| Background: In August 2009, China formally established the National Essential Medicine System to reduce drug costs, protect the basic health care interests of the masses, and promote rational drug use. Essential medicine system could be successfully implemented or not, depending on the doctor’s prescribing practices. However, the proportion of prescribed essential medicines is still relatively low under the influence of personal factors and external environmental factors, which hinders the smooth implementation of essential medicine system. Therefore, it is necessary to analyze the essential medicines prescribing behavior objectively and scientifically, and find out the main factors influencing the essential medicines prescribing behavior of physicians, so that provides the theoretical basis for the future behavior intervention, increases the prescription rate of essential medicines, and further promotes the smooth implementation of essential medicine system.Objective: To understand the status of prescribing essential medicines among second-level hospitals in Anhui province. Find out subjective and objective factors based on the information-motivation-behavioral skills model, and construct essential medicines prescribing behavior model. To fit essential medicines prescribing behavior path graph by SEM and describe the relationship between the various factors quantitatively. Put forward corresponding countermeasures and suggestions aimed at the relevant factors affecting the behavior of essential drugs, and provide a theoretical basis for the promotion of prescribing essential medicines.Methods: The stratified random sampling method was conducted by researchers to select samples of participants. Three prefecture-level cities in Anhui, namely Bengbu, Hefei and Tong ling were randomly selected as sampling frame, and three second-level hospitals from each city were randomly selected. All the eligible doctors in each of sampled hospital were asked to complete the survey. The eligible participants had to be practicing physicians with prescription right, being present on the day when the research was administered. The Epidata3.1 was used for data entry to establish the database and the SPSS 16.0 was applied to performing the statistical analysis. AMOS 7.0 was used to examine structural equation model, and assess goodness of fit of designed IMB model.Results:(1) Chronbach’s alpha coefficient for all the measurements of IMB model was 0.891. KMO index was 0.857, P <0.05. The main factors explained 59.4% of the total cumulative variance.(2) A total of 765 questionnaires were sent out, and 732 valid questionnaires were recovered. The respond rate was 95.7%.(3) An average age of 732 participants was 37.7 years old, and the range of age was 22 to 67 years. The majority of participants were male(63.7%) and predominantly owned bachelor degree(58.7%).The surgeons, physicians, gynecologists, and pediatricians accounted for 37.3%, 27.9%, 11.6%, and 8.2%, respectively. In terms of work experience, 26.1% had experience more than twenty years, and 43.7% less than ten years. Daily average outpatient in the 50 cases following accounted for80.3%.More than half participants(52.2%) reported an income of 2,001- 4,000 Chinese Yuan.(4) The participants had access to more information about essential medicines concept and policy, and the average accuracy rate was 90.64%. The average score of essential medicines prescribing-related motivation and behavioral skillswas45.33±7.31 and 19.83±3.47, respectively.46.1% participants reported that they had prescribed essentialmedicines with the proportion of <60%in last one month.(5) The initial model was constructed based upon the IMB model and the goodness of fit index and the index was adjusted to obtain the final model. The final fit indexes were: ?2/df=4.216, GFI=0.953, CFI=0.945 and RMSEA=0.066.(6) In the final model, information directly predicted behavioral skills for 0.084, and the direct effect on essential medicines prescribing behavior was 0.040. Motivation directly predicted behavioral skills for 0.771, and the indirect effect on the essential medicines prescribing behavior was 0.365.Behavioral skills directly predicted essential medicines prescribing behavior for 0.474.Suggestions: It is urgent to carry out a variety of integrated interventions from the information, motivation, behavioral skills to promote essential medicines prescribing behavior by strengthening essential medicines and rational drug use-related education and training and actively exploring various forms of publicity to improve physician acceptance for essential medicines. The other interventions including targeting behavioral skills training, improving the ability to identify the essential medicines, encourage doctors to recommend colleagues to prescribe essential medicines, and ensuring the availability of essential medicines are needed. |