Survivor Bias in a Pooled Longitudinal Study of CTS and Related Work Disability

CTS 和相关工作障碍的汇总纵向研究中的幸存者偏差

• 批准号: 9307090
• 负责人: Carisa Harris Adamson
• 金额: $ 3.8万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-07-01 至 2017-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9307090
• 关键词: Survivor; Bias; Pooled; Longitudinal; Study

DESCRIPTION (provided by applicant): Carpal tunnel syndrome (CTS), the most common peripheral entrapment neuropathy, results from compression of the median nerve under the transverse carpal ligament at the wrist. It is an important occupational health problem because the frequency and severity of CTS cause more disability than virtually any other upper extremity disorder (Foley 2007). The ultimate goal of this research proposal is to reduce the incidence as well as the severity of CTS in working populations by elucidating the causes of both. Few prospective epidemiologic studies have the potential to assess the exposure-response relationships between biomechanical and psychosocial risk factors and either CTS or subsequent disability using rigorous case criteria and individual level exposure estimates. Even fewer are prospective with sufficient power to assess incidence and address sources of healthy worker survivor effect (HWSE) bias. HWSE has been shown to cause downward bias in occupational studies, and is a particular challenge in studies of musculoskeletal disorders such as CTS that are accompanied by discomfort (Arigghi 1994; Eisen and Robins 2001; Eisen 2005). There have been recent attempts to address the HWSE, using g-estimation in a mortality study of chronic heart and lung disease (Chevrier 2012), and inverse proportional treatment weights in a study of occupational asthma (Dumas 2013). However, there are no published longitudinal studies of CTS or subsequent work disability that address HWSE bias. In fact, other than a single cross-sectional study of garment workers (Punnett 1996), HWSE has not been taken into account when studying musculoskeletal disorders (Punnett 2004). If the workers exposed to repetitive and forceful movements consequently who develop CTS (including median nerve symptoms) are more likely to leave such work then those who remain exposed will be at lower risk. Such self-selection based on health status can reverse the causal direction and result in underestimation of the association between biomechanical risk factors, CTS, and subsequent work disability. The objective of this research is to estimate the association between biomechanical exposures, work psychosocial exposures, and CTS or CTS-related work disability while adjusting for downward bias from the HWSE. To address right truncation, inverse proportional censoring weights will be used in the analysis. To address left truncation, a sensitivity analysis will be performed by varying the date of hire criterion for entry into the pooed cohort. Adjustment for confounding from time-varying variables will utilize inverse proportional treatment weights. This approach to reducing HWSE bias has not been performed in a study of CTS or subsequent work disability because most studies of musculoskeletal disorders are not of adequate size or complexity. The findings of this research will help clarify the relationships between biomechanical, work psychosocial risk factors and the severity of CTS and CTS-related work disability that can facilitate the development of comprehensive prevention strategies.

描述（由申请人提供）：腕管综合征（CTS）是最常见的周围卡压性神经病，由腕部腕横韧带下的正中神经受压引起。这是一个重要的职业健康问题，因为 CTS 的频率和严重程度导致的残疾几乎比任何其他上肢疾病都多（Foley 2007）。这项研究计划的最终目标是通过阐明 CTS 的原因来降低工作人群中 CTS 的发病率和严重程度。 很少有前瞻性流行病学研究有可能使用严​​格的病例标准和个体水平暴露估计来评估生物力学和心理社会危险因素与 CTS 或随后的残疾之间的暴露-反应关系。拥有足够权力评估发病率并解决健康工人幸存者效应（HWSE）偏差来源的人就更少了。 HWSE 已被证明会导致职业研究中的向下偏差，并且对于伴有不适的肌肉骨骼疾病（例如 CTS）的研究来说是一个特殊的挑战（Arigghi 1994；Eisen 和 Robins 2001；Eisen 2005）。最近有人尝试解决 HWSE，在慢性心肺疾病死亡率研究中使用 g 估计（Chevrier 2012），并在职业性哮喘研究中使用反比例治疗权重（Dumas 2013）。然而，目前还没有针对 CTS 或随后的工作障碍的纵向研究来解决 HWSE 偏见。事实上，除了针对制衣工人的一项横断面研究（Punnett 1996）外，在研究肌肉骨骼疾病时并未考虑 HWSE（Punnett 2004）。如果那些经常进行重复性剧烈运动而出现 CTS（包括正中神经症状）的工人更有可能离开此类工作，那么那些仍处于暴露状态的工人面临的风险就会较低。这种基于健康状况的自我选择可能会扭转因果方向，并导致低估生物力学危险因素、CTS 和随后的工作残疾之间的关联。 本研究的目的是评估生物力学暴露、工作社会心理暴露和 CTS 或与 CTS 相关的工作残疾之间的关联，同时调整 HWSE 的向下偏差。为了解决右截断问题，分析中将使用反比例审查权重。为了解决左截断问题，将通过改变进入队列的雇用日期标准来进行敏感性分析。对时变变量的混杂因素的调整将利用反比例处理权重。这种减少 HWSE 偏差的方法尚未在 CTS 或随后的工作残疾的研究中采用，因为大多数肌肉骨骼疾病的研究规模或复杂性不够。这项研究的结果将有助于阐明生物力学、工作心理社会风险因素与 CTS 的严重程度以及与 CTS 相关的工作残疾之间的关系，从而有助于制定全面的预防策略。