Department of Orthopaedic Surgery Operations, University of Utah School of Medicine, 590 Wakara Way, Salt Lake City, UT, 84108, USA,
Clin Orthop Relat Res. 2013 Nov;471(11):3466-74. doi: 10.1007/s11999-013-3097-1.
In 2012, the American Orthopaedic Foot & Ankle Society(®) established a national network for collecting and sharing data on treatment outcomes and improving patient care. One of the network's initiatives is to explore the use of computerized adaptive tests (CATs) for patient-level outcome reporting.
QUESTIONS/PURPOSES: We determined whether the CAT from the NIH Patient Reported Outcome Measurement Information System(®) (PROMIS(®)) Physical Function (PF) item bank provides efficient, reliable, valid, precise, and adequately covered point estimates of patients' physical function.
After informed consent, 288 patients with a mean age of 51 years (range, 18-81 years) undergoing surgery for common foot and ankle problems completed a web-based questionnaire. Efficiency was determined by time for test administration. Reliability was assessed with person and item reliability estimates. Validity evaluation included content validity from expert review and construct validity measured against the PROMIS(®) Pain CAT and patient responses based on tradeoff perceptions. Precision was assessed by standard error of measurement (SEM) across patients' physical function levels. Instrument coverage was based on a person-item map.
Average time of test administration was 47 seconds. Reliability was 0.96 for person and 0.99 for item. Construct validity against the Pain CAT had an r value of -0.657 (p < 0.001). Precision had an SEM of less than 3.3 (equivalent to a Cronbach's alpha of ≥ 0.90) across a broad range of function. Concerning coverage, the ceiling effect was 0.32% and there was no floor effect.
The PROMIS(®) PF CAT appears to be an excellent method for measuring outcomes for patients with foot and ankle surgery. Further validation of the PROMIS(®) item banks may ultimately provide a valid and reliable tool for measuring patient-reported outcomes after injuries and treatment.
2012 年，美国矫形足踝外科学会（®）成立了一个全国性网络，用于收集和共享治疗结果数据，并改善患者护理。该网络的一项举措是探索使用计算机自适应测试（CAT）进行患者层面的结果报告。
问题/目的：我们确定 NIH 患者报告的测量信息系统（PROMIS®）（®）物理功能（PF）项目库的 CAT 是否为患者的物理功能提供了高效、可靠、有效、精确和充分涵盖的点估计。
在获得知情同意后，288 名平均年龄为 51 岁（范围 18-81 岁）的患者接受了常见足部和踝关节疾病的手术，他们完成了一项基于网络的问卷调查。通过测试管理时间来确定效率。通过人员和项目可靠性估计来评估可靠性。有效性评估包括专家审查的内容有效性和针对 PROMIS®疼痛 CAT 的结构有效性以及基于权衡感知的患者反应。通过患者物理功能水平的测量误差标准（SEM）来评估精度。仪器覆盖率基于人员-项目图。
平均测试管理时间为 47 秒。人员可靠性为 0.96，项目可靠性为 0.99。与疼痛 CAT 的结构有效性的 r 值为-0.657（p<0.001）。在广泛的功能范围内，精度的 SEM 小于 3.3（相当于 Cronbach 的 alpha 大于 0.90）。关于覆盖范围，天花板效应为 0.32%，没有地板效应。
PROMIS®PF CAT 似乎是一种测量足踝手术后患者结果的极好方法。对 PROMIS®项目库的进一步验证最终可能为测量受伤和治疗后患者报告的结果提供一种有效和可靠的工具。
