Improving the Quality of Private Sector Health Care in West Bengal

National Rural Health Mission, Government of West Bengal, India, Liver Foundation, West Bengal, World Bank

The rural healthcare market in much of the developing world is composed largely of informal private providers. These private providers often have little to no certifiable medical training. Recent studies in India using medical vignettes (or hypothetical medical situations) to measure clinical competence and direct observations of doctor-patient interactions to measure clinical practice highlight the poor quality of care that most patients receive-a problem that is clearly relevant beyond India and affects most low-income countries worldwide. For instance: In rural India, standardized patients presenting with chest pain and (on further questioning) radiating pain in the arm are (correctly) diagnosed with a heart attack in less than 25 percent of cases. Across 8 low and middle-income countries, health care providers completed the four necessary vital statistics for new patients in less than 4 percent of interactions: health care providers in the public sectors of many developing countries routinely spend less than 1 minute per patient. To address these deplorably low standards in both medical knowledge and practice, the Liver Foundation in Kolkata has been working with private rural health care providers through capacity building activities to improve quality in the private sector. The program consists of multiple-week training to private rural health care providers on the basis of a well-developed curriculum in the district of Birbhum, West Bengal. This study aims to assess the impact of this training program using a randomized evaluation, in which providers are randomly assigned to the treatment, i.e. the Liver Foundation's training program, or the control, i.e. no such training. As an independent outside evaluation team, we will run a baseline survey for all providers (through a third party data collection agency), monitor the application of and compliance in the Liver Foundation's training intervention, and conduct a final endline study. By comparing the treatment and control groups on a variety of measures developed to capture competence in provider knowledge and practice, we can rigorously assess whether such a training program for informal rural health care providers is an effective means of improving provider medical knowledge and practice in the short run. It is worth noting that this study will not be able to capture long run effects , such as price or location changes, on health care for the rural poor.

The evaluation of the Liver Foundation training program for rural health care providers consists of four phases. In the first phase, the Liver Foundation will comb blocks in Birbhum district and invite 300-400 rural health care providers to participate in the program. In the second phase, all recruited providers will convene at a center in Birbhum where we, the evaluation team JPAL, will conduct a baseline study. We will then randomly allocate participating providers into two groups: the Treatment group, which will receive the Liver Foundation training within the next twelve months, and the Control group, which will not. In the third phase, the program inductees will be trained according to the curriculum put forth by the Liver Foundation. Finally, in the fourth phase, we will run an extensive endline survey to evaluate how training has affected the (1) clinical practice, (2) clinical competence, and (3) prices charged and caseload of the providers who have received the training relative to their non-trained counterparts. We detail each phase in turn. Phase I: Provider Census in Birbhum District The Liver Foundation will identify between 300 to 400 representative private rural health care providers within the Birbhum district who are interested in the training program provided by the Liver Foundation and willing to participate in the necessary randomized selection process. All providers will be told that if they attend the baseline study, they will receive the medical training; however, how soon they will receive it will be determined through a lottery. Phase II: Baseline Study and Randomization of Providers into Training Intervention Bringing the providers in batches to a center established for this purpose in Birbhum, the evaluation team will run a baseline study on all 300 participants over the course of three weeks. The baseline study will consist of two parts: Baseline provider survey: this survey asks basic questions about training, socioeconomic background, work practices, etc. Vignette testing: Teams of trained vignette testers will test each doctor's competence through a series of five medical vignettes. After having completed the above two steps of the baseline survey, the evaluation team will randomly allocate these providers into two groups: the "Treatment" group, which will receive the training, and the "Control" group, which will not. Note that those individuals selected for the control group will be informed that they will receive training at a later time due to limited resources; thus we will phase them in one year later, after the evaluation of the first treatment group is complete. Phase III: Training Intervention The primary objective of the Liver Foundation's training program will be capacity building through theoretical information in health sciences, training in proper referral techniques during emergencies, and maternal and child health care priorities in rural areas. The training will be in the form of two three-hour classes per week and will continue for nine months, with three mid-term exams administered every three months as per the Liver Foundation curriculum. During this time, the evaluation team will only collect the rosters of the attendees of the training classes and conduct sporadic random visits to the sessions to gain a better sense of the training methods and attendance rates. Phase IV: Endline survey After the intervention has run its course for at least nine months, we will wrap up the evaluation with an extensive endline survey. Unlike the baseline survey, which can be completed in the one central location at which the providers meet, the endline survey must be done at each providers' own place of work. We expect that the process of locating and evaluating each of the 300 providers at his or her place of work will require approximately three months.

Of the 304 participants in the study, half will be randomized into the treatment arm in which they will receive training from the Liver Foundation.The objective of the Liver Foundation's training program will be capacity building through information in health sciences, training in referral techniques during emergencies, and maternal and child health care priorities in rural areas. The training will be in the form of two three-hour classes per week and will continue for nine months, with three mid-term exams administered every three. During this time, the evaluation team will only collect the rosters of the attendees of the training classes and conduct sporadic random visits to the sessions to gain a better sense of the training methods and attendance rates.

Half of the sample will be randomized into the control group, for whom there will be no training. They will be phased into training the following year. For the current year, the control group will experience no difference in their routine, aside from some (infrequent) opportunities to participate in other public health activities as provided by the Liver Foundation to any provider involved in their various programs.

This group will not receive any training from the Liver Foundation; they will simply be evaluated at baseline and endline.

Clinical Practice will be assessed in two ways. First, we will send Standardized Patients, or trained professionals who present with a pre-developed set of symptoms (of one of the three conditions specified) to a provider without the provider's knowledge that he/she is acting. Second, the evaluating team will sit in the providers' clinics after the completion of the intervention, observing provider-patient interactions for a full day. Information will be obtained for each provider-patient interaction on (a) consultation length, (b) history taking, (c) examinations performed, (d) information given to the patient, (e) medicines dispensed/prescribed and (f) prices charged.

Clinical competence will be assessed through the use of medical vignettes, which have been developed and used in a number of countries. In these medical vignettes, providers are presented with a series of symptoms related to a particular disease or medical condition and are evaluated on their ability to diagnose correctly and make the proper recommendations. The impact of the program in terms of clinical competence will be computed using a difference-in-difference estimator, which is the gain in the treatment group minus the gain in the control group in the completion of checklist items and diagnosis rates for the cases considered.

Clinical prices, or the consultation fees charged to patients, and caseload, or the number of patients a health practitioner sees per day, will be assessed using both direct observation and the standardized patients. In both cases, prices will be noted and recorded to assess the impact of the treatment.

Inclusion Criteria: must be a private rural healthcare practitioner in West Bengal Exclusion Criteria: cannot live outside of three specified districts of West Bengal