Patient Mortality Linked to End-of-Rotation Transitions

Norra MacReady

December 06, 2016

End-of-rotation transitions in care may be associated with a significant increase in mortality risk among hospital patients, according to the results of a new analysis.

The analysis included all patients admitted before an end-of-rotation transition who died or were discharged within 7 days after the staff transition.

However, a similar association was not seen in an alternative analysis restricted to patients admitted within 2 days before a transition, suggesting that "this increased risk may be limited to longer-stay, complex patients with greater comorbidities or those discharged soon after the transition," the authors write in the December 6 issue of the Journal of the American Medical Association.

Impetus for the study came from the recognition that, despite a growing body of evidence suggesting that the risk for errors and adverse events increases during shift-to-shift handoffs, data on risks associated with the end-of-rotation handoff are scarce, write Joshua L. Denson, MD, a fellow in the Division of Pulmonary Sciences and Critical Care Medicine at University of Colorado in Aurora, and colleagues.

The researchers conducted a retrospective cohort study of adults discharged from 10 university-affiliated US Veterans Health Administration hospitals between July 1, 2008, and June 30, 2014. They included patients discharged from a general medicine team, intensive care unit, coronary care unit, step-down unit, or subspecialty team, such as cardiology or pulmonology.

In the main analysis, they studied patients admitted any time before an end-of-rotation handoff who died or were discharged within 1 week after the transition (the transition group). The control group consisted of all other discharged patients.

In an alternate analysis, the transition group was defined as patients admitted 1 or 2 days before the staff transition, regardless of length of stay. The control group in this analysis consisted of patients admitted 2 weeks later on days not associated with a service change.

"The primary outcome for all analyses was adjusted in-hospital mortality rate," the authors write. Thirty-day and 90-day mortality and readmission rates were secondary outcomes, measured from date of admission. Patients who stayed in the hospital more than 33 days were excluded from the analysis.

The main study included 230,701 patient discharges. The patient cohort was 95.8% male, with a mean age of 65.5 years and a mean length of stay of 4.2 days (standard deviation [SD], 4.7; median, 3.0 days; interquartile range [IQR], 1.0 - 5.0). Of the patient discharges, 25,938 (11.2%) were exposed to intern-only transitions; 26,456 (11.5%) were exposed to resident-only transitions; and 11,517 (5.0%) were exposed to intern-plus-resident transitions.

In an unadjusted analysis, all-cause in-hospital mortality in the main study was 907 (3.5%) among patients exposed to intern-only transitions, compared with 4112 (2.0%) among the controls (risk difference, 1.5 percentage point [95% confidence interval (CI), 1.3 - 1.7 percentage points]; odds ratio [OR], 1.77 [95% CI, 1.58 - 1.98]; P < .001). Among patients exposed to resident-only transfers, all-cause mortality was 868 (3.3%), compared with 4151 (2.0%) control patients (risk difference, 1.3 percentage points [95% CI, 1.0 - 1.5 percentage points]; OR, 1.64 [95% CI, 1.46 - 1.83]; P < .001). With patients exposed to intern-plus-resident transitions, all-cause mortality was 458 (4.0%), compared with 4561 (2.1%) among the control patients (risk difference, 1.9 percentage points [95% CI, 1.5 - 2.3 percentage points]; OR, 1.95 [95% CI, 1.63 - 2.33]; P < .001).

After adjustment for age, sex, race or ethnicity, length of stay, calendar month and year, hospital site, and Elixhauser comorbidity index, the adjusted ORs for in-hospital mortality among patients exposed to transitions were significantly higher in all three groups. Specifically, adjusted ORs for those exposed to intern-only, resident-only, and intern-plus-resident transitions were 1.12 (95% CI, 1.03 - 1.21; P = .008), 1.07 (95% CI, 0.99 - 1.16; P = .09), and 1.18 (95% CI, 1.06 - 1.33; P = .003), respectively.

The adjusted ORs for 30-day mortality among patients exposed to intern, resident, and intern-plus-resident transitions were 1.17 (95% CI, 1.13 - 1.22; P < .001), 1.11 (95% CI, 1.04 - 1.18; P = .002), and 1.21 (95% CI, 1.12 - 1.31; P < .001), respectively.

The corresponding adjusted ORs for 90-day mortality were 1.14 (95% CI, 1.10 - 1.19; P < .001), 1.10 (95% CI, 1.05 - 1.16; P = .001), and 1.17 (95% CI, 1.11 - 1.23; P < .001), respectively.

Unadjusted 30-day and 90-day readmission rates were significantly higher among all the transition groups, but those differences were no longer significant after adjustment.

The authors note that increased mortality both in-hospital and after discharge consistently involved the transition of interns, with or without residents. This was seen in a comparison of patient mortality from July 1, 2008, to June 30, 2011, and July 1, 2011, when the Accreditation Council for Graduate Medical Education (ACGME) duty regulations went into effect, to June 30, 2014. Before the duty-hour regulations, the adjusted ORs for mortality were 1.06 (95% CI, 0.90 - 1.25; P = .47) for patients exposed to intern-only transitions, 1.01 (95% CI, 0.89 - 1.13; P = .92) for patients exposed to resident-only transitions, and 1.10 (95% CI, 0.83 - 1.45; P = .50) for patients exposed to intern-plus-resident transitions. After the duty-hour regulations began, the ORs were 1.16 (95% CI, 1.08 - 1.25; P < .001), 1.13 (95% CI, 0.99 - 1.27; P = .06), and 1.24 (95% CI, 1.11 - 1.39; P < .001), respectively.

In a difference-in-differences analysis comparing risk within each physician group before and after duty regulations went into effect, the adjusted ORs were 1.11 (95% CI, 1.02 - 1.21; P = .01) for transitions involving interns only and 1.17 (95% CI, 1.02 - 1.34; P = .03) for those involving interns plus residents. For residents only, the OR was 1.06 (95% CI, 0.99 - 1.15; P = .10).

Interns are "the most inexperienced residents, increasing the risk for misinformation transfer and potential errors," the authors write. Also, the ACGME duty-hour rules now mean that transitions occur more frequently because of shorter rotation schedules, which increase the risk for errors even more. Also, shorter hours mean that interns now have less time to complete the same duties, "leaving them with limited time to prepare for service change."

Less Complex Patients at Lower Risk

In the alternate analysis, there were no significant differences between the transition and control groups in any category. For example, unadjusted in-hospital mortality was 359 (2.5%) among patients exposed to intern-only transitions and 369 (2.5%) in the control group (risk difference, –0.1 percentage point [95% CI, –0.4 to 0.3 percentage point]; OR, 0.98 [95% CI, 0.87 - 1.10]; P = .68). Similar patterns were seen in all the other calculations for mortality and hospital readmission rates.

Several things may account for these findings, Vineet M. Arora, MD, and Jeanne M. Farnan, MD, from the Department of Medicine, University of Chicago, Illinois, write in an accompanying editorial.

Patients who remain in the hospital over a service change may be sicker or differ in other important ways from patients who can be discharged sooner. For example, they may have "social or financial issues preventing safe discharge." Complex, longer-stay patients "are more likely to die for any reason, regardless of the service change."

Still, the editorialists write, "given well-documented content omissions during handoffs, it is plausible that poor information exchange during service change contributed to increased in-hospital mortality."

However, other factors may also come into play. The patient may receive less attention from the admitting team or the receiving team, or the receiving team may cling to initial clinical impressions "even if new information emerges that warrants further workup."

Questions also arise as to why transitions were associated with increased mortality at 30 and 90 days, but not more hospital readmissions, Dr Arora and Dr Farnan write. "Although these questions remain unanswered, it is reasonable to suggest that service changes may be risky for patients and to explore ways to improve them." They recommended more verbal communication among teams and greater involvement of patients or their caregivers in the transition process.

The study authors, Dr Arora, and Dr Farnan have disclosed no relevant financial relationships.

JAMA. 2016;316:2204-2213, 2193-2194. Abstract, Editorial

For more news, join us on Facebook and Twitter

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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:

processing....