Secondary Logo

Institutional members access full text with Ovid®

Share this article on:

Overachieving Municipalities in Public Health: A Machine-learning Approach

Chiavegatto Filho, Alexandre Dias Portoa,b; dos Santos, Hellen Geremiasa; do Nascimento, Carla Ferreiraa; Massa, Kaioa; Kawachi, Ichirob

doi: 10.1097/EDE.0000000000000919
Social Epidemiology

Background: Identifying successful public health ideas and practices is a difficult challenge towing to the presence of complex baseline characteristics that can affect health outcomes. We propose the use of machine learning algorithms to predict life expectancy at birth, and then compare health-related characteristics of the under- and overachievers (i.e., municipalities that have a worse and better outcome than predicted, respectively).

Methods: Our outcome was life expectancy at birth for Brazilian municipalities, and we used as predictors 60 local characteristics that are not directly controlled by public health officials (e.g., socioeconomic factors).

Results: The highest predictive performance was achieved by an ensemble of machine learning algorithms (cross-validated mean squared error of 0.168), including a 35% gain in comparison with standard decision trees. Overachievers presented better results regarding primary health care, such as higher coverage of the massive multidisciplinary program Family Health Strategy. On the other hand, underachievers performed more cesarean deliveries and mammographies and had more life-support health equipment.

Conclusions: The findings suggest that analyzing the predicted value of a health outcome may bring insights about good public health practices.

From the aDepartment of Epidemiology, School of Public Health of the University of Sao Paulo, Sao Paulo, SP, Brazil

bDepartment of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, Boston, MA.

Submitted February 19, 2018; accepted September 6, 2018.

Data and code are available in eAppendix9.

Supported by a grant from the Lemann Foundation (Harvard Brazil Research Fund) and FAPESP (grant number: 17/09369-8).

The authors report no conflicts of interest.

Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com).

Ma and Lin have contributed equally to this article.

Correspondence: Alexandre D. P. Chiavegatto Filho, Department of Epidemiology, School of Public Health, University of Sao Paulo, 715 Av Dr Arnaldo, Sao Paulo, SP, Brazil 01246-904. E-mail: alexdiasporto@usp.br.

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.