Responsibility for prison healthcare has shifted from the prison service to the NHS. However, NHS methodologies for healthcare needs assessment are not suited to a prison setting for various reasons, including the small size and high turnover of individual prisons' populations; differences between types of prison; the lack of routine numerical data on prisoners' health status and needs; and the difference in culture between the health and prison services. These differences could impede the development of evidence-based planning.

Hull prison, Eastern Hull primary care trust and a team from Hull University's department of public health and primary care collaborated on a pilot project to assess to what degree the prison was adhering to authoritative guidance as a means of identifying unmet need. The approach combined assessing adherence to guidance with a proactive, pragmatic but nonprescriptive programme of identifying unmet health need through other means.

The small size and high turnover of the prison population advises against using epidemiological approaches (traditionally used by public health professionals looking at their local population).

The researchers also suggested that a specialised epidemiological approach would need to be specialist-led, but that this is likely to separate needs-assessment from planning. There is also a lack of IT in prisons, making the collection of quantitative data difficult.

The researchers realised that they would need to take a different view of 'evidence' and 'data', recognising the value of 'softer' information such as expert opinion, professional judgement and audit results.

Generating credible evidence to help NHS and prison staff understand how well they were meeting standards set out in the guidance was a key activity for the partnership.

Facilitated small group discussions of four to eight people were held, each with an explicit purpose (such as IT). These largely occurred between members of the prison health steering group, and 'expert witnesses' (appropriate specialists) were invited where necessary.

The group chose to draw on some of the findings of the prison's 'satisfaction survey' of prisoners rather than seek the prisoner perspective directly or through advocates. This approach led to a widely owned action plan, which included intentions to change some services, and bids to address service shortfall. It also commissioned evaluations of recent service changes and logged intentions to fill 'evidence gaps' - especially around how best to capture the user perspective.