Preparedness of primary & secondary care health facilities for the management of non-communicable diseases in tribal population across 12 districts in India

Non-communicable diseases (NCDs) have emerged as a public health challenge for the poorest billion population1. NCDs and injuries accounted for 800,000 deaths among people below 40 yr among the poorest populations globally1. India is undergoing an epidemiological transition with the rising burden of NCDs. Although the burden initially increased in more urbanized and developed States, all States have transitioned to higher NCD mortality than infectious and other causes in the past decade2. NCDs account for more than half of disability adjusted life years (DALYs) in India3.

According to the 2011 census, the scheduled tribe (ST) population constituted 8.7 per cent of the 104 million in India4. The literacy of India was 67 per cent in 2011 compared to 59 per cent among the ST population4. ST communities live in about 15 per cent of the country’s areas in various ecological and geo-climatic conditions ranging from plains and forests to hills and inaccessible regions5. The STs below the poverty line exceed the national average6. Furthermore, the health system has poor infrastructure and an inadequate workforce in tribal areas6.

Health indicators of the ST population were worse than the national average for most indicators, such as infant mortality rate and institutional delivery6. In addition to poor reproductive child health and infectious disease burden, evidence also suggests high burden of various risk factors such as tobacco use, alcohol use and hypertension7-9 Poor insurance coverage and challenges in seeking healthcare due to difficult geographical terrain and low literacy lead to poor health outcomes10. The percentage of individuals (15-49 yr) who visited health facilities/camps in the three months preceding the survey was lower (11.5%) among STs rather than in other castes (23.2%)11. The high burden of NCDs might cause complications and mortality among the tribal population due to poor access to healthcare and inadequate healthcare services. Therefore, this study was conducted to describe the availability of various resources in the health system for patients with NCDs.

The Government of India launched the National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke (NPCDCS) intending to prevent and control NCDs through awareness generation, behaviour and lifestyle changes, early diagnosis and treatment of hypertension, diabetes and three cancers12. Health systems strengthening at the primary care level are essential to implement NCD interventions among the indigenous populations. The six building blocks of the health system are health services, health workforce, health information system, essential medical products, vaccines and technologies, health financing system, leadership and governance. There are limited data regarding the status of resources required to implement NCD interventions in the tribal areas in India. Therefore, this study was planned to survey healthcare facilities to map the existing resources, identify the gaps to inform policy decisions. The objectives were to describe the infrastructure (drugs, equipment and information systems) and human resources available for screening of five NCDs (hypertension, diabetes, cancer cervix, breast cancer and oral cancer) and treatment of hypertension and diabetes in the primary and secondary care facilities in the selected tribal districts of India. The survey included services which were part of the NPCDCS programme guidelines. We also estimated the knowledge level among the study participants and described the doctors’ prescription practices to manage hypertension and diabetes.

Material & Methods

Study design and population: A cross-sectional survey was conducted in a phased manner between 2015 and 2018 after obtaining approval from the Institutional Ethics Committee of ICMR-Regional Medical Research Centre and ICMR-National Institute of Epidemiology, Chennai. A written informed consent was obtained from all participants prior to the start of the survey. The survey was part of a comprehensive study on NCDs in tribal populations with multiple components. The sites were covered in a phased manner, with 2-3 sites surveyed yearly for various aspects of NCDs. The results were shared with the local authorities soon after the survey to initiate the action. Eight northeast States (Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim and Tripura) and four other States (Andaman and Nicobar Islands, Himachal Pradesh, Madhya Pradesh and Orissa) were included in the study (Table I). One district per State was selected for the survey with a 50 per cent or more tribal population. Sikkim is a small State; hence, all facilities in the State were included. Public sector health facilities and doctors working in the surveyed facilities were included in the study.

Sample size and sampling procedure: The study districts were selected to ensure the representation of the various parts of the country. All northeastern States were included due to the high proportion of the tribal population. Within States such as Assam, where all districts are not tribal, a district with at least 50 per cent tribal population was selected. On the mainland, four States were chosen, considering the geographic representation and presence of different tribes. In each State, one district with a tribal population >50 per cent was selected while prioritizing the districts under the NPCDCS national programme for NCDs12. At least one doctor was included from each facility, depending on the availability on the survey day. All the doctors assigned to the medicine outpatient (OP) on the survey day, were included in the study. All district, sub-district hospitals and CHCs were included. In a few sites where the number of PHCs was more than 10, 10 PHCs were randomly selected, and in the remaining sites, all PHCs were selected.

Data collection: Data on the health facilities and providers were collected using structured questionnaires. The field investigators were trained at the selected study districts. The field staff was graduates and hired locally through an advertisement as per the State rules. The field supervisor monitored data collection. The data in the medical OP were collected in context to the provider survey in addition to those pertaining to services, human resources, equipment and drugs from doctors, nurses, laboratory technicians and pharmacists. The data collection tool included components which were part of the NPCDCS programme.

Data was also collected on the knowledge of doctors about diagnosis guidelines and treatment practices for hypertension and diabetes, various categories of human resources and their level of training related to managing NCDs, information on screening five diseases (hypertension, diabetes, breast cancer, cervical cancer and oral cancer), hypertension and diabetes treatment services and documentation practices. In each facility, equipment such as blood pressure (BP) monitors, glucometer, height scale, glucometer strips and semi-autoanalyzers were assessed. The availability of medications prescribed for NCDs at the pharmacy on the day of the survey was also evaluated.

Operational definitions of terms describing diseases type or study centres are provided as Supplementary material.

Data analysis: The proportions for various indicators were analyzed using MS Excel 16.0 and Epi Info software 7.2.4. The critical indicators for different health system pillars, namely; service delivery, human resources, health information systems, equipment and drugs were analyzed. The indicators included the proportion of facilities with protocols, the proportion providing screening and OP services, the proportion with various cadres of staff, and the proportion with the availability of equipment, laboratory services and drugs to treat hypertension and diabetes. Sikkim was considered as one district for analysis. The key indicators were stratified into two categories, northeastern and other districts. Northeastern districts included Sikkim, East Kameng, Dhemaji, Senapati, East Garo Hills, Lunglei, Mokokchung and Dhalai. The other districts included Nicobar, Kinnaur, Mandla and Koraput. The indicators were also stratified based on two types of facilities. District and sub-district/taluk hospitals were combined in one category as secondary care facilities and PHCs and CHCs were combined in another category as primary care facilities consistent with the norms in the State health departments. The knowledge indicators included the proportion of doctors who correctly reported the cut off for diagnosing hypertension/diabetes and respective prescription practices.

Results

Screening and treatment services: NPCDCS was operational at all the study sites. State- and district-level nodal officers provided an overview of the programme-related activities. The programme envisaged screening and management of the major NCDs such as diabetes/hypertension at the CHC level and management of cardiovascular diseases at the district hospitals. The programme primarily provided funding for awareness-related activities, additional human resources, training, drugs and consumables at CHCs and district hospitals.

Twelve districts were included in the study with ST population above 50 per cent (except Sikkim)13. Of the 177 facilities surveyed across the 12 districts (Table I), 156 PHCs/CHCs and 21 district/sub-district hospitals were included in this study. Most of the facilities were functioning in government-owned buildings and had electricity connections. Among 14 CHCs, two CHCs did not have drinking water facilities for patients. There was an AYUSH clinic located at 14 CHCs. Overall, one-third of the PHCs conducted screening for hypertension or diabetes. In the northeastern districts, less than one in five PHCs conducted screening compared to nearly half of the facilities in other districts. The screening for three cancers was operational in less than one in 10 primary care health facilities across all districts (Table II). Only 17 per cent of the PHCs had treatment protocols for the diagnosis and treatment of hypertension, and three per cent for diabetes. The district-level facilities and the majority (96%) of the primary care facilities provided OP treatment for hypertension. A large proportion (82%) of PHCs/CHCs and hospitals had outpatient department services for diabetes (Table II).

Human resource: Overall, 151/156 (97%) of the primary care facilities had doctors, and 78 per cent had staff nurses. Three fourths of the primary care facilities had pharmacists and laboratory technicians (Table III). Most hospitals had doctors, nurses, pharmacists and laboratory technicians. However, nurses were assigned various duties in addition to NCD related clinical responsibilities such as screening and follow up. Most facilities did not have a clearly defined task-sharing between different types of staff. Counsellors were appointed under NPCDCS in half of the hospitals in northeastern districts and 20 per cent in other districts. Only a few, about six per cent of the primary care facilities had counsellors and only a few staff in every category were trained for NCD-related activities (Table III).

Equipment and drugs: Sphygmomanometer for blood pressure measurement was available in most health facilities, but the availability of automated digital BP monitors ranged from 35 to 43 per cent in various facilities (Table IV). Only one in 10 PHCs and one-fourth of the hospitals had automated digital BP monitors in the surveyed northeastern districts. A weighing scale was available in most (96%) PHCs/CHCs and 86 per cent of the district hospitals. Glucometer was available in all except three hospitals and 83 per cent of the PHCs/CHCs. Three-fourths of the primary care facilities had glucometers in the northeastern districts compared to 97 per cent in the other four districts. Semi-autoanalyzer was available in 24 per cent and 90 per cent of PHCs and secondary care hospitals. Electrocardiogram was available in 20 per cent of the PHCs/CHCs and 17/21 hospitals.

The pharmacy registers were reviewed for the availability of commonly used hypertension and diabetes drugs. Among antihypertensives, amlodipine was available in three-fourths and atenolol in one third of the health facilities (Table IV). Among diabetes drugs, metformin was available in nearly half and glimepiride in at least one-fourth of the health facilities. The availability of drugs such as amlodipine was lower in northeastern districts compared to other districts (67 vs. 97% at the PHC level). Pharmacists were not trained to estimate the requirement of drugs and reported interruptions in the supply of drugs, including stock-outs.

Health information systems: NCD patients were seen in the OP departments, and the screening register was maintained in only 60/177 health facilities. Only 40 per cent (71/177) of the facilities prepared and submitted monthly reports. Less than one-fifth (28/177) of the facilities maintained separate registers for follow up patients with diabetes or hypertension. Of the facilities which maintained follow up registers for NCD, 13/28 (46%) facilities were from northeastern states. Most health facilities did not maintain the line list of NCD patients under care. Hence, it was difficult to assess the patient load.

Knowledge and practices of providers: A total of 213 doctors were surveyed, of these 149 (70%) were below 40 yr of age, and 157 (74%) were males. Only 15 were AYUSH doctors, and 14 doctors had post-graduate (MD) qualifications. Most of the doctors were treating patients for hypertension, and 167 doctors also treated patients for diabetes.

Three-fourths of the doctors were aware of the correct cut-off for the diagnosis of hypertension, and only 18-39 per cent knew the cut-off for fasting or postprandial glucose for diagnosis of diabetes. One-third of the doctors routinely did anthropometry for patients with hypertension or diabetes. Nearly 60 per cent of the patients were referred for laboratory investigations such as creatinine, lipid and urine albumin. Almost three-fourths (70%) reported amlodipine as the most commonly prescribed drug for hypertension. Other commonly prescribed drugs were ACEI/ARB (44%; angiotensin converting enzyme inhibitor/angiotensin receptor blockers), atenolol or other beta-blockers (41%). Metformin (59%) was the most widely prescribed drug, followed by various sulfonylureas (49%) for diabetes treatment.

Discussion

The present study documents the preparedness of health facilities to manage hypertension and diabetes predominantly in the tribal districts of India. There were significant gaps in the availability of a trained workforce, drugs for hypertension and diabetes, laboratory services and lack of clinical protocols for service delivery as envisaged under NPCDCS. The programme’s implementation was in its early phases at all sites. However, there was variation in the availability of resources based on the States’ initiatives for NCDs. Most facilities had at least one doctor; however, there were gaps in the availability of paramedical personnel. Doctors were inadequately trained to manage these diseases in the primary care setting. The lack of information systems limited our understanding of service demand and the extent of follow up among patients on treatment.

The present study documented the gaps in programme implementation in tribal areas, but given the poor emphasis on NCDs, the health system preparedness is inadequate in the non-tribal areas as well14. The National NCD Survey14 during 2017-2018 highlighted the gaps in public and private sector facilities. A small proportion of the rural primary care facilities were fully equipped with drugs, diagnostics and other required resources14. Our data suggest that an intensified focused mission mode approach will be required to accelerate the health system strengthening for NCD management in tribal areas. In the past, most health programmes in these populations prioritized maternal and child health and infectious diseases6. Given the investments in maternal health, child health and infectious diseases, there has been an improvement in the health indicators in the tribal population in the past decades. However, the pace of progress was slower compared to non-tribal populations6. In recent years, attention was drawn to the growing numbers of NCDs among tribal regions of India, as several studies reported a high burden of hypertension, tobacco use, alcohol use and cancers7-9,15. The interventions and models from non-tribal areas may not be replicable and scalable for the tribal populations due to geographical constraints, health-seeking behaviours and availability of limited resources. The linkages between the health system and community are essential to generate awareness and maintain a continuum of care. Hence, one may need to learn to address the burden of NCDs in India and elsewhere from ongoing community-centric initiatives in tribal areas.

A randomized control trial16 is ongoing in a tribal district of Gadchiroli, India, to estimate the effectiveness of community-based interventions for screening and managing cardiovascular diseases. The model is proposed to reduce stroke mortality in rural areas where access to healthcare is difficult16. Jan Swasthya Sahyog, an NGO based in Chhattisgarh that caters to tribal districts in the region, is also working on several community-based interventions for NCDs. One such initiative encompassed formation of peer support groups for various chronic diseases, and this improved the adherence to treatment17. The models from these initiatives can be analyzed, and best practices can be scaled.

The findings of the present study suggest that, doctors were inadequately trained, and non-physician health workers did not play a significant role in the NCD interventions. Similar challenges have been reported in studies from India and other low- and middle-income countries (LMICs)18. A survey from Tanzania reported a lack of trained workforce, proper reporting and data management systems18. Tribal districts are poorly connected; hence, travelling to distant places may not be feasible except for serious illness. Provision of healthcare closer to home involving non-physician health workers would thus be the most viable strategy to enhance the utilization of services. Systematic reviews have documented the effectiveness of non-physician healthcare worker interventions to improve hypertension and diabetes control. A nurse-led intervention led to a mean decline of −5.23 mmHg systolic and −2.92 mmHg diastolic BP, respectively19. Similarly, non-physician health workers’ interventions led to an average reduction of −36.26 mg/dl (−52.60 to −19.92, I2 = 78%) in the fasting blood sugar20. In the South-East Asian Region, Bhutan documented the feasibility of engaging non-physician health workers to detect and manage diabetes in remote hilly areas21. In 2016, the WHO developed the HEARTS package22, which recommends task-sharing involving all health workers in managing hypertension and diabetes. The Government of India’s health and wellness centres initiative for every 5000 populations involving a community health officer (a specially trained nurse) will facilitate scaling NCD services closer to home, especially in remote areas23. In addition to the inadequate workforce, lack of simplified treatment protocol and training were the other barriers to good quality care. The WHO HEARTS guidelines recommend a simple treatment protocol that can make it easier for doctors to manage patients in primary healthcare settings22. The availability of an adequately trained workforce is essential to improving NCD care in the tribal districts.

The present study documented inadequate availability of drugs and equipment. A multi-country study conducted in eight LMICs reported a similar scenario. Automated BP monitors were available in only 10 per cent of the facilities, and low-cost antihypertensives and diabetes medications were unavailable in all primary care facilities24. Antihypertensive drugs are cheap and readily available in India25. However, forecasting, budgeting and appropriate distribution are required to ensure adequate availability in most peripheral health facilities26. High-quality professional BP monitors should be procured and available up to the PHC level. The WHO has issued a technical specification for BP monitors, which the states and districts can use to plan procurements27. The national programme NPCDCS provides a budget for drugs and consumables for NCDs12. The training of district- and state-level managers in programme planning and management is a must to improve the overall functioning of the NCD programme.

The present study was not without its limitations. The information system in most districts did not capture the NCD patient visits fully; therefore, the adequacy of drugs and human resources could not be assessed. The second limitation was that we could not interview all cadres of healthcare workers.

Overall, the present survey will serve as a baseline to measure the progress in the NCD services delivered in the healthcare system over time under the NPCDCS programme. The challenge of inadequate human resources can be addressed by adding a health workforce at all levels. Furthermore, strategies such as task-sharing and team-based care approaches can help overcome the workforce related challenges. All categories of staff should be trained to detect and manage NCDs and the doctors should be provided with standard treatment protocols. The information system should be streamlined to track the patients, procurement and distribution of drugs and devices should be strengthened, and adequate uninterrupted supply should be ensured.