|Home | About | Journals | Submit | Contact Us | Français|
Effective improvements in health care require methods to evaluate professional practice. Azeem Majeed, Helen Lester, and Andrew Bindman examine the assessment of quality
The quality of services provided by primary care doctors varies widely, and there is often a large gap between optimal primary care services and actual practice.1 This quality gap can have serious health consequences, including deaths from medical errors, increased rates of complications in chronic disease, hospital admissions for adverse drug reactions and interactions, and outbreaks of potentially preventable infectious diseases such as measles. It also has large financial costs for the healthcare system, national governments, and society, as well as affecting patients' quality of life.
The reasons for the quality gap are not always within the doctors' control. Sometimes the cause can lie with the public—for example, parents who refuse to allow their child to receive the measles, mumps, and rubella vaccine because of concerns about side effects. Even when the doctor and patient agree to follow a healthcare plan that meets the highest standard for quality, structural barriers related to the design or financing of healthcare systems can prevent the timely receipt of that service—for example, screening mammography for an appropriately aged woman. Nevertheless, the focus of this article and others in the series is on measuring the performance of doctors. Causes of the quality gap that lie with the doctor include being unaware of best practice and the latest guidance on managing a condition or being wary about using certain interventions, such as warfarin to reduce the risk of cerebrovascular disease, because of the fear of adverse events.
The Institute of Medicine defines quality as: “The degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.”2 To measure how well health services meet this goal, a range of performance indicators (sometimes described as quality indicators or quality measures) have been developed.3
Indicators are measurable elements of practice for which there is evidence or consensus that they reflect quality and hence help change the quality of care provided. Indicators are often based on routinely collected data, data from electronic medical records, and sometimes data from surveys.4
In England in the 1990s, the use of performance indicators initially developed ad hoc, with different regions developing their own indicators. The introduction of performance indicators was accompanied by various other quality improvement initiatives including a series of national service frameworks, which set out objectives for the health service, and the establishment of the National Institute for Clinical Excellence (now the National Institute for Health and Clinical Excellence), which provides guidance on promoting good health and preventing and treating ill health.
During the past decade, the development and implementation of performance indicators has been largely driven by an increased interest in the quality of care and the arrival of computerised administrative and clinical databases that, for the first time, could provide routine information on quality. Performance indicators have become increasingly sophisticated—for example, moving in the UK from relatively simple indicators based on administrative or claims data to more sophisticated measures based on clinical information from electronic medical records. In the United States, the development of quality measurement was initially driven by the rapidly increasing costs of health care and purchasers' need to know they were getting value for money. Other important factors were the desire to make performance data available publicly and developments in health informatics, which have reduced the cost of producing performance indicators while steadily increasing their sophistication.5
In April 2004, the UK government took the bold step of introducing standardised performance indicators across the country and linking performance to general practitioners' pay.6 The quality and outcomes framework in the resulting new contract for general practitioners includes a range of performance measures for clinical, organisational, and other areas (such as cervical screening and contraceptive services) and also patient experience. Early results suggest that most general practices achieved high scores across the different parts of the framework. However, as indicators within the framework change and thresholds for achievement alter, we may begin to see greater variation between practices in measured quality of care.
Public disclosure of performance data is becoming increasingly common in both the UK and the US. In the UK, practice data from the quality and outcomes framework is published online, giving the public access to standardised information on general practices for the first time (www.qof.ic.nhs.uk/). At present, this does not seem to have a large role in how patients choose their general practice, although public disclosure of performance data has been shown to encourage provider organisations to improve quality.7 However, use may change as the range of information on general practices increases and patients become more skilled at using the internet to view performance data.
Performance monitoring systems and performance indicators have largely been developed in a manner that makes them unique to each country. This means it can be difficult to compare quality of care and to transfer performance indicators directly between different health systems, clinical practices, and cultures.8 However, although some indicators will inevitably be country specific, others, such as glycated haemoglobin concentrations in diabetic people or percentage of patients with coronary heart disease prescribed statins, need not be, and could be designed in a way that makes international comparisons possible.
Quality indicators can also be used to benchmark performance across countries to gain insights into what is achievable and how to improve quality.9 However, this requires investment in information systems to support measurement of quality. For example, it would have been difficult to implement a system of performance indicators throughout the UK without widespread computerisation of medical records in primary care. Furthermore, although the practice of primary care has many similarities in different countries, differences in the way in which clinical data are collected and coded complicates comparisons.10 For example, the UK uses Read codes, the US uses international classification of disease (ICD-9) codes, and many other countries use international classification of primary care (ICPC) codes. An internationally accepted set of data standards for coding diagnoses and other clinical data is needed as a first step towards routine comparisons of quality of health care across countries.
Information systems that can produce comparable information on process and outcomes of care are also needed to enable international comparisons. The need for substantive baseline data to compare change in quality against pre-existing trends, and the development of supporting educational strategies for health professionals are also issues that other countries may wish to consider if introducing a national system of performance indicators.
Finally, using performance indicators also has some potential adverse consequences. These include doctors declining to accept patients who could be difficult to manage; overtreatment of patients who may not benefit greatly from an intervention; and neglect of areas not covered by performance monitoring. Doctors may also have to spend more time on collecting the performance data and less on dealing with patients. However, despite the pitfalls,11 performance measures in primary care are here to stay and will be used increasingly for quality improvement and performance management in the UK, Europe, US United States, and elsewhere. Other articles in this series will discuss how quality measures have been used in the United States; the patient perspective on measuring quality; whether quality of care is determined by more than what is measurable; and future directions in measuring quality in primary care.
This is the first article in a series looking at use of performance indicators in the UK and elsewhere.
This series is edited by Azeem Majeed, professor of primary care, Imperial College London (email@example.com) and Helen Lester, professor of primary care, University of Manchester (firstname.lastname@example.org).
Competing interests: AM's department has received funding for work on developing methods of measuring quality of care from the Department of Health and Dr Foster Intelligence. HL provides academic advice to the BMA and employers' negotiating teams on the development of the quality and outcomes framework.
Contributors and sources: AM has research interests in the measurement of healthcare quality using administrative and clinical databases. HL has written about pay for performance and has a long research interest in health quality and inequalities. AB does performs research on the impact of polices on low-income persons' access to and quality of care in the United States. This article was based on previous reviews of the measurement of healthcare quality completed by the authors.
Provenance and peer review: Commissioned; externally peer reviewed.