Dybdahl Torben, Andersen Morten, Søndergaard Jens, Kragstrup Jakob, Kristiansen Ivar Sønbø
Research Unit of General Practice, Institute of Public Health, University of Southern Denmark, Winsløwparken 19, 3rd, 5000, Odense C, Denmark.
Eur J Clin Pharmacol. 2004 Nov;60(9):667-72. doi: 10.1007/s00228-004-0797-1. Epub 2004 Oct 21.
To analyse associations between indicators for adoption of new drugs and to test the hypothesis that physicians' early adoption of new drugs is a personal trait independent of drug groups.
In a population-based cohort study using register data, we analysed the prescribing of new drugs by Danish general practitioners. Angiotensin-II antagonists, triptans, selective cyclo-oxygenase-2 antagonists and esomeprazol were used in the assessment. As indicators of new drug uptake, we used adoption time, cumulative incidence, preference proportion, incidence rate and prescription cost and volume. For each measure, we ranked the general practices. Ranks were pair-wise plotted, and Pearson's correlation coefficient ( r) was calculated. Next, we analysed the correlation between ranks across different drug classes.
For all indicators, the general practitioners' adoption of one group of drugs was poorly associated with adoption of others ( r</=0.49), indicating that early adoption of one type of drugs is not associated with early adoption of another. For all drug groups, adoption time adjusted for practice size was only weakly associated with other indicators ( r: -0.56 to -0.27). Indicators, based on cost and volume of drugs, were highly correlated ( r: 0.96-0.99), and the others correlated reasonably well ( r: 0.51-0.91).
Within drug groups, indicators of drug adoption, except for adoption time, correlate reasonably well. However, the theory that physicians' early adoption of new drugs is a personal trait independent of the type of drug could not be confirmed. The notion of the early-drug-adopting general practitioner may be mistaken.
分析新药采用指标之间的关联，并检验医生早期采用新药是一种独立于药物组别的个人特质这一假设。
在一项基于人群的队列研究中，我们利用登记数据对丹麦全科医生的新药处方进行了分析。评估中使用了血管紧张素II拮抗剂、曲坦类药物、选择性环氧化酶-2拮抗剂和埃索美拉唑。作为新药采用的指标，我们使用了采用时间、累积发病率、偏好比例、发病率以及处方成本和数量。对于每一项指标，我们对全科医疗进行了排名。将排名进行两两绘制，并计算皮尔逊相关系数（r）。接下来，我们分析了不同药物类别之间排名的相关性。
对于所有指标，全科医生对一组药物的采用与对其他药物的采用之间关联较弱（r≤0.49），这表明早期采用一种药物与早期采用另一种药物无关。对于所有药物组，根据医疗规模调整后的采用时间与其他指标仅存在微弱关联（r：-0.56至-0.27）。基于药物成本和数量的指标高度相关（r：0.96 - 0.99），其他指标之间的相关性也较好（r：0.51 - 0.91）。
在药物组内，除采用时间外，药物采用指标之间的相关性较好。然而，医生早期采用新药是一种独立于药物类型的个人特质这一理论未能得到证实。早期采用药物的全科医生这一概念可能是错误的。
