I usually don’t wake up to hate mail in my inbox. What prompted this deluge is a recent paper that evaluates the impact of a training program for informal health care providers (providers without any formal medical training) in the state of West Bengal, India (paper summary). Training improved the ability of informal providers to correctly manage the kind of conditions they may see in their clinics, but it did not decrease their overuse of unnecessary medicines or antibiotics.
Lead Economist, Development Research Group - World Bank
This paper complements two others on rural health care in India. The first shows that 77 percent of rural primary care is provided by informal providers in the state of Madhya Pradesh. The second (ungated) demonstrates that people receive the same quality of care between informal providers and public-sector clinics in the same setting; it also shows that the same doctor provides higher quality care in his private clinic relative to the public one.
The hate mail I received would lead one to believe that simply studying quality among informal providers is illegal or that finding an equivalence of care between the public sector and informal providers is blasphemous. These are surface markers of disquiet. The deeper discomfort is, I believe, because our findings substantially complicate an existing narrative that pins all the blame for poor health care on the private sector and on informal providers.
Health care in rural India
Let’s begin with some conventional facts:
- Most primary care in rural India is provided by informal providers; 77 percent in rural Madhya Pradesh, 50-80 percent in other states
- Even when there is a public-sector clinic in the village, the majority of primary care visits are to informal providers.
These two facts lay out the fundamental conundrum: Either, (Option 1) people are making the best of what they have (one person’s “quack” is another’s only hope). Or, (Option 2) they are being systematically fooled by hundreds of thousands of informal providers running wild in the Indian countryside.
Under Option 1, we are in a standard economics framework where more choice is good. Under Option 2, we are not, and a number of regulatory and other policy responses may be required. What do the data say?
- In rural Madhya Pradesh, there is at least one informal provider in every village; usually there are two to three. In contrast, there are virtually no qualified doctors and very few public clinics. In the 203 villages in West Bengal where the informal providers in our study came from, there were only 11 public primary health care centers.
- Even when there is a public sector clinic, the quality of care is very similar in the public sector and among informal providers. This is because in a majority of cases when patients go to the public clinics, they are not seen by the doctor, but by some other clinic functionary who is operating in the doctor’s place at the time of the visit. When they do see the doctor, they receive higher quality care, but no one knows when the doctor will be there, or how long they may need to wait, or even whether the doctor’s post is filled.
Both these descriptive statistics support Option 1. Furthermore, market features suggest that people are making reasonable decisions:
- Market feature #1: Direct price—private providers who charge higher fees are also more likely to correctly manage cases.
- Market feature #2: When people travel farther (an implicitly higher price), the technical quality they access is higher—even when we condition on the qualification of the provider they visit. The poor travel farther to access the same quality—because they get lower prices if they can make it to dense markets.
Again, these two key market features support the idea that people are willing to pay for higher technical quality. But what about unnecessary medicines and antibiotics? This question leads to the third market feature:
- Market feature #3: Providers who give more unnecessary medicines are also able to charge more. The curious part of health care—you can both get what you need and what you don’t need—points to a poorly functioning market.
But here is the troubling part: In Madhya Pradesh and West Bengal, fully trained (MBBS) doctors are the worst culprits when it comes to giving unnecessary medicines and antibiotics—and even more so when they are in public-sector clinics. Given the huge concern about antibiotic resistance in India, this finding comes as a real surprise and turns the usual narrative of blame on its head. The fully trained MBBS doctors and the public sector are equally or even more to blame for the rampant overuse of antibiotics in India.
Will training informal providers help?
Given these findings, can training improve the quality of care given by informal providers? That is what we attempt to answer in our recent paper in Science. The paper, which shows that training improved correct treatment rates but did not decrease the use of unnecessary medicines, is unusual for three reasons:
- It is the first triple-blind study of the impact of training in primary care. We sent in standardized patients to providers who received training and those who didn’t. Training implementers did not know what patients we would send in and therefore could not tailor their training to the evaluation (Blinding 1). Standardized patients did not know whether they were going to providers who received the training (Blinding 2). The providers did not know they were going to see standardized patients (Blinding 3). We therefore claim that our study comes close to capturing the impact of training on generalized improvements in primary care.
- It benchmarks the performance of control and treatment providers against the 11 public primary clinics in the 203 villages our informal providers came from. Training closed half the gap in correct treatment between informal providers and the public sector. Attendance in the training program was 56 percent and distance—clearly a policy variable—was a strong correlate of attendance. This suggests that increasing the attendance to 100 percent would pretty much close that gap. But public primary clinics were more likely to give unnecessary medicines and antibiotics than the informal providers (market feature #3) and training had no effect on the use of unnecessary medicines and antibiotics.
- Perhaps most unusual was that The Liver Foundation, our partner organization heading the training, has been able to put in place this program and continue it even in the face of sustained attack and opposition from the Indian Medical Association. Even more laudable, the West Bengal Government supported the evaluation, funding part of it from the National Rural Health Mission, and on the basis of the evaluation, has decided to scale up the training to more than 3,000 informal providers and beyond. This is a remarkable case of a state government funding research, acting on the basis of the research findings, and waiting till the research is completed before choosing to act.
What do the findings imply?
The West Bengal government should scale up this training—as they have. These studies do not discuss whether informal providers can be mainstreamed into the delivery of health care. This is a confusing landscape, and if we want to solve all the problems at once (regulatory, legal, access, quality, cost, insurance), we are in for a wild ride.
The government recognizes this, and seems to have adopted a stance of strategic ambiguity: Although informal providers do not officially exist, state and local governments frequently work with them when required. In the area where we conducted the study, there is an association of informal providers who are in regular contact with the chief medical officer and who hold medical camps in popular fairs. In Mumbai, local hospitals have rosters of informal providers in the slums who help with detection and treatment of tuberculosis patients.
The ubiquity of informal providers challenges the very notion of a modern state and puts into question repeated health policy documents, dating back at least to the J.P. Bhore committee report of 1946, which eschewed the use of non-physician clinicians in favor of a utopian public sector with fully trained and altruistically minded doctors.
That model has failed entirely. Alternatives will arise and we will, in our usual democratic chaos, move forward.
But we should start by acknowledging the fact that, at least in this case, the quacks are not the crooks.
- Das, Jishnu, Abhijit Chowdhury, Reshmaan Hussam and Abhijit V. Banerjee. 2016. The Impact of Training Information Healthcare Providers in India: A Randomized Controlled Trial. Science. Vol. 354 (6308)
- Das, Jishnu and Aakash Mohpal. 2016. Socioeconomic Status and Quality of Health Care: New Evidence from Linked Surveys of Providers and Households in Rural India. Health Affairs. Vol. 35(10): 1764-1779.
- Das, Jishnu, Alaka Holla, Aakash Mohpal and Karthik Muralidharan. 2016. “Quality and Accountability in Healthcare Delivery: Audit-Study Evidence from Primary Care in India.” Forthcoming, American Economic Review.