# Preplanned Studies: Estimating the Incidence of Tuberculosis — Shanghai, China, 2025−2050

View author affiliations
• ### Summary

What is already known on this topic?

Despite the impressive achievements in eliminating tuberculosis (TB), the TB burden is still heavy in China. By 2010, China halved the prevalence and mortality reported in 1990, but China is still one of 30 high-TB burden countries in the world.

What is added by this report?

A dynamic transmission model including both rifampin resistant TB (RR-TB) and relapse of pulmonary TB was created. The TB incidence of Shanghai in 2025 and 2035 was predicted, and sensitively analysis of reducing transmission, treating latent TB infection (LTBI), and reducing the recurrence rate was conducted.

What are the implications for public health practice?

Screening for latent TB infections should be carried out regularly in high-risk groups and areas using tuberculin skin testing and/or interferon gamma release assays.

•  [1] World Health Organization. The end TB strategy. Geneva: WHO; 2014. https://www.who.int/tb/strategy/en/. [2] Wang LX, Zhang H, Ruan YZ, Chin DP, Xia YY, Cheng SM, et al. Tuberculosis prevalence in China, 1990-2010; a longitudinal analysis of national survey data. Lancet 2014;383(9934):2057 − 64. https://www.sciencedirect.com/science/article/abs/pii/S0140673613626392. [3] Li XW, Yang Y, Liu JM, Zhou F, Cui W, Guan L, et al. Treatment outcomes of pulmonary tuberculosis in the past decade in the mainland of China: a meta-analysis. Front Med 2013;7(3):354 − 66. http://dx.doi.org/10.1007/s11684-013-0257-3CrossRef [4] World Health Organization. Global tuberculosis report 2020. Geneva, Switzerland: WHO; 2020. https://apps.who.int/iris/handle/10665/336069. [5] Houben RMGJ, Dodd PJ. The global burden of latent tuberculosis infection: a re-estimation using mathematical modelling. PLoS Med 2016;13(10):e1002152. http://dx.doi.org/10.1371/journal.pmed.1002152CrossRef [6] Shanghai Municipal Health Commission. Epidemic situation of notifiable infectious diseases in Shanghai in 2019. Shanghai; 2020. http://wsjkw.sh.gov.cn/yqxx/20200703/ce47739b555e4b70bea9f704678524ee.html. (In Chinese). [7] Cui XJ, Gao L, Cao B. Management of latent tuberculosis infection in China: Exploring solutions suitable for high-burden countries. Int J Infect Dis 2020;92 Suppl:S37 − 40. http://dx.doi.org/10.1016/j.ijid.2020.02.034CrossRef [8] Gao L, Zhou F, Li XW, Jin Q. HIV/TB co-infection in mainland China: a meta-analysis. PLoS One 2010;5(5):e10736. http://dx.doi.org/10.1371/journal.pone.0010736CrossRef [9] Tobe RG, Xu LZ, Song PP, Huang Y. The rural-to-urban migrant population in China: gloomy prospects for tuberculosis control. Biosci Trends 2011;5(6):226 − 30. http://dx.doi.org/10.5582/bst.2011.v5.6.226CrossRef [10] Cao SY, Gan Y, Wang C, Bachmann M, Wei SB, Gong J, et al. Post-lockdown SARS-CoV-2 nucleic acid screening in nearly ten million residents of Wuhan, China. Nat Commun 2020;11(1):5917. http://dx.doi.org/10.1038/s41467-020-19802-wCrossRef
• FIGURE 1.  The flow diagram of tuberculosis model considering rifampicin resistance and recurrence.

S referred to people who are not infected with Mycobacterium tuberculosis (M. tb); E referred to people infected with M. tb but were not yet infectious; I referred to patients with infectious pulmonary tuberculosis (TB); IU1 referred to the undetected and drug sensitive population; IU2 referred to the undetected and rifampicin resistant population; IF1 referred to the detected and drug sensitive population; IF2 refers to the detected and rifampicin resistant population; and R referred to TB patients who have been successfully treated. β1 and β2 are the transmission rates of infectious drug-sensitive TB cases and rifampin resistant TB (RR-TB) cases. κ is the progressive rate from the exposed to the infectious; ρ is the progressive rate from drug-sensitive TB to RR-TB; σ is the detection rate of the infectious; γ1 and γ2 are the successful treatment rates of detected patients with infectious drug-sensitive TB and RR-TB, respectively; ω is the disease recurrence rate from the recovered population; τ1 and τ2 are the drug resistance rates of new patients and recurrent patients, respectively; and μ0 is the natural mortality rate, while μ1 and μ2 are the fatality rates of TB in infectious drug-sensitive TB cases and RR-TB cases, respectively.

FIGURE 2.  Predictions of the incidence of pulmonary TB in Shanghai with different parameters. (A) Reducing the values of parameters β1, β2 and ω. (B) Reducing the values of parameter κ. (C) Reducing the values of parameter ω.

TABLE 1.  Definitions and estimated values of parameters.

 Parameter Definition Estimated value Λ Constant recruitment of the population 486,245 β1 Transmission rate of infectious drug-sensitive TB cases 8.69×10−12 β2 Transmission rate of infectious RR-TB cases 2.52×10−10 k Progressive rate from the exposed to the infectious 1.65×10−3 τ1 Drug resistance rate of new patients 2.39×10−2 σ Detection rate of the infectious 6.20×10−1 ρ Progressive rate from drug-sensitive TB to RR-TB 8.85×10−2 γ1 Treatment successful rates of infectious drug-sensitive cases 9.14×10−1 γ2 Treatment successful rates of infectious RR-TB cases 8.99×10−1 ω Recurrence rate from recovered 5.94×10−3 τ2 Drug resistance rate of recurrent patients 1.25×10−1 μ0 Natural mortality rate 7.46×10−3 μ1 Fatality rates of TB in infectious drug-sensitive TB cases 9.32×10−3 μ2 Fatality rates of TB in infectious RR-TB cases 2.26×10−2 Abbreviations: TB=tuberculosis; RR-TB=rifampin resistant TB.

 Citation:

###### 通讯作者: 陈斌, bchen63@163.com
• 1.

沈阳化工大学材料科学与工程学院 沈阳 110142

Turn off MathJax
Article Contents

## Estimating the Incidence of Tuberculosis — Shanghai, China, 2025−2050

View author affiliations

### Summary

What is already known on this topic?

Despite the impressive achievements in eliminating tuberculosis (TB), the TB burden is still heavy in China. By 2010, China halved the prevalence and mortality reported in 1990, but China is still one of 30 high-TB burden countries in the world.

What is added by this report?

A dynamic transmission model including both rifampin resistant TB (RR-TB) and relapse of pulmonary TB was created. The TB incidence of Shanghai in 2025 and 2035 was predicted, and sensitively analysis of reducing transmission, treating latent TB infection (LTBI), and reducing the recurrence rate was conducted.

What are the implications for public health practice?

Screening for latent TB infections should be carried out regularly in high-risk groups and areas using tuberculin skin testing and/or interferon gamma release assays.

• 1. School of Public Health, Peking University, Beijing, China
• 2. Chinese Center for Disease Control and Prevention, Beijing, China
• 3. Center for Intelligent Public Health, Institute for Artificial Intelligence, Peking University, Beijing, China
• 4. Center for Drug Abuse Control and Prevention, National Institute of Health Data Science, Peking University, Beijing, China
###### doi: 10.46234/ccdcw2020.266
• Tuberculosis (TB) is a global public health problem. The World Health Organization (WHO) proposed that the incidence rate of TB should be reduced to less than 55/100,000 population by 2025, less than 10/100,000 by 2035, and to eliminate TB by 2050 (incidence rate <1/100,000) (1). Based on the directly observed treatment, short-course (DOTS) strategy nationwide, China halved the prevalence and mortality of TB in 2010 as compared to 1990 (2), and the cure rate of TB has been reached 92.9% in 2013 (3). The TB incidence rate fell 43.1% from 1990 (130/100,000) to 2010 (74/100,000) (4). Despite impressive achievements, China is still one of the 30 high-TB burden countries in the world. In 2019, there were about 833,000 new TB cases in China with a TB incidence rate of 58/100,000 (4). China also has the highest latent TB infection (LTBI) burden globally with approximately 350 million infections that are at risk for active TB disease (5).

Shanghai is one of the areas with the best implementation of TB control measures in China, but the incidence rate was still above 25/100,000 in 2019 (6). Shanghai failing to reach the target by 2035 would indicate a high likelihood of failure for other areas in China. We established a dynamic TB model to estimate the predicted incidence in Shanghai and the impact of different prevention and control measures.

According to the natural progressive history of pulmonary TB, the overall population was divided into 7 classes: S referred to people who are not infected with Mycobacterium tuberculosis (M. tb); E referred to people infected with M. tb but were not yet infectious; I referred to patients with infectious pulmonary TB; IU1 referred to the undetected and drug sensitive population; IU2 referred to the undetected and rifampicin resistant population; IF1 referred to the detected and drug sensitive population; IF2 refers to the detected and rifampicin resistant population; and R referred to TB patients who have been successfully treated. In this model, we made the following assumptions: 1) the population was evenly mixed, and contact between all individuals was equally likely; 2) patients were likely to infect susceptible population and the recovered after contact with them; 3) all detected pulmonary TB cases were reported to National Notifiable Disease Reporting System; and 4) the total population of the system was relatively stable. The population supplement was due to births in each year, and population loss was due to natural deaths from each group and deaths due to pulmonary TB from patients.

The equations of the model are as follows:

 $${\left\{ {\begin{array}{*{20}{l}} {\dfrac{{{\rm{dS}}}}{{{\rm{dt}}}}{\rm{ = \Lambda - }}{{\rm{\beta }}_{\rm{1}}}{\rm{(}}{{\rm{I}}_{{\rm{U1}}}}{\rm{ + }}{{\rm{I}}_{{\rm{F1}}}}{\rm{)S - }}{{\rm{\beta }}_{\rm{2}}}{\rm{(}}{{\rm{I}}_{{\rm{U2}}}}{\rm{ + }}{{\rm{I}}_{{\rm{F2}}}}{\rm{)S - }}{{\rm{\mu }}_{\rm{0}}}{\rm{S}}}\\ {\dfrac{{{\rm{dE}}}}{{{\rm{dt}}}}{\rm{ = }}{{\rm{\beta }}_{\rm{1}}}{\rm{(}}{{\rm{I}}_{{\rm{U1}}}}{\rm{ + }}{{\rm{I}}_{{\rm{F1}}}}{\rm{)S + }}{{\rm{\beta }}_{\rm{2}}}{\rm{(}}{{\rm{I}}_{{\rm{U2}}}}{\rm{ + }}{{\rm{I}}_{{\rm{F2}}}}{\rm{)S - \kappa E - }}{{\rm{\mu }}_{\rm{0}}}{\rm{E}}}\\ {\dfrac{{{\rm{d}}{{\rm{I}}_{{\rm{U1}}}}}}{{{\rm{dt}}}}{\rm{ = \kappa (1 - }}{{\rm{\tau }}_{\rm{1}}}{\rm{)(1 - \sigma )E + \omega (1 - }}{{\rm{\tau }}_{\rm{2}}}{\rm{)(1 - \sigma )R - \rho }}{{\rm{I}}_{{\rm{U1}}}}{\rm{ - \sigma }}{{\rm{I}}_{{\rm{U1}}}}{\rm{ - }}\left( {{{\rm{\mu }}_{\rm{0}}}{\rm{ + }}{{\rm{\mu }}_{\rm{1}}}} \right){{\rm{I}}_{{\rm{U1}}}}}\!\!\!\!\!\!\\ {\dfrac{{{\rm{d}}{{\rm{I}}_{{\rm{U2}}}}}}{{{\rm{dt}}}}{\rm{ = \kappa }}{{\rm{\tau }}_{\rm{1}}}{\rm{(1 - \sigma )E + \omega }}{{\rm{\tau }}_{\rm{2}}}{\rm{(1 - \sigma )R + \rho }}{{\rm{I}}_{{\rm{U1}}}}{\rm{ - \sigma }}{{\rm{I}}_{{\rm{U2}}}}{\rm{ - }}\left( {{{\rm{\mu }}_{\rm{0}}}{\rm{ + }}{{\rm{\mu }}_{\rm{2}}}} \right){{\rm{I}}_{{\rm{U2}}}}}\\ {\dfrac{{{\rm{d}}{{\rm{I}}_{{\rm{F1}}}}}}{{{\rm{dt}}}}{\rm{ = \kappa (1 - }}{{\rm{\tau }}_{\rm{1}}}{\rm{)\sigma E + \omega (1 - }}{{\rm{\tau }}_{\rm{2}}}{\rm{)\sigma R + \sigma }}{{\rm{I}}_{{\rm{U1}}}}{\rm{ - \rho }}{{\rm{I}}_{{\rm{F1}}}}{\rm{ - }}{{\rm{\gamma }}_{\rm{1}}}{{\rm{I}}_{{\rm{F1}}}}{\rm{ - }}\left( {{{\rm{\mu }}_{\rm{0}}}{\rm{ + }}{{\rm{\mu }}_{\rm{1}}}} \right){{\rm{I}}_{{\rm{F1}}}}}\\ {\dfrac{{{\rm{d}}{{\rm{I}}_{{\rm{F2}}}}}}{{{\rm{dt}}}}{\rm{ = \kappa }}{{\rm{\tau }}_{\rm{1}}}{\rm{\sigma E + \omega }}{{\rm{\tau }}_{\rm{2}}}{\rm{\sigma R + \sigma }}{{\rm{I}}_{{\rm{U2}}}}{\rm{ + \rho }}{{\rm{I}}_{{\rm{F1}}}}{\rm{ - }}{{\rm{\gamma }}_{\rm{2}}}{{\rm{I}}_{{\rm{F2}}}}{\rm{ - }}\left( {{{\rm{\mu }}_{\rm{0}}}{\rm{ + }}{{\rm{\mu }}_{\rm{2}}}} \right){{\rm{I}}_{{\rm{F2}}}}}\\ {\dfrac{{{\rm{dR}}}}{{{\rm{dt}}}}{\rm{ = }}{{\rm{\gamma }}_{\rm{1}}}{{\rm{I}}_{{\rm{F1}}}}{\rm{ + }}{{\rm{\gamma }}_{\rm{2}}}{{\rm{I}}_{{\rm{F2}}}}{\rm{ - \omega R - }}{{\rm{\mu }}_{\rm{0}}}{\rm{R}}} \end{array}} \right.}$$

The model involves 7 classes and 14 parameters. Each equation represents the change rate of the number of people in each class in unit time, and the right side includes the moving in and out of items that lead to the change of class population. The unit time of this model is one year.

Λ is the constant recruitment in the system. β1 and β2 are the transmission rates of infectious drug-sensitive TB cases and rifampin resistant TB cases (RR-TB). κ is the progressive rate from the exposed to the infectious; ρ is the progressive rate from drug-sensitive TB to RR-TB; σ is the detection rate of the infectious; γ1 and γ2 are the successful treatment rates of detected patients with infectious drug-sensitive TB and RR-TB, respectively; ω is the disease recurrence rate from the recovered population; τ1 and τ2 are the drug resistance rates of new patients and recurrent patients, respectively; and μ0 is the natural mortality rate, while μ1 and μ2 are the fatality rates of TB in infectious drug-sensitive TB cases and RR-TB cases, respectively. The transmission diagram is shown in Figure 1.

We collected the reported incidence of pulmonary TB in Shanghai from 2004 to 2017 provided by the Public Health Science Data Center, of which 2004−2012 were used as training data and 2013−2017 years were used as test data. The values of the parameters were determined by the reports of earlier studies and adjusted according to TB data, then the incidence of pulmonary TB in Shanghai was estimated for the near future. The incidence of pulmonary TB was numerically defined as the number of new reported cases of pulmonary TB within each year as a proportion of the number of average annual population.

The parameters in the model were adjusted to simulate the effect of three different TB prevention and control strategies. We reduced the values of parameters β1, β2, and ω to simulate reducing the probability of infection or reinfection of susceptible and recovered patients (assuming 60% of recurrent patients are due to reinfection) to simulate strengthening personal protection and isolation of active cases during contagious period. The effect of preventive treatment on LTBI cases was evaluated by reducing the rate of progression (κ) of the exposed group to the infectious groups. We reduced the recurrence rate (ω) of the recovered group to study the impact of recurrence rate on the TB epidemic.

We set the initial values of the model classes as S (0) = 14,453,131, E (0) = 3,834,319, IU1 (0) = 6,462, IU2 (0) = 333, IF1 (0) = 7,011, IF2 (0) = 361, R (0) = 48,194, and the values of parameters are shown in Table 1. The first curve of each panel in Figure 2 shows our prediction of the incidence of pulmonary TB in Shanghai under current strategies. We predicted that the estimated incidence of pulmonary TB in Shanghai will continue to decline from 2004 to 2050. In 2025, the incidence of TB in Shanghai was estimated to be 24.27/100,000, which will achieve the WHO’s goal in 2025 (<55/100,000). However, the incidence was estimated to be 20.81/100,000 in 2035, still far from the goal set for 2035 (<10/100,000).

Figure 2 shows the impact of 3 different prevention and control strategies on pulmonary TB in Shanghai. The incidence will decrease slightly with the values of parameters β1, β2 and ω reduced (Figure 2A). The incidence of pulmonary TB in Shanghai in 2035 will be 19.69/100,000 when the parameters dropped by 50%. Reducing the progressing rate (κ) of the exposed group to the infectious groups, the incidence of pulmonary TB in Shanghai will decrease significantly (Figure 2B). In 2035, the incidence will be 11.55/100,000 with the parameters κ dropped by 50%. The incidence of pulmonary TB in Shanghai will be slightly decreased by reducing their recurrence rate (ω) (Figure 2C). With the recurrence rate reduced by 50%, the incidence of pulmonary TB in Shanghai will be 19.08/100,000 in 2035.

Reference (10)

 Citation:

/