Introduction
The global COVID-19 pandemic overwhelmed and disrupted health services worldwide, negatively impacting the delivery of TB services in many countries. Globally, there were 18% fewer TB cases diagnosed in 2020 compared to 2019 [
1], and reports from diverse countries have measured significantly longer delays in TB diagnosis during the pandemic [
2‐
5]. Modeling studies suggest that the pandemic has caused substantial increases in TB-associated mortality by delaying TB diagnoses or causing diagnoses to be missed completely [
6,
7].
Despite the abundance of quantitative evidence of the pandemic’s impact on TB diagnoses, there have been few published reports describing the ways in which the pandemic affected people’s pathway to diagnosis from their own perspectives. Qualitative studies exploring patients’ experience with TB during the pandemic have reported patients’ concerns around the financial and social costs of lockdowns, their ability to continue accessing TB medications, and their fear of infection [
8‐
11]. However, these studies have not identified ways in which the pandemic had impacted patients’ experience obtaining a TB diagnosis.
Understanding how the COVID-19 pandemic affected people’s pathways to TB diagnosis is important for improving the resiliency of health systems so that future public health emergencies do not have such a damaging impact on TB detection. We therefore sought to understand the perspective of people with TB on how the pandemic affected their ability to be diagnosed with TB. We performed a mixed methods study in Peru, a country with a high TB burden, which experienced serious morbidity and mortality as a result of the COVID-19 pandemic.
Methods
Study design
Prior to the pandemic, we designed a sequential explanatory mixed-methods study to understand barriers to prompt TB diagnosis in Lima, Peru. However, after enrolling half the cohort of TB patients during the first year of the pandemic, we realized that the pandemic’s effect on the health system was causing delays in diagnosis. Therefore, we delayed enrollment of the second half of the cohort until a year after the study start date in hopes that health services would have normalized, allowing us to quantify the effect of the acute stage of the pandemic. We performed quantitative analysis of surveys to assess differences in TB diagnostic experiences between the two periods and qualitative analysis of interviews to explain these differences. Although the original study design was sequential, we used a convergent approach to integrating quantitative and qualitative information on the impact of the pandemic since this was not an original study objective and the quantitative results did not inform qualitative data collection on this topic.
Study setting
Our study took place in Carabayllo, a district on the periphery of Peru’s capital Lima. Peru has an estimated TB incidence of 116 per 100,000 population and the second-highest TB burden in the Americas. Peru’s health system was greatly impacted by the COVID-19 pandemic [
12], with health care visits for other conditions decreasing by 64% [
13] and TB diagnoses decreasing by 25% in 2020 [
1]. The first case of COVID-19 was detected in March of 2020, with increasing cases by April. The first wave of pandemic peaked in August 2020 and the second wave in March-April 2021 [
14]. Vaccines became available in February 2021, and half the population had received at least one dose by September [
15]. As of May 2022, Peru had experienced the highest reported cumulative population mortality rate from COVID-19 in the world, with deaths concentrated during the first and second waves, prior to widespread vaccine availability [
16].
Quantitative data collection and analysis
We enrolled a convenience sample of 100 adults who had recently enrolled for TB treatment at the 12 Ministry of Health primary care facilities in Carabayllo, purposively distributing enrollments across the different facilities. Patients were eligible for enrollment if they were ≥18 years old and had started TB treatment within the last month. We enrolled half the sample during November 2020–April 2021 (period 1) and half during October 2021–February 2022 (period 2). Period 1 was during the first year of the pandemic in Peru, encompassing the first and second waves of COVID-19. Period 2 was after widespread vaccine availability.
Structured surveys asked participants when they first felt sick with their current episode of TB and about each visit to the health system until the point where they were diagnosed with TB. For each visit, participants were asked the date of the visit and the amount of money that they spent on transport, medical procedures, and drugs. For each participant, we calculated (1) total diagnostic delay, defined as the number of weeks between symptom onset and diagnosis, (2) delay before contact with the health system, defined as the weeks between symptom onset and the first visit to a health facility, and (3) delay after contact with the health system, defined as the weeks between the first visit to a health facility and diagnosis. We also calculated total out-of-pocket expenditures, using an exchange rate of 3.7 Peruvian soles to 1 USD. We used a Wilcoxon rank-sum test to assess differences in delay and expenditures between periods, sexes, and age groups (18-34 years old versus ≥35 years old). Analysis was performed in SAS v9.4 (SAS Institute, Cary, NC).
Qualitative data collection and analysis
We recruited participants for in-depth interviews based on the delays that they reported during the surveys, balancing the sample in terms of long versus short delays before contact with the health system, and long versus short delays after first contact with the health system. The study team member who conducted the survey recruited participants for one-time interview lasting approximately 1 h. One or two study team members conducted each interview in Spanish in the health center or the patient’s home during April–May 2021 and November 2021–January 2022. All interviewers (DA, EA, HC, SF, GM, JR, IT) were female Peruvian nurse technicians who were trained in interviewing and had no prior relationship with the participants. We interviewed 16 participants from period 1 (participants #1-16) and 10 from period 2 (participants #17-26). Twelve individuals declined interviews upon recruitment; none withdrew during the interview.
The interview guide specified probes based on the information reported during the survey, asking participants about initial symptoms, what prompted them to seek care in each instance, and what happened during each health facility visit. After recounting this pathway to diagnosis in detail, participants were asked either what factors they think led to prompt care-seeking or diagnosis or what factors they think delayed their care-seeking or diagnosis. They were also asked what interventions they believed could encourage prompt diagnosis for others in the future. We did not specifically ask about COVID-19. We planned to interview 30 patients but stopped after 26 because no new themes were emerging relating to factors that facilitated or delayed diagnosis, which we considered a sign of data saturation. Interviews were audio recorded, transcribed, and checked by the interviewer for fidelity with the help of field notes; transcripts were not returned to participants.
We used an inductive content analysis approach [
17] to understand the impact of the COVID-19 pandemic on TB diagnosis. Two authors (AKM, CMY) with experience in qualitative research open-coded content related to COVID-19 or the pandemic in five transcripts (in Spanish), resolved discrepancies by consensus, and applied the resulting codebook to the remaining transcripts. Coding was done manually in Microsoft Word. We analyze the coded data, grouping themes into higher-level categories using an iterative approach. Findings were not shared directly with participants but will be disseminated via public community-oriented presentations. Strategies to ensure analytic rigor included having two coders independently performing the initial open-coding to develop the codebook, discussing interpretations with co-authors who performed the interviews (DA, SF, IT) as well as co-authors who were not part of the interview process (AKM, LL, CMY), and using rich verbatim data from patient interviews to illustrate findings.
Discussion
The COVID-19 pandemic impacted multiple steps in the pathway to care for TB patients in Lima, causing longer delays in TB diagnosis during the first year of the COVID-19 pandemic compared to after. People diagnosed during the first year of the pandemic also spent more money on accessing care, with participants describing out-of-pocket payments for COVID-19 treatments and attention in the private sector. Within the health system, the focus on the pandemic led to potentially incorrect COVID-19 diagnoses and delayed consideration of TB, as well as reduced access to hospitals and pulmonologists. As the health system recovered in the second year of the pandemic delays in TB diagnosis returned to pre-pandemic levels [
18], although confusion of TB and COVID-19 signs and symptoms still complicated the diagnostic process.
Our qualitative findings show how the pandemic’s impact on Peru’s health system forced people affected by TB to suffer through prolonged illness, repeated attempts to access diagnostic services, and misdiagnoses. While we are not aware of other qualitative studies focused on the pandemic’s impact on patients’ experiences during the TB diagnostic process, patients who were already receiving treatment for TB have reported difficulty accessing health services for treatment monitoring [
10]. In addition, a survey of health care workers from 64 countries found that most felt that people with TB and HIV faced greater challenges accessing health services during the pandemic, with reduced mobility and health facility closures frequently stated as reasons [
19]. While the pandemic’s effects on TB-associated mortality have not yet become clear [
20], increased deaths from cardiovascular disease [
21,
22], neonatal deaths [
23], and deaths attributable to delays in accessing medical care during the pandemic [
24] have been reported in various countries. Thus, it is an important lesson for future public health emergencies that changes to health system priorities and procedures should consider who may be harmed as well as who will benefit.
Our finding that the initial phase of the COVID-19 pandemic was associated with longer delays in TB diagnosis is consistent with reports from other countries that showed that patients experienced longer overall delays during the pandemic compared to before [
2‐
5]. In most studies that distinguished between delay before and after contact with the health system, the former was far longer than the latter; in these studies, the average delay after entering the health system was generally under a week, even during the pandemic, and overall delay was driven by the delay in accessing health services [
2,
3,
5]. However, in our study, many patients experienced substantial delays in receiving a TB diagnosis after accessing health services—an average of 4 weeks during the first year of the pandemic and 2 weeks in the period after. Interestingly, a study from Burkina Faso found that the average time to TB diagnosis once people entered the health system decreased substantially during the pandemic, with possible explanations including increased access to molecular diagnostics and increased TB screening as a means of ruling out a COVID-19 diagnosis [
25]. Thus, it is possible for a public health emergency such as the COVID-19 pandemic to have beneficial effects for people with other health conditions if the emergency response ultimately strengthens the health system.
The difficulty of differentiating TB from COVID-19 symptoms and the instances of potentially incorrect COVID-19 diagnoses reported by participants in our study suggests the importance of integrated evaluation in places where both conditions are common. Models for integrated screening have been reported from multiple countries [
26‐
28], and the Peruvian Ministry of Health has recently established an integrated evaluation algorithm for people with respiratory symptoms. Implementing integrated programs requires not only training providers in these algorithms, but also investment in robust testing capacity for both conditions. In particular, access to rapid SARS-CoV-2 testing is necessary so that a negative test can quickly help doctors focus on alternative diagnoses.
The major limitation of our study is that it was not originally designed to assess the impact of the COVID-19 pandemic on TB diagnosis, so this question was not specifically probed during interviews. However, the fact that so many patients described ways in which the pandemic impacted their pathway to diagnosis without being asked underscores the importance of our findings. Another limitation is that quantitative analyses of diagnostic delay are dependent on patient recall of when symptoms started, and recall can be imperfect, particularly when symptoms started a long time before diagnosis. However, the in-depth interviews conducted corroborated the details of the pathway to care for a large subset of patients, suggesting that recall of this major life experience is accurate immediately following diagnosis. Finally, we did not collect data on many patient characteristics (e.g. income, specific symptoms) in the quantitative survey, limiting our ability to assess differences between patient groups.
In conclusion, our findings serve as a warning about unintended negative effects of health system responses to the COVID-19 pandemic on people affected by TB. In addition, they suggest several ways in which services can be improved. The information dissemination methods used to rapidly create high public awareness about COVID-19 symptoms could be used to improve TB awareness and promote care-seeking behavior. In addition, training doctors in both the public and private sectors to use clear diagnostic algorithms for people with respiratory systems can help avoid delay in considering a TB diagnosis. As countries exit the acute phase of the COVID-19 pandemic, it is important to rebuild health systems to restore and improve services for TB and other conditions while maintaining capacity to manage COVID-19.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.