Does the quality of healthcare providers that people use vary systematically by socioeconomic status in rural India? A new study by Jishnu Das, Senior Visiting Fellow at CPR, and Aakash Mohpal combines unique data (collected between 2009 and 2011) on the quality of providers and primary care visits among 23,275 households in rural Madhya Pradesh to examine this question. Their paper titled Socioeconomic Status and Quality of Care in Rural India: New Evidence from Provider and Household Surveys was published in a special issue of the journal Health Affairs.

Why is the research important?

Equitable health systems ensure that the poor and the rich receive the same quality of care. Assessing equity in healthcare and identifying disadvantaged populations is the first step towards improving health outcomes for those who need it most. Yet, the lack of data on who provides healthcare in rural India or how rural populations use such providers hampers a systematic approach to the question.

What does the research do?

For 100 villages in rural Madhya Pradesh, Das and Mohpal (a) provide the first ‘counts’ of the availability of healthcare providers and their qualifications; (b) measure the quality of available care using specially developed tests of medical knowledge known as ‘medical vignettes’ and; (c) survey all 23,275 households and link household characteristics to the quality of providers they visited.

These unique data allow the researchers to understand the nature of the disadvantage in rural India as it relates to the use of healthcare. It answers the key questions:

How many health care providers can a rural household access in this region and what are their qualifications?
What types of health care providers provide the bulk of primary care in rural Madhya Pradesh?
How does quality of care vary by household’s socioeconomic status?
How was the research conducted?

The study was completed in three phases:

In the first phase, Das and Mohpal visited all villages and convened focus groups to identify all providers that households accessed for primary care. These providers included both those who were in the village as well as those who were in markets close to the village–typically on the main road or highway nearby. For all providers who we identified, they completed a short questionnaire that included information on their demographic, practice and clinic characteristics.
In the second phase, they surveyed all 23,275 households and asked about morbidity in the household. For those who had fallen sick in the last month, they asked if they visited a provider and which provider they visited. Das and Mohpal recorded 19,331 primary care visits in the last month and were able to match household characteristics to provider characteristics for 18,850 primary care visits (98%).
In the third phase, they returned to a large sample of healthcare providers and administered medical vignettes to assess their knowledge. Using their performance on this test as a measure of quality, they examined the link between village and/or household socioeconomic status (SES) and the quality of care that people receive.
What were the key findings?

The average village in their sample could access 11 healthcare providers and 49 percent of these providers had no formal medical training. Usage data were even more striking: 77 percent of all primary care visits were to providers without any formal medical training. Only 11 percent of all primary care visits were to the public sector and only 4 percent were to providers with an MBBS degree.
Providers of average quality in the selected sample were able to correctly diagnose 5 key conditions 47.3% of the time and correctly treat these conditions 68% of the time. Because in some cases correct treatment could include ‘referrals to a higher level’, conditions could be correctly treated without a correct diagnosis. Providers with an MBBS degree had higher correct diagnosis and correct treatment rates, relative to those with alternate qualifications (AYUSH) and those without any medical training.
When it came to equity, there was a key difference between the `village’ and the ‘household’ as the unit of analysis. Low SES (socio-economic status) households living in low SES villages use low quality care. But low SES households living in high SES villages use higher quality care. In fact, when Das and Mohpal compared low and high SES households living in the same village, they found no difference in the quality of care received. These findings show that where people live matters more than who they are.
There is striking evidence in this population–the majority of which is poor and illiterate–that households can assess the quality of health care providers and actively seek out higher quality. When patients travel farther, they access higher quality care. And low SES households travel farther than high SES households to access the same quality care.
Interpreting the findings:

Despite significant increases in budgetary allocations through the National Rural Health Mission, the vast majority of households surveyed between 2009 and 2011 still relied on private sector providers without formal medical training for their primary care needs. However, two features of the rural landscape and household behavior in this region limit health inequity in the system.

First, villages in this region are ‘integrated’ rather than ‘enclaved’. That is, instead of a situation where most high SES households live in high SES villages and low SES households live in low SES villages, the researchers found that a significant fraction of low SES households live in high SES villages. This means that they have access to higher quality providers, either in the village, or close to the village.

Second, low and high SES households living in the same village visit providers of similar quality. This is not because they visit the same providers, but because low SES households are willing to travel farther to access higher quality. And by doing so, they can reach more competitive markets where prices are lower.

Consequently, most of the inequity in this system arises from disparities between larger, higher SES villages that are also well connected and small, low SES villages that are scattered, far from roads and can access only low quality providers. In some of these villages, even walking 2 hours will not bring the patient to a higher quality provider.

Way forward:

At the outset, Das and Mohpal note that providing public care in scattered rural outposts is a very costly option. Even if the government were to staff these posts, the number of patients would be so low that doctors may effectively provide care to only 5-6 patients a day. Options that are worth pursuing include (a) training informal sector providers who practice in every village, and (b) providing some kind of medical transport that allows households from rural and scattered villages to visit providers in larger towns and cities.

The full paper can be accessed here.