【作者】 涂晓晨；

【导师】 卢祖洵；

【作者基本信息】 华中科技大学 ， 社会医学与卫生事业管理， 2013， 博士

【摘要】 研究目的本研究主要针对中国目前合理用药研究的不足以及全球合理用药研究关注的焦点，对武汉市某三级甲等医院全年门诊患者电子处方信息进行分析，从而为了解大型综合医院合理用药现状提供可靠的有代表性参考数据。研究方法本研究首先通过文献研究了解国内外合理用药现状，掌握世界卫生组织推荐的WHO/INRUD处方指标分析方法以及ATC/DDD药物利用分析方法，在此基础上抽取某三级甲等医院2012年近百万门诊患者的电子处方信息，对医院门诊病人处方用药情况进行总体分析。同时对上呼吸道感染、5岁以下儿童腹泻、高血压和糖尿病四种常见病处方用药情况进行分病种针对性分析。本研究主要分析指标为门诊患者单次就诊用药个数、处方费用、注射剂使用比例、抗生素使用比例、激素使用比例、抗病毒药物使用比例以及抗高血压和降糖药物的使用频度等。在统计分析过程中，使用均数、标准差、中位数、构成比及率等对门诊患者基本特征和处方指标进行描述性分析，根据患者基本特征和就诊医生职称不同采用2检验、 t检验等对各处方指标进行分组比较，最后通过多因素Logistic回归模型分析门诊患者单次就诊用药个数和处方费用的主要影响因素。本研究中所有相关数据处理工作都应用SAS9.2软件进行统计分析。结果1、门诊患者总体用药情况。本研究主要对990029位门诊患者处方用药情况进行统计分析，患者平均年龄为38.77，标准差为19.97，男性和女性患者分别占44.79%和55.21%，自费和医保患者分别占95.45%和4.55%。门诊患者就诊医务人员职称分布主要为主任医师、副主任医师及主治医师，分别占31.76%、31.51%及31.77%。门诊患者单次就诊平均用药个数为2.50，抗生素使用比例为24.31%，激素使用比例为17.33%，注射剂使用比例为23.70%，单次就诊平均处方费用为581.14元，中位数为293.4元。患者性别、年龄、付费方式、就诊科室和医生职称不同，其处方指标有明显差异。多因素Logistic回归分析显示，与女性患者相比，男性患者单次就诊发生高处方费用和过多联合用药的可能性增加，OR值分别为1.52和1.09。随着年龄的升高，门诊患者发生高处方费用和过多联合用药的可能性增加。与医保患者相比，自费患者单次就诊发生高处方费用和过多联合用药的可能性增加，OR值分别为2.00和1.52。不同科室之间，内科门诊患者发生高处方费用和过多联合用药的可能性最大。不同职称医生之间，副主任医师开具的处方发生高处方费用和过多联合用药的可能性最大。2、门诊上呼吸道感染患者用药情况。本研究中共有31909位门诊上呼吸道感染患者，门诊患者单次就诊平均用药个数为2.56，抗生素使用比例为64.05%，激素使用比例为16.62%，抗病毒药物使用比例为10.45%，注射剂使用比例为43.89%，单次就诊平均处方费用为254.25元，中位数为171.40元。患者性别、年龄、付费方式和就诊医生职称不同，其处方指标有明显差异。3、门诊5岁以下儿童腹泻患者用药情况。本研究中共有1749位5岁以下儿童腹泻患者，门诊患者单次就诊平均用药个数为2.52，抗生素使用比例为25.52%，抗病毒药物使用比例为5.91%，注射剂使用比例为25.50%，单次就诊平均处方费用为102.29元，中位数为79.80元。患者年龄和就诊医生职称不同，其处方指标有明显差异。不同性别和付费方式患者的处方指标差别不明显。4、高血压患者用药情况。本研究中共有28788位高血压患者，门诊患者单次就诊平均用药个数为3.28，单次就诊平均处方费用为530.84元，中位数为370.55元。研究中共有6大类40种抗高血压药物，钙拮抗剂、肾素-血管紧张素系统抑制剂和β-受体阻滞剂是最常用的三类抗高血压药物，使用频率分别为35.85%、30.28%和20.07%。氨氯地平、非洛地平、硝苯地平、厄贝沙坦和培哚普利是使用频度最高的5种抗高血压药物，其限定日费用在40种抗高血压药物中的排序分别为21、23、27、18和22，属于经济适中类药物。5、糖尿病患者用药情况。本研究中共有8986位糖尿病患者，门诊患者单次就诊平均用药个数为2.88，单次就诊平均处方费用为701.02元，中位数为491.1元。研究中共有7大类18种降糖类药物，胰岛素、磺脲类促分泌药和葡萄糖苷酶抑制剂是最常用的三类降糖药物，使用频率分别为26.43%、18.27%和15.47%。胰岛素注射液、格列美脲、格列齐特、西格列汀和二甲双胍是使用频度最高的5种降糖药物，其限定日费用在18种降糖药物中的排序分别为4、17、16、9和14，属于经济适中类药物。结论和建议门诊患者单次就诊平均处方费用过高，处方用药指标与合理水平尚有差距，医师职称类型和患者付费方式对医生处方行为有显著影响，上呼吸道感染患者治疗过程中存在过度使用抗生素和激素现象，小儿腹泻、高血压和糖尿病治疗基本符合临床治疗指南。有必要扩大国家基本药物制度在三级医院的实施范围，降低患者门诊疾病经济负担。医院应该对医师处方行为进行定期监督评价，尤其是对上呼吸道感染患者处方重点关注，采取相应干预措施控制药物不合理使用。另外，有必要以前瞻性设计现场调查的方式对门诊患者处方指标进行研究。创新和局限性首先，本次研究对象选择和研究内容具有创新性。合理用药是目前卫生政策领域研究的重点，本研究是国内首次抽取年门诊量超过百万的三级甲等特大规模医院整年的门诊患者电子处方信息进行全面的统计分析。其次，研究方法和研究视角具有创新性。本研究在综合分析门诊患者处方合理用药的基础上首次同时对多种常见病用药情况分病种研究。另外，与已有研究只分析处方用药不同，本研究首次立足大样本数据对医生处方行为的影响因素进行了分析。同时，本研究还发现了一些比较有意义的创新性结论，如本研究首次发现了在三级甲等医院，副主任医师开具处方中处方用药个数以及药品费用明显高于其他医师。另外医保患者处方用药个数和费用远低于自费患者的发现也为进一步了解医生处方行为提供了新的研究思路和依据。研究不足和局限性。首先，本研究的现场具有一定的特殊性，研究结论在推广到其他医院时需要考虑到医院级别和规模对数据结果的影响。另外，本研究为回顾性分析，由于缺少患者病史、疾病严重程度等重要数据，在分析处方用药是否合理时只能根据临床指南从宏观层面加以评价，而不能针对单张处方进行具体分析。最后由于缺少患者服药时间等信息，本研究在计算ATC/DDD药物利用指标是只能初步计算用药频度，限制了进一步对DDD/100病人/天进行更详细的分析。更多还原

【Abstract】 ObjectiveRational drug use has attracted global attentions now. However, reliable data aboutrational drug use is in short supply in China. This study aims to provide reliable andrepresentative data about rational drug use in outpatients in large general hospitals byanalyzing electronic prescriptions of outpatients obtained from a tertiary level hospital inWuhan City.MethodsWe conducted comprehensive reviews of literature on rational drug use to understandthe current research situation and master the standard analysis methods recommended bythe World Health Organization (WHO). The WHO/INRUD prescription indicators analysismethod and the ATC/DDD drug utilization analysis method were applied in this study.Nearly one million electronic prescriptions of outpatient in a tertiary level hospital inWuhan City in2012were collected and analyzed. In addition, medicine use in upperrespiratory tract infections outpatients, children less than5years old with acute diarrhea,hypertensive and diabetics were analyzed separately. The main measurements used in thisstudy to assess rational drug use were average number of drugs per encounter, averageprescription cost, percentage of encounters with an injection prescribed, percentage ofencounters with an antibiotic prescribed, percentage of encounters with a hormoneprescribed, percentage of encounters with an antiviral drug prescribed, frequency ofantihypertensive and hypoglycemic agents prescribed, and so on. In the present study, descriptive analysis was carried out for demographics data and prescribing indicators.2Test, Student’s t-test and ANOVA were conducted to compare prescribing indicatorsbetween groups due to different outpatient socio-demographic characteristics and doctortitle. Multivariate logistic regression analysis was used to analyze determinant factors ofaverage number of drugs per encounter and average prescription cost. SAS9.2was used forall data analysis.Results1. Medicine use in all outpatients.990,029outpatients’ prescriptions were analyzed in the study. The average age of alloutpatients was38.77, standard deviation was19.97. Of the outpatients,44.79%weremales.95.45%were private patients. Outpatients were mainly treated by attending doctors,associate chief physicians and chief physicians; the proportion was31.76%,31.51%and31.77%, respectively. The average number of drugs per prescription was2.50, thepercentage of injection prescriptions was23.70%, the percentage of prescriptionscontaining antibiotics or hormone was24.31%and17.33%, respectively. The averageprescription cost was581.14and the median was293.4. Prescribing patterns, in terms ofaverage prescription cost, the number of drugs per prescriptions and the percentage ofprescriptions with antibiotics, injections, or hormones, were significantly different due tooutpatient’s age, gender, modes of payment, clinical departments and doctors’ titles. Theresults of the multivariate logistic regression analysis showed that males were more likelyto be prescribed with high cost prescriptions and to have excessive drug combination usethan females, the OR was1.52and1.09. The probability of being prescribed with high costprescriptions and to have excessive drug combination use increased among the outpatientswho are older. Compared with outpatients insured by medical insurance, private patientswere greater likelihood to be prescribed with high cost prescriptions and to have excessivedrug combination use. The OR was2.00and1.52. In terms of clinical departments,outpatients of the internal medicine were more likely to be prescribed with high costprescriptions and to have excessive drug combination use than that in other departments.And associate chief physicians had greater odds to prescribe high cost prescriptions and to use excessive drug combination than their counterparts.2. Medicine use in outpatients with upper respiratory tract infections31,909prescriptions of outpatients with upper respiratory tract infection were analyzedin the study. The average number of drugs per prescription was2.56, the percentage ofprescriptions with antibiotics, hormones, antiviral agents and injections were64.05%,16.62%,10.45%and43.89%, respectively. The average prescription cost was254.25andthe median was171.40. Prescribing patterns were significantly different due to outpatient’sage, gender, modes of payment, and doctors’titles.3. Medicine use in children less than5years old with acute diarrheaA total of1,749children under five years old suffered from diarrhea in the study. Theaverage number of drugs per prescription was2.52, the percentage of prescriptions withantibiotics, antiviral agents and injections were25.52%,5.91%and25.50%, respectively.The average prescription cost was102.29and the median was79.80. With regard to variousprescribing indicators, there were significant differences among outpatients with differentage and different doctors’ titles. While there were no obvious statistics difference betweenmales and females and among outpatients with different modes of payment.4. Medicine use in outpatient with hypertension28,788prescriptions of outpatients with hypertension were analyzed in the study. Theaverage number of drugs per prescription was3.28. The average prescription cost was530.84and the median was370.55. There were a total of40kinds of antihypertensive drugsin six categories. Calcium antagonists, Rennin-angiotensin system inhibitors and β-receptorblockers were most commonly used, with the frequency of utilization were35.85%,30.28%and20.07%, respectively. Amlodipine, Felodipine, Nifedipine, Irbesartan andPerindopril were the top five kinds of antihypertensive agents which were most frequentlyused, and their rankings of defined daily cost (DDDc) among40kinds of antihypertensivedrugs were21,23,27,18and22, respectively. Thus, the five most frequently usedantihypertensive agents were economic affordable.5. Medicine use in outpatients with diabetes 8,986prescriptions of outpatients with diabetes were analyzed in the study. Theaverage number of drugs per prescription was2.88. The average prescription cost was701.02and the median was491.1. There were a total of18kinds of hypoglycemic agents inseven categories. Insulin, sulfonylurea secretagogues drugs and glucosidase inhibitors weremost commonly used; with the frequency of utilization were26.43%,18.27%and15.47%,respectively. Insulin injection, Glimepiride, Gliclazide, Sitagliptin and Metformin were thetop five kinds of hypoglycemic drugs which were most frequently used, and their rankingsof defined daily cost (DDDc) among18kinds of antihypertensive drugs were4,17,16,9and14, respectively. Thus, the five most frequently used hypoglycemic agents wereeconomic affordable.Conclusions and suggestionsThe average prescription cost for an outpatient encounter is relatively too high.Prescribing indicators such as percentage use of antibiotics are still higher than therecommended level. There are significant associations between doctors’ prescribingbehaviors and their titles, as well as outpatients’ payment modes. It is worth noting thatthere is irrational usage of antibiotics and hormones among outpatients with upperrespiratory tract infection. However, the medical treatments for children less than5yearsold with acute diarrhea, hypertensive and diabetics are in line with the clinical guidelines ingeneral. It is necessary to implement essential drug system in outpatient department oftertiary level hospitals to alleviate outpatients’ economic burden. The managementdepartment of hospital should assess and supervise doctors’ prescribing behaviors regularly,especially for prescriptions of outpatients with upper respiratory tract infection, and takemeasures to control the irrational drug use. In addition, it is needed to conduct prospectivestudies to investigate outpatient prescribing indicators.Innovations and limitationsFirst of all, participants and research contents of the study is innovative. Rational druguse is a hot research field of health policy, and our study firstly and comprehensivelyanalyzed all outpatients’ electronic prescriptions collected from a tertiary level hospitalwhose outpatients exceed one million in a year. Second, the research method and perspective in this current study are innovative. In addition to comprehensively analyzerational drug use among outpatients, this study preliminary explored rational drug useamong various common diseases. Moreover, previous researches only investigated rationaldrug use, while our study explored the factors influencing doctors’ prescribing behavior onbasis of large samples as well. Meanwhile, our study had some innovative and interestingfindings. For example, associate chief physicians had greater odds to prescribe high costprescriptions and to use excessive drug combination than their counterparts. In addition, thenumber and cost of drug use of outpatients insured by medical insurance were much lessthan those of private patients, which may provide a new perspective to understand doctors’prescription behaviors.There are still some limitations needed to be pointed out. First, thestudy was carried out in a relatively particular hospital, and may not be representative ofoutpatients from different levels and scales of hospitals. Second, this study was aretrospective analysis, and patient’s disease history, severity of illness and some otherimportant data were not collected, so we could only analyze rational drug usemacroscopically under the clinical guidelines instead of analyzing a single prescription indetail. Finally, due to the lack of information on taking medicine of participants, the studyonly analyzed the frequency of drug use, a preliminary indicator of ATC/DDD system, andit limited to analyze other prescription indicators such as DDD/100patients/day further.更多还原