The performance of routine immunization in selected African countries during the first six months of the COVID-19 pandemic

Introduction following the declaration of the COVID-19 pandemic on 11 March 2020, countries started implementing strict control measures, health workers were re-deployed and health facilities re-purposed to assist COVID-19 control efforts. These measures, along with the public concerns of getting COVID-19, led to a decline in the utilization of regular health services including immunization. Methods we reviewed the administrative routine immunization data from 15 African countries for the period from January 2018 to June 2020 to analyze the trends in the monthly number of children vaccinated with specific antigens, and compare the changes in the first three months of the COVID-19 pandemic. Results: thirteen of the 15 countries showed a decline in the monthly average number of vaccine doses provided, with 6 countries having more than 10% decline. Nine countries had a lower monthly mean of recipients of first dose measles vaccination in the second quarter of 2020 as compared to the first quarter. Guinea, Nigeria, Ghana, Angola, Gabon, and South Sudan experienced a drop in the monthly number of children vaccinated for DPT3 and/or MCV1 of greater than 2 standard deviations at some point in the second quarter of 2020 as compared to the mean for the months January-June of 2018 and 2019. Conclusion countries with lower immunization coverage in the pre-COVID period experienced larger declines in the number of children vaccinated immediately after the COVID-19 pandemic was declared. Prolonged and significant reduction in the number of children vaccinated poses a serious risk for outbreaks such as measles. Countries should monitor coverage trends at national and subnational levels, and undertake catch-up vaccination activities to ensure that children who have missed scheduled vaccines receive them at the earliest possible time.


Introduction
The World Health Organisation (WHO) African Region adopted a strategic plan on immunization (2014-2020) that aims to improve immunization coverage in all countries in the region; to complete interruption of poliovirus transmission and ensure virus containment; and to attain measles elimination and control other vaccine-preventable diseases [1]. Subsequently, countries have developed national multi-year comprehensive plans for immunization as well as annual program plans which reflect these ambitions and targets. Countries regularly monitor vaccination coverage, disease trends and other programmatic indicators at national and subnational levels in order to determine progress towards the set milestones and targets. During major natural disasters, humanitarian crises, armed conflicts and largescale disease epidemics, national health systems struggle to maintain the provision of routine health services. Community access and utilization of health services also declines in these conditions. The experiences of the various countries have demonstrated the adverse effects of conflicts and large-scale epidemics on health service delivery and access [2][3][4][5]. During the 2014-2015 Ebola outbreak in West Africa, it was estimated that antenatal care coverage decreased by 22%, while there was an 8% decline in health facility delivery services, and 13% reduction in postnatal care services [5,6]. The Ebola outbreak also significantly affected measles vaccination coverage rates particularly in Guinea and Liberia where decreases in routine immunization coverage led to high measles incidence that persisted for two years after the end of the Ebola outbreak. Liberia and Guinea experienced a 30% and 33% respective decline in the number of children receiving measles doses in 2014 and a decline of 25% and 26% respectively during 2015 [7]. In addition to measles, Guinea also experienced an outbreak of circulating vaccine derived polio virus (cVDPV) type-2 following the program gaps created during the Ebola outbreak [8].
On March 11, 2020 the COVID-19 outbreak was declared a pandemic by WHO with a total of 118,000 cases reported in 114 countries globally. At the time, nine countries in the WHO African region had reported a total of 47 confirmed COVID-19 cases, with no deaths [9]. By 31 March 2020, 42 countries in the region had reported 3766 COVID-19 cases and 95 COVID-19 deaths [10]. By 30 June 2020, all 47 countries in the region have been affected with a cumulative total of 303,986 COVID-19 cases and 6,155 deaths [11]. As the COVID-19 pandemic continued and increasingly strict control measures including lock-downs and social distancing were imposed in various countries, disruptions of regular preventive and curative health services were reported. Fear of contracting COVID-19 in health facilities, reassignment of health workers, closure of health facilities and stock-out of supplies contributed to this disruption. Routine childhood immunization is among the essential health services that faced disruptions. In addition, previously scheduled mass vaccination campaigns against measles, yellow fever and polio were postponed in a number of countries [12][13][14]. Statistical models developed taking into consideration the COVID-19 pandemic have attempted to quantify the number of additional deaths expected as a result of the reduced coverage of various child and maternal health interventions. One model projects as much as a 45% increase in under-five childhood deaths and a 39% increase in maternal death per month if coverage of basic life-saving interventions are extensively disrupted [15].
Considering the risk of increased deaths from preventable causes in the context of COVID-19, the World Health Organisation (WHO) developed guidance for countries to ensure the continuity of essential health interventions including immunization, including the appropriate measures to prevent COVID-19 transmission in health care settings [16][17][18]. Countries monitor the administrative vaccination coverage for each antigen by documenting the number of doses of vaccine delivered in each service delivery site during a defined period, usually one month. These figures are then progressively aggregated at the district, provincial and national levels. The national EPI program shares the compiled country data with the WHO as a monthly report detailing the monthly number of children vaccinated by antigen and by district [19]. This paper examines the actual routine immunization program performance in selected countries in the African Region by comparing the number of children vaccinated in the early months of the COVID-19 pandemic to the number vaccinated in the months prior to the arrival of COVID-19 in the countries.

Methods
We selected 15 countries in the African Region, seven of which are countries with sustained high immunization coverage in previous years, while the rest are countries with low program coverage as evidenced by the WHO-UNICEF annual vaccination coverage estimates (WUENIC) for 2019 and the immunization program maturity grading conducted for countries in the region [20,21]. In addition, the selection considered countries of different sizes and from different geographic sub-regions of the continent. We reviewed the administrative reporting data from the routine immunization programs for the period from January 2018 to June 2020 to analyze the trends in the monthly number of children vaccinated with specific antigens. We compared the number (monthly average and statistical deviation) of children who received vaccine doses before and after the onset of the COVID-19 pandemic. For this analysis, we reviewed the number of children who received BCG vaccine, the first and third doses of Diphtheria-Pertussis-Tetanus containing vaccine (DPT1 and DPT3) and the first and second doses of measles containing vaccine (MCV1 and MCV2). We also calculated the incidence of COVID-19 at the end of June 2020, using official country reports to the WHO as of 30 June 2020, against the UNPD official population estimates for each country.

Results
The 2019 WHO/UNICEF coverage estimates used for the selection of the countries are indicated in Table 1, indicating one subset of countries with low coverage, and the remaining countries having high coverage. The number of COVID-19 cases reported in the respective countries is indicated in Table 2, along with the intensity of COVID-19 transmission as of as of the end of June 2020. With the exception of Eritrea and Rwanda having sporadic cases, the other countries had either community transmission or clusters of cases. The highest incidence among this group of countries was in Gabon, followed by the Central African Republic (CAR), Ghana, Guinea and Senegal (Table 2). Most countries in the region detected their first cases of COVID-19 in March 2020.
The completeness of district reporting of immunization data for the first 6 months of the years 2018 -2020 was >95% in all countries except for South Sudan, which had a completeness of 93% and 96% in 2018 and 2019, while completeness for the first half of 2020 was 91%. The aggregate number of children vaccinated by month in these 15 countries is shown in Figure 1. In April and May of 2020, the number of children vaccinated with DPT1, DPT3 and MCV1 declined as compared to the first quarter of the year. The lowest number of children vaccinated with DPT 1 and DPT 3 doses were in the month of April, while May had the lowest number for MCV1 doses. The number of children vaccinated with DPT3 and MCV1 showed an increase in June as compared to April and May.
At country level, the difference in the mean number of children vaccinated monthly with DPT3 in the months April -June 2020 as compared to January -March 2020 showed a range from a decline of 52% in Guinea to an increase of 6% in Chad. Thirteen of the 15 countries showed a decline in the monthly average of vaccinated children, with 6 countries (Angola, Gabon, Guinea, Nigeria, Burundi and Senegal) having more than 10% decline (Table 3). For the first dose of measles vaccine (MCV1), nine of the 15 countries had a lower monthly mean for April -June as compared to the first quarter of 2020, with Gabon, Guinea, Nigeria and Burundi having a decline in excess of 10%. The change in the mean number of children receiving MCV1 during the second quarter ranged from a decline of 53% in Guinea to an increase of 13% in Chad as compared to the first quarter of 2020 (Table 3) (Table 4 and Table  5). Regarding the second dose of measles vaccine (MCV2) provided in the second year of life, Eritrea and Ghana reported a decline in the monthly number of children vaccinated, with the number in April 2020 exceeding 2 standard deviations as compared to the mean for the months January -June of 2018 and 2019 (Table 6). Rwanda, Nigeria and Angola were excluded from this analysis for different reasons though they have MCV2 in their immunization schedules. Rwanda introduced MCV2 starting in 2015, but the country started reporting MCV2 as part of the monthly data shared with WHO only in 2020. Nigeria started providing MCV2 in late 2019, while the monthly data for Angola had wide discrepancies that rendered the trends in 2020 difficult to interpret as compared to the mean for previous years.

Discussion
This analysis has found that the reduction in the number of children vaccinated through the routine immunization programs in 15 countries during the early period of the COVID-19 pandemic varies considerably. Countries with previously high immunisation coverage, such as Senegal, Rwanda and Eritrea have managed to maintain the levels of service delivery according to the reported cumulative number of children vaccinated for the January-June 2020 period. On the other hand, countries with lower program coverage like Gabon, Guinea, Angola and South Sudan experienced larger declines in the number of children vaccinated immediately after the COVID-19 pandemic was declared. Where the reduction in the number of children vaccinated has persisted, as in Guinea, the risk for outbreaks such measles, is expected to persist and increase with time. The absence of large declines in the number of children vaccinated in countries with chronically low coverage such as CAR, Chad, DR Congo may be explained by the differing extent to which countries have been affected by COVID-19 or by the absence of lengthy or strict societal lockdown and movement restrictions. By June 2020, the number of children vaccinated per month does not appear to have recovered to pre-pandemic levels in CAR, Guinea and South Sudan, while for Nigeria, Angola and Ghana only a transient reduction in the number of children was observed.
Countries with weaker health systems are particularly vulnerable to health service disruptions caused by outbreaks, natural disasters, protracted armed conflict or civil disturbances [4]. During the ebola outbreak of 2014 -2015, Sierra Leone, Liberia and Guinea experienced significant decline in health service delivery [6]. The reduction in measles vaccination documented during the Ebola outbreak in West Africa are comparable to the declines with the COVID-19 outbreak in Guinea. In the two years following the vaccination coverage declines associated with the Ebola outbreak, Liberia, Guinea and Sierra Leone experienced various measles outbreaks that led to a sustained increase in measles incidence despite mass vaccination campaigns and outbreak response immunization efforts [7]. Furthermore, in addition to the disruptions in routine immunization service delivery, many countries have postponed previously scheduled follow-up measles immunization campaigns. These COVID-19 induced delays will further exacerbate the accumulation of susceptible young children, and risks causing measles outbreaks [13]. WHO guidelines recommend a risk-benefit analysis be conducted and the implementation of mass vaccination activities with appropriate infection control measures when the risk of outbreaks is high and further postponement is deemed likely to contribute to outbreaks and deaths from vaccine preventable diseases [18].
The desire to decrease COVID-19 propagation in health facilities as well as the repurposing of health workers may have led to hesitation to continue the usual health facility-based routine immunization services in the early days of the COVID-19 pandemic. However, a benefit-risk analysis has shown that the deaths prevented by sustaining routine childhood immunisation in Africa far outweigh the excess risk of COVID-19 deaths associated with vaccination clinic visits, especially for the vaccinated children [22]. Countries experiencing declines in the coverage of essential health services in the first half of 2020 should ensure the continuity of essential services by implementing the necessary COVID-19 prevention measures and assuring the public about the safety of service delivery. WHO has published guidance on the continuity of essential health services, conducting routine immunization services as well as the considerations when implementing mass vaccination campaigns during the COVID-19 pandemic [17,18]. These guidelines emphasize the need to maintain the provision of immunization services with appropriate precautions to prevent the spread of COVID-19 and a case-by-case assessment of the national COVID-19 and VPD situation. In addition, in August 2020, WHO developed a draft guideline for catch-up vaccination in order to ensure eligible individuals who miss routine vaccine doses for any reason can be identified and vaccinated at the earliest opportunity [23].
In the face of the COVID-19 pandemic, Ethiopia conducted a national measles campaign in July 2020, while DR Congo conducted measles outbreak response and mop-up vaccination in the months of March -June 2020. Both countries undertook these exercises, taking into account the local COVID-19 transmission, and implementing specific COVID-19 prevention measures. There is no evidence to date that these mass vaccination activities contributed to further spread of COVID-19 in the respective countries [24].
Our study has limitations. The analysis considers only 3 months (April-June) of routine immunization administrative coverage data from 15 of 47 countries in the African Region, after the initial detection of confirmed COVID-19 cases in March 2020. Therefore, the trends represented over these three months are only an indication of the early changes in the patterns of immunization coverage. In some countries, administrative coverage data may also include data from periodic intensive service delivery efforts organized in the form of mini-campaigns in order to close service gaps. As a result, the data may reflect the performance beyond the day-to-day service delivery. In addition, only a limited sample of all antigens in routine immunization programs was considered for our analyses. DTP1, DTP3, MCV1 and MCV2 were considered since these antigens are often considered to measure immunization program performance.

Conclusion
The impact of the COVID-19 pandemic on the provision of routine childhood vaccination services in the first three months following the pandemic has varied by country according to the administrative data available. A universal decline was not observed in all countries studied. Countries should monitor trends in the number of children vaccinated at national and subnational levels, and those that experience extensive and persistent declines should undertake catch-up vaccination activities to ensure that children who have missed scheduled vaccine doses receive them at the earliest possible time. Mass vaccination campaigns may be undertaken in the context of COVID-19, in order to prevent the accumulation of nonvaccinated children and thus avert outbreaks, but only after ensuring that benefit-risk analysis is done, and with the appropriate infection prevention measures. Since countries with low immunization coverage have experienced greater COVID-19 related declines in the number of vaccinated children, they should update their outbreak risk assessments, maintain active disease surveillance and prepare to take measures to mitigate the risk for outbreaks.

Competing interests
The authors declare no competing interests.

Authors' contributions
BM and RL conceptualised the study and wrote the original draft. AN, BN, JC compiled the data sets. BM and RL did the analysis. All authors contributed to reviewing and editing of the manuscript and have approved the final version. All the authors have read and agreed to the final manuscript. Table 1: WHO-UNICEF coverage estimates for DPT 3 and MCV1 for 2019 in countries selected for the study Table 2: COVID-19 incidence and transmission status in selected countries as of end June 2020 Table 3: change in the mean number of monthly DPT 3 and MCV1 doses provided in the first two quarters of 2020 Table 4: comparison of the monthly number of DPT 3 vaccinated children in January -June 2020 against the monthly mean for the first half of 2018 and 2019 Table 5: comparison of the monthly number of MCV 1 vaccinated children in January -June 2020 against the monthly mean for the first half of 2018 and 2019 Table 6: comparison of the monthly number of MCV2 vaccinated children during January -June 2020 against the monthly mean for the first half of 2018 and 2019 Figure 1: aggregate number of children vaccinated with specific antigens in the selected countries in the African Region. January -June 2020