Health problems lead to considerable productivity loss at work among workers with high physical load jobs

Abstract

Objective

To assess the feasibility and validity of two instruments for the measurement of health-related productivity loss at work.

Study Design and Setting

A cross-sectional study was conducted in two occupational populations with a high prevalence of health problems: industrial workers (n = 388) and construction workers (n = 182). We collected information on self-reported productivity during the previous 2 weeks and during the last work day with the Health and Labor Questionnaire (HLQ) and the Quantity and Quality instrument (QQ), with added data on job characteristics, general health, presence of musculoskeletal complaints, sick leave, and health-care consumption. For construction workers, we validated self-reported productivity with objective information on daily work output from 19 work site observations.

Results

About half the workers with health problems on the last working day reported reduced work productivity (QQ), or 10.7% of all industrial workers and 11.8% of all construction workers, resulting in a mean loss of 2.0 hr/day per worker with reduced work productivity. The proportion of workers with reduced productivity was significantly lower on the HLQ: 5.3% of industrial workers and 6.5% of construction workers. Reduced work productivity on the HLQ and the QQ was significantly associated with musculoskeletal complaints, worse physical, mental and general health, and recent absenteeism. The QQ and HLQ questionnaires demonstrated poor agreement on the reporting of reduced productivity. Self-reported productivity on the QQ correlated significantly with objective work output (r = .48).

Conclusion

Health problems may lead to considerable sickness presenteeism. The QQ measurement instrument is better understandable, and more feasible for jobs with low opportunities for catching up on backlogs.

Introduction

In economic evaluations of health care interventions, it is widely recommended to consider all costs and savings relative to the benefits of the intervention. From a societal perspective, this also includes productivity costs, that is, the costs of production loss due to illness and associated disability [1]. Loss of productivity is traditionally measured by illness-related absence from work [2]. Even when employees are present at work, however, they may experience a decreased productivity caused by functional limitations due to health problems. The phenomenon that workers turn up at work, despite health problems that should prompt absence from work, is referred to as sickness presenteeism. A study across the Swedish workforce demonstrated that during a period of 12 months about 37% of all workers experienced sickness presenteeism [3]. In economic evaluations of health care interventions, the additional impact of primary effect measures (usually clinical and health outcomes) on associated indirect costs are seldom included.

Although sickness presenteeism may lead to substantial economic losses, few studies have estimated the decrease in productivity of workers with health problems. Among health insurance claim processors, it was shown that workers who used sedating antihistamines experienced on average 8% reduction in daily work output in the three days after receipt of the prescription, relative to the regular number of claims per day handled by these workers [4]. At a credit card company, sickness presenteeism accounted for higher productivity losses than sickness absence among telephone customer service operators with migraine, whereby job productivity was measured by handle time per call and time unavailable for calls [5]. Because objective measures of productivity at work are rarely available or are difficult to access, other studies have used self-reports to estimate the decrease in productivity that is associated with health problems at work. About 50% of migraine patients reported at least two work days lost per month [6]. Osteoarthritis patients with health complaints during work time reported about 9% mean loss in productivity [7]. In two studies, the prevalence of sickness presenteeism and the impact on worker productivity were estimated. Brouwer et al. [8] reported that, on an average day, 7% of the workers in a trade company experienced health problems while being at work, with an estimated productivity loss of 13% per worker with health problems. Among computer users, ∼8% reported reduced productivity due to musculoskeletal symptoms; the mean productivity loss was ∼15% for women and 13% for men [9].

Several questionnaires have been developed to measure sickness presenteeism. The Work Limitations Questionnaire measures time, physical, mental–interpersonal, and output demands of the job [10]; it has been validated with objective work productivity data [11]. Other questionnaires have focused directly on output performance by asking about work efficiency in the past 1 or 2 weeks on a 10-point numerical rating scale [6], or the average level of functioning during a period with health problems [12]. The Health and Labor Questionnaire (HLQ) asks for the number of hours needed to compensate for lost work due to health problems during the previous 2 weeks [13], whereas in the Quality and Quantity questionnaire (QQ) the quantity and quality of the work performed on the last working day can be reported on a 10-point numerical rating scale [8].

Our primary objective was to evaluate the HLQ and QQ for the measurement of productivity loss at work in occupational populations with an established high prevalence of health problems. Although these questionnaires have been used in several cost-effectiveness studies, reliability and validity studies are scarce. Our secondary objective was to analyze the influence of individual characteristics, work-related risk factors, and general health on self-reported productivity at work.