Healthy China: Effects of Three Major Immunization Interventions on Measles Control — China, 1952–2021

Measles, an acute respiratory infectious disease caused by the measles virus, was responsible for millions of deaths annually before the introduction of the measles vaccine (MV) (1). Widespread availability of measles vaccines and the initiation of the Expanded Program on Immunization, approved at the 27th World Health Assembly in 1974 (2), have significantly improved global measles control. In China, measles has been a statutorily notifiable disease since the establishment of the National Notifiable Diseases Reporting System (NNDRS) in 1950 (3). From 1950 to 1986, data were collected by subnational institutions of disease control and aggregated by the national institution using paper documentation. In 1987, an electronic documentation system replaced the paper-based system (3). In 2004, China upgraded the NNDRS to enable direct reporting via the Internet (4).

The measles vaccine was first introduced in China as a liquid formulation in 1965. In 1978, the National Expanded Program on Immunization (EPI) was launched, initiating a one-dose routine measles vaccination schedule (5). Between 2003 and 2009, 27 out of 31 provincial-level administrative divisions (PLADs) in China carried out unsynchronized provincewide supplementary immunization activities to combat measles, targeting approximately 185.7 million children (6). In 2010, China executed synchronized nationwide supplementary immunization activities against measles.

The epidemiology of measles has been described at subnational levels across four stages; however, a quantitative evaluation of the effects of major interventions at the national level has not yet been conducted. Interrupted time series (ITS) analyses are commonly employed to assess the impact of public health interventions, as they offer quantitative comparisons before and after interventions and evaluate both short-term changes and long-term trends. In this study, we utilized ITS analyses to examine the influence of three significant MV interventions at both the national and regional levels in China. These findings will enhance our understanding of the effects of previous immunization strategies and provide valuable evidence for the future development of immunization policies and strategies.

We obtained measles incidence data from the Public Health Science Data Center, along with provincial population data and per capita gross domestic product (GDP) data from the National Bureau of Statistics. We divided the study period from 1952 to 2021 into four stages based on the history of measles control in China:

Stage 1: Pre-vaccine (1952–1964) — Prior to the introduction of the MV, measles was a nearly ubiquitous childhood illness.

Stage 2: Pre-EPI (1965–1977) — Following MV introduction but before the EPI, MV availability was limited, and coverage rates remained low in China.

Stage 3: EPI (1978–2008) — During the EPI period, a standard routine immunization schedule was implemented to vaccinate a defined target population, resulting in gradually increased coverage.

Stage 4: Post-SIAs (2009–2021) — After the implementation of national and subnational Supplementary Immunization Activities (SIAs), MV coverage in children remained high.

The study period involved four stages that included three interventions as the focus of our research: 1) vaccine introduction in 1965, 2) the implementation of the EPI in 1978, and 3) the initiation of SIAs in 2009. The year 2009 was presumed to be when SIAs were conducted nationwide, taking into account the timeline of subnational and national SIAs.

Based on the economic status in China, four economic regions by PLADs were identified: western China (Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, Xinjiang); central China (Shanxi, Anhui, Jiangxi, Henan, Hubei, Hunan); eastern China (Beijing, Tianjin, Hebei, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, Hainan); and northeastern China (Liaoning, Jilin, Heilongjiang). Eastern China exhibits the highest GDP level, followed sequentially by central China, western China, and northeastern China (7).

A single-group ITS analysis was conducted on annual measles incidence data to assess the impact of three interventions on measles control in China from 1952 to 2021. In the ITS model, the outcome variable Y represents the annual incidence of measles, with X₁ representing the year (0, 1, 2, ..., n by year). X₂ signifies the first intervention (0 before 1965 and 1 after 1965), while X₃ denotes a trend variable following the first intervention (0 before 1965, with the value of 0, 1, 2, …, n by year after 1965). The second intervention is illustrated by X₄ (0 before 1978 and 1 after 1978), and X₅ corresponds to a trend variable following the second intervention (0 before 1978, with values of 0, 1, 2, …, n by year after 1978). X₆ represents the third intervention (0 before 2009 and 1 after 2009), and X₇ is a trend variable after the third intervention (0 before 2009, with values of 0, 1, 2, …, n by year after 2009). Lastly, X₈ denotes per capita GDP. The ITS model equation was as follows:

In the given model, β₀ represents the intercept term, signifying the average measles incidence at the beginning of the study. Meanwhile, β₁ indicates the slope or trend of the incidence before the introduction of the measles vaccine, reflecting the average direction of measles incidence over time. β₂ denotes the change in incidence after the measles vaccine was introduced in 1965, and β₃ signifies the slope change following the vaccine’s introduction. Additionally, β₄ represents the change subsequent to the EPI implementation, and β₅ corresponds to the trend after the second intervention, illustrating the long-term impact of EPI implementation. Furthermore, β₆ demonstrates the short-term change resulting from SIAs; while β₇ indicates long-term trends after SIAs. Lastly, β₈ represents the association between per capita GDP and incidence.

The current study employed ITS analyses to assess the impact of three primary interventions on measles incidence in China. Segmented regression within a quasi-Poisson model was utilized, with annual measles cases as the dependent variable. In this model, annual measles incidence functioned as the response variable, accompanied by an offset component representing the total population for the given year. Additionally, GDP per capita was incorporated into the model to estimate its association with alterations in the level of measles incidence.

In this study, Microsoft Excel 2019 (Microsoft Corporation, Redmond, WA, USA) was employed to construct the measles incidence database. Statistical analyses were performed using the Statistical Analysis System (SAS, version 9.4; SAS Institute Inc, Cary, NC, USA).

Figure 1 shows the temporal patterns of measles incidence in China from 1952 to 2021. The annual reported measles incidence significantly decreased from 189.44 cases per 100,000 individuals in 1952, with a maximum of 1,432.41 cases per 100,000 in 1959, to 0.04 cases per 100,000 in 2021. From 1952 to 1978, incidence remained high, ranging from 157.51 to 1,432.41 cases per 100,000, and exhibited periodic peaks every 3–5 years. A decline in incidence became evident after 1978, accompanied by the disappearance of periodic peaks. Incidence consistently remained low after 1987, falling below 10 cases per 100,000 in most years, and reached a nadir in 2012 (0.45 cases per 100,000). In 2015, incidence increased to 3.09 cases per 100,000, but subsequently declined to a historical low in 2021 (0.04 cases per 100,000). The annual measles incidence in China decreased by more than 99.99% from its peak in 1959 to 2021.

Figure 1. 

Annual reported measles incidence in China, 1952 to 2021.

Abbreviation: EPI=expanded program on immunization; SIAs=supplementary immunization activities.

Between 1952 and 2021, a total of 111,925,154 measles cases were reported, averaging 1,599,316 cases annually and a cumulative average incidence of 153.45 per 100,000 population. Approximately 90% of these cases occurred during the pre-vaccine and pre-EPI stages. The average incidence rates for the pre-vaccine, pre-EPI, EPI, and post-SIAs stages were 615.06, 450.91, 31.15, and 1.62 per 100,000, respectively. This data demonstrate a declining trend in measles incidence from the pre-vaccine stage to the post-SIAs stage (Table 1).

According to analyses examining the spatial distribution of measles incidence across different stages and throughout the entire study period (1952–2021), high incidences were observed in eastern, central, and western China. The incidence was highest in western China (252.58/100,000) and lowest in eastern China (4.61/100,000). In the pre-vaccine era, eastern China displayed high incidence rates, peaking at 1,599.07/100,000. During the pre-EPI stage, most high-incidence PLADs were located in central China, but the highest incidence (848.96/100,000) was observed in western China. High incidence rates were reported in both western and central China during the EPI stage, with the highest rate of 200.74/100,000 occurring in western China. Finally, in the post-SIAs stage, most high incidences were observed in western China, peaking at 9.17/100,000.

In the pre-vaccine stage, measles incidence increased by 5.9% per year [incidence rate ratio (IRR)=1.059, P=0.016], as demonstrated in our Poisson model (Table 2). Following the introduction of the measles vaccine during the pre-EPI stage, incidence decreased by 13.9% (IRR=0.861, P<0.001) each year compared to the pre-vaccine stage. This resulted in a short-term decrease in measles incidence by 16.1% (IRR=0.839, P=0.432).

The introduction of measles vaccine led to significant declines in incidence over time. In the EPI stage, compared to the pre-EPI stage, incidences further decreased by 16.1% (IRR=0.839, P=0.002) each year. Additionally, a short-term level decrease of 4.1% in 1978 (IRR=0.959, P=0.901) was observed.

After the implementation of national and subnational SIAs in China (beginning in 2009), the post-SIAs stage (2009–2021) saw an average annual decrease in the incidence of 54.6% (IRR=0.454, P=0.132), and the short-term level of incidence decreased by 87.9% (IRR=0.121, P=0.355) compared to the EPI stage (1978–2008). Our model indicated an IRR of 1.0002 (P=0.007) with GDP, suggesting a slight increase in incidence associated with GDP.

The results of the ITS Poisson model for the four economic regions were consistent with those observed in China. In the pre-vaccine era, the annual incidence trend increased by 6.7% (IRR=1.067, P=0.005) in eastern China, 6.7% (IRR=1.067, P=0.015) in northeastern China, 5.5% (IRR=1.055, P=0.050) in central China, and 5.6% (IRR=1.056, P=0.044) in western China.

Following the introduction of the measles vaccine in 1965, incidence rates decreased in the short term by 29.5% (IRR=0.705, P=0.124) and 62.00% (IRR=0.380, P=0.002) in eastern and northeastern China. In contrast, incidence rates increased by 6.6% (IRR=1.066, P=0.795) and 3.5% (IRR=1.035, P=0.889) in central and western China. When compared to the pre-vaccine era (1952–1964), incidence rates during the pre-EPI stage (1965–1977) decreased by 16.5% (IRR=0.835, P<0.001), 23.2% (IRR=0.768, P<0.001), 11.7% (IRR=0.883, P=0.001), and 13.0% (IRR=0.870, P<0.001) in eastern, northeastern, central, and western China, respectively.

Following the implementation of EPI in 1978, incidence rates during the EPI stage (1978–2008) showed annual decreases of 16.7% (IRR=0.833, P=0.010), 16.0% (IRR=0.840, P=0.159), 19.7% (IRR=0.803, P<0.001), and 12.5% (IRR=0.875, P=0.006) in eastern, northeastern, central, and western China, respectively. Short-term level reductions were observed in central (IRR=0.754, P=0.399) and western China (IRR=0.966, P=0.913), whereas increases were found in eastern (IRR=1.118, P=0.779) and northeastern China (IRR=3.340, P=0.061).

Following the SIAs, the incidence of the disease decreased by 85.7% (IRR=0.143, P=0.370), 74.6% (IRR=0.254, P=0.662), 88.7% (IRR=0.113, P=0.431), and 94.0% (IRR=0.060, P=0.301) in the eastern, northeastern, central, and western regions of China, respectively. In the post-SIAs phase, the annual incidence reduction was 52.9% (IRR=0.471, P=0.119), 61.9% (IRR=0.381, P=0.248), 67.5% (IRR=0.325, P=0.126), and 47.9% (IRR=0.521, P=0.256) in the eastern, northeastern, central, and western regions of China.