EDITOR,—Moulding et al eloquently describe those theories derived from social and behavioural sciences that may help to bring about change in clinical practice through the use of guidelines.¹ Occasionally, the behavioural approach of providing incentives related to specific actions may be all that is required to achieve the desired change.² Early in the winter of 1996 the health authority notified general practitioners (GPs) in Cornwall and the Isles of Scilly that a pharmaceutical company was offering to supply pneumococcal vaccine at a favourable discount, giving a profit margin to the general practice of £6.15 per dose. This notification also advised GPs as to which groups of patients were deemed by the Department of Health to be at risk of invasive pneumococcal disease and therefore required immunisation. The table shows how the prescribing of pneumococcal vaccine has changed over time. During the winter of 1996 GPs displayed an obvious greater enthusiasm for prescribing this particular immunisation.

Although this simple approach of a single mail shot did not achieve the same dramatic results as seen in the multifaceted campaign reported by McDonald et al,³ and though we were unable to ensure that immunisation was offered only to the at risk groups, we would suggest that in this instance compensation had a positive effect on clinical behaviour.⁴