The Role of Manual Occupation in the Aetiology of Dupuytren’s Disease in Men in England and Wales

INTRODUCTION

The role of occupation in the pathogenesis of Dupuytren’s disease remains unclear (Bennett, 1982; Early, 1962; Liss and Stock, 1996). In a study of 901 cases of Dupuytren’s disease, it was concluded that, in certain cases, it is precipitated and/or aggravated by “hard work” (Mikkelsen, 1978). Conversely, in a book dedicated to Dupuytren’s contracture the conclusion reached was that manual labour is not an initiating or aggravating factor (Hueston, 1963).

We have studied the relationship between the incidence of “significant” Dupuytren’s disease and occupational social class in men in England and Wales. The term “significant Dupuytren’s disease” is used to denote disease that is sufficiently prominent or troublesome to require consultation with a General Practitioner (GP).

PATIENTS AND METHODS

The data used in this analysis were compiled from data collected in the Fourth National Morbidity Survey (MSGP4) undertaken by the Royal College of General Practitioners and the Office of National Statistics between September 1991 and September 1992. The MSGP4 was a prospective survey of patients’ consultations with GPs over a 12-month period. The total population surveyed was 502,493 people, equating to 1% of the population of England and Wales. The sample was representative of the population of England and Wales in terms of age, sex, marital status, tenure of housing, economic position, occupation and location of housing in an urban or rural area (McCormick et al., 1995). Social classes were defined as 1 (Professional), 2 (Managerial and Technical), 3N (Skilled Non-manual), 3M (Skilled Manual), 4 (Partly Skilled), 5 (Unskilled). For men, classes 3M, 4 and 5 include predominantly manual occupations. Data on Dupuytren’s disease were identified and retrieved from the database using the International Statistical Classification of Diseases (9th Revision) code 728.6. Incidence rates were defined as “first-ever” consultations for Dupuytren’s disease, consultation rates were defined as all consultations, and cumulative incidence was defined as the sum of successive age-specific incidence rates. Rates were age standardized to the standard European population using the Direct Method.

RESULTS

One hundred and sixty-nine new cases of Dupuytren’s disease were recorded from the survey population of 502,493, giving an incidence rate of 34.3 (SD, 5.3) per 100,000 men. Age-standardized incidence rates, as calculated for individual social classes, showed no significant differences in incidence (Fig 1). When occupational social classes were grouped into “non-manual” (1, 2 and 3N) and “manual” (3M, 4 and 5) groups, cumulative incidence rates were found to be significantly higher in the non-manual than in the manual group beyond 65 years of age (Fig 2). The age-specific incidence of significant Dupuytren’s disease in men increased with age until 60–69, and then declined. There was then a further increase with age, with a second peak at 80–89. Consultation rates were no different across occupational social classes, with the exception of social class 5 (Unskilled), which had a significantly higher rate.

DISCUSSION

Despite evidence that known risk factors for Dupuytren’s disease, such as smoking (Burge et al., 1997), are more prevalent in lower social classes, it is unclear whether the disease itself is associated with occupation and social class. Studies of the relationship between Dupuytren’s disease and manual labour have failed to reach a consensus (Bennett, 1982; Early, 1997; Hueston, 1963; Liss and Stock, 1996; Mikkelsen, 1978). Conclusions have varied depending on the populations surveyed, the type of manual work involved, the size of these populations and the controls used. The data of the present study suggest that men from social classes engaging in manual labour in England and Wales are at no higher risk of developing significant Dupuytren’s disease than those in non-manual occupations. We wondered if a true social class effect might be masked by any tendency for men from lower social classes to under-utilize primary health care. However, this is an unlikely explanation, as consultation rates per person with Dupuytren’s disease were actually higher in social class 5, and were no different for classes 3M and 4, than those for other social classes. Furthermore, analysis of general consultation rates by occupational social class in the MSGP4 study (McCormick et al., 1995) showed that individuals from social classes 3N, 3M, 4 and 5 consulted more frequently with their GP.

The variation in the age-specific incidence indicates a bimodal age distribution of significant Dupuytren’s disease. Whether this is a true effect, or an artefact due to some men waiting until after retirement to seek medical help, is uncertain.

In the present study, the diagnoses were made by GPs and it is possible that some cases of callosities in the hand and other contractures of the finger could have been mistakenly diagnosed as Dupuytren’s disease. However, this would not have decreased the difference in incidence that was found to exist between the manual and non-manual groups, since callosities in the hand and post-traumatic finger contractures are more common in manual workers.

Contrary to the view that manual occupation may predispose a person to Dupuytren’s disease, we found that it more common in the elderly who had non-manual occupations (Fig 2). This finding needs cautious interpretation for two reasons. Firstly, it is not possible to extrapolate our findings to the incidence of sub-clinical or milder forms of Dupuytren’s disease. Secondly, more refined epidemiological studies are required to investigate associations with individual occupations and time spent in different types of manual work. For both medical and legal reasons, such studies should be undertaken.