When Patients Are Hospitalized in Bunches, Mortality Rises

LOS ANGELES, Jan. 10 -- Older patients admitted to a large academic medical center on busy shifts may have a greater mortality risk than those who come to the hospital on less frenetic days, suggested investigators here.

This was the bottom line of a study of admissions to the general medicine service of the University of California San Francisco's Moffitt-Long Hospital, a 525-bed tertiary care center, over three years.

"Our findings suggest that higher house staff workload on admitting days -- when fewer backup resources are available -- increases resource use and may increase inpatient mortality," reported Michael Ong, M.D., Ph.D., of UCLA, and UCSF colleagues, in the Jan. 8 Archives of Internal Medicine.

The finding emerged from a retrospective cohort analysis of 5,742 adults, a majority of Medicare age, admitted from July 1, 1998, to June 30, 2001.

The study included only patients admitted to the general medicine service and intensive care unit without cardiovascular, neurologic, or cancer-related primary diagnoses. Those hospitalized for short-stay procedures or elective chemotherapy were also excluded.

The house staff teams were composed of one attending, one resident, and one or two interns. Patients were admitted when one team was scheduled to admit patients in 41.7% of the cases and when two teams were admitting in 36.5% of the cases.

The investigators found that each additional admission by a team of house staffers on a patient's admission day increased that patient's length of stay by 3.09% (95% CI 2.22% to 3.96%) and total costs by 2.31% (95% CI 1.29% to 3.33%).

The teams averaged 5.8 admissions per call cycle with a daily census of 10.1 patients. The total daily census for the general medical service averaged 44.8 patients with a mean of 7.8 admissions and 7.8 discharges each day.

The mortality risk increased substantially when days with three or fewer admissions were compared with days of nine or more admissions. Whereas four to six admissions had about a 50% increased mortality risk, 10 to 12 admissions was associated with more than 100% increased mortality risk and 13 to 15 admissions was associated with more than a 250% increased risk.

During the study, 5.5% of the inpatients died and 7.6% were readmitted to the same hospital within 30 days of discharge. Their median length of stay was four days with a median total cost of ,319.

With each additional admission on a patient's admission day, the researchers reported in the multivariate analysis:

• Inpatient mortality increased 1.09-fold (95% CI 1.02 to 1.15),
• Length of stay was 3.09% longer (2.22% to 3.96%),
• Total costs were 2.31% higher (1.29% to 3.33%), and
• 30-day admission was not significantly affected (adjusted odds ratio 0.97, 95% CI 0.92 to 1.02).

Although 13 to 15 admissions was associated with more than a 250% increased risk, at the busiest -- 16 or more admissions -- the increased mortality risk was no longer significant (P>0.05). This may reflect the general medicine service's method of redistributing patients to less busy teams after the busiest admitting days, the authors suggested.

In the multivariate analysis, additional findings included:

• Each extra admission on a patient's discharge day had a significant effect on total costs (difference 1.51%, 95% CI 0.70% to 2.32%) but not on length of stay (difference 0.09%, 95% CI -0.59% to 0.77%),
• Each additional discharge on a patient's admission day significantly decreased the length of stay by 4.69% (95% CI -3.38% to -5.97%) and total costs by 3.63% (95% CI -2.08% to -5.16%), and
• The number of discharges on a patient's discharge day had no significant impact on length of stay or total costs.

The significant effect from busier admission days but not discharge days for patients may have due to the importance of early, accurate diagnosis and workup, the researchers said.

"Admission workup activity is extensive," they wrote, "more admissions reduce the time spent by teams on any one admitted patient, potentially leading to inaccurate initial clinical assessment or pushing workup activity onto subsequent days, leading to longer lengths of stay and higher total costs."

However, each additional patient on the team's average census during the patient's hospitalization was associated with earlier discharge (difference -5.30%, 95% CI -4.54% to -6.07%) and lower total costs (difference -5.11%, 95% CI -4.20% to -6.00%).

Dr. Ong and colleagues said they believe this counterintuitive finding represents adaptation by teams or the entire inpatient service.

"Unlike admission volume, which cannot be modified by teams," they wrote, "team average census is a work measure that teams can control by adapting their daily tasks, such as skipping teaching conferences to perform patient care tasks, to meet workload demands."

They also found that each additional admission assigned to a house staff team increased inpatient mortality risk (odds ratio 1.09, 95% confidence interval 1.02 to 1.15).

While the study was not large enough to be powered to show a mortality difference, the results have important implications for residency training programs and hospital administrators, the investigators said.

"Balancing the clinical and economic outcomes with available resources and the educational impact of changes in the organization of house staff teams will be important tasks for training programs and teaching hospitals in coming years," they wrote.

"High workload may also increase pressure to discharge patients quickly or focus attention of social workers and discharge planners on teams with higher patient volumes," they added.

On the other hand, increases in the total number of team admissions in a patient's month of admission -- reflecting long-term workload -- did increase length of stay and costs slightly though significantly (0.53% and 0.40%, respectively).

"This finding suggests that internal efficiency can be increased in the short-term, but fatigue may accumulate within teams over time," the authors wrote.

Results were similar in analyses of patients who had been redistributed to other teams or those with intensive care unit stays.

Findings were adjusted for patient sociodemographic factors, diagnosis-related severity, intensive care unit stays, and diagnoses of HIV, cancer, and pneumonia, as well as changes over the academic year and discontinuities in care due to team personnel switches.

Because the study included only a single academic medical center, the findings may not be applicable to smaller institutions with limited resources or to those that include surgical or pediatric services or have different house staff training systems.