Rapid Responses to:

PRIMARY CARE: Peter Murchie, Neil C Campbell, Lewis D Ritchie, Julie A Simpson, and Joan Thain Secondary prevention clinics for coronary heart disease: four year follow up of a randomised controlled trial in primary care BMJ 2003; 326: 84 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Secondary Prevention Intervention - Consider Cluster Randomisation

Ameet Bakhai, Duolao Wang   (15 January 2003)

Individual GPs might not see the impact of their hard work

Martin Ashton-Key, Martin Ashton-Key, and Thomas J. Scanlon   (22 January 2003)

Secondary Prevention Intervention - Consider Cluster Randomisation 15 January 2003
Ameet Bakhai, Senior Research Fellow Clinical Trials & Evaluations Unit, Royal Brompton Hospital, London, SW3 6NP, Duolao Wang Send response to journal: Re: Secondary Prevention Intervention - Consider Cluster Randomisation	Dear Colleagues, We are encouraged by the results of the study by Murchie et al. [1], which showed how patient adherence to secondary prevention advice may be improved with downstream benefits in terms of clinical outcomes. However as it will have been difficult to keep nurse practioners and patients truly separated at the same practices (as seen by the number of patients that crossed over treatment groups subsequently) we would suggest that some contamination biases will have been present. Such biases are hard to capture or limit unless entire practices were randomised to the intervention or control arm, which was not the case here. Indeed after the first year, a large proportion of the patients (55%) had exposure to secondary prevention clinics and more importantly more may have had the same practioners caring for them earlier. We wonder if it would have been a better trial design approach to have considered cluster randomisation techniques from the onset and performed appropriate statistics on this basis. Failure to control for organisational contamination effects can lead to biased results, often over estimating the statistical significance of the observed effects and increasing the chances of spuriously significant findings [2,3]. This would have more directly answered the key questions raised by this trial of what number of nurse practitioners, resources and commitment are needed to achieve such beneficial results? Furthermore a prospective cost-effectiveness analysis would have been particularly interesting as there was a sizeable life years gained difference, boding well for formal economic evaluation. References : 1. Murchie P, Campbell NC, Ritchie LD, Simpson JA, Thain J. Secondary prevention clinics for coronary heart disease: four year follow up of a randomised controlled trial in primary care. BMJ. 2003 Jan 11;326(7380):84. 2. Campbell MK, Grimshaw JM. Cluster randomised trials: time for improvement. BMJ 1998; 317:1171-1172. 3. Torgeson T. Contamination in trials : is cluster randomisation the answer? BMJ 2001 ;322 :355-357. Competing interests:   None declared
Individual GPs might not see the impact of their hard work 22 January 2003
Martin Ashton-Key, Specialist Registrar in Public Health Kent, Surrey and Sussex Public Health Training Scheme, Martin Ashton-Key, and Thomas J. Scanlon Send response to journal: Re: Individual GPs might not see the impact of their hard work	Editor - Murchie et al’s paper provides good evidence that primary care teams can reduce coronary heart disease by tackling risk factors where pharmacological treatments exist, and where behaviour can be modified.1 However, a key issue is whether individual GPs or other primary care staff will actually see the fruits of their labours; this will influence how primary care implements secondary prevention measures. The data presented show that over the 4.7 years study period there were 25 fewer deaths in the intervention group in the 19 practices that took part in the study; in other words most practices would have seen just one less death from coronary heart disease as a result of their efforts. If the intervention were offered to all eligible subjects, including the control group, the inference might be that these numbers would double. The estimated impact on mortality and coronary events per individual GP in this study are of the same order as we have estimated using published data 2 for GPs in Brighton and Hove if they implement many of the secondary prevention initiatives recommended in the National Service Framework for Coronary Heart Disease.3 The response of local GPs on viewing these data was a mixture of surprise and disappointment. While secondary prevention initiatives in primary care reduce CHD in a wider population, at a practice level many GPs and primary care teams may not notice any discernible impact despite all their hard work. It is important to be honest with health care staff if we are to engage them in this agenda; primary care teams need to appreciate that they have a crucial role to play in reducing coronary heart disease even if they do not witness the rewards of their efforts. 1. Murchie P, Campbell NC, Ritchie LD, Simpson JA, Thain J. Secondary care prevention clinics for coronary heart disease: four year follow up of a randomised controlled trial in primary care. BMJ 2003;326:84-87. 2. McColl A, Roderick P, Gabbay J, Smith H, Moore M. Performance indicators for primary care groups: an evidence based approach. BMJ 1998;317:1354-1360. 3. Department of Health. National Service Framework for Coronary Heart Disease. London: DoH, 2000. Martin Ashton-Key Specialist Registrar in Public Health Kent, Surrey and Sussex Public Health Training Scheme Thomas J. Scanlon Consultant in Public Health Brighton and Hove City Primary Care Trust Competing interests:   None declared