Mortality Lower When Inpatients See Their Own PCP in Hospital

Norra MacReady

November 13, 2017

A primary care physician (PCP)'s prior knowledge of a patient has a significant effect on outcomes when that patient is hospitalized, data from a new study suggest.

In a retrospective analysis of more than 560,000 Medicare patients, hospital care by a PCP who had previously seen the individual as an outpatient was associated with a greater chance of being discharged home and a lower 30-day mortality compared with care delivered by a hospitalist or other generalist who had never met the patient before, Jennifer P. Stevens, MD, MS, and colleagues report today in JAMA Internal Medicine.

These findings may be especially important for patients with multiple illnesses and those who require a complex level of care, Seth Landefeld, MD, coauthor of an accompanying editorial, said in an interview with the journal.

The effect on patient survival was particularly surprising, Dr Stevens, from the Center for Healthcare Delivery Science, Beth Israel Deaconess Medical Center, Boston, Massachusetts, said in the interview. "This challenges us to take another look at the hospitalist model and re-evaluate the full-scale adoption of hospitalist coverage in 75% of hospitals."

However, she cautioned that prospective studies were necessary to confirm the data.

Random Sample of Patients

Dr Stevens and colleagues studied a random 20% sample of Medicare beneficiaries 66 years of age and older who were admitted to acute care hospitals from January 1 through December 31, 2013. To be eligible for study inclusion, patients had to have been admitted for 1 or more of the 20 most common diagnosis-related groups (DRGs) and to have had at least one encounter with an outpatient clinician in the 365 days before admission.

The primary exposure of interest was the type of physician who cared for the patient in the hospital: the patient's own PCP, a hospitalist, or another covering generalist physician (defined as a physician specializing in general practice, internal medicine, family medicine, or geriatrics) who had not previously met the patient. Hospitalists were used as the reference group.

The final cohort consisted of 560,651 patients (59.1% women), with a mean age of 80 years (standard deviation, 8), admitted to 4535 hospitals throughout the United States. In 59.7% of cases, the admitting physician was a hospitalist. PCPs and other outpatient generalists were the admitting physicians in 14.2% and 26.1% of cases, respectively. Overall, 10.6% of patients died within 30 days of admission.

For the entire study population, the median length of stay was 5 days (interquartile range [IQR], 3 - 7 days). Patients treated by PCPs had a slightly longer median length of stay of (5 days; IQR, 4 - 7 days) than those treated by hospitalists or other generalists (5 days; IQR, 3  -7 days for both types of physicians). However, the difference was "to the tune of hours," Dr Stevens said in the interview.

Patients cared for by PCPs were more likely to be discharged home (68.5%) compared with those cared for by hospitalists (64.0%) and other generalists (62.1%) (test of difference, P < .001). Compared with hospitalist care, PCP care was associated with an adjusted odds ratio (AOR) of being discharged home of 1.14 (95% confidence interval [CI], 1.11 - 1.17; P < .001), while for other generalist care the AOR was 0.94 (95% CI, 0.92 - 0.96; P < .001).

Seven-day readmission rates among patients cared for by PCPs, hospitalists, and other generalists were 11.1%, 11.6%, and 12.0%, respectively. In adjusted analysis, PCP and hospitalist care were associated with lower 7-day and 30-day readmission rates than care provided by other generalists (P < .001 for both comparisons).

Unadjusted 30-day mortality was lowest among patients cared for by PCPs, followed by the rates in those cared for by hospitalists and other generalists (8.6%, 10.8%, and 11.0%, respectively; test of difference, P < .001).

"These findings persisted in adjusted analyses for PCPs (AOR, 0.94; 95% CI, 0.91 - 0.97; P < .001) and other generalists (AOR, 1.09; 95% CI, 1.07 - 1.12; P < .001), with hospitalists as the reference group," the authors write.

This pattern also remained significant in a sensitivity analysis that accounted for the complexity of the admission and patient disease burden. "For example, patients cared for by PCPs had the lowest 30-day mortality across all 3 physician groups at all quantiles of DRG complexity and patient comorbidities," the authors write.

All in all, "we found that [for the best outcomes], the physician either had to be an expert on the patient or on the hospital they were working in," Dr Stevens said in the interview. Care suffered when it was provided by a physician who had neither longitudinal knowledge of the patient nor expertise in hospital-based practice.

The Importance of Continuity

In their editorial, Lisa Willett, MD, and Dr Landefeld, both from the Department of Medicine at the University of Alabama in Birmingham, point out that these findings demonstrate the importance of continuity of care, which itself may be "a proxy for what might improve care, a combination of deep knowledge of a patient and his or her situation with a therapeutic trusting relationship, and easy access to care."

In the interview, Dr Landefeld added that an important challenge to address is: "How do we build a system that supports primary care physicians providing that level of comprehensive care, where they really gain two things: one is the continuity of a relationship with the patient, and the other is a deep knowledge of the context of that patient where they live?" He offered the Veterans Health Administration and the Kaiser Permanente system as two examples of organizations that have risen to the challenge, saying, "we need to think about doing that more broadly in the United States."

The take-home message, Dr Stevens said, "is that a prior relationship with a physician has a meaningful impact for patient outcomes, particularly those outside the walls of the hospital. The idea that many funders are interested in care models that extend both in the inpatient and outpatient setting means that simply efficiency-related outcomes may not be the be-all and end-all."

The authors and editorial writers have disclosed no relevant financial relationships.

JAMA Intern Med. Published online November 13, 2017. Full text, Editorial

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: