Context: General practitioners (GPs) and generalist hospital doctors provide the majority of palliative and terminal care in the UK. Studies have revealed problems with symptom control and communication in these settings and inadequate training for clinical students and junior doctors.
Objectives: To investigate the training of GPs in Wales in palliative medicine throughout their careers, with a focus on the Welsh Valleys, an area of social deprivation and high levels of chronic ill health. To compare these data with those previously obtained from a survey of GPs in East Anglia. To develop regression models that enable the prediction of less well trained medical students and GPs.
Methods: A postal questionnaire survey of a random sample of GPs, stratified by practice location (Valleys or elsewhere in Wales). Responders were invited to recall their training in five topics of palliative care (pain control, other symptom control, use of syringe drivers, communication skills and bereavement care) during four career stages (clinical students, junior doctors, GP registrars and GP principals).
Results: The response rate was 67.6%. Available data enabled evaluation of generalizability and response bias. Contrary to an initial hypothesis, no significant differences were found between Valleys and non-Valleys responders' reported training, although the study was adequately powered. As medical students, 27% reported receiving no training in any topic, 75% no training in bereavement care and 50% no training in communication skills. Training varied across medical schools and was more common for more recent graduates. As junior doctors, 25% reported no training in any topic, 75% no training in bereavement care and 77% no training in communication skills. The GP registrar year provided significantly more coverage of communication, bereavement and syringe drivers than the combined preceding 6 years of general professional training. As GP principals, a high level of training is reported in all topics. The training experience of GPs in Wales is very similar to that previously reported by GPs in East Anglia: this lends support to the generalizability of these data. Logistic regression analysis indicated that the only predictor of less common training as medical students was having qualified less recently. The main predictors of less common training as GP principals was having become a GP more recently and not being a GP trainer.
Conclusions: There is still some way to go in ensuring that medical students, junior hospital doctors and GPs are all adequately trained for their important role in caring for dying patients.