Impact of the COVID-19 pandemic on tuberculosis control in Indonesia: a nationwide analysis of programme data and health system vulnerabilities

Background

Coronavirus disease (COVID-19) caused by the SARS-CoV-2 virus, has resulted in over 6.8 million reported deaths as per February 9, 2023 ¹, and the actual excess COVID-19 death toll may be triple that amount.² The COVID-19 pandemic severely impacted almost every aspect of global health, including the delivery of essential services for other major health conditions such as tuberculosis (TB), the leading cause of death from a single infectious agent other than SARS-CoV-2.³ Mitigating the transmission, morbidity, and mortality of TB involves patient-centred approaches to accessible and high-quality TB diagnostic and treatment services.^4,5 According to WHO estimates, TB case notifications declined from 7.1 million in 2019 to 5.8 million (18% decline) in 2020 ⁶, with pandemic-related disruptions in health services being especially exacerbated in low-and middle-income countries (LMIC) in sub-Saharan Africa ^7–11 and Asia ^12,13, because of the vulnerabilities in their health systems.¹⁴ Only three countries accounted for 67% of the global reduction in case reporting: India, Indonesia and the Philippines, each experiencing major COVID-19 epidemics through 2020. Global TB mortality increased for the first time in more than a decade; from an estimated 1.4 million deaths in 2019, 1.5 million in 2020, to 1.6 million in 2021.¹⁵ Reduced case-finding and hence treatment coverage during the past two years likely increased transmission rates and WHO modelling forecasted worsening TB incidence and mortality in the coming few years.⁶

Indonesia has the second-highest TB burden in the world ⁶, and reported the highest number of confirmed COVID-19 cases and deaths in the Southeast Asia region.¹ A recent nationwide analysis of patient-level data from the National TB Control Programme (NTCP) in Indonesia reported that during the years immediately preceding the COVID-19 pandemic (2017-2019) notified cases of drug-susceptible TB increased from 429 219 to 523 614 (from 167 to 196 cases per 100 000), with stable treatment success rates and mortality.¹⁶ Despite significant progress in rolling out national health insurance towards achieving universal health coverage ^17,18, the quality of health care remains highly variable across its decentralised public health system.¹⁸ The 2018 Public Health Development Index (PHDI), a composite indicator constructed by the Ministry of Health expressing coverage and equity in health services, and health status ¹⁹, ranged from 35% in Paniai district, Papua province to 75% in Gianyar district, Bali province. No studies have yet examined the impact of the COVID-19 pandemic on TB services and outcomes in Indonesia at the subnational level in association with key health system indicators.

Reports to date, including both the WHO Global TB Report ¹⁵ and the Global Burden of Disease Study ²⁰, applied extrapolated and modelled TB burden estimates at national levels. However, rigorous large-scale, sub-national analyses of patient databases in high-burden TB countries are required to more fully understand the impacts of the COVID-19 pandemic on TB control, services and disease outcomes in the context of metrics of health system resilience.¹² In the present study, we sought to conduct a comprehensive nationwide analysis of the intertwined COVID-19 and TB epidemics in Indonesia. We specifically sought to understand the impacts on the NTCP, by examining TB case notification, treatment and mortality rates before (2016-2019) and during the pandemic (2020-2021), in association with metrics of healthcare system resilience at the district level, across all 514 districts and 34 provinces of Indonesia.

Methods

Results

Table 1 summarises the general profiles of the 514 districts analysed in this study. The cumulative burden of COVID-19 incidence and mortality rate as per 31 December 2021 was 139 (IQR: 59-280) and 4 (1-9) per 100 000 population, respectively. Health facilities with a GeneXpert machine was median 0.5 (IQR: 0.2-0.9) per 100 000 population, health facility with TB microscopy smear service was median 4 (IQR: 3-6) per 100 000 population, and the number of doctors was median 8 (IQR: 5-18) per 100 000 population. The per capita domestic expenditure was median USD 675 (IQR: 574-777).

Tuberculosis case notification rate and mortality rate before (2016-2019) and during the COVID-19 pandemic (2020-2021)

Table 2 summarises the number of reported TB cases and deaths, and changes in TB case notification rate and mortality rate from 2016 to 2021. The median case notification per quarter gradually increased from quarter 1 of 2016 through quarter 1 of 2020, sharply decreased in quarter 2 of 2020 (after the start of COVID-19 pandemic in Indonesia in early March 2020), and then increased from quarter 1 through 4 in 2021 (Supplementary Figure 2A). The median case notification rate in 2019 (before the COVID-19 pandemic), was 172 (1QR 129-244) per 100 000 population. The median case notification rate during the COVID-19 pandemic was 119 (IQR 87-170) in 2020 and 142 (IQR 99-204) in 2021. The total number of new TB cases decreased from 565 669 in 2019, to 393 323 (−31%) in 2020, and increased again to 473 006 (+20%) in 2021.

View this table:

• View inline
• View popup
• Download powerpoint

Table 2. Changes in tuberculosis case notification rate and mortality over time

The decline in case notification rate was highly heterogeneous across districts (Figure 1 and Supplementary Table 1). The top five districts experiencing the largest case notification rate decline from 2019 to 2020 were Pasaman Barat in West Sumatra province (case notification rate ratio 0·15, 95% CI: 0·13-0·18), Kapuas Hulu in West Kalimantan province (0·16, 95% CI: 0·13-0·20), Yalimo in Papua province (0·19, 95% CI: 0·09-0·44), Sorong Selatan in West Papua province (0·21, 95% CI: 0·29-0·47), and Solok in Maluku province (0·22, 95% CI: 0·48-0·65) (Supplementary Table 1). Compared with the first pandemic year 2020, we observed an improvement in the second pandemic year 2021, i.e., fewer districts maintained a significant decline in TB case notification rate (80% in 2020 to 66% in 2021) (Figure 3A).

• Download figure
• Open in new tab

Figure 1: Maps showing case notification rate ratio in 2020 and 2021, compared to 2019 by district.

A smaller CNR ratio (i.e., closer to 0) (red) indicates that fewer TB cases have been notified during the pandemic compared to before the pandemic, i.e., greatest negative impact of the COVID-19 pandemic on the national TB programme, whereas a greater CNR ratio (i.e., greater than 1) (blue) indicates that more TB cases have been notified during the pandemic compared to before the pandemic, i.e., smallest negative impact of the COVID-19 pandemic on the national TB programme.

The total number of reported deaths increased from 13,059 in 2019, to 14,148 (8%) in 2020, and decreased to 12,016 (−15%) in 2021. The median number of deaths per quarter sharply increased in quarter 1 of 2020 (start of COVID-19 pandemic), then gradually decreased from quarter 2 of 2020 through quarter 4 of 2021 (Supplementary Figure 2B). The median annual mortality rate sharply increased during the first COVID-19 pandemic year 2020. The median mortality rate was 4·2 (IQR 2·0-7·4) per 100 000 population in 2019, 5·0 (IQR 3·1-7·5) in 2020 and 4·1 (IQR 2·5-6·8) in 2021 (Table 2). Similar to the case notification rate, the significant increase in mortality rate was highly heterogeneous across districts (Figure 2 and Supplementary Table 2). The top five districts experiencing the largest increase in mortality rate from 2019 to 2020 were Barru in South Sulawesi province (mortality rate ratio 16·9, 95% CI: 3·9-73·9), Lebong in Bengkulu province (16·5, 95% CI: 3·7-72·9), Langkat in North Sumatra province (15·9, 95% CI: 6·7-37·7), Kota Manado (15·4, 95% CI: 3·4-69·2), and Kepulauan Talaud in North Sulawesi province (11·7, 95% CI: 2·4-58·3) (Supplementary Table 2). Compared with the first pandemic year 2020, we observed an improvement in the second pandemic year 2021, i.e., fewer districts maintained a significant increase in mortality (20% in 2020 to 13% in 2021) (Figure 3B).

• Download figure
• Open in new tab

Figure 2: Maps showing mortality rate ratio in 2020 and 2021, compared to 2019 by district.

A greater mortality rate ratio (i.e., greater than 1) (red) indicates more reported deaths during the pandemic compared to before the pandemic, i.e., greatest negative impact of the COVID-19 pandemic on the national TB programme, whereas a smaller mortality ratio (i.e., closer to 0) (blue) indicates fewer deaths during the pandemic compared to before the pandemic, i.e., smallest negative impact of the COVID-19 pandemic on the national TB programme.

• Download figure
• Open in new tab

Figure 3:

Proportion of districts within their province impacted based on decrease of case notification rate (A) and increase of mortality rate (B) during the COVID-19 pandemic years 2020 and 2021, compared to the pre-pandemic year 2019.

Districts were categorised as impacted if they have a statistically significant decrease in case notification rate and significant increase in mortality rate in 2020 and 2021, compared to 2019, respectively.

Treatment coverage of TB cases before (2019) and during the COVID-19 pandemic (2020-2021)

Due to incomplete data, data from only 503 of 514 districts in 2019-2021 were included in this analysis. The overall proportion of TB cases who started treatment decreased from 98% in 2019 to 91% in 2020 (−7%), and 84% (−7%) in 2021 (Table 3). The median proportion of TB cases who started treatment significantly decreased from 100% in 2019 to 95% in 2020, and 88% in 2021. The top five districts experiencing the largest decrease in treatment coverage from 2019 to 2020 were Keerom in Papua province (treatment coverage ratio 0·3, 95% CI: 0·2-0·4), Bombana in Southeast Sulawesi province (0·4, 95% CI: 0·3-0·4), Padang Pariaman in West Sumatra province (0·5, 95% CI: 0·4-0·5), Seluma in Bengkulu province (0·5, 95% CI: 0·3-0·7), and Nias in North Sumatra province (0·5, 95% CI: 0·4-0·6) (Supplementary Table 3). Compared with the first pandemic year 2020, we observed an improvement in the second pandemic year 2021, i.e., more districts had a significant increase in TB treatment coverage (24% in 2020 to 48% in 2021) (Figure 4).

View this table:

• View inline
• View popup
• Download powerpoint

Table 3. Tuberculosis treatment coverage in 503 districts in 2019-2021

• Download figure
• Open in new tab

Figure 4:

Proportion of districts within their province impacted based on decrease in treatment coverage during the COVID-19 pandemic years 2020 and 2021, compared to the pre-pandemic year 2019.

Districts were categorised as impacted if they have a statistically significant decrease in treatment coverage in 2020 and 2021, compared to 2019.

Discussion

Our analysis identified a gradual increase in TB case notifications from 2016 through early 2020, corroborating a recently published analysis covering 2017 and 2019.¹⁶ After the onset of the COVID-19 pandemic in Indonesia in early March 2020, the TB case notification rate decreased by 31% for 2020, signifying one of the most dramatic drops among the high-burden TB countries reported by WHO.⁶ Likewise, there was a significant decrease in TB treatment coverage (7%) and an increase in reported deaths for any reason (8%). Although we did control for COVID-19 mortality in our regression models, the data did not allow us to ascertain cause of death, and estimate mortality attributable to TB, COVID-19, or to any interruptions in vital health services for other health conditions. Nonetheless, these findings affirm substantial impacts of the COVID-19 pandemic on national TB programmes, as reported for several of the TB high-burden, resource-limited countries in sub-Saharan Africa ^7–11 and Asia.^12,13 The findings further suggest that the second pandemic year saw signs of recovery in the public health system in many, but not all, districts, with attenuated impacts on TB notifications (80% of districts impacted in 2020 vs 66% in 2021) and/or reported deaths (20% of districts impacted in 2020 vs 13% in 2021), whereas there was continued decline in TB treatment coverage overall (24% of districts impacted in 2020 vs 48% in 2021). Extended reductions in treatment coverage may seriously escalate TB transmission in the coming years.⁶

Even without pandemic disruptions of TB case finding, diagnostic and treatment services, patient pathways to reach a TB diagnosis and access effective treatment are already complex, often involving long delays and visits to multiple health care providers.^5,28,29 The social and economic shocks of the pandemic deeply exacerbated the fraught path to recovery, correlating with the vulnerability of vital health services delivery. Indeed, we observed substantial heterogeneity of the COVID-19 impacts between districts regarding all three TB study outcomes, and several local health system factors were identified that could offer, at least partially, explanations for the substantial differences observed between districts.

The study findings suggested that districts with a higher COVID-19 incidence and lower resources in the health system were experiencing the most detrimental impacts on TB notification, treatment coverage, and reported death. Specifically, this was true for districts with fewer per capita health facilities with TB GeneXpert diagnostic services, doctors and primary health care centres and per capita domestic expenditure. In Indonesia’s decentralised public health system, resource allocations are mostly under the responsibility of local governments at district level, and as such are highly dependent on local government resources, policy strategies and priorities. Considerable disparities in available financial resources between districts, with per capita domestic expenditure ranging from USD 290-1572, may have contributed to the heterogeneous impact on the overall performance of the TB programme. The pandemic forced local governments to reallocate already scarce health resources to the COVID-19 response ³⁰, which, nationally, included an approximately 30% decrease in total TB financing in 2020 compared to 2019. Moreover, a recent review of the Indonesian healthcare system suggested that the existing medical work force was largely insufficient and not evenly distributed to deal with the COVID-19 pandemic.³¹ Therefore, our data highlight the need for both national and sub-national governments to make targeted investments in vital health infrastructure in districts who have shown to be most vulnerable to shocks like COVID-19.

Several additional factors, which our analysis was not able to capture, may have further exacerbated the pandemic impacts on TB case finding and diagnostic and treatment services. First, high rates of infection ³² as well as deaths ³³ among frontline health care workers resulted in substantial absenteeism and task-shifting, especially during the early pandemic. Notably, a recent report estimated that Indonesian healthcare workers had a five times higher chance of death from COVID-19 than the general population.^33,34 Second, the pandemic affected patient health-seeking behaviour as well as access to essential health services, because of fear of contracting COVID-19 and extra costs for personal protective equipment ³⁵, and government-mandated local lockdowns.³⁶ Third, emerging data suggest that biological interactions between both pathogens can result in shared dysregulation of immune responses and a dual risk for COVID-19 severity and TB disease progression or poor outcomes.^37,38 The interplay between these highly prevalent infections warrants further research.

This study had several limitations. This study was based on routine district health office reports, with the potential for gaps in quality control as well as underreporting, which were likely exacerbated by the COVID-19 pandemic. A previous inventory study in 2016 estimated that only half of incident TB cases were detected and reported to the NTCP, with underreporting estimated at 31% for government facilities and 75% for private facilities.³⁹ Private health facilities are key entry points for a large proportion of patients seeking health care in Indonesia, including for TB.^5,40 Additionally, the SITB registers drug-susceptible TB cases, and we could not assess the burden of multidrug/rifampicin-resistant (MDR/RR) cases. Although Indonesia is listed as one of the 30 high-burden countries for drug-resistant TB, incidence estimates at the district-level are lacking and may be underreported, due to limited diagnostic capacities in many remote, under-resourced districts. We were also not able include recent information on district-level HIV and diabetes mellitus prevalence.²⁰

In conclusion, Indonesia and other high-burden TB countries, will need to intensify efforts to identify the missed cases with undiagnosed TB, close to the most affected communities, in order to prevent more excess TB reported deaths. Our analysis affirms the notion that people whose access to quality health care services was already limited prior to the pandemic experienced the most acute and severe hardships during the pandemic. Poor resilience of the public healthcare system will, directly and indirectly, exacerbate the shocks and impacts of events like the COVID-19 pandemic. Preparing for the inevitable next pandemic requires fortifying those systems, while prioritising those locations that are the most vulnerable.

Contributors

RLH and IRFE were the principal investigators of this study. HS designed the analysis, did the analysis with assistance from EP, and had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. HS, EP, DD, TTP, EL, SS, and SMD contributed to data collection and verification. IRFE and RLH contributed to funding acquisition. HS, IRFE and RLH drafted the paper. All authors critically revised the manuscript for important intellectual content and all authors gave final approval for the version to be published.

Conflict of interests

We declare no competing interests.

Data sharing statement

After publication, the datasets used for this study will be made available to others on reasonable requests to the corresponding author and Sub-Directorate of Tuberculosis Ministry of Health of Indonesia, including a detailed research proposal, study objectives and statistical analysis plan.

Funding

This study was funded by the Wellcome Africa Asia Programme Vietnam (106680/Z/14/Z).