A framework for evaluating health system surveillance sensitivity to support public health decision-making for malaria elimination: a case study from Indonesia

Abstract

Background The effectiveness of a surveillance system to detect infections in the population is paramount when confirming elimination. Estimating the sensitivity of a surveillance system requires identifying key steps in the care-seeking cascade, from initial infection to confirmed diagnosis, and quantifying the probability of appropriate action at each stage. Using malaria as an example, a framework was developed to estimate the sensitivity of key components of the malaria surveillance cascade. Methods Parameters to quantify the sensitivity of the surveillance system were derived from monthly malaria case data over a period of 36 months and semi-quantitative surveys in 46 health facilities on Java Island, Indonesia. Parameters were informed by the collected empirical data and estimated by modelling the flow of an infected individual through the system using a Bayesian framework. A model-driven health system survey was designed to collect empirical data to inform parameter estimates in the surveillance cascade. Results Heterogeneity across health facilities was observed in the estimated probability of care-seeking (range = 0.01–0.21, mean ± sd = 0.09 ± 0.05) and testing for malaria (range = 0.00–1.00, mean ± sd = 0.16 ± 0.29). Care-seeking was higher at facilities regularly providing antimalarial drugs (Odds Ratio [OR] = 2.98, 95% Credible Intervals [CI]: 1.54–3.16). Predictably, the availability of functioning microscopy equipment was associated with increased odds of being tested for malaria (OR = 7.33, 95% CI = 20.61). Conclusions The methods for estimating facility-level malaria surveillance sensitivity presented here can help provide a benchmark for what constitutes a strong system. The proposed approach also enables programs to identify components of the health system that can be improved to strengthen surveillance and support public-health decision-making.

Freedom From Infection (FFI): A paradigm shift towards evidence-based decision-making for malaria elimination.

“…The analysis was conducted using Bayesian hierarchical model fitting with the program JAGS [47], interfaced with the statistical environment R [48] via the package rjags [49]. To inform the parameters in both the state and the observation processes, we used a mix of expert-based priors and empirical data combined with the semi-qualitative data collected via health centre interviews [15]. These are described in the following paragraphs and summarised in Table 1.…”

“…Seventy percent of the Indonesian population resides on Java Island, and persistent endemic malaria remains [42]. Magelang and Kulon Progo are neighbouring districts with different malaria transmission histories and surveillance policies [15]. Magelang district in the central Java province was certified malaria-free in 2014.…”

“…and modified by [52] and [15] (please refer to the latter citation for more details on the methodological approach). For the intercept 𝛼 𝑆𝐸𝐸𝐾 all the 𝛽s coefficients in equation ( 15), we chose diffuse normal priors centred at zero, corresponding to a null hypothesis of no-effect for each covariate.…”

“…The probability of testing for malaria at health facilities can vary depending on the characteristics of the national malaria monitoring protocols [53], and the intrinsic capabilities of the given health facility. Following the approach we have previously proposed [15], we modelled 𝑃 𝑇𝐸𝑆𝑇 𝑗 as 𝑙𝑜𝑔𝑖𝑡 (𝑃 𝑇𝐸𝑆𝑇 𝑗 ) = 𝛼 𝑇𝐸𝑆𝑇 + ∑ 𝛽 𝑘 𝑋 𝑗,𝑘 𝑇 𝑘=1 (16) Similar to equation (15), and based again on previous findings [15], we used as explanatory variables whether the health centre has functioning microscopy equipment (𝑚𝑖𝑐𝑟𝑜𝑠𝑐𝑜𝑝𝑦_𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛), whether functioning counting meters were available (𝑚𝑖𝑐𝑟𝑜𝑠𝑐𝑜𝑝𝑦_𝑚𝑒𝑡𝑒𝑟𝑠), whether there was a recent stockout of microscopy materials (𝑚𝑖𝑐𝑟𝑜𝑠𝑐𝑜𝑝𝑦_𝑠𝑡𝑜𝑐𝑘𝑜𝑢𝑡), whether a copy of the national malaria treatment guidelines or standard operating procedures on malaria case management were available for staff in the facility (𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡_𝑠𝑜𝑝𝑠), and whether the facility received supervisory visits from a district health officer or consultant in the last year (𝑠𝑢𝑝𝑒𝑟𝑣𝑖𝑠𝑖𝑜𝑛). In addition, to capture the seasonality of malaria cycle, we added the month (1: January to 12: December) with a cubic spline effect.…”

“…This approach involves describing the decision route of an infected person from the development of symptoms to the case being diagnosed. Each branch of the scenario tree is parameterised by the corresponding probability of an individual moving to the next stage [2,[15][16][17]. A basic tree structure would include the probability of developing symptoms, seeking care, being suspected of infection, being tested, and the test correctly diagnosing the infection [15].…”