Adams Alyce S, Zhang Fang, LeCates Robert F, Graves Amy Johnson, Ross-Degnan Dennis, Gilden Daniel, McLaughlin Thomas J, Lu Christine, Trinacty Connie M, Soumerai Stephen B
Department of Ambulatory Care and Prevention, Harvard Medical School, Harvard Pilgrim Health Care, Boston, Massachusetts, USA.
Arch Intern Med. 2009 Apr 27;169(8):750-6. doi: 10.1001/archinternmed.2009.39.
Prior authorization is a popular, but understudied, strategy for reducing medication costs. We evaluated the impact of a controversial prior authorization policy in Michigan Medicaid on antidepressant use and health outcomes among dual Medicaid and Medicare enrollees with a Social Security Disability Insurance designation of permanent disability.
We linked Medicaid and Medicare (2000-2003) claims for dual enrollees in Michigan and a comparison state, Indiana. Using interrupted time-series and longitudinal data analysis, we estimated the impact of the policy on antidepressant medication use, treatment initiation, disruptions in therapy, and adverse health events among continuously enrolled (Michigan, n = 28 798; Indiana, n = 21 769) and newly treated (Michigan, n = 3671; Indiana, n = 2400) patients.
In Michigan, the proportion of patients starting nonpreferred agents declined from 53% prepolicy to 20% postpolicy. The prior authorization policy was associated with a small sustained decrease in therapy initiation overall (9 per 10,000 population; P = .007). We also observed a short-term increase in switching among established users of nonpreferred agents overall (risk ratio, 2.88; 95% confidence interval, 1.87-4.42) and among those with depression (2.04; 1.22-3.42). However, we found no evidence of increased disruptions in treatment or adverse events (ie, hospitalization, emergency department use) among newly treated patients.
Prior authorization was associated with increased use of preferred agents with no evidence of disruptions in therapy or adverse health events among new users. However, unintended effects on treatment initiation and switching among patients already taking the drug were also observed, lending support to the state's previous decision to discontinue prior approval for antidepressants in 2003.
预先授权是一种常见但研究不足的降低药物成本的策略。我们评估了密歇根医疗补助计划中一项有争议的预先授权政策对同时参加医疗补助和医疗保险且被社会保障残疾保险指定为永久性残疾的参保者使用抗抑郁药物及健康结局的影响。
我们将密歇根州和对照州印第安纳州双重参保者的医疗补助和医疗保险（2000 - 2003年）索赔数据进行了关联。使用中断时间序列和纵向数据分析，我们估计了该政策对持续参保患者（密歇根州，n = 28798；印第安纳州，n = 21769）和新接受治疗患者（密歇根州，n = 3671；印第安纳州，n = 2400）使用抗抑郁药物、开始治疗、治疗中断及不良健康事件的影响。
在密歇根州，开始使用非首选药物的患者比例从政策实施前的53%降至政策实施后的20%。预先授权政策总体上与治疗开始的小幅持续下降相关（每10000人中有9人；P = 0.007）。我们还观察到，总体上非首选药物的现有使用者（风险比，2.88；95%置信区间，1.87 - 4.42）以及抑郁症患者（2.04；1.22 - 3.42）换药的短期增加。然而，我们没有发现新接受治疗患者的治疗中断或不良事件（即住院、急诊就诊）增加的证据。
预先授权与首选药物使用增加相关，且没有证据表明新使用者的治疗中断或不良健康事件增加。然而，也观察到了对已经在服用药物的患者开始治疗和换药的意外影响，这支持了该州此前在2003年停止对抗抑郁药物进行预先批准的决定。
