Can vaccination coverage be improved through reducing the missed opportunities for immunization? Results from the evaluation in Zhejiang province, east China

ABSTRACT

Background: To evaluate the missed opportunities for vaccination (MOV) in Zhejiang province by using the global methodology from World Health Organization (WHO).

Methods: Based on the WHO Planning Guide to Reduce Missed Opportunities for Vaccination (MOV) and Methodology for the Assessment of MOV, 33 health facilities from 11 cities in Zhejiang province were selected. For each health facility, exit investigations for 20 caregivers of children aged 0–23 months and knowledge, attitudes, and practices (KAP) surveys for 10 health workers was implemented. A MOV was determined based on the child’s age on the date of investigation, eligibility for various vaccines. The prevalence of MOV was calculated and the risk factors of MOV were explored.

Results: There were 660 completed exit investigations of caregivers of children aged 0–23 months and 330 health worker KAP investigations. Of the 658 children with documented vaccination records, 12.6% were still under-vaccinated. Among these under-vaccinated children, 54.2% still had a MOV. Children’s age and their previous vaccination behavior, as well as caregivers’ relationship to children and education level had a significant impact on the incidence of MOV.

Conclusion: The high proportions of visits with MOV in Zhejiang province suggested that interventions to reduce MOV in health service settings may be a potential quick win for improving coverage and equity. National immunization programs should explore the tailored efforts to improve health worker practices by making better use of existing health service contacts.

Keywords:

• Evaluation
• Vaccination coverage
• Missed opportunities
• Methodology

Introduction

Expanded program on immunization (EPI), which was launched in 1974, has contributed to the improvements in childhood health and survival worldwide.¹ Chinese EPI was established in 1978 with 4 vaccines, and it continues with 11 vaccines before 7 years of age update. Despite Chinese government has set the goal to achieve and to maintain 90% coverage for each vaccine in the EPI’s schedule and to achieve full immunization for 90% of children 12–23 months of age, a lower coverage can still be observed at peripheral level like district or town level.^2,3 There are many reasons for the lower coverage from the previous reports, among which are the missed opportunities for vaccination (MOV).^4–7

A MOV is defined as any contact with health services by a target person who is eligible for vaccination (unvaccinated or partially vaccinated/not up-to-date, and free of contraindications) but who does not receive all the vaccinations eligible.⁸ One previous study indicated that the prevalence of MOV was 67% among the subpopulation of child and woman who were eligible for vaccination at the time of contact.⁹

The interventions to reducing MOV had been explored in many countries,^10–13 which responded to the need for a coordinated strategy. A methodology to evaluate and reduce MOV drafted by World Health Organization (WHO) has been published, in collaboration with multiple immunization partners.¹⁴ This methodology build on the protocol from Pan American Health Organization, by simplifying the health facility sampling strategy, incorporating qualitative methodologies, and emphasizing implementation of interventions and a follow-up component.¹⁵ Specifically, the evaluation concludes with the participation of all local vaccination providers in a brainstorming session, which aims to synthesize all the available preliminary data and to build consensus and advocacy for an endorsed and funded work plan to reduce MOV, as part of the aim to strengthen health systems and local EPI.

Field work using the new MOV methodology has now been completed in 11 countries across four of the six WHO regions (African, South East Asian, Eastern Mediterranean and Western Pacific regions). In order to ensure that this global methodology is relevant to the Chinese context, and to help institutionalize efforts to reduce MOV, Zhejiang provincial center for disease control (CDC) and prevention piloted this methodology in eleven counties in 2019. The objective of this paper was to document the experiences and lessons learned from Zhejiang province for other areas who might be interested in using the MOV strategy to address persistent vaccination coverage gaps.

Methods

Study area

Zhejiang province is on the east coast of China. It covers an area of 101,800 km², with a population of 71 million (2017 census).Zhejiang is one of the most densely populated provinces in China. The annual population growth of Zhejiang province is around 1%, with an estimated 709,225 births in 2019. Administratively, it is divided into 11 cities, 90 counties, and 1319 towns.

Study design

Our study was based on WHO Planning Guide to Reduce Missed Opportunities for Vaccination (MOV) and Methodology for the Assessment of MOV,¹⁴ which was a ten-step mixed-methods approach that triangulates quantitative and qualitative data from a broad range of interview sources including caregivers, health workers, and healthcare administrators and managers. It included the plan for MOV assessment, preparation for field work or funding, data collection, data analysis and identification of key themes, brainstorm on proposals of intervention or action plan, debriefing to partners, implementation of interventions, monitoring the progress, impact evaluation, incorporating into a long-term health system. The evaluation included the quantitative arm and the qualitative arm. The quantitative arm aimed to assess the frequency of MOV and its possible reasons through the voluntary exit investigation with caregiver and the self-administered health worker knowledge, attitudes, and practices (KAP) surveys, while the qualitative arm aimed to find the interventions to reduce MOV by the focus group discussions with caregivers and health workers, as well as in-depth interviews with healthcare administrators.

Data collection questionnaire

Prior to data collection, the generic caregiver exit investigation and health worker investigation were adapted to Chinese context. Exit investigation included vaccination history, awareness of routine vaccination, and quality of vaccination service. Health worker investigation included KAP of vaccination, with an additional part on vaccination practice and decision making for health worker.

Sample size and participant selection

Since the results are not intended to be representative for a wide range of geographic regions, the WHO methodology recommends a simplified quota sampling strategy. The purposive selection of district, town. and health facilities was utilized on the basis of the third dose of diphtheria-tetanus-pertussis combined vaccine (DTP), which was close to the provincial average level of the coverage of the third dose of DTP. An effort was made to include all health facilities in both rural and urban settings. Data collection was implemented in 11 districts selected from each city of Zhejiang province, with three health facilities per district.

Children aged 0–23 months who attended the selected health facilities for any type of service on the day of evaluation were included in the primary analysis. To assess coordination of vaccination delivery within health facilities, all available health workers were eligible and were randomly selected for the KAP survey, regardless of their involvement in routine vaccination service.

WHO’s methodology recommends a sample size of 600 eligible children and 300 health workers.¹⁴ In this study, each data collection team was assigned to complete 20 exit investigations and 10 health worker KAP investigations per health facility. If a caregiver was accompanied by more than one age-eligible child, the youngest child was selected for the survey. However, as the service size was different among health facilities, data collection teams continued to investigate caregivers and health workers for the entire duration of their visit to the health facility, even if the health facility quota had been met, to compensate for the lower patient volume in the smaller facilities.

Data collection

Eleven data collection team were built from the health workers in 11 CDCs at city level and were trained during the 3 days immediately preceding field work. To avoid compromising the candor of responses, we assigned the health workers to conduct the field work in the geographical area where they normally work. Field work was undertaken during August 16–20, 2019 and was conducted on paper-based data collection method. All these data were later entered into the database using Epi-data software.

Data collectors were positioned at the exits of each selected health facility. They approached caregivers accompanied by a child 0–23 months of age when they were leaving and asked if they were willing to participate in the study. Exit interviews lasted approximately 15 minutes. Every caregiver was asked for the child’ s vaccination card, from which the vaccination dates were recorded. If the vaccination card was present, it was the sole source of vaccination dates. If the vaccination card was not available, the data collector noted the child’s demographic information, which was used to search for the child’s vaccination data in Zhejiang provincial immunization information system (ZJIIS) later.³ If the caregiver did not have the vaccination card and no vaccination record could be found in ZJIIS, we considered this child did not receive any vaccinations. Verbal report of vaccination dates was not accepted in our study. Photocopies of the vaccination card and ZJIIS were taken for subsequent data validation and cleaning. According to Chinese vaccination schedule, vaccinations scheduled before 24 months of age were included:¹⁶ one dose of Bacillus-Calmette-Guérin vaccine (BCG), three doses of diphtheria-tetanus-pertussis combined vaccine (DTP), three doses of poliovirus live attenuated vaccine (PV), two doses of measles-containing vaccine (MCV), three doses of hepatitis B virus vaccine (HepB), one dose of hepatitis A virus vaccine (HepA), two doses of meningococcal polysaccharide vaccine type a (MenV-a) and one dose of Japanese encephalitis vaccine (JEV). During the exit interview, caregivers were asked the main motivation for bringing their children to the health facility. The responses were categorized as 1) medical consultation, 2) vaccination visit, 3) healthy child visit or developmental checkup, 4) child is accompanying adult, 5) other or 6) no reason. Health workers completed a self-administered KAP investigation in approximately 20 minutes, which were either self-administered on paper and subsequently entered into the database.

Data analysis

Data were analyzed using Stata (version 11.0, College Station, Texas). We produced frequency distributions for each variable to explore themes within the caregiver exit and health worker KAP investigations. A MOV was determined based on the child’s age on the date of investigation, eligibility for various vaccines (according to the EPI schedule), and presence of potential contraindications (as recorded on the vaccination card). Only children with either documented evidence of vaccination dates or a blank vaccination card (indicating no vaccines had been given) and who were eligible for vaccination were included in the calculation of MOV. The prevalence of MOV among children in need of immunization was calculated, excluding those already up-to-date or with valid contraindications. The MOV was further cross-tabulated by motivation for visit, as well as other factors from child and caregiver aspects. These categories allowed for grace periods based on the Chinese EPI policies and previous timeliness studies in Zhejiang province.

Results

There were 660 completed exit investigations of caregivers of children aged 0–23 months and 330 health worker KAP investigations, of which only two children did not have documented vaccination records. There was a high response rate with only three refusals among caregivers and no refusals among health workers.

Among 658 children with documented vaccination records, 50.3% were male and 64.1% were under 12 months of age and 95.3% had received at least one dose of vaccination. Most of the caregivers were children’s mothers (76.1%) and had an education level of senior middle school and above (74.0%). Over 90% of the caregivers were investigated at the public health facility, and the main reasons for visiting health facility included medical consultation (47.3%) and vaccination (30.9%). The majority of the surveyed caregivers (90.7%) had an available vaccination record at visit, however, only 21.3% of the surveyed caregivers reported that they were asked for checking the vaccination card when they visited the health facility (Table 1).

Table 1. Characteristics of investigated caregivers of children with documented vaccination records (N = 658)

Child demographics	n	%	Caregiver demographics	n	%	Health facility	n	%
Gender			Gender			Type of Health Facility
Male	331	50.3	Male	525	79.8	Public	605	91.9
Female	327	49.7	Female	133	20.2	Private	53	8.1
Age			Relation to child			Reason to visit		0.0
0–11	422	64.1	Mother	501	76.1	Medical consultation	311	47.3
12–23	236	35.9	Father	127	19.3	Vaccination	203	30.9
Ever vaccinated			Other	30	4.6	Developmental checkup	136	20.7
Yes	627	95.3	Education level			Child is accompanying adult	5	0.8
No	31	4.7	< Senior middle school	171	26.0	Other or no reason reported	3	0.5
			≥ Senior middle school	487	74.0	Child has vaccination card		0.0
						Available at visit	597	90.7
						Not available at visit	61	9.3
						Did health staff ask for vaccination card?		0.0
						Yes	140	21.3
						No	518	78.7

Of the 330 health workers investigated, 82.1% were female and 56.1% were nurses. Most of them worked in public health facilities, and only 6.4% of all surveyed health workers had not subsequently received on-the-job training on vaccination. Majority of the health workers felt their knowledge of vaccination was insufficient or out of date. Of the surveyed health workers, 12.7% incorrectly identified the low-grade fever as a contraindication for any vaccination and 50.0% of the respondents indicated that vaccination status should only be assessed at wellness or routine visits. Over 90% of the health workers would advise caregivers to keep the vaccination card safe. The majority of health workers (82.4%) reported that there was insufficient staff offering vaccination services at their facilities, as well as 45.8% felt that they did not have enough vaccine supply for all children seeking vaccination services (Table 2).

Table 2. Characteristics and knowledge, attitudes, and practices of investigated health workers (N = 330)

Heath worker‘s demographics	n	%	Health worker’s knowledge, attitudes, practices	n	%
Gender			I feel my knowledge of vaccination is insufficient or out of date
Male	59	17.9	Agree	249	75.5
Female	271	82.1	Disagree	81	24.5
Professional Training			What are contraindications for any vaccine?
Clinician	67	20.3	Local reaction to previous dose	291	88.2
Nurse	185	56.1	Low grade fever	42	12.7
Community Health Worker	78	23.6	Seizures under medical treatment	197	59.7
Type of Service			When should vaccination status be assessed？
Public	313	94.8	Child’s wellness/routine visit	165	50.0
Private	17	5.2	Consultation for any illness	107	32.4
Ever trained in vaccination			When a child is accompanying an adult for any reason	58	17.6
Yes	309	93.6	What instructions do you give caregivers when you issued a vaccination card
No	21	6.4	Keep this card safe	298	90.3
			Bring this card to all visits to the health facility	253	76.7
			There is sufficient staff offering immunization services at this facility
			Agree	58	17.6
			Disagree	272	82.4
			There is enough vaccine supply
			Agree	179	54.2
			Disagree	151	45.8

Prior to the visit, 12.6% (83/658) of those investigated children were still under-vaccinated (missing at least one dose for which they were age-eligible) (Table 3). Of the children eligible for at least one vaccination, 54.2% still had a MOV when they left the health facilities and did not receive at least one of the vaccines for which they were eligible. The probability of MOV was higher among children‘s visiting for a non-vaccination purpose (57.8% among non-vaccination visit vs. 42.1% among vaccination visit). Of the number of vaccine doses eligible, 36.6% doses were missed, with 12.9% among vaccination visit and 42.3% among non-vaccination visit. Children‘s age and their previous vaccination behavior, as well as the relationship between caregivers and children and caregiver’s education level were the significant risk factors of the incidence of MOV (Table 4).

Table 3. Missed opportunities for vaccination by reason for visit

		Child-based MOV				Dose-based MOV
	Total children with documented vaccination dates	No. of children with 1+ eligible doses due		Proportion of eligible children with 1+ MOV		No. of eligible doses among children with 1+ eligible doses due	Proportion of total eligible doses missed
	n	n	%	n	%	n	n	%
Vaccination visit	409	19	4.6	8	42.1	31	4	12.9
Non-vaccination visit
Medical consultation	135	24	17.8	13	54.2	38	11	28.9
Healthy child visit or developmental checkup	68	12	17.6	5	41.7	23	9	39.1
Child is accompanying adult	29	19	65.5	11	57.9	41	20	48.8
Other	11	5	45.5	4	80.0	12	6	50.0
No reason reported	6	4	66.7	4	100.0	16	9	56.3
Non-vaccination visit total	249	64	25.7	37	57.8	130	55	42.3
Total	658	83	12.6	45	54.2	161	59	36.6

Table 4. Missed opportunities for vaccination stratified by child and caregiver factors

Child demographics	n	MOV	%	X^2	p	Caregiver demographics	n	MOV	%	X^2	p
Gender				0.1773	0.157	Gender				0.0014	0.971
Male	331	24	7.3			Male	525	36	6.9
Female	327	21	6.4			Female	133	9	6.8
Age				8.1408	0.004	Relation to child				15.1254	0.001
0–11	422	20	4.7			Mother	501	27	5.4
12–23	236	25	10.6			Father	127	11	8.7
Ever vaccinated				12.6533	<0.001	Other	30	7	23.3
Yes	627	38	6.1			Education level				23.9007	<0.001
No	31	7	22.6			< Senior middle school	171	24	14.0
						≥ Senior middle school	487	21	4.3

As part of the field work, the new WHO methodology prescribes a process to develop an action plan to reduce MOV. Health workers suggested a multi-pronged approach focusing on innovative trainings including non-immunization staff for the first time and incorporating a MOV supportive supervision plan. Caregivers suggested that the curative health staff were required to screen all eligible children for vaccination status and to record any missing dose. Furthermore, the curative health staff should refer child with MOV to the nearest vaccination clinic and make the notification to the vaccination staff of the relevant vaccination clinic.

Discussion

The first field experiences with using the new WHO MOV methodology in Zhejiang province had shown that the proposed methodology provided a breadth of actionable information, while focusing on the primary reasons for MOV. Our results indicated that many unvaccinated children did make contact with health facilities and yet did not receive the recommended vaccine doses. Our findings also pointed to two major barriers to full vaccination of eligible children: a lack of coordination between vaccination and curative health services and incomplete vaccination during vaccination visits.

There was a solicit evidence of the high uptake of vaccination service, as indicated by over 95% of the surveyed children with receiving at least one dose of vaccination. However, a substantial proportion of children were still not receiving all the eligible vaccination for during the health facility visits, even during the vaccination visits. Since the percentage of children already up-to-date prior to the visits was low in this study, we assumed that a larger proportion of children visiting health facilities at any given time were likely to be eligible for vaccination.

Although the guidelines of vaccination service of Zhejiang province required that health workers check the child’s vaccination card at every health facility visit for any reason, few health workers might review the vaccination cards for vaccination eligibility, but used it largely to record or verify demographic data as they did not know sufficient knowledge of the regulation of checking vaccination status at every visit. Rodewald L¹⁷ found that screening the immunization histories of children visiting primary care facilities, irrespective of the purpose of the visit, and placed an MOV sticker on the charts of children in need of vaccination; specifying the required vaccine doses. It would help the vaccination workers identify under-vaccinated children and then use other ways to recall these children to receive the needed vaccines. We recommended that the vaccination status of age-appropriate child should be checked at all health facility visit for any reason. It would be more effective and efficient when these interventions were incorporated into a system improvement plan, which could result in increases in both the timeliness of vaccination and the general coverage.¹⁸ Additionally, our finding indicated that the check of vaccination status would become more important when encountered an older child since the proportion of children with MOV increased with age.

The results indicated the necessity of the coordination of vaccination with curative health service. Over half of children at non-vaccination visits missed the opportunities to get vaccination. Therefore, it might be feasible and possible to increase the vaccination coverage and timeliness simply by better using the existing contacts for other health service purpose. We recommended that the policy makers should explore tailored efforts to improve health worker practices and to increase vaccine delivery by making better use of existing health service contacts to improve childhood vaccination. For example, vaccination status checking should be added into the standard clinical practice and maintaining vaccination cards should be required for caregivers at every health service encounter. As we known, the vaccination cards had the dual function of serving as a reminder to caregivers of the vaccination schedule, as well as a way of communication between the health worker and the caregiver.¹⁹ Hence, we needed to encourage the caregiver to well retain the vaccination card and bring it to every health service encounter for checking up the child’s vaccination status.

The current review on MOV by Anelisa Jaca²⁰ recommended that use of education on provider or provider promoting should be considered as interventions on reducing the MOV and improve the vaccination coverage. Health workers must be given the appropriate tools and resources to address vaccination gaps. In order to reduce MOV and increase vaccination coverage, it was imperative that health workers should have the capacity to properly review vaccination cards and provide recommendations of catch-up vaccination to eligible children.^21,22 We suggested that all health workers, not only immunizing staff, should be able to correctly review a vaccination card for eligibility to ensure the synergy between vaccination and other health services. Health facilities should consider leveraging other health workers to assist with vaccination services, such as instituting a “triage” station where vaccination status can be checked. It would help to alleviate the pressure on the vaccination staff, allowing them to spend more time on administering vaccine doses or on interpersonal communication. Additionally, the issue of human resource constraints emerged from the results, with over 80% of health workers indicating that there was insufficient staff offering vaccination services at their facility. Where possible, local ministries of health should consider a functional review of the human resources available in the health system and consider re-alignment of staff or hiring more staff to fill needed roles, as appropriate.

There are few true contraindications to vaccination. Children with low-grade fever, a cold, diarrhea, vomiting, or other mild illness can safely and effectively be vaccinated.^23,24 However, both health workers and caregivers would hesitate or delay to vaccinate a child with mild illness because of the false or unnecessary contraindications, leaving many children unprotected. According to the current findings,²⁰ the MOV could be reduced by using provider education and this method would be useful to address the problem on false contraindications. We recommended that the national health ministry issue the policies to clarify the use of contraindications and monitor its implementation to reduce the delay vaccination. In recent years, the vaccine management law of China as well as the regulations on vaccination and the national recommended vaccination schedule on children under 7 years of age was released or updated successively. Many vaccination staff might be under pressure to master the new requirements from the updated regulations. We suggested that the national or local immunization program prepare the unified training materials and courseware, and organize a series of training courses target all vaccination staff to help them overcome these barriers.

Other impact factors found in this study were also found in our previous studies. First, children in the care of mothers would have a lower incidence of MOV, which was consistent with other studies. To our knowledge, mother was the main caregiver in the family and they would pay more attention to their children including well-baby care and vaccination than father or other legal caregivers.^16,25 Hence, mothers might get more knowledge and have a more positive attitude toward vaccination service that could induce a lower incidence of MOV. Second, higher education background was negatively associated with the incidence of MOV, which was similar to the reports from other settings.^26,27 Higher education background would help mother communicate with the health workers smoothly and have a positive influence on vaccination practice through the better understanding and accepting the knowledge or policy.

Limitations

Our study was subjected to several limitations. First, due to the nonrandom sampling strategy, the results of these assessments were not intended to be provincially representative. The presented results should be viewed as a programmatic assessment whose intent was to diagnose the major program issues and implement the actions to reduce MOV. Second, our study included the children visited the health facilities and did not account for the children within communities. The selection bias would be induced and our findings would underestimate the MOV as we only account for the MOV to attendees of health facilities with retrievable recorded vaccination dates. Third, the cross-sectional methodology was useful for finding the clues associated with MOV. However, these associations might not be causal and the interventions designed to reduce them might therefore not eliminate the associated MOV.

Conclusions

The high proportions of visits with MOV in Zhejiang province suggested that interventions to reduce MOV in health service settings would be a potential quick win for improving the coverage and the equity. The results revealed both supply- and demand-side barriers to fully immunize children who already had access to health services. National immunization programs should explore the tailored efforts to improve health worker practices by making better use of existing health service contacts.

Author Contributions

Y.H. and YP C. conceived and designed the experiments; H.L. and YP.C. performed the experiments; Y.H. and Y.W. analyzed the data; Y.W. contributed reagents/materials/analysis tools; Y.H. wrote the paper.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed

Ethics approval and consent to participate

This study was approved by the ethical review board of Zhejiang provincial CDC. Written informed consent was obtained from a legal caregiver of each eligible child and health workers enrolled in this study.

Acknowledgments

The authors would like to thank the immunization staff from 11 CDCs at city level for their investigation and data collection.

Additional information

Funding

This study was funded by medical and health science and technology project of Zhejiang province [Grant number: 2020KY522];This study was funded by medical and health science and technology project of Zhejiang province [Grant number: 2020KY52].