Estimating malaria incidence from routine health facility-based surveillance data in Uganda

Area of Science:

• Epidemiology
• Public Health
• Health Informatics

Background:

• Accurate malaria incidence measurement is crucial for tracking progress and targeting interventions.
• Health management information system (HMIS) data offer case counts but struggle with defining population denominators due to undefined catchment areas and variable care-seeking behaviors.
• This study addresses the challenge of quantifying malaria incidence using HMIS data by adjusting the population denominator based on travel time to health facilities.

Purpose of the Study:

• To estimate malaria incidence using HMIS data by adjusting the population denominator to account for travel time to health facilities.
• To compare HMIS-derived incidence estimates with gold standard measures from prospective cohorts.
• To evaluate the utility of HMIS data for tracking malaria incidence trends.

Main Methods:

• Utilized outpatient data from two Ugandan public health facilities (2011-2014).
• Modeled the relationship between patient travel time and probability of attendance using Poisson generalized additive models.
• Generated a weighted population denominator incorporating travel time and probability of attendance to estimate malaria incidence in children aged 6 months to 11 years.

Main Results:

• A total of 48,898 outpatient visits were analyzed.
• HMIS incidence correlated with cohort incidence over time (Kihihi: r=0.64; Nagongera: r=0.34).
• HMIS incidence, even with adjusted denominators, underestimated cohort incidence (e.g., Kihihi: 1.1 vs. 1.7 cases per person-year).

Conclusions:

• Malaria incidence estimated using HMIS data, even with travel time adjustments, underestimates cohort-measured incidence.
• HMIS surveillance data represent a promising and scalable resource for monitoring relative changes in malaria incidence over time.
• Incorporating village of residence data to estimate population denominators enhances the utility of HMIS for malaria surveillance.