Intended for healthcare professionals

Primary Care

Association between practice size and quality of care of patients with ischaemic heart disease: cross sectional study

BMJ 2003; 326 doi: (Published 15 February 2003) Cite this as: BMJ 2003;326:371
  1. Azeem Majeed, professor of primary care (a.majeed{at},
  2. Jeremy Gray, directorb,
  3. Gareth Ambler, statisticianc,
  4. Kevin Carroll, specialist registrar in public health medicineb,
  5. Andrew B Bindman, professor of medicine, epidemiology, and biostatisticsd
  1. a Primary Care Research Unit, School of Public Policy, University College London, London WC1H 9QU,
  2. b Battersea Research Group, Bolingbroke Hospital, London SW11 6HN,
  3. c Medical Statistics Unit, Research and Development Directorate, University College London Hospitals NHS Trust, London NW1 2LT,
  4. d Departments of Medicine, Epidemiology, and Biostatistics, University of California San Francisco, San Francisco, CA 94118, USA
  1. Correspondence to: A Majeed
  • Accepted 5 November 2002

Proportionally fewer inpatients die in hospitals that do more operations than in hospitals that do fewer.1 Similar associations between outcome and the size of hospitals have been found in other studies. An association between size and outcome may also be important in primary care settings, where most patients with chronic illnesses are managed. If large practices or those that treat more people provide better care, this could have important implications for the organisation of primary care services. We looked for an association in patients with ischaemic heart disease because the management of this disease is an international priority.2

Participants, methods, and results

From September 2000 to May 2001, we identified patients diagnosed as having ischaemic heart disease using paper and computerised medical records in four primary care groups in southwest London (69 general practices; population 382 188). Seven general practices did not take part.3

We recorded patients as hypertensive if their blood pressure was more than 140/85 mm Hg. We classed cholesterol concentrations greater than 5 mmol/l as high and defined patients with a body mass index (weight (kg)/height (m2)) of 30 or greater as obese. We extracted information on treatment with cardiovascular drugs from computerised records. Fifteen practices were unable to supply some data and were excluded from some of the analyses.

We calculated the proportion of patients in each practice whose risk factors were assessed or controlled; who were taking aspirin, statins, β blockers, or angiotensin converting enzyme inhibitors; or who had had revascularisation treatment. To examine the association of practice size and volume of cases with quality of care, we used a logistic population averaged generalised estimating equation model, adjusted for age and sex, that allowed for clustering within practices.

Practice size varied from 1265 to 13 147 patients (mean 5762). In total, 6888 people had ischaemic heart disease; the number of cases in individual practices varied from 12 to 326 (mean 111) and prevalence varied from 0.45% to 4.37% (mean 1.96%).

Only records of cholesterol concentrations showed an improvement with increasing number of cases of ischaemic heart disease. An increase of 10 in the number of cases was associated with a 6% increase in the odds of recording (table). On average, a practice with 200 patients with ischaemic heart disease would have recorded cholesterol concentrations for 69% of patients registered with the practice compared with 56% in a practice with 100 cases.


Most aspects of the management of ischaemic heart disease in primary care were not associated with the number of cases managed. We also found no association between practice size and the quality of care. This suggests that the trend in the NHS towards larger general practices by itself has little impact on the quality of chronic disease management in primary care.

Although recent developments in the NHS have cast doubt on the future of smaller practices, both patients and the doctors seem happy with smaller practices. Smaller practices are seen as more accessible and achieve higher levels of patient satisfaction. 4 5 The NHS should reconsider how it can improve the quality of care provided by general practices, without relying on the presumed benefits of consolidating them into larger units. Other initiatives—for example, the use of disease facilitators, local incentive schemes, expansion in specialist services, and the development of general practitioners with special interests—need to be evaluated to see if they can achieve this objective.

Univariate associations between number of cases (adjusted for practice size) and practice size and management of ischaemic heart disease in 62 general practices, southwest London, September 2000 to May 2001

View this table:


We thank Rumana Omar for advice on the analysis of clustered data.

Contributors: All the authors planned the study. GA did the statistical analysis. AM wrote the paper and received comments from the other authors. AM and JG are guarantors.


  • Funding The data collection for this study was funded by Battersea; Balham, Tooting, and Wandsworth; East Merton and Furzedown; and Putney and Roehampton primary care groups. An additional grant was received from Merton Sutton and Wandsworth Health Authority. The Battersea Research Group is a primary care research network funded by the Department of Health. AM holds a primary care scientist award, funded by the NHS Research & Development Directorate.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.