RESULTS
We identified 10,429 unique patients who met study inclusion criteria. Among these, we excluded 3,596 patients who had missing information leaving us with an overall sample size of 6,893 patients with an inpatient or outpatient encounter (Figure 1). Among these patients, 4,018 were hospitalized, and 720 had a history of cancer. Table 1 reports patient characteristics overall and by mortality status among hospitalized patients (primary analysis). Among the 4,018 hospitalized patients, the average age was 64, 1,844 (46%) were female, 1,060 (26%) were Black, 374 (9%) had a history of cancer, and 1,045 (26%) died.
Table 2 reports multivariable results for the primary and secondary analyses. For the primary analysis, patients with a history of cancer had 33% increased adjusted odds of mortality compared to patients without a history of cancer (OR = 1.33, 95% CI = 1.05, 1.69). This effect appeared to be largely driven by the increased risk associated specifically with lung cancer. Patients with lung cancer had 144% increased adjusted odds of mortality compared to patients without a history of cancer (OR = 2.44, 95% CI = 1.05, 4.39). Patients with kidney cancer, bladder, cancer, hematological cancer, and all genitourinary cancers did not have an increased mortality risk compared to patients without a history of cancer. Further, patients with prostate cancer had 71% increased adjusted odds of mortality compared to patients with other cancers (OR = 1.71, 95% CI = 1.02, 2.92). Finally, patients with kidney cancer (OR = 0.41, 95% CI = 0.18, 0.95) and those with all genitourinary cancers had lower odds of hospitalization compared to other cancers. We did not observe any other statistically significant effects.