METHODS

Every institution in the United Kingdom that undertakes animal experimentation is required to nominate an individual (the holder of certificate of designation) who oversees all aspects of the procedures.⁵ They, numbering 296 in 1999, are each registered with the Home Office. We devised a single page questionnaire in which respondents were asked to enumerate the employees at their institution by job category and species handled. Three broad job categories were considered: animal technicians, researchers/scientists, and others (maintenance, security, etc). Species were grouped into small (rats, mice, guinea pigs, hamsters, ferrets, and rabbits) and large (cats, dogs, sheep, cows, primates, horses, goats, and pigs) animals. Groups for insects and other animals were also included. The Home Office delivered the questionnaire with a circular during October 1999 and anonymous responses were returned to the authors in a reply envelope. Two follow up mailings were conducted in early 2000. The estimated range of the total population, our denominator, is based on the information obtained and the likely characteristics of the non-responding populations. The lower estimates reflect the actual responses for small animals and assume that none of the non-responders used them at their sites. The higher estimates assume that the distribution among non-responders is the same as the responders. Quarterly SWORD reports over 12 months following the survey identified the number of new cases of occupational asthma and rhinitis reported among laboratory staff which were attributed to laboratory animals, insects, or birds. We calculated numerators (with 95% confidence intervals based on a Poisson distribution) using the sampling fraction for four consecutive quarters. Incidence rates were calculated using the higher estimates of the denominator population. The data were analysed using STATA (Stata Corporation, Texas).

RESULTS

Most of the surveyed sites were commercial organisations (38%) or places of higher education (34%). A total of 207 questionnaires (70% of those eligible) were returned; information regarding the breakdown of species for small and large animals was available for 200 responses. Table 1 shows the lower and higher estimated numbers of employees grouped by species and job category. Between 12 000 and 17 300 employees were working with small mammals, in most cases scientists working with mice or rats. Much smaller numbers work with larger animals, birds, insects, or fish.

Using the higher estimates of the denominators, and among those working with small animals, we calculated a current, annual incidence rate of 1.56 cases/1000 persons for occupational asthma and a figure of 2.54/1000 persons for occupational rhinitis. The rates for those working with birds or fish were higher, but based on a small number of cases.

DISCUSSION

This is the first attempt in 20 years to determine the size of the population occupationally exposed to laboratory animals. Over this period major changes have taken place in the number and type of animal experiments conducted in the UK. Our results suggest that the number of exposed employees has halved since 1981. Unlike the earlier estimate, we have been able also to estimate the numbers working with individual species. The higher of our estimates of those working with small animals is in close agreement with the number of current personal licence holders (animal technicians and researchers, n = 13 700) recorded by the Home Office in 1999. In 2000 there were 1900 animal technicians who were members of the Institute of Animal Technology,⁶ a figure also close to our lower estimate. In future, these more readily available statistics may serve as a useful indicator for these populations. Our estimated rates of occupational asthma and rhinitis are high; but nonetheless are likely to underestimate the true figures because the numerator depends on cases reported by chest and occupational physicians to SWORD. Rates among newer employees, who are probably at particularly high risk, may also be higher.