Where: Olive View-UCLA Medical Center, a public teaching hospital in Sylmar, Calif.
The issue: Covering an inpatient census that exceeded teaching service limits and reducing payment denials from Medi-Cal (the California Medicaid program).
In 2007, inpatient admissions were on the rise at Olive View-UCLA, while the number of patients that housestaff could treat had declined, due to new limits from the Accreditation Council for Graduate Medical Education.
“They changed the number of admissions they were allowed to take from 12 per resident to 10 per resident in a night, and they changed the total team census cap from 24 to 20, so we needed an additional service to take admitted patients,” said Scott Lundberg, MD, associate program director of the hospital's internal medicine residency program.
The hospital was also dealing with a significant rate of payment denials (29% in 2007) from its largest payer, Medi-Cal. Because Medi-Cal reimburses hospitals per diem, it conducts retrospective chart reviews and denies payment for any days deemed unnecessary.
Because of the per diem payments, the hospital had not yet followed the path of many academic hospitals and instituted a hospitalist service. “It's been demonstrated over and over again that hospitalists are very cost-effective and very good solutions for hospitals that deal mostly with private insurance or DRG [Diagnosis-Related Group]-based reimbursement, where length of stay is a very critical factor and where each procedure or note that a doctor writes is billed as a revenue source. Our reimbursement system is very different…so a lot of the traditional advantages of hospitalists didn't apply,” said Dr. Lundberg.
But in December 2007, hospital leaders decided to institute a hospitalist service to see whether it could be a cost-effective solution to their capacity and payment issues.
How it works
The service started with one hospitalist and one nurse practitioner, and expanded six months later to also include a senior resident. The hospitalist is paged for all daytime admissions and decides whether to take the patient or give him or her to the teaching service.
“The idea was that they would try to select for things like low-risk chest pain, syncope, things that generally are shorter length of stay, while triaging some of the more complex patients to the teaching services,” said Dr. Lundberg.
The hospitalist takes the simpler patients with the goal of treating and releasing them faster. “One of the problems of a traditional teaching service is that just by the way that a schedule is structured, it's very hard for them to turn patients around in less than two days. They admit, they take call all night, they meet with their attending the next day, the student talks to the intern, the intern talks to the resident, then the resident talks to the attending,” said Dr. Lundberg.
“For patients who really just need to stay overnight or maybe two days and be discharged quickly, that process doesn't allow for that kind of quick decision-making and turnaround,” he added.
The implementation of the hospitalist service was associated with an overall decrease in payment denials (from 29% to 27.4%), and increased admissions to the hospital, according to an analysis by Dr. Lundberg and colleagues, published in the August Academic Medicine.
They also looked specifically at some of the patients targeted by the hospitalist service–those with low-risk chest pain–and found shorter length of stay (from 2.48 to 1.92 days) and lower payment denial rates (from 43.8% to 31.8%) after the hospitalist service began. The total salary cost of the service in the first year was $310,000, while an additional $1,430,000 in hospital revenue was attributable to its existence, the study reported.
Dr. Lundberg believes the success stemmed from the hospitalist service's ability to move quickly. “Besides that, hospitalists in general, being focused on inpatient care, just naturally tend to be more attentive to issues of utilization review and length of stay,” he said.
How others benefit
Due to the limited number of patients involved, and the deliberate assignment of more complex patients to the teaching service, the study wasn't able to assess the hospitalist service's impact on quality of care. But patient satisfaction with the hospitalists was higher than that of patients on the hospital's other medical services.
The service has also gotten a positive response from the residents and hospitalists who staff it. “Especially for folks who are relatively recently out of training, it's nice, because it offers a fairly predictable schedule. It's mostly shifts. It offers a chance to do some teaching without having a lot of administration responsibility,” said Dr. Lundberg.