A Bayesian approach for estimating typhoid fever incidence from large-scale facility-based passive surveillance data

Abstract

Decisions about typhoid fever prevention and control are based on estimates of typhoid incidence and their uncertainty. Lack of specific clinical diagnostic criteria, poorly sensitive diagnostic tests, and scarcity of accurate and complete datasets contribute to difficulties in calculating age-specific population-level typhoid incidence. Using data from the Strategic Typhoid Alliance across Africa and Asia program, we integrated demographic censuses, healthcare utilization surveys, facility-based surveillance, and serological surveillance from Malawi, Nepal, and Bangladesh to account for under-detection of cases. We developed a Bayesian approach that adjusts the count of reported blood-culture-positive cases for blood culture detection, blood culture collection, and healthcare seeking-and how these factors vary by age-while combining information from prior published studies. We validated the model using simulated data. The ratio of observed to adjusted incidence rates was 7.7 (95% credible interval [CrI]: 6.0-12.4) in Malawi, 14.4 (95% CrI: 9.3-24.9) in Nepal, and 7.0 (95% CrI: 5.6-9.2) in Bangladesh. The probability of blood culture collection led to the largest adjustment in Malawi, while the probability of seeking healthcare contributed the most in Nepal and Bangladesh; adjustment factors varied by age. Adjusted incidence rates were within or below the seroincidence rate limits of typhoid infection. Estimates of blood-culture-confirmed typhoid fever without these adjustments results in considerable underestimation of the true incidence of typhoid fever. Our approach allows each phase of the reporting process to be synthesized to estimate the adjusted incidence of typhoid fever while correctly characterizing uncertainty, which can inform decision-making for typhoid prevention and control.

Keywords: incidence estimation; passive surveillance; reporting pyramid; typhoid fever.

FIGURE 1

Flowchart of typhoid disease and observation process, and adjustment method to estimate the true number of cases. The pyramid (left) illustrates the different steps in the observation process for reporting typhoid incidence, with details on how parameters are estimated at each step. The flowchart (right) illustrates the corresponding Bayesian framework for each step of the observation process and which datasets and variables are used for adjustment. Adjustments for blood culture sensitivity are shown in purple, the probability of receiving a blood culture test is shown in red, and the probability of seeking healthcare is shown in blue. Variable definitions: $\lambda$, typhoid incidence rate; $\varphi$, a probability estimated in the model; S, sensitivity of blood culture; B, blood culture collection; H, healthcare seeking; a, age category; c, site. Abbreviation: BC, blood culture [Colour figure can be viewed at wileyonlinelibrary.com]

FIGURE 2

Estimated typhoid incidence based on simulated data: Full model vs simplified approach. The typhoid incidence per 100 000 person‐years of observation was estimated from simulated data based on a true incidence of 1000 typhoid infections per 100 000 person‐years (dashed horizontal black line). Data were simulated for low and high probabilities of seeking healthcare, receiving a blood culture diagnostic test, and antibiotic use. Scenarios were as follows: (1) low probability of seeking care, high probability of being tested, and low prior antibiotic usage; (2) low probability of seeking care, high probability of being tested, and high prior antibiotic usage; (3) high probability of seeking care, high probability of being tested, and low prior antibiotic usage; and (4) high probability of seeking care, high probability of being tested, and high prior antibiotic usage. Each simulation was performed sampling 735; 1000; and 2000 individuals from the population for the hypothetical healthcare utilization survey. Estimated “true” values are shown for models that did (red) and did not (blue) account for variation in blood culture sensitivity and variation in typhoid incidence among those who did or did not seek care and were or were not tested [Colour figure can be viewed at wileyonlinelibrary.com]