Frequencies of diseases presenting to General Practitioners according to patients’ occupation

General practitioner morbidity recording schemes may offer an opportunity to describe the
frequencies of a wide range of conditions according to occupation.
The aim of the study was to examine the feasibility of assembling information nationally, on
the frequencies and distribution of ill-health presenting to general practitioners, in relation to
occupation.
Schemes were identified by literature search and discussions with expert personnel, including
the HSE. Information was requested, by structured interview, from the scheme owners and
operators, by meetings, telephone and post. Relevant peer review papers were also studied
for details of methods.
In none of the Schemes is occupation routinely recorded, and such information would have to
be collected by an additional procedure.
Together the four main ongoing Schemes can provide a population of sufficient size to detect
differences in frequency of ill-health in most occupations. Statistical power to detect rare
effects would be limited in the smallest occupational groups.
The criteria for data entry are broadly similar in the Schemes, and differences in completeness
of entry at the individual doctor or practice level are likely to be more important than
differences in criteria.
All Schemes monitor completeness of recording, in various ways, and feed the results back to
the practices, so there is active encouragement of the quality of recording.
All Schemes use Read codes for diagnostic classification.
The Schemes differ in their use of independent ethics advisory committees. The main ethical
concern is that there should be no possibility of identification of individual subjects. In all
schemes, identities are coded in the practices before transmission of the clinical record. One
concern is that adding occupation to the clinical record might enable identification of
individuals.
All the Schemes are willing to discuss the possibilities for a combined study, subject to
practical and ethical considerations.
A possible design for a pilot study is suggested.