Text size


Front Page
Special Edition - Response to Cochrane Review Nov

NHS Health Check - Response to the Cochrane Review

This eBulletin is produced by NHS Diabetes and Kidney Care in conjunction with the Department of Health to support the NHS Health Check Learning Network.   This is a useful tool for anyone working in health or social care with an interest in the prevention of vascular disease and vascular risk assessment and management.  To subscribe or unsubscribe, please contact Eleanor Kent-Dyson. 

In response to demand from NHS Health Check commissioners, we have put together a further assessment of the Cochrane review published on 17th October 2012.

The Cochrane review is a systematic review of the published literature on randomised trials of general health checks.  The principal end point reviewed was mortality, for which nine trials provided evidence.  A meta-analysis of these trials showed no difference in deaths overall, or from cardiovascular disease or cancer, between intervention and control groups.

The review also comments on comparisons of morbidity between intervention and control groups, although this was not the primary intention of the review and no meta-analysis was possible.  Some trials showed differences, but the review authors did not consider these useful, partly as a result of the poor quality of the evidence in many individual trials.

There are some major difficulties in interpreting the results of this review:

(a)     There was no specification of what constitutes a ‘general health check’, its content or its objectives.  It is clear that there was significant heterogeneity between different trials.  Some interventions included relevant measures such as blood pressure and cholesterol, but not all.  In keeping with the ‘general health check’ terminology, most were not well focused on specific enquiries and tests but included non-specific searches for any abnormal finding including those suggestive of cancer.

(b)     Nor was there any specification of what follow up action would be invoked in response to any identified abnormality, and the trials considered differed.  It is notable that in most the action triggered by an identified lifestyle risk was merely brief lifestyle advice.

(c)     Most of the trials considered are old, dating from as long ago as the 1960s when understanding of health risks and particularly how they may be modified was at an early stage of development.  Most of the risk reduction measures now used post-date these trials.

(d)     These older trials contribute disproportionately to the overall mortality results, because by definition all of the trials with longer follow up and thus more deaths are older.  The newer trials that are more likely to reflect current practice have such brief follow up periods that it is unlikely that they would have much prospect of showing a significant effect.

(e)     Few of the trials appear to have looked at the effect of a sustained programme of either checks or risk reduction.  The effect of a single piece of lifestyle advice on mortality many years later is likely to be so diluted by other factors as to be undetectable.

A few trials were based on vascular risk factors and reasonably recent, although still not based on protocols that would match current NICE guidance.  The OxCheck trial reported significant improvements in risk in the study group compared with controls.  This study was not designed to look at mortality and none of the referenced publications report on it; the review authors say that they contacted the authors of the study to enquire about death rates and were able to include them in their analysis despite them never having been published.  This unusual approach is unlikely to be considered appropriate in other circumstances.

The other relevant and recent trial that looked at cardiovascular risk is the ongoing Inter99 trial, which has already reported significantly improved mental health and physical health amongst the study group compared with controls, with no detectable mental distress resulting from the intervention.  Self-reported health was significantly better in the study group, which is known to correlate with mortality.  This trial has not reported on mortality to date as sufficient follow up has not yet accrued.

The NHS Health Check Programme was designed as a specified set of questions and investigations focused strictly on a set of related (‘vascular’) conditions, followed by a range of risk reduction measures known to be effective.  The basis for both aspects has been firmly grounded in NICE guidance.  None of the trials reviewed by Krogsbøll et al would match this protocol, even the Inter99 study that most closely resembles it.  The Cochrane review authors conclude that their “…results do not support the use of general health checks…”; the key point is that NHS Health Checks are not ‘general health checks.’

Overall, the review conclusions can have little if any relevance to NHS Health Checks.  What the review does, however, is underline very clearly the need to evaluate the effectiveness of the programme at national level, including outcomes.

It should be noted that assessments and tests were included in the NHS Health Check only if there was cost effectiveness evidence to support their inclusion. The economic modelling also only included interventions for which there is good evidence of their cost effectiveness (see page 9 of the Economic Modelling Report). The relevant NICE guidance is shown in table 1 below (page 10 of the report).

Table 1:

Intervention offered

Existing Guidance

Brief exercise intervention

NICE Guidance PHI002 “Four commonly used methods to increase physical activity”, March 2006

Multi-component weight loss programmes

NICE clinical guideline CG43 “Obesity”, December 2006

IGR intensive lifestyle management

NICE clinical guideline CG43 “Obesity”, December 2006 and Health Technology Assessment 2004; Vol 8: No. 21

Stop Smoking Services

NICE guidance PHI001 “Brief interventions and referral for smoking cessation in primary care and other settings”, March 2006

Anti-hypertensives for those with hypertension

NICE clinical guideline 34 “Management of hypertension in adults in primary care: partial update”, June 2006

Statins for primary prevention

NICE technology appraisal 94 “Statins for the prevention of cardiovascular events”, January 2006

In addition, the following NICE Guidance have been released since the economic analysis was carried out:

Cardiovascular risk assessment and the modification of blood lipids for the primary and secondary prevention of cardiovascular disease - Clinical guidelines, CG67 - Issued: May 2008

Preventing type 2 diabetes: population and community-level interventions in high-risk groups and the general population - Public health guidance, PH35 - Issued: May 2011

Alcohol-use disorders - preventing the development of hazardous and harmful drinking - Public health guidance, PH24 - Issued: June 2010


All operational queries about the NHS Health Check programme, or questions about the Learning Network should be directed to Eleanor Kent-Dyson.  Policy queries should continue to be directed to the NHS Health Check team at the Department of Health using their dedicated mailbox .