Variation in local policies and guidelines for cholesterol management: national surveyBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7069.1368 (Published 30 November 1996) Cite this as: BMJ 1996;313:1368
- Rosalind Raine, senior registrara,
- Allison Streetly, senior lecturerb,
- Alan Maryon Davis, senior lecturerb
- a Directorate of Public Health and Health Policy, Lambeth, Southwark, and Lewisham Health Authority, London SE1 7NT
- b Department of Public Health Medicine, UMDS, St Thomas's Hospital, London SE1 7EH
- Accepted 6 September 1996
Raised serum cholesterol concentrations are known to be an important risk factor for coronary heart disease. In 1993 an authoritative systematic review of effective cholesterol management was disseminated to health authorities and boards throughout the United Kingdom.1 The review recommended that population cholesterol screening should be discouraged and cholesterol lowering treatment targeted at those patients at highest overall risk of coronary heart disease according to a range of risk factors. In 1994 we conducted a national survey to assess variations in local cholesterol management policies and clinical guidelines and the extent to which they reflected the recommendations.
Methods and results
A questionnaire was sent to the director of public health (or chief administrative medical officer) of all 151 health authorities or boards in the United Kingdom. Respondents were also asked to submit copies of local clinical guidelines. The responses were analysed using EPIINFO 6.1.
Completed questionnaires were received from 142 (94%) of the authorities. Only 70 reported the existence of a local written policy or guidelines covering cholesterol management, though 26 reported that policies were under development. Of the 70 with existing policies 55 reported that they included clinical criteria for cholesterol testing in general practice, referral to a dietitian or lipidologist, and hyperlipidaemia management—that is, they included clinical guidelines. However, only 34 reported a collaborative approach to developing such policies—for example, with family health services authorities, general practitioners, laboratories, and pharmacists. Only 13 had procedures for monitoring the implementation of the policies.
Of the 40 guidelines submitted, 37 advocated selective testing of individuals at high overall risk of coronary heart disease (as recommended in the review bulletin), rather than screening whole populations. Nevertheless, there was much variation in the risk factors specified and how they were defined. Seven guidelines discussed the concept of “high overall risk” but failed to specify any risk factors. Three guidelines did not mention pre-existing coronary heart disease, two did not mention a family history of hyperlipidaemia or premature coronary heart disease, and one suggested that cholesterol testing should be extended to “women without risk factors.” Only one guideline proposed the use of a composite risk factor scoring system. The age range recommended for testing also varied greatly: the lower limit varied from 16 to 35 years and the upper from 55 to 70. The cholesterol lowering management criteria also showed striking variation (table 1). Although 35 of the 40 submitted guidelines gave clinical criteria for cholesterol lowering drug treatment, there was much variation in the advice given.
This study revealed great variation and inconsistency in local criteria for cholesterol testing and treatment throughout the United Kingdom, despite the previous widely disseminated review recommendations. Only about half of responding districts had developed policies or guidelines, and only about half of those had done so collaboratively.
In view of the high potential cost of cholesterol management for large numbers of patients there is a need to make priorities based on clear criteria for both testing and treatment.2 For testing, this requires an explicit definition of each risk factor and of the various combinations comprising “high overall risk.” For treatment, it requires explicit criteria based on overall risk status (not merely serum cholesterol concentration), agreed in accordance with current evidence of cost effectiveness. Clear, explicit guidelines, developed collaboratively with those who will be using them, have been shown to facilitate, albeit not to guarantee, more consistent practice.3 All health authorities, through their directors of public health working with general practitioners, physicians, and lipidologists, should ensure that suitable local policies and guidelines for cholesterol management are agreed, disseminated, and monitored.
We thank members of the cardiovascular working group of the Faculty of Public Health Medicine on whose behalf the study was undertaken, and Mary Matthews of the health promotion office at the faculty for her administrative help.
Conflict of interest None.