Hospitalizations for Tuberculosis in the United States in 2000: Predictors of In-Hospital Mortality

doi:10.1378/chest.126.4.1079

Chest

Volume 126, Issue 4, October 2004, Pages 1079-1086

https://doi.org/10.1378/chest.126.4.1079 Get rights and content

Study objectives:

Despite curative therapy, mortality remains high for hospitalized patients with tuberculosis (TB) in the United States. The purpose of this study was to describe the characteristics of hospitalized patients with TB and to identify patient characteristics associated with in-hospital mortality.

Design, setting, and patients:

Using the 2000 Nationwide Inpatient Sample, representing 20% of US hospital admissions, we identified 2,279 hospital admissions with a primary diagnosis of TB (International Classification of Diseases, ninth revision, codes, 010.xx to 018.xx).

Measurements and results:

Mortality was the main outcome measure. Logistic regression analyses were performed including age, gender, race, insurance status, income, Deyo-adapted Charlson comorbidity index (DCI), HIV status, hospital admission source, and hospital characteristics as explanatory variables. A disproportionate number of patients hospitalized with TB were men (64%), nonwhite (72%), lived in areas with median incomes of < $35,000 (50%), and had publicly funded health insurance (49%) or no health insurance (17%). The mortality rate for patients hospitalized for TB was greater than that for non-TB hospital admissions (4.9% vs 2.4%, respectively; p < 0.001). Patients with TB who died during hospitalization were older (mean age, 65.1 vs 49.4 years, respectively; p < 0.001), had greater comorbid illness (DCI, 1.1 vs 0.55, respectively; p < 0.001), required longer hospitalizations (19.9 vs 13.9 days, respectively; p < 0.001), and accumulated substantially higher charges ($79,585 vs $31,610, respectively; p < 0.001) than did patients with TB who were alive at hospital discharge. In a multivariable analysis, older age, comorbid illnesses, and emergency department admissions were independently associated with mortality. The total charges for TB hospitalizations in the United States in 2000 exceeded $385 million.

Conclusions:

Despite public health efforts, patients who are hospitalized with TB are frequently admitted through emergency care settings, have a high risk of in-hospital mortality, and incur substantial hospital charges. To improve TB health outcomes, more vigorous clinical management and prevention strategies should especially target older patients and those with comorbid medical conditions.

Section snippets

Study Design and Population

Using an administrative database, we conducted a cross-sectional study characterizing hospitalizations for TB in the United States in 2000. Patients selected for this study had a primary discharge diagnosis of TB, using the International Classification of Diseases, ninth revision (ICD-9), hospital discharge codes 010.xx to 018.xx.

Database

We used the 2000 Nationwide Inpatient Sample (NIS), a database of hospital inpatient stays, developed by the Healthcare Cost and Utilization Project. As the largest

Results

In this 20% sample of US hospitalizations, representing 7,450,992 hospital admissions, there were 2,279 hospitalizations with a primary hospital discharge diagnosis of TB. The mean patient age on hospital admission was 50.2 years. Demographic characteristics of these patients, compared to all other hospital admissions (7,448,713 hospital admissions), are shown in Table 1. Unlike hospital admissions for primary reasons other than TB, the majority of hospital admissions for TB were in men (64%)

Discussion

This is the largest study, to date, to provide nationally representative estimates of outcomes of hospitalized patients with TB in the United States. Since the observations in this study are not limited to specific geographic areas or subgroups of patients, the external validity of these results is increased. Using a national database, we identified patients hospitalized with TB in the United States in 2000 along with predictors of in-hospital mortality. There were 2,279 TB hospitalizations in

References (43)

VK Rao et al.
The impact of comorbidity on mortality following in-hospital diagnosis of tuberculosis
Chest
(1998)
DJ Shulkin et al.
The cost of caring for patients with tuberculosis: planning for a disease on the rise
Am J Infect Control
(1995)
ME Charlson et al.
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
J Chronic Dis
(1987)
RA Deyo et al.
Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases
J Clin Epidemiol
(1992)
C Perez-Guzman et al.
Does aging modify pulmonary tuberculosis?
Chest
(1999)
P Pompei et al.
Relating patient characteristics at the time of admission to outcomes of hospitalization
J Clin Epidemiol
(1991)
C Dye et al.
Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country: WHO Global Surveillance and Monitoring Project
JAMA
(1999)
American Thoracic Society
Center for Disease Control and Prevention. Infectious Diseases Society of America. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis
Am J Respir Crit Care Med
(2003)
NN Hansel et al.
Quality of life in tuberculosis: patient and provider perspectives
Qual Life Res
(2004)
NN Hansel et al.
Hospitalized tuberculosis patients and predictors of mortality [abstract]
Am J Respir Crit Care Med
(2002)

L Singleton et al.

Long-term hospitalization for tuberculosis control: experience with a medical-psychosocial inpatient unit

JAMA

(1997)

C Greenaway et al.

Delay in diagnosis among hospitalized patients with active tuberculosis: predictors and outcomes

Am J Respir Crit Care Med

(2002)

RE Brown et al.

Health-care expenditures for tuberculosis in the United States

Arch Intern Med

(1995)

Z Taylor et al.

Causes and costs of hospitalization of tuberculosis patients in the United States

Int J Tuberc Lung Dis

(2000)

LS Rosenblum et al.

Effect of HIV infection and tuberculosis on hospitalizations and cost of care for young adults in the United States, 1985 to 1990

Ann Intern Med

(1994)

DWJ Hosmer et al.

Goodness-of-fit testing for the logistic regression model when the estimated probabilities are small

Biom J

(1998)

Centers for Disease Control and Prevention

Tuberculosis morbidity among US-born and foreign-born populations: United States, 2000

MMWR Morb Mortal Wkly Rep

(2002)

J Grange et al.

Tuberculosis in the disadvantaged groups

Curr Opin Pulm Med

(2001)

MF Cantwell et al.

Tuberculosis and race/ethnicity in the United States: impact of socioeconomic status

Am J Respir Crit Care Med

(1998)

Centers for Disease Control and Prevention

Tuberculosis morbidity: United States 1995

MMWR Morb Mortal Wkly Rep

(1996)

SP Patil et al.

In-hospital mortality following acute exacerbations of chronic obstructive pulmonary disease

Arch Intern Med

(2003)

Cited by (58)

Current in-hospital management for patients with tuberculosis in a high-income country: a retrospective cohort study
2022, Clinical Microbiology and Infection
Citation Excerpt :
Multivariate logistic regression was performed to predict the factors associated with in-hospital mortality. Factors related to mortality in previous studies, including age [5,17–19], underlying illness [17–19], KPS score [20] and body mass index [21] were forced into the final model. Age was categorized into four groups (<65, 65–74, 75–89 and > 89 years) [22].
In Japan, most cases of tuberculosis (TB) occur among individuals aged 65 years or older. However, data on in-hospital adverse events (AEs) associated with TB management, especially in high-income nations with an ageing population, are scarce. The present study aimed to scrutinize the current TB unit practices, incidence of in-hospital AEs and predictors of in-hospital mortality.
This retrospective cohort study was conducted at a tertiary care centre in Tokyo, Japan from 2012 to 2017. Inpatients with the diagnosis of TB and aged >18 years were included. Quality of in-hospital care and factors associated with in-hospital mortality were investigated using multivariate logistic regression analysis.
In total, 448 patients were enrolled. The in-hospital mortality rate was 16.7% (75/448). Miliary/disseminated TB was common (59/448, 13.2%), especially in those who died (17/75, 22.7%). Factors independently associated with in-hospital mortality were a low Karnofsky performance status score on admission (score: 40-10, adjusted odds ratio (aOR) 25.65, 95% CI 5.63–116.92 and score: 70-50, aOR 9.47, 95% CI 2.07–43.3), age over 89 years (aOR 3.68, 95% CI 1.08–12.46), Charlson Co-morbidity Index >5 (aOR 3.56, 95% CI 1.37–9.21), development of any health-care-associated infection (aOR 2.95, 95% CI 1.35–6.41), and development of any drug-related AE leading to discontinuation of anti-TB agents (seven patients were unable to resume treatment with anti-TB agents before death) (aOR 2.29, 95% CI 1.02–5.11).
In-hospital AEs (i.e. health-care-associated infection and drug-related AEs), as well as patient-related variables, were associated with in-hospital mortality among TB patients.
In-hospital mortality of pulmonary tuberculosis with acute respiratory failure and related clinical risk factors
2021, Journal of Clinical Tuberculosis and Other Mycobacterial Diseases
Citation Excerpt :
Tuberculosis is preventable and curable disease but has substantial mortality of 34 per 100,000 population [1]. In-hospital mortality of PTB patients range 2–17% and dramatically increased to 40–60% in patients who developed acute respiratory failure [2–8]. Acute respiratory failure caused by PTB is rare but PTB with complicated comorbidities and sepsis can lead to ARF and death.
Data on acute respiratory failure (ARF) in pulmonary tuberculosis (PTB) patients is limited. This study aims to investigate in-hospital mortality, its clinical risk factors and the accuracy of the existing scoring system in predicting in-hospital mortality.
An observational prospective cohort study involving PTB patients with ARF in tertiary hospital, between January 2017 and December 2018, was conducted. The in-hospital mortality was predicted using the National Early Warning Score 2 (NEWS2), quick Sequential Organ Failure Assessment (qSOFA) and CRB-65. Regression models were run to analyze the clinical risk factors for in-hospital Mortality. Sensitivity and specificity of scoring systems were calculated using a Wilson score interval.
A total of 111 subjects were included. Most of subjects were hypoxemic type respiratory failure (68.5%), advanced lesions (62.2%), new cases (70.3%) and pneumonia co-infection (72.1%) patients. Invasive mechanical ventilation was utilized for 29.73% of cases. There were 53 (47.75%) in-hospital mortality cases and its risk factors were intensive phase treatment (3.34 OR; CI95% 1.27–8.78), P/F ratio < 100 (OR 4.30; CI 95% 1.75–10.59) and renal insufficiency (4.09 OR; CI95% 1.46–11.49). The sensitivity and specificity of NEWS2 ≥ 6, qSOFA ≥ 2 and CRB-65 ≥ 2 were 62.26% and 67.24%; 60.38% and 72.41%; 41.51% and 84.48% respectively.
Most of PTB with ARF were new cases, advanced lesion and hypoxemic type respiratory failure. Intensive phase treatment, severe hypoxemia and renal insufficiency are independent predictors of in-hospital mortality in PTB patients with ARF. NEWS2, qSOFA and CRB-65 scores were poor to predict the in-hospital mortality.
Pulmonary vs. extra-pulmonary tuberculosis hospitalizations in the US [1998–2014]
2020, Journal of Infection and Public Health
Few studies have explored the relative burden and trends in pulmonary (PTB) vs. extra-pulmonary (EPTB) tuberculosis in the United States using a nationally representative sample.
This study examined trends in hospitalization rates, length-of-stay (LOS), in-hospital mortality and inflation-adjusted charges, for PTB vs. EPTB using the Nationwide/National Inpatient Sample (NIS) from 1998 to 2014. Descriptive and multivariable analyses (linear, negative binomial and logistic) were utilized adjusting for demographics, co-morbidity and hospital characteristics.
During the study period there were a survey-adjusted, estimated 258,631 PTB (75.5%), 76,476 EPTB (22.3%) and 7552 concurrent PTB and EPTB (2.2%) discharges. Whites accounted for 27.6% of PTB, 21.9% of EPTB and 17.6% of concurrent discharges; and self-pay or no insurance accounted for 22.2%, 18.4%, and 25.9%, respectively. EPTB was more common among blacks (22.5%), and combined TB more common among Hispanics (24.8%). Mean LOS was 11.4 days, 13.2 days, and 19.5 days; with mean nominal charges of $48,031, $62,255, and $89,364 for PTB, EPTB and combined TB respectively. Inpatient mortality for all three groups was approximately 5.7%. Miliary TB and TB of meninges and central nervous system were positively associated with mortality (odds ratios of 2.44 and 2.11, respectively), as was alcohol abuse (OR 1.21). Trend analyses showed decreased hospitalizations for all TB types, no change in LOS trends, decreased mortality for PTB and ETB and increased charges for PTB and ETB from 1998 to 2014. Increased utilization, higher charges and higher risk of mortality (to some extent) among the EPTB cases warrant improved methods for screening, diagnosis and treatment.
Though rates of TB hospitalization are declining, EPTB is becoming relatively more common and is more costly compared to pulmonary TB. Screening methods that focus on identification of ETB contrary to current practice guidelines are needed to aid ETB case finding.
Coexistence of dementia with smear-positive pulmonary tuberculosis is associated with patient in-hospital mortality
2019, Respiratory Investigation
Citation Excerpt :
Previous studies have shown that the physical characteristics and comorbidities of specific patient conditions are related to TB mortality. These conditions include human immunodeficiency virus (HIV) coinfection [5–8], malignancies [9], chronic lung [9], heart [9], and renal [7,10] diseases, cavitation [11], anemia [11], malnutrition [3,10], low body weight [6,12], old age [2,4–6,13], smoking habit [13], respiratory failure [4,10,13], and decreased activity of daily living (ADL) [4,14]. Japan has the highest levels of life expectancy [15].
The impact of dementia on the survival of patients with pulmonary tuberculosis (TB) remains unclear. This study sought to describe the risk factors influencing in-hospital mortality in patients with pulmonary TB and comorbid dementia.
A 9-y, medical record-based retrospective study of hospitalized adult patients with newly diagnosed, smear-positive, non-multidrug-resistant pulmonary TB without human immunodeficiency virus infection was performed. Clinical presentations, biochemical tests, radiographic findings, and clinical outcomes were collected. Variables were compared between groups. Statistically significant (p-value < 0.05) variables were entered into a multivariate stepwise logistic regression model. Survival analysis was performed using the Kaplan-Meier method, and groups were compared by log-rank test.
Of the 279 enrolled patients (178 men; median age, 76 y), the mortality rate was 12.2% (34/279). Univariate analysis showed a higher frequency of dementia in patients who died in hospital than that in surviving patients. Multivariate stepwise logistic analysis showed that dementia was significantly associated with higher rates of in-hospital mortality (odds ratio, 3.20; 95% confidence interval, 1.15–8.88, p = 0.026). In addition, subgroup survival curves showed that dementia was associated with reduced survival rates, even after adjusting for age (log-rank test, p = 0.0007).
The comorbidity of dementia with pulmonary TB was associated with patient in-hospital mortality. Medical practitioners should be aware of dementia in patients with smear-positive pulmonary TB to identify high-mortality groups.
Sociodemographic and clinical risk factors associated with in-hospital tuberculosis mortality in Türkiye, 2008-2018
2024, Tuberkuloz ve Toraks
Deaths from tuberculosis: differences between tuberculosis-related and non-tuberculosis-related deaths
2023, Frontiers in Public Health

View all citing articles on Scopus

View full text

Chest

Clinical InvestigationsINFECTIONSHospitalizations for Tuberculosis in the United States in 2000: Predictors of In-Hospital Mortality

Study objectives:

Design, setting, and patients:

Measurements and results:

Conclusions:

Section snippets

Study Design and Population

Database

Results

Discussion

Chest

Am J Infect Control

J Chronic Dis

J Clin Epidemiol

Chest

J Clin Epidemiol

Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country: WHO Global Surveillance and Monitoring Project

JAMA

Center for Disease Control and Prevention. Infectious Diseases Society of America. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis

Am J Respir Crit Care Med

Quality of life in tuberculosis: patient and provider perspectives

Qual Life Res

Hospitalized tuberculosis patients and predictors of mortality [abstract]

Am J Respir Crit Care Med

Long-term hospitalization for tuberculosis control: experience with a medical-psychosocial inpatient unit

JAMA

Delay in diagnosis among hospitalized patients with active tuberculosis: predictors and outcomes

Am J Respir Crit Care Med

Health-care expenditures for tuberculosis in the United States

Arch Intern Med

Causes and costs of hospitalization of tuberculosis patients in the United States

Int J Tuberc Lung Dis

Effect of HIV infection and tuberculosis on hospitalizations and cost of care for young adults in the United States, 1985 to 1990

Ann Intern Med

Goodness-of-fit testing for the logistic regression model when the estimated probabilities are small

Biom J

Tuberculosis morbidity among US-born and foreign-born populations: United States, 2000

MMWR Morb Mortal Wkly Rep

Tuberculosis in the disadvantaged groups

Curr Opin Pulm Med

Tuberculosis and race/ethnicity in the United States: impact of socioeconomic status

Am J Respir Crit Care Med

Tuberculosis morbidity: United States 1995

MMWR Morb Mortal Wkly Rep

In-hospital mortality following acute exacerbations of chronic obstructive pulmonary disease

Arch Intern Med

Clinical Investigations
INFECTIONS
Hospitalizations for Tuberculosis in the United States in 2000: Predictors of In-Hospital Mortality