In collaboration with Operation ASHA (OpASHA), an Indian NGO that operates 200 community-based DOTS (“Directly Observed Treatment Short Course”) centers in India, we propose to conduct a randomized controlled trial to compare the effectiveness of different types of financial incentives to encourage referrals from current TB patients to new suspects. We will vary (1) the conditionality of the incentives (“unconditional” and “conditional” incentives), (2) whether outreach is led by patients or OpASHA’s health workers (“patient outreach” and “patient provides names” conditions), and (3) whether the new suspects know who named them (“known referrer” and “anonymous referrer” conditions).
The “unconditional” incentive will consist of Rupees (Rs.) 150 offered to current patients for each new suspect they induce to come to the center and get tested for TB. When the reward is fixed, each current patient has an incentive to refer any person who may be willing to be tested, irrespective of whether they know or believe they have TB. The “conditional” incentive treatment will consist of Rs. 100 for each new suspect who comes to the center and gets tested, plus a Rs. 150 bonus if the new suspect tests positive for TB. Comparison of the effect of conditional and unconditional incentives will allow us to determine whether the subjects take advantage of incentives that do not depend on test results, and whether they have concrete information about their contacts’ health.
To separate the effect of current patients’ networks from their ability to deliver information and persuade new suspects to get tested, we will vary whether outreach is led by patients or OpASHA’s health workers. In the “patient outreach” treatment, current patients will be given a set of referral cards and will be told that they will receive the reward if new suspects present at the center with the card and get tested for TB; in the “patient provides names” treatment, current patients will be invited to provide names and contact information of people whom they believe should get tested, and will receive the reward if the new suspect, who will be approached by OpASHA’s health workers, comes to the center and gets tested.
Finally, there will be two versions of the “patient provides names” treatment. In the “known referrer” version, the health workers will reveal to the new suspects who named them, whereas in the “anonymous referrer” version, the name of the referrer will not be revealed to the new suspects. This will allow us to determine the extent to which social stigma is a barrier to referrals. In fact, if social costs are the reason why current patients are reluctant to approach others, anonymity should remove them resulting in more and perhaps more effective referrals.