Published 27 August 2009, doi:10.1136/bmj.b3081
Cite this as: BMJ 2009;339:b3081

Editorials

The effects of payment by results

Cannot necessarily be predicted by theory and past experience

In the linked study (doi:10.1136/bmj.b3047), Farrar and colleagues assess the effect of payment by results—a fixed tariff payment system based on case mix that reimburses acute care hospitals for the type and number of patients treated—on volume, cost, and quality of care in acute care hospitals in England for 2003-4 and 2005-6.¹ By linking provider income to activity, the tariff is expected to provide incentives for higher output and lower costs per patient. The study found that in hospitals where payment by results was implemented the length of stay decreased and the proportion of day cases increased, but indicators of quality of care changed little.

Perhaps the most exciting part of forming health policies is that you do not know what will happen once a change in policy occurs. The Medicare prospective payment system in the United States in 1982-3 shares many similarities with the payment by results programme assessed by Farrar and colleagues. When the system was developed in 1982 it was guided by theory and experience in several states. Unfortunately these are not always a reliable guide, and sometimes the unexpected can happen. For example, it was thought that the number of readmissions would increase because hospitals would be paid for two admissions instead of one for any patient who was discharged and then readmitted. In 1984, the New England Journal of Medicine published an article showing the rate of readmissions before implementation of the system and forecasting an increase in the readmission rate.² However, 25 years later the same journal reported no such increase.³

In the Medicare system the response of hospital managers was unexpected. The payment system was based on statistical averages—some patients would stay longer than average and others shorter. Hospital managers interpreted the legislation to mean that the average was the upper limit and told doctors and patients that patients’ stays were limited to a certain number of days. The response was unpredicted, although in retrospect understandable.

Evaluations like those by Farrar and colleagues are needed because patients, doctors, and hospitals do not always respond in ways that programme designers anticipate. This is true even when there is already considerable international experience, as the authors describe. Many countries have instituted a programme similar to payment by results over the past 25 years; however, experiences in one country do not always predict those in another.

From the US perspective, the reduction in length of stay in England after the introduction of payment by results was relatively small. In England the average length of stay fell by 0.08 of a day more than in Scotland, but in the US it fell by more than one day. The greater response in the US could be the result of differences between the programmes, the fact that the length of stay in England fell during the intervening 25 years, or different behavioural responses by English and American hospital managers. This is what makes policy making and international comparisons so interesting. You need to expect the unexpected.

Farrar and colleagues found no improvements in quality of care. This is a fairly common result. From the beginning the Medicare prospective payment system did not include an adjustment for quality of care—all hospitals in a geographical area received the same payment, regardless of quality of care. One reason is that it was difficult to measure quality of hospital care in 1982, and this remains true today. As a result, the findings of no real difference in quality of care in England, the US, and elsewhere are not surprising.

Recently the Medicare programme proposed policy changes to encourage improvements in quality of care, and it has been evaluating a pay for performance programme that rewards hospitals that provide higher quality of care. So far, few of these programmes have shown much success.^{4 5 6} More recently, the Medicare programme has targeted specific activities, such as nosocomial infections and readmissions for certain conditions for special attention. These activities are just beginning, and it is unclear how much effect, if any, they will have.

In the US an unexpected effect of the Medicare prospective payment system became known as "discharged sicker and quicker." The shorter length of stay meant that patients were being discharged with higher levels of acute illness and were more likely to need nursing home care or home health care after discharge. This resulted in unexpectedly rapid increases in rates of such care. A follow-up study could examine whether this also occurred in England.

Cite this as: BMJ 2009;339:b3081

¹ Johns Hopkins University, Hampton House, 624 N Broadway, Baltimore, MD 20815, USA

Research, doi:10.1136/bmj.b3047

---

Competing interests: None declared.

Provenance and peer review: Commissioned; not externally peer reviewed.

References

1. Farrar S, Yi D, Sutton M, Chalkley M, Sussex J, Scott A. Has payment by results affected the way that English hospitals provide care? Difference-in-differences analysis. BMJ 2009;339:b3047.[Abstract/Free Full Text]
2. Anderson GF, Steinberg EP. Hospital readmissions in the Medicare population. N Engl J Med 1984;311:1349-53.[Abstract]
3. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 2009;360:1418-28.[Abstract/Free Full Text]
4. Wolff JI, Starfield B, Anderson GF. Prevalence, expenditures and complications of multiple chronic conditions in the elderly. Arch Intern Med 2002;162:2269-76.[Abstract/Free Full Text]
5. Lee TH. Pay for performance, version 2.0. N Engl J Med 2007;357:531-3.[Free Full Text]
6. Epstein AM. Paying for performance in the United States and abroad. N Engl J Med 2007;355:406-8.[CrossRef][Web of Science]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    StumbleUpon    Technorati    What's this?

Relevant Articles

Expect the unexpected

Fiona Godlee
BMJ 2009 339: b3571. [Extract] [Full Text]

Has payment by results affected the way that English hospitals provide care? Difference-in-differences analysis

Shelley Farrar, Deokhee Yi, Matt Sutton, Martin Chalkley, Jon Sussex, and Anthony Scott
BMJ 2009 339: b3047. [Abstract] [Full Text] [PDF]

Online poll

Find out more on doc2doc >>