• What do hospital readmissions measure?

    From Associations Between Reduced Hospital Length of Stay and 30-Day Readmission Rate and Mortality: 14-Year Experience in 129 Veterans Affairs Hospitals, by Peter Kaboli, et al. (Annals of Internal Medicine, 2012) and based on 1997 to 2010 data:

    Results: For all medical diagnoses combined, risk-adjusted mean hospital LOS [length of stay] decreased by 1.46 days from 5.44 to 3.98 days, or 2% annually (P < 0.001). Reductions in LOS were also observed for the 5 specific common diagnoses, with greatest reductions for acute myocardial infarction (2.85 days) and community-acquired pneumonia (2.22 days). Over the 14 years, risk-adjusted 30-day readmission rates for all medical diagnoses combined decreased from 16.5% to 13.8% (P < 0.001). Reductions in readmissions were also observed for the 5 specific common diagnoses, with greatest reductions for acute myocardial infarction (22.6% to 19.8%) and chronic obstructive pulmonary disease (17.9% to 14.6%). All-cause mortality 90 days after admission was reduced by 3% annually. Of note, hospitals with mean risk-adjusted LOS that was lower than expected had a higher readmission rate, suggesting a modest tradeoff between hospital LOS and readmission (6% increase for each day lower than expected).

    Limitations: This study is limited to the Veterans Health Administration system; non–Veterans Affairs admissions were not available. No measure of readmission preventability was used.

    One wonders to what extent index hospitalizations and readmissions occurred outside the VA (e.g., in a non-VA hospital and funded by Medicare, Medicaid, or private health insurance). Would including such data, to the extent available, matter?

    The paper also includes this passage that conveys the probability of readmission and suggests that a modest fraction of them are preventable:

    In studies of Medicare patients, 30- day readmission rates range from 8% (8) to 21% (10), depending on diagnosis, with annual estimated costs of $17.4 billion (10). A recent Centers for Medicare & Medicaid Services Hospital Quality Alliance study reported readmission rates of 21% and 25% for pneumonia and heart failure (HF), respectively (11). The debate on readmissions centers on understanding which readmissions may be prevented by modifying care during the index admission or transitions of care and which are unavoidable events for patients with complex medical conditions. In 1 study using retrospective chart review, 8.8% of readmissions were considered preventable (12). A recent randomized, controlled trial using a packet of nursing and pharmacist discharge services to reduce readmissions found a reduction that was not statistically significant but estimated that 15% of readmissions could be prevented using their intervention (13). Regardless, hospital readmission is not an optimal outcome for patients or providers but may be financially advantageous for hospitals.

    To the extent readmissions reflect poor and modifiable quality, this is not very encouraging. An accompanying editorial by Eugene Oddone and Morris Weinberger throws more cold water.

    [L]ess than 20% of the sample of older, chronically ill veterans fell into one of [the examined disease] groups, meaning that taking a disease specific approach in an attempt to further reduce hospital readmission rates is not likely to lower the overall rate to a large extent. In this study, no single diagnostic category accounted for more than 5% of hospitalizations. […]

    [H]ospital readmission rates may be a poor measure of quality of care because of the complexity of factors that cause them and the poor correlation among those factors. It’s not only the quality of care during the index hospitalization or the quality of the handoff to post discharge care that influences readmission rates. Rather, many important factors affect when and how often patients are hospitalized, including access to post discharge care, ability to purchase evidence-based medications or services prescribed at discharge, disease and disease severity, socioeconomic status, community resources, and social support (3). So, if we focus intensely on a few of these factors, will readmission rates decrease? Not always. Sixteen years ago, we showed that improving the handoff from hospital to primary care and enhancing veterans’ access to primary care systematically increased 30-day hospital readmission rates by 26%, but patients were much more satisfied with their care (4).

    [P]unishing hospitals alone will not necessarily lower the rates, and it may lead to potentially perverse practices (for example, funneling sick patients to short-stay units when higher levels of care are needed).

    I’m not taking sides. Not yet, anyway.


    • The real solution is to stop paying hospitals per admission. No one cares how often I go to the public library, because the library is funded on the first day of each fiscal year.

      In any hospital, about 80% of expenses are fixed for the year in terms of salaries, equipment, utilities, liability insurance, and any bonds or mortgages for the hospital building.

      For this reason, it is not productive to decrease the number of admissions. If we succeed, then hospitals wiill just charge more for the admissions that remain.

      This has already happened. The Statistical Abstract of the US has numerous tables which show the decline in hospital beds and the relative decline in admissions since 1970. Yet we spend more than ever on hospitals, thanks in large part to upcoding the admissions that still occur.

      I grant that it will be a huge task to get hospitals onto global budgets.
      But there is no shortage of intelligent people both in hospitals and in academia to accomplish this.

      Bob Hertz,
      The Health Care Crusade

      • I’d still expect some potentially preventable readmissions to occur if hospitals are not paid by the admission. It’s plausible that allowing some potentially preventable readmissions is cheaper than reducing the rate to zero. (I’m thinking only of those preventable by modification of hospital factors. There are many that might be preventable by modification of other factors outside the control of the hospital. Those would also still occur independent of how hospitals are paid.)

    • In the last year of my father’s life, he was readmitted six times to hospitals. He fell and broke bones, his bowels stopped, his heart almost stopped, etc.

      All the admissions were to the Mpls VA hospital. Outside of cooking some extra meals, and maybe some additional shifts for part-time nurses, what did this really cost? He nver required expensive drugs that were not already in the pharmacy.

      In a user fee environment where 90% of Americans reside, this would have cost about $120,000. In the VA global budgeting environment, no one really knows or cares about the tiny marginal cost that my father imposed.

      Global budgets require an adult willingness to assess enough in taxes to cover the normal expenses of public hospitals. Instead, Americans have preferred to gamble on paying insurance companies per admission, and then trying to hold down admissions. Then it turns out that taxes go up anyways because low-income workers cannot afford the insurance.

      It is better to send tax revenues straight to public hospitals, and leave insurers out of it.