Education And Debate US and UK health care: a special relationship?

Why is the grass greener?

BMJ 2005; 330 doi: (Published 24 March 2005) Cite this as: BMJ 2005;330:727
  1. Barbara Starfield (bstarfie{at}, university distinguished professor1
  1. 1 Johns Hopkins School of Public Health, 624 North Broadway, Baltimore, MD 21205, USA

As well as learning from each other's existing systems, US and UK researchers have potential for collaborative research into improving health care


Almost a century ago, the UK and US health systems diverged in a dramatic way. In the United Kingdom, parliament legislated on health insurance; subsequent seminal governmental reports and actions, such as the Dawson report on primary care after the first world war and the enactment of the NHS, set the stage for the current health services system. Despite several notable attempts at reform in the United States, particularly in the early 1930s, mid-1960s, and early 1990s, the US healthcare system remains much like it was a century ago, mired in a market oriented private system with private financing. The system makes only a nod towards public responsibility, mainly in the form of guaranteed financial access for elderly people (Medicare) and a fragmented and unstable system of financing and direct service for care of poor people (Medicaid and community health centres, respectively).

As a result, the United States lacks system-wide health policy making. By contrast, the NHS assumes responsibility for national health policy, the quality of care (for example, the National Institute for Clinical Excellence), and banning direct to consumer advertising of prescription medicines. The United Kingdom has a national director for primary care who, besides being responsible for policy in various aspects of primary care, is the country's official representative in international deliberations on primary care. The United States has no comparable position.

Despite these enormous differences between the two health systems, there remain possibilities for two-way learning. These fall into several groups: the importance of primary care as the infrastructure of a health system; information technologies for health services; quality of services; and equity in the attainment of health of populations.

Primary health care

World Health Organization documents on primary health care characterise it as an approach directed at maximising public responsibility for a health services system. The United Kingdom is a prime example of a national approach to achieving the cardinal features of primary care:

  • Person focused care over time

  • Assured first contact access

  • Comprehensiveness (in the sense of providing care for the most common health conditions and referral for conditions that are too uncommon to maintain competence)

  • Coordination of care when people are seen elsewhere.

The principles of family orientation, community centredness, and cultural competence are often included as critical features but rarely achieved well anywhere.

The United States, in contrast, has only sporadically shown interest in primary care as a core of the health services system. A relentlessly increasing focus on specialty care dominates, despite evidence that it is not associated with improvements either in average levels of health of the population or in the distribution of health across population groups.1 Nevertheless, the potential for joint learning is considerable.

Unified force for change

In the United Kingdom, the general practice community has a major influence on action to improve care. In the United States, the professional constituency for primary care is fragmented into family medicine, general internal medicine, and general paediatrics. US primary care professionals have much to learn from the successes of a united front in the United Kingdom.

Embedded Image

The United States has much to learn about equitable health care


Referral rates

Referral rates from primary care to specialist care in the United States are three times those in the United Kingdom,2 3 even after differences in morbidity burdens in the two populations have been taken into account. Patients need to visit specialists for several reasons: second opinion, need for special diagnostic and therapeutic procedures, and ongoing care that requires special expertise.4 The United Kingdom has been experimenting with ways of reducing visits to specialists, including outreach visits to general practices by specialists and more extensive telecommunications.1 The National Institute for Clinical Excellence guidelines include at least some criteria for referral to specialists. On the other side of the Atlantic, some innovative clinical programmes have been successful in improving clinical management by better coordination between primary care physicians and specialists,5 6 although the precise mechanisms by which they do so are largely unexplored.

Information systems

Doctors in the United Kingdom are far ahead of their US counterparts in the use of computers for clinical practice. As a result of a landmark report of the US National Committee on Vital and Health Statistics,7 greater attention is being devoted to considering a national health information infrastructure with three electronic components: medical records, population information systems, and records kept by individuals about their health. Important barriers to the adoption of such a system are the cost and the absence of a public health infrastructure to coordinate the developments.

Characterisation of ill health

Medical research has focused on increasing knowledge about specific diseases. Although big improvements in average levels of health have resulted from these increases in knowledge,8 9 the nature of disease is changing, with greater recognition of the phenomena of comorbidity and multimorbidity.10 New models of influence on health make it clear that diseases have no single causes (or even determinants), except, perhaps, for rare Mendelian dominant conditions. The multiple influences all interact with each other, many in as yet unknown ways, so that disease is not randomly distributed in the population.11 US researchers lead British researchers in developing methods to characterise case mix and morbidity burden in individuals and, by aggregation of individual data, in subpopulations and populations.12

This new model of illness greatly increases the potential for tailoring interventions to individuals, and subpopulations, according to their health needs. Given the recognition of the importance of primary care in the United Kingdom, greater attention to these new possibilities for understanding and management of ill health seems appropriate.

Population based clinical data systems

The United States has a remarkable clinical data system (the Medicare database) for the population over age 65. The utility of this database has been proved many times over, particularly with regard to documenting and understanding variations in physicians' practices in different areas of the country. The most recent of these demonstrations concerns the use of specialists and the differences in both processes and outcomes of care associated with different specialist: population ratios.1315 Could the United Kingdom develop a similar system, even in the absence of the claims forms that provide the basis for the clinical data? The Prescribing Analysis and Cost (PACT) database within the NHS provides national data on drugs16 but does not seem to have been as well used for health services research as the Medicare data.

Equity in health

The United States has expressed a commitment to reducing inequities in health, in the form of one of the two national goals in Healthy People 2010.17 As is characteristic for a US national document, it does not pose any mechanisms for achieving equity or for understanding its genesis. The landmark Black report18 provided the world with a well reasoned analysis of the issue of social differences, and the subsequent Acheson report19 added to the basis for a national commitment to action. No such analysis has been conducted in the United States, which also devotes little attention to disparities by social class, making the problems of the poor and working class white subpopulation virtually invisible. With the commitment of WHO to working towards national and international equity, the United States could learn a great deal about potential strategies within the country, taking the lead from the impressive research efforts and policy decisions in the United Kingdom.

Common challenges

The considerable track record of both US and UK health services researchers makes potential collaborations an exciting possibility. The large differences between the two health systems increase the potential to shed light on new approaches. At least two areas pose new challenges: assessment and monitoring of quality of care and potential of teamwork, particularly in primary care.

Assessing quality of care

Despite over 20 years of interest in assessing the quality of care through its impact on outcomes (health status), most efforts are still focused on processes of care that are thought to be related to desired outcomes. Most clinical guidelines are of this type. Several years of experience in the implementation of evidence based medicine suggest a need for a reassessment. Firstly, the evidence base for most guidelines is inadequate. Even the most elegant randomised clinical trials lack assurance of generalisability of the evidence. Secondly, most trials, even the best ones, are not conducted under conditions of real practice. As a result, the selection of candidates for most disease oriented trials ensures that people with one or more other illnesses—the majority of people (especially the elderly population)—are not included in the trials. Thirdly, clinical trials are generally not designed to ascertain variability in response to the intervention, even when it is possible that certain sub-groups of the population differ in their responsiveness, and some assumptions of trials may make the conclusions inappropriate.20

Moreover, methods of ascertaining quality of care by its effect on health have not been implemented in any national health system. For example, the simple method of asking patients whether their health is improved after care, which proved promising in a short term evaluation,21 has apparently not been considered as a method for holding health services accountable for what they do. Well tested tools for assessing changes in health status exist but are not used.

Both the United States and the United Kingdom use techniques of evaluation that are derived frommarketing techniques to ascertain satisfaction. Too great a focus on patient satisfaction will detract from a more concerted effort to gain evidence from clinical practices, thus depriving both practitioners as well as consumers of a rational rather than a preference basis for decision making. It could be argued that need, rather than demand, should be the primary criterion for providing services, in the interest of better outcomes and more equitable distribution of resources.

Summary points

Organisational differences between the United States and United Kingdom present many possibilities for joint learning

The NHS primary care system offers many learning opportunities

The United States is ahead of the United Kingdom in research to understand the multiple determinants of disease but lags behind on equity in health

The United Kingdom could make better use of national patient data

Potential of teamwork

Both UK and US organisations employ teams of practitioners to provide primary care. In US managed care, it is common for nurse practitioners to act as frontline providers of care, with primary care doctors as back up. Because of the proprietary nature of managed care data, little is written about how decisions are made about who does what, and its impact in terms of quality and outcomes of care. This issue could well serve as a basis for a bi-national collaborative evaluation, with benefit to both countries.

This is the third in a series of articles in which we asked experts in UK and US healthcare systems to identify opportunities for learning between the two countries


  • Contributors and sources This article reflects BS's experience in health services research and health policy, which derives from personal and policy contacts in many industrialised and developing nations.

  • Funding This work was supported in part by Grant No 6 U30 CS 00189-05 S1 R1 of the Bureau of Primary Health Care, Health Resources and Services Administration, Department of Health and Human Services, to the Primary Care Policy Center for the Underserved at Johns Hopkins University.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
View Abstract

Log in

Log in through your institution


* For online subscription