Intended for healthcare professionals


Assessing the options available to Lord Darzi

BMJ 2008; 336 doi: (Published 20 March 2008) Cite this as: BMJ 2008;336:625
  1. Martin Roland, director
  1. 1National Primary Care Research and Development Centre, University of Manchester, Manchester M13 9PL
  1. m.roland{at}

Time to look at both the system and the patient as a whole

The Department of Health in the United Kingdom is undertaking a major review of National Health Service policy (“the Darzi review”). Which options are most likely to produce an effective, efficient, and patient centred health service? The UK has made major progress in reducing avoidable deaths and improving care for chronic diseases.1 2 However, many areas still have considerable room for improvement.

Strong primary care is associated with reduced costs and improved outcomes.3 The review therefore needs to strengthen primary care. The UK system of universal registration with a single general practice promotes equity, provides a “medical home” for coordinating care, and is an effective mechanism for holding providers to account for the quality of care provided. This must be retained, even though patients may occasionally consult other practitioners—for example, a doctor near their workplace.

Much has been written about “polyclinics”—facilities that group together primary care practitioners with diagnostic and specialist services. Despite lack of clarity about their purpose, several are already planned. Part of the debate about polyclinics relates to a perceived need to merge small practices in highly deprived areas where the quality of medical services is low. Providing good premises and facilities for such practices could make a big difference to care and would enable the practices to pool resources to provide a wider range of services. However, whether practices should merge within such facilities is a separate question.

The NHS goal of providing patient choice in primary care is not realised in many parts of the country where patients have little real choice of practice. Increased patient choice requires more high quality practices, not the small number of large practices that some polyclinic models suggest. We know that patients in small practices rate their care more highly in terms of both access and continuity.4 Indeed, although small practices show more variation in quality, on average, they achieved slightly higher levels of clinical quality than larger practices in the quality and outcomes framework.5 One approach would be to use polyclinics to provide extra community based resources but to use them to house general practitioner practices only when local buildings or access is poor. Additional practices are still needed to increase patient choice.

Some models for polyclinics include a greater role for specialists working in the community (bringing services “closer to home”), and government policies are already moving specialists out of hospitals and training primary care staff to take on new specialist roles. However, specialists may be less efficient when deployed outside hospitals.6 A recent evaluation of NHS closer to home demonstration sites suggests that these new services improve access when provided as an additional resource. However, their economic justification is much less clear when they are designed to substitute for existing hospital based services. Specialists and practitioners have raised concerns about the quality and safety of some of these new services, although there is little hard evidence of problems, and patients rate them highly.7 The justification for moving specialist services into the community is greatest where patients have long travelling times, and the benefits are small in many urban areas. Training new types of practitioner to provide specialist services can provide effective additional capacity, but only if quality control is good.

Better support for self management is needed for people with long term conditions. Practitioners need to provide patients with better information about their care and involve them more in decisions about care. These are areas where NHS performance is poor.8 Existing NHS models of self management need to change so that they are better integrated with existing services, include support and training for practitioners, and organise services around the needs of patients.9

The greatest demand on the future NHS will be to provide high quality coordinated care for patients with multiple chronic diseases. Recent NHS initiatives have increased the range of providers in both primary care (for example, walk-in centres) and secondary care (for example, independent sector treatment centres). This has the potential to worsen coordination of care—an area in which UK performance is already poor compared with other countries.10

Several changes are necessary to improve continuity and coordination of care. Firstly, the general practitioner contract should be modified to incentivise personal continuity of care as well as access, especially as three recent studies suggest that many patients in England value personal continuity more highly than rapid access to care.11 12 13. Secondly, financial incentives for hospitals to admit patients as emergencies need to be reduced. Instead, hospitals should have incentives to work with providers within the community to provide high quality alternatives to admission. Thirdly, new approaches to commissioning health care are needed. Patients with multiple conditions need better coordination of care between the various sectors of the NHS, and involving hospital specialists in joint commissioning with general practitioners would provide better integration than a policy that focuses exclusively on commissioning led by primary care.

Vigorously pursued policies may deliver on their stated goals but have other unintended effects. For example, strategies designed to reduce waiting times to see general practitioners have made it more difficult for patients to book in advance,14 and an emphasis on seeing non-urgent new patients within target times can result in other patients coming to harm because their follow-up appointment is delayed.15 Now is the time to look at both the system and the patient as a whole. That is the challenge for the Darzi review.


  • Competing interests: MR receives research funding from the Department of Health. The opinions are those of the author and not of the Department of Health.

  • Provenance and peer review: Commissioned; not externally peer reviewed.