Comparison of community and referral intensive care unit patients in a tertiary medical center: evidence for referral bias in the critically ill.

OBJECTIVE

To determine the existence of referral bias in the critically ill by comparing the clinical and epidemiologic characteristics of community (Olmsted County, MN residents) and referral (non-Olmsted County residents) patients admitted to the intensive care unit.

DESIGN

Retrospective, cohort study.

SETTING

Academic tertiary care medical center.

PATIENTS

Patients admitted to the medical and surgical intensive care units at Mayo Medical Center from 1995 to 2004.

INTERVENTION

None.

MEASUREMENTS AND MAIN RESULTS

Residency status, demographics, Acute Physiology and Chronic Health Evaluation III score, intensive care unit admission diagnosis and treatment status, intensive care unit and hospital mortality, length of stay, and travel distances to Mayo Clinic. Referral patients with a medical intensive care unit admission were more severely ill, had greater mortality rates and length of stay and were more likely to receive an active intensive care unit intervention compared with community patients (p < 0.0001). Referral and community patients who had a surgical intensive care unit admission had similar severity of illness, length of stay, and intensive care unit mortality rate. Hospital mortality rate was lower in the referral surgical patients compared with community surgical patients (p = 0.0001). When adjusted for severity of illness, intensity of treatment, and admission source, community and referral medical intensive care unit patients had a similar risk of hospital death, whereas referral surgical patients had a lower risk of hospital death compared with community patients. Referral patients who had a medical intensive care unit admission and traveled greater distances to Mayo Clinic had greater mortality rates and length of stay; those who had a surgical intensive care unit admission and traveled greater distances had lower mortalities and length of stay.

CONCLUSIONS

Patients who resided outside of our local community and who had medical admissions to the intensive care unit were more severely ill, had greater mortality rates, and had longer length of stay compared with community patients. Our findings support the existence of referral bias in critically ill medical patients at our tertiary medical center.