I’m talking about the big stuff. If we were to scrap the NHS and design a new service from the ground up, what would an evidence-based health service look like? Who should fund services? Who should provide services? Who should be accountable to patients?
It seems that many developing countries actually face this challenge of building a health service from scratch. I hope they have a better grasp of the evidence than me.
There are probably some issues on which the evidence would be clear; universal coverage would be an obvious starting point. Public financing solves a lot of problems (demand-side socialism, as Tony Culyer puts it). Beyond that I’m left wondering. A number of sources summarise some of the evidence on NHS reforms, such as Cam Donaldson’s Credit Crunch Health Care, but I can’t find anything comprehensive. I’d really like it if somebody could recruit a PhD student to review all the evidence for the design of a health service and summarise where the evidence is strong and where it is weak. Obviously this would require a definition of objectives and decision rules. Simultaneously maximising health, equality of health and access to services seems a relatively uncontentious goal. So what kind of service would best provide this?
We have some idea of what the service shouldn’t look like, and that’s what they have in the United States. Little else is certain. The debate over the changes to the NHS brought in by the Health and Social Care Bill 2010 was relatively light on evidence, though a few issues were brought to the table. In a poll on the Academic Health Economists’ Blog, a majority voted that the Health and Social Care Bill should have been rejected. I’d like to think that this was at least partly informed by the readers’ understanding of the evidence. However, for someone to reject a bill does not mean that they can offer a better alternative, even if they do have some idea about the evidence. Maybe if we had a better grasp of the evidence we’d have a better idea of what we want our health service to look like.