OUP user menu

Selecting indicators for the quality of cardiac care at the health system level in Organization for Economic Co-operation and Development countries

Ulla M. Idänpään-Heikkilä, Laura Lambie, Soeren Mattke, Vin McLaughlin, Heather Palmer, Jack V. Tu
DOI: http://dx.doi.org/10.1093/intqhc/mzl028 39-44 First published online: 5 September 2006

Abstract

Background. Cardiovascular (CV) diseases are major causes of morbidity and death in adults in the world. Major differences have been reported in the management strategies and the outcome of CV diseases within and between countries. To better understand and address these differences, there is a need for quantitative information on patient management, outcome, and prognosis.

Objective. This article describes the development of a set of quality indicators for cardiac care and summarizes work undertaken by the Cardiac Care Panel of the OECD Health Care Quality Indicators Project.

Methods. A list of 61 potential indicators was identified through a literature search, review of national measurement systems, and nomination from countries participating in the project. The Cardiac Care Panel then used a modified Delphi process developed originally by RAND to select indicators. Panel members individually rated each indicator on a scale of 1–9 for scientific soundness and importance. All indicators receiving scores of 7 or more for both importance and soundness were included in the final set.

Results. Seventeen cardiac indicators were selected for the final set of indicators from the following areas: acute coronary syndromes, cardiac interventions, secondary prevention, and congestive heart failure.

Conclusions. The final set of 17 indicators selected by the Cardiac Care Panel constitutes a comprehensive set of measures for the most relevant domains of CV care. Nevertheless, gaps remain in the area of primary prevention and in particular in areas with rapidly changing technology and improving treatment options.

  • acute myocardial infarction
  • chronic heart failure
  • coronary artery bypass surgery
  • health care
  • percutaneous coronary intervention
  • quality indicators

In 2003, the Organization for Economic Co-operation and Development (OECD), an intergovernmental economic research institution, launched its Health Care Quality Indicators (HCQI) Project to identify and implement quality indicators, which can be used for international comparisons of the performance of health care systems. Cardiac care was chosen as one of the priority areas for indicator selection because of the high disease burden of cardiovascular (CV) diseases in industrialized countries [1]. They are the major causes of morbidity and death in adults in the world, second in frequency only to mental health disorders.

Progress in new technologies has been of enormous importance in preventing and treating cardiac diseases. There is an abundance of data on effective measures to reduce cardiac mortality. Still, major differences have been reported in the management strategies and outcome of CV diseases [2–5]. There is considerable potential to further reduce CV morbidity and improve patients’ chances of survival. To better understand and address these differences, there is a need for quantitative information on patient management, outcome, and prognosis. Most indicators have so far been constructed for use on institutional level, and there is a lack of measures intended for reporting quality on national system level.

The aim of this study was to make the content of the work of the Cardiac Care Panel of the OECD HCQI Project available to a broader audience. This article describes the selection of a set of quality indicators for cardiac care for comparison and benchmarking between nations and summarizes work undertaken between 2003 and 2004. The Cardiac Care Panel members were five leading experts in the quality of care from five different OECD countries. The panel members had experience in clinical work, research, and policy making.

Methods

Identification of potential indicators

The time and resource constraints on the project dictated that only existing indicators from around the world be reviewed rather than a set be developed de novo. An initial list of 61 potential indicators was identified through a literature search, review of national measurement systems, and nomination from countries participating in the project. A more detailed description of the procedure and methods is found in the paper by Mattke et al. in this special issue and in the OECD technical paper [6].

Review process

To ensure that the indicators chosen cover the most relevant domains of CV care, we developed a conceptual framework that contained the domains’ primary prevention of cardiac disease, secondary prevention of cardiac disease, acute coronary syndromes (ACS), cardiac interventions including percutaneous coronary interventions (PCI) and coronary artery bypass grafting (CABG), and congestive heart failure (CHF).

ACS, such as acute myocardial infarction (AMI) and unstable angina pectoris, are the leading cause of mortality among the CV diseases [2,3]. They cause a major burden of disease and are the single most costly disease for the health system. Despite the promotion of primary prevention and the great improvement in treatment options, successful secondary prevention after ACS remains a cornerstone in reducing mortality and preventing future non-fatal attacks. Both fatal and non-fatal AMIs are 4–7 times more common in patients with previously diagnosed coronary disease [7].

Successful secondary prevention after ACS using effective methods would thus have a significant effect on mortality and prevent non-fatal attacks [8–11]. That the health care system can improve the prognosis of patients who have experienced an ACS has abundant evidence/background. The rate with which appropriate secondary prevention measures, such as β-blocker treatment, are implemented after an ACS event is indicative of quality and provides information about the quality of the care for coronary patients in a country.

On the treatment of ACS, it has been shown in many studies that revascularization by initiating thrombolytic therapy or performing PCI within a few hours after the onset of symptoms greatly reduces mortality [4,12–14].

Despite the abundant evidence and the recommendations, both early treatment and secondary prevention after ACS remain underutilized in many countries, and there is still considerable potential to further reduce coronary heart disease morbidity and mortality and improve patients’ chances of survival.

Elective PCI and CABG are effective treatment measures for ACS and stable coronary artery disease, but the invasive nature of these interventions brings with it operative risk, which may lead to clinically significant complications and even death [15–19]. PCI and CABG should achieve the goal of revascularization of coronary arteries without major clinical complications. Still, there are large differences in post-operative and longer-term mortality between institutions, pointing to differences in the quality of care.

Finally, CHF is a common disease with high morbidity and mortality and substantial implications for resource use. The disease burden associated with CHF is expected to increase markedly with the ageing of the population and the decrease in case fatality rates associated with ACS. Several interventions have been shown to improve patients’ prognosis. For example, appropriate treatment with drugs, such as angio-tensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers, and β-blockers can reduce the risk of death and the combined risk of death and hospitalization [20–22].

Indicators representing all five domains were found among the candidate indicators. It was decided that the Primary Care Panel, one of the four other panels of the HCQI Project, should evaluate the indicators for primary prevention of cardiac diseases.

To select indicators, the panellist used a modified Delphi process developed originally by the RAND Corporation [23]. This process was carried out over a period of 6 months using a series of telephone conferences and email discussions. Panel members individually rated each indicator on a scale of 1–9 for scientific soundness and importance. Panellists were also asked to assess whether the feasibility of an indicator was likely, possible, or unlikely. Median scores and measures of disagreement for scientific soundness and importance for each indicator were calculated and reported back to the panellists, who discussed the ratings and were allowed to change them based on the discussion. Indicators receiving final scores of 7–9 were regarded as robust, 4–6 as equivocal, and 1–3 as weak. All indicators receiving scores of 7 or more for both importance and soundness were included in the final set. In addition, a few indicators that received scores of 4–6 for either or both dimensions were retained if the panellists considered the indicators essential in contributing to the overall balance and comprehensiveness of the final set. If feasibility was considered unlikely, an indicator was dropped, unless strong conceptual reasons existed to retain it, leading to some conceptually appealing indicators being left out, because the panel determined that it would be difficult to define and collect them internationally. The details of the review of relevance, soundness, and feasibility for the individual indicators have been described in the report of the Cardiac Care Panel [6].

Results

Using the above-described rules in the review process, the panel selected 17 indicators for the following areas: ACS, cardiac interventions, secondary prevention, and CHF. The indicators chosen are presented below. A full description of the indicators and their background can be found in Lambie et al. [6].

Indicators on secondary prevention of cardiac disease

Of the nine proposed secondary prevention indicators, the panel selected the four summarized in Table 1. There was high agreement on all four indicators. National guidelines and several expert groups and task forces strongly recommend them for secondary prevention of subsequent CV events in patients discharged after AMI (Table 1).

View this table:
Table 1

Indicators on secondary prevention of coronary heart disease

IndicatorDefinition of the indicatorSource
Aspirin on discharge after acute MINumerator: those prescribed aspirin at dischargeJCAHO1, CMS2, US National Quality Report
Denominator: discharged patients with AMI without aspirin contraindications
ACE inhibitors at discharge after AMINumerator: those prescribed an ACE inhibitor at dischargeJCAHO, CMS, US National Quality Report
Denominator: discharged patients with left ventricular systolic dysfunction and without ACE inhibitor contraindications
β-Blockers at discharge after AMINumerator: those prescribed a β-blocker at hospital dischargeJCAHO, CMS, US National Quality Report
Denominator: discharged patients with AMI without β-blocker contraindications
Statin treatment after a cardiac eventNumerator: people who attend primary care; who have had a cardiac event and who have been prescribed a statinNew Zealand Guidelines on the assessment and management of cardiovascular risk
Denominator: people who attend primary care and who have had a cardiac event

ACS

All selected indicators for ACS were related to the treatment of AMI because the panel agreed that data on unstable angina would be too difficult to collect accurately and compare internationally. Highly rated indicators included the timing of revascularization treatment following AMI, treatment on admission to hospital, and mortality following AMI. In addition, in-hospital mortality rate after AMI had already been selected in previous proceedings of the HCQI Project.

Readmission rates for AMI, although highly rated, were excluded, because of probable difficulties in interpreting data, because some countries readmit for elective revascularization.

The indicators chosen on acute coronary care are summarized in Table 2.

View this table:
Table 2

Indicators on acute coronary syndromes

IndicatorDefinitionSource
Timing of thrombolytics for patients with AMINumerator: number of minutes from time of arrival at hospital to time of administration of the thrombolyticHCFA CCP1
Denominator: patients with confirmed AMI receiving thrombolytics and having adequate documentation of the time of arrival and the time of administration of the thrombolytic
Timing of emergent PTCA for patients with AMINumerator: the time in minutes from arrival at the hospital until the beginning of the PTCAHCFA CCP
Denominator: all patients with confirmed AMI receiving a PTCA within 12 hours after arrival at the hospital and having adequate documentation of the time of arrival and the time of the PTCA
Aspirin at admission to hospital for AMINumerator: number who received aspirin within 24 hours before or after hospital arrivalJCAHO2
Denominator: hospitalized AMI patients without aspirin contraindications
One-year mortality following AMINumerator: number of deaths in any setting that occurred within 1 year of hospital admission for a primary (principal) diagnosis of AMINMDS3
Denominator: number of unique individuals hospitalized with a primary diagnosis of AMI

Cardiac interventions including PCI and CABG

Six indicators related to PCI and CABG, initially collected or suggested by the panellists, were selected for cardiac interventions. PCI and CABG rates, which were suggested as important for providing context to those indicators, are already routinely collected and published as part of OECD Health Data. Indicators on complications other than mortality received low ratings for scientific soundness.

The indicators on coronary interventions are summarized in Table 3.

View this table:
Table 3

Indicators on cardiac interventions

IndicatorDefinition of the indicatorSource
CABG in-hospital mortality rateNumerator: number of deaths per 100 discharges with procedure code for CABG in any field. Age 40 years and olderAHRQ HCUP1
Denominator: all non-maternal/non-neonatal discharges with procedure code for CABG in any field. Age 40 years and older
One-year mortality rate following CABGNumerator: number of people who have had a CABG operation who have died after 1 year of discharge of a CABGNMDS2
Denominator: number of people who have been discharged from hospital who have had a CABG operation
CABG re-operation within 6 months of dischargeNumerator: number of unique individuals undergoing CABG re-operations within 6 months of dischargeMeasure proposed by panel members
Denominator: number of unique individuals discharged following a CABG operation
PTCA in-hospital mortalityNumerator: number of deaths in hospital in patients with PTCAMeasure proposed by panel members
Denominator: number of PTCA performed
Same-day CABG surgery rate after PTCANumerator: number of unique individuals who have had a CABG within 24 hours following a PTCAMeasure proposed by panel members
Denominator: number of unique individuals who have had a PTCA
Repeat PTCA within 30 days of dischargeNumerator: number of unique individuals having a second PTCA performed within 30 days of dischargeMeasure proposed by panel members
Denominator: number of PTCA performed

CHF

There was broad consensus about the three indicators selected for CHF (Table 4). The indicators are recommended in several national indicator sets developed from consensus processes and already in use. Readmission rate was dropped because of data reliability problems.

View this table:
Table 4

Indicators on congestive heart failure (CHF)

IndicatorDefinition of the indicatorSource
Proportion of patients with CHF receiving ACE inhibitor on dischargeNumerator: number of individual patients with a principal diagnosis of CHF (ICD-9 428, ICD-10 I50) who are prescribed an ACE inhibitor at dischargeJCAHO1, CMS2, US National Quality Report3
Denominator: number of individual patients discharged with a principal diagnosis of CHF
Rate of β-blocker prescription at hospital discharge for CHFNumerator: number of individual patients with a diagnosis of CHF (ICD-9 428, ICD-10 I50) who are prescribed a β-blocker at dischargeAHRQ HCUP4
Denominator: number of individual patients discharged with a diagnosis of CHF
CHF in-hospital mortality rateNumerator: number of deaths per 100 discharges with principal diagnosis code for CHFAHRQ HCUP
Denominator: number of discharges with principal diagnosis code for CHF, exclude discharges with cardiac procedure codes in any field

Discussion

The final set of 17 indicators selected by the Cardiac Care Panel, in combination with the four indicators for primary prevention left to the Primary Care Panel, constitutes a comprehensive set of measures for the most relevant domains of CV care. As cardiac care is a field in which the conceptualization of quality and the development of indicators are well advanced, it proved to be possible to select the set of measures from existing sources. Most of the proposed measures are currently in use in national policy planning and provider comparisons, underscoring their operational feasibility.

A potential threat to the validity of the indicators as indicators for comparison is the possible variation in the diagnostic criteria that are used, for instance concerning AMI. The prognosis and risk of recurrence is different in patients with established AMI and acute coronary attack, although both groups benefit from the measures represented by the indicators. Some problems with the diagnosis and coding of CHF have also been identified in the administrative data sets. The relatively higher number of indicators on coronary interventions might depend as much on easier access to data about results as on its importance as a domain.

The Cardiac Care Panel recommends interpreting the indicators in the context of information on CV mortality in different age and risk groups. Some of the information is currently collected in OECD Health Data, but further refinement may become necessary.

Nevertheless, gaps remain in primary prevention and particularly in areas with rapidly changing technology and improving treatment options. No indicator on absolute CV risk, which would capture the additive and multiplicative effects of the various risk factors for CV disease [24], was selected, because the panel decided that this concept was not yet universally accepted and thresholds were still under discussions. The issue may be reconsidered in the future. The panel also agreed on the importance of measuring and comparing blood pressure control but rejected a corresponding indicator because of the difficulty in collecting the required data internationally.

In the areas of new technologies, the changing standards of good practice make it difficult to select indicators at the moment. Among those areas that should be considered are cardiac transplantation and ventricular assist devices. Those treatment options constitute the definite treatment of CHF, a disease of increasing importance in industrialized countries. Also, the treatment of cardiac arrhythmias with percutaneous interventions, devices, such as pacemakers and implantable cardioverter-defibrillators, and drugs has become a growing component of cardiac care. Given the high cost of those interventions, indicators for those treatments should be added in the future. Finally, coronary stenting has developed from an experimental treatment to a standard approach in a matter of a few years, and its growing relevance suggests the need to include measures in this area. Despite those limitations, however, the panel is confident that the selected set of indicators can provide policy makers and researchers with solid comparative information about the quality of cardiac care in industrialized countries.

Footnotes

  • This article describes the key recommendations made by the Cardiac Care Panel of the OECD Health Care Quality Indicators Project. It reflects the opinion of the authors and not an official position of the OECD, its member countries, or institutions participating in the project. The full report of the panel proceedings can be found at http://www.oecd.org/dataoecd/28/35/33865450.pdf

References

View Abstract