Rapid Responses to:

GENERAL PRACTICE: Gene Feder, Chris Griffiths, Sandra Eldridge, and Matthew Spence Effect of postal prompts to patients and general practitioners on the quality of primary care after a coronary event (POST): randomised controlled trial BMJ 1999; 318: 1522-1526 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

It is time for less talk and more action

Alastair D Short   (17 June 1999)

Several concerns

Claire Shevels   (24 June 1999)

Are postal prompts really ineffective?

Richard Edwards   (7 July 1999)

The POST Study: a failure of Hospitals not GPs

Hugh Bethell   (10 August 1999)

Efficacy of Postal Prompts in secondary prevention of coronary heart disease

Gene Feder   (13 August 1999)

It is time for less talk and more action 17 June 1999
Alastair D Short Send response to journal: Re: It is time for less talk and more action	Sir If the patients in Feder's study (1) had been doctors I suspect that a lot more would have been on drugs for secondary prevention of their ischaemic heart disease. The prescribing of aspirin, B blockers and statins is neither difficult nor time consuming. The recording of risk factors is quick and easy. (2) All these patients must have been in contact with medical care either for follow up or "sick lines". At a time when the future of general practice is under debate (3,4) the current move towards nurse led care for some of the most interesting and worthwhile parts of general practice must be critically evaluated. You need to be able to prescribe! Bradley and Cupples (5) suggest a register, but that exists already in the practices. Only the GPs know who all the patients are and it is from the practice that such initiatives should run, although to be most effective all medical staff have to be committed to prevention wherever they happen to meet the patient. Ideally, there has to be a way in to the system from both primary and secondary care. If there is a resource need then that should be identified and argued on robust data. The Health Service knows these Patients. I doubt if they want to die unnecessarily and the response rate to Feder's study was good, suggesting that the patients want such intervention. It is time for less talk and more action or in five years time will another study demonstrate that nothing has happened? The challenge to change persists. (6) Yours sincerely Dr Alastair D Short General Practitioner Anniesland Medical Practice 778 Crow Road Glasgow G13 1LU No conflict of interest 1 Feder G, Griffiths C, Eldridge S, Spence M.Effect of postal prompts to patients and general practitioners on the quality of primary care after a coronary event (POST): randomised controlled trial. BMJ 1999;318:1522-6 2 McKinlay J, Short A.D., An Audit of Secondary Prevention in Patients with established Coronary Heart Disease. Health Bulletin 56(2) March 1998;pp 592-601 3 Lipman T, Is there a clinical future for the general practitioner? (letters) BMJ 1999;318:1420 4 Hennell T, Role of general practitioners in NHS must not be undervalued. (letters) BMJ 1999;318 1420 5 Bradley F, Cupples M, Reducing the risk of recurrent coronary heart disease (Ed) BMJ 1999;318:1499 6 Short AD, West B Early Management of Myocardial infarction, The Challenge is to Change. BMJ 1994;308:1159
Several concerns 24 June 1999
Claire Shevels, 3rd Year Medical Student University of Newcastle-upon-Tyne Send response to journal: Re: Several concerns	24/6/1999 Editor, Feder et al in their study examined postal prompts in secondary prevention of coronary heart disease in primary care which is an important issue. However the results of their study are not unexpected. As the authors have stated in their introduction a multifaceted approach has already been shown to be more effective than a single intervention. We wondered why the authors then go on to test a single method. Methods used in the study also leave some questions unanswered. Firstly we recognise that the authors have adjusted for practice size, number of partners, training status and number of practice nurses, and have identified smoking, diabetes and beta blocker prescription as variables between each practice. However cultural and socio-economic factors have not been addressed. Since in an area such as Hackney there are likely to be pockets of severe deprivation this may be related to particular practices and influence results. Secondly, the authors have excluded patients who died within six months of discharge. We realise there may have been difficulties obtaining data regarding these patients but it might have been useful to know the distribution of deaths between the two groups. We feel that the authors did achieve their objectives but were unrealistic in the principle outcome measure that they specified. The postal prompts were effective in increasing patient consultation rates but the authors instead identified their principle outcome measure as beta- blocker prescription. We feel this is a separate issue which may be more related to GP training, awareness and availability of the pre-existing East London guidelines. Finally, the study showed the intervention group of GPs recorded giving lifestyle advice more often than control group GPs. However this did not have an effect on patients self-reported lifestyle changes. Therefore we agree with the authors that this may be an implication for future research into why GP consultations regarding lifestyle changes are apparently ineffective. Louise Li, Lesley Maher, Louisa Pollock, Claire Shevels, Ben White (b.w.white@ncl.ac.uk) 3rd year medical students Department of Epidemiology and Public Health Medical School, University of Newcastle-upon-Tyne.
Are postal prompts really ineffective? 7 July 1999
Richard Edwards, Lecturer in Public Heatlh Medicine Department of Epidemiology and Public Health, University of Newcastle upon Tyne Send response to journal: Re: Are postal prompts really ineffective?	We read the POST study with great interest. Given the evidence presented in this well designed cluster randomised trial, we were a little surprised that the investigators concluded that postal prompts to general practitioners have a 'marginal role' in improving the secondary prevention of coronary heart disease. The authors found that for all except one of the measures of risk factor recording and advice were significantly increased, some such as recorded cholesterol measurement dramatically so. Also, for their principal prescribing outcome measures (B-blockers and cholesterol lowering drugs), the odds ratio was non-significantly raised to 1.7 - which, if real, seems clinically significant. The failure to detect a statistically significant difference may reflect the absence of a real effect or simply a type 2 error due to insufficient power. However, we agree that the overall level of prescribing of B-blockers and cholesterol lowering drugs in both arms of the study was disappointing. We believe that the finding that postal prompts were effective in influencing a range of process measures suggests that rather than dismissing the intervention as ineffective, we should be exploring further the reasons for the more modest effect on prescribing. The challenge then is to devise interventions or introduce policies which address barriers to the implementation of evidence based practice. For example, anecdotal evidence from discussions with GPs in Northumberland suggest that the failure to prescribe Statins for cholesterol lowering is rarely lack of knowledge of best practice. Rather GPs express concerns about the cost implications of long term prescribing of relatively expensive drugs to a significant proportion of their practice population. If such systematic barriers exist then no amount of prompting, postal or otherwise, is likely to bring about the adoption of best practice. Richard Edwards Lecturer in Public Health Medicine Department of Epidemiology and Public Health University of Newcastle upon Tyne Paul Murphy Primary Care Information Manager Northumberland Health Authority Morpeth
The POST Study: a failure of Hospitals not GPs 10 August 1999
Hugh Bethell Send response to journal: Re: The POST Study: a failure of Hospitals not GPs	The POST study1 (BMJ 5th June 1999) indicates that there are considerable deficiencies in the management of myocardial infarct and angina patients discharged from a district hospital and, unfairly, implies that it is the general practitioners who are failing. Most GPs will continue hospital-initiated treatment for patients with chronic diseases. The small proportion of patients in the POST study on beta-blockers and lipid-lowerers must reflect hospital management which may not include a cardiologist's input. In most district hospitals acute infarct patients are admitted under the general physician on call whose main interest may be chest disease, endcrinology, renal disease, gastroenterology or just possibly cardiology (and there are still district hospitals without a cardiologist or physician with an interest in cardiology). A proportion of these patients may never be seen by a cardiologist. We have investigated this problem by a survey of all the coronary care units in the UK. In January 1998 we circulated all the coronary care units (CCU) in the UK with a questionnaire enquiring whether infarct patients were admitted to a CCU, ITU or a general ward, under whose care they were admitted and whether those admitted under a general physician were reviewed by the cardiologist. We also asked whether patients had access to cardiac rehabilitation and what routine follow-up arrangements were made. 325 CCUs were identified and 244 (75%) replied, 3 of which had ceased to exist. The doctor in charge of the CCU was a cardiologist or physician with an interest in cardiology in 206 (85%). In only 8 (3%) were patients admitted to a general ward. In 96 (40%) patients were admitted under the care of either the cardiologist or physician with an interest in cardiology. In 84 (35%) admission was shared by the duty physician and the cardiologist and 60 (25%) was under the sole care of the general physician. In 18 (6%) patients did not have access to a cardiologist's opinion. Follow-up was provided by a cardiac clinic in 51 (21%), medical out- patients in 103 (32%) and a specific cardiac rehabilitation clinic in 42 (17%). In 81 (34%) outpatient follow up was provided by the admitting physician who was sometimes the cardiologist. In those units which responded the great majority of infarct patients were admitted under a cardiologist or had access to a cardiological opinion (which is not the same thing as getting it), but still an appreciable minority did not. If general practitioners are to provide a high standard of long term care for their coronary patients they need to start from the best vantage point. To ensure this we believe that all infarct patients should have access to a cardiological opinion (and where possible receive it) to ensure that they are discharged on the most appropriate medication and receive further investigation when needed. Hugh Bethell, Robin Graham & Marilyn Wallwork and the secondary prevention & rehabilitation advisory committee. 1. Feder G, Griffiths C, Eldridge S, Spence M. Effect of postal prompts to patients and general practitioners on the quality of primary care after a coronary event (POST): randomised controlled trial. Brit Med J 1999;318:1522-6.
Efficacy of Postal Prompts in secondary prevention of coronary heart disease 13 August 1999
Gene Feder, Professor of Primary Care Dept of General Practice and Primary Care, St Bartholomew''s and the Royal London School of Medicine Send response to journal: Re: Efficacy of Postal Prompts in secondary prevention of coronary heart disease	We were pleased to note the interest shown by correspondents following the publication of the POST trial1 and would like to respond. Clare Shevels and colleagues ask: why we tested a "single intervention" when a "multi-faceted approach" is more likely to be effective; whether differences in practice deprivation between intervention and control practices may have confounded our results; and, what was the distribution of deaths between intervention and control groups before 6 months. They and Richard Edwards question our choice of prescribing beta-blockers and lipid lowering agents as main outcome measures saying that we were unrealistic in expecting change in prescribing rates and should take heart from the increased consultation and risk factor recording rates. Our intervention is, in fact, multi-faceted because it adds to a pre- exisiting intensive programme of guidelines facilitation,2 targets both clinicians and patients, and provides recording templates for intervention practices for use during consultations.3 Mean deprivation scores are the same for intervention and control practices: 41.7 (range: 34.9,47.6) and 42.8 (34.8,46.9) respectively, deaths before 6 months are reported in the trial profile flow diagram (www.bmj.com/cgi/content/full/318/7197/1522/DC1). We welcome the improvement in consultation and risk factor recording rates for this group of patients, but think that studies of health service interventions need to measure outcomes that make a difference to patients' morbidity or mortality. Consultation rates and risk factor recording remain process measures of care, while prescribing of beta blockers and lipid lowering drugs are intermediate outcome measures. We believe that our trial has an essentially negative result because we did not show changes in these outcomes. Gene Feder, Professor of Primary Care Chris Griffiths, Senior Lecturer Sandra Eldridge, Statistician Matthew Spence, Research Officer Reference List 1. Feder G, Griffiths C, Eldridge S, Spence M. Effect of postal prompts to patients and general practitioners on the quality of primary care after a coronary event (POST): randomised controlled trial. BMJ 1999;318:1522-1526. 2. Mott S, Feder G, Griffiths CJ, Donovan S. Coronary heart disease in general practice guidelines. Practice based audit: results from a dissemination and implementation programme. Journal of Clinical Effectiveness 1998;3:1-4. 3. Feder G, Griffiths C, Highton C, Eldridge S, Spence M, Southgate L. Do clinical guidelines introduced with practice based education improve care of asthmatic and diabetic patients? A randomised controlled trial in general practices in east London. BMJ 1995;311:1473-1478.