Time to Progress Through the Latent Tuberculosis Infection Care Cascade in Primary Care

Abstract

Objectives:

Although treatment of latent tuberculosis infection (LTBI) is highly effective for preventing tuberculosis disease, diagnosis requires several steps. A care cascade is a helpful framework for identifying the clinical events necessary for successful LTBI treatment, yet the timing of progression through these events is poorly characterized. We calculated the timing of an LTBI care cascade and assessed whether timing was associated with increased cascade progression.

Methods:

We conducted an observational study through the Tuberculosis Epidemiologic Studies Consortium using electronic health records to calculate the timing of steps in the LTBI care cascade in the United States from 2020 through 2022 (N = 693 254 patients). Using receiver operating characteristic curves, we identified optimal maximum times (OMTs) associated with cascade progression, and we calculated risk ratios to compare outcomes between patients who advanced above and below OMTs.

Results:

The median (IQR) time from most recent test order to test result was 3 (1-4) days; from most recent test order to imaging order, 7 (4-18) days; from imaging order to treatment prescription, 17 (6-40) days; from treatment prescription to treatment start, 0 (0-1) days; from treatment start to treatment completion, 116 (103-120) days; and for the entire cascade, 150 (133-189) days. Risk ratios for cascade progression ranged from 1.10 to 2.81, demonstrating that patients completing steps below the OMTs showed marginally higher likelihoods of progression through the care cascade.

Conclusions:

These findings highlight the importance of sustained follow-up and caution against early classification of patients as lost to care, supporting public health strategies for health systems, programs, health care providers, and care navigators that emphasize long-term engagement to maximize the preventive effect of LTBI treatment.

Keywords

latent tuberculosis infection care cascade time primary care

Tuberculosis (TB) is the leading cause of death due to a single infectious agent, with 1.25 million deaths globally in 2023.¹ An estimated 24% of the global population, or 1.8 billion people, may have latent TB infection (LTBI).^1,2 In the United States, up to 13 million people are estimated to have LTBI,³ and the progression of LTBI to TB disease accounts for >80% of all TB disease diagnoses domestically.⁴ Because LTBI treatment can reduce the risk of progression to TB disease by as much as 90%,⁵ treatment of people with LTBI is crucial to TB elimination goals in the United States.⁶

The care cascade model can be used to identify gaps in care and inform decisions on prioritizing health care resources.⁷ Successfully completing the LTBI care cascade involves several sequential steps: identifying and screening individuals at high risk of TB exposure, testing for LTBI, excluding TB disease, prescribing treatment, initiating treatment, and completing treatment.⁸ Patients may be lost to follow-up at any of these steps. Despite recommendations,⁹ a meta-analysis highlighted substantial drop-offs due to inadequate testing, lack of treatment initiation, and low completion rates among those who started treatment.¹⁰ More recent studies on US cascades identified similar trends in public health¹¹ and primary care.^8,12-15

While prior analyses have identified important gaps in care, none have focused on the timing between all steps in the LTBI care cascade and the impact of timing on progression to subsequent steps. To our knowledge, no prior study has investigated the timing of full LTBI care cascades in primary care. Existing studies on care cascade timing for other infectious diseases, such as HIV and hepatitis C virus, found differences in timing across subgroups and comorbidities but ultimately concluded no straightforward relationship between timing and patient outcomes.^16-18 Prior research on timing in TB care has primarily examined delays from symptom onset to TB disease diagnosis,^19-24 which differs from asymptomatic LTBI and does not encompass all cascade steps. Additionally, these studies typically occur in the context of mass public health campaigns in developing countries, where access to primary care is limited.

Understanding timing between cascade steps can facilitate monitoring of patient progress and prompt high-yield interventions to improve uptake of TB prevention. For example, prolonged delays between diagnosis and treatment initiation may signal weak links in the care cascade where programmatic support or engagement would improve continuation of care. Thus, we evaluated the time between critical steps in the LTBI care cascade in primary care to identify whether certain time frames optimize completion of those steps. We also sought to determine which optimal maximum times (OMTs) between steps, such as time to imaging following a positive test result, were most predictive for cascade progression and might offer the highest yield for systems-level improvements.

Methods

Study Population and Data Collection

We conducted a longitudinal study through the Tuberculosis Epidemiologic Studies Consortium (TBESC-III), a partnership between the Centers for Disease Control and Prevention (CDC) and 4 primary care networks in the United States (eTable 1 in the Supplement). Electronic health record (EHR) data on patient demographic characteristics, clinic visits, TB diagnostics, imaging, prescriptions, and International Classification of Diseases codes were collected from each site for a minimum of 1 year from 2020 through 2022 and mapped to a common data dictionary. Each site chose study start and end dates to best reflect the existing clinical protocols. We excluded patients who did not have a clinic visit during the study period; were born in the United States; indicated English as their primary language if country of birth was unavailable; or had any TB or LTBI tests, diagnosis, or treatment before the study. To maximize sample sizes, we did not require progression through all prior cascade steps for inclusion in subsequent steps (eFigure 1 in the Supplement).

Time From Test Order to Test Result

We calculated the time from each patient’s most recent interferon gamma release assay (IGRA) order date with a positive or negative test result to the date that the result was reported (result date). Tuberculin skin tests were excluded because our patients largely originate from countries where bacillus Calmette-Guérin vaccination is common, and guidelines recommend IGRA testing in this context.²⁵ IGRAs that were indeterminate or failed were also excluded. If a patient had LTBI treatment, we used the most recent positive IGRA test result before the earliest LTBI treatment.

Time From Positive Test Order to Imaging Order

Analyzing imaging records in EHR data presents challenges. Patients undergo imaging for many reasons unrelated to TB. Health care providers may also opt to use previously administered imaging instead of ordering new imaging. Absent reviewing health care provider notes, which was beyond the scope of this study, we could not determine which imaging records were LTBI related; thus, we included all pulmonary images (x-rays and computed tomography scans) in the analysis. We calculated the time from each patient’s most recent positive IGRA order date to the closest pulmonary imaging order date and assumed that the imaging closest in time to the positive IGRA order date was LTBI related. We excluded any imaging that occurred >180 days before or after the IGRA order date because clinical practice generally requires imaging to exclude TB disease within 6 months of testing before starting LTBI treatment. Because health care providers could rely on prior imaging rather than reordering, the imaging order may have preceded the positive test result order.

Time From Imaging Order to Treatment Prescription

We analyzed individual prescriptions for isoniazid, rifampin, rifabutin, and rifapentine to identify regimens for LTBI and exclude regimens for TB disease and other conditions.⁸ We calculated the time from each patient’s earliest LTBI treatment prescription date to the closest prior pulmonary imaging order date. We assumed that the closest imaging to any LTBI treatment prescription was LTBI related; we excluded imaging performed >180 days before the earliest LTBI prescription, consistent with clinical practice not relying on imaging >6 months. Because positive test order date and the LTBI treatment prescription date were independently matched to the closest imaging order date, they did not always match to the same record.

Time From Treatment Prescription and Start

A patient starts treatment once the EHR system records that an LTBI prescription has been dispensed. We calculated the time from each patient’s earliest prescription date to the earliest prescription pickup date for any drug in an LTBI regimen.

Time From Treatment Start to Treatment Completion

A patient completes treatment once they finish all pills in their regimen.^26,27 We used prescription pickup dates, the number of pills dispensed, and the expected pill consumption frequency (eg, daily, weekly) to approximate regimen completion dates, and we calculated the time to the earliest prescription pickup date. For people with multiple regimens, we calculated the time from their earliest prescription pickup of any regimen to the end of their first complete regimen (Figure 1).

Figure 1.

Sample selection of dates for steps in the care cascade for patients with latent tuberculosis infection (LTBI), 4 primary care health networks, United States, 2020-2022. Hypothetical demonstration of how the timing of each step in the care cascade was calculated, when multiple possible events could be used for each patient. This shows a timeline of possible LTBI-related events in a patient’s record. (A) Most recent positive interferon gamma release assay (IGRA) test order date prior to any LTBI treatment (TX) within the study period was used for the time from test ordered to test result. (B) Most recent positive IGRA test result was matched to the closest image order date. (C) Earliest LTBI TX prescription date within the study period was matched to the closest image order date (not necessarily the same imaging as selected in B). (D) Earliest LTBI TX prescription date within the study period was matched to the earliest LTBI prescription pickup date at the pharmacy. (E) Earliest LTBI prescription pickup date at the pharmacy was matched to the end of the first complete regimen. Incomplete attempts at the regimen are captured, and any subsequent complete regimens are excluded. (F) Most recent positive IGRA test result was matched to the end of the first complete regimen.

Because of nuances in EHR data, some calculations resulted in times outside the expected minimum and maximum durations (eTable 2 in the Supplement). These calculations occurred because of duplicate records, multimonth bulk pill dispensing, or other data quality issues. In such cases, we recalculated the completion date based on when the minimum medication dosage was achieved. Patients remaining whose data fell outside the minimum and maximum were dropped.^26,27

Time to Progress Through the Entire Cascade

We calculated the time from each patient’s most recent positive IGRA order date before their earliest LTBI treatment to their first treatment completion date (Figure 1). The calculated time for the entire cascade may be different than the sum of all the individual times between steps because people may have had multiple tests, imaging, and treatment regimens.

Receiver Operating Characteristic Curves

For each pair of cascade steps (eg, test order to imaging order), we defined a patient as progressing to the next cascade step if they completed the subsequent step in the cascade (eg, treatment prescription). We constructed receiver operating characteristic (ROC) curves by iterating through all possible time thresholds and plotting the true-positive rate against the false-positive rate that a patient completing a given cascade step at or below each time threshold progresses to the next cascade step.²⁸ We selected OMTs for each pair of cascade steps by maximizing the difference between the true-positive rate and false-positive rate. In addition to ROC curves that predict progression to the next cascade step, we constructed ROC curves that predict completion of the entire cascade and calculated area under the curve (AUC), risk ratios (RRs) for progression, and 95% CIs.

Stratifications by Age, Race, and Health Insurance

For each pair of cascade steps, we defined each patient’s age and health insurance status based on their age and health insurance status at the initial cascade step. We classified patients who self-identified as Hispanic as Hispanic only and those who identified with multiple races as multiple races. We used the Kruskal-Wallis H test to analyze data, with P < .05 considered significant. Given the small sample sizes, further modeling was not possible.

Statistical Analysis

We conducted statistical analysis and visualizations using Python 3.11.5 with established libraries.^29-32

Ethical Considerations

This activity was reviewed by CDC, deemed research not involving human subjects, and conducted consistent with applicable federal law and CDC policy (see eg, 45 CFR part 46; 21 CFR part 56; 42 USC §241[d], 5 USC §552a, 44 USC §3501 et seq).

Results

Time Between Care Cascade Steps

A total of 693 254 patients had a visit during the study period, of whom 375 707 met our study criteria. Of 16 910 patients with data on time from test order to test result, the median (IQR) time was 3 (1-4) days overall (Figure 2A) and 3 (2-4), 3 (3-3), 3 (2-4), and 0 (0-3) days for sites A through D, respectively (eFigure 2 in the Supplement). Medical record review of selected outliers indicated that patients experiencing excess times (>365 days; n = 16) were often pregnant or diagnosed with acute conditions that made managing LTBI a secondary priority. Hence, many of these ordered tests were not performed until much later.

Figure 2.

Time between all steps in the cascade of care for patients with latent tuberculosis infection (LTBI), 4 primary care health networks (N = 16 190), United States, 2020-2022. (A) Time between nonregimen steps, with time restricted between 0 and 180 days. Not represented in the box plots are 166 records for test ordered to test result with time beyond 180 days and 224 records for test ordered to image ordered with negative time. (B) Time between steps stratified by regimen. Line, median; box, IQR; whiskers, 1.5 x IQR; circles, outliers. (C) Summary statistics table with quartiles and averages for time between all steps. Statistics for test ordered to image ordered exclude records with negative time. Abbreviations: 3HP, 3 months of isoniazid and rifapentine; 3HR, 3 months of isoniazid and rifampin or rifabutin; 4R, 4 months of rifampin or rifabutin; 6H/9H, 6 or 9 months of isoniazid; TX, treatment.

Of 2016 patients with data on time from test order to imaging order, 221 completed chest imaging before the LTBI test order date, and 74 with time >180 days were dropped. Excluding all negative times, the median (IQR) days from test order to imaging order was 7 (4-18) days overall and 6 (3-14), 7 (3-13), 17 (9-31), and 11 (5-21) days for sites A through D, respectively (eFigure 3 in Supplement).

Of 972 patients with data on time from imaging order to treatment prescription, 45 with time >180 days were dropped. The median (IQR) time from imaging order to treatment prescription was 17 (6-40) days overall and 33 (15-68), 8 (3-21), 12 (4-38), and 16 (7-33) days for sites A through D, respectively (eFigure 4 in the Supplement).

Of 1069 patients with data on time from treatment prescription to treatment start, the median (IQR) time was 0 (0-1) days overall and 0 (0-0), 1 (0-6), 0 (0-1), and 0 (0-0) days for sites A through D, respectively (eFigure 5 in the Supplement).

Overall, 419 patients started and completed LTBI treatment. The median (IQR) time from treatment start to treatment completion was 90 (87-104) days for 3HR (3 months of isoniazid and rifampin or rifabutin), 84 (79-86) days for 3HP (3 months of weekly isoniazid and rifapentine), 117 (113-121) days for 4R (4 months of rifampin or rifabutin), 186 (178-211) days for 6H/9H (6 or 9 months of isoniazid), and 148 (124-215) days for multiple regimens (Figure 2B, eFigure 6 in the Supplement). Patients with multiple regimens had larger IQRs of times to complete treatment, but most completed treatment more quickly than those on a single 6H/9H regimen.

Overall, 395 patients progressed through the entire cascade. Regimen completion was the longest step and largely drove the total time. Patients undergoing 3HR, 3HP, and 4R (short-course regimens) completed the cascade in a median (IQR) time of 121 (102-137), 120 (98-177), and 152 (137-182) days, respectively. Patients undergoing 6H/9H and multiple regimens had a median (IQR) time of 277 (225-384) and 193 (167-307) days to complete the care cascade (Figure 2B, eFigure 7 in the Supplement).

OMTs for Cascade Progression

ROC curves identified OMTs for each pair of sequential cascade steps that was associated with cascade progression (Figure 3). Patients who received their test result within 1 day (identified OMT) of their test order had an RR of 1.11 (95% CI, 1.07-1.16; AUC, 0.56) of progressing to imaging order, meaning that they were 11% more likely to have imaging ordered than those who received their test result >1 day after their test order. Patients with imaging ordered within 21 days of test order had an RR of 1.30 (95% CI, 1.12-1.50; AUC = 0.53) for treatment prescription, and patients prescribed treatment within 32 days of imaging order had an RR of 1.18 (95% CI, 1.09-1.26; AUC = 0.60) for starting treatment. Patients who picked up their LTBI medication on the same day that it was prescribed had an RR of 1.11 (95% CI, 0.93-1.32; AUC = 0.52) for treatment completion. Site-stratified ROC curves showed variability in OMTs. All RRs were >1.0 and most were significant, but some AUC values were <0.5 (eTable 3 in the Supplement).

Figure 3.

Receiver operating characteristic (ROC) curves and cumulative incidence curves for time between latent tuberculosis infection care cascade steps, 4 primary care health networks, United States, 2020-2022. (A) ROC curves for time between 2 care cascade steps to predict progression to the next care cascade step or full cascade completion. The OMT to progression or completion was calculated by maximizing the difference between the TPR and FPR for each pair of cascade steps. The gray shaded circle on the ROC curve represents the TPR and FPR used for that OMT. The gray dashed line represents the line when TPR = FPR. (B) Cumulative incidence curves for time between 2 care cascade steps between patients who progressed to the next step or completed the cascade (black line) and patients who did not progress to the next step or did not complete the cascade (gray line). Vertical dotted line represents the OMT identified by ROC curves. (A, B) The columns, from left to right, represent time from test ordered to test result; time from test ordered to imaging ordered; time from imaging ordered to treatment (TX) prescribed; and time from TX prescribed to TX started. In each part, the top row represents progression to the next step in the care cascade, and the bottom row represents full cascade completion. AUC, area under the curve; FPR, false-positive rate; OMT, optimal maximum time; RR, risk ratio; TPR, true-positive rate.

ROC curves constructed to predict full cascade completion yielded similarly mixed findings. Patients who received their test result within 1 day of their test order had an RR of 1.37 (95% CI, 1.14-1.65; AUC = 0.53) for cascade completion. Patients with imaging ordered within 15 days of the test order had an RR of 1.21 (95% CI, 0.96-1.51; AUC = 0.52), and patients with treatment prescribed within 147 days of their imaging order had an RR of 2.81 (95% CI, 0.77-10.30; AUC = 0.48).

Stratifications by Age, Race, and Health Insurance Status

We found no significant differences in timing by various age groups, except for time from treatment prescription to treatment start (P = .004; eTable 4 in the Supplement). Among racial groups, the timing for all steps was significantly different except for time to progress through the entire cascade (P = .37). We found significant differences in all steps in the care cascade for patients with various types of health insurance, except for time from imaging order to treatment prescription (P = .69) and for the entire cascade (P = .82).

Discussion

Here, we present the most complete picture to date of time frames from LTBI screening to treatment completion in 4 US primary care networks. Our findings suggest that OMTs may exist between all cascade steps for maximizing cascade progression, although further investigation is warranted. Significant RRs but low AUC values for these OMTs suggest that, while a shorter time was correlated with an increased likelihood of progression to subsequent cascade steps, the correlation was weak, and many people still progress through the care cascade despite longer times between cascade steps. On a population level, individuals who require less time than the OMT to complete a given cascade step are more likely to progress to the next cascade steps than individuals who exceed the OMT. Inevitably in any care cascade, a proportion of patients will never progress regardless of timing, which explains the low AUC and significant RRs. Certain cascade steps, such as treatment prescription and treatment completion, had drop-off rates of 40% to 50%,⁸ which complicate the model’s ability to predict individual outcomes.

When predicting full cascade completion, ROC curves revealed similarly low AUC values despite RRs >1.0. Taken together, the stepwise and full cascade ROC curve results suggest that while shorter times are associated with an increased likelihood of cascade progression or completion, the low AUC values reflect substantial overlap in outcomes across time frames, with many patients progressing through and completing the care cascade despite delays. These findings suggest that timing has only a minor association with whether a patient will continue through the care cascade. As such, health systems, programs, health care providers, and care navigators are encouraged to continue outreach with all patients, including those who have long lags since completing the previous step.

Limitations

This analysis had several limitations. First, our study enrolled only patients who were non–US-born or indicated a primary language preference other than English if country of birth was missing, and participating TBESC-III primary care networks may not be representative of all primary care settings. Consequently, results may not be generalizable to other US populations for whom LTBI screening is recommended, such as individuals in congregate settings. Second, while we observed significant differences in timing related to sociodemographic factors, the extent to which these differences were also influenced by site is unclear. The demographic distribution of the populations served by various clinical sites differed, but sample sizes were too small to consider site or clinic factors in models with demographic characteristics. Third, we similarly lacked information on how specific clinic facilities may affect timing. For example, some but not all clinics had on-site imaging, a laboratory, or a pharmacy. Patients who were served at clinics that lacked these services may have faced additional barriers that contributed to longer times to complete specific cascade steps or not completing them at all. Finally, because our study relied solely on standardized variables and excluded free-text fields that may contain more context, we lacked insight into patient and health care provider motivations behind observed timings. We could not ascertain whether delays in care were driven by patients, health care providers, health systems, or other social and structural barriers. Previous studies examining delays between TB disease symptoms and diagnosis suggest that patient and health system factors contribute equally to extended delays.^20,21,33 Moreover, data routinely captured in EHR systems, such as International Classification of Diseases codes, may be incomplete or subject to coding inaccuracies. EHR systems are also specific to our sites, and patients who received care or filled prescriptions outside their health care provider network did not have those relevant records captured.

Conclusion

Understanding gaps in the LTBI care cascade is critical to informing the development and implementation of targeted interventions to maximize screening and treatment of disproportionately affected populations. While we identified OMTs, the effect of time on the likelihood of cascade progression and completion is unclear. Our findings suggest that patients with prolonged delays may still progress through the care cascade at rates that are comparable with those who have shorter times. Therefore, we encourage public health programs and systems-level interventions to prioritize sustained, long-term engagement with patients to support continued progression and completion of the care cascade.

Supplemental Material

sj-docx-1-phr-10.1177_00333549261447813 – Supplemental material for Time to Progress Through the Latent Tuberculosis Infection Care Cascade in Primary Care

Supplemental material, sj-docx-1-phr-10.1177_00333549261447813 for Time to Progress Through the Latent Tuberculosis Infection Care Cascade in Primary Care by Angus Zhuo-Long Wu, Kaylynn Aiona, Matthew Murrill, Masahiro Narita, Jacek Skarbinski and Kathryn Winglee in Public Health Reports®

Footnotes

Acknowledgements

The authors acknowledge the contributions to this article from Jagadheeswari Adhimurthy, MS, Kumar Batra, Juan (Antonio) Hernandez, Kuan-Chieh Huang, Jihming Lin, MS, Taylor Moore, MPH, and Bhumika Sharma, MBA, at the Peraton Corporation, and from Julie Espey, MPH, Danique Gigger, MPH, Mehabuba Rahman, MPH, Preeti Ravindhran, MPH, Laura Vonnahme, PhD, MPH, Thara Venkatappa, PhD, MPH, and Jonathan Wortham, MD, at the Centers for Disease Control and Prevention. They also acknowledge all TBESC-III sites, site staff, and study participants.

ORCID iDs

Angus Zhuo-Long Wu, BS

Kaylynn Aiona, PhD, MPH

Matthew Murrill, MD, PhD

Jacek Skarbinski, MD

Kathryn Winglee, PhD

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by funding from the Centers for Disease Control and Prevention Tuberculosis Epidemiologic Studies Consortium (contracts 75D30121C12877 to K.A., 75D30121C12878 to J.S., 75D30121C12880 to M.N., and 75D30121C12879 to M.M.).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated and analyzed during the current study are not publicly available because of health information protections.

Disclaimer

The findings and conclusions in this article are those of the authors and do not necessarily represent the views of CDC. References in this article to any specific commercial products, process, service, manufacturer, or company do not constitute its endorsement or recommendation by the US government or CDC.

Supplemental Material

Supplemental material for this article is available online. The authors have provided these supplemental materials to give readers additional information about their work. These materials have not been edited or formatted by Public Health Reports’s scientific editors and, thus, may not conform to the guidelines of the AMA Manual of Style, 11th Edition.

References

World Health Organization. Global Tuberculosis Report 2024. 2024. Accessed November 12, 2024. https://iris.who.int/bitstream/handle/10665/379339/9789240101531-eng.pdf?sequence=1

Kiazyk

Ball

. Latent tuberculosis infection: an overview. Can Commun Dis Rep. 2017;43(3-4):62-66. doi:10.14745/ccdr.v43i34a01

Miramontes

Hill

Yelk Woodruff

, et al. Tuberculosis infection in the United States: prevalence estimates from the National Health and Nutrition Examination Survey, 2011-2012. PLoS One. 2015;10(11):e0140881. doi:10.1371/journal.pone.0140881

Lin

Melendez-Torres

. Critical interpretive synthesis of barriers and facilitators to TB treatment in immigrant populations. Trop Med Int Health. 2017;22(10):1206-1222. doi:10.1111/tmi.12938

Zenner

Beer

Harris

Lipman

Stagg

van der Werf

. Treatment of latent tuberculosis infection: an updated network meta-analysis. Ann Intern Med. 2017;167(4):248-255. doi:10.7326/M17-0609

Stop TB USA Tuberculosis Elimination Plan Committee. A call for action on the tuberculosis elimination plan for the United States. March 22, 2010. Accessed May 14, 2026. https://stacks.cdc.gov/view/cdc/12076

Socías

Volkow

Wood

. Adopting the “cascade of care” framework: an opportunity to close the implementation gap in addiction care? Addiction. 2016;111(12):2079-2081. doi:10.1111/add.13479

Vonnahme

Ravindhran

Espey

, et al. Latent tuberculosis infection care cascade outcomes in primary care clinics in the Tuberculosis Epidemiologic Studies Consortium-III. Ann Epidemiol. 2026;118:110081. doi:10.1016/j.annepidem.2026.110081

US Preventive Services Task Force; Mangione

Barry

Nicholson

, et al. Screening for latent tuberculosis infection in adults: US Preventive Services Task Force recommendation statement. JAMA. 2023;329(17):1487-1494. doi:10.1001/jama.2023.4899

10.

Alsdurf

Hill

Matteelli

Getahun

Menzies

. The cascade of care in diagnosis and treatment of latent tuberculosis infection: a systematic review and meta-analysis. Lancet Infect Dis. 2016;16(11):1269-1278. doi:10.1016/S1473-3099(16)30216-X

11.

Holzman

Perry

Saleeb

, et al. Evaluation of the latent tuberculosis care cascade among public health clinics in the United States. Clin Infect Dis. 2022;75(10):1792-1799. doi:10.1093/cid/ciac248

12.

Vonnahme

Raykin

Jones

, et al. Using electronic health record data to measure the latent tuberculosis infection care cascade in safety-net primary care clinics. AJPM Focus. 2023;2(4):100148. doi:10.1016/j.focus.2023.100148

13.

Cochran

Tibbs

Haptu

Paradise

Bernardo

Tierney

. Scaling up latent tuberculosis infection testing and treatment for non–US born patients in a federally qualified community health center. J Immigr Minor Health. 2023;25(6):1482-1487. doi:10.1007/s10903-023-01514-0

14.

Prater

Holzman

Shah

. Programmatic effectiveness of latent tuberculosis care cascade in a community health center. J Immigr Minor Health. 2021;23(3):566-573. doi:10.1007/s10903-020-01047-w

15.

Shapiro

Gupta

Lan

Kim

. Latent tuberculosis screening cascade for non–US-born persons in a large health system. Open Forum Infect Dis. 2023;10(7):ofad303. doi:10.1093/ofid/ofad303

16.

Valencia

Lazarus

Ceballos

, et al. Differences in the hepatitis C virus cascade of care and time to initiation of therapy among vulnerable subpopulations using a mobile unit as point-of-care. Liver Int. 2022;42(2):309-319. doi:10.1111/liv.15095

17.

Kamali

Shumbusho

Barnhart

, et al. Time to complete hepatitis C cascade of care among patients identified during mass screening campaigns in rural Rwanda: a retrospective cohort study. BMC Infect Dis. 2022;22(1):272. doi:10.1186/s12879-022-07271-z

18.

Supervie

Marty

Lacombe

Dray-Spira

Costagliola

; FHDH-ANRS CO4 Study Group. Looking beyond the cascade of HIV care to end the AIDS epidemic: estimation of the time interval from HIV infection to viral suppression. J Acquir Immune Defic Syndr. 2016;73(3):348-355. doi:10.1097/QAI.0000000000001120

19.

El Halabi

Palmer

McDuffie

, et al. Measuring health-care delays among privately insured patients with tuberculosis in the USA: an observational cohort study. Lancet Infect Dis. 2021;21(8):1175-1183. doi:10.1016/S1473-3099(20)30732-5

20.

Sreeramareddy

Panduru

Menten

Van den Ende

. Time delays in diagnosis of pulmonary tuberculosis: a systematic review of literature. BMC Infect Dis. 2009;9:91. doi:10.1186/1471-2334-9-91

21.

Park

Min

, et al. Timely pulmonary tuberculosis diagnosis based on the epidemiological disease spectrum: population-based prospective cohort study in the Republic of Korea. JMIR Public Health Surveill. 2024;10:e47422. doi:10.2196/47422

22.

Min

Kim

, et al. Time delays and risk factors in the management of patients with active pulmonary tuberculosis: nationwide cohort study. Sci Rep. 2022;12(1):11355. doi:10.1038/s41598-022-15264-w

23.

Shewade

Govindarajan

Thekkur

, et al. MDR-TB in Puducherry, India: reduction in attrition and turnaround time in the diagnosis and treatment pathway. Public Health Action. 2016;6(4):242-246. doi:10.5588/pha.16.0075

24.

Storla

Yimer

Bjune

. A systematic review of delay in the diagnosis and treatment of tuberculosis. BMC Public Health. 2008;8:15. doi:10.1186/1471-2458-8-15

25.

Division of Tuberculosis Elimination, National Center for HIV, Viral Hepatitis, STD, and TB Prevention, Centers for Disease Control and Prevention. Clinical testing guidance for tuberculosis: interferon gamma release assay. May 9, 2024. Accessed March 6, 2026. https://www.cdc.gov/tb/hcp/testing-diagnosis/interferon-gamma-release-assay.html

26.

Sterling

Njie

Zenner

, et al. Guidelines for the treatment of latent tuberculosis infection: recommendations from the National Tuberculosis Controllers Association and CDC, 2020. MMWR Recomm Rep. 2020;69(1):1-11. doi:10.15585/mmwr.rr6901a1

27.

Division of Tuberculosis Elimination, National Center for HIV, Viral Hepatitis, STD, and TB Prevention, Centers for Disease Control and Prevention. Treatment for latent tuberculosis infection. February 13, 2020. Accessed December 6, 2024. https://www.cdc.gov/tb/hcp/treatment/latent-tuberculosis-infection.html

28.

Park

Goo

. Receiver operating characteristic (ROC) curve: practical review for radiologists. Korean J Radiol. 2004;5(1):11-18. doi:10.3348/kjr.2004.5.1.11

29.

Virtanen

Gommers

Oliphant

, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020;17(3):261-272. doi:10.1038/s41592-019-0686-2

30.

Pedregosa

Varoquaux

Gramfort

, et al. Scikit-learn: machine learning in Python. JMLR. 2011;12(85):2825-2830.

31.

Davidson-Pilon

. lifelines: survival analysis in Python. J Open Source Softw. 2019;4(40):1317. doi:10.21105/joss.01317

32.

Waskom

. seaborn: statistical data visualization. J Open Source Softw. 2021;6(60):3021. doi:10.21105/joss.03021

33.

Sreeramareddy

Qin

Satyanarayana

Subbaraman

Pai

. Delays in diagnosis and treatment of pulmonary tuberculosis in India: a systematic review. Int J Tuberc Lung Dis. 2014;18(3):255-266. doi:10.5588/ijtld.13.0585

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.76 MB