Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Managing to make a difference

Background and intervention
Results
Comment

Ten years ago the very first issue of Bandolier concerned itself with the topic of getting research into practice, or GRiP. Getting a grip on management and quality of healthcare services has become a core issue for many health services since then. The trouble is that quality is an awfully difficult thing to measure, and outcomes are hardly ever affected by a single technology. Most of what we do is to use complex packages of care with many technologies.

Things do get better, though. Bandolier 82, for instance, showed how handwashing practice in a Geneva hospital was improved to reduce hospital acquired infection. Bandolier 100 examined improvements in treatment after heart attacks in South Derbyshire. But there are few reports about system-wide changes that make a difference. Now we have one [1] from the Veterans Affairs administration in the USA that shows that major changes can make major differences.

Background and intervention

After criticism of the service in the early-mid 1990s, the VA launched a major re-engineering of its health care system, aiming to use better information technology, measurement and reporting of performance, integration of services and realigned payment policies. The years 1994 and 1995 were used as baselines, and the present study [1] examined a number of quality of care markers over succeeding years to investigate how the re-engineering made a difference. It also examined comparable national data from the Medicare programme.

Indicators chosen were of various sorts. In preventive medicine they included cancer screening and vaccination. Outpatient care was examined looking at diabetes care, for instance, with annual eye examinations and glycosylated haemoglobin (HbA1c) measurements. For inpatient care indicators included aspirin immediately after heart attack and at discharge, and smoking cessation counselling in hospital. In all 17 indicators were used.

Results

About 48,000 patients were included in the baseline years, with more patients in subsequent years. Performance for 13 indicators where information was available for all years rose substantially, with statistically highly significant improvements in 12 of them. Examples for influenza vaccination and screening for breast and cervical cancer are shown in Figure 1, and for outpatient diabetes and aspirin on discharge after a heart attack in Figure 2.

Figure 1: Changes in VA healthcare - 1


















Figure 2: Changes in VA healthcare - 2

















For 11 indicators, comparison was possible with Medicare in the years 1997-1999 and 2000-2001. In each case quality performance was higher in the VA than Medicare, sometimes substantially so.

Comment

Of course there are all sorts of limitations with this sort of retrospective study. It could have been that there was a general change in US healthcare, and that the VA improvement simply mirrored this change. The comparison with Medicare implies that this was not the case. There may have been sampling differences, or age structure differences, though this was unlikely. And perhaps the indicators chosen just happened to be easy targets.

The most likely reason for the change was the operational reorganisation and implementation of quality management principles inside the VA. In this, performance contracts held managers accountable for meeting improvement targets, with data gathering and monitoring performed by an independent agency, and made public. This was made possible by instituting an integrated, comprehensive electronic medical records system. Not mentioned were league tables, star ratings, sackings, or direct political oversight, though apparently members of Congress kept a beady eye on progress.

There are lessons here. The most important is the knowledge that it is possible to make big changes in complex healthcare systems in a relatively short time and in a number of areas at the same time. Interestingly, the authors report that the budget of the VA was "essentially flat" between 1995 and 2000, while the number of patients increased by 40%. Food for thought for us all, and the hope that at least one organisation has seen gain. What was not reported was the amount of pain inside the organisation to achieve it.

Reference:

  1. AK Jha et al. Effect of the transformation of the Veterans Affairs health care system on the quality of care. New England Journal of Medicine 2003 348: 2218-2227.

previous or next story