Influence of physically demanding occupations on the development of osteoarthritis of the hip: a systematic review

We identified a total of 6070 records and read 273 potentially relevant reports. After exclusion of all reports that did not fulfil our inclusion criteria, we included a total of 13 eligible studies including 7 cohort studies [2937] and 6 case–control studies [3843] with a total of 15 publications (Fig. 1). Cross-sectional studies with an exposure to occupations of at least 10 years to ensure the association between occupational exposure and the development of HOA were not identified.

Fig. 1figure 1

Flow chart [44] to describe identification and selection of included studies

Participants

All eligible studies were conducted in Europe, predominantly in Scandinavian countries (Sweden, Denmark, Finland, and Iceland), two studies in the UK and one in Germany. Most studies included working age participants with a mean or median age of over 36 years at time of diagnosis. The proportions of men showed a broad variation across occupational groups. Most studies recruited more men and four studies [38, 39, 42, 43] recruited only men (Supplementary Tables S7 and S8).

Exposure

Information on occupation was based on registry data [29, 3137] in Scandinavian countries, occupational titles [32] and interviews [38, 39, 43] or questionnaires [4042] on occupational history. Some studies summarized different occupational groups, others were more specific and differentiated between degrees or periods of exposure. This review focuses on results on the risk to develop HOA in physically demanding occupational groups traditionally dominated by men (farming, construction, driving, metal work, storage and transportation).

We identified nine occupational sectors with physically demanding workloads and resulting high risk to develop HOA. These sectors considered occupations in the following:

•agriculture, fishery or forestry (12 studies): Three of these studies [38, 42, 43] differentiated between exposure lengths. Studies reported results on the exposure of farmers or summarized farmers and forest workers [40, 43] farmers and agricultural workers [38] or agricultural and fishery workers [34] (Supplementary Table S9).

•construction (9 studies): Five studies reported the risk for construction workers [29, 31, 37, 40, 43], three studies summarized exposure of comparably demanding occupations [34, 39, 41] and one study reported more specialized occupations [33] (Supplementary Table S10).

•driving vehicles with exposure to whole body vibration (8 studies): Two studies reported on the risk of any occupation with exposure to whole body vibration [33, 40], others reported results for professional motorists [31, 34, 39] or differentiated between driving different vehicles [36, 37, 42] (Supplementary Table S11).

•metal work (6 studies): Of these, three studies reported the risk of general occupations in metal work [31, 40, 43], others summarized metal and machinery workers [34] or reported more specific information on sheet-metal workers [33] and furnace men, smiths and workers doing metal processing [37] (Supplementary Table S12).

•sales (5 studies): Of these, four studies reported the risk of working as service or shop worker [3441], whereas one [37] summarized very heterogeneous demanding occupations for store and warehouse workers (Supplementary Table S13).

•gastronomy (4 studies): All of these studies summarized different occupations with comparable exposure in gastronomy and hotels [34, 36, 37, 40] (Supplementary Table S14).

•food production (4 studies): Three studies summarized comparable demanding occupations [37, 40, 43], whereas one study [36] reported the risk of less demanding occupations as baker, pastry cook or confectionery maker (Supplementary Table S15).

•storage transportation (3 studies): All studies investigated the exposure of postmen [38, 41, 44], one study added results for storage and transport workers (Supplementary Table S16).

•health care (3 studies): Studies reported results for health-care assistants [29], medical doctors [36] and one study reported combined results for nurses and environmental officers [34] (Supplementary Table S17).

Finally, three studies [34, 37, 41] investigated the risk of HOA of men in unskilled or basic level labours in comparison to managers, professionals or low-exposure blue-collar workers (Supplementary Table S18).

Diagnosis

Follow-up periods between exposure and diagnoses are mainly available from cohort studies and ranged from 3 to 28 years.

Various criteria were used for the diagnosis of HOA:

•Disability pension or sick leave due to HOA [31, 34, 43]

•Implantation of a total hip replacement (THR) [33, 36, 41]

•Radiological diagnostic criteria or THR [39, 42]

•Clinical or radiological diagnostic criteria [29, 35, 37, 38, 40]

•Disability pension due to low back disorders [43]

A high validity of the diagnosis of HOA was assumed for disability pension due to HOA, THR and radiological imaging compared to a clinical diagnosis and back disorders.

Quality assessmentCohort studies

The seven cohort studies scored 12 to 15 points on a scale ranging from 0 to 19 points. No study could provide accurate and reliable quantitative data on exposure such as frequency or duration of occupation. All studies adjusted results for age; three used other important confounders (body mass index or education and physical workload factors) [31, 33, 34] (Supplementary Table S7).

Case–control studies

The six case–control studies scored between 5 to 10 points on a scale of 0 to 15. Three studies received less than 3 out of 6 points for the selection of the study population [3840] due to deficits in the representativeness of the cases and the selection of controls. Two studies [40, 41] had a low response rate. Only one study received full points for an adequate case definition by consideration of people with a THR [41] and one study received no points due to the low validity of diagnostic criteria [43]. Only one studies excluded pre-existing HOA al the baseline [35]. Only two studies received points for including a long enough follow-up of more than 10 years for the endpoint to occur [42, 43]. Four studies were downgraded due to low validity of data acquisition [3942] and two studies [

留言 (0)

沒有登入
gif