Survival analysis with semi-supervised predictive clustering trees.

Many clinical studies follow patients over time and record the time until the occurrence of an event of interest (e.g., recovery, death, …). When patients drop out of the study or when their event did not happen before the study ended, the collected dataset is said to contain censored observations. Given the rise of personalized medicine, clinicians are often interested in accurate risk prediction models that predict, for unseen patients, a survival profile, including the expected time until the event. Survival analysis methods are used to detect associations or compare subpopulations of patients in this context. In this article, we propose to cast the time-to-event prediction task as a multi-target regression task, with censored observations modeled as partially labeled examples. We then apply semi-supervised learning to the resulting data representation. More specifically, we use semi-supervised predictive clustering trees and ensembles thereof. Empirical results over eleven real-life datasets demonstrate superior or equivalent predictive performance of the proposed approach as compared to three competitor methods. Moreover, smaller models are obtained compared to random survival forests, another tree ensemble method. Finally, we illustrate the informative feature selection mechanism of our method, by interpreting the splits induced by a single tree model when predicting survival for amyotrophic lateral sclerosis patients.

Community action for health in India: evolution, lessons learnt and ways forward to achieve universal health coverage.

The role of civil society and community-based organizations in advancing universal health coverage and meeting the targets of the 2030 Agenda for Sustainable Development has received renewed recognition from major global initiatives. This article documents the evolution and lessons learnt through two decades of experience in India at national, state and district levels. Community and civil society engagement in health services in India began with semi-institutional mechanisms under programmes focused on, for example, HIV/AIDS, tuberculosis, polio and immunization. A formal system of community action for health (CAH) started with the launch of the National Rural Health Mission in 2005. By December 2018, CAH processes were being implemented in 22 states, 353 districts and more than 200 000 villages in India. Successive evaluations have indicated improved performance on various service delivery parameters. One example of CAH is community-based monitoring and planning, which has been continuously expanded and strengthened in Maharashtra since 2007. This involves regular, participatory auditing of public health services, which facilitates the involvement of people in assessing the public health system and demanding improvements. At district level, CAH initiatives are successfully reaching "last-mile" communities. The Self-Employed Women's Association, a cooperative-based organization of women working in the informal sector in Gujarat, has developed community information hubs that empower clients to access government social and health sector services. CAH initiatives in India are now being augmented by regular activities led and/or participated in by civil society organizations. This is contributing to the democratization of community and civil society engagement in health. Additional documentation on CAH and the further formalization of civil society engagement are needed. These developments provide a valuable opportunity both to improve governance and accountability in the health sector and to accelerate progress towards universal health coverage. Lessons learnt may be applicable to other countries in South-East Asia, as well as to most low- and middle-income countries.