When Kira Ryskina works in Penn Presbyterian Medical Center, she usually sees her patients every day.
When it comes time for them to leave, some are too sick to go directly home or need rehabilitation in a nursing home. She tells them that another doctor will see them there.
But Ryskina, an internist and health policy researcher at the University of Pennsylvania’s Perelman School of Medicine, began to wonder how true that was. Her own experience working in nursing homes, plus feedback from patients and their families, told her that patients might not see doctors as quickly as many would expect.
Her curiosity led to a study published Monday in the journal Health Affairs that supported her theory that nursing home patients can’t be sure of seeing a doctor. Her team’s analysis of records for 2.4 million Medicare patients discharged from hospitals to nursing homes from January 2012 to October 2014 found that 10.4 percent never saw a doctor, nurse practitioner or physician assistant. Nearly 72 percent of patients were examined by a doctor or advanced practitioner within four days, but there was considerable variability among nursing homes. Smaller and rural nursing homes were the slowest. Ryskina saw little evidence that nursing homes were doing a good job of triaging patients to assure that the sickest ones were seen the most quickly.
Of the group that never saw a doctor, 28 percent were readmitted to a hospital and 14 percent died within 30 days of admission to the nursing home. Among patients who saw a doctor at least once, 14.3 percent were readmitted to a hospital and 7.2 percent died within that time period.
From a policy standpoint, these numbers are important because some modern payment systems, including Medicare's, hold hospitals accountable for what happens after their patients leave. There can be penalties if patients need to return to the hospital too quickly.
From a patient perspective, quick readmissions and excess deaths can be a sign that patients weren't ready to leave the hospital or needed more attentive care.
Ryskina said there is no official guideline for how quickly newly admitted patients should see a doctor, but she’d start by suggesting 48 hours. “I think we need some empirical data to support this,” she said. Medicare’s current rules specify that patients be seen within 30 days of admission, likely a holdover from days when many nursing home residents received “custodial” care, she said. The patients in her study were not long-term nursing home residents.
Physicians are rarely on site at nursing homes every day. Ensuring quicker assessments would be challenging for the facilities, Ryskina said. “I think reimbursement is a big barrier,” she said.
About 20 percent of hospital patients are discharged to nursing homes for further medical care. The rest go home or to hospice. Ryskina said it is often the most vulnerable patients who go to nursing homes, where they can receive physical therapy for joint replacements, intravenous antibiotics, or skilled wound care.
While she assumes that 100 percent of nursing home patients should be seen by a doctor, Ryskina said it is possible that some patients arrived on a weekend and were readmitted to the hospital so quickly that a doctor didn’t have a chance to examine them. However, the patients in the group that never saw a doctor had a median stay of 11 days. “I think it’s concerning,” she said.
Patients in rural nursing homes waited the longest on average to see a doctor: 8.1 days compared with 3.2 in other facilities.
Geographically, doctors saw patients the fastest — 2.2 days — in nursing homes in the Northeast. The average wait was 5.3 days in the Midwest. Patients in large facilities saw doctors faster than those in small ones.
Ryskina said growing numbers of doctors are specializing in nursing home care or in the needs of complex patients at risk for further hospitalization. Telemedicine could also improve access to physicians.