Specialist Referrals During a Hospitalization

While it is the case that the biggest issue in controlling health spending in the United States is unit prices, there are still opportunities to address potentially inappropriate utilization.  A study reported in Journal of the American Medical Association Network Open examines use of specialty consultations during hospitalizations.   (JAMA Article)   The Medicare population was used for the study.  On average, a Medicare patient receives 1 to 3 consultations per admission.  In addition to the cost of the actual consultation, they can lead to further testing, some of which can have its own dangers, and to additional specialty visits following discharge.  The authors sought to identify the practice style of different hospitalist physicians and see if those who used more specialty consultations had any better outcomes, specifically all-cause 30-day mortality and readmission rates within 7 and 30 days of discharge.  They also looked at rates of use of outpatient specialty care following discharge.  Admissions in the fee-for-service Medicare population which involved a hospitalist during 2013 and 2014 were analyzed.  Hospitalists were ranked on their likelihood and degree of specialty care referrals during an admission, factoring out a potential facility-specific component to such use, and compared to other hospitalists at the same facility.  Typical adjustments based on patient characteristics and other factors were performed.  Over 700,000 admissions at 737 hospitals attended to by around 14,600 doctors were included in the final analysis.

Across all hospitalists there were 1.2 specialty consultations per admission.  The top 25% of hospitalists had 1.4 consultations per admission on average.  High-consultation using hospitalists tended to have more non-white patients, more dual-eligible patients and more patients with high-severity health status.  After adjustment, patients cared for by high-consulting doctors had 4% longer lengths of stay and 11% higher Part B spending during the stay.  These patients were also 4% less likely to be discharged to home and 7% more likely to see a specialist within 90 days of discharge.  There was no difference in 7 day or 30 day readmission rates, nor was there any difference in mortality rates.  The results suggest that while more specialty consultations during an admission raises utilization and spending, they do not result in improvement on some common measures of quality outcomes.  It is certainly possible that some hospitalists aren’t using consultations when they should, but the study indicates that it is more likely there is overuse than underuse.  As with most things, the best approach is probably to present the data to the physicians and help them understand why there is variance and identify specific steps to ensure that referrals to specialists are truly needed.