The Application Of Drug Sales Surveillance System In Rural Area Of Hubei Province

Objectives1. To understand the status of self-medication behavior and attitude of people in rural areas of Hubei province, investigate the baseline information of pharmacies, the status of the drug sales management and willingness of staffs in pharmacies for drug sales surveillance in rural areas of Hubei province, to explore the feasibility of the establishment of drug sales surveillance system in rural areas.2. To establish framework for drug sales surveillance system, develop an electronic drug sales surveillance system for infectious disease, establish a drug sales surveillance system in rural areas, and improve the early warning capacity of infectious diseases in the rural areas.3. To explore the most suitable early outbreak detection algorithms and its optimum threshold for drug sales surveillance, compare the timeliness of drug sales data source and outpatient volume datasource in health centers, and evaluate the early warning capacity of drug sales surveillance system for infectious diseases.Methods1. Purposively sample strategy and stratified sample strategy method was used in the study. Self-designed questionnaire was used to investigate the status of self-medication behavior and attitude of rural residents, basic information of pharmacies and pharmacy staffs’willingness to participate the drug sales surveillance project.2. Establishing the framework for durg sales surveillance system based on literature review and experts consulting meetings. Selecting the most common used drugs as surveillance drugs by analyzing the historical sales data, and then developing the Intergarted Surveillance System (ISS system) which contains an electronic drug sales surveillance system cooperated with Fudan University and the FPX Company. Conducting drug sales surveillance in Qianjiang and Shayang county in Hubei province with the established system. Analyzed drug sales data reported in the ISS system from April1, to December31,2012.3. Sales data of respiratory related drugs and influenza-like illness cases in Qianjiang was collected by ISS system from April1, to December31,2012. Comparing the moving average method, exponentially weighted moving average method, and cumulative sum control chart method with simulated data. Correlations between the daily sales data of respiratory related drug and daily outpatient volume of influenza-like illness cases visited the health center was calculated in different time lags to compare the timeliness and evaluate the early warning capability of drug sales surveillance system.Results1. More than60%rural residents who suffered less severe symptoms would choose self-medication immediately or after1-3days if there were no alleviation of symptoms.100%county pharmacies has computer, and82.35%of them have access to the internet.71.79%township pharmacies equipped with computers, the internet penetration rate was48.72%.94.12%county pharmacies and51.28%township pharmacies had electronic drug sales management system.16.67%pharmacy staffs were skilled on computer operation and use a variety of office software,62.50%of them were on the general level of computer operation.79.17%of them thought that carring out the drug sales surveillance was meaningful,41.67%of them thought it was very meaningful for infectious diseases surveillance and control.87.50%of them would like to participate in the drug sales surveillance project. Reasons for dislike to participate in the project includes:worring about the leak of business information, involving heavy workload, lacking of technical equipment, difficulty to perform, fear of failing in the project work, lacking of bonus to support enthusiasm for work and related policies support.2.103,780records had been reported by23pharmacies from April1st to December31st,2012.59,855of which were reported by county pharmacies. The total missing report rate was1.69%, zero report rate was1.94%, incidence report rate was96.38%, timely report rate was86.18%. The consistency rates of categories and sales volume reported in the first consistency check was63.19%and59.72%respectively. In the second consistency check, the rate was85.96%and79.21%respectively.288,835.5packing units were reported,51.5packing units were reported per pharmacy per day on average. Compound cold medicines, cough suppressants and antibiotics were main categories reported in the system; compound cold medicines accounted more than40%of total drug sales volume.3. The weighted moving average method and cumlative sum control chart-C3method is more suitable for early warning applicated on drug sales surveillance system. With the most suitable parameters and thresholds, the sensitivity of moving average method was100%, the specificity was93.60%, and the Youden’s index was0.94, the sensitivity of C3was100%, the specificity was93.60%, and the Youden’s index was0.94. The drug sales surveillance system could detect outbreak2days earlier than outpatient surveillance in health centers.Conclusions1. The rural areas have the basic hardware facilities to carry out the drug sales surveillance, pharmacy staffs would like to participate in drug sales surveillance. Also they are able to operate the computer. So, it is possible to carry out drug sales surveillance in rural areas.2. The study successfully constructed and implemented a drug sales surveillance system in the rural areas of Hubei Province. The completeness, timeliness, accuracy and consistency of reported data was higher, and the quality of data was reliable. It was indicated that there were seasonal fluctuations on the trend of daily drug sales volume, the drug sales surveillance could be used for early detection of infectious disease outbreaks.3. The exponentially weighted moving average method and cumlative sum control chart-C3method is more suitable for early warning in drug sales surveillance system. Drug sales surveillance system has higher timeliness than outpatient surveillance in health centers.Innovations1. We firstly applied the drug sales surveillance system in rural area, study the feasibility of drug sales surveillance in rural areas, and established a drug sales surveillance system in rural Hubei successfully.2. Assessed the moving average model, exponentially weighted moving average model and CUSUM control chart method in drug sales surveillance with simulated data.3. Evaluate the early warning capacity of drug sales surveillance system in rural areas for the first time.