Parast Layla, Doyle Brian, Damberg Cheryl L, Shetty Kanaka, Ganz David A, Wenger Neil S, Shekelle Paul G
RAND Corporation, 1776 Main Street, PO Box 2138, Santa Monica, CA, 90407-2138, USA,
J Gen Intern Med. 2015 Mar;30(3):359-64. doi: 10.1007/s11606-014-3150-0. Epub 2015 Jan 7.
The expanded use of clinical process-of-care measures to assess the quality of health care in the context of public reporting and pay-for-performance applications has led to a desire to demonstrate the value of such efforts in terms of improved patient outcomes. The inability to observe associations between improved delivery of clinical processes and improved clinical outcomes in practice has raised concerns about the value of holding providers accountable for delivery of clinical processes of care. Analyses that attempt to investigate this relationship are fraught with many challenges, including selection of an appropriate outcome, the proximity of the outcome to the receipt of the clinical process, limited power to detect an effect, small expected effect sizes in practice, potential bias due to unmeasured confounding factors, and difficulties due to changes in measure specification over time. To avoid potentially misleading conclusions about an observed or lack of observed association between a clinical process of care and an outcome in the context of observational studies, individuals conducting and interpreting such studies should carefully consider, evaluate, and acknowledge these types of challenges.
在公开报告和按绩效付费应用的背景下，临床护理过程指标的广泛使用旨在评估医疗保健质量，这引发了人们想要证明此类努力在改善患者预后方面的价值的愿望。在实践中，无法观察到临床护理过程的改善与临床预后改善之间的关联，这引发了人们对要求医疗服务提供者对临床护理过程负责的价值的担忧。试图研究这种关系的分析面临许多挑战，包括选择合适的预后指标、预后指标与接受临床护理过程的时间接近程度、检测效应的能力有限、实践中预期效应量较小、由于未测量的混杂因素导致的潜在偏差以及由于测量指标规范随时间变化而产生的困难。为避免在观察性研究中就临床护理过程与预后之间观察到或未观察到的关联得出潜在的误导性结论，开展和解释此类研究的人员应仔细考虑、评估并认识到这类挑战。
