Nearly 5.8 million people in India die from Non-Communicable Diseases (NCDs) every year, according to the World Health Organization (WHO). While there is a long list of NCDs, the most prominent ones with high mortality and prevalence rates are Cardiovascular Diseases, Chronic Respiratory Diseases, Diabetes Mellitus and Cancer.
Together, NCDs account for over 60% of all deaths per year. India’s share of the world’s NCD mortality burden is about 15%. WHO estimates that every year, there are 41 million NCD-related deaths globally. The NCD Alliance estimates that about 18 million of these deaths are of women. Of these, the Alliance notes, the biggest killer is Cardiovascular disease, with 8.6 million women succumbing worldwide annually. Women in Low and Middle-Income Countries (LMICs), already facing disparities in health access and outcomes, suffer disproportionately from the burden of NCDs.
Given this volume of annual NCD deaths, India was the first country to draw up a national plan under WHO’s Global Action Plan for the prevention and control of NCDs. The plan emphasizes prevention and a 25% reduction in mortality by 2025. Incidentally, the Sustainable Development Goals also aim at a one-third reduction in NCD mortality by 2030 (Goal 3.4).
While the mortality numbers and goals/targets to reduce them are telling, it is worthwhile to dig deeper into the national and sub-national prevalence and mortality data, as well as into other significant indicators that describe and measure the burden of NCDs. For example, while the number of years lost (YLLs — Years of Life Lost) is an indicator to describe premature death due to disease, Disability Adjusted Life Years (DALYs) also take into account loss of well-being and productivity due to the disease-induced disability. Even if YLLs show a decline, DALYs can be an indicator of the social and economic cost of suffering from chronic NCDs. DALYs are higher if the chronic illnesses set in earlier, as seen in the case of Diabetes Mellitus in India.
Data for all these measures can come from a variety of sources. National disease surveillance programs, and periodic population & household-based health surveys are some of the means of collecting data on disease prevalence and outcomes. As an example of data available on NCDs, the rest of this article will focus on two main NCDs in India — Cardiovascular diseases and Diabetes Mellitus. The discussion will also demonstrate the need for high-quality health-related Open Data and discover the areas where such data is available and where it is not.
Prevalence, Mortality and DALYs
The following visuals show the prevalence and mortality numbers for Diabetes and Cardiovascular diseases (with additional links to interactive visualizations where readers can hover over the maps for state-level numbers and the bars for male/female numbers). Note that the two data components in the visual are from different data sources: the prevalence data by state (in the maps) is from the National Family and Household Survey (NFHS 2020, also known as NFHS-5). Researchers are able to use government data readily because other researchers like @Pratapvardhan on Github — an open data portal — clean up the data and arrange it in downloadable files on Github and other such open source platforms. Otherwise, we are left with reports put out by the government agencies in pdf files that need to be machine-read and extracted into a database. One page at a time. The gender numbers for prevalence per 100,000 population are from the Global Burden of Disease Datasets (2019).
*Click here for the interactive visualization of Diabetes prevalence in India.
Across India, 25% of women and 27% of men reported suffering from diabetes. There are significant regional variations, with the southern states showing higher prevalence for both men and women. On the bars below the maps, we can see near similar prevalence for both men and women across the top ten states. Some regions vary. In Punjab, for example, we see a higher prevalence per 100,000 women than for men.
If the government’s target (set in 2010) of a 25% reduction in mortality by 2025 is to be met, how many fewer deaths would we need to see? Across the country, the year 2010 saw 203,158 deaths due to diabetes. A 25% reduction would mean approximately fifty thousand fewer deaths in 2025. However, deaths have in fact increased over the years since 2010, with 272,281 deaths reported in 2019. In effect then, by 2025, the country would have to reduce mortality by nearly a hundred thousand deaths. It seems like a tall order indeed.
The country would have to see about 58,000 fewer deaths per year due to diabetes alone. When we add up deaths due to each NCD that have to be prevented to meet the target, the challenge gets steeper. In addition to lowering mortality, DALYs also indicate the need for sustained care over the lifetime of the patient to ensure that the adverse impact of the disease is mitigated (thus reducing DALYs). With improved diagnosis and treatment over the years, DALYs should see a downward trend. An upward trend would imply inaccessible or un-accessed testing and treatment.
The next set of maps shows the prevalence of cardiovascular diseases. Since NFHS 2020 did not have a question on cardiovascular diseases in particular, only the Global Burden of Disease (GBD) dataset is used to report prevalence per 100,000 population.
*Click here for the interactive visualization of Cardiovascular prevalence in India
There is significant state-level variation in DALYs per 100,00 population. A couple of states like Punjab and Tamil Nadu show a higher number of DALYs, with a nearly 2000 point difference upward from the median value of approximately 4000 DALYs per 100,000 population. In particular, women’s DALYs have been increasing in states like Punjab, and men’s DALYs have shown modest decreases. This is important to note because, if Disability Adjusted Life Years are increasing, women are not seeking or receiving optimum care for cardiovascular diseases and the disease burden for women is increasing in some states.
There were 2.3 million deaths in 2015 due to cardiovascular diseases. In 2019, the number of deaths increased to 2.57 million. If the Government of India’s 25% reduction by 2025 is to be achieved, we need to see a reduction of more than half a million deaths compared to the 2015 numbers (the number of deaths will have to be 1.71 million or fewer to achieve the target).
Why is this data important?
Both sets of indicators have demonstrated the importance of robust, high-quality data across years (and decades!) so that disease prevalence and mortality are tracked and, health policy and implementation focus on prevention and reduction of mortality. However, taking it one step forward, the data can also be analysed along important indicators like gender (just one example here). Both disease patterns have shown that prevalence and mortality for women are almost at par with men. In fact, mortality from diabetes for women is marginally higher than for men.
If India were to meet the targets of mortality reduction as well as Sustainable Development Goals by 2025 and 2030 respectively, ideally, the reduction needs to happen across all categories including gender. The key question to be asked at this stage, therefore, is: are men and women equally seeking care for their chronic health conditions? Other socio-economic categories such as social group, income levels, religion and the intersection of all of these are equally important but will not be discussed here.
Gender as a Key Differentiator
A study conducted by the All India Institute of Medical Sciences (New Delhi) examined more than 2 million outpatient records and the hospital’s patient management system covering patients from four states in 2016. The study concluded that only 37% of the patients were women and the proportions did not match the population ratios, nor the prevalence data on chronic illnesses. They estimated that approximately 400,000 women patients were missing (i.e. did not seek care) from their hospital in 2016. Researchers involved in the study observed that the greatest disparity in seeking tertiary care was seen in cardiology.
Optimum care for these chronic illnesses requires continuous resources and access. Narasamma’s experience with both diabetes and cardiovascular disease gives us a glimpse into the challenges. In her younger days, 69-year-old Narasamma used to work as a domestic help in the city of Bangalore. After her children grew up, she and her husband moved back to their village, about 70 km from Bangalore.
In 2011, Narasamma sought out medical help at a private hospital in Bangalore when she started experiencing weakness and various aches and pains. She was diagnosed with diabetes and cardiovascular disease and had to undergo surgery for heart disease. Medicine required for her chronic ailments is not available in the village where she lives, or even in the government hospital in a nearby town. So Narasamma travels every month to Bangalore to buy the medicine she needs.
As recently as 2019, the Government of India reported that more than 80% of the population in India is not covered by any type of insurance (it is reportedly changing now with some government schemes like Ayushman Bharat - enrolling larger numbers into the government health scheme). The utilization of such new schemes for NCD care remains to be studied. One such preliminary study (Working Paper 014) by the National Health Authority assessed claims for cardiac care between 2019 and 2020 under the government’s health insurance scheme and found that only 30% of those who accessed care for cardiac conditions were women.
Luckily for Narasamma, she was able to tap into government insurance funds for her heart surgery, though her family had to pay about 80,000 rupees (about 1,400 US dollars at 2011 exchange rates) due to the complexity of the surgery. Without the insurance, the family would have had to shell out 6,400 US dollars. However, for the medicine that she needs every day, she has to purchase them in the city. This is in addition to the insulin and other medicine she has to buy for diabetes.
Both studies concluded that regardless of cost or insurance status, barriers to access for women were many. Women’s access to care has to be improved if a reduction in mortality is to be achieved. Removing such barriers to access is not easy — women’s participation in the formal labour force is at its lowest at 17%, which translates to a lack of wages or steady income. A majority may participate in the informal workforce which is plagued by inconsistent incomes, lack of social safety nets, insurance and seasonality.
After retirement, Narasamma is left with a government pension of 1,200 rupees (approximately 15 US dollars). It is hardly enough to pay for all the costs she incurs to manage her two chronic ailments. Her diabetes medicines, including insulin, cost her 3,700 rupees per month (about 50 US dollars). Heart medicine is about 1,000 rupees (13 US dollars). If the doctor at the hospital orders any tests like an electrocardiogram (ECG), she has to pay extra, along with the consultation fee of 1,200 rupees (15 US dollars).
Narasamma travels to the city every month to get the medicine at the diabetes hospital and at the larger pharmacies in the city (for her heart medicines). She has to take two buses to get to the city to meet up with her children who then help her get to the hospital and pharmacy. For this, she spends another 500 rupees (7 US dollars) on bus tickets. In total, Narasamma spends about 5,200 rupees (about 70 US dollars) per month.
Inability to pay, prioritization of the health needs of the breadwinner in the family, lack of transport, and lack of support to visit a far-off hospital for long-term care (which a chronic NCD requires) are some of the factors hindering consistent treatment of NCDs in women. Treatment of diabetes, for example, costs about 6,000 rupees ($80 at current rates) per year in rural areas and 10,000 rupees (approx. $135) in urban areas. Treatment for cardiovascular diseases can be even higher, especially if inpatient care is required. The hospitals providing NCD care — especially for cardiology — are unavailable in smaller towns and rural areas, thus requiring frequent travel to hospitals in far-off towns and cities. All these barriers are often insurmountable to women with limited or no financial means to pay for sustained treatment of a chronic illness.
*Click here for the interactive visualization of the cost of Diabetes and Cardiovascular care in India.
It is clear that Narasamma would not be able to afford the steep financial burden to treat her chronic ailments. In theory, the government hospitals near her that are responsible for treating her and providing free or subsidized medicines often do not have the resources to support her treatment consistently. “They don’t always have the medicine I require,” she says. “The pharmacies are smaller, and the government hospitals are not always stocked with these medicines.” She has no choice but to seek treatment and medicines in the city which requires her to travel there every month.
Narasamma’s son is a cab driver and lives in the city with his family. He pays for her medicine each month. “I have to arrange for the money once I know she is coming into town to get them,” he said. There are occasions when he had to borrow money to pay for the medicines. Sometimes he has taken loans from banks, but these days it is more often from friends and family. It was especially difficult during the COVID-19 period when his own income fell, so he had to borrow money to buy medicine for his mother.
Without such a family support system, Narasamma’s chronic illnesses would go untreated or under-treated. “I am happy with the level of care provided,” Narasamma says of the private hospital she frequents. “They treat us well and prescribe good medicines. Earlier, they used to give the medicine free because they had a donor [for low-income patients], but now we buy them. But it is good because my health is stable with their treatment.” It is possible to be treated there only with her son’s financial support. She is lucky that she has a place to stay when she is in the city, and her children accompany her to the hospital or pharmacy. Without such support to help with the logistics of travel, stay, and finances, patients would find it challenging to access treatment and medicines as consistently as Narasamma has been able to over the last decade.
Those unable to maintain such access contribute to increasing DALYs and mortality numbers due to these chronic NCDs. Measuring such access/lack of access and the resultant fall out of undertreatment and lack of treatment needs more robust healthcare utilization data from all providers across the country. Who is falling through the cracks and how can we count them?
New Frontiers in Health Care Utilization Data in India
To achieve the targets of reducing mortality and prevention of NCDs, healthcare providers, governments and policymakers need data on who is seeking and receiving healthcare, where and how and, the outcomes of care to assess the quality of healthcare being provided. Utilization studies are few and far in between and data is not consistent, periodic or comprehensive. This is especially true for private/non-government hospitals where utilization data is not aggregated or shared for wider use by researchers and policymakers. This is a significant lacuna in the data available since 70% of India’s citizens seek health care in private doctors’ clinics/private hospitals or other informal providers. Self-reported data in surveys conducted by the government is usually the only available data on healthcare utilization in the country.
Data on health care utilization/cost of care/source of care (public, private, other) along with anonymized demographic indicators will give a window into health care utilization and help us study barriers to care for NCDs. Currently, while we have prevalence data, DALYs calculated by region and by gender, we have no reliable data on who is able to access sustained care for NCDs. As they are chronic diseases, patients need review and medication for long periods of time, often for their lifetime. With the current state of utilization data (even if aggregate, and not patient-wise — since Electronic Health Records (EHRs) are not widespread in government hospitals and are used only in high-end private hospitals in large cities) — it is not possible to study any trends and patterns. Even where EHRs exist, due to concerns of privacy and confidentiality and a lack of robust regulatory frameworks governing health care data (or any other data), sharing of data hardly ever happens. If de-identified patient data (with patient consent under the best case scenario) is stored, curated into datasets and shared (even with a cost involved), it can be of immense value to assess patterns of disease, treatment pathways, the effectiveness of treatment and, health disparities at various intersectionalities.
If we have to assess whether women (or any other vulnerable categories of citizens such as the disabled, the socially disadvantaged/marginalized groups) are seeking treatment for NCDs, and the reasons they are not doing so, we need to have access to data from health care providers in all settings: government, private; rural, urban, OPD and in-patient. Even if all settings are not primed to provide data due to imperfect and/or manual record keeping, a start can be made by collecting from as many providers as possible in each state. Data aggregators can also be trained to convert paper data to electronic data, thus creating datasets from paper records at hospitals and other healthcare settings.
How to set and adhere to regulatory frameworks to collect and use anonymized patient care data ethically and without harm to patients is an issue that India as a country has to deal with sooner or later. Allowing this lacuna to be a show-stopper in developing and using datasets is counterproductive to the cause of health care for all. Instead, all stakeholders have to push for the development of such frameworks that will allow us to move forward with collecting and using healthcare utilization data. If we were to pull off this high-stakes project of patient data for ethical use, it would indeed be a new frontier in healthcare policy and planning in India, hopefully leading to better delivery of healthcare to all sections of the population.
Research for this article and the preparation of visuals in Tableau were a part of a Data Journalism Fellowship (2021–2022) awarded to the author by Equal Measures 2030 (EqualMeasures2030.org) and the Tableau Foundation (USA).
It was then published as part of Open Heroines’ writing grants series, intended to elevate the voices of women and non-binary people in the open spaces.
Lalita Pulavarti is the Director at 3H Catalyst, where she sees an exciting opportunity to use her education, skills and experience as a trainer to design and implement upskilling programs in India. She has over twenty years of experience spanning social research, gender, grassroots development and, Accountability and Transparency in Governance. Find her on Twitter, @Lalita_Pulavart.
