Arthritis Res Ther. 2013 Oct 3;15(5):R144. doi: 10.1186/ar4327.
Disability and Physical Function (PF) outcome assessment has had limited ability to measure functional status at the floor (very poor functional abilities) or the ceiling (very high functional abilities). We sought to identify, develop and evaluate new floor and ceiling items to enable broader and more precise assessment of PF outcomes for the NIH Patient-Reported-Outcomes Measurement Information System (PROMIS).
We conducted two cross-sectional studies using NIH PROMIS item improvement protocols with expert review, participant survey and focus group methods. In Study 1, respondents with low PF abilities evaluated new floor items, and those with high PF abilities evaluated new ceiling items for clarity, importance and relevance. In Study 2, we compared difficulty ratings of new floor items by low functioning respondents and ceiling items by high functioning respondents to reference PROMIS PF-10 items. We used frequencies, percentages, means and standard deviations to analyze the data.
In Study 1, low (n = 84) and high (n = 90) functioning respondents were mostly White, women, 70 years old, with some college, and disability scores of 0.62 and 0.30. More than 90% of the 31 new floor and 31 new ceiling items were rated as clear, important and relevant, leaving 26 ceiling and 30 floor items for Study 2. Low (n = 246) and high (n = 637) functioning Study 2 respondents were mostly White, women, 70 years old, with some college, and Health Assessment Questionnaire (HAQ) scores of 1.62 and 0.003. Compared to difficulty ratings of reference items, ceiling items were rated to be 10% more to greater than 40% more difficult to do, and floor items were rated to be about 12% to nearly 90% less difficult to do.
These new floor and ceiling items considerably extend the measurable range of physical function at either extreme. They will help improve instrument performance in populations with broad functional ranges and those concentrated at one or the other extreme ends of functioning. Optimal use of these new items will be assisted by computerized adaptive testing (CAT), reducing questionnaire burden and insuring item administration to appropriate individuals.
残疾和身体功能（PF）结果评估在测量功能状态的下限（非常差的功能能力）或上限（非常高的功能能力）方面能力有限。我们试图确定、开发和评估新的下限和上限项目，以便更广泛和更精确地评估 NIH 患者报告的结果测量信息系统（PROMIS）的 PF 结果。
我们使用 NIH PROMIS 项目改进协议，结合专家审查、参与者调查和焦点小组方法进行了两项横断面研究。在研究 1 中，PF 能力较低的受访者评估了新的下限项目，PF 能力较高的受访者评估了新的上限项目，以评估其清晰度、重要性和相关性。在研究 2 中，我们将低功能受访者对新下限项目的难度评分与高功能受访者对新上限项目的难度评分与参考 PROMIS PF-10 项目进行了比较。我们使用频率、百分比、平均值和标准差来分析数据。
在研究 1 中，低功能（n=84）和高功能（n=90）受访者主要为白人、女性，年龄 70 岁，具有一些大学学历，残疾评分为 0.62 和 0.30。31 个新下限和 31 个新上限项目中，超过 90%的项目被评为清晰、重要和相关，其余 26 个上限和 30 个下限项目用于研究 2。研究 2 的低功能（n=246）和高功能（n=637）受访者主要为白人、女性，年龄 70 岁，具有一些大学学历，健康评估问卷（HAQ）评分为 1.62 和 0.003。与参考项目的难度评分相比，上限项目的难度评分被评定为高出 10%至 40%以上，下限项目的难度评分被评定为低出 12%至近 90%。
这些新的下限和上限项目大大扩展了身体功能的可测量范围的极限。它们将有助于提高具有广泛功能范围的人群以及功能处于一个或另一个极端的人群的仪器性能。通过计算机自适应测试（CAT），这些新的下限和上限项目将得到最佳利用，从而减少问卷负担并确保向适当的个人进行项目管理。
