Making the most of the Myocardial Ischaemia National Audit Project (MINAP)

Br J Cardiol 2009;16:159–61 Leave a comment
Click any image to enlarge
Authors:

The Myocardial Ischaemia National Audit Project (MINAP) represents one of the largest observational databases of acute coronary syndrome (ACS) events.1-3 Since its inception in 2000, it has accumulated rich data (including timing and method of admission, emergency and subsequent treatments, and long-term mortality data through linkage to the UK Statistics Authority) for over 650,000 ACS events from all acute hospitals (n=228) in England and Wales (figure 1). Initially designed to monitor standards set by the National Service Framework for Coronary Heart Disease4 with the generation of annual reports of hospital-level ST elevation myocardial infarction (STEMI) performance,5 the provision of contemporary online performance analyses has facilitated improvements in the care of ACS patients.6 Moreover, MINAP is more than a resource for the purposes of audit, it is also a key research tool for the evaluation of cardiovascular care and outcomes.7,8 Although it is primarily focused on clinical need, its research potential has been recognised by several grant-giving bodies, and a committee (the MINAP Academic Group [MAG]) dedicated to overseeing MINAP research has been established.3 The Clinical Performance Group (University of Leeds), a multi-disciplinary team comprising clinical cardiologists, health service researchers and health economists draws on MINAP data to investigate clinical care at multiple levels (patient, population, process and healthcare professional).

Missing data

Figure 1. Computed tomography (CT) sagittal reconstruction, two-chamber view. The subepicardial myocardium is thin and normally compacted with a thicker non-compacted subendocardial layer in the anterior wall and apex. Note the artefact from the right ventricular (RV) pacemaker tip
Figure 1. Computed tomography (CT) sagittal reconstruction, two-chamber view. The subepicardial myocardium is thin and normally compacted with a thicker non-compacted subendocardial layer in the anterior wall and apex. Note the artefact from the right ventricular (RV) pacemaker tip

There are, however, justified concerns with regard to MINAP data relating to data quality and completeness of ascertainment. These concerns reflect, in some cases, difficulties experienced by some hospitals with data collection. Systematic differences between patients with and without information recorded may bias the estimated performance of a hospital. Moreover, reliable inferences of ACS care cannot be made concerning events not submitted to the database. Fortunately, there are many ways of handling missing data (albeit the best solution is to prevent its occurrence!). Our Clinical Performance Group is studying methods (such as data imputation, the substitution of some value for missing data) by which MINAP ‘data missingness’ biases may be overcome. Preliminary work suggests hospital-level data missingness (such as the failure to submit a particular variable relating to the patient or their management) may relate to early mortality. These inferences are consistent with findings from the Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC/AHA Guidelines (CRUSADE) National Quality Improvement Initiative9 and Prospective Registry Evaluating Myocardial Infarction: Events and Recovery (PREMIER) study.10 The annual health check undertaken by the Healthcare Commission includes two indicators pertaining to the submission of MINAP data: first, whether a trust has at least 90% completion across the key fields in MINAP, and second, whether it takes part in the annual MINAP data validation exercise.11 Data missingness may, therefore, be both a performance indicator and a health outcome measure.

Feedback to hospitals

Complete and comprehensive data may also be enhanced through the provision of data and analyses to the hospitals who submit them. To date, available case-analysis has permitted the appraisal of evidence-based ACS care across England and Wales. Using statistical process control methods, we have demonstrated that funnel plots may be applied to the MINAP data set to allow visual comparison of performance data derived from hospitals (figure 2).12 Through this methodology variation is readily identified, permitting units to appraise their practices so that effective quality improvement may take place. Anonymised real-time provision of analyses (such as funnel plots) to units submitting data may be one method by which hospitals get feedback. In addition to cross-sectional evaluation of care,13 the shear quantity of data available from MINAP permits longitudinal analyses. For example, it is feasible to evaluate contemporary care practices consistent with national guidelines for ACS management, identify hospital characteristics predictive of adherence to guidelines, and assess whether adherence to guidelines is associated with mortality rates.

Figure 2. CT five-chamber view again showing prominent trabeculations along the lateral and anteroapical wall of the left ventricle
Figure 2. CT five-chamber view again showing prominent trabeculations along the lateral and anteroapical wall of the left ventricle

Indicators of performance

Adherence to ACS guidelines is associated with improvement in outcome.14,15 The CRUSADE database and National Registry of Myocardial Infarction (NRMI) demonstrated a positive association between hospitals that perform well with respect to process measures and survival following acute myocardial infarction (AMI).16,17 Whether these findings are upheld in the UK is yet to be determined, but possible through data from MINAP. Furthermore, with recommendations of primary percutaneous coronary intervention (PCI) for STEMI and early PCI for non-ST elevation myocardial infarction (NSTEMI), the UK ACS performance indicator (attainment of a 60 minute door-to-needle thrombolysis time threshold) seems soon to be no longer appropriate for the evaluation of acute cardiac services. The development and selection of novel and composite indicators18 that strongly predict outcome is required.19 Our group are presently developing and evaluating ACS performance indicators applicable to MINAP.

Performance (such as revascularisation times and attainment of evidence-based drugs on discharge) must be carefully analysed and represented because variation is attributable to many factors.20 Analysis after case-mix adjustment reflects hospital process performance or ‘quality of care’, which is the basis of medical institution profiling21,22 necessary for clinical governance, resource allocation and economic/workforce planning. Fair comparison of hospitals’ performance, therefore, requires careful consideration of case-mix. This is possible through the development of ACS risk models (scores), of which many exist. Using MINAP data, our group recently developed a risk score that discriminated in-patient death for STEMI,23 and externally validated the Global Registry of Acute Coronary Events (GRACE), Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrillin Therapy (PURSUIT), Global Utilization of Steptokinase and t-PA for Occluded Coronary Arteries (GUSTO-I), Simple Risk Index (SRI) and Evaluation of the Methods and Management of Acute Coronary Events (EMMACE) risk scores.24 However, there are many challenges to the development and application of ACS risk models, and isolated case-mix adjustment can lead to the erroneous conclusion that an unbiased comparison between hospitals then follows (the case-mix fallacy).25 Moreover, although MINAP has 118 data fields (not all have to be collected), it does not collect all the predictor variables used in common ACS risk scores. This is one of the weaknesses of MINAP data, and measures to overcome this constitute part of our Clinical Performance Group research programme.

Risk scores

Case-mix adjustment (far-point testing) is only one of the many uses for a validated risk model. Risk models also represent a simple, convenient method of determining the risk characteristics of a patient (near-point testing). Consequently, they facilitate clinical decision making so that patients may receive timely evidence-based therapies. For ACS, early and accurate risk stratification is essential, as the benefits of more aggressive and costly treatments are seen mainly in those at higher risk of adverse clinical events.26-29 In turn, this improves outcomes and optimises resource usage. Traditionally, logistic regression techniques have been used to generate risk scores for medical practice, but MINAP data are extensive and complex and require more sophisticated analyses to optimise model development. Indeed, building good models is not a simple process and a phrase attributed to George Box is often cited: “all models are wrong, but some are useful”.30 The wealth of data available from MINAP will permit the development of a range of near-point and far-point risk models that can be used by healthcare professionals, policy makers and epidemiologists alike.

Making the most…

MINAP has accumulated a vast quantity of contemporary ACS data that allow the investigation of cardiovascular services and outcome throughout England and Wales. This national resource is now in a position to be used for cardiovascular research, but would not have been possible without the assistance of all the hospitals in England and Wales who have contributed data to MINAP. If you wish to make the most out of MINAP, applications for data may be accessed from: http://www.rcplondon.ac.uk/CLINICAL-STANDARDS/ORGANISATION/PARTNERSHIP/Pages/MINAP-.aspx

Alternatively, there are opportunities for MINAP research within the Clinical Performance Group, University of Leeds.

Acknowledgements

The authors gratefully acknowledge funding from the British Heart Foundation. MINAP is funded by the Healthcare Commission.

Conflict of interest

None declared.

References

  1. Birkhead JS. Responding to the requirements of the national service framework for coronary disease: a core data set for myocardial infarction. Heart 2000;84:116–17.
  2. Birkhead JS, Pearson M, Norris RM. The national audit of myocardial infarction: a new development in the audit process. J Clin Excellence 2002;4:379–85.
  3. Royal College of Physicians. Myocardial Ischaemia National Audit Project (MINAP). Available from: http://www.rcplondon.ac.uk/college/ceeu/ceeu_ami_home.htm [accessed January 2009].
  4. DH coronary heart disease policy team. Coronary heart disease national service framework. Building for the future. Progress report 2007. London: The Stationery Office, 2008.
  5. Royal College of Physicians. How the NHS manages heart attacks. Seventh public report of the Myocardial Infarction National Audit Project. London: Clinical Effectiveness and Evaluation Unit, 2008.
  6. Birkhead JS, Walker L, Pearson M, Weston C, Cunningham AD, Rickards AF. Improving care for patients with acute coronary syndromes: initial results from the National Audit of Myocardial Infarction Project (MINAP). Heart 2004;90:1004–09.
  7. Horne S, Weston C, Quinn T et al. The impact of pre-hospital thrombolytic treatment on re-infarction rates: analysis of the Myocardial Infarction National Audit Project (MINAP). Heart 2009;95:559–63.
  8. Birkhead JS, Weston C, Lowe D. Impact of specialty of admitting physician and type of hospital on care and outcome for myocardial infarction in England and Wales during 2004-5: observational study. BMJ2006;332:1306–11.
  9. Dunlay SM, Alexander KP, Melloni C et al. Medical records and quality of care in acute coronary syndromes: results from CRUSADE. Arch Intern Med 2008;168:1692–8.
  10. Schelbert EB, Rumsfeld JS, Krumholz HM et al. Ischaemic symptoms, quality of care and mortality during myocardial infarction. Heart 2008;94(2):e2.
  11. Quality Care Commission. http://www.cqc.org.uk/guidanceforprofessionals/healthcare/nhsstaff/annualhealthcheck2008/09/qualityofservices/exis/participationinheartdiseaseaudits.cfm
  12. Gale C, Roberts A, Batin P, Hall A. Funnel plots, performance variation and the Myocardial Infarction National Audit Project 2003-2004. BMC Cardiovascular Disorders 2006;6:34.
  13. Lawrance RA, Dorsch MF, Sapsford RJ et al. Use of cumulative mortality data in patients with acute myocardial infarction for early detection of variation in clinical practice: observational study. BMJ2001;323:324–7.
  14. Fox KA, Steg PG, Eagle KA et al. Decline in rates of death and heart failure in acute coronary syndromes, 1999-2006. JAMA 2007;297:1892–900.
  15. Granger CB, Steg PG, Peterson E et al. Medication performance measures and mortality following acute coronary syndromes. Am J Med 2005;118:858–65.
  16. Bradley EH, Herrin J, Elbel B et al. Hospital quality for acute myocardial infarction: correlation among process measures and relationship with short-term mortality. JAMA 2006;296:72–8.
  17. Peterson ED, Roe MT, Mulgund J et al. Association between hospital process performance and outcomes among patients with acute coronary syndromes. JAMA 2006;295:1912–20.
  18. Williams SC, Koss RG, Morton DJ, Loeb JM. Performance of top-ranked heart care hospitals on evidence-based process measures. Circulation 2006;114:558–64.
  19. Weston CFM. Performance indicators in acute myocardial infarction: a proposal for the future assessment of good quality care. Heart 2008;94:1397–401.
  20. Lilford R, Mohammed MA, Spiegelhalter D, Thomson R. Use and misuse of process and outcome data in managing performance of acute medical care: avoiding institutional stigma. Lancet 2004;363:1147–54.
  21. Marshall EC, Spiegelhalter DJ. Reliability of league tables of in vitro fertilisation clinics: retrospective analysis of live birth rates. BMJ 1998;316:1701–04.
  22. Austin C. A comparison of Bayesian methods for profiling hospital performance. Medical Decision Making 2002;March/April:163–72.
  23. Gale CP, Manda SO, Batin PD, Weston CF, Birkhead JS, Hall AS. Predictors of in-hospital mortality for patients admitted with ST-elevation myocardial infarction: a real-world study using the Myocardial Infarction National Audit Project (MINAP) database. Heart 2008;94:1407–12.
  24. Gale CP, Manda SO, Weston CF, Birkhead JS, Batin PD, Hall AS. Evaluation of risk scores for risk stratification of acute coronary syndromes in the Myocardial Infarction National Audit Project (MINAP) database. Heart 2009;95:221–7.
  25. Mohammed MA, Deeks JJ, Girling A et al. Evidence of methodological bias in hospital standardised mortality ratios: retrospective database study of English hospitals. BMJ 2009;338:b780.
  26. Cannon CP, Weintraub WS, Demopoulos LA et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban.N Engl J Med 2001;344:1879–87.
  27. Morrow DA, Antman EM, Snapinn SM, McCabe CH, Theroux P, Braunwald E. An integrated clinical approach to predicting the benefit of tirofiban in non-ST elevation acute coronary syndromes. Application of the TIMI Risk Score for UA/NSTEMI in PRISM-PLUS. Eur Heart J 2002;23:223–9.
  28. Morrow DA, Cannon CP, Rifai N et al. Ability of minor elevations of troponins I and T to predict benefit from an early invasive strategy in patients with unstable angina and non-ST elevation myocardial infarction: results from a randomized trial. JAMA 2001;286:2405–12.
  29. Poole-Wilson PA, Pocock SJ, Fox KA et al. Interventional versus conservative treatment in acute non-ST elevation coronary syndrome: time course of patient management and disease events over one year in the RITA 3 trial. Heart 2006;92:1473–9.
  30. Box GEP, Drapes NR. Empirical model building and response surfaces. USA: John Wiley & Sons, Inc., 1987.
THERE ARE CURRENTLY NO COMMENTS FOR THIS ARTICLE - LEAVE A COMMENT