Inequity of use of implantable cardioverter defibrillators in England: retrospective analysisBMJ 2005; 330 doi: https://doi.org/10.1136/bmj.38337.635648.82 (Published 24 February 2005) Cite this as: BMJ 2005;330:454
- Julie Parkes (), MRC clinical training fellow1,
- Deborah L Chase, research fellow1,
- Andrew Grace, consultant cardiologist2,
- David Cunningham, technical director, central cardiac audit database3,
- Paul J Roderick, senior lecturer1
- 1 Health Care Research Unit, University of Southampton, Southampton SO16 6YD,
- 2 Papworth Hospital, Cambridge,
- 3 NHS Information Authority, Tavistock House, London
- Correspondence to: J Parkes
- Accepted 14 December 2004
Sudden cardiac death occurs in approximately 100 000 people annually in the United Kingdom and can be prevented by implantable cardioverter defibrillators (ICDs).1 Rates of implantation of ICDs in England have been increasing but lag behind those in other western European countries and North America. The National Institute for Clinical Excellence has recommended indications for use in patients with ventricular arrhythmias and proposed an annual implantation rate of 50 per million population.2 We present data on current use, geographical and social equity, and barriers to care in the provision of ICDs in England.
Methods and results
We used the national pacemaker and ICD database of implantations done in 1998-2002 to derive national rates. We used a dataset from 1998 to 2000, in which improvements in the quality and completeness of data had been made, to derive rates of implantation of new ICDs by English health region. We calculated indirect age and sex standardised ratios of ICD use by health region by using regional population data divided into five year age bands (0 to 85+ years). We assessed equity by using proxy measures for need for ICDs—namely, regional standardised mortality ratios for ischaemic heart disease—and population fifths of deprivation determined using the Townsend index at census ward level and 1991 census data. We did a national postal questionnaire survey of all recorded ICD centres in England to establish perceived barriers to implantation.
The crude rate of implantation of new ICDs in England rose from 12.4 (95% confidence interval 11.5 to 13.5) per million in 1998 to 30 (28.7 to 31.7) per million in 2002. Regional standardised ratios of use ranged from 0.6 to 1.25 (1). Significant regional differences in standardised rates of implantation existed (χ2 for heterogeneity, P = 0.005), although we found no consistent geographical pattern. Differences between implantation and need in five out of eight regions (95% confidence intervals for standardised ICD implantation and standardised mortality ratio for ischaemic heart disease did not overlap) suggested inequity. A significant inverse relation existed between standardised ICD implantation and fifths of deprivation (P = 0.005, test for trend using a Poisson regression model), ranging from 1.09 to 0.85 (least to most deprived), indicating that an inverse care law may be operating.
The survey response rate was 74% (26/35). The three most commonly perceived barriers to care for patients eligible for an ICD were identification of patients and referral to implanting centres, staff capacity, and funding for treatment. All of the respondents recorded that they expect to see a large increase in demand for ICDs in the future.
Use of ICDs varies between English health regions, and use is not commensurate with need. Although incomplete data could be contributing, an inverse care law seems to be operating. This, along with the slow diffusion of the technology and setting of services predominantly in larger tertiary centres, is similar to the pattern previously seen for coronary revascularisation.3 4
Demand for ICDs will probably increase in the future, particularly in view of expanding indications with randomised evidence of the benefits of ICDs in post-myocardial infarction patients with a low left ventricular ejection fraction.5 Planned expansion of implanting centres and resources are needed to tackle low levels of referral, geographical and social inequity, and the expected increase in demand for ICDs. Strategies should include referral guidelines and targeted education to ensure appropriate identification and referral of eligible patients. These analyses highlight the value of robust national data to inform service development and the need for adequate resources to collect and analyse such information.
What is already known on this topic
Implantation of implantable cardioverter defibrillators in England lags behind most western European and North American countries
What this study adds
An inverse care law seems to be operating on implantation of new implantable cardioverter defibrillators in England
Demand for implantable cardioverter defibrillators is likely to increase in the near future, and a pressing need exists to tackle any inequity and perceived barriers to care
This article was posted on bmj.com on 4 February 2005: http://bmj.com/cgi/doi/10.1136/bmj.38337.635648.82
We thank Scot Harris for statistical support and Morag Cunningham, administrator of the national pacemaker and ICD database.
Contributors JP led the project, cleaned the dataset, did the analyses, constructed the questionnaire and conducted the survey, wrote the first draft of the paper, and is the guarantor. DLC did the analyses, conducted the survey, and helped in writing the paper. AG helped with the survey, commented on drafts, and provided clinical perspective and support to the study. DC is project leader of the national pacemaker and ICD database,provided the 1998-2000 dataset in which data quality and completeness had been improved, and commented on drafts of the paper. PJR oversaw the project, supervised JP and DLC, supplied epidemiology expertise for analyses, and commented on drafts of the paper.
Funding AG and JP are grant holders of HTA grant 93/23/04 (a review of the evidence on the effects and costs of implantable cardioverter defibrillator (ICD) therapy in different patient groups, and modelling of cost effectiveness and cost utility for these groups in a UK context). Professor Martin Buxton is the principal investigator of this study. DLC is funded by an NHS South East Research and Development Fellowship.
Conflict of interest None declared.
Ethical approval Not needed for this study, as it used aggregated anonymised data and no patient contact.