Segal R, Hepler C D
Am J Hosp Pharm. 1982 Nov;39(11):1891-7.
A cognitive model of the drug-prescribing process, which incorporates prescribers' attitudes (valences) about treatment outcomes and beliefs (instrumentalities) about drug effects, was tested. Twelve physicians were interviewed to identify common outcomes associated with drug therapy; six distinct outcomes were identified. Then 50 family practice residents were given a hypothetical description of a mildly hypertensive patient and three treatment choices. The physicians completed a questionnaire that identified (1) the instrumentalities of each of the alternative treatments for each of the six outcomes and (2) the valences for each outcome. The physicians were also asked to respond to an open-ended question on how they would treat the patient described in the case. Forty-six physicians proposed 39 treatments that corresponded to the three treatments for which instrumentalities and valences were available. The drug-choice model correctly predicted 28 of 39 (72%) therapeutic decisions, which was significantly more than would be expected by chance (p less than 0.01). Both valences and instrumentalities appear important for predicting prescribing intention. Pharmacists can use this model in their attempts to influence prescribing.
对一个药物处方过程的认知模型进行了测试，该模型纳入了开处方者对治疗结果的态度（效价）以及对药物效果的信念（手段）。对12名医生进行了访谈，以确定与药物治疗相关的常见结果；共确定了6种不同的结果。然后，向50名家庭医学住院医生提供了一名轻度高血压患者的假设描述以及三种治疗选择。医生们完成了一份问卷，问卷确定了（1）六种结果中每种替代治疗的手段，以及（2）每种结果的效价。还要求医生们回答一个开放式问题，即他们将如何治疗病例中描述的患者。46名医生提出了39种治疗方案，这些方案与有手段和效价值的三种治疗方案相对应。药物选择模型正确预测了39个治疗决策中的28个（72%），这显著高于随机预期（p小于0.01）。效价和手段对于预测处方意图似乎都很重要。药剂师在试图影响处方时可以使用这个模型。
