Guidelines have been defined as 'systematically developed statements to assist practitioner and patient decisions about appropriate healthcare for specific clinical circumstances' with a purpose to 'make explicit recommendations with a definite intent to influence what clinicians do'.
Cardiovascular medicine is an area that has long been characterised by clinical guidelines.
This trend shows no sign of abating with the December 2005 release of the Joint British Societies' Guidelines on Prevention of Cardiovascular Disease (CVD) in Clinical Practice 3 - abbreviated to 'JBS 2'.
The 'joint societies' in question are a number of fee-charging and sponsorship-dependent professional groups with a stated interest in CVD or diabetes.
JBS 2 follows and develops the comprehensive recommendations encompassed some seven years ago by its predecessor, JBS 1. This original document had a major influence on shaping the critically important policy document, the NSF for CHD.
Running to 52 pages of dense text, JBS 2 in its current form is not suited for use by GPs working in a busy real-time environment - the document is more of a reference level resource. It would be more useful to GPs if JBS 2 was eventually made available in a short form.
The intended scope of this guideline is to address 'all atherosclerotic CVD. The target population is adequately described and although specific clinical questions are not highlighted, these can be inferred from the discussions on lifestyle interventions and risk factor management.
The guideline has been developed under the direction of the respective professional groups, and given the specialist nature of these it may have been desirable to have incorporated non-specialist opinion as well.
There does not appear to be have been any stakeholder involvement through either seeking the views and preferences of patients, or piloting the recommendations among intended users, hence the recommendations are specifically those from the perspective of the authoring organisations.
The methodology underpinning these guidelines is not specifically described.
In a clinical trial report, it is essential to state the methodology, and it is important that readers of guidelines are able to assess the reproducibility of the guideline process so that the distinction between opinion and systematic appraisal is maintained.
Although reference is made to searches for systematic reviews and meta-analyses and the role of evidence-based medicine is discussed, it is not possible to determine how these processes have been adopted in relation to the conclusions.
The information is presented with reasonable clarity and the recommendations are specific and unambiguous. Much of the document is referenced, although the recommendations have not been graded, thereby diminishing the quality of the guideline.
The document is written in a narrative style conveyed in the first person plural ('we recommend', 'we believe').
Key recommendations are easily identifiable, and where appropriate the tools required for application are provided, for example risk-estimation charts.
The potential impact of the guideline on workload and resources is not evaluated, lending a slightly idealistic slant to the recommendations.
The importance of cost-effectiveness considerations in ultimately deciding priorities in practice is recognised, and this permits a licence for moderation within actual practice.
Relevance to GPs
Many of the JBS 2 recommendations will be familiar to GPs.
The most significant recommendation is that clinicians adopt a primary prevention threshold of 20 per cent CVD risk over 10 years.
The subsequent recommendation that all patients over the age of 40, with the exception of those with diabetes or receiving treatment for BP, should be offered opportunistic risk assessment at least every five years has enormous ramifications for primary care. Unfortunately, the basis and considerations relevant to this important recommendation are not given.
Given that just under half the UK population (some 28 million people) are over 40 years old, a wide divergence between the ideal and the reality of general clinical practice is highlighted. With increasing competing demands on GP resources it would have been reassuring to ensure explicit consideration of these.
It would also have been useful to include a discussion of adequacy of the performance of risk assessment tools and the clinical implications of such an opportunistic screening programme.
The cholesterol targets first advocated within the British Hypertension Society (BHS) guidelines 2004 also appear within JBS 2, although it is acknowledged that there is no direct trial evidence to support cholesterol targets and they are a matter of judgment.
The relevant trials describing both primary and secondary prevention of CVD with statin therapy are displayed on two graphs; the second plot highlights the benefits of treatment to lower targets and is probably unnecessary.
Trial evidence indicates that the benefits of treating to lower targets in patients with established disease is small and constrained by an increase in side-effects.
The ultimate test of a guideline's validity is whether it leads to the health gains and costs predicted for it.
The lipid management chapter is the longest and most detailed, partly due to the considerable emphasis on trials such as GREACE, TNT, PROVE-IT, REVERSAL, IDEAL, to name a few.
There is useful narrative describing the decrease in lipid parameters following an MI or other cardiac event. More importantly, practical advice on taking measurements in this situation is given.
The hypertension guidance will be familiar to those who have read the earlier BHS guidance. This has been updated in the light of the ASCOT study, and the criteria for co-prescribing diuretics and beta-blockers now takes into consideration a patient's risk of developing diabetes.
The ABCD algorithm remains, with an acknowledgement that it is based more on theory rather than on direct evidence. This algorithm will now be superseded by the recent joint BHS/NICE hypertension draft guidance where the additional change extends mostly to promoting calcium-channel antagonists over diuretics in the over-55 category.
JBS-2 recommends that the least expensive anti-hypertensive drug be used in the absence of a 'compelling indication'. The table listing compelling indications does not indicate the strength of evidence making the choice 'compelling', and in some cases the choice appears to be based on a single study.
The hypertension chapter concludes with a useful discussion on the management of BP in the post-stroke scenario.
The concept of metabolic syndrome is introduced within the chapter on diabetes. The National Cholesterol Education Programme definition is highlighted as being clinically useful, although this use is not specified.
Throughout the guideline the various metabolic syndrome criteria that are cited are often inconsistent, in particular different HDL levels are specified.
Reference to the current vigorous scientific debate around the validity of the metabolic syndrome concept is omitted, although the inclusion of the concept might be assumed to indicate its tacit acceptance.
As well as BMI, the waist circumference criteria used to define the metabolic syndrome is included, which is justifiable as their role in predicting diabetes is arguably more important than their relation to CVD.
Probably most useful of all is the algorithm for assessing impaired fasting glucose and impaired glucose tolerance. However, the accompanying narrative confuses the issue rather than clarifying it.
BP targets for patients with diabetes are also discussed. The optimum BP target is recommended as 130/80mmHg, while the audit target is recommended as 140/80mmHg. Clearly the use of the same threshold for diastolic BP means that an audit 'margin' is only present for systolic BP. The confusion is further compounded by the recommendation in the final chapter of a different audit standard for patients with hypertension and diabetes of 145/80mmHg. This is probably an error.
The results of the PRO-ACTIVE study are cited to support the role of oral anti-diabetic drugs in reduc-ing cardiovascular events. Unfortunately, because this was a negative study the secondary end point is employed to support this assertion and the figure is also incorrectly quoted.
The role of ethnicity in CVD is dealt with rather ambivalently. The recommendation to increase risk factor weightings in South Asians by 40 per cent is maintained, but elsewhere the observation is made that the incidence of CHD among South Asians has been reported to be no higher after accounting for the prevalence of diabetes.
This would seem to advance the view that an excess risk is not present after accounting for the prevalence of a key risk factor, ignoring the evidence for the role of insulin resistance in this patient group.
This point is previously acknowledged within the narrative of the diabetes chapter. The preceding reference to variations in cardiovascular risk among different ethnic groups within the risk assessment chapter also appears self-contradictory. Applying a guideline validation instrument (the 'AGREE' tool) I found that the JBS 2 guidelines achieved an average score of 45 per cent.
The areas of identified weakness were in relation to the methodology employed to construct the guideline, a lack of transparency in the formulation of the recommendations and non-disclosure of conflicting interests.
Lack of validation
Many bodies such as SIGN, the International Osteoporosis Society and the Royal College of Physicians now subscribe to the use of validation tools. Therefore the lack of adoption by the Joint British Societies is regrettable.
JBS 1 was a seminal document of its time when guideline development had not evolved to its current state.
While JBS 2 adopts a similarly contemporary opinion of the evidence base, the opportunity to also update the methodology used to construct and convey the recommendations has been missed, and the guideline can appear both paternalistic and anachronistic at times.
Publications, such as JBS 2, that employ differing chapter authors can be prone to suffering from lack of cohesion, and greater editorial scrutiny should reduce the number of apparent inconsistencies in the future.
Implicit within many of the recommendations in clinical guidelines are social and scientific value judgments - guideline developers must take into account public attitudes on bioethical issues as these are no longer the preserve of healthcare professionals.
The JBS 2 guidelines remain attractive as a single point of reference, but bear in mind that in some situations there may be trade-offs between evidence, practicality and opinion.
- Dr Minhas is a GP with a special interest in cardiology in Gillingham, Kent.
- Cardiovascular medicine is an area characterised by clinical guidelines.
- JBS 2 follows and develops the comprehensive recommendations encompassed by JBS 1.
- The information is presented with reasonable clarity and the recommendations are specific and unambiguous.
- The most significant recommendation is that clinicians adopt a primary prevention threshold of 20 per cent CVD risk over 10 years.
- GPs need to remember that there may be trade-offs between evidence, practicality and opinion.
- Lohr K N, Field M J. A provisional instrument for assessing clinical practice guidelines. In: Field M J, Lohr K N (eds). Guidelines for clinical practice. From development to use. National Academy Press (Washington DC, 1992).
- Howard RSA, Wilson M C, Tunis S R et al. Evidence-Based Medicine Working Group. Users' guide to the clinical literature VIII. How to use clinical practice guidelines. Are the recommendations valid? JAMA 1995; 274: 570-4.
- British Cardiac Society, British Hypertension Society, Diabetes UK, HEART UK, Primary Care Cardiology Society, The Stroke Association. JBS 2: Joint British Societies' guidelines on prevention of cardiovascular disease in clinical practice. Heart 2005; 91: 1-52.