Noben Cindy Y G, Evers Silvia M A A, Nijhuis Frans J, de Rijk Angelique E
CAPHRI School of Public Health and Primary Care, Department of Health Services Research, Faculty of Health, Medicine and Life Sciences, Maastricht University, Maastricht, The Netherlands.
BMC Public Health. 2014 Feb 4;14:115. doi: 10.1186/1471-2458-14-115.
Health impairments can result in disability and changed work productivity imposing considerable costs for the employee, employer and society as a whole. A large number of instruments exist to measure health-related productivity changes; however their methodological quality remains unclear. This systematic review critically appraised the measurement properties in generic self-reported instruments that measure health-related productivity changes to recommend appropriate instruments for use in occupational and economic health practice.
PubMed, PsycINFO, Econlit and Embase were systematically searched for studies whereof: (i) instruments measured health-related productivity changes; (ii) the aim was to evaluate instrument measurement properties; (iii) instruments were generic; (iv) ratings were self-reported; (v) full-texts were available. Next, methodological quality appraisal was based on COSMIN elements: (i) internal consistency; (ii) reliability; (iii) measurement error; (iv) content validity; (v) structural validity; (vi) hypotheses testing; (vii) cross-cultural validity; (viii) criterion validity; and (ix) responsiveness. Recommendations are based on evidence syntheses.
This review included 25 articles assessing the reliability, validity and responsiveness of 15 different generic self-reported instruments measuring health-related productivity changes. Most studies evaluated criterion validity, none evaluated cross-cultural validity and information on measurement error is lacking. The Work Limitation Questionnaire (WLQ) was most frequently evaluated with moderate respectively strong positive evidence for content and structural validity and negative evidence for reliability, hypothesis testing and responsiveness. Less frequently evaluated, the Stanford Presenteeism Scale (SPS) showed strong positive evidence for internal consistency and structural validity, and moderate positive evidence for hypotheses testing and criterion validity. The Productivity and Disease Questionnaire (PRODISQ) yielded strong positive evidence for content validity, evidence for other properties is lacking. The other instruments resulted in mostly fair-to-poor quality ratings with limited evidence.
Decisions based on the content of the instrument, usage purpose, target country and population, and available evidence are recommended. Until high-quality studies are in place to accurately assess the measurement properties of the currently available instruments, the WLQ and, in a Dutch context, the PRODISQ are cautiously preferred based on its strong positive evidence for content validity. Based on its strong positive evidence for internal consistency and structural validity, the SPS is cautiously recommended.
健康损害可导致残疾并改变工作生产力，给员工、雇主乃至整个社会带来相当大的成本。有大量工具可用于衡量与健康相关的生产力变化；然而，其方法学质量仍不明确。本系统评价对通用的自我报告工具中测量与健康相关的生产力变化的测量属性进行了批判性评估，以推荐适用于职业和经济健康实践的工具。
系统检索了PubMed、PsycINFO、Econlit和Embase数据库中的研究，这些研究需满足以下条件：（i）工具测量与健康相关的生产力变化；（ii）目的是评估工具的测量属性；（iii）工具为通用型；（iv）评分是自我报告的；（v）可获取全文。接下来，基于COSMIN要素进行方法学质量评估：（i）内部一致性；（ii）信度；（iii）测量误差；（iv）内容效度；（v）结构效度；（vi）假设检验；（vii）跨文化效度；（viii）效标效度；（ix）反应度。建议基于证据综合得出。
本评价纳入了25篇评估15种不同通用自我报告工具测量与健康相关的生产力变化的信度、效度和反应度的文章。大多数研究评估了效标效度，无一评估跨文化效度，且缺乏测量误差方面的信息。工作限制问卷（WLQ）评估最为频繁，其内容效度和结构效度分别有中等强度和强的正面证据，而信度、假设检验和反应度方面有负面证据。较少被评估的斯坦福出勤量表（SPS）在内部一致性和结构效度方面有强的正面证据，在假设检验和效标效度方面有中等强度的正面证据。生产力与疾病问卷（PRODISQ）在内容效度方面有强的正面证据，其他属性方面缺乏证据。其他工具大多质量评级为一般到较差，证据有限。
建议根据工具内容、使用目的、目标国家和人群以及现有证据做出决策。在有高质量研究准确评估现有工具的测量属性之前，基于其在内容效度方面的强正面证据，谨慎推荐WLQ；在荷兰的背景下，谨慎推荐PRODISQ。基于其在内部一致性和结构效度方面的强正面证据，谨慎推荐SPS。
