van Dijk Liset, Heerdink Eibert R, Somai Dinesh, van Dulmen Sandra, Sluijs Emmy M, de Ridder Denise T, Griens Anna M G F, Bensing Jozien M
NIVEL, Netherlands Institute for Health Services Research, Utrecht, The Netherlands.
BMC Health Serv Res. 2007 Apr 10;7:51. doi: 10.1186/1472-6963-7-51.
Many patients experience difficulties in following treatment recommendations. This study's objective is to identify nonadherence risk profiles regarding medication (antidepressants, antihypertensives, and oral hypoglycemics) from a combination of patients' socio-demographic characteristics, morbidity presented within general practice and medication characteristics. An additional objective is to explore differences in nonadherence among patients from different general practices.
Data were obtained by linkage of a Dutch general practice registration database to a dispensing registration database from the year 2001. Subjects included in the analyses were users of antidepressants (n = 4,877), antihypertensives (n = 14,219), or oral hypoglycemics (n = 2,428) and their GPs. Outcome variables were: 1) early dropout i.e., a maximum of two prescriptions and 2) refill nonadherence (in patients with 3+ prescriptions); refill adherence < 80% was considered as nonadherence. Multilevel modeling was used for analyses.
Both early dropout and refill nonadherence were highest for antidepressants, followed by antihypertensives. Risk factors appeared medication specific and included: 1) non-western immigrants being more vulnerable for nonadherence to antihypertensives and antidepressants; 2) type of medication influencing nonadherence in both antihypertensives and antidepressants, 3) GP consultations contributing positively to adherence to antihypertensives and 4) somatic co-morbidity influencing adherence to antidepressants negatively. There was a considerable range between general practices in the proportion of patients who were nonadherent.
No clear risk profiles for nonadherence could be constructed. Characteristics that are correlated with nonadherence vary across different types of medication. Moreover, both patient and prescriber influence adherence. Especially non-western immigrants need more attention with regard to nonadherence, for example by better monitoring or communication. Since it is not clear which prescriber characteristics influence adherence levels of their patients, there is need for further research into the role of the prescriber.
许多患者在遵循治疗建议方面存在困难。本研究的目的是从患者的社会人口学特征、全科医疗中呈现的发病率以及药物特征的组合中,识别关于药物(抗抑郁药、抗高血压药和口服降糖药)的不依从风险概况。另一个目的是探讨来自不同全科医疗的患者在不依从方面的差异。
通过将荷兰全科医疗注册数据库与2001年的配药注册数据库相链接来获取数据。纳入分析的受试者为抗抑郁药使用者(n = 4877)、抗高血压药使用者(n = 14219)或口服降糖药使用者(n = 2428)及其全科医生。结局变量为:1)早期停药,即最多两张处方;2)续方不依从(处方数≥3张的患者);续方依从性<80%被视为不依从。采用多水平模型进行分析。
抗抑郁药的早期停药和续方不依从率均最高,其次是抗高血压药。风险因素似乎因药物而异,包括:1)非西方移民在抗高血压药和抗抑郁药不依从方面更易受影响;2)药物类型影响抗高血压药和抗抑郁药的不依从情况;3)全科医生会诊对抗高血压药的依从性有积极作用;4)躯体合并症对抗抑郁药的依从性有负面影响。不同全科医疗中不依从患者的比例存在相当大的差异。
无法构建明确的不依从风险概况。与不依从相关的特征因不同类型的药物而异。此外,患者和开处方者都会影响依从性。特别是非西方移民在不依从方面需要更多关注,例如通过更好的监测或沟通。由于不清楚哪些开处方者特征会影响其患者的依从水平,因此需要进一步研究开处方者的作用。