Jensen Rikke Krüger, Lauridsen Henrik Hein, Andresen Andreas Duch Kiilerich, Mieritz Rune Mygind, Schiøttz-Christensen Berit, Vach Werner
Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark.
Nordic Institute of Chiropractic and Clinical Biomechanics, Odense, Denmark.
Clin Epidemiol. 2020 Aug 19;12:891-905. doi: 10.2147/CLEP.S263646. eCollection 2020.
To develop a self-administered diagnostic screening questionnaire for lumbar spinal stenosis (LSS) consisting of items with high content validity and to investigate the diagnostic value of the questionnaire and the items.
A self-reported diagnostic LSS screening questionnaire was developed based on items from the existing literature describing key symptoms of LSS. The screening questionnaire (index test) was to be tested in a cohort of patients with persistent lumbar and/or leg pain recruited from a Danish publicly funded outpatient secondary care spine clinic with clinicians performing the reference test. However, to avoid unnecessary collection of data if the screening questionnaire proved to be of limited value, a case-control design was incorporated into the cohort design including an interim analysis. Additional cases for the case-control study were recruited at two Danish publicly funded spine surgery departments. Prevalence, sensitivity, specificity and diagnostic odds ratio (OR) were calculated for each individual item, and AUC (area under the curve) was calculated to examine the performance of the full questionnaire.
A 13-item Danish questionnaire was developed and tested in 153 cases and 230 controls. The interim analysis was not in favour of continuing the cohort study, and therefore, only results from the case-control study are reported. There was a positive association for all items except the presence of back pain. However, the association was only moderate with ORs up to 3.3. When testing the performance of the whole questionnaire, an AUC of 0.72 was reached with a specificity of 20% for a fixed sensitivity of 95%.
The items were associated with LSS and therefore have some potential to identify LSS patients. However, the association was not strong enough to provide sufficient accuracy for a diagnostic tool. Additional dimensions of symptoms of LSS need identification to obtain a reliable questionnaire for screening purposes.
开发一份具有高内容效度项目的腰椎管狭窄症（LSS）自我管理诊断筛查问卷，并调查该问卷及其项目的诊断价值。
基于现有文献中描述LSS关键症状的项目，制定了一份自我报告的LSS诊断筛查问卷。该筛查问卷（指标测试）将在一组患有持续性腰腿痛的患者中进行测试，这些患者是从丹麦公共资助的门诊二级护理脊柱诊所招募的，由临床医生进行参考测试。然而，为了避免在筛查问卷被证明价值有限时不必要的数据收集，在队列设计中纳入了病例对照设计，包括中期分析。病例对照研究的额外病例在丹麦两个公共资助的脊柱外科部门招募。计算每个单独项目的患病率、敏感性、特异性和诊断比值比（OR），并计算曲线下面积（AUC）以检验完整问卷的性能。
制定了一份包含13个项目的丹麦语问卷，并在153例病例和230例对照中进行了测试。中期分析不支持继续进行队列研究，因此，仅报告病例对照研究的结果。除了背痛的存在外，所有项目都存在正相关。然而，这种相关性仅为中等程度，OR值最高为3.3。在测试整个问卷的性能时，在固定敏感性为95%的情况下，AUC达到0.72，特异性为20%。
这些项目与LSS相关，因此有一定潜力识别LSS患者。然而，这种相关性不够强，无法为诊断工具提供足够的准确性。需要识别LSS症状的其他维度，以获得一份用于筛查目的的可靠问卷。
