Medicine of the Future in America

Hospitalizations for Tuberculosis: Outcomes

Comorbid illness was common in patients with TB, with 29% having a DCI score of 1, 9% having a score of 2, and 4% having a score of > 3. Seventy-five patients (3.3%) were identified as having HIV infection.
Over half of the patients with TB (57%) were admitted to the hospital through the emergency department, and Table 2 shows the characteristics of the hospitals to which TB patients were admitted. Compared to non-TB patients, those with TB were more likely to be admitted to urban hospitals (92% vs 84%, respectively; p < 0.001) and those designated as teaching hospitals (61% vs 43%, respectively; p < 0.001). 
The length of hospital stay was considerably longer for hospital admissions for TB compared to other hospital admissions (mean, 14.2 ± 18.8 vs 4.6 ± 6.7 days, respectively; p < 0.001; median, 9 vs 3 days, respectively). Additionally, the total charges were almost 2.5 times higher for hospital admissions for TB compared to all other hospital admissions (mean total charges, $33,985 ± 52,414 vs $13,770 ± 29,336, respectively; p < 0.001; median total charges, $18,975 vs $7,123, respectively). While two thirds of the patients (69.5%) were routinely discharged to home, 4.9% of those with a primary diagnosis of TB died during the hospitalization, compared to 2.4% of non-TB patients (relative mortality risk, 2.08; 95% CI, 1.74 to 2.49) [Table 3].

Compared to patients with TB who were alive at hospital discharge, patients who died during hospitalization were older (mean age, 65.1 ± 16.5 years vs 49.4 ± 20.8 years, respectively; p < 0.001), had a higher level of comorbid illness (mean DCI, 1.1 ± 1.0 vs 0.55 ± 0.81, respectively; p < 0.001), required longer hospitalizations (mean length of stay, 19.9 ± 20.5 days vs 13.9 ± 18.7 days, respectively; p < 0.001), and accumulated substantially higher charges (mean charges, $79,585 ± 107,873 vs $31,610 ± 46,578, respectively; p < 0.001) [Fig 1], Logistic regression analyses estimating the unadjusted and adjusted ORs of risk of in-hospital mortality are presented in Table 4. The Hosmer-Lemeshow test showed that the multivariate models demonstrated a good fit, and there was no evidence for collinearity among the independent variables. Increasing age, comorbidity, and having Medicare insurance were statistically significant predictors of mortality in bivariate analyses. In multivariate analyses, age (OR, 1.03 per year of age increase; p = 0.001) and comorbidity (OR, 1.59; p < 0.001) remained strong predictors of mortality. Medicare insurance status was no longer a statistically significant predictor of mortality, after adjustment. Similarly, HIV status, gender, race, and income did not significantly predict mortality.
Fig1
Figure 1. Comparison of patients with TB who died or were alive at hospital discharge. Comparison of age, DCI, length of stay (LOS), and total charges for patients with TB who died in the hospital (112 patients; black columns) and for those who were alive at hospital discharge (2,163 patients; white columns). p < 0.001 for all comparisons.
Table 2—Admission Source and Hospital Characteristics, NIS, United States, 2000

Variables TB Admissions (n = 2,279) Non-TB Admissions (n = 7,448,713) p Value t
Admission source < 0.001
Emergency department 56.7 40.8
Routine 37.5 54.4
Another hospital 3.4 3.2
Another facility/including long-term care 1.8 1.6
Court/law enforced 0.6 0.1
Hospital characteristics
Location < 0.001
Rural 7.7 15.6
Urban 92.3 84.4
Teaching status < 0.001
Nonteaching 38.6 56.6
Teaching 61.4 43.4
Region of United States < 0.001
Northeast 19.7 20.2
Midwest 17.6 23.1
South 37.2 37.7
West 25.5 19.0

Table 3—Outcomes of Hospitalized Patients With and Without TB

Variables TB patients (n = 2,279) Non-TB patients (n = 7,448,713)
Length of stay,t d 14.2 ± 18.8 4.6 ± 6.7
Total charges{ $33,985 ± $52,414 $13,770 ± $29,336
Disposition{
Routine 69.8 77.3
Short-term hospital 4.7 2.5
Another facility/intermediate care/SNF 11.9 10.8
Home health care 7.0 6.2
Against medical advice 1.8 0.8
Died 4.9 2.4

Table 4 —Predictors of In-Hospital Mortality Among Patients Hospitalized for TB

Predictors Unadjusted OR (95% CI) p Value Adjusted ORt(95% CI) p Value Adjusted OR{(95% CI) p Value
egA 1.04(1.03-1.05) < 0.001 1.03(1.01-1.05) 0.001 1.03(1.01-1.05) 0.001
Sex| 1.21 (0.82-1.78) 0.340 0.96(0.61-1.52) 0.873 1.02 (0.64-1.62) 0.929
Race^
Black 0.77 (0.46-1.30) 0.335 1.44 (0.78-2.65) 0.244 1.32 (0.71-2.47) 0.383
Hispanic 0.49 (0.27-0.90) 0.022 0.95 (0.47-1.91) 0.876 0.89 (0.43-1.82) 0.750
Asian/Pacific Islander 0.67 (0.34-1.32) 0.251 1.15 (0.56-2.37) 0.709 1.03 (0.49-2.15) 0.943
Other 0.35 (0.11-1.17) 0.089 0.72 (0.21-2.50) 0.607 0.72 (0.21-2.51) 0.605
DCI# 1.97(1.63-2.39) < 0.001 1.59(1.26-1.99) < 0.001 1.57(1.25-1.99) < 0.001
Diagnosis of HIV** 1.09 (0.39-3.04) 0.867 0.83 (0.19-3.71) 0.811 0.89 (0.20-4.02) 0.884
Insurancett
Medicare 4.67 (2.48-8.78) < 0.001 1.74 (0.79-3.84) 0.169 1.79 (0.81-3.93) 0.149
Medicaid 1.97 (0.97-3.98) 0.059 1.75 (0.79-3.87) 0.167 1.46 (0.65-3.28) 0.354
Self-pay 1.24 (0.54-2.84) 0.614 1.40 (0.49-3.99) 0.532 1.16(0.40-3.33) 0.785
Other 0.62 (0.20-1.95) 0.416 0.37 (0.08-1.74) 0.209 0.33 (0.07-1.53) 0.156
Median income{{
$25,000-$34,999 1.27 (0.73-2.20) 0.392 1.02 (0.52-1.97) 0.958 0.93 (0.48-1.83) 0.844
$35,000-$44,999 1.04 (0.57-1.89) 0.899 0.87 (0.43-1.77) 0.697 0.82 (0.40-1.69) 0.591
$45,000+ 0.99 (0.55-1.76) 0.964 0.77 (0.37-1.58) 0.472 0.76 (0.37-1.57) 0.461
Admission source§§
Emergency department 1.75 (1.12-2.73) 0.013 2.38 (1.40-4.05) 0.001
Another hospital 2.45 (0.98-6.11) 0.055 2.50 (0.85-7.34) 0.095
Other facility/long-term care/SNF 2.45 (0.71-8.43) 0.157 1.25 (0.27-5.84) 0.778
This entry was posted in Tuberculosis and tagged Epidemiology, health-care costs, hospitalization, mycobacterium, outcomes, tuberculosis.
Copyright © 2012 Medicine of the Future in America www.perfexis.org