Rapid Responses to:

GENERAL PRACTICE: U N Premaratne, J A C Sterne, G B Marks, J R Webb, H Azima, and P G J Burney Clustered randomised trial of an intervention to improve the management of asthma: Greenwich asthma study BMJ 1999; 318: 1251-1255 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Greenwich asthma study

Mike Thomas   (12 May 1999)

Designing studies to test guidelines and educational interventions in primary care

Chris Griffiths   (16 May 1999)

Greenwich asthma study

Helen Parnell   (12 July 1999)

Greenwich asthma study 12 May 1999
Mike Thomas Send response to journal: Re: Greenwich asthma study	Editor It is disappointing that the Greenwich asthma study failed to show improved outcomes from the establishment of a local asthma resource centre to support practice nurses treating asthma in primary care (1). While the study does demonstrate that an untargeted approach of this nature is unlikely to be effective, it remains possible that different organisation and specific targeting of similar interventions could be effective. Several points in relation to this study could have a bearing on this. Firstly, The intervention appears to have been initiated from and based in the secondary care sector. The project was supervised by a respiratory physician and based at a hospital, and no mention of involvement of GPs or Practice Asthma Nurses in the establishment of the asthma resource centre or the planning of the intervention is mentioned. It is possible that involvement at the outset of the project of interested and motivated primary care workers, particularly those with direct experience of local needs, could have resulted in a different design for the intervention. Secondly, the intervention was directed only at the practice asthma nurses and not at the GPs. It is to be hoped that most Practice based asthma clinics are multi-disciplinary and involve both professional groups. At the present moment however, Practice nurses are unable to prescribe or initiate pharmacological treatments without the involvement and approval of their GP colleagues. If it was felt that there was a general under prescribing of inhaled prophylactic medication in the district, then it would have been appropriate to target the GPs in these practices. Thirdly, while we are told that the control and intervention practices were stratified by whether or not the practice nurse had attended an asthma training course, there is no sub-group analysis on this variable in the outcome measures. There are now several excellent and detailed training programmes for primary care asthma nurses, at the end of which the nurses should be well educated in the management of asthma in primary care according to the BTS guidelines (2). It is to be hoped and expected that the nurses who had attended such a programme should be well versed in such management, and so the care provided would be less affected by the specialist nurse support; those who had not undergone such training however may have been helped and supported more effectively by the intervention. Fourthly, the presence in the practice of a GP with a particular interest and involvement in the asthma clinic may make the additional support provided less effective, and an analysis of outcomes by this variable could possibly show a greater effect from the intervention in some groups. There remains considerable variation at the practice level in the organisation and provision of care to asthmatic patients in primary care, and this will need to be taken into account when allocating resources in this area. It remains possible that a local asthma resource centre could improve treatment and quality of life for asthma patients in primary care, but it does appear that greater targeting of support and more precise assessment of needs, possibly from within the local primary care community itself, are needed in future projects of this nature. Mike Thomas General Practitioner Minchinhampton Surgery Minchinhampton Stroud Glos GL6 9JF 1 Premaratne UN, Sterne JAC, Marks GB, Webb JR, Azima H, Burney PGJ. Clustered randomised trial of an intervention to improve the management of asthma: Greenwich asthma study. BMJ 1998;318:1251-5 2 British Thoracic Society Guidelines on the management of asthma. Thorax 1993;48:1-24S
Designing studies to test guidelines and educational interventions in primary care 16 May 1999
Chris Griffiths, Senior Lecturer Dept General Practice and Primary Care, Queen Mary and Westfield College, London Send response to journal: Re: Designing studies to test guidelines and educational interventions in primary care	The Editor The British Medical Journal London 15/05/99 Designing studies to test guidelines and educational interventions in primary care Dear Sir Premaratne et al. have amassed an impressive amount of data in the Greenwich asthma study.1 However, their choice of outcome measures made it unlikely that they would be able to detect an effect of their intervention. We believe they had an unreal expectation of the impact of the intervention in primary care, perhaps reflecting the absence of primary care researchers in the study team. Designing large scale studies to test guidelines and educational interventions is not easy, particularly in inner city general practice.2 When testing whether nurses improve care of asthma patients in general practice, appropriate primary outcomes are measures of asthma control and health service use in the patients who consulted with the nurses, rather than in the wider population of asthmatics. It seems harsh to judge the efficacy of the study nurses by their lack of effect on patients they did not see, although it is arguable that asking asthma patients to attend review sessions is part of their role. Whilst mean square root of quality of life of the total asthmatic population is clearly an important (secondary) outcome, it is one whose significance will be opaque to many readers. Even the secondary outcomes chosen by the authors (steroid prescribing, A&E attendance, hospital admission) do not use as a denominator the patients who consulted during the study, nor is it clear that the study was sufficiently powered to detect differences for these variables. Focusing on the questionnaires or clinical records of these patients or a pre-determined high risk group would have told the authors much more about the efficacy of the intervention. Most studies demonstrating the efficacy of education and guided-self management have been carried out in hospital or out-patient populations, i.e. among asthmatics with more severe disease. It is perhaps not surprising that an intervention judged on its efficacy among all asthmatics registered in general practice, even those on no medication, might fail to show a benefit. Practice nurse intervention is not necessarily appropriate for all asthmatics. Little data is presented about the delivery of a complex health service intervention. For instance, interactive educational methods may be more effective than simple information transfer when implementing asthma guidelines.3 A qualitative analysis unpicking 'the black box' of this educational programme would have been instructive.4 Without this it may be too early to suggest, as the authors do, that doctors might be more effective than nurses in the role of educating asthma patients. Should the results of the study influence government policy? To conclude that primary care nurses trained by hospital specialist nurses are ineffective would be premature until other studies with more focused outcome assessment and cost effectiveness data have been reported. Reference List 1. Premaratne UN, Sterne JA, Marks GB, Webb JR, Azima H, Burney PG. Clustered randomised trial of an intervention to improve the management of asthma: Greenwich asthma study. BMJ 1999;318:1251-1255. 2. Griffiths CJ, Feder G. Clinician education - a key to implementing asthma guidelines? Quality in Health Care 1999;(in press) 3. Clark NM, Gong M, Schork MA, et al. Impact of education for physicians on patient outcomes. Pediatrics 1998;101:831-836. 4. Bradley F, Wiles R, Kinmonth AL, Mant D, Gantley M. Development and evaluation of complex interventions in health services research: case study of the Southampton heart integrated care project (SHIP). The SHIP Collaborative Group. BMJ 1999;318:711-715. Yours sincerely Gene Feder, Senior Lecturer Chris Griffiths, Senior Lecturer Gill Foster, Research Assistant Shamoly Ahmed, Research Assistant Dorcas Maclaren, Researcher Yvonne Carter, Professor of General Practice
Greenwich asthma study 12 July 1999
Helen Parnell Send response to journal: Re: Greenwich asthma study	Editor Premaratne et al's studyhighlights the difficulties encountered in implementing British Thoracic Society's guidelines for asthma in Greenwich,¹ an inner city area which has a high incidence of urban deprivation with all its attendant problems known to have an impact on asthma and quality of life.² This study was taking place at a time that was seeing an explosion of asthma education for practice nurses and it would have been useful to compare asthma knowledge and services within the control and intervention groups. The authors acknowledge that there was an initiative by the FSHA targeting the under use of inhaled steroids in general practice, affecting GP prescribing and that an unusually high turnover of Practice nurses within the intervention group had an impact on the service provided . In view of these influences it seems unlikely that the control group was untainted and the study results become difficult to interpret. Only 26% of registered asthmatic patients were seen by nurses in the intervention practices. The project therefore failed to target a large number of asthmatics (74%). We do not know if these patients were being seen by GP's, failed to keep appointments, were too mild for their asthma to be a problem, or were more severe/uncontrolled and more likely to be admitted to hospital or attend A&E. Conclusions have been drawn therefore from a relatively small sample of the total number of asthmatics available, a group the authors suggest had relatively mild disease compared to other studies. Although stated that the intervention was integrated across both primary and secondary care there is no evidence to suggest that it was anywhere but in the primary sector. There is no mention of any input involving secondary care staff or patients. Previous studies have illustrated that those asthmatics attending A&E in inner city areas are a notoriously difficult target for asthma care and are often poor at attending for follow up.³ No mention is made of how many of those seen in general practice as part of the intervention were repeated A&E attenders or vice versa. What this study does illustrate are the difficulties encountered in providing health care within inner city areas, where staff can become demoralised and face tremendous difficulties in targeting care effectively. It poses more questions than it answers and highlights the need for rigorous protocols in such studies which need to target specific, easily measured outcomes. Helen Parnell RGN Dip N Clinical Nurse Specialist Dr NT Cooke Bsc MD FRCP Consultant Chest Physician The Airways Clinic St Helier Hospital Wrythe Green Lane Carshalton Surrey SM5 1AA Rreferences 1 Premaratne UN, Sterne JAC, Marks GB, Webb JR, Azima H, Burney PGJ. Clustered randomised trial of an intervention to improve the management of asthma: Greenwich asthma study. BMJ 1999;318:1251-1255 (12 June). 2 Burr L. Social Deprivation and Asthma. Respiratory Medicine 1997. Vol 91 : 603-8. 3 Garrett JE, Mulder J, Wong-Toi H. Characteristics of asthmatics using an urban accident and emergency department. NZ Med J 1988:101: 359- 361.