6), and this trend is also seen in lesser developed countries like Bangladesh (7)(8)(9)(10). About 20% of people in the UK (4), 42% in the USA (5), 48% in Australia (6) and 76% of Japanese people (7) use CAM for their primary health care services. In developing countries such as India (70%) (8), Pakistan (70-80%) (9), CAM usage is widespread. A recent study showed that in Southeast Asia 20-97% of people use CAM (11). In Bangladesh, at least 70% of the population uses various forms of CAM for their primary health care (12). Different clinical or medical methods or techniques like Ayurvedic, Unani, Homeopathy, and Naturopathy are widely practised addressing medical needs in Bangladesh (13). Ayurvedic is a Hindu system of medicine and Unani is a traditional Muslim system of medicine originating in India and Greece, respectively (13,14).
The present study aims to examine the use and perception of CAM among patients receiving treatment in a tertiary care CAM health facility in Dhaka, Bangladesh. To address the objective following the research questions (RQs) were formulated. Sample size determination and procedure The ideal sample size was calculated using the standard formula for a cross-sectional study.
Where n is the appropriate sample size, P is the approximate proportion of CAM use (0.615); taken from a related study conducted in Nepal (15).Q = 1-P is the likelihood of those that do not use CAM, i.e. (1-0.615), Z = the value of the test statistics corresponding to the 95% con dence interval (1.96), and d = the degree of accuracy/standard error (0.05). This results in approximately 364 participants required as a minimum sample size. The nal recruitment of 1,183 participants greatly reduced the risk of sampling error.

Data collection tools and techniques
A structured questionnaire was used to gather information about demographic characteristics, socioeconomic characteristics, and preferences of consumers/patients for CAM usage. Patients were surveyed after completing their CAM healthcare visit. Before conducting the survey, the study objectives, aims, and bene ts were explained to patients. Patients attending the hospital were rst told about the intent of the research and asked to participate by responding to the questionnaire (S1).
Data were obtained either via a face-to-face interview with undereducated or through the selfadministration of a questionnaire for those who were literate. Two skilled medical students at the hospital administered the questionnaires to patients.

Data analysis
Data from the completed questionnaires were coded and analysed using the Statistical Package for Social Sciences (SPSS) for Windows, Version 26. Categorical and continuous variables were expressed in frequency, percentage, mean and standard deviations, respectively. Frequency tabulation was used to summarise basic details such as demographic and socioeconomic characteristics and their relation to CAM use. The obtained data were analysed descriptively. Also, socio-demographic data were inferentially analysed. Usage of CAM was checked by comparing CAM users' socio-demographic data with non-users, using chi-square at 0.05 levels of signi cance. To nd potential socio-demographic predictors of CAM use, a logistic regression model was used. Unadjusted odds ratios (ORs) with 95% con dence intervals were calculated using independent variables from the bivariate analysis with a p-value less than 0.05 in the initial univariate analysis. In the multivariable analysis, demographic characteristics with p-values less than 0.05 in the univariate analysis were entered to obtain adjusted ORs with a 95% con dence interval. All the data were entered into Microsoft Excel, used to generate descriptive statistics, and transferred into SPSS (version 26) for further statistical analysis.

Participant's characteristics
This study included a total of 1,183 patients (  Seventeen different types of health problems were identi ed. Gastrointestinal problems, skin problems, respiratory diseases, and menstrual disorders were the four most common illnesses for which CAM was utilised ( Figure 1). Of the total 67% of the patients said they had tried CM before trying CAM to solve their problem.
Utilisation of CAM  Note: OR = Odd ratio, CI = Con dence interval Most patients used more than one type of CAM treatment. The most frequently CAM therapies were Ayurveda (48%), Unani (45%) and Neuropathy (7%).
The majority of patients reported multiple reasons for using CAM. The most common reasons were: 20.2% believed that CAM could control their illness/disease, 17.7% believed that CAM had fewer side effects than CM, and 16.5% used CAM because people advised them to do so ( Figure 2).
Satisfaction with CAM treatment was very high (78%), and patients were willing to advise others to use CAM (65%). About 19% were willing to combine CAM with conventional treatment.

Discussion
This study aimed to examine the sociodemographic characteristics linked with CAM use in Bangladesh, as well as the relevant indications for CAM usage and patients' perceptions of CAM. The ndings showed that socioeconomic characteristics (age, education, marital status, occupation, and income status) were positively connected to the use of CAM. The preference for CAM was highest among middle-aged patients, which is similar to the ndings of previously published studies elsewhere (4)(5)(6)20). This could be due to their health-seeking behaviours or more likely to seek out treatments that will help them improve their health (13).
Patients with no or school education are more likely to use CAM than those with higher levels of education. However, prior studies have shown that those with higher education and more economically a uent are more inclined to use CAM (4,6,15,17). This could be because well-educated and nancially secure patients are more motivated to look into alternative remedies and ways to cope with their sickness and medication side effects. However, in Bangladesh, poor and non-educated patients often have insu cient money to purchase advanced medical treatment from well-equipped conventional/modern hospitals or clinics (21,22). The study ndings suggest that married patients are more likely to utilised CAM. However, a prior study indicated that single respondents were more likely to utilise CAM than married participants (13). This disparity in results can be attributed to contextualisation and cultural backgrounds. Another factor could be that married women often rely on their husbands for health-seeking behaviour in a male-dominated and patriarchal society like Bangladesh (23).
Patients who own their business or are formal employees are much less likely to use CAM, contradicting previous ndings (6). This might be because the out-of-pocket model of payment is the norm for CAM treatment. This study also found that patients with relatively high-incomes were more likely to use CAM, which is consistent with prior studies conducted in the USA (5), Ethiopia (17), and Nepal (15), in contrast to a study in India (8) and Pakistan (9). This might be because of the tendency of people with high incomes to seek out alternatives for their health care treatments and well-being. Although a recent study reported that along with the income distribution both the lowest and highest socioeconomic groups showed a strong preference for CAM in China (24).
In this study the most common reasons for using CAM were 1. effectiveness of CAM, 2. fewer side effects, 3. easily available and 4. cheaper. These ndings are con rmed by previous studies (8,9,20). The placebo effect is an essential component of CAM treatment. The development of CAM treatment is predictable on a broad base of quality research. There is momentum now to expand beyond basic clinical and experimental research to a joint public health program alongside CM. Interestingly, most patients (78.3%) were satis ed with CAM, and they want to recommend it to others (65.7%) which is in line with attitudes expressed elsewhere in Ethiopia (17) and Bangladesh (12). This may be because of their positive belief that CAM is less harmful than conventional medicine. Many consumers believe that CAM is equally reliable in terms of its scienti c basis with CM which could convince them towards utilising CAM more regularly (13).

Strength and limitations
This is the rst-ever study among Bangladeshi healthcare consumers on their attitudes towards CAM.
The study does however have several limitations. The major one is that the study population was from an out-patient department of a medical college hospital which specialises in CAM. Moreover, as this study was cross-sectional, so it cannot attribute causality to any of its associated factors. A nationwide population-based study is required to be undertaken to understand the exact prevalence, patterns, and perception among the general population of CAM use.

Conclusions
This study examined how patients in Bangladesh use CAM and how they perceive it. Ayurveda and Unani are the most common CAM practices in Bangladesh. Overall, there was a strong link between CAM utilisation and socioeconomic position. Common motivations for embracing CAM include belief in its ability to control disease, resulting in fewer adverse effects, a lack of faith in pharmaceutical treatments, and lower cost than CM. Furthermore, most surveyed patients are satis ed with CAM therapies and are prepared to recommend them to others. Reasons for using CAM * multiple responses

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CAMQuestionnaire.docx