|| Checking for direct PDF access through Ovid
Escalating costs and increasing pressure to improve health services have driven a trend toward contracting with the private sector to provide traditionally state-run services. Such contracting is seen as an opportunity to combine theorized advantages of contracting with the efficiency of the private sector. There is still a limited understanding of the preconditions for successful use of contracting and the resources needed for their appropriate use and sustainability. This study assesses the large-scale contracting of 294 non-governmental organizations (NGOs) for delivery of basic health services in Uttar Pradesh, a state with almost 170 million in India. Due to high rates of discontinuation or non-renewal of contracts based on poor performance in the project, a better method for selecting partners was requested. Data on characteristics of the NGOs (intake data) and performance/outcome monitoring indicators were combined to identify correlations. The results showed that NGOs selected were generally small but well-established, had implemented at least two large projects, and had more non-health experience than health experience. Bivariate regressions of outcome score on each input variable showed that training experience, proposal quality and having ‘health’ contained in the objectives of the organization were statistically significant predictors of good performance. Factors relating to financial capacity, staff qualification, previous experience with health or non-health projects, and age of establishment were not. A combined training plus proposal score was highly predictive of outcome score (β = 1.37, P < 0.001). The combined score was found to be a much better predictor of outcome scores than a total score used to select NGOs (β = 0.073, P = 0.539). The study provides valuable information from large-scale contracting. Conclusions on criteria for selecting NGOs for providing basic health care could guide other governments choosing to contract for such services.