Today’s Managing Health Care Cost Indicator is 19.6%
Source |
There’s been a lot of talk about preventing hospital readmissions -- and the Affordable Care Act has some early incentives for hospitals that can lower readmission rates, and some painful later penalties for those which continue to have higher readmission rates.
One in five Medicare beneficiaries discharged from a hospital is readmitted within 30 days. (It’s 24% among the disabled under 65, and 19% among those over 65). MedPAC suggested that the cost of preventable Medicare hospitalizations could be $12 billion per year. There are innumerable demonstration projects to lower hospital readmission rates. They often include better discharge planning and patient education, medication reconciliation, post-discharge phone calls and appointments, and home visits. Some of the more innovative include home visits by pharmacists, since improper use of medication is a frequent cause of readmissions.
There’s a disappointing review of the published literature in the late October Annals of Internal Medicine. The authors examined over 4000 articles, and did in depth review of about a tenth of them. 43 articles met inclusion criteria because they and compared outcomes between an intervention group and a control group. Even so, most of the studies met less than half of the Cochrane Collaborative criteria regarding unbiased clinical trial reports.
Of the 43 trials reported, only 7 were randomized, and most were quite small.
The authors’ conclusion:
We did not identify a discrete intervention or bundle of interventions that appears to reliably reduce re-hospitalization.
Hospital readmissions are the largest opportunity in Medicare and in disabled populations, and those with chronic disease. In employer-insured populations under age 65, many of the readmissions are either related to mental health, or are planned (scheduled chemotherapy) or desirable (organs becoming available for those requiring transplants).
The special sauce to substantially reduce hospitalizations still hasn’t been identified.