Universal healthcare coverage and health service delivery before and during the COVID-19 pandemic: A difference-in-difference study of childhood immunization coverage from 195 countries.

BACKGROUND: Several studies have indicated that universal health coverage (UHC) improves health service utilization and outcomes in countries. These studies, however, have primarily assessed UHC's peacetime impact, limiting our understanding of UHC's potential protective effects during public health crises such as the Coronavirus Disease 2019 (COVID-19) pandemic. We empirically explored whether countries' progress toward UHC is associated with differential COVID-19 impacts on childhood immunization coverage. METHODS AND
FINDINGS: Using a quasi-experimental difference-in-difference (DiD) methodology, we quantified the relationship between UHC and childhood immunization coverage before and during the COVID-19 pandemic. The analysis considered 195 World Health Organization (WHO) member states and their ability to provision 12 out of 14 childhood vaccines between 2010 and 2020 as an outcome. We used the 2019 UHC Service Coverage Index (UHC SCI) to divide countries into a "high UHC index" group (UHC SCI ≥80) and the rest. All analyses included potential confounders including the calendar year, countries' income group per the World Bank classification, countries' geographical region as defined by WHO, and countries' preparedness for an epidemic/pandemic as represented by the Global Health Security Index 2019. For robustness, we replicated the analysis using a lower cutoff value of 50 for the UHC index. A total of 20,230 country-year observations were included in the study. The DiD estimators indicated that countries with a high UHC index (UHC SCI ≥80, n = 35) had a 2.70% smaller reduction in childhood immunization coverage during the pandemic year of 2020 as compared to the countries with UHC index less than 80 (DiD coefficient 2.70; 95% CI: 0.75, 4.65; p-value = 0.007). This relationship, however, became statistically nonsignificant at the lower cutoff value of UHC SCI <50 (n = 60). The study's primary limitation was scarce data availability, which restricted our ability to account for confounders and to test our hypothesis for other relevant outcomes.
CONCLUSIONS: We observed that countries with greater progress toward UHC were associated with significantly smaller declines in childhood immunization coverage during the pandemic. This identified association may potentially provide support for the importance of UHC in building health system resilience. Our findings strongly suggest that policymakers should continue to advocate for achieving UHC in coming years.

Introduction

Achieving universal health coverage (UHC) is a pivotal target of the United Nations’ Sustainable Development Goal 3 (SDG-3) [1]. Under UHC, it is envisioned that populations will have access to essential health services across the full spectrum of care, ranging from promotion to prevention, treatment, rehabilitation, and palliative care, without financial hardship [1]. Many studies to date have provided indications that UHC strategies improve health service coverage, utilization, and outcomes [2-6]. While the causal pathway remains contested, it is argued that UHC’s emphasis on expanding pre-pooled funding mechanisms leads to a reduction in financial barriers to accessing necessary care and thereby results in improvements to population health [2]. Individual country studies further show that UHC helps to reduce inequalities in access to health services and increase utilization across sociodemographic groups, particularly for people with limited financial resources [3-6].

Despite this large evidence base, a robust quantitative assessment of the effects of UHC on health system performance and outcomes has proved challenging for 2 reasons [2]. First, disaggregated and standardized official data for public health, health system, and other pertinent indicators are scarce. Second, there exist many systems-wide contextual factors that may confound the relationship. Hence, the confluence of data and methodological constraints limit our ability to draw firm causal conclusions. For example, an observational study in Indonesia looked at a set of key health indicators, including maternal mortality ratio, infant mortality rate, and life expectancy, and found that UHC interventions achieved preliminary success in improving health equity and service access [7]. Another study from Indonesia used survival analysis to examine child cancer outcomes under UHC and reported significant improvements—especially among disadvantaged socioeconomic groups—in cancer survival and treatment failure [8]. However, the presence of multiple context-dependent confounders limited the generalizability of these findings to other similar healthcare settings.

Further, the extant research on UHC has largely focused on its peacetime impacts on health and health systems, which limits our understanding of UHC’s potential contributions to countries’ preparedness and response capacities during public health crises [9,10]. Studies examining the role of UHC in mitigating the health impacts of the Coronavirus Disease 2019 (COVID-19) pandemic are small in number and have generally focused on COVID-19 outcomes or the first wave of the pandemic in early 2020 [11-13]. Furthermore, given the paucity of empirical research in this area, UHC has generally not been considered integral to assessments of countries’ preparedness and response capacities [9]. This is potentially a significant oversight given that countries’ progress toward UHC requires not only overall health system strengthening but also sustainable pre-pooled funding mechanisms [2], both of which in theory should make countries more resilient to external shocks and more agile when responding to public health crises [10].

Establishing the causal effects of UHC on population health through an experimental study design is extremely challenging. However, an unforeseen public health crisis can serve as a natural experiment. The ongoing COVID-19 pandemic offers an opportunity to examine the role of UHC in safeguarding population health during a public health crisis. The COVID-19 pandemic itself needs little introduction; since early 2020, it has imposed severe burdens on countries’ health systems, affecting the delivery of essential health services in varying but significant ways. With this in mind, we used a quasi-experimental difference-in-difference (DiD) design to compare differences in childhood immunization coverage based on countries’ progress toward UHC, a proxy for countries’ health system resilience, before and during the COVID-19 pandemic.

Immunization coverage, particularly the coverage of essential routine vaccines, is a good outcome measure to gauge the protective effects of UHC before, during, and after a crisis like the COVID-19 pandemic. First, immunization is considered an essential health service across all healthcare settings [14]. Second, vaccine coverage is an easily accessible and robust indicator at the global level; all countries have national immunization programs seeking to provide universal access to essential routine vaccines, especially for vulnerable populations, and report coverage data annually, which is further subject to verification processes for data quality [14]. Lastly, childhood immunizations—as represented by DTP-3 coverage among children and HPV vaccination coverage among teens—are a key input for the World Health Organization (WHO)’s UHC essential service coverage index and a specific target of SDG-3 because of the availability and ubiquity of data on childhood vaccine coverage [1]. The goal of this study was to examine the relationship between UHC and health service delivery during the COVID-19 pandemic shock. We hypothesized that greater progress toward UHC, represented by higher UHC Service Coverage Index (UHC SCI) 2019 values, would safeguard countries’ ability to provide essential health services and minimize disruptions to service delivery during public health crises like the COVID-19 pandemic.

Methods

Data

National immunization data and trends were derived from the WHO/UNICEF Joint Estimates of National Immunization Coverage [15]. These data include annual vaccination coverage—in both absolute numbers and percentages—by country and type of vaccine. The dataset includes 195 countries and 14 childhood vaccines between 1997 and 2020. Details of data collection and calculation are described elsewhere [16,17]. For our analysis, we used a percentage coverage estimate, specifically the number of children who received a specific vaccine dose during a reported year (the numerator) divided by the number of children who were eligible to receive the vaccine during that year (the denominator). We merged the national immunization data with the UHC SCI 2019 obtained from the Institute for Health Metrics and Evaluation (IHME) [18]. UHC SCI 2019 is a robust and comparable measurement framework that measures countries’ health system–effective coverage. Details of its framework and measurement are provided elsewhere [19]. In summary, the UHC index is a weighted aggregate of 23 indicators against 5 types of health services—promotion, prevention, treatment, rehabilitation, and palliation—and 5 age groups—newborn, children under 5 years, children and adolescents between 5 and 19 years, adults between 20 and 64 years, and older adults with age 65 years or more—across the life course. The index ranges from 0 to 100, with 100 indicating the highest effective service coverage. We used the World Bank’s 2020 classification to assign each country to 1 of 4 income groups—high, upper-middle, lower-middle, or low income [20]. To control for countries’ preparedness for an epidemic/pandemic, we incorporated data from the 2019 Global Health Security Index (GHSI) [21]. GHSI 2019 is an assessment of countries’ health security and related capabilities necessary to prepare for future outbreaks including epidemics and pandemics. Details of its framework and measurement are provided elsewhere [22]. GHSI includes 37 indicators across 6 categories—prevention, detection and reporting, rapid response, health system, compliance with international norms, and risk environment. The index ranges from 0 to 100, with 100 indicating the highest preparedness capabilities. The Supporting information presents the full list of countries included in the analysis (Table A in S1 File), as well as the compiled data from the aforementioned sources and the information on how to access the original data (S1 Text). Our study did not require institutional review board (IRB) approval as all of the data used in the study were publicly available and did not include any private, identifiable information.

Data analysis

To test the study hypothesis, we adopted a quasi-experimental research design and conducted a DiD analysis [23]. DiD approaches are typically used to assess the causal effect of a policy or program by comparing the treatment group to a control group before and after an intervention, wherein a clear temporal cutoff pre- and post-intervention exists. DiD models have already been used to provide preliminary evidence of the effects of COVID-19 on several different health outcomes, including but not limited to neonatal outcomes [24], birth outcomes [25], or healthcare utilization rates [26].

DiD analyses need to satisfy 3 assumptions: first, a parallel pre-trend between the treatment and control groups prior to the intervention; second, no external spillovers of the outcome across the 2 groups; and third, that the intervention is unrelated to the outcome at baseline [23]. We demonstrated the parallel pre-trend in Fig 1 and presented empirical evidence of this parallel relationship in the Supporting information (Table B in S1 File). Further, a large body of literature demonstrates that one country’s demand for essential vaccines (COVID-19 vaccines excluded) does not affect other countries’ vaccine supplies due to the generally sufficient global manufacturing of essential vaccines [27]; rather, vaccine shortages are generally due to country-level logistics management, accessibility of health facilities, and health financing and policy issues [28]. Lastly, even in resource-limited countries with poor health service provisioning and lower UHC index values, childhood immunization programs are prioritized and perform relatively well compared to other service areas [1,14].

Fig 1

Change in overall vaccine coverage (A) and vaccine-specific coverage (B) globally by UHC Service Coverage Index 2019 (UHC SCI ≥80 vs. UHC SCI <80) between 1997 and 2020.

Abbreviations: BCG = Bacille Calmette-Guérin; DTP1 = diphtheria, tetanus toxoid, and pertussis containing vaccine—first dose; DTP3 = diphtheria, tetanus toxoid, and pertussis containing vaccine—third dose; HEPB3 = third dose; HEPBB = hepatitis B vaccine—birth dose; HIB3 = Haemophilus influenzae type B containing vaccine; MCV1 = measles containing vaccine—first dose; MCV2 = measles containing vaccine—third dose; PCV3 = pneumococcal conjugate vaccine—third dose; POL3 = polio containing vaccine—third dose; RCV1 = rubella containing vaccine—first dose; ROTAC = rotavirus vaccine—second or third dose; UHC SCI = UHC Service Coverage Index.

We divided countries into 2 groups based on their UHC index value, using a cutoff of 80. Countries with high UHC index values, defined as UHC SCI 2019 ≥80, were assigned to the treatment group, while the rest of the countries (UHC SCI 2019 <80) were assigned to the control group. The cutoff value of 80 was based on the published literature on this index, where UHC SCI ≥80 was operationalized to define the highest level of service coverage provision [19,29]. This was further confirmed through a visual inspection (Fig B in S1 File) of the distribution of the UHC SCI 2019 data, which showed that the data were roughly divided into 2 groups around this cutoff value. For robustness checks, we tested the alternative assumption that countries that made less progress toward UHC would be less able to manage disruptions to the delivery of essential health services during the pandemic. Accordingly, we used a lower cutoff value of 50 for the UHC index, which was close to the bottom quartile (UHC SCI 2019 <48.9) of the UHC SCI 2019 distribution, and assigned countries to the treatment group if UHC SCI 2019 <50 and to the control group if UHC SCI 2019 ≥50. The list of countries in their corresponding UHC SCI 2019 category is available in Table 1.

Table 1

Countries with UHC SCI 2019 ≥80 and UHC SCI 2019 <50.

Countries with UHC SCI 2019 ≥80 (n = 35)	Andorra, Australia, Austria, Belgium, Canada, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy, Japan, Kuwait, Luxembourg, Malta, Monaco, Netherlands, New Zealand, Norway, Portugal, Qatar, Republic of Korea, San Marino, Singapore, Slovenia, Spain, Sweden, Switzerland, United Kingdom, United States
Countries with UHC SCI 2019 <80 and UHC SCI 2019 ≥50 (n = 100)	Albania, Algeria, Antigua and Barbuda, Argentina, Armenia, Bahamas, Bahrain, Bangladesh, Barbados, Belarus, Belize, Bhutan, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Brunei, Bulgaria, Cabo Verde, Cambodia, Chile, China, Colombia, Cook Islands, Costa Rica, Croatia, Cuba, Cyprus, North Korea, Dominican Republic, Ecuador, Egypt, El Salvador, Eswatini, Gabon, Georgia, Grenada, Guatemala, Honduras, Hungary, Iran, Iraq, Jamaica, Jordan, Kazakhstan, Kenya, Kyrgyz Republic, Latvia, Lebanon, Libya, Lithuania, Malawi, Malaysia, Maldives, Mauritania, Mauritius, Mexico, Montenegro, Morocco, Namibia, Nicaragua, Oman, Palestine, Panama, Paraguay, Peru, Philippines, Poland, Moldova, North Macedonia, Romania, Russia, Rwanda, Saint Kitts and Nevis, Saint Lucia, Sao Tome and Principe, Saudi Arabia, Serbia, Seychelles, Slovakia, South Africa, Sri Lanka, Sudan, Suriname, Syria, Thailand, Tonga, Trinidad and Tobago, Tunisia, Turkey, Uganda, Ukraine, United Arab Emirates, Tanzania, Uruguay, Venezuela, Vietnam, Zambia, Zimbabwe
Countries with UHC SCI 2019 <50 (n = 60)	Afghanistan, Angola, Azerbaijan, Benin, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Comoros, Congo, Cote d’Ivoire, Democratic Republic of the Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Fiji, Gambia, Ghana, Guinea, Guinea Bissau, Guyana, Haiti, India, Indonesia, Kiribati, Lao People’s Democratic Republic, Lesotho, Liberia, Madagascar, Mali, Marshall Islands, Federated States of Micronesia, Mongolia, Mozambique, Myanmar, Nepal, Niger, Nigeria, Niue, Pakistan, Palau, Papua New Guinea, Saint Vincent and the Grenadines, Samoa, Senegal, Sierra leone, Solomon Islands, Somalia, South Sudan, Tajikistan, Timor Leste, Togo, Turkmenistan, Tuvalu, Uzbekistan, Vanuatu, Yemen

For the DiD analysis, we leveraged the COVID-19 pandemic to introduce a “prepost” variable wherein the years prior to 2020 (2010 to 2019) were defined as “pre” (0) and the pandemic year of 2020 was defined as “post” (1). While suboptimal for measuring the long-term treatment effects, DiD models only require one observation of data posttreatment to yield preliminary evidence of the treatment effect [23]. Unlike in a typical DiD design, the COVID-19 pandemic happened to countries on both ends of the UHC spectrum. While atypical, our study still falls within the DiD framework given our assumption that health system resilience against COVID-19, the treatment effect of interest, was present at different levels within the 2 groups and was present prior to the COVID-19 pandemic.

The primary outcome was childhood immunization coverage, and the analysis was conducted for both overall vaccine coverage and specific vaccine coverage. We excluded the yellow fever vaccine (YFV) from the analysis because YFV is only administered in a limited number of high-risk countries, and the data are therefore not balanced between our stratified UHC categories. We also excluded the first dose of the inactivated polio vaccine (IPV-1) because IPV-1 was only introduced into the dataset after 2015, and the data were collected irregularly. After these exclusions, our analysis included 12 different childhood vaccines: bacille Calmette–Guérin (BCG); the first and third dose of diphtheria, tetanus toxoid, and pertussis containing vaccine (DTP1, DTP3); the birth dose of hepatitis B vaccine (HEPB-3); the third dose of hepatitis B containing vaccine (HEPBB); the third dose of Haemophilus influenzae type B containing vaccine (HIB3); the first and second doses of measles containing vaccine (MCV1, MCV2); the third dose of pneumococcal conjugate vaccine (PCV3); the third dose of polio containing vaccine (POL3); the second or third dose of rotavirus vaccine (ROTAC); and the first dose of rubella containing vaccine (RCV1).

We first used linear regression to estimate the base model without the DiD term (Eq 1) to illustrate the difference in immunization coverage between the pre- and post-COVID-19 period, as well as between the countries in the treatment and control groups. In the equation below, “Prepost” is the dummy variable that divides the time period into the pre-pandemic and pandemic periods, as explained above. “Treatment” is the dummy variable to indicate a country’s assignment to either the treatment or control group.

We then used linear regression to estimate a DiD model for the effect of UHC on childhood immunization coverage pre-pandemic and during the pandemic (Eq 2).

The DiD estimator is represented by an interaction term between the Prepost and Treatment dummy variables in a regression model, and the coefficient β4 quantifies the causal impact of UHC on health system resilience after adjusting for all covariates. In all the analyses performed, we controlled for the following covariates as represented by the vector Z in the equation: calendar year, countries’ income group as per the World Bank classification [20], geographical location based on countries’ memberships to the WHO regional offices [30], and countries’ preparedness for an epidemic/pandemic as represented by the GHSI [31].

We used linear regression because our outcome variable was continuous (range: 0 to 100) and because the pre-period trend in immunization coverage (2010 to 2019) appeared to be linear. Even if some countries’ immunization coverage for certain vaccines were close to 100%, this should have not affected our analysis, which aimed to quantify the reduction in coverage, which brings coverage down from 100%. However, we limited our interpretation of the results to only the DiD coefficient (β4) because the limitation of our linear regression methodology was that predicted values could be obtained beyond a realistic range (0 to 100).

The analysis was initially conceptualized and planned in September 2021, conducted between September 2021 and January 2022, and was revised based on peer review in April 2022. As part of the revisions, 2 additional analyses were conducted to ensure the robustness of findings. First, we tested whether shortening the pre-period to 2015 to 2019 would change the DiD estimate. Second, we replicated the same analysis using a sliding scale of cutoff values between 50 and 80 with step increments of 5 to evaluate the threshold of UHC SCI 2019 where significant resilience occurs. Throughout the analysis, all statistical tests are performed two-sided, and we use the threshold of 0.05 (P < 0.05) for stating the statistical significance. All analyses were conducted using R software (version 3.6.3), and the data and code used in the analysis are available in a public repository detailed in the Supporting information (S1 Text). This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (S1 STROBE Checklist).

Results

A total of 38,139 country-year observations were included in the collated data, of which 1,658 took place after the COVID-19 pandemic began. Fig 1 shows the mean childhood immunization coverage rate overall and by type of vaccine from 1997 to 2020. Prior to 2010, many countries, especially those with lower UHC index values, showed rapid improvements in immunization coverage each year; we thus performed the DiD analysis on the data from 2010 onward to fulfill the requisite parallel pre-trend assumption. This criterion resulted in a total sample size of 20,230, of which 1,658 (8.2%) observations took place during the pandemic.

Of the 18,572 observations over the pre-pandemic period, those countries with a UHC index greater than or equal to 80 (N = 3,223) had an average childhood immunization coverage rate of 92.6% (standard deviation [SD] 9.8), whereas those with a UHC index less than 80 (N = 15,349) had an average immunization rate of 86.6% (SD 16.0). Of the 1,658 observations that took place during the pandemic year of 2020, countries with the high UHC index (UHC SCI 2019 ≥80) (N = 232) had an average childhood immunization coverage rate of 92.0% (SD 9.0), and countries with the UHC index lower than 80 (N = 1,426) had an average coverage rate of 82.3% (SD 16.0). Summaries of countries’ characteristics in the treatment and control groups based on the cutoff values of 80 and 50 are provided in Table 2 and the Supporting information (Tables C in S1 File).

Table 2

Summary characteristics of the countries included in the analysis (N = 195) based on the cutoff value of 80 for the UHC SCI 2019.

	UHC Index <80 (N = 160)	UHC Index > = 80 (N = 35)	Overall (N = 195)
UHC SCIa 2019
Mean (SDb)	54.6 (11.9)	88.0 (4.60)	60.6 (16.8)
World Bank Income group
High	26 (16.3%)	35 (100%)	61 (31.3%)
Upper-Middle	54 (33.8%)	0 (0%)	54 (27.7%)
Lower-Middle	49 (30.6%)	0 (0%)	49 (25.1%)
Low	30 (18.8%)	0 (0%)	30 (15.4%)
WHO Region
Africa	47 (29.4%)	0 (0%)	47 (24.1%)
Americas	33 (20.6%)	2 (5.7%)	35 (17.9%)
Eastern Mediterranean	20 (12.5%)	2 (5.7%)	22 (11.3%)
Europe	27 (16.9%)	26 (74.3%)	53 (27.2%)
South East Asia	11 (6.9%)	0 (0%)	11 (5.6%)
Western Pacific	22 (13.8%)	5 (14.3%)	27 (13.8%)
Global Health Security Index 2019
Mean (SDb)	36.1 (11.2)	58.7 (13.8)	40.2 (14.6)

aUHC SCI: Universal Health Coverage Service Coverage Index.

bStandard deviation.

The results of the adjusted base model and the DiD model are shown in Table 3, and the unadjusted base model and the DiD model are provided in the Supporting information (Table A in S2 File). Across both the treatment and control groups, the global childhood immunization coverage rate in 2020 was 2.72% lower than the average of the period from 2010 to 2019 when all other covariates were held constant (base model coefficient for Prepost = −2.72%; 95% CI: −3.50, −1.95; p-value < 0.001). Using the DiD model, we found that countries with a high UHC index (UHC SCI 2019 ≥80) had a 2.70% smaller reduction in overall childhood immunization coverage during the pandemic year of 2020 (DiD model coefficient for Prepost * Treatment = 2.70; 95% CI: 0.75, 4.65; p-value = 0.007) as compared to the control group. In other words, when controlling for all other covariates, countries with high UHC index values experienced almost no decline in immunization coverage (DiD model combined effect size for Prepost and Prepost * Treatment = −0.41; 95% CI: −3.18, 2.36) as compared to a significant decline (DiD model coefficient for Prepost = −3.11; 95% CI: −3.93, −2.29; p-value < 0.001) in coverage among countries with low UHC index values (UHC SCI <80) (Fig 2).

Table 3

DiD analysis results of countries with high UHC index values (UHC SCI ≥80) vs. all other countries (UHC SCI <80) in childhood immunization coverage in pre-pandemic period (2010–2019) compared with pandemic period (2020)—Adjusted for calendar year, pandemic preparedness, country income group, geographic region, and vaccine types.

	Base model			DiD model
Variable	Coefficient	95% CIa	p-value	Coefficient	95% CIa	p-value
Intercept	47.49	(−91.18, 186.16)	0.502	48.22	(−90.43, 186.87)	0.495
Year	0.02	(−0.05, 0.08)	0.651	0.02	(−0.05, 0.08)	0.659
GHSIb 2019	0.06	(0.04, 0.08)	<0.001	0.06	(0.04, 0.08)	<0.001
World Bank Income Group (Reference category: Low)
Lower-middle	7.11	(6.44, 7.78)	<0.001	7.11	(6.44, 7.78)	<0.001
Upper-middle	11.43	(10.69, 12.17)	<0.001	11.42	(10.68, 12.16)	<0.001
High	17.02	(16.20, 17.85)	<0.001	17.01	(16.19, 17.84)	<0.001
WHO Region (Reference category: Americas)
Europe	2.62	(2.01, 3.23)	<0.001	2.61	(2.01, 3.22)	<0.001
Western Pacific	0.01	(−0.68, 0.69)	0.986	0.01	(−0.68, 0.69)	0.981
Eastern Mediterranean	−0.38	(−1.11, 0.35)	0.309	−0.37	(−1.10, 0.36)	0.315
Southeast Asia	3.87	(2.90, 4.83)	<0.001	3.87	(2.91, 4.83)	<0.001
Africa	−0.73	(−1.43, −0.03)	0.042	−0.72	(−1.43, −0.02)	0.043
Vaccine type (Reference category: BCGo)
DTP1c	0.59	(−0.28, 1.46)	0.183	0.59	(−0.28, 1.46)	0.184
DTP3d	−3.85	(−4.72, −2.98)	<0.001	−3.85	(−4.72, −2.98)	<0.001
HEPB3e	−4.64	(−5.52, −3.76)	<0.001	−4.64	(−5.52, −3.76)	<0.001
HEPBBf	−10.98	(−12.11, −9.86)	<0.001	−10.97	(−12.09, −9.85)	<0.001
HIB3g	−4.95	(−5.83, −4.07)	<0.001	−4.95	(−5.83, −4.07)	<0.001
MCV1h	−4.63	(−5.50, −3.76)	<0.001	−4.63	(−5.50, −3.76)	<0.001
MCV2i	−10.73	(−11.66, −9.81)	<0.001	−10.73	(−11.66, −9.80)	<0.001
PCV3j	−11.54	(−12.56, −10.52)	<0.001	−11.53	(−12.55, −10.51)	<0.001
POL3k	−3.93	(−4.80, −3.06)	<0.001	−3.93	(−4.80, −3.06)	<0.001
RCV1l	−3.22	(−4.16, −2.28)	<0.001	−3.22	(−4.15, −2.28)	<0.001
ROTACm	−14.57	(−15.74, −13.39)	<0.001	−14.58	(−15.75, −13.40)	<0.001
DiD variables
Pre/Post	−2.72	(−3.50, −1.95)	<0.001	−3.11	(−3.93, −2.29)	<0.001
UHC SCIn 2019 ≥80	−3.74	(−4.50, −2.97)	<0.001	−3.92	(−4.69, −3.14)	<0.001
Pre/Post * UHCn SCI ≥80				2.7	(0.75, 4.65)	0.007

aConfidence interval.

bGlobal Health Security Index.

cDiphtheria, tetanus toxoid, and pertussis containing vaccine—first dose.

dDiphtheria, tetanus toxoid, and pertussis containing vaccine—third dose.

eHepatitis B vaccine—third dose.

fHepatitis B vaccine—birth dose.

gHaemophilus influenzae type B containing vaccine.

hMeasles containing vaccine—first dose.

iMeasles containing vaccine—third dose.

jPneumococcal conjugate vaccine—third dose.

kPolio containing vaccine—third dose.

lRubella containing vaccine—first dose.

mRotavirus vaccine—second or third dose.

nUHC Service Coverage Index.

oBacille Calmette–Guérin.

Fig 2

Differential drop in childhood immunization coverage (%) during the COVID-19 pandemic between the countries with high UHC index (UHC SCI 2019 ≥80) and the rest (UHC SCI 2019 <80).

Abbreviation: UHC = Universal healthcare coverage; SCI = Service coverage index.

As a robustness check, we first used a lower threshold of UHC SCI <50 to group countries into treatment and control groups (Supporting information Table Z in S2 File for unadjusted analysis, Table AA in S2 File for adjusted analysis). In this analysis, we found that the adjusted DiD coefficient was not significant (coefficient −0.56; 95% CI: −1.98, 0.86; p-value = 0.44), suggesting that there was no differential impact of a lower threshold of UHC on childhood immunization coverage during the pandemic. This finding may imply that achieving a certain level of UHC might be critical for the protective benefit of UHC to take effect. Subsequently, when we repeated the same analysis using the sliding threshold of UHC SCI between 50 and 80 with step increments of 5, we observed a significant association between UHC and a differential drop in immunization coverage during the pandemic from the threshold value of 65 and above (Supporting information Fig A and Table A in S3 File). Changing the pre-period from 2010 to 2019 to 2015 to 2019 did not change the overall findings (Supporting information Tables B–E in S3 File).

The DiD results using the original threshold for the UHC index (treatment: UHC SCI 2019 ≥80; control: UHC SCI 2019 <80) for each individual childhood vaccine are presented in Table 4, and full regression results are available in the Supporting information (Tables D–Y in S2 File). None but one (RCV1; coefficient 4.55; 95% CI: 0.15, 8.96; p-value = 0.043) of the DiD coefficients were statistically significant (i.e., no p-values < 0.05), but the DiD coefficient was positive with a p-value < 0.20 for some vaccines, namely, MCV-1 (coefficient 4.26; 95% CI: −1.30, 9.82; p-value = 0.13) and HEPB-3 (coefficient 5.47; 95% CI: −1.45, 12.38; p-value = 0.12). The absence of statistical significance of these DiD coefficients could be due to the smaller sample sizes resulting in insufficient statistical power to yield significance.

Table 4

Adjusted DiD analysis results of countries with high UHC index values (UHC SCI ≥80) vs. all other countries (UHC SCI <80) in immunization coverage by type of vaccine in pre-pandemic period (2010–2019) compared with pandemic period (2020).

Vaccine type	DiDa coefficient*	95% Confidence Interval	p-value
All vaccines	2.7	(0.75, 4.65)	0.007
BCGb	−2.19	(−13.94, 9.57)	0.716
DTP-1c	2.46	(−1.41, 6.32)	0.213
DTP-3d	2.36	(−2.74, 7.47)	0.364
HEPB3e	5.02	(−1.11, 11.14)	0.108
HEPBBf	−0.23	(−22.42, 21.97)	0.984
HIB-3g	1.74	(−4.27, 7.76)	0.57
MCV1h	4.05	(−1.01, 9.10)	0.117
MCV-2i	3.77	(−3.39, 10.93)	0.303
PCV-3j	−2.00	(−11.41, 7.40)	0.676
POL-3k	3.13	(−1.77, 8.02)	0.211
ROTACl	0.72	(−11.38, 12.82)	0.907
RCV1m	4.55	(0.15, 8.96)	0.043

*All analyses are adjusted for calendar year, pandemic preparedness, country income group, geographic region, and vaccine types (results full adjusted analyses and unadjusted analyses are available in the Supporting information (Tables A–AA in S2 File).

aDifference-in-difference.

bBacille Calmette–Guérin.

cDiphtheria, tetanus toxoid, and pertussis containing vaccine—first dose.

dDiphtheria, tetanus toxoid, and pertussis containing vaccine—third dose.

eHepatitis B vaccine—third dose.

fHepatitis B vaccine—birth dose.

gHaemophilus influenzae type B containing vaccine.

hMeasles containing vaccine—first dose.

iMeasles containing vaccine—third dose.

jPneumococcal conjugate vaccine—third dose.

kPolio containing vaccine—third dose.

lRotavirus vaccine—second or third dose.

mRubella containing vaccine—first dose.

Discussion

In this paper, we found preliminary evidence suggesting UHC’s potential contribution to safeguarding health service delivery against external shocks. Using the DiD methodology to measure the differences in immunization coverage before and during the first year of the COVID-19 pandemic, we observed significantly smaller decline in childhood immunization coverage among countries with greater progress toward UHC.

Our findings suggest that countries with greater progress toward UHC were able to mitigate the decline in childhood immunization coverage during the pandemic year of 2020, even after adjusting for vaccine type and countries’ income level, geographic region, and health emergency preparedness and response capacity. This finding may provide potential support to our hypothesis that greater progress toward UHC safeguard countries’ ability to provide essential health services, as measured by the proxy of childhood immunizations, during external shocks.

With our robustness checks, we observed that the association between UHC index and change in immunization coverage was only observed among countries with strong UHC performance (UHC SCI 2019 ≥80). Although the same significant impact was not observed when the analysis was performed on each individual vaccine, this lack of statistical significance is likely due to insufficient statistical power stemming from the inherently small sample size of countries in the world. Nonetheless, to the best of our knowledge, our study is one of the first attempts to quantify the possible contribution of UHC in safeguarding health system performance during a public health crisis. Our findings not only add to the current literature showing the tangible benefits of UHC on health and health systems, thus strengthening the basis of policy dialogue and advocacy to promote UHC, but also underscores the relevance and importance of future research on the protective effects of UHC during public health crises.

There were 2 obstacles to overcome in this analysis. First, studies relying on cross-sectional summary health statistics have shown that countries’ progress toward UHC might be irrelevant to COVID-19-related health outcomes [9]. However, summary health statistics may not be appropriate in analyses seeking to examine the complicated mechanisms underpinning countries’ pandemic preparedness and response capacity, and more advanced methodologies are needed to test for causality. Second, establishing empirical evidence of the pandemic’s impact on countries’ service delivery is currently difficult due to the innate delay in the release of pertinent national statistics. Most annual global health statistics require at least a year before their release; the delay will likely be longer for the 2020 data due to the pandemic. Timely evidence, however, is critically important to inform policies aimed at improving countries’ health system resilience both now and before the next public health crisis strikes. For this reason, we used currently available data on routine childhood immunization as a starting point and plan to follow up in the future with more complete data on other types of essential health services. Even though our data source for routine childhood immunization is widely used for research exploring country-level immunization coverage [32-34], we acknowledge uncertainty in the data’s estimates [17,35]. These uncertainties include country-level estimates not accounting for pockets of low immunization coverage within countries, varying reporting mechanisms and data quality across countries, and differing methods used in the model- and consultation-based adjustments. Even with these limitations, we believe the data source was optimal for testing our hypothesis, which concerns immunization coverage trends rather than point estimates.

We believe that the study can be further improved in 3 ways. First, our findings may still be subject to unobserved confounding factors. Due to data unavailability, we were not able to include several potential confounders as covariates in our analysis, such as countries’ average number of healthcare workers per capita, the size of the catchment population for each type of vaccine, the resources allocated for immunization activities, and countries’ vaccine availability. Additionally, for simplification, we assumed that all covariates included in our analysis were time-fixed within our analytic timeframe (2010 to 2020) after observing no major change in any of the variables included in our analysis. However, the inclusion of more confounders, including time-varying covariates, may improve future results.

Second, once the immunization coverage data for subsequent years becomes available, the same hypothesis can be tested using a comparative interrupted time series analysis. In this analysis, the one-time change in post-pandemic immunization coverage attributed to UHC’s protective effect, which is analogous to the DiD coefficient in our study, can be quantified by the interaction term between the indicator variables defining the pre-/post-periods and the treatment and control groups based on the UHC SCI index. The subsequent year-over-year change in post-pandemic immunization coverage attributed to UHC’s protective effect can be quantified by introducing a 3-way interaction term between the prepost variable, the treatment variable, and the year variable centered to the introduction of the shock so that 2020 will be coded with a value of 1, 2021 with a value of 2, and so forth. Using this approach, we will be able to not only quantify the protective effect of UHC in the face of an external shock, but also observe the effect of UHC on the speed of health system recovery in subsequent years. Toward this end, we aim to monitor the release of data so that we can further evaluate the role of UHC in supporting countries to better respond to—and recover from—public health crises.

Using public health indicators other than immunization coverage to serve as a proxy measure of essential health service delivery can further strengthen our findings. We believe that the HIV continuum of care indicators, such as CD4+ count and viral load, or other indicators related to maternal and newborn health could be robust candidates. Further, the use of key health indicators, such as neonatal, under 5, or maternal mortality rates, would enable us to quantify the role of UHC in safeguarding population health in times of crisis rather than just during peacetime, which has been demonstrated by many prior studies [2-6]. These analyses cannot be performed in the foreseeable future due to the dearth of post-pandemic data available for these indicators. However, once the data becomes available, we strongly believe that replicating this analysis using alternative indicators can validate and reinforce our findings.

Countries’ health system resilience against public health emergencies has been studied almost exclusively under the framework of global health security (GHS), with no role of UHC discussed [9]. This clear separation of investigations has precluded the opportunity to examine the potential contribution of UHC in strengthening health system resilience against external shocks; GHS and UHC policies should likely complement each other [9]. In view of the several major outbreaks over the past decade, it is important to understand the synergistic impact of investments in GHS and UHC on countries’ health system resilience against external shocks [36,37]. Toward this end, our study provides preliminary evidence on the role of UHC in supporting countries’ ability to deliver essential health services in the face of external shocks like the COVID-19 pandemic and sets the stage for future research in this area. Further, our findings strongly suggest that policymakers should continue to advocate for policies aimed at achieving UHC in coming years.

References to the datasets used for the analysis.

(DOCX)

Click here for additional data file.

Contains additional descriptive statistics (Tables A-C and Figs A and B).

Paragraph A. Data source, compiled dataset, link to the repository. Table A. Complete list of 195 countries included in the dataset with their UHC SCI 2019 in alphabetical. Fig A. Change in overall vaccine coverage (A) and vaccine specific coverage (B) globally by UHC Service Coverage Index 2019 (UHC SCI <50 vs. UHC SCI ≥50) between 1997 and 2020. Table B. Results of ordinary-square linear regression analysis to assess the parallel pre-trend assumption before the COVID-19 pandemic. Fig B. Histogram of the distribution of UHC SCI 2019 with the cutoff value of 80 for the treatment vs. control group marked with dotted line. Table C. Summary characteristics of the countries included in the analysis (N = 195) based on the cutoff value of 50 for the UHC Service Coverage Index 2019.

(DOCX)

Click here for additional data file.

Contains additional results of the regression analysis (Tables A-AA).

Table A. Regression analysis results of countries with high UHC index values (UHC SCI ≥80) vs. all other countries (UHC SCI <80) in childhood immunization coverage in pre-pandemic period (2010–2019) compared with pandemic period (2020)—Base model with no DiD interaction term (Unadjusted). Table B. Difference-in-difference regression analysis of bacille Calmette–Guérin (BCG) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table C. Difference-in-difference regression analysis of bacille Calmette–Guérin (BCG) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table D. Difference-in-difference regression analysis of the first dose of diphtheria and tetanus toxoid and pertussis containing vaccine (DTP1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table E. Difference-in-difference regression analysis of the first dose of diphtheria and tetanus toxoid and pertussis containing vaccine (DTP1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table F. Difference-in-difference regression analysis of the third dose of diphtheria and tetanus toxoid and pertussis containing vaccine (DTP3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table G. Difference-in-difference regression analysis of the third dose of diphtheria and tetanus toxoid and pertussis containing vaccine (DTP3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table H. Difference-in-difference regression analysis of the third dose of hepatitis B containing vaccine (HEPB3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table I. Difference-in-difference regression analysis of the third dose of hepatitis B containing vaccine (HEPB3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table J. Difference-in-difference regression analysis of the birth dose of hepatitis B containing vaccine (HEPBB) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table K. Difference-in-difference regression analysis of the birth dose of hepatitis B containing vaccine (HEPBB) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table L. Difference-in-difference regression analysis of the third dose of Haemophilius influenza B containing vaccine (HIB3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table M. Difference-in-difference regression analysis of the third dose of Haemophilius influenza B containing vaccine (HIB3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table N. Difference-in-difference regression analysis of the first dose of measles containing vaccine (MCV1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table O. Difference-in-difference regression analysis of the first dose of measles containing vaccine (MCV1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table P. Difference-in-difference regression analysis of the second dose of measles containing vaccine (MCV2) after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table Q. Difference-in-difference regression analysis of the second dose of measles containing vaccine (MCV2) after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table R. Difference-in-difference regression analysis of the third dose of pneumococcal conjugate vaccine (PCV3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table S. Difference-in-difference regression analysis of the third dose of pneumococcal conjugate vaccine (PCV3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table T. Difference-in-difference regression analysis of the third dose of polio containing vaccine (POL3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table U. Difference-in-difference regression analysis of the third dose of polio containing vaccine (POL3) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table V. Difference-in-difference regression analysis of the second or third dose of rotavirus vaccine (ROTAC) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table W. Difference-in-difference regression analysis of the second or third dose of rotavirus vaccine (ROTAC) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table X. Difference-in-difference regression analysis of the first dose of rubella containing vaccine (RCV1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Unadjusted. Table Y. Difference-in-difference regression analysis of the first dose of rubella containing vaccine (RCV1) coverage after COVID-19 pandemic by UHC SCI 2019 (≥80 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, and geographic region. Table Z. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (<50 vs. the rest)—Unadjusted. Table AA. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (<50 vs. the rest)—Adjusted for calendar year, pandemic preparedness, country income group, geographic region, and vaccine types.

(DOCX)

Click here for additional data file.

Contains further sensitivity analyses as suggested by reviewers (Tables A-E, Fig A).

Fig A. Adjusted difference-in-difference coefficient from the analysis replicated with a range of cutoff values threshold (50–80) for UHC SCI 2019. Table A. Adjusted difference-in-difference coefficient from the analysis replicated with a range of cutoff values threshold (50–80) for UHC SCI 2019. Table B. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (> = 80 vs. the rest) from 2015 to 2020—Unadjusted. Table C. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (> = 80 vs. the rest) from 2015 to 2020—Adjusted for calendar year, pandemic preparedness, country income group, geographic region, and vaccine types. Table D. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (<50 vs. the rest) from 2015 to 2020—Unadjusted. Table E. Difference-in-difference regression analysis of overall immunization coverage after COVID-19 pandemic by UHC SCI 2019 (<50 vs. the rest) from 2015 to 2020—Adjusted for calendar year, pandemic preparedness, country income group, geographic region, and vaccine types.

(DOCX)

Click here for additional data file.

Table A. STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies.

(DOCX)

Click here for additional data file.

25 Jan 2022

Dear Dr Tozan,

Thank you for submitting your manuscript entitled "The Role of Universal Healthcare Coverage in Safeguarding Health Service Delivery in Times of Public Health Crises: A Difference-in-Difference Analysis of Childhood Immunization Coverage Before and During the COVID-19 Pandemic" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by Jan 27 2022 11:59PM.

Login to Editorial Manager here: https://www.editorialmanager.com/pmedicine

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

18 Apr 2022

Dear Dr. Tozan,

Thank you very much for submitting your manuscript "The Role of Universal Healthcare Coverage in Safeguarding Health Service Delivery in Times of Public Health Crises: A Difference-in-Difference Analysis of Childhood Immunization Coverage Before and During the COVID-19 Pandemic" (PMEDICINE-D-22-00227R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to three independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by May 09 2022 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

From the academic editor: From the academic editor: The biggest concern is that the UHC index is highly correlated with national income (and health system spend)--so we may be actually assessing whether rich countries maintained immunization rather than whether UHC per se had an impact. To assess the latter, the authors could do an additional analysis of low-middle and low income countries only. Or split analysis into high income, middle income, and low-income. In any case, as written, I am not sure we can conclude much about UHC as the protective factor.

Other editorial comments:

1. Title: Please revise your title according to PLOS Medicine's style.

-Your title must be nondeclarative and not a question.

-It should begin with main concept if possible.

-"Effect of" should be used only if causality can be inferred, i.e., for an RCT. Please place the study design ("A randomized controlled trial," "A retrospective study," "A modelling study," etc.) in the subtitle (ie, after a colon).

-Please use “sentence case capitalization” for the title. Please capitalize the first word of the subtitle. For example: “Universal healthcare coverage and childhood immunization coverage before and during the COVID-19 pandemic: A difference-in-difference analysis”

2. Data availability statement: The Data Availability Statement (DAS) requires revision. Please update the data availability statement to indicate the where the full data sets and analysis code may be accessed.

For each data source used in your study:

a) If the data are freely or publicly available, note this and state the location of the data: within the paper, in Supporting Information files, or in a public repository (include the DOI or accession number).

b) If the data are owned by a third party but freely available upon request, please note this and state the owner of the data set and contact information for data requests (web or email address). Note that a study author cannot be the contact person for the data.

c) If the data are not freely available, please describe briefly the ethical, legal, or contractual restriction that prevents you from sharing it. Please also include an appropriate contact (web or email address) for inquiries (again, this cannot be a study author).

3. Throughout: Please avoid referring to “effect of universal healthcare coverage on childhood immunization coverage” and similar language that strongly implies causality. Please soften language to allow for the possibility of alternate explanations - identified associations may potentially provide support for a causal relationship.

4. Abstract: Background: The final sentence should clearly state the study question.

5. Abstract: Methods and Findings: Please ensure that all numbers presented in the abstract are present and identical to numbers presented in the main manuscript text. Please include a brief mention of the inclusion criteria for the 180 countries, and please explicitly describe the main outcome measures. Please briefly describe how countries’ income group, geographical region, and pandemic preparedness were categorized/assessed. Please describe the threshold of UHC SCI >/=80) to establish the treatment and control groups, in terms of what this metric represents, and the rationale for selecting it.

6. Abstract: Methods and Findings: Please quantify the main results with 95% CIs and p values. Please refer to “statistically non-significant” findings. Please mention the important variables that are adjusted for in the analyses.

7. Abstract: Methods and Findings: In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology.

8. Abstract: Conclusions: * Please address the study implications without overreaching what can be concluded from the data; the phrase "In this study, we observed ..." may be useful. * Please interpret the study based on the results presented in the abstract, emphasizing what is new without overstating your conclusions. Please remove the “Funding” statement from the abstract.

9. Author summary: At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

10. In-text references: Please use numbers within square brackets to refer to references within the main text. Please place brackets prior to sentence punctuation. Where multiple references are indicated, please do not include spaces within brackets.

11. Main text: Please include line numbers running continuously throughout with the revised version.

12. Introduction: Please provide more rationale and importance of the examination of childhood immunization rates specifically as representative of health service delivery, in general (for example, the public health implications of disruption in delivery of essential childhood vaccinations).

13. Method: In the manuscript text, please indicate: (1) the specific hypotheses you intended to test, (2) the analytical methods by which you planned to test them, (3) the analyses you actually performed, and (4) when reported analyses differ from those that were planned, transparent explanations for differences that affect the reliability of the study's results. If a reported analysis was performed based on an interesting but unanticipated pattern in the data, please be clear that the analysis was data-driven.

14. Methods: Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis-- including those made in response to peer review comments-- should be identified as such in the Methods section of the paper, with rationale.

15. Methods: Please ensure that the study is reported according to the STROBE guideline, and include the completed STROBE checklist as Supporting Information. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

The STROBE guideline can be found here: http://www.equator-network.org/reporting-guidelines/strobe/

When completing the checklist, please use section and paragraph numbers, rather than page numbers.

16. Methods: Please provide additional detail for the UHC Service Coverage Index and 2019 Global Health Security Index (GHSI) in terms of what they represent and how they are incorporated into the analyses. Please also mention how geographical location was assigned/categorized.

17. Methods: Page 6-7: “We hypothesized that greater progress towards UHC…” The use of “greater progress” in the text seems to imply changes in UHC SCI over time, rather than UHC SCI index at a single time point (2019). Please clarify.

18. Methods: Page 8: “For the DiD analysis, we leveraged the COVID-19 pandemic to introduce a “pre-post” variable wherein the years prior to 2020 were defined as “pre” (0) and the pandemic year of 2020 was defined as “post” (1).” Please specify the years.

19. Methods: Please specify the significance level used (eg, P<0.05, two-sided) and the statistical test used to derive a p value. When a p value is given, please specify the statistical test used to determine it.

20. Methods: Ethical approval: Please provide the name(s) of the institutional review board(s) that provided or waived ethical approval for the study.

21. Results: Please clarify “SD” in the text, at first use.

22. Results: Please quantify all results from the analyses with 95% CIs and p values. Where adjusted analyses are presented, please explicitly mention the covariates adjusted for, and please also provide the results of unadjusted analyses.

23. Results: Page 12: “Using the DiD model, we found that countries a high UHC index (UHC SCI 2019 80) prevented a 2.93% reduction in overall childhood immunization coverage during the pandemic year of 2020 (p value= 0.01).” Please temper the use of language implying causality throughout, and please describe this association more clearly, avoiding interpretations such as “prevented” in the sentence. We suggest moving “This finding supports our hypothesis that greater progress towards UHC safeguard countries’ ability to provide essential health services, as measured by the proxy of childhood immunizations, during external shocks.” to the Discussion.

24. Discussion: Please present and organize the Discussion as follows: a short, clear summary of the article's findings; what the study adds to existing research and where and why the results may differ from previous research; strengths and limitations of the study; implications and next steps for research, clinical practice, and/or public policy; one-paragraph conclusion.

25. References: Please use the "Vancouver" style for reference formatting, and see our website for other reference guidelines https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references

26. Figures and Tables: Please ensure each figure and table (including supporting information tables) has a title and descriptive legend. Please define all abbreviations used in the figure/table within the legend.

27. Table 2 and Table 3: Please include the covariates adjusted for in the legend, and please also present results from unadjusted analyses. Please present 95% CIs in addition to p values. When a p value is given, please specify the statistical test used to determine it.

28. S1 Table: “Full dataset used for this analysis can be found here: LINK” Please provide the link to access the full data set, with a reference. Please update the data availability statement section of the manuscript submission system accordingly.

29. Table S2-1: Please specify the reference groups. Please mention the variables adjusted for in the analyses. Please also provide the 95% CIs, and results of unadjusted analyses.

30. Table S2-2 through S2-14: Please mention the variables adjusted for in the analyses. Please also provide the 95% CIs, and results of unadjusted analyses. Please refer to p values as p<0.001 where applicable.

31. Tables: We suggest including a table showing the characteristics of the countries included, summarizing UHC SCI (with thresholds of 80 and 50), income group, geographical location, GHSI.

Comments from the reviewers:

Reviewer #1: This paper provides a clearly written account of an effort to understand the relationship between universal healthcare coverage and declines in immunization coverage during the pandemic. Overall, the topic is important and the logic of the paper is sound. However, some details, which would be helpful to the reader, are omitted (I've outlined these below).

1. Abstract: This sentence is unclear: "The DiD estimators indicated that countries with a high UHC index (UHC SCI380) prevented a 2.93% reduction in childhood immunization coverage during the pandemic year of 2020 as compared to the control group (p-value=0.01)." I think my confusion comes from the mixing of causal and descriptive language. Did having high UHC prevent a 2.93% reduction or was the reduction among countries with high UHC 2.93% less than the reduction among countries with low UHC?

2. The paper models immunization coverage (which is a proportion) as a function of covariates and DiD variables using linear regression. Given that immunization coverage for some vaccines is near 1 in some groups, linear regression may not be appropriate (that is, it would be easy to obtain predictive values above 1). Some justification for the use of a linear model should be included here.

3. Did the model perform any type of weighting for population?

4. Page 9, last few sentences: how were covariates included in the model (e.g., linear terms, splines, indicator variables?) and how were functional forms chosen?

5. Page 9, last sentence: how was pandemic preparedness operationalized? Specifically, was UHC part of this definition?

6. Page 12, lines 2-4: was the -2.94% computed from the model without the interaction term? This analysis should be described in the methods section. Moreover, I would change "statistically significant and negative" to "lower", or simply use the second part of this sentence that is easier to understand.

7. How was the summary immunization coverage metric computed (both overall and for individual vaccines? Some details should be included in the Methods. (I see that the text references the WHO/UNICEF Joint Estimates. It would suffice here to give a rough overview of what the immunization coverage metric is and how to interpret it).

8. The paper used the UHC SCI estimates from 2019. Is there any evidence that the pandemic affected UHC? If so, what are the implications for the results?

9. How were specific vaccines that are not typically administered in one or more countries handled? (e.g., BCG in the US)

Reviewer #2: The authors describe an analysis of the influence of universal health coverage (UHC) on the level of disruption to childhood immunisation due to the COVID-19 pandemic. They used a difference-in-difference approach to examine the relative change for each country given their achieved UHC. They found a higher UHC had a protective effect for routine immunisation services in the pandemic. The paper was generally very well written, made some strong points and I enjoyed reading it.

I have some comments on improvements:

The threshold for the UHC cutoff feels a little arbitrary and I think a small amount of work would help justify this (even with the sensitivity analysis at UHC=50). For example, Figure S1-2 would suggest a cutoff of 75 would split the distributions into two more distinct groups. You could perform a Kolmogorov-Smirnov test on a range of cutoff values to find the most distinct split. In this way, the threshold itself may also be informative.

It would be worth discussing the other indicators used in UHC to highlight the value added by this analysis and to illustrate that high UHC does not necessarily only depend on high immunisation rates. Otherwise, the argument may appear circular.

Whilst vaccination coverage may be easily accessible it is subject to large uncertainties in reporting, denominator and effective impact. Coverage clustering leads to zero dose children this while high numbers of doses are delivered there may still be pockets of low immunisation status which will influence the perceived benefits. Similarly, coverage estimates are based on population estimates that are themselves subject to substantial uncertainty. Finally, reporting mechanisms vary substantially between countries. It would be good to discuss and acknowledge these uncertainties.

Over the years 2010-2019 most countries saw an improvement in achieved coverage- in line with this, the reduction seen for 2020 may appear more stark compared to the end of the time period rather than the average. Would it be feasible to compare a shorter initial time period to capture this variation ie comparing 2015-2019 to the pandemic year.

Can you comment on the code availability. The link in the document does not appear to load.

Reviewer #3: This study explores if the Covid 19 pandemic impact in 2020 on immunization coverage was less in 180 countries with stronger UHC. It compares the coverage trends of 14 childhood vaccines between 2010 and 2020 between countries with higher UHC service coverage index score compared to low score (cut off 80 SCI). The comparison was compared using calendar year, income group, geographical region and preparedness index using global health security index (GHSI).

The topic is relevant and timely. It adds evidence that stronger health systems are more resilient and that investments in UHC is important in advance of next pandemic

Suggest minor revisions for publication in PLOS Medicine.

General questions are:

1. Why was the analysis controlled for by pandemic preparedness? In the IHR the index includes coverage. Could it be that preparedness could be well functioning in a country but health systems performing badly? I wonder if by controlling for it, the very thing you want to study also equals out because GHSI and SCI are linked. Table 3 shows no significance of any individual vaccine in the regression analysis. The explanation in the text instead says it is because of insufficient statistical power. Can you explain?

2. The conclusion in the abstract to " strongly suggest policymakers to achieve UHC even during a public health crisis". Would suggest "achieve UHC in the coming years". The study did not study the possibility to achieve UHC during the pandemic.

3. Data is used from the end of 2020, but we know that the situation of impact on essential immunization was very dynamic during all of 2020, where Africa was extraordinary resilient compared to other Regions. Would it not have been good in include data on the epidemiologic situation or lockdown measures, which was very impactful reasons for impact on immunization services?

4. How was the "average coverage rate " calculated? Was it an average of all 14 vaccines by country?

Minor comments in addition:

1. In table 1, why do you now mention the countries with less than 80 but more than 50 UHC SCI?

2. HPV is not included in the Fig 1B. Was it used in the analysis? HPV coverage in 2020 was mainly affected by lockdowns and school closure which is less directly related to UHC.

3. I find it a bit odd to describe in the discussion what equation to use in the next paper. Consider to avoid using the equation and explain plainly what is suggested.

Any attachments provided with reviews can be seen via the following link:

[LINK]

6 May 2022

Submitted filename: (Final) Response to editors and reviewers.docx

Click here for additional data file.

22 Jun 2022

Dear Dr. Tozan,

Thank you very much for re-submitting your manuscript "Universal Healthcare Coverage and Health Service Delivery in Times of Public Health Crises: A Difference-in-Difference Study of Childhood Immunization Coverage Before and During the COVID-19 Pandemic" (PMEDICINE-D-22-00227R2) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by one of the reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Jun 29 2022 11:59PM.

Sincerely,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

1. Title: Please indicate the number of countries in the title.

2. Short title: Please specify the “COVID-19 pandemic” or similar.

3. Data availability statement: Thank you for including the github dataset. Please provide information on how to access the WHO/UNICEF Joint Estimates of National Immunization Coverage, The IHME UHC Service Coverage Index, World Bank Classification for income groups, and the GHSI for 2019. (The references provided in “S1 Dataset used for the analysis” might be useful).

4. Abstract: Line 24-25: We suggest “...quantified the relationship between UHC and childhood immunization coverage before and during the COVID-19 pandemic.”

5. Abstract: Please combine the Methods and Results sections into one section, “Methods and findings”.

6. Abstract: Line 27: It may be helpful to mention that only 12 vaccines were considered in the final analyses.

7. Abstract: Line 35: Here, and throughout the manuscript, please replace “less reduction” with “smaller reduction” or similar.

8. Abstract: Line 37: Rather than “control group” it would be helpful to explain these are the countries with UHC less than 80.

9. Abstract: Lines 35-38: For both the analysis with UHC SCI > or = 80 and 50, please give the numbers of countries that fell above and below these thresholds.

10. Abstract: Line 42: Please replace “less declines” with “smaller declines” or similar.

11. Abstract: Line 43: Please reword this to avoid drawing conclusions about causality. “...may potentially provide support for the importance of UHC in building health system resilience.” or similar.

12. Author summary: Line 62-64: Please use “smaller decline” here. Please also describe the control group.

13. Author summary: Line 65-66: Please remove this point.

14. Introduction: Line 101: Here and throughout, please remove spaces from within brackets [9,10].

15. Introduction: Line 128-131: Please revise throughout to remove emphasis on causal implications. “This goal of the study was to examine the relationship between UHC and health service delivery during the COVID-19 pandemic shock.” or similar may be useful.

16. Methods: Hypothesis: Please move the hypothesis to the end of the Introduction.

17. Methods: Line 140: We suggest “These data include…” or “This dataset includes…” or similar.

18. Methods: Line 143: Please remove spaces from within reference brackets.

19. Methods Line 164: The text indicates that compiled data is presented in paragraph 1 of S1 file. Please make it clear that this paragraph contains information to access the full dataset as well as the information about the original data.

20. Methods: Line 185-187: “Lastly, even in resource-limited countries with poor health service provisioning and lower UHC index values, childhood immunization programs are prioritized and perform relatively well compared to other service areas.” Please provide a reference for this sentence.

21. Methods: Line 218-226: Please provide a brief explanation for why these particular vaccine/dose combinations were selected as indicators (e.g. why the third PCV dose, but the birth dose of Hep B?) if possible.

22. Methods: Data analysis: Please specify the significance level used (eg, P<0.05, two-sided) and the statistical tests used to derive a p value.

23. Results: Line 277: Here and throughout, it may be helpful to consistently refer to UHC index less than or greater/equal to 80 (similar to lines 273-276) rather than switch to “treatment” and “control” groups.

24. Results: Line 296-299: Although this is implied, it would be helpful to please also present here the Pre-post coefficient for UHC SCI greater than equal to 80 and less than 80, with 95% CIs and p values, to support that those countries with an index of 80 and greater experienced no significant change while the decline was significant in countries with an index of less than 80.

25. Results: Line 303: Please use “not significant” here.

26. Results: Line 306-310: “Subsequently, when we repeated the same analysis using the sliding threshold of UHC SCI between 50 and 80 with step increments of five, we observed a significant association between UHC and a differential drop in immunization coverage during the pandemic from the threshold value of 65 and above (Supporting Information Figure S3-1).” Please quantify these results in Figure S3-1, with the difference-in-difference coefficients, 95% CIs, and p values. Please indicate in the legend the tests used to determine significance.

27. Results: Line 340: Please use “absence of statistical significance” here.

28. Discussion: Line 365-368: We suggest rewording to: “Using the DiD methodology to measure the differences in immunization coverage before and during the first year of the COVID-19 pandemic, we observed significantly less decline in childhood immunization coverage among countries with greater progress toward UHC.” or similar.

29. Discussion: Line 369: We suggest revising to: “Our findings suggest that countries with greater progress towards UHC were able to mitigate the decline in childhood immunization coverage…”

30. Discussion: Line 375: Please revise to: “With our robustness checks, we observed that the association between UHC index and change in immunization coverage was only observed among countries with strong UHC performance…”

31. Discussion: Line 379-381: Please revise to: “Nonetheless, to the best of our knowledge, our study is one of the first attempts to quantify the possible contribution of UHC in safeguarding health system performance during a public health crisis.”

32. Discussion: Please be sure that the Discussion is organized as follows: a short, clear summary of the article's findings; what the study adds to existing research and where and why the results may differ from previous research; strengths and limitations of the study; implications and next steps for research, clinical practice, and/or public policy; one-paragraph conclusion.

33. References: Please use the "Vancouver" style for reference formatting, and see our website for other reference guidelines https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references

-Please pay particular attention to the journal title abbreviations (e.g. The Lancet should be Lancet in Reference 2 and throughout).

-Please provide complete details for reference 12, reference 13, reference 15, reference 18.

-Please change the journal title abbreviation to BMJ Glob Health for reference 17.

-Please change the journal title abbreviation to Lancet Infect Dis for reference 27.

-Please change the journal title abbreviation to Lancet Glob Health for reference 29.

-Please change the journal title abbreviation to BMC Int Health Hum Rights for reference 34.

-Please change the journal title abbreviation to PLOS Glob Public Health for reference 35.

34. Table 1: It might be helpful to also include a section of the table listing those countries that fall between UHC SCI 80 and 50, as these would be the ones to “switch” from control to treatment groups between the analyses.

35. Figure 1: Please define all abbreviations used in the legend.

36. Supporting information Table S1-3: We suggest moving the table of summary characteristics for the countries to the main text of the manuscript.

37. Supporting information Tables S3-4: Please make it clear in the title and/or legend that for these analyses you are replicating using a pre period of 2015-2019.

38. Supporting Information 4: STROBE Checklist: We suggest including this as a separate document.

Comments from Reviewers:

Reviewer #1: The authors have addressed my previous comments.

Any attachments provided with reviews can be seen via the following link:

[LINK]

23 Jun 2022

Submitted filename: Response to editors.docx

Click here for additional data file.

29 Jun 2022

Dear Dr Tozan,

On behalf of my colleagues and the Academic Editor, Margaret Kruk, I am pleased to inform you that we have agreed to publish your manuscript "Universal Healthcare Coverage and Health Service Delivery in Times of Public Health Crises: A Difference-in-Difference Study of Childhood Immunization Coverage from 180 Countries Before and During the COVID-19 Pandemic" (PMEDICINE-D-22-00227R3) in PLOS Medicine.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.

In the meantime, please log into Editorial Manager at http://www.editorialmanager.com/pmedicine/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

Please also address the following editorial points:

-Title: Please capitalize only the first word of the title, and first word of the subtitle (after the colon). Please shorten the title to: “Universal healthcare coverage and health service delivery before and during the COVID-19 pandemic: A difference-in-difference study of childhood immunization coverage from 180 countries”

-Introduction: Line 85: Please use “particularly for people with limited financial resources” or similar.

-Introduction: Line 95: Please revise to: “among disadvantaged socioeconomic groups” or similar.

-Methods: Line 153-154: Please mention the fiscal year used for the World Bank classification of income groups.

-Methods: Line 193-194: Please revise to: “...countries that made less progress towards UHC…”

-Table 3: Please correct two typos for the p values (p=0) indicated for Southeast Asia under “WHO region” in the table.

-Figure 2: Please change the y axis label to: “Drop in childhood immunizaiton coverage (%) during 2020” or similar, and please change the legend accordingly. Please indicate in the legend if the bars represent 95% CIs, or other.

-Reference 15: Please change the journal title abbreviation to: Pediatr Res.

-Reference 28: Please change the journal title abbreviation to: Int Health.

-Reference 31: Please revise the author, correct the weblink and revise format. Please also correct this in file S1.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with medicinepress@plos.org. If you have not yet opted out of the early version process, we ask that you notify us immediately of any press plans so that we may do so on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper.

Sincerely,

Caitlin Moyer, Ph.D.

Associate Editor

PLOS Medicine