Vaccinations are recognized as a feasible, cost-effective public health intervention for reducing the burden and associated mortality of many infectious diseases. The purpose of this study was to evaluate the effectiveness of potential interventions to improve the uptake of vaccines among children and adolescents.
We performed a literature search until December 2020. Eligible studies were identified using Cochrane Central Register of Controlled Trials, MEDLINE, PubMed, and other sources. We included studies conducted on children and adolescents aged 5 to 19 years. Studies comprised of hospitalized children and those with comorbid conditions were excluded. Two authors independently performed the meta-analysis.
Findings from 120 studies (123 articles), of which 95 were meta-analyzed, reveal that vaccination education may increase overall vaccination coverage by 19% (risk ratio [RR], 1.19; 95% confidence interval [CI], 1.12–1.26), reminders by 15% (RR, 1.15; 95% CI, 1.11–1.18), interventions for providers by 13% (RR, 1.13; 95% CI, 1.07–1.19), financial incentives by 67% (RR, 1.67; 95% CI, 1.40–1.99), and multilevel interventions by 25% (RR, 1.25; 95% CI, 1.10–1.41). The impact of school-based clinics and policy and legislation on overall vaccination coverage is still uncertain, and no impact of a multicomponent intervention on overall vaccination coverage was found.
Educational interventions, reminders, provider-directed interventions, financial incentives, and multilevel interventions may improve vaccination coverage among school-aged children and adolescents.
What’s Known on This Subject:
Vaccination coverage has plateaued in the past decade. In 2020, 20 million children globally were unprotected, and 10 countries accounted for 62% of unvaccinated children. Existing interventions can improve vaccination coverage in school-aged children and adolescents.
What This Study Adds:
There are various interventions to improve vaccination coverage among school-aged children and adolescents. However, more studies are needed that assess the effectiveness of interventions in low- and middle-income countries.
Vaccinations are recognized as feasible and cost-effective public health interventions that are instrumental in preventing and reducing the burden of many infectious diseases.1–3 Increased vaccination coverage has contributed to developing herd immunity, and consequently, led to a drastic reduction in the incidence and transmission of vaccine-preventable diseases.4 From the years 1990–2017, global mortality for children aged <5 years from vaccine-preventable diseases reduced from 5.1 to 1.8 million.5 The World Health Organization estimates that 2 to 3 million deaths can be prevented every year by ensuring timely vaccinations against diphtheria, tetanus, whooping cough, and measles.6
Despite many strides made to promote vaccination, coverage has plateaued in the past decade.7 As of 2020, ∼20 million children globally remain unprotected, with 10 countries accounting for 62% of unvaccinated children.7 The likelihood that a child will be fully vaccinated by 5 years of age is <20%.7 Vaccination coverage is also well below the desired level among adolescents, and this gap continues to exist worldwide, despite existing recommendations for adolescent immunization.8 Adolescence is a pivotal age bracket for the promotion of preventive measures and catching up on missed vaccinations, boosting of waning immunity, and primary immunization with new vaccines.9 In 2020, however, only 15% of girls between 9 and 14 years of age received a human papillomavirus (HPV) vaccination globally, and the likelihood that a girl will be fully vaccinated by the time she is 14 is <10%.7
Adolescent vaccination schedules include 2 doses of quadrivalent meningococcal vaccine (MCV); 1 dose of tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis, adsorbed (Tdap) vaccine; 2 or 3 doses of HPV vaccine; and 1 annual influenza vaccine.9 Additionally, catch-up vaccinations can be provided as needed for hepatitis B; hepatitis A; measles, mumps, and rubella (MMR); poliovirus; and varicella.9 A knowledge gap exists around which interventions can be effectively implemented to improve the uptake and coverage of vaccination among school-aged children and adolescents. In this systematic review, we evaluated the effectiveness of potential interventions to improve the uptake of vaccines among children and adolescents aged 5 to 19 years.
Methods
Objectives
The objective of this systematic review is to assess the effectiveness of various interventions to improve vaccination coverage among school-aged children and adolescents. We followed Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.10 The PRISMA checklist is presented in Supplemental Table 1.
Studies and Participants
We included primary studies, including randomized controlled trials (RCTs) (individual and cluster), quasi-experimental studies, and controlled before and after studies (CBAs) from both high-income countries (HICs) and low- and middle-income countries (LMICs) that evaluated interventions to improve vaccination coverage compared with no intervention or standard of care. We included studies that had multiple intervention arms and only selected the arm that met the eligibility criteria. We also included studies with a wider age group that were inclusive of our target population of children and adolescents aged 5 to 19 years. We excluded studies that had comparisons or interventions without a control group; qualitative studies on vaccine intent, beliefs, knowledge, and misinterpretations; and studies focusing on vaccine preparations, antibody titers, vaccine administration, and cost effectiveness. Studies performed specifically in children and adolescents who were hospitalized or had existing comorbidities were excluded.
Interventions
We included studies that were comprised of any intervention designed to improve vaccination coverage among school-aged children and adolescents. The following interventions were identified in the included studies of this review:
Educational interventions: education about vaccines delivered through counseling sessions, printable information materials, audio-visual aids, and the Internet;
Reminders: vaccination reminders sent through mailed letters, telephone calls, text messages, and e-mails;
Interventions for providers: interventions targeting providers, including physician training and reminders;
School-based clinics: school-based vaccination programs to vaccinate students on school premises;
Financial incentives: provision of cash and gift vouchers;
Policy and legislative interventions: specific policy and legislation, including state-level legislation and public-private partnerships;
Multicomponent interventions: a combination of interventions to improve coverage; and
Multilevel interventions: interventions delivered simultaneously at different levels (eg, to schools, providers, clinics).
Outcome Measures
The primary outcome of this review was vaccination coverage, and secondary outcomes were series completion of vaccines and missed opportunity for vaccination (MOV).
Search Methods
A comprehensive search strategy was devised, and eligible studies were identified in the following electronic reference libraries: the Cochrane Central Register of Controlled Trials, MEDLINE, PubMed, CINAHL, Web of Science, and Embase (see Supplemental Table 2 for the search strategy). We also screened the reference list of included studies and relevant systematic reviews, and searched Google Scholar by title of included studies and reviewed the first 50 hits. The last date of the search was December 2020.
Data Collection and Analysis
Studies were imported into the Covidence software program,11 and screened by 2 reviewers for eligibility. Any disagreements were resolved through a discussion with a third reviewer until a consensus was reached. Data were extracted from the included studies for study design, participants, sample size, description of intervention and control arms, outcomes, and quality assessment indicators.
We conducted a meta-analysis using RevMan version 5.4 software.12 We report dichotomous data using risk ratios (RRs) with 95% confidence intervals (CIs). Heterogeneity among studies was determined visually from forest plots, and statistical heterogeneity was tested using standard χ2 test and reported as I2, Q, and τ2. Sensitivity analysis was conducted for the primary outcome on the basis of study quality by removing studies judged to be at high risk of bias for sequence generation and attrition. Subgroup analysis was performed for the primary outcome on the basis of age, study setting, type of intervention, and vaccine.
Quality Assessment
Two reviewers independently assessed the quality and risk of bias of each study. For RCTs, the Cochrane risk-of-bias tool was used to assess quality,13 and each study was judged as having high, low, or unclear risk of bias for selection of participants, blinding, attrition bias, and reporting bias. For non-RCTs, we used the Cochrane Effective Practice and Organization of Care criteria,14 and studies were judged as having high, low, or unclear risk of bias on the basis of additional criteria of baseline characteristics, outcomes, contamination, and selective reporting. The overall quality of evidence was summarized in accordance with Grading of Recommendations, Assessment, Development and Evaluation criteria,15 and a grade of either high, moderate, low, or very low was assigned to the overall evidence.16
Results
Results of Search
The search identified a total of 60 472 records after removing duplicates (Fig 1), and the full text of 1139 studies was screened for eligibility. A total of 120 studies (123 articles)17–136 were included, of which 95 (103 articles) were included in the meta-analysis. The characteristics of the included studies are detailed in Supplemental Table 3.
Description of Included Studies
Of the 120 studies included,17–136 81 were RCTs,18–98 38 were quasi-experimental trials,99–136 and 1 was a CBA.17 The quality of the included studies is summarized in Fig 2. All studies were conducted in HICs, including the United States, Canada, United Kingdom, Italy, Israel, Sweden, Australia, Belgium, and the Netherlands. The included studies focused on interventions to improve coverage of HPV; influenza; Tdap; tetanus-diphtheria (Td) booster; meningococcal (A, B, C, and conjugate); diphtheria, pertussis, and tetanus (DPT); hepatitis A and B; MMR; and varicella and polio vaccines. The included studies assessed the following intervention types:
Educational interventions (34 studies)
Reminders (32 studies)
Intervention for providers (20 studies)
School-based clinics (14 studies)
Financial incentives (4 studies)
Policy and legislative interventions (8 studies)
Multicomponent interventions (3 studies)
Multilevel interventions (6 studies)
Risk-of-bias summary. A, Risk-of-bias summary for RCTs. B, Risk-of-bias summary for quasi-randomized studies.
Risk-of-bias summary. A, Risk-of-bias summary for RCTs. B, Risk-of-bias summary for quasi-randomized studies.
Educational Interventions Compared With No Education
Authors of 34 studies (29 meta-analyzed studies)19,23,25,28,29,31,32, 34,38,41,43,45,46,52,54,57,60,61,65,66,71–74, 91,95,104,105,110,119,121,125,126,130 assessed educational interventions delivered through print (leaflets, pamphlets, brochures, posters, and flipcharts), audio-visual aids (videos, radio, PowerPoint presentations, digital health games, and mHealth apps), web sites, and counseling (individual or group). These educational interventions were delivered at schools, health facilities, communities, and through the Internet and targeted to youth or parents. The forest plots for all the outcomes and the subgroup and sensitivity analyses are available in Supplemental Information.
Primary Outcome
The analysis showed that overall, educational interventions may improve vaccination coverage by 19% (RR, 1.19; 95% CI, 1.12–1.26; 29 studies; 88 230 participants; low-quality evidence) compared with the control group (Fig 3). Subgroup analysis by types of educational intervention reveals that printed educational materials (RR, 1.20; 95% CI, 1.07–1.36; 3 studies; 1064 participants), audio-visual interventions (RR, 1.27; 95% CI, 1.15–1.39; 4 studies; 2542 participants), counseling (individual or group) (RR, 2.63; 95% CI, 2.19–3.16; 4 studies; 1826 participants), and multicomponent educational interventions (RR, 1.16; 95% CI, 1.06–1.27; 14 studies; 51 955 participants) may improve overall vaccination coverage. However, web sites (RR, 1.01; 95% CI, 0.96–1.06, 4 studies, 30 843 participants) may not have any effect on overall vaccination coverage.
Forest plot for the effect of educational intervention on vaccination coverage. aIntervention: web site and lesson (varicella). bIntervention: web site and lesson (HPV). cIntervention: web site only (HPV). dIntervention: web site only (Tdap). eIntervention: web site and lesson (Tdap). fIntervention: web site only (MCVA). gIntervention: web site and lesson (MCVA). hIntervention: web site and lesson (MCVB). iIntervention: web site and lesson (MCVC). jIntervention: web site only (MCVC). kIntervention: media education. lIntervention: community mobilization. mDPT. nHPV. oMCVB. pMCVC. qIntervention: parent and adolescent (HPV). rIntervention: parent-only group (HPV). sIntervention: parent only (Tdap). tIntervention: parent and adolescent (Tdap). uIntervention: parent and adolescent (MCVA). vIntervention: parent only (MCVA). wIntervention delivered to oldest girls (HPV). xIntervention delivered to youngest girls. yIntervention: gain framed. zIntervention: loss framed. IV, inverse variance.
Forest plot for the effect of educational intervention on vaccination coverage. aIntervention: web site and lesson (varicella). bIntervention: web site and lesson (HPV). cIntervention: web site only (HPV). dIntervention: web site only (Tdap). eIntervention: web site and lesson (Tdap). fIntervention: web site only (MCVA). gIntervention: web site and lesson (MCVA). hIntervention: web site and lesson (MCVB). iIntervention: web site and lesson (MCVC). jIntervention: web site only (MCVC). kIntervention: media education. lIntervention: community mobilization. mDPT. nHPV. oMCVB. pMCVC. qIntervention: parent and adolescent (HPV). rIntervention: parent-only group (HPV). sIntervention: parent only (Tdap). tIntervention: parent and adolescent (Tdap). uIntervention: parent and adolescent (MCVA). vIntervention: parent only (MCVA). wIntervention delivered to oldest girls (HPV). xIntervention delivered to youngest girls. yIntervention: gain framed. zIntervention: loss framed. IV, inverse variance.
Subgroup analysis by setting reveals that educational interventions may improve overall vaccination coverage in schools (RR, 1.12; 95% CI, 1.01–1.24; 10 studies; 25 505 participants), health facilities (RR, 1.52; 95% CI, 1.19–1.93; 8 studies; 3810 participants), and communities (RR, 1.37; 95% CI, 1.14–1.65; 8 studies; 28 404 participants). Web-based education may not have any effect on overall vaccination coverage (RR, 1.01; 95% CI, 0.96–1.06; 3 studies; 30 511 participants). The subgroup analysis by age group reveals that educational interventions may improve vaccination coverage among children aged 4 to 10 years (RR, 2.10; 95% CI, 1.03–4.31; 3 studies; 850 participants) and adolescents aged 11 to 18 years (RR, 1.12; 95% CI, 1.06–1.19; 21 studies; 84 906 participants); however, educational interventions may not have any effect among young adults aged ≥19 years (RR, 1.45; 95% CI, 0.82–2.56; 5 studies; 2474 participants).
Sensitivity analysis after removing quasi-experimental studies and RCTs with a high risk for bias revealed that multicomponent interventions (previously significant) may not have any effect on vaccination coverage (RR, 1.09; 95% CI, 0.98–1.20). There was no change in conclusions for any other estimates.
Secondary Outcomes
Educational interventions may improve overall vaccination series completion (RR, 1.62; 95% CI, 1.17–2.24; 11 studies; 20 673 participants; low-quality evidence) and may improve series completion for HPV (RR, 2.30; 95% CI, 1.46–3.62; 8 studies; 2374 participants). However, educational interventions may not have any effect on series completion hepatitis B doses (RR, 1.10; 95% CI, 0.80–1.53; 2 studies; 17 515 participants). Educational interventions may lead to reductions in MOV by 7% (RR, 0.93; 95% CI, 0.89–0.98; 1 study; 3176 participants; low-quality evidence).
Reminders Compared With No Reminders
Primary Outcome
The analysis showed that reminders may improve vaccination coverage by 15% (RR, 1.15; 95% CI, 1.11–1.18; 27 studies; 450 125 participants; low-quality evidence) compared with the control group (Fig 4). Subgroup analysis by type of reminders reveals that vaccination coverage may improve when reminders are sent through text messages (RR, 1.10; 95% CI, 1.05–1.16; 8 studies; 75 698 participants), telephone calls (RR, 1.05; 95% CI, 1.02–1.07; 6 studies; 271 253 participants), mailed letters (RR, 1.16; 95% CI, 1.07–1.26; 4 studies; 76 087 participants), both mailed and telephone reminders (RR, 1.33; 95% CI, 1.26–1.40; 4 studies; 19 347 participants) and telephone calls, e-mails, and text messages (RR, 1.73; 95% CI, 1.30–2.31; 1 study; 929 participants). E-mails only (RR, 1.09; 95% CI, 1.00–1.19; 1 study; 3545 participants) and e-mails and text messages (RR, 1.01; 95% CI, 0.88–1.16; 2 studies; 534 participants) might not have any effect on overall vaccination coverage.
Forest plot for the effect of reminders on vaccination coverage. aPolio. bDPT. cMMR. dInfluenza. eTdap. fMCV conjugate. gHPV. hIntervention: telephone reminders (HPV). iIntervention: mailed reminders (HPV). jIntervention: telephone reminders (Tdap). kIntervention: mailed reminders (Tdap). lIntervention: telephone reminders (MCV conjugate). mIntervention: mailed reminders (MCV conjugate). nIntervention: text reminders. oIntervention: motivational text messages. pIntervention: self-regulatory text messages. qInfluenza: Colorado and New York, 3 reminders. rInfluenza: Colorado and New York, 1 reminder. sInfluenza: Colorado and New York, 2 reminders. tTd booster. uDiphtheria-tetanus-acellular pertussis. vHepatitis B. wIntervention: 1 reminder. xIntervention: 2 reminders. yIntervention: 3 reminders. zIntervention: autodialer reminders. aaRate of immunization. bbReceipt of second immunization. IV, inverse variance.
Forest plot for the effect of reminders on vaccination coverage. aPolio. bDPT. cMMR. dInfluenza. eTdap. fMCV conjugate. gHPV. hIntervention: telephone reminders (HPV). iIntervention: mailed reminders (HPV). jIntervention: telephone reminders (Tdap). kIntervention: mailed reminders (Tdap). lIntervention: telephone reminders (MCV conjugate). mIntervention: mailed reminders (MCV conjugate). nIntervention: text reminders. oIntervention: motivational text messages. pIntervention: self-regulatory text messages. qInfluenza: Colorado and New York, 3 reminders. rInfluenza: Colorado and New York, 1 reminder. sInfluenza: Colorado and New York, 2 reminders. tTd booster. uDiphtheria-tetanus-acellular pertussis. vHepatitis B. wIntervention: 1 reminder. xIntervention: 2 reminders. yIntervention: 3 reminders. zIntervention: autodialer reminders. aaRate of immunization. bbReceipt of second immunization. IV, inverse variance.
Subgroup analysis by age group reveals that reminder interventions may improve vaccination coverage among children aged 4 to 10 years (RR, 1.57; 95% CI, 1.20–2.06; 2 studies; 637 participants) and adolescents aged 11 to 18 years (RR, 1.14; 95% CI, 1.11–1.18; 23 studies; 448 859 participants). The subgroup analysis by setting reveals that reminder interventions may improve vaccination coverage when delivered in facilities (RR, 1.16; 95% CI, 1.12–1.20; 19 studies; 394 600 participants) and communities (RR, 1.19; 95% CI, 1.15%–1.23; 4 studies; 36 292 participants) but may not have any effect in school-based settings (RR, 1.05; 95% CI, 1.00–1.11; 4 studies; 20 926 participants).
Sensitivity analysis after removing quasi-experimental studies and RCTs with a high risk for bias revealed that reminders in the form of text messages (RR, 1.03; 95% CI, 1.00–1.06) may not have any effect on vaccination coverage. There was no change in the effect estimates for any of the other interventions.
Secondary Outcomes
The analysis showed that reminders may improve overall vaccination series completion by 24% (RR, 1.24; 95% CI, 1.14–1.35; 15 studies; 84 912 participants; low-quality evidence). Series completion may increase for HPV by 26% (RR, 1.26; 95% CI, 1.15–1.38; 14 studies; 84 173 participants). However, reminders may not have any effect on the completion of the hepatitis B series (RR, 1.10; 95% CI, 0.99–1.24; 1 study; 739 participants). Reminders led to a reduction in MOV by 5% (RR, 0.95; 95% CI, 0.93–0.98; 2 studies; 7681 participants; high-quality evidence).
Interventions for Providers Compared With No Intervention
Primary Outcome
The analysis revealed that interventions targeting providers may improve vaccination coverage by 13% (RR, 1.13; 95% CI, 1.07–1.19; 9 studies; 50 365 participants, low-quality evidence) compared with the control group (Fig 5). The subgroup analysis by intervention reveals that reminders delivered to providers through prompting/alert systems or text messages may improve coverage (RR, 1.13; 95% CI, 1.07–1.20; 6 studies; 32 997 participants), whereas educational interventions (eg, physician training delivered through physician educators, PowerPoint presentations, individual or group training sessions, webinars, feedback systems) may not have any effect on vaccination coverage (RR, 1.09; 95% CI, 0.85–1.40; 3 studies; 17 368 participants). The subgroup analysis by age group reveals that interventions for providers may improve uptake of vaccination in children aged 4 to 10 years (RR, 1.22; 95% CI, 1.10–1.35; 1 study; 660 participants) and adolescents aged 11 to 18 years (RR, 1.12; 95% CI, 1.06–1.19; 8 studies; 49 705 participants). Subgroup analysis by setting was not performed because all studies were done within health facilities.
Forest plot for the effect intervention for providers on vaccination coverage. aIntervention: educational text message. bIntervention: conventional text message. cIntervention Continuity Research Network (CORNET) practices (HPV). dIntervention: CORNET practices (influenza). eIntervention: Greater Rochester Practice-Based Research Network (GR-PBRN) practices (influenza). fIntervention: GR-PBRN practices (Tdap). gIntervention: CORNET practices (Tdap). hIntervention: GR-PBRN practices (MCV conjugate). iIntervention: CORNET practices (MCV conjugate). jIntervention: GR-PRN practices (HPV). kIntervention: single reminder prompt. lIntervention: elaborated reminder prompt. IV, inverse variance.
Forest plot for the effect intervention for providers on vaccination coverage. aIntervention: educational text message. bIntervention: conventional text message. cIntervention Continuity Research Network (CORNET) practices (HPV). dIntervention: CORNET practices (influenza). eIntervention: Greater Rochester Practice-Based Research Network (GR-PBRN) practices (influenza). fIntervention: GR-PBRN practices (Tdap). gIntervention: CORNET practices (Tdap). hIntervention: GR-PBRN practices (MCV conjugate). iIntervention: CORNET practices (MCV conjugate). jIntervention: GR-PRN practices (HPV). kIntervention: single reminder prompt. lIntervention: elaborated reminder prompt. IV, inverse variance.
Sensitivity analysis by removing quasi-experimental studies and RCTs at high risk for bias revealed that educational interventions (previously insignificant) delivered to providers may improve vaccination coverage (RR, 1.10; 95% CI, 1.06–1.15).
Secondary Outcomes
The analysis showed that interventions for providers probably will not have any effect on HPV series completion (RR, 1.05; 95% CI, 0.95–1.16; 3 studies; 7386 participants; moderate-quality evidence). We are uncertain of the effect of provider interventions on overall MOV (RR, 0.90; 95% CI, 0.81–1.01; 3 studies; 93 424 participants; very-low-quality evidence).
School-Based Clinics Compared With Standard of Care
Primary Outcome
We are uncertain that school-based clinics may improve overall vaccination coverage (RR, 1.31; 95% CI, 1.18–1.46; 14 studies; 1 446 947 participants; very-low-quality evidence). On the basis of the subgroup analysis, we are uncertain that school-based clinics have an effect on vaccination coverage among children aged 4 to 10 years (RR, 1.30; 95% CI, 1.16–1.45) and participants adolescents aged 11 to 18 years (RR, 1.29; 95% CI, 1.10–1.50). We are also uncertain of the effect on HPV series completion (RR, 1.46; 95% CI, 1.01–2.11), influenza vaccine (RR, 1.21; 95% CI, 1.15–1.27), MCV conjugate vaccine (RR, 1.55; 95% CI, 1.04–2.31), Tdap vaccine and Td booster (RR, 1.48; 95% CI, 0.95–2.29), and hepatitis B vaccine (RR, 0.62; 95% CI, 0.40–0.95) uptake. Subgroup analysis was not performed for setting and type of intervention because all studies were performed in school-based clinics.
Secondary Outcomes
We are uncertain of the effect of school-based clinics on overall vaccination series completion (RR, 1.80; 95% CI, 0.96–3.39; 4 studies; 39 911 participants; very-low-quality evidence). None of the authors of the included studies reported MOV for this comparison.
Financial Incentives Compared With No Intervention
Primary Outcome
The analysis revealed that financial incentives of cash and gift vouchers probably improve vaccination coverage for HPV by 67% (RR, 1.67; 95% CI, 1.40–1.99; 2 studies; 1188 participants; low-quality evidence) compared with the control group. Subgroup analysis for age, setting, type of intervention, and vaccine was not performed because both included studies were performed among adolescents aged 11 to 18 years in health facilities and were concerned with the uptake of HPV vaccine.
Secondary Outcomes
The analysis revealed that financial incentives may improve HPV vaccine series completion (RR, 2.44; 95% CI, 1.80–3.32; 2 studies; 1188 participants; low-quality evidence) compared with the control group. None of the authors of the included studies reported the outcome of MOV under this comparison.
Policy and Legislative Interventions Compared With No Intervention
Primary Outcome
We are uncertain of the effect of policy and legislation interventions, such as state-level vaccine mandates and public-private partnership programs, on vaccination coverage (RR, 1.19; 95% CI, 1.04–1.36; 3 studies; 84 913 participants; very-low-quality evidence) compared with the control group (Fig 6). On the basis of the subgroup analysis, we are uncertain that state-level policy and legislation (RR, 1.31; 95% CI, 0.93–1.85; 2 studies; 43 413 participants) and public-private partnerships (RR, 1.01; 95% CI, 0.99–1.03; 1 study; 41 500 participants) have an effect on vaccination coverage. Similarly, we are uncertain about the effect of policy and legislation on vaccination coverage among children aged 4 to 10 years (RR, 1.01; 95% CI, 0.99–1.03; 1 study; 41 500 participants) and adolescents aged 11 to 18 years (RR, 1.31; 95% CI, 0.93–1.85; 2 studies; 43 413 participants). Subgroup analysis for study setting was not performed because all included studies were done in a community-based setting.
Forest plot for the effect of policy and legislation on vaccination coverage. IV, inverse variance.
Forest plot for the effect of policy and legislation on vaccination coverage. IV, inverse variance.
Sensitivity analysis after removing quasi-experimental studies and RCTs with a high risk for bias revealed that state-level policy and legislation (previously insignificant) may have an impact on vaccination coverage (RR, 1.10; 95% CI, 1.08–1.12). There was no change in conclusions for any other estimates.
Secondary Outcomes
The analysis revealed that policy and legislation may improve hepatitis B vaccination series completion by 91% (RR, 1.91; 95% CI, 1.69–2.16; 1 study; 982 participants; low-quality evidence) and may probably reduce MOV by 20% (RR, 0.80; 95% CI, 0.78–0.82; 1 study; 28 050 participants; moderate-quality evidence).
Multicomponent Interventions
Primary Outcome
The analysis showed that multicomponent interventions had no effect on vaccination coverage for HPV (RR, 1.05; 95% CI, 0.90–1.23; 1 study; 973 participants; moderate-quality evidence) compared with the control group. Subgroup analysis for age, setting, type of intervention, and vaccine was not performed because there was only 1 study meta-analyzed in adolescents aged 11 to 18 years done in health facilities for the uptake of HPV vaccine.
Secondary Outcomes
The analysis revealed that multicomponent interventions may improve HPV vaccine series completion by 84% (RR, 1.84; 95% CI, 1.20–2.80; 1 study; 337 participants; low-quality evidence). None of authors of the included studies reported the outcome of MOV under this comparison.
Multilevel Interventions Compared With No Intervention
Primary Outcome
The analysis revealed that multilevel interventions may improve vaccination coverage by 25% (RR, 1.25; 95% CI, 1.10–1.41; 6 studies; 160 916 participants; low-quality evidence) compared with the control group (Fig 7). Multilevel interventions given as an educational intervention (RR, 2.01; 95% CI, 1.87–2.16; 1 study; 547 participants) and education in combination with reminder interventions (RR, 1.36; 95% CI, 1.11–1.65; 1 study; 11 326 participants) may improve vaccination coverage. Subgroup analysis by age reveals that multilevel interventions may improve coverage among children aged 4 to 10 years (RR, 1.08; 95% CI, 1.07–1.08; 1 study; 87 665 participants) and adolescents aged 11 to 18 years (RR, 1.29; 95% CI, 1.08–1.55; 5 studies; 73 251 participants). We found that such interventions may be effective in facility (RR, 1.14; 95% CI, 1.01–1.30; 5 studies; 160 369 participants) and community-based settings (RR, 2.01; 95% CI, 1.87–2.16; 1 study; 547 participants). After sensitivity analysis, there was no change in conclusion for any of the estimates.
Forest plot for the effect of multilevel interventions on vaccine coverage. aReceipt of the first dose of the HPV vaccine (medical record review). bReceipt of the first dose of the HPV vaccine (parent reported). cIntervention: clinician-focused decision support. dIntervention: family-focused decision support. eIntervention: health information technology. fIntervention: postcards. IV, inverse variance.
Forest plot for the effect of multilevel interventions on vaccine coverage. aReceipt of the first dose of the HPV vaccine (medical record review). bReceipt of the first dose of the HPV vaccine (parent reported). cIntervention: clinician-focused decision support. dIntervention: family-focused decision support. eIntervention: health information technology. fIntervention: postcards. IV, inverse variance.
Secondary Outcomes
The analysis revealed that multilevel interventions may not have any effect on HPV vaccine series completion (RR, 0.95; 95% CI, 0.82–1.11; 3 studies; 49 990 participants; low-quality evidence) and will probably lead to reductions in MOV for HPV series completion (RR, 0.92; 95% CI, 0.86–0.99; 1 study; 16 529 participants; moderate-quality evidence).
Discussion
In this systematic review, we consolidate data on the effectiveness of interventions to improve vaccination coverage among children and adolescents aged 5 to 19 years. We summarize the findings of 120 studies from 123 articles and assessed educational, reminder, provider, school-based clinic, financial incentive, policy and legislation, multicomponent, and multilevel interventions for improving vaccination coverage. All included studies were performed in HICs. Our findings reveal that educational interventions, reminders, interventions for providers, financial incentives, and multilevel interventions may improve overall vaccination coverage among children and adolescents. Multicomponent interventions, on the other hand, might not have any effect on overall vaccination coverage, and we are uncertain of the effect of school-based clinic and policy and legislation interventions. We also performed subgroup analyses by type of intervention, participant age, study setting, and vaccine administered. Summaries or our review findings and the subgroup analyses are shown in Tables 1 and 2.
Summary of the Effect Estimates and 95% CIs
| Outcomes | Education, 34 studies | Reminders, 32 studies | Provider, 20 studies | School-Based Clinics, 13 studies | Financial Incentives, 4 studies | Policy and Legislative, 8 studies | Multicomponent, 3 studies | Multilevel, 6 studies |
|---|---|---|---|---|---|---|---|---|
| Primary | ||||||||
| Overall vaccination coverage | 1.19 (1.12–1.26) | 1.15 (1.11–1.18) | 1.13 (1.07–1.19) | 1.31 (1.18–1.46) | 1.67 (1.40–1.99) | 1.19 (1.04–1.36) | 1.05 (0.90–1.23) | 1.25 (1.10–1.41) |
| HPV | 1.21 (1.11–1.31) | 1.14 (1.09–1.20) | 1.18 (1.01–1.38) | 1.46 (1.01–2.11) | 1.67 (1.40–1.99) | 1.10 (1.08–1.12) | 1.05 (0.90–1.23) | 1.29 (1.08–1.55) |
| Influenza | 1.82 (1.03–3.22) | 1.04 (1.01–1.06) | 1.09 (1.01–1.17) | 1.21 (1.15–1.27) | — | 1.01 (0.99–1.03) | — | 1.08 (1.07–1.08) |
| Tdap and Td booster | 2.25 (1.11–4.57) | 1.24 (1.13–1.35) | 1.05 (1.00–1.10) | 1.48 (0.95–2.29) | — | — | — | — |
| DPT | 0.95 (0.62–1.46) | 1.13 (0.74–1.72) | — | — | — | — | — | — |
| MCVA | 1.73 (1.05–2.86) | — | — | — | — | — | — | — |
| MCVB | 7.56 (0.67–85.35) | — | — | — | — | — | — | — |
| MCVC | 18.61 (4.32–80.21) | — | — | — | — | — | — | — |
| MCV conjugate | — | 1.36 (1.26–1.48) | 1.05 (0.99–1.11) | 1.55 (1.04–2.31) | — | — | — | — |
| Hep A | — | — | — | — | — | — | — | — |
| Hep B | 1.33 (1.01–1.73) | 1.19 (0.62–2.29) | — | 0.62 (0.40–0.95) | — | 1.57 (1.45–1.70) | — | — |
| Varicella | 3.59 (0.15–87.20) | — | — | — | — | — | — | — |
| MMR | 0.95 (0.91–0.99) | 2.26 (1.23–4.13) | 1.37 (1.04–1.81) | — | — | — | — | — |
| Polio | — | 1.45 (0.88–2.39) | — | — | — | — | — | — |
| Secondary | ||||||||
| Overall series completion | 1.62 (1.17–2.24) | 1.24 (1.14–1.35) | 1.05 (0.95–1.16) | 1.80 (0.96–3.39) | 2.44 (1.80–3.32) | 1.91 (1.69–2.16) | 1.84 (1.20–2.80) | 0.95 (0.82–1.11) |
| HPV | 2.30 (1.46–3.62) | 1.26 (1.15–1.38) | 1.05 (0.95–1.16) | 1.76 (0.85–3.66) | 2.44 (1.80–3.32) | — | 1.84 (1.20–2.80) | 0.95 (0.82–1.11) |
| Hep B | 1.09 (0.93–1.28) | 1.10 (0.99–1.24) | — | 1.94 (1.28–2.93) | — | 1.91 (1.69–2.16) | — | — |
| MOV | 0.93 (0.89–0.98) | 0.95 (0.93–0.98) | 0.90 (0.81–1.01) | — | — | 0.80 (0.78–0.82) | — | 0.92 (0.86–0.99) |
| Outcomes | Education, 34 studies | Reminders, 32 studies | Provider, 20 studies | School-Based Clinics, 13 studies | Financial Incentives, 4 studies | Policy and Legislative, 8 studies | Multicomponent, 3 studies | Multilevel, 6 studies |
|---|---|---|---|---|---|---|---|---|
| Primary | ||||||||
| Overall vaccination coverage | 1.19 (1.12–1.26) | 1.15 (1.11–1.18) | 1.13 (1.07–1.19) | 1.31 (1.18–1.46) | 1.67 (1.40–1.99) | 1.19 (1.04–1.36) | 1.05 (0.90–1.23) | 1.25 (1.10–1.41) |
| HPV | 1.21 (1.11–1.31) | 1.14 (1.09–1.20) | 1.18 (1.01–1.38) | 1.46 (1.01–2.11) | 1.67 (1.40–1.99) | 1.10 (1.08–1.12) | 1.05 (0.90–1.23) | 1.29 (1.08–1.55) |
| Influenza | 1.82 (1.03–3.22) | 1.04 (1.01–1.06) | 1.09 (1.01–1.17) | 1.21 (1.15–1.27) | — | 1.01 (0.99–1.03) | — | 1.08 (1.07–1.08) |
| Tdap and Td booster | 2.25 (1.11–4.57) | 1.24 (1.13–1.35) | 1.05 (1.00–1.10) | 1.48 (0.95–2.29) | — | — | — | — |
| DPT | 0.95 (0.62–1.46) | 1.13 (0.74–1.72) | — | — | — | — | — | — |
| MCVA | 1.73 (1.05–2.86) | — | — | — | — | — | — | — |
| MCVB | 7.56 (0.67–85.35) | — | — | — | — | — | — | — |
| MCVC | 18.61 (4.32–80.21) | — | — | — | — | — | — | — |
| MCV conjugate | — | 1.36 (1.26–1.48) | 1.05 (0.99–1.11) | 1.55 (1.04–2.31) | — | — | — | — |
| Hep A | — | — | — | — | — | — | — | — |
| Hep B | 1.33 (1.01–1.73) | 1.19 (0.62–2.29) | — | 0.62 (0.40–0.95) | — | 1.57 (1.45–1.70) | — | — |
| Varicella | 3.59 (0.15–87.20) | — | — | — | — | — | — | — |
| MMR | 0.95 (0.91–0.99) | 2.26 (1.23–4.13) | 1.37 (1.04–1.81) | — | — | — | — | — |
| Polio | — | 1.45 (0.88–2.39) | — | — | — | — | — | — |
| Secondary | ||||||||
| Overall series completion | 1.62 (1.17–2.24) | 1.24 (1.14–1.35) | 1.05 (0.95–1.16) | 1.80 (0.96–3.39) | 2.44 (1.80–3.32) | 1.91 (1.69–2.16) | 1.84 (1.20–2.80) | 0.95 (0.82–1.11) |
| HPV | 2.30 (1.46–3.62) | 1.26 (1.15–1.38) | 1.05 (0.95–1.16) | 1.76 (0.85–3.66) | 2.44 (1.80–3.32) | — | 1.84 (1.20–2.80) | 0.95 (0.82–1.11) |
| Hep B | 1.09 (0.93–1.28) | 1.10 (0.99–1.24) | — | 1.94 (1.28–2.93) | — | 1.91 (1.69–2.16) | — | — |
| MOV | 0.93 (0.89–0.98) | 0.95 (0.93–0.98) | 0.90 (0.81–1.01) | — | — | 0.80 (0.78–0.82) | — | 0.92 (0.86–0.99) |
Hep, hepatitis; —, not applicable.
Summary of the Effect Estimates and 95% CIs From the Subgroup Analyses Based on Age and Setting
| Outcomes | Education, 34 studies | Reminders, 32 studies | Provider, 20 studies | School-Based Clinics, 13 studies | Financial Incentives, 4 studies | Policy and Legislative, 8 studies | Multicomponent, 3 studies | Multilevel, 6 studies |
|---|---|---|---|---|---|---|---|---|
| Overall vaccination coverage | 1.19 (1.12–1.26) | 1.15 (1.11–1.18) | 1.13 (1.07–1.19) | 1.31 (1.18–1.45) | 1.67 (1.40–1.99) | 1.19 (1.04–1.36) | 1.05 (0.90–1.23) | 1.25 (1.10–1.41) |
| By age, y | ||||||||
| 4–10 | 2.10 (1.03–4.31) | 1.57 (1.20–2.06) | 1.22 (1.10–1.35) | 1.22 (1.12–1.33) | — | 1.01 (0.99–1.03) | — | 1.08 (1.07–1.08) |
| 11–18 | 1.12 (1.06–1.19) | 1.14 (1.11–1.18) | 1.12 (1.06–1.19) | 1.38 (1.19–1.61) | — | 1.31 (0.93–1.85) | — | 1.29 (1.08–1.55) |
| ≥19 | 1.45 (0.82–2.56) | 1.02 (0.86–1.21) | — | — | — | — | — | — |
| By study setting | ||||||||
| School | 1.12 (1.01–1.24) | 1.05 (1.00–1.11) | — | — | — | — | — | — |
| Community | 1.37 (1.14–1.65) | 1.19 (1.15–1.23) | — | — | — | — | — | 2.01 (1.87–2.16) |
| Facility | 1.52 (1.19–1.93) | 1.16 (1.12–1.20) | — | — | — | — | — | 1.14 (1.01–1.30) |
| Web | 1.01 (0.96–1.06) | — | — | — | — | — | — | — |
| Outcomes | Education, 34 studies | Reminders, 32 studies | Provider, 20 studies | School-Based Clinics, 13 studies | Financial Incentives, 4 studies | Policy and Legislative, 8 studies | Multicomponent, 3 studies | Multilevel, 6 studies |
|---|---|---|---|---|---|---|---|---|
| Overall vaccination coverage | 1.19 (1.12–1.26) | 1.15 (1.11–1.18) | 1.13 (1.07–1.19) | 1.31 (1.18–1.45) | 1.67 (1.40–1.99) | 1.19 (1.04–1.36) | 1.05 (0.90–1.23) | 1.25 (1.10–1.41) |
| By age, y | ||||||||
| 4–10 | 2.10 (1.03–4.31) | 1.57 (1.20–2.06) | 1.22 (1.10–1.35) | 1.22 (1.12–1.33) | — | 1.01 (0.99–1.03) | — | 1.08 (1.07–1.08) |
| 11–18 | 1.12 (1.06–1.19) | 1.14 (1.11–1.18) | 1.12 (1.06–1.19) | 1.38 (1.19–1.61) | — | 1.31 (0.93–1.85) | — | 1.29 (1.08–1.55) |
| ≥19 | 1.45 (0.82–2.56) | 1.02 (0.86–1.21) | — | — | — | — | — | — |
| By study setting | ||||||||
| School | 1.12 (1.01–1.24) | 1.05 (1.00–1.11) | — | — | — | — | — | — |
| Community | 1.37 (1.14–1.65) | 1.19 (1.15–1.23) | — | — | — | — | — | 2.01 (1.87–2.16) |
| Facility | 1.52 (1.19–1.93) | 1.16 (1.12–1.20) | — | — | — | — | — | 1.14 (1.01–1.30) |
| Web | 1.01 (0.96–1.06) | — | — | — | — | — | — | — |
—, not reported/analyzed.
The majority of the included studies (n = 81) in our review were RCTs, followed by quasi-experimental trials (n = 38) and CBA (n = 1). Among the RCTs, we judged the majority to be at low risk of bias for random sequence generation and attrition bias and at high risk of bias for blinding because of the nature of interventions that were evaluated. Most studies were judged to be at unclear risk of bias for selective reporting because of lack of information on prepublished protocols and trial registrations. Among the quasi-experimental trials, we judged all to be at high risk of bias for sequence generation, allocation concealment, and prevention of knowledge of an intervention. The majority were judged to be at low risk of bias for attrition and contamination and unclear risk of bias for selective reporting. Overall, the quality of evidence for the outcomes ranged from high quality to very low quality. Significant reasons for downgrading the outcomes were study limitations, inconsistencies (high heterogeneity), and imprecision.
Our systematic review is a comprehensive evaluation of the existing data on the effectiveness of interventions to improve vaccination coverage among school-aged children and adolescents. We included all types of potential interventions aimed at improving vaccination coverage recommended by the World Health Organization for children and adolescents globally and in all settings. However, all the included studies in our review were performed in HICs. In concordance with recent reviews,137,138 the current findings reveal that providing vaccine information and education may improve coverage. Similarly, reminder and recall interventions, including telephone and autodialer calls, mailed letters, postcards, text messages, e-mails, and a combination of telephone calls, text messages, and e-mails, may also improve vaccination coverage.1,3 Our results were similar to the findings presented by Leung et al,139 who suggested that interventions delivered to providers through provider-specific education and reminders may improve vaccine uptake. Evidence suggests that provider recommendation is a key determinant in improving coverage. Our results also aligned with Abdullahi et al137 and Lee and Robinson140 in that financial incentives and the implementation of mandates in the form of vaccine-friendly policy and legislation have the potential to improve coverage. The difference between our review and these others is the targeted age groups wherein previous reviews included either children or adolescents (aged 11–19 years) and ours included all studies targeting both school-aged children and adolescents (ie, aged 5–19 years). We also evaluated the combined effect of all these interventions and performed a subgroup analysis based on age groups (children and adolescents).
Improving vaccination coverage among children and adolescents is a public health priority as vaccination is an effective and proven intervention to decrease the burden of preventable diseases. We suggest that education, reminders, interventions for providers, school-based clinics, financial incentives, policy and legislation, and multicomponent and multilevel interventions have the potential to improve overall vaccination coverage in this age group. However, because the existing evidence is from HICs, future studies are needed in LMIC settings to assess whether these interventions can be effective. Because we rated most of the outcomes in our review to be low-/very-low-quality evidence, more rigorous evaluations are needed, particularly in the domains of financial incentives, policy and legislative interventions, and multicomponent and multilevel interventions. We could not meta-analyze all the included studies because the outcomes were not reported consistently. Future studies should report consistent measures to allow for more streamlined comparisons of the effect of the various interventions evaluated.
Improving coverage requires an integrated approach that ranges from individual-level interventions to vaccine mandates at the national level. Interventions should be cost effective and feasible to ensure long-term benefit for children and adolescents.
Ms Siddiqui screened the search results and the retrieved articles against the inclusion criteria, appraised the quality of articles, extracted the data, completed the data analysis, drafted the initial manuscript, and reviewed and revised the final manuscript; Drs Padhani and Aliani screened the search results and the retrieved articles against the inclusion criteria, appraised the quality of articles, extracted the data, and completed the data analysis; Drs Salam and Das designed the study, coordinated and supervised the data collection, drafted the initial manuscript, and critically reviewed and modified the manuscript; Drs Lassi and Bhutta designed the study, coordinated and supervised the data collection, and critically reviewed and modified the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: This work was supported by a grant from the International Development Research Centre (#109010-001). The funder did not participate in the work. Core funding support was also provided by the SickKids Centre for Global Child Health in Toronto, and the Center of Excellence for Women & Child Health at the Aga Khan University in Karachi.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest relevant to this article to disclose.
- CBA
controlled before and after study
- CI
confidence interval
- DPT
diphtheria, pertussis, and tetanus
- HIC
high-income country
- HPV
human papillomavirus
- LMIC
low- and middle-income country
- MCV
meningococcal vaccine
- MMR
measles, mumps, and rubella
- MOV
missed opportunity for vaccination
- PRISMA
Preferred Reporting Items for Systematic Reviews and Meta-analyses
- RCT
randomized controlled trial
- RR
risk ratio
- Td
tetanus-diphtheria
- Tdap
tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis, adsorbed
Comments