摘要
目的阐明Hurdle模型原理及其在计数资料中的应用。方法利用logit-Poisson Hurdle模型分析两周内居民就诊次数的影响因素。结果 Hurdle模型logistic回归部分结果表明,性别、年龄和经济收入是决定居民是否去医院就诊的主要因素,零截尾Poisson部分结果表明两周内患病次数、由疾病而导致活动天数减少是影响两周内居民就诊次数的主要因素。结论 Hurdle模型是处理零频数过多计数资料一种有效可行的方法 ,其参数估计准确,分析结果切合实际、解释合理。
Objective To explore the application of Hurdle model in Health Service Utilization.Methods Analyse the impacting factors about the number of residents＇ treatment within two weeks applying logit-Poisson Hurdle model.Results The part of logistic revealed gender,age and income of the residents become the factors that deciding whether or not to treatment.And the part of zero-truncated Poisson indicated illness（number of illness in past 2 weeks） and actdays（number of days of reduced activity in past 2 weeks due to illness or injury） were related with the number of visit doctor in past 2 weeks.Conclusion Hurdle model was a feasible and effective method to deal with the count data with excess zeros,its parameter estimation was accurate,the analysis result was in line with reality and easy to explain.
出处
《中国卫生统计》
CSCD
北大核心
2010年第2期149-151,共3页
Chinese Journal of Health Statistics
基金
山西省自然基金项目(200911005-2)
