Decrease in Tuberculosis Cases During COVID-19 Pandemic

Abstract and Introduction

Abstract

We analyzed a pharmacy dataset to assess the 20% decline in tuberculosis (TB) cases reported to the US National Tuberculosis Surveillance System (NTSS) during the coronavirus disease pandemic in 2020 compared with the 2016–2019 average. We examined the correlation between TB medication dispensing data to TB case counts in NTSS and used a seasonal autoregressive integrated moving average model to predict expected 2020 counts. Trends in the TB medication data were correlated with trends in NTSS data during 2006–2019. There were fewer prescriptions and cases in 2020 than would be expected on the basis of previous trends. This decrease was particularly large during April–May 2020. These data are consistent with NTSS data, suggesting that underreporting is not occurring but not ruling out underdiagnosis or actual decline. Understanding the mechanisms behind the 2020 decline in reported TB cases will help TB programs better prepare for postpandemic cases.

Introduction

The coronavirus disease (COVID-19) pandemic has affected many areas of public health, including tuberculosis (TB) prevention and response activities.^[1] TB cases reported to the US National Tuberculosis Surveillance System (NTSS) in 2020 decreased 20% compared with the average number of cases reported during 2016–2019.^[2] Although some annual decline is expected on the basis of public health investments in TB control and prevention, TB incidence decreased an average of only 2%–3% annually during the previous 10 years.^[3] A decline of nearly 20% raises concern that TB cases are being left undetected or unreported to public health agencies. A sharp decline in TB incidence in 2020 is possible, potentially because of control efforts undertaken to combat the COVID-19 pandemic or reduced immigration, leading to fewer cases among persons newly arriving in the United States from regions with higher TB incidence. We therefore sought to determine the extent to which this decline is actual, a surveillance artifact caused by underreporting, or representative of delayed or missed TB diagnoses. Understanding the underlying cause will help TB programs better allocate resources and prepare for TB cases after the pandemic. Analysis of TB-related trends in data sources unlikely to be affected by public health disruptions is a critical way to evaluate the mechanisms behind the reported decline.

One such source is the IQVIA (https://www.iqvia.com) prescription dataset, which captures >88% of all outpatient prescription activity in the United States, including retail, mail, and long-term care channels. These data have been used in public health to answer a variety of questions, including estimating costs of HIV preexposure prophylaxis (PrEP);^[4] analyzing the demographics of persons who have been prescribed PrEP;^[5] identifying opioid prescription patterns;^[6–8] and assessing naloxone, antibiotic, and hydroxychloroquine prescriptions.^[9–12] Pharmacy data are particularly valuable for TB disease because of the unique drug regimens used to treat TB. Initial treatment for newly diagnosed drug-susceptible TB disease typically consists of 4 drugs: isoniazid, rifampin, ethambutol, and pyrazinamide.^[13] All 4 drugs are taken in the first 2 months. If drug susceptibility testing results do not demonstrate resistance to isoniazid or rifampin, this intensive phase is typically followed by a continuation phase consisting of just isoniazid and rifampin for an additional 2–4 months or longer, depending on response to treatment. Rifampin is used to treat multiple diseases, including TB, Neisseria meningitidis, Haemophilus influenzae, leprosy, and endocarditis. Rifampin and isoniazid (alone or in combination) are also used to treat latent TB infection (LTBI). Ethambutol is used to treat TB and nontuberculous mycobacteria. Isoniazid is used to treat TB disease and LTBI and is rarely used for other diseases. Pyrazinamide is only used to treat TB disease. Thus, we focused on individual isoniazid and pyrazinamide prescriptions, because these prescriptions should generally indicate TB treatment even in the absence of information on concurrent prescriptions.^[14]

Although most US TB cases are treated in public health clinics that dispense their own medication, some cases are treated by private providers or by clinics that have their TB medications filled by retail pharmacies. Therefore, although it would not be practical to determine overall US TB disease incidence on the basis of outpatient pharmacy dispensing data, assessing the trend in dispensing of TB drugs is possible. We first determined whether the trends in IQVIA's TB medication prescription data correlated with trends observed in NTSS data before the pandemic (pre-2020). Once this correlation was established, we compared changes in IQVIA TB prescription data with changes in TB cases reported to NTSS in 2020 to assess potential underreporting of TB cases to public health.