A cost-effectiveness analysis of three components of a syndromic surveillance system for the early warning of epidemics in rural China.

Background: Syndromic surveillance systems (SSSs) collect non-specific syndromes in early stages of disease outbreaks. This makes an SSS a promising tool for the early detection of epidemics. An Integrated Surveillance System in rural China (ISSC project), which added an SSS to the existing Chinese surveillance system for the early warning of epidemics, was implemented from April 2012 to March 2014 in Jiangxi and Hubei Provinces. This study aims to measure the costs and effectiveness of the three components of the SSS in the ISSC project.

Methods: The central measures of the cost-effectiveness analysis of the three components of the syndromic surveillance system were: 1) the costs per reported event, respectively, at the health facilities, the primary schools and the pharmacies; and 2) the operating costs per surveillance unit per year, respectively, at the health facilities, the primary schools and the pharmacies. Effectiveness was expressed by reporting outputs which were numbers of reported events, numbers of raw signals, and numbers of verified signals. The reported events were tracked through an internal data base. Signal verification forms and epidemiological investigation reports were collected from local country centers for disease control and prevention. We adopted project managers' perspective for the cost analysis. Total costs included set-up costs (system development and training) and operating costs (data collection, quality control and signal verification). We used self-designed questionnaires to collect cost data and received, respectively, 369 and 477 facility and staff questionnaires through a cross-sectional survey with a purposive sampling following the ISSC project. All data were entered into Epidata 3.02 and exported to Stata for descriptive analysis.

Results: The number of daily reported events per unit was the highest at pharmacies, followed by health facilities and finally primary schools. Variances existed within the three groups and also between Jiangxi and Hubei. During a 15-month surveillance period, the number of raw signals for early warning in Jiangxi province (n = 36) was nine times of that in Hubei. Health facilities and primary schools had equal numbers of raw signals (n = 19), which was 9.5 times of that from pharmacies. Five signals were confirmed as outbreaks, of which two were influenza, two were chicken pox and one was mumps. The cost per reported event was the highest at primary schools, followed by health facilities and then pharmacies. The annual operating cost per surveillance unit was the highest at pharmacies, followed by health facilities and finally primary schools. Both the cost per reported event and the annual operating cost per surveillance unit in Jiangxi in each of the three groups were higher than their counterparts in Hubei.

Conclusions: Health facilities and primary schools are better sources of syndromic surveillance data in the early warning of outbreaks. The annual operating costs of all the three components of the syndromic surveillance system in the ISSC Project were low compared to general government expenditures on health and average individual income in rural China.