This retrospective cohort study quantified 12-month all-cause healthcare expenditures for commercially insured adults with tuberculosis in the United States and identified clinical predictors of higher costs. Using administrative claims, the authors report substantial mean and median costs and show that specific comorbidities and comorbidity burden are associated with markedly increased expenditures.
The analysis used the Merative (TM) MarketScan (R) Commercial Claims Database covering years 2013 to 2018. Participants were adults aged 18 years or older with a diagnosis of TB disease identified in claims. The final analytic sample comprised 303 individuals. The study used de-identified secondary data obtained under license and was determined to be exempt from review by the UCSF Institutional Review Board.
The primary outcome was total 12-month all-cause healthcare costs calculated from the date of TB diagnosis, incorporating outpatient, inpatient, and pharmacy claims. Reported summary statistics for costs were:
These figures indicate a skewed distribution with a subset of patients incurring very high costs relative to the median.
Key baseline characteristics and comorbidities in the cohort included a median age of 46 years and 158 individuals (52%) who were male. Reported comorbidities included:
The study grouped several conditions (HIV, HBV, hepatitis C, alcohol use disorder, drug use disorder) and also used Charlson comorbidity scores to quantify overall comorbidity burden. Median 12-month costs were substantially higher among persons with any listed comorbidity compared with those without comorbidity ($11,930 [IQR $4,194 to $36,073] vs $3,385 [IQR $1,506 to $8,609]; p<0.001).
Adjusted cost ratios (aCR) were estimated using a Gamma generalized linear model to identify independent predictors of higher 12-month costs. The principal findings were:
These adjusted estimates identify HIV and substance use-related diagnoses, as well as overall comorbidity burden, as key drivers of higher healthcare spending in this commercially insured TB population.
The authors note that the increased costs associated with HIV coinfection were driven predominantly by pharmacy expenditures; the pharmacy-specific adjusted cost ratio for HIV was reported as 16.4. The study calculated total costs as the sum of outpatient, inpatient, and pharmacy claims, but specific adjusted ratios for inpatient and outpatient components were not detailed beyond this finding about pharmacy costs.
The study concludes that healthcare costs for persons with TB who are commercially insured are substantial and are further amplified by the presence of comorbidities such as HIV and drug use disorder. The authors suggest that enhanced screening, improved care coordination, and focused management of TB and high-risk comorbid conditions could yield significant cost savings. These implications are framed in the context that most prior cost estimates emphasized public-sector care and that costs among commercially insured populations warrant attention.
The analysis used licensed, de-identified commercial claims data that are not publicly available; access was provided through the study institution under license and restrictions apply. As a claims-based retrospective study, limitations inherent to administrative data apply (for example, case identification and comorbidity ascertainment by claims codes). The article is presented as a preprint and has not undergone peer review; the source notes this should not be used to guide clinical practice. The authors declared no competing interests and reported that the work was exempt from further IRB review at UCSF.