Related Experiment Video
Updated: Jul 13, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Performance of predictive algorithms in estimating the risk of being a zero-dose child in India, Mali and Nigeria
Arpita Biswas1, John Tucker2, Sebastian Bauhoff3
1Center for Research on Computation and Society, Harvard University John A Paulson School of Engineering and Applied Sciences, Cambridge, Massachusetts, USA.
Insights
Predictive algorithms can effectively identify children missing routine vaccinations (zero-dose). This approach offers a low-cost, efficient method to target interventions and improve global vaccination equity.
Area of Science:
- Global Health
- Public Health
- Epidemiology
Background:
- Many children in low- and middle-income countries miss routine vaccinations, termed 'zero-dose' (ZD).
- Effective and efficient strategies to identify and target ZD children are lacking.
Purpose of the Study:
- To evaluate the efficacy of predictive algorithms in identifying ZD children using routine administrative data.
- To compare algorithmic targeting with traditional geographical approaches.
Main Methods:
- Supervised learning algorithms were applied to administrative data from India, Mali, and Nigeria.
- Predictor variables included health system contact and geographical data.
- Performance was assessed using specificity, sensitivity, and F1 Score, with feature importance analysis.
Main Results:
- Cost-sensitive Ridge classification demonstrated high specificity in identifying ZD children.
- Adding predictors beyond an initial set of seven did not significantly improve performance.
- Algorithmic targeting showed a 9.1% lower exclusion error than geographical targeting and identified ZD children in more areas.
Conclusions:
- Predictive algorithms can effectively and affordably identify ZD children.
- This approach can guide interventions to reduce ZD prevalence and improve vaccination equity.
Introduction:
Many children in low-income and middle-income countries fail to receive any routine vaccinations. There is little evidence on how to effectively and efficiently identify and target such 'zero-dose' (ZD) children.
Methods:
We examined how well predictive algorithms can characterise a child's risk of being ZD based on predictor variables that are available in routine administrative data. We applied supervised learning algorithms with three increasingly rich sets of predictors and multiple years of data from India, Mali and Nigeria. We assessed performance based on specificity, sensitivity and the F1 Score and investigated feature importance. We also examined how performance decays when the model is trained on older data. For data from India in 2015, we further compared the inclusion and exclusion errors of the algorithmic approach with a simple geographical targeting approach based on district full-immunisation coverage.
Results:
Cost-sensitive Ridge classification correctly classifies most ZD children as being at high risk in most country-years (high specificity). Performance did not meaningfully increase when predictors were added beyond an initial sparse set of seven variables. Region and measures of contact with the health system (antenatal care and birth in a facility) had the highest feature importance. Model performance decreased in the time between the data on which the model was trained and the data to which it was applied (test data). The exclusion error of the algorithmic approach was about 9.1% lower than the exclusion error of the geographical approach. Furthermore, the algorithmic approach was able to detect ZD children across 176 more areas as compared with the geographical rule, for the same number of children targeted.
Interpretation:
Predictive algorithms applied to existing data can effectively identify ZD children and could be deployed at low cost to target interventions to reduce ZD prevalence and inequities in vaccination coverage.
Related Concept Videos
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
01:12Relative Risk
01:29Bias in Epidemiological Studies
01:52Regression Toward the Mean
01:17z Scores and Area Under the Curve
01:12Analysis of Population Pharmacokinetic Data