|REPORT FROM THE FIELD
|Year : 2015 | Volume
| Issue : 1 | Page : 98-103
Challenges in conducting community-based trials of primary prevention of cardiovascular diseases in resource-constrained rural settings
Twinkle Agrawal1, Farah Naaz Fathima1, Shailendra Kumar B Hegde2, Rajnish Joshi3, Nallasamy Srinivasan4, Dominic Misquith1
1 Department of Community Health, St John's Medical College, Koramangala, Bangalore, Karnataka, India
2 Department of Community Medicine, S R M Medical College, Potheri, Kattankulathur, Kanchipuram, Tamil Nadu, India
3 Department of Medicine, All Institute of Medical Sciences, Bhopal, Madhya Pradesh, India
4 Raja Muthiah Medical College, Annamalainagar, Chidambaram, Tamil Nadu, India
|Date of Web Publication||19-May-2017|
Department of Community Health, Robert Koch Bhavan, St John's Medical College, St John's National Academy of Health Sciences, Sarjapura Road, Koramangala, Bangalore 560 034, Karnataka
Cardiovascular diseases account for almost half of all deaths from noncommunicable diseases, and almost 80% of these deaths occur in low- and middle-income countries such as India. The PrePAre (Primary pREvention strategies at the community level to Promote treatment Adherence to pREvent cardiovascular disease) trial was a primary prevention trial of community health workers aimed at improving adherence to prescribed pharmacological and nonpharmacological therapies in cardiovascular diseases. It was conducted at three geographically, culturally and linguistically diverse sites across India, comprising 28 villages and 5699 households. Planning and implementing large-scale community-based trials is filled with numerous challenges that must be tackled, while keeping in mind the local community dynamics. Some of the challenges are especially pronounced when the focus of the activities is on promoting health in communities where treating disease is considered a priority rather than maintaining health. This report examines the challenges that were encountered while performing the different phases of the trial, along with the solutions and strategies used to tackle those difficulties. We must strive to find feasible and cost-effective solutions to these challenges and thereby develop targeted strategies for primary prevention of cardiovascular diseases in resource-constrained rural settings.
Keywords: Cardiovascular diseases, challenges, community-based primary prevention trials, implementation
|How to cite this article:|
Agrawal T, Fathima FN, Hegde SB, Joshi R, Srinivasan N, Misquith D. Challenges in conducting community-based trials of primary prevention of cardiovascular diseases in resource-constrained rural settings. WHO South-East Asia J Public Health 2015;4:98-103
|How to cite this URL:|
Agrawal T, Fathima FN, Hegde SB, Joshi R, Srinivasan N, Misquith D. Challenges in conducting community-based trials of primary prevention of cardiovascular diseases in resource-constrained rural settings. WHO South-East Asia J Public Health [serial online] 2015 [cited 2019 Jul 22];4:98-103. Available from: http://www.who-seajph.org/text.asp?2015/4/1/98/206628
| Introduction|| |
Cardiovascular diseases (CVDs) account for almost half of all deaths from noncommunicable diseases, and almost 80% of these deaths occur in low- and middle-income countries such as India. Effective low-cost methods are needed to prevent CVDs. Design and implementation of large-scale community-based trials in low-income rural settings will help to build a robust evidence base for methods to prevent CVDs. The authors have recent experience of such an undertaking, and use this article to list the challenges faced during the implementation of the trial and the strategies that were used to mitigate them.
In 2008-2009, the United Health Group and the National Heart, Lung, and Blood Institute collaborated to support a global network of centres of excellence to help reduce the burden of chronic diseases in low- and middle-income countries. St John’s Research Institute, Bangalore, was chosen as one of the 11 centres of excellence. Under this programme, the PrePAre (Primary pREvention strategies at the community level to Promote treatment Adherence to pREvent cardiovascular disease) trial was designed to test the hypothesis that households with individuals at risk for a cardiovascular event would have improved risk factor levels if they were counselled and monitored by community health workers (CHWs), when compared with households not receiving such visits.
Approval for the trial was obtained from the institutional ethics committees of the participating institutes in India (Rajah Muthiah Medical College, Annamalainagar, Tamil Nadu, India; St John’s Medical College, Bangalore, India; Mahatma Gandhi Institute of Medical Sciences, Sevagram, India) and from the ethics committees of the funding institutes (Population Health Research Institute, Hamilton, Canada; National Heart, Lung, and Blood Institute, Bethesda, United States of America) and the Health Ministry Screening Committee, Government of India. The study protocol was registered with Clinical Trial Registry of India (CTRI/2012/09/002981).
The PrePAre trial was a multicentre, household-level cluster-randomized trial with 1:1 allocation to intervention and control arms, conducted at three geographically, culturally and linguistically diverse sites across India, comprising 5699 households from 28 villages: three villages in Cuddalore district, Tamil Nadu, 15 villages in Kolar district, Karnataka and 10 villages in Wardha district. A short summary of the trial methodology is presented here, and further details have been published elsewhere.
CHWs performed door-to-door surveys and helped to organize village-level preintervention clinics over a period of six months from October 2011. Households were randomized during April 2012. Households randomized to the intervention arm received one household visit by a CHW every two months for 12 months from May 2012 to March 2013. During each visit, CHWs ascertained and reinforced adherence to prescribed pharmacological and nonpharmacological therapies, and measured and ascertained whether blood pressure values met the preset targets. Households randomized to the standard care arm received clinic-based care and did not receive visits from CHWs.
To evaluate the study outcomes, independent trained personnel conducted follow-up household surveys at six months (October 2012), 12 months (April 2013) and 18 months (October 2013) after the first CHW intervention visit. The primary outcome measures were systolic blood pressure and adherence to medication. Secondary outcome measures were INTERHEART risk score, waist-hip ratio and body mass index. The final follow-up visits were completed between October and November 2013. As of today, the trial, including data analysis, stands completed. The investigating team is now in the process of reporting the findings of the trial. However, the monthly clinics continue to provide care to patients, despite the completion of the trial.
| Challenges encountered|| |
The challenges encountered and the strategies undertaken to address these difficulties are described according to the phases of the trial (identification and training of CHWs, house listing and baseline survey, clinic visits, intervention and follow-up visits).
Identification and training of community health workers
Community meetings were held in the 28 villages taking part in the study to obtain community consent for participation in the study. Stakeholders in each village were requested to identify persons from the village who had completed matriculation, were acceptable to the community members and were willing to work as CHWs. An objective selection process was used to recruit a total of 18 individuals.
House listing, enumeration and baseline survey
Community-level consent was obtained before initiating the activities of the trial (see [Table 1]). Subsequently, individual consent and family-level consent were also obtained. Some individuals and families did not readily consent to participate in the study because of the long follow-up period. The investigators spoke personally with these individuals to inform them of the benefits of the trial. Some individuals, despite the best efforts of the CHWs, did not give their consent to participate in the study; however, the number of refusals was negligible.
|Table 1: Challenges faced and mitigating stategies undertaken during various phases of the PrePAre trial|
Click here to view
Households in rural India are not prenumbered. All households had to be identified with serially numbered stickers printed on a 2-inch-square piece of flex-sheet. After seeking permission from the household members, the CHWs pasted the stickers on the upper right corner of the door of each house. Some household members refused permission to have the stickers pasted on their doors. The CHWs explained to the members of the household the utility of such numbering and the need for the stickers to remain pasted for at least 18 months (see [Table 1]).
Loss of previously pasted stickers made it difficult for CHWs to identify households at subsequent visits. In some cases, stickers peeled off as a result of monsoon rains. This had not been anticipated. Social maps had been drawn by the CHWs, in which the numbers that had been used to identify the households were marked. About 20% of all stickers had to be repasted as a result of the stickers peeling off.
Some CHWs found it difficult to effectively engage all the households in a village because of social or political conflicts. In certain villages, CHWs belonging to lower castes were not allowed in the higher-caste households. Many such conflicts were successfully resolved through dialogue. Where dialogue failed, households were excluded from the study.
Safety of female CHWs also needed attention. Most of the CHWs in the trial were women for whom this was their first formal job. On their way around the villages, they occasionally encountered verbal sexual harassment. Repeated community meetings (with village leaders and youth groups) and peer-group discussions (with other CHWs working in other projects in similar circumstances) were held to share coping strategies. Additionally, CHWs were asked to always work in pairs or in groups, and to avoid working at night.
Two very critical challenges were locked households and nonavailability of all individuals over the age of 35 years in the study villages. The latter was more pronounced in the economically productive groups. In order to address these two issues, the CHWs made at least three visits to those households at three different times, including evening visits, to make sure individuals in those houses were included in the baseline survey. In many cases, the CHWs sought a face-to-face meeting with the subject that had been pre-arranged by a phone call. It took multiple visits by CHWs to complete this task because of locked households; therefore, this increased the time taken and the cost of the baseline survey.
Many individuals had migrated to nearby cities and visited their rural roots only occasionly, but were listed in the administrative records as residing in the village in order to qualify for benefits of state-sponsored schemes. In some instances, this led to incorrect registrations and duplications. Investigators had to explain to the household members that such individuals could not be included in the trial and that such a non-inclusion would have no impact on their benefits from state-sponsored schemes. After clarification, duplicated and wrong entries were deleted.
Another very important challenge was reaching remote villages. The investigators personally visited each village, making a note of the distance and the availability of public transport or any other form of transport to and from the village. The investigators discussed this issue with village elders and addressed it through different modes of transport in different villages. In some villages, private auto-rickshaws were hired; in others, villagers who owned vehicles agreed to escort the CHWs to and from the village, both at the time of the survey and at the time of intervention.
Data collection during the baseline survey also threw up some interesting challenges. The standard of living index was difficult to assess in some families, and the CHWs had to corroborate the answers by asking multiple family members. During the enumeration and house-listing phase, CHWs collected the age of every individual in the household. However, at the time of the baseline survey, age was one variable where discrepancies were found. Hence, the CHWs were directed to collect documentary evidence regarding the age of an individual in all such cases during the baseline survey. Some people who consumed tobacco and alcohol hesitated to give details regarding their intake. The CHWs explained to them the study objectives, and the harmful effects of tobacco and alcohol, and how this trial would benefit them in avoiding tobacco and alcohol, thus convincing them to be forthcoming and give information on their smoking and alcohol habits. Measurement of waist and hip circumferences was difficult and was addressed as explained in [Table 1].
Only 70% of those identified as being at intermediate or high risk for CVD attended the village clinics (see [Table 1]). The reason given for nonattendance was that the individuals perceived themselves to be healthy. Some people accused the programme of creating fear in the minds of the people regarding CVDs since they did not have the disease but had only risk factors. The concept of risk was difficult for them to understand. Most individuals identified as having a high CVD risk were asymptomatic and were in the high-risk category because of age, behavioural issues (tobacco use) and abnormal measurements (waist circumference). Many of these individuals did not believe that they were at risk for CVD.
Individuals who were economically productive had to take time out of their occupation to attend clinics; in order to suit their needs, the clinics started functioning very early in the day, as early as 06:00 in some villages, to allow working people to attend clinics before leaving for work. In some villages, the clinics were held on Sundays. Despite these efforts, clinic attendance was not 100%. Health was of a much lower priority when compared with their occupational commitment. Disease becomes a priority when individuals feel either pain or infirmity; since risk factors do not cause any pain or infirmity, occupation and other commitments are more important. Disease would have been a priority; prevention was not.
Some people diagnosed with diabetes or hypertension did not attend clinics because they did not want other members in the village know that they had these diseases. There was a social stigma associated with these diseases. The CHWs and the investigators identified such individuals and counselled them.
At one of the sites, the clinic schedules were disrupted as a result of torrential rains leading to flooding. On another occasion, the clinics were disrupted because of political unrest due to impending local elections. On both occasions, the baseline survey was delayed. These events were totally unanticipated and thus there was a delay in the completion of activities.
The public health care system in rural India does not run clinics below the level of a subcentre. In order to replicate this situation, after randomization had been carried out, regular monthly clinics were held at subcentre locations, in order to review the patients’ medical conditions and to provide drug prescriptions. These clinics were called postrandomization clinics. This resulted in a drop in attendance at existing clinics. Also, most clients in the postrandomization clinics were from the village where the subcentre was located. Despite this limitation, the investigators did not set up village-level clinics, since this would introduce a bias by introducing facilities not normally available. The postrandomization clinics were held at subcentres. There was a change in local governance after the trial began and hence the investigators had to obtain fresh approvals from the local bodies to establish the clinics at the subcentres.
In India generally, the health of male family members is given more importance, while that of female family members is neglected. This was also seen in the clinic attendance, since more men attended the clinics compared with women. CHWs were told to counsel the women appropriately, and reinforce the fact that their health is equally important.
During the course of the trial, a nongovernmental organization started distributing free drugs at one of the sites as a part of their field activity. However, at all sites in this trial, patients were allowed to obtain medicines from any place, provided they adhered to trials prescriptions for both pharmacological management as well as lifestyle modifications. A bigger problem, however, was patients taking herbal medicines and advice from unqualified and unlicensed practitioners. Apart from counselling such patients, it was not possible to do any more.
Intervention (see [Table 1]) used in the trial has been explained in detail elsewhere. During intervention, 10 simple and focused lifestyle modification messages in local languages (no tobacco, dietary modification and physical activity domains) were placed on common household articles (such as mirrors, salt containers, drug pouches, mobile phone covers, television sets, key chains etc.). These articles were given to the intervention households. Some of the nonintervention households as well as nonstudy households also demanded these articles from the CHWs, which could lead to data contamination. The CHWs explained the design and purpose of the project to such individuals and gently declined to hand over the objects to them.
Also, the CHWs were instructed to educate strictly only those individuals who belonged to the intervention households. All interested individuals belonging to the nonintervention households were requested by CHWs to attend the subcentre-level clinics.
In some intervention households, the CHWs were not able to meet the individuals because of their occupational commitments. In such situations, the CHWs fixed a face-to-face meeting with the subject that had been pre-arranged by phone calls.
Migration was an important problem faced during interventions. Permanent migration and seasonal migration due to occupational or social issues were encountered. Effort was made to contact such individuals and include them in the study; however some were ultimately lost to follow-up. In large community-based trials lasting for a long period of time, losing study subjects due to migration is an important issue that needs to be addressed at the time of planning.
Randomized households were followed up at six, 12 and 18 months by a group of independent evaluators who were blinded to the randomization (see Table 1). The evaluators found it difficult to trace the study participants because of similar names and missing household stickers. Tracing was much more difficult among the nonintervention households. To help the evaluators, the CHWs who had done the baseline survey accompanied the new team to locate the study households, but stayed away from the households during the evaluation, and care was taken to ensure that the evaluators were blinded. A short list of the most important lessons learnt from our experience is presented in Box 1.
| Strengths and limitations|| |
This report lists the various challenges encountered and the strategies used to address them during the course of implementing a community-based primary prevention programme. The strengths and limitations of the trial on which this report is based have not been mentioned here since that is the subject of discussion in another article.
The strength of this report is that it not only lists the challenges but also lists the solutions to those challenges. These solutions were customized to suit the needs of the local communities and hence they have immense practical value. The other strength of this article is that it classifies the challenges according to the sequential phases of the trial. Thus any other investigator planning to conduct a community-based primary prevention trial in a resource-constrained rural setting can relate to this article.
The major limitation of this article is a lack of quantitative assessment of the various challenges. The investigators documented the challenges, but not the quantity in terms of numbers. For example, it would have been more useful to report the proportion of people that were not available at home at the time of visit of the CHWs or the additional number of visits made by the CHWs to reach out to the people.
| Conclusions|| |
Planning and implementing large-scale community-based programmes is filled with numerous challenges that must be tackled while keeping in mind the local community dynamics. Some of the challenges are especially pronounced when the focus of the activities is on promoting health. Health is not considered a priority, disease is. Despite these challenges, we must strive to find feasible and cost-effective solutions to address these challenges in the best possible way and thereby thwart the threat posed by CVDs.
Source of Support: The PrePAre trial was funded by a grant from the UnitedHealth Group and the National Institutes of Health, United States of America. No additional funding was received for the preparation of this report.
Conflict of Interest: None declared.
Contributorship: TA and FNF conceived, planned and revised the manuscript. SKBH and NS wrote the first draft and reviewed the manuscript. RJ and DM reviewed and gave final approval for the manuscript.
| References|| |
Fathima FN, Joshi R, Agrawal T, Hegde SB, Xavier D, Misquith D, et al. Rationale and design of the Primary Prevention strategies at the community level to Promote Adherence of treatments to prevent cardiovascular diseases trial number (CTRI/2012/09/002981). Am Heart J. 2013;166:4-12.