About this Impact evaluation
More than 9 million people around the world suffer from tuberculosis (TB) each year. India has the highest incidence rate in the world. While TB is curable, the challenge is being able to detect it early and ensuring that patients comply with the 6 month treatment. The DOTS (Directly Observed Treatment Short Course) system introduced by WHO, has been designed to increase accessibility of treatment and monitoring of patients through community-based DOT centers and health workers. However, as experienced by Operation ASHA, an NGO in Delhi that runs DOTS centers, monitoring community health workers (counselors) is a difficult task. Particularly in remote areas, many studies have found problems with attendance and motivation of public health staff. The NGO has consequently designed a financial incentive scheme, where counselors are initially rewarded for detecting new patients, and after the number of detections reaches a saturation point, they get penalized for drop-outs. The incentives of course come in addition to a base salary. This study is a randomised controlled trial of the incentive scheme in slums of 15 cities in 3 states of India. The study will evaluate the causal links between incentives, counselor's commitment, performance and satisfaction with their job, patients well being and the over-all cost-effectiveness of Operation ASHA's centers.
Website : http://s01.opasha.org/
This paper investigates whether financial incentives provided to health workers may encourage them to detect new tuberculosis cases and improve treatment adherence. This paper reports experimental evidence from India on the effect of health workers' performance-based incentives on patient detection and treatment default.
The tuberculosis (TB) epidemic is recognised as one of the key development challenges by the international community. Ending it by 2030 is one of the Sustainable Development Goals adopted in 2015. With 2 million people infected, India has one of the highest incidences of TB in the world. About one thousand Indians die of TB every single day.
Cheap, widely available drugs exist to cure the vast majority of TB cases. The biggest challenge to contain the spread of the disease is to detect cases in a timely manner and ensure that patients complete the entire course of the six-month treatment. The DOTS (Directly Observed Treatment'Short course) system, which stipulates that an independent observer watch the TB patient take her pills three times a week for at least two months, has led to significant progress in the treatment of TB. Yet early detection is still hampered by stigma, lack of information and poor outreach of the health system in remote communities. Treatment adherence is made difficult by remoteness, drug side-effects, and commitment issues as symptoms of the disease disappear after a few weeks of treatment. Fighting TB in remote communities is essentially a service-delivery challenge, and health workers play a crucial role in addressing it.
This design allows to evaluate the impact incentives have on a series of outcomes including (i) detection - number of patients enrolled in the DOTS system; (ii) default - number of patients leaving the DOTS system during the course of their treatment; (iii) health workers' effort or motivation and job satisfaction; (iv) patients' characteristics, satisfaction and health status.
Outcomes of interest are measured by a combination of administrative data and about 5,000 comprehensive health workers and patient surveys
Using a field experiment conducted in the urban slums of northern India, this study examines whether financial incentives provided to social workers and agents performing multiple tasks do increase performance.
Theory of change
The basic assumption underlying this evaluation of performance-based incentives is that in the absence of constant monitoring, health worker motivation towards their work is low. Based on past studies, if the issue of health worker motivation could be resolved, then the long-term impact on tuberculosis patients in India could be substantial. The introduction of financial incentives'the primary input provided to the community heakth worker (CHW) in the treatment group'leads to higher wages for CHWs who perform better in terms of increasing detection of new tuberculosis patients and treatment compliance among existing patients.
Operation ASHA, an NGO delegated by the Government of India to operate local TB treatment centres, initially hired 75 health workers to' cater to more than 2,500 patients in five Indian states'Chhattisgarh, Delhi, Madhya Pradesh, Punjab and Uttar Pradesh.
These TB health workers were randomly assigned to one of four treatment arms. They either received (i) financial incentives based on patient detection for six months and incentives based on treatment adherence subsequently or (ii) financial incentives based on patient detection for six months and a fixed salary subsequently or (iii) a fixed salary for six months and incentives based on treatment adherence subsequently or (iv) a fixed salary for the whole duration of the experiment.
Results point to a 24 per cent increase in reported detections induced by the provision of detection-based incentives in the first phase (an additional 1.6 detections per centre per month, significant at the 10 per cent level, from a control mean of 6.6). The number of defaults however also increased over the same time period (an increase by 0.1, significant at the 5 per cent level, from a control mean of 0.1). Health workers' survey answers suggest that this could be due to health workers reallocating their effort towards the rewarded task (early detection) and to the detriment of other non-rewarded activities (treatment compliance), in line with the multitasking theory. There is no detectable impact of the default-based incentives introduced in the second phase, arguably because of the perceived difficulty to prevent defaults which makes effort less rewarding.
Qualitative interviews of a subset of these health workers and their peers further complement these results, for example by highlighting the difficulties of default prevention relative to new patient detection. These interviews further suggest the need to be attentive towards the presence of other local competing incentive schemes that may already be in place.
Implications for policy and practice
These findings suggest three policy conclusions. First, performance-based incentives may boost health worker performance and improve health outcomes. Second, incentives may not be effective in improving all outcomes. The link between effort and observable performance must be tight for health workers to respond to incentives. Third, in a multitasking environment, performance-based incentives may have undesired impacts on non-incentivised outcomes. Incentives should be carefully structured to take into account the complexity of the work expected from the workers in the public-service sector.