
==== Front
eClinicalMedicine
EClinicalMedicine
eClinicalMedicine
2589-5370
Elsevier

S2589-5370(24)00397-3
10.1016/j.eclinm.2024.102818
102818
Articles
Effectiveness and implementation of decentralized, community- and primary care-based strategies in promoting hepatitis B testing uptake: a systematic review and meta-analysis
Kim Thanh Van abc
Pham Trang Ngoc Doan ad
Phan Paul a
Le Minh Huu Nhat aef
Le Quan a
Nguyen Phuong Thi g
Nguyen Ha Thi g
Nguyen Dan Xuan ah
Trang Binh a
Cao Chelsea a
Gurakar Ahmet c
Hoffmann Christopher J. choffmann@jhmi.edu
aij∗∗
Dao Doan Y ddoa1@jhmi.edu
ac∗
a Center of Excellence for Liver Disease in Viet Nam, Johns Hopkins School of Medicine, Baltimore, MD, USA
b Department of Epidemiology, Pham Ngoc Thach University of Medicine, Viet Nam
c Division of Gastroenterology and Hepatology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, MD, USA
d School of Public Health, University of Illinois at Chicago, Chicago, IL, USA
e International Ph.D. Program in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
f Research Center for Artificial Intelligence in Medicine, Taipei Medical University, Taipei, Taiwan
g University of Health Sciences, Vietnam National University Ho Chi Minh City (VNUHCM-UHS), Binh Duong, Viet Nam
h Boston University School of Public Health, Boston, MA, USA
i Division of Infectious Diseases, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, MD, USA
j Department of Health, Behavior, and Society, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
∗ Corresponding author. Division of Gastroenterology & Hepatology, Department of Medicine, Johns Hopkins School of Medicine, Ross 908, 600 N. Wolfe Street, Baltimore, MD 21287, USA. ddoa1@jhmi.edu
∗∗ Corresponding author. Division of Infectious Diseases, Department of Medicine, Johns Hopkins School of Medicine, CRB II 1M11, 1550 Orleans St, Baltimore, MD 21231, USA. choffmann@jhmi.edu
12 9 2024
10 2024
12 9 2024
76 10281813 7 2024
20 8 2024
21 8 2024
© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Summary

Background

Expanding chronic hepatitis B (CHB) testing through effective implementation strategies in primary- and community-care setting is crucial for elimination. Our study aimed to determine the effectiveness of all available strategies in the literature and evaluate their specifications and implementation outcomes, thereby informing future programming and policymaking.

Methods

We conducted a systematic review and meta-analysis (PROSPERO CRD42023455781), searching Scopus, Embase, PubMed, and CINAHL databases up to June 05, 2024, for randomized controlled trials investigating primary- and community-care-based implementation strategies to promote CHB testing. Studies were screened against a priori eligibility criteria, and their data were extracted using a standardized protocol if included. ROB-2 was used to assess the risk of bias. Implementation strategies' components were characterized using the Behavior Change Wheel (BCW) framework. Random-effect models were applied to pool the effectiveness estimate by strategy. Mixed-effect meta-regression was employed to investigate if effectiveness varied by the number of strategy's BCW components.

Findings

7146 unique records were identified. 25 studies were eligible for the review, contributing 130,598 participants. 19 studies were included in the meta-analysis. No studies were conducted in low-and-middle-income countries. Implementation outcomes were reported in only ten studies (40%). Community-based strategies included lay health workers-led education (Pooled Risk Difference = 27.9% [95% Confidence Interval = 3.4–52.4], I2 = 99.3%) or crowdsourced education on social media (3.1% [−2.2 to 8.4], 0.0%). Primary care-based strategies consisted of electronic alert system (8.4% [3.7–13.1], 95.0%) and healthcare providers-led education (HCPs, 62.5% [53.1–71.9], 27.5%). The number of BCW-framework-driven strategy components showed a significant dose-response relationship with effectiveness.

Interpretation

HCPs-led education stands out, and more enriched multicomponent strategies had better effectiveness. Future implementation strategies should consider critical contextual factors and policies to achieve a sustainable impact towards hepatitis B elimination targets.

Funding

Tran Dolch Post-Doctoral Fellowship in Hepatology, 10.13039/100012304 Johns Hopkins University School of Medicine , Baltimore MD, USA.

Keywords

Effectiveness
Implementation
Hepatitis B
Decentralization
Primary care
Testing
Elimination
==== Body
pmc Research in context

Evidence before this study

Developing implementation strategies to expand chronic hepatitis B (CHB) testing is pivotal to addressing the impediment to the CHB treatment and elimination. We conducted a systematic review using PubMed, Embase, CINAHL, and Scopus for evidence on implementation strategies to expand CHB testing to community and primary care settings. Our search encompassed studies in English published until June 05, 2024, using key terms related to the hepatitis B virus, testing, primary health care, community care, clinical trials, and intervention models. We found six relevant systematic reviews, which focused on the effectiveness of optimizing different steps in the viral hepatitis care continuum of some implementation strategies (e.g., mHealth intervention, health worker intervention) on specific populations (e.g., high-risk or migrant populations in high-income countries). However, a notable knowledge gap persists concerning the effectiveness coupled with implementation features of strategies aimed explicitly at CHB testing in primary and community care settings worldwide.

Added value of this study

This study advances the CHB elimination efforts with actionable insights on the real-world effectiveness and implementation challenges of various strategies to promote CHB testing in primary and community care settings. It highlights the considerable evidence gap in low- and middle-income countries (LMICs). Among the assessed strategies, CHB education delivered by healthcare providers, coupled with decentralized testing, emerged as particularly effective. Notably, the specification of implementation strategies is considered essential. Strategies enriched with a greater number of framework-driven mechanisms, such as those exemplified by Lay Health Workers-driven CHB education, demonstrated larger effect sizes on CHB testing. However, it is concerning that less than half of the studies adequately evaluated the implementation outcomes of these strategies. This underreporting may hinder the generalizability of the findings and applicability of the strategies in various settings to achieve sustainable impact.

Implications of all the available evidence

Hepatitis B stands at a crossroads between a future of continued inequal access to care and a path towards rapid expansion of program scale-up for testing and subsequent linkage to care. To catalyze for the latter, our study identifies effective strategies, such as CHB education delivered by healthcare providers and decentralized testing in primary care settings. These strategies have demonstrated real world effectiveness in increasing testing rates for early detection and treatment initiation of CHB. Another important implication is the recognition of the evidence gap in LMICs, highlighting the need for targeted research and interventions in these settings. Addressing this gap is essential for ensuring equitable access to CHB testing and care services worldwide. Lastly, strategies can be synergistically bundled and integrated with policy and health system-level factors to maximize impact and sustainability. Taken together, the findings of the study provide valuable insights for policymakers, healthcare providers, and researchers involved in CHB elimination efforts.

Introduction

Despite global efforts and World Health Organization and United Nations calls for progress, chronic hepatitis B (CHB) remains a substantial global burden. In 2019, there were close to 300 million people living with CHB, accounting for a worldwide prevalence of 3.84%.1 In the same year, 1,525,800 people developed acute hepatitis B, and 821,100 CHB-associated deaths occurred.1 Deaths from CHB annually exceed the number of deaths from HIV, TB, or malaria.2

In 2020, the global elimination targets of a 30% reduction in new infections and a 10% reduction in mortality were unmet, primarily due to gaps in the CHB care cascade.1 Diagnosis and treatment coverage in 2019 were estimated at 10% and 2%, respectively.1 For low and middle income countries (LMICs), the inequities in access to testing and treatment are even more amplified.3 Rapid acceleration and expansion of diagnosis and treatment are demanded in the later stage of the elimination endeavor from 2020 to 2030.

Testing is the essential first step in the CHB care cascade. Despite the availability of accurate tests and global calls to expand CHB testing, delivery progress has been insufficient over the past years.1 Effective strategies are needed and need to be used at scale for implementation to successfully bridge from policies to practice to reach the millions of people living with undiagnosed CHB. Decentralized testing in community- and primary-care settings have the potential to reduce the coverage gaps, as exemplified with robust evidence in HIV and HCV literature.4,5

Various strategies have been tested and implemented to optimize the general CHB care cascade. Some systematic reviews attempted to evaluate the effectiveness of specific strategies, such as mHealth or education, in some particular populations, such as high-risk or migrant people.6, 7, 8, 9, 10, 11 However, a notable knowledge gap persists in understanding the effectiveness of implementation strategies aimed explicitly at CHB testing in the context of primary and community care settings. Importantly, further synthesis of strategy specifications (i.e., components or mechanisms) and their implementation outcomes appears necessary to strengthen the contextualization of meta-analytic effectiveness findings.12

Taken together, we sought to fill this gap by conducting a systematic review and meta-analysis to 1) determine the effectiveness of community-based and primary care-based implementation strategies for promoting CHB testing in the general population, 2) evaluate the specifications of implementation strategies and its implementation outcomes, and 3) examine whether the enrichment of strategies’ components is related to an increase in their effectiveness. Through this study, we aim to provide actionable insights into real-world effectiveness and implementation challenges, thereby guiding future CHB testing programs toward achieving elimination goals.

Methods

Conceptual framework

Practically, we employed the Behavior Change Wheel (BCW) framework to characterize components within implementation strategies into strategy functions and sources of behavior to be included in qualitative and quantitative analyses.13 This application facilitated the deeper understanding of mechanisms behind each implementation strategy, which were central to the second and third aims of our study. Strategy functions represent the tools employed in implementation strategies to influence healthcare providers (HCPs) or patients in ordering or obtaining CHB testing (the “what”). Targeted sources of behavior refer to why HCPs or participants decide to order or obtain CHB testing (the “why”). We chose the BCW because of its high validity, as evidenced by cross-validation in different settings, and its transparent coding guideline.14 The definitions of BCW framework's components are provided in Table 1.Table 1 Definition of implementation strategy mechanisms delineated through Behavior Change Wheel framework.

Mechanism	Definitions	
Strategy functions: the tools employed in implementation strategies to influence HCPs or participants in ordering or obtaining CHB testing (the “what”)	
 Education	Increasing knowledge or understanding	
 Persuasion	Using communication to induce positive or negative feelings or stimulate action	
 Incentivization	Creating an expectation of reward and incentives.	
 Coercion	Creating an expectation of punishment or cost.	
 Training	Imparting knowledge and skills.	
 Restriction	Using rules or laws to reduce the opportunity to engage in the target behavior (or to increase the target behavior by reducing the opportunity to engage in competing behaviors)	
 Environmental restructuring	Changing the physical or social context	
 Modelling	Providing a role model or an example for people to aspire to or imitate.	
 Enablement	Increasing means/reducing barriers to increase capability (beyond education and training) or opportunity (beyond environmental restructuring.	
Sources of behavior: the reasons why HCPs or participants decide to order or obtain CHB testing (the “why”)	
 Physical Capability	Having the physical skills, strength, or stamina to order HBV tests (for health care workers) or obtain HBV tests.	
 Psychological Capability	Having the knowledge, psychological skills, strength, or stamina to order/obtain HBV testing.	
 Physical Opportunity	The environment allows or facilitates ordering/obtaining HBV testing in terms of time, triggers, resources, locations, physical barriers, etc.	
 Social Opportunity	Interpersonal influences, social cues, and cultural norms (from peers and networks) allow or facilitate ordering/obtaining HBV testing, etc.	
 Reflective Motivation	Self-conscious planning and evaluation (beliefs about what is good and bad) about HBV testing	
 Automatic Motivation	More intrinsic, such as emotional reactions, desires (wants & needs), impulses, inhibitions, drive states, and reflex responses regarding ordering and obtaining HBV testing.	

Eligibility criteria

We (TVK, TNDP, CJH, DYD) developed preliminary criteria for study selection. Based on the team's experience, publications considered most relevant to the preliminary criteria were identified. TVK piloted the preliminary criteria on those publications and looked for more relevant publications through reference lists and citation tracing. During this pilot phase, the eligibility criteria were finalized during whole-team discussions. The eligibility criteria were as follows.

Population

We retrieved studies conducted on populations who were recommended for CHB testing (e.g., people who were born or lived in high prevalence countries, institutionalized populations, vulnerable or marginalized populations, people who inject drugs, people who live with HIV, or males who have sex with males). These populations also included children and pregnant women. We excluded studies focused solely on blood donors, or cancer patients.

Settings

The primary focus of this study is on primary care and community care settings. Primary care settings encompass healthcare facilities that serve as the initial point of contact for individuals seeking care for general or specific health concerns within the health system. On the other hand, the community care setting extends beyond clinical environments and is primarily provided outside the traditional clinic setting.

Strategies

Our study defines implementation strategies as a compilation, or “bundle,” of implementation interventions that aim to promote CHB testing.15 We included studies using community-based and primary care-based implementation strategies. This included but was not restricted to health education (e.g., by lay health workers (LHWs) or healthcare providers (HCPs), different group sizes, and different types of delivery), active or passive navigation to testing services, financial incentivization to physicians or test takers, simplified testing (e.g., home-based or community-based testing, point-of-care (POC) testing, self-testing, rapid diagnostic testing, dried blood spot testing), application of eHealth or mHealth, electronic reminder or best practice alert systems, and health policies.

Comparison

Studies compared different implementation strategies or periods with and without the strategy of interest.

Outcomes

The primary outcome includes HBsAg testing uptake in patients (self-reporting or confirmed diagnosis by chart validation or proof of testing). When studies reported more than one measure, confirmed HBsAg testing was selected for analysis due to less risk of misclassification biases that affect self-reporting, including recall bias, social desirability bias, and others.

The secondary outcome is the uptake of tests for diagnosing hepatitis B immunity (i.e., anti-HBs antibody and anti-HBc antibody) and linkage to care such as hepatitis B vaccination and treatment evaluation (e.g., viral load testing or hepatologist consultation). Implementation outcomes such as acceptability, adoption, appropriateness, costs, feasibility, fidelity, penetration and sustainability were also collected when available.16

Time point/timing

Eligible studies published until the day of the final search were included. Also, no restrictions were made on timing of outcome assessment.

Study designs

Primary studies included experimental designs with a randomization component, specifically including randomized controlled trials and cluster randomized trials.

Information sources and search strategies

An information specialist translated the eligibility criteria into search terms relevant to each database (Scopus, Embase, Pubmed, CINAHL). Additionally, MeSH (Medical Subject Headings) terms were extracted from the preliminary publications using the Yale MeSH Analyzer and added to the search terms.17 The research team then reviewed and edited the search strategies. The included databases and the search strategies are shown in Supplemental Material 1. The final search was conducted on June 05, 2024.

Manual searches were conducted by tracing the reference lists of the included studies and relevant systematic reviews. We also traced and screened studies that cited the included studies and related studies using the “Cited by” and “Related articles” functions on PubMed and Google Scholar. As mentioned in detail in the next section, two independent reviewers conducted the study tracing and conformed to the similar voting and conflict resolution.

Selection process

We managed all search results on Covidence (https://www.covidence.org/), including screening, extraction, and quality assessment. Duplicate studies were excluded by built-in algorithms deployed by Covidence or manually by reviewers during the study.

First, two independent reviewers (TVK, PP, BT, CC, QL, HN, PN, DN, MHNL) screened the titles and abstracts against the eligibility criteria. Studies voted “included” or “maybe included” by consensus were carried to the next step, while studies voted “not included” by both reviewers were excluded. Discrepancies in voting were resolved by two-way discussions, or in case any conflicts remained, by seeking a third opinion and casting a third vote (TNDP, CJH, DYD). Next, full texts were uploaded onto Covidence. Two independent reviewers screened the full text against the eligibility criteria using a similar voting and conflict resolution system in full-text screening. The number of studies screened was assigned to TVK as the first reviewer and other members (PP, BT, CC, QL, HN, PN, DN, MHNL) as second reviewers.

Data extraction and risk of bias assessment

A data extraction form, which had been reviewed, piloted, and edited, was used to extract relevant information from the included studies. The form comprised data fields for study identification, methods, populations, implementation strategies, and outcomes. Details of the form are presented in Supplemental Material 2. Behavior Change Wheel and TIDieR checklist were used to guide the design of implementation strategy-related data fields.14,18 Other data fields were developed based on suggestions in the Cochrane handbook.19 When relevant information was unavailable in the texts, the fields were entered as “not reported.” The risk of bias assessment form was developed using the ROB-2 system for individual RCTs and its variant for CRTs.20

The data extraction and risk of bias assessment were conducted simultaneously. Two independent reviewers extracted data using the form. Each data field was compared head-to-head between the two reviewers. The same voting and conflict resolution were performed to achieve consensus for each data field.

We used the BCW framework to characterize the mechanisms of the implementation strategies. The coding system for strategy functions and targeted sources of behavior follows the steps described by Michie et al.‘s guidebook.14 Extracted implementation characteristics were thematically matched with the behavioral change techniques defined in Step 7 of the guidebook. After that, we linked those behavioral techniques to corresponding behavior targets and implementation strategy functions. All team members underwent group training and were familiar with this coding process (Supplemental Material 3).

Synthesis methods and effect measure

The outcome measure was the proportion of CHB testing among those who were recommended for CHB testing. We have used the term CHB testing rather than HBsAg or HBV testing to indicate the role of HBsAg testing in identifying individuals with surface antigenemia for further assessment and possible CHB treatment. The effect measure of effectiveness was the risk difference (RD), which represents the absolute incremental increase in the proportion tested. When possible, effect estimates from individual- and cluster-randomized trials employing the same implementation strategies were pooled in the same meta-analysis. The grouping by implementation strategy was specified a priori based on prior literature and overall descriptions by the authors of the included studies.6, 7, 8, 9, 10, 11 As we anticipated considerable between-study heterogeneity, a random-effects model was used to pool effect sizes. The Higgins & Thompson's I2 statistic was used to quantify the heterogeneity.21,22 The synthesis results were presented in a forest plot.

We conducted a mixed-effects meta-regression analysis to investigate if the enrichment of strategies' mechanisms is related to an increase in strategies' effectiveness, or in other words, whether the strategy effectiveness increased by the number of BCW components (i.e., the number of targeted sources of behavior or the number of strategy functions involved) in multicomponent strategies.23 Publication bias would be assessed using funnel plots and Egger's test, and the corresponding adjustment would be performed if the heterogeneity was small to moderate (<25%) to avoid adding further bias.24

For individual RCTs, RDs and their 95% confidence interval were computed using extracted raw values. Adjusted effect estimates reported in the RCTs would be used if multiple adjustment analyses were mentioned in their a priori analysis plans. For cluster-randomized trials, if the authors calculated the effect estimates using cluster-effect-adjusted univariate or multivariate analyses (e.g., design-based model, mixed-effects models, or generalized estimating equations), their effect estimates were extracted directly. Effect estimates presented in relative risks (RR) or odds ratios were converted to RDs (conversion formulas shown in the Supplemental Material 4). In one study, a reported covariate- and cluster-adjusted incidence rate ratio was assumed to approximate RR and then converted to RD.25

Some cluster-randomized trials did not report accounting for survey design in the outcome analysis, potentially leading to overestimated precisions. We applied an approximately correct analysis to these trials, by which their observed sample sizes were reduced to effective sample size26 (Supplemental Material 4). Intra-cluster correlation was assumed to be 0.03, which was based on one of the included CRT.25 The meta-analysis did not include studies that only reported the changes between baseline and study.

The alpha significance level was 0.05. All data were processed and analyzed using R version 4.3.2 (2023-10-31). The protocol had been registered on PROSPERO (CRD42023455781).

Role of the funding source

The sponsor of TVK's postdoctoral fellowship had no involvement in data collection, analysis, interpretation, writing of the manuscript and the decision to submit.

Results

Study selection

Our search strategy identified 8652 results, of which 7146 remained after removing duplicates (Fig. 1). After screening titles and abstracts, 25 studies were included in the review, none of which involved children or pregnant women (Table 2).25,27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 Each study's characteristics are presented in Supplemental Table S1. The study by Fung et al. was excluded due to combining self-reported CHB screening uptake and intent to CHB screening.51 Eight studies on pregnant women were excluded for non-randomized design, no CHB testing outcome, or already having positive HBsAg results as an entry criteria.52, 53, 54, 55, 56, 57, 58, 59 One pediatric study was excluded because of its non-randomized design and the setting of tertiary care oncology.60 Studies excluded at full-text screening stage are shown in Supplemental Table S2. Nineteen studies were included in five meta-analyses and two meta-regression because their implementation strategies and outcomes could be directly compared to at least one other included study.25,27,28,30, 31, 32,34,35,37,38,40, 41, 42,44,46,47,50,61Fig. 1 The flow of study inclusion.

Table 2 Trials of implementation strategies meeting inclusion criteria.

Study	Study design/setting	Study population	Implementation strategy arm vs. control arma	N of strategy arm vs. control arm	%Female	Mean age (SD)	Risk difference (95% CI)	
Ahmadi 2019	CRT/Community	People with substance-related disorders in Iran	CHB education by LHWs vs. Usual practice	50 vs. 50	100%b	28.6 (6.5)	34.0% (12.6–55.4)	
Bastani 2015	CRT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	543 vs. 580	65.3%	45.5 (12.5)	17.1% (7–31.7)	
Bottero 2015	RCT/Primary care	Migrant patients visiting PCCs in France	Point-of-care CHB testing vs. Usual practice	162 vs. 162	38.0%	38 (12.2)	27.2% (19.9–34.5)	
Chak 2018	RCT/Primary care	HCPs (to order CHB tests for Asian patients in the US)	CHB testing electronic reminder vs. Usual practice	1542 vs. 1568	53.3%	42.8 (14.7)	9.0% (6.6–11.3)	
Chak 2020	RCT/Primary care	HCPs (to order CHB tests for Asian patients in the US)	CHB testing electronic reminder vs. Usual practice	2599 vs. 2590	53.3%	51.5 (20.8)	2.9% (2–3.9)	
ChenMSJr 2013	RCT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	130 vs. 130	59.6%	NA	10.8% (2.5–19.1)	
Fitzpatrick 2019	RCT/Community	Males who have sex with males in China	Crowdsourced CHB education vs. Usual practice	280 vs. 276	0.0%b	25.5 (7)	2.3% (−2.9 to 10.9)	
Flanagan 2019	CRT/Primary care	HCPs/Migrant patients visiting PCCs in the UK	Financial incentivization vs. Attention control	58,512 vs. 31,738	51.8%	NA	4.6% (0.5–16.3)	
Hsu 2013	RCT/Primary care	HCPs/Asian patients visiting PCCs in the US	CHB testing electronic reminder vs. Usual practice	88 vs. 87	62.9%	40.8 (12.3)	34.1% (24.1–44.1)	
Juon 2014	CRT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	220 vs. 226	NA	45.1 (13.5)	23.6% (10.1–40.4)	
Khalili 2022	CRT/Primary care	HCPs/Asian patients visiting PCCs in the US	CHB education on digital platforms vs. Attention control	270 vs. 182	63.9%	56.8 (16.9)	30.7% (15.5–48.6)	
Ma 2017	CRT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	1131 vs. 1206	56.2%	55.4 (15.1)	78.1% (73.7–82.5)	
Ma 2018	CRT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	972 vs. 862	58.0%	51.6 (13.5)	91.3% (80.4–93.3)	
Richens 2010	RCT/Primary care	Patients with sexual health problems in the UK	Computer-assisted self-interviewing vs. Attention control	795 vs. 779	NA	NA	0.6% (−3.1 to 4.2)	
Rosenberg 2010	RCT/Primary care	Patients with mental health problems in the US	CHB education by HCPs with decentr. testing vs. Usual practice	95 vs. 93	47.9%	46.5 (8.9)	57.6% (46–69.1)	
Sahajian 2011	CRT/Community	Shelter residents in France	CHB education by HCPs with decentr. testing vs. Usual practice	211 vs. 793	NA	NA	67.2% (56.1–78.3)	
Sequeira-Aymar 2022	CRT/Primary care	HCPs (to order CHB tests for migrant patients in Spain)	CHB testing electronic reminder vs. Usual practice	3445 vs. 2784	NA	NA	2.2% (0.6–4.1)	
Shireman 2020	CRT/Community	Community health workers in the US	Training modalities of LHWs vs. Usual practice	481 vs. 523	53.5%	46.2 (14.4)	27.9% (18.9–36.8)	
Taylor 2009	RCT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	231 vs. 229	37.2%	NA	2.6% (−0.3 to 5.5)	
Taylor 2011	CRT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	95 vs. 123	55.0%	NA	4.6% (−2.6 to 11.8)	
Taylor 2013	RCT/Community	Asian community members in the US	CHB education by LHWs vs. Attention control	125 vs. 125	50.0%	NA	8.8% (3.7–13.9)	
VanderVeen 2014	RCT/Community	Turkish community members in the Netherlands	CHB education on digital platforms vs. Attention control	472 vs. 496	50.2%	33.6 (5.2)	−1.5% (−9.0 to 5.8)	
Wong 2022	RCT/Primary care	Patients visiting PCCs in China	Crowdsourced CHB education vs. Usual practice	376 vs. 376	57.3%	42.7 (11.2)	4.3% (−4.7 to 11.9)	
Xiao 2021	RCT/Community	Asian community members in Australia	CHB education by LHWs vs. Attention control	26 vs. 28	NA	NA	8.0% (−6.1 to 22)	
Ye 2023	CRT/Community	Males who have sex with males in China	Financial incentivization vs. Attention control	160 vs. 162	0.0%b	29.5 (12.6)	35.2% (24.1–46.3)	
Abbreviation: CRT, Clustered Randomized Trial; RCT, Randomized Controlled Trial; CHB, Chronic hepatitis B; LHWs, Lay health workers; HCPs, Healthcare providers; SD, Standard deviation; PCCs, Primary care clinics; decentr. testing, decentralized testing; NA, Not available.

a Attention controls are controls that received an intervention designed to mimic the attention and contact time of the strategy group, without providing the content directly related to the outcome of hepatitis B testing; Usual practice refers to the standard practice that participants would normally receive outside of the study.

b These studies either recruited only men or women.

Study characteristics

The characteristics of the studies are summarized in Table 3. The 25 studies contributed a total of 130,598 participants, with 34,925 participants included in all the meta-analyses. Ten studies (40.0%) were conducted at primary care clinics, while 12 (48.0%) were clustered randomized trials. Four studies (16.0%) were conducted in upper-middle-income countries, and 21 (84%) were from high-income countries. 56.0% (14/25) and 16.0% (4/25) of studies worked solely with migrant Asian populations and those living in Asian territories, respectively. The most common implementation strategy to promote CHB testing in community care settings was health education by laypeople (10 studies, 67.0%). In contrast, in primary care settings, it was efforts to increase CHB testing by clinicians through electronic reminders (4 studies, 40.0%). Five out of 25 studies had strategies aimed at clinicians. Ten studies (40.0%) reported implementation outcomes such as fidelity, adoption, contamination, or cost-effectiveness. Nine studies (36.0%) were considered at high risk of bias (Table 3 and Supplemental Figures S1 & S2).Table 3 Summary estimates of studies’ characteristics by setting.

Characteristic	Overall, n (col %)	Settings	
Community care, n (col %)	Primary care, n (col %)	
Total	25 (100%)	15 (100%)	10 (100%)	
Study design				
 Clustered Randomized Trials	12 (48.0%)	9 (60.0%)	3 (30.0%)	
 Randomized Controlled Trials	13 (52.0%)	6 (40.0%)	7 (70.0%)	
Country				
 China	3 (12.0%)	2 (13.3%)	1 (10.0%)	
 Iran	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Australia	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Canada	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 France	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 Netherlands	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Spain	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 United Kingdom	2 (8.0%)	0 (0.0%)	2 (20.0%)	
 United States and Canadaa	13 (52.0%)	8 (53.3%)	5 (50.0%)	
Race(s) in the study sample				
 Multi-racial	7 (28.0%)	2 (13.3%)	5 (50.0%)	
 Only Asian (migrant)	14 (56.0%)	10 (66.7%)	4 (40.0%)	
 Only Asian (native)	4 (16.0%)	3 (20.0%)	1 (10.0%)	
Implementation strategies				
 Computer-assisted interviewing of high-risk people	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Crowdsourced CHB education on social media	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 CHB electronic reminder	4 (16.0%)	0 (0.0%)	4 (40.0%)	
 Financial incentivization	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 CHB education by HCWs with decentralized testing	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 CHB education by LHWs	10 (40.0%)	10 (66.7%)	0 (0.0%)	
 CHB education on digital platforms	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 Point-of-care CHB testing	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Training modalities of LHWs	1 (4.0%)	1 (6.7%)	0 (0.0%)	
Actors of implementation				
 Administrators	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 Community members	11 (44.0%)	10 (66.7%)	1 (10.0%)	
 Digital applications	3 (12.0%)	1 (6.7%)	2 (20.0%)	
 Electronic health record system	3 (12.0%)	0 (0.0%)	3 (30.0%)	
 Health professionals	4 (16.0%)	1 (6.7%)	3 (30.0%)	
 Peer members	2 (8.0%)	2 (13.3%)	0 (0.0%)	
Action targeting on healthcare providers				
 Healthcare Providers	5 (20.0%)	0 (0.0%)	5 (50.0%)	
 No	20 (80.0%)	15 (100%)	5 (50.0%)	
Action targeting on patients or community members				
 Community health workers	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 General community members	10 (40.0%)	10 (66.7%)	0 (0.0%)	
 Males who have sex with males	2 (8.0%)	2 (13.3%)	0 (0.0%)	
 Patients visiting primary care clinics	5 (20.0%)	0 (0.0%)	5 (50.0%)	
 Patients with mental health problems	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Patients with sexual health problems	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Patients with substance-related disorders	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Shelter residents	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 No	3 (12.0%)	0 (0.0%)	3 (30.0%)	
Mode of delivery				
 Distanceb	3 (12.0%)	2 (13.3%)	1 (10.0%)	
 Electronic health record system prompts	4 (16.0%)	0 (0.0%)	4 (40.0%)	
 Face-to-face, group	10 (40.0%)	10 (66.7%)	0 (0.0%)	
 Face-to-face, individual	8 (32.0%)	3 (20.0%)	5 (50.0%)	
Location of implementation				
 Churches	3 (12.0%)	3 (20.0%)	0 (0.0%)	
 Community-based organizations	5 (20.0%)	5 (33.3%)	0 (0.0%)	
 Drop-in centers	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Homes	3 (12.0%)	3 (20.0%)	0 (0.0%)	
 Mental health clinics	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Primary care clinics	7 (28.0%)	0 (0.0%)	7 (70.0%)	
 Sexual health clinics	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Shelters	1 (4.0%)	1 (6.7%)	0 (0.0%)	
 Social media	2 (8.0%)	1 (6.7%)	1 (10.0%)	
 Websites	1 (4.0%)	1 (6.7%)	0 (0.0%)	
Reporting implementation outcomes				
 Yes	10 (40.0%)	6 (40.0%)	4 (40.0%)	
 No	15 (60.0%)	9 (60.0%)	6 (60.0%)	
Time at outcome assessment				
 After 12 months	3 (12.0%)	1 (6.7%)	2 (20.0%)	
 After 6 months	10 (40.0%)	9 (60.0%)	1 (10.0%)	
 After 3 months	3 (12.0%)	1 (6.7%)	2 (20.0%)	
 After 1 month	4 (16.0%)	2 (13.3%)	2 (20.0%)	
 Within 1 week	5 (20.0%)	2 (13.3%)	3 (30.0%)	
Type of outcome report				
 Self-reporting testing	6 (24.0%)	5 (33.3%)	1 (10.0%)	
 Confirmed testing	19 (76.0%)	10 (66.7%)	9 (90.0%)	
Overall risk of bias				
 High	9 (36.0%)	8 (53.3%)	1 (10.0%)	
 Low	6 (24.0%)	1 (6.7%)	5 (50.0%)	
 Some concerns	10 (40.0%)	6 (40.0%)	4 (40.0%)	
Abbreviation: CHB, Chronic hepatitis B; HCWs, Health care workers; LHWs, Lay health workers.

a One study was conducted in both the United States and Canada.

b In these studies, the educational materials were delivered through digital or printed media.

Publication bias was performed for the implementation strategy of CHB education by lay health workers. The funnel plot shows a slight asymmetry (Supplemental Figure S3); however, Egger's regression test failed to detect significant asymmetry, with a p-value of 0.430. Publication bias was not performed for other strategies due to the small number of studies and/or large heterogeneity (I2 > 75.0%) in each meta-analysis.24

Table 4 shows the summary statistics for strategies functions and sources of behavior characterized by the BCW framework. In the community care setting, most implementation strategies involved education and persuasion as strategy functions (84.0% and 76.0%, respectively) and psychological capability and reflective motivation as targeted sources of behavior (84.0% and 84.0%, respectively). In contrast, environmental restructuring to increase physical opportunity had a central role in the primary care setting (100% and 90.0%, respectively). However, no strategies involved the function of coercion or restriction. No policy-level factors were also involved in all strategies. BCW's components in each study are presented in Table 5.Table 4 Summary estimates for Behavior Change Wheel's components of implementation strategies by setting.

Characteristic	Overall, n (% out of total)	Settings	
Community care, n (% out of total)	Primary care, n (% out of total)	
Total	25 (100%)	15 (100%)	10 (100%)	
Sources of behavior				
 Physical Capability	1 (4.0%)	0 (0.0%)	1 (10.0%)	
 Psychological Capability	21 (84.0%)	15 (100%)	6 (60.0%)	
 Physical Opportunity	14 (56.0%)	5 (33.3%)	9 (90.0%)	
 Social Opportunity	11 (44.0%)	10 (66.7%)	1 (10.0%)	
 Reflective Motivation	21 (84.0%)	15 (100%)	6 (60.0%)	
 Automatic Motivation	13 (52.0%)	7 (47.7%)	6 (60.0%)	
Strategy functions				
 Education	21 (84.0%)	15 (100%)	6 (60.0%)	
 Persuasion	19 (76.0%)	15 (100%)	4 (40.0%)	
 Incentivization	3 (12.0%)	1 (6.7%)	2 (20.0%)	
 Coercion	0 (0.0%)	0 (0.0%)	0 (0.0%)	
 Training	6 (24.0%)	1 (6.7%)	5 (50.0%)	
 Restriction	0 (0.0%)	0 (0.0%)	0 (0.0%)	
 Environmental Restructuring	16 (64.0%)	6 (40.0%)	10 (100%)	
 Modelling	4 (16.0%)	3 (20.0%)	1 (10.0%)	
 Enablement	14 (56.0%)	10 (66.7%)	4 (40.0%)	
Note: One study may have more than one behavior targets or strategy functions, so the column numbers or proportion may not be mutually exclusive.

Table 5 Characterization of Behavior Change Wheel's components in each study.

Study ID	Implementation strategy	Strategy functionsa	Sources of behavior targets	
Education	Persuasion	Training	Modelling	Enablement	Incentivization	Environment
Restructure	Physical
Capability	Psycho
Capability	Physical
Opportunity	Social
Opportunity	Reflective
Motivation	Automatic
Motivation	
Fitzpatrick 2019	Crowdsourced education	X	X			X		X		X		X	X		
Wong 2022	Crowdsourced education	X	X			X		X		X		X	X		
Ye 2023	Financial incentivization	X	X			X	X	X		X		X	X	X	
Flanagan 2019	Financial incentivization	X				X	X	X		X	X		X		
Sahajian 2011	Education by HCPs	X	X			X		X		X	X	X	X	X	
Rosenberg 2010	Education by HCPs	X	X			X	X	X		X	X		X	X	
Bastani 2015	Education by LHWs	X	X		X	X		X		X	X	X	X	X	
Ahmadi 2019	Education by LHWs	X	X		X	X				X	X	X	X	X	
Ma 2018	Education by LHWs	X	X			X		X		X	X	X	X		
Ma 2017	Education by LHWs	X	X			X		X		X		X	X		
Juon 2014	Education by LHWs	X	X		X					X		X	X		
Taylor 2013	Education by LHWs	X	X			X				X		X	X		
ChenMSJr 2013	Education by LHWs	X	X			X				X			X	X	
Xiao 2021	Education by LHWs	X	X							X			X		
Taylor 2011	Education by LHWs	X	X							X			X		
Taylor 2009	Education by LHWs	X	X							X			X		
VanderVeen 2014	Digitalized education	X	X			X				X			X	X	
Khalili 2022	Digitalized education	X	X	X	X	X		X		X	X		X	X	
Sequeira-Aymar 2022	Electronic reminder	X	X	X				X	X	X	X		X	X	
Hsu 2013	Electronic reminder	X		X				X		X	X		X	X	
Chak 2020	Electronic reminder			X				X			X			X	
Chak 2018	Electronic reminder			X				X			X			X	
Bottero 2015	Point-of-care testing							X			X				
Shireman 2020	Training modalities	X	X	X						X	X	X	X	X	
Richens 2010	CAPI							X			X				
Abbreviation: HCPs, Healthcare providers; LHWs, Lay health workers; CAPI, Computer-assisted Personal Interviewing.

a No studies had Coercion and Restriction in their implementation strategies.

CHB education by lay health workers (LHWs)

Ten studies, totaling 7451 participants, investigated the effectiveness of the strategy in which LHWs without clinical training provided CHB- or liver cancer-related education in the community.27,28,32,35,37,38,44,46,47,61 Three studies had some concerns about bias, while the other seven showed high risks of bias. When pooled together in the meta-analysis, the LHWs-driven implementation strategy showed an increase of 27.9% (95% CI, 3.4–52.4%) in CHB testing uptake with large heterogeneity (I2 = 99.3%) compared with control arms (Fig. 2). Most studies were conducted on migrant Asian populations, except for the study by Ahmadi et al., which focused on a high-risk population (patients with substance-related disorders) living in Iran.27Fig. 2 Forest plots of individual and pooled effect estimates in promoting CHB testing of implementation strategies. Moderate to large heterogeneity but with consistency of direction toward benefits can be observed for CHB education by lay health workers, CHB testing electronic reminder, CHB education by healthcare providers coupled with decentralized testing. Abbreviation: CHB, Chronic hepatitis B; RD, Risk Difference.

All ten studies included education and persuasion as functions in their strategy, hence increasing patients' psychological capability to change CHB testing behavior (Table 5). Three studies restructured the environments to optimize the physical and social opportunities.28,37,38 Ma et al., in 2017 and 2018 employed navigation services (e.g., transportation, language translation, scheduling appointments) and a community participatory approach (e.g., engaging community members in planning, development, and implementation) to alleviate any physical or social barriers for the target sample.37,38

Implementation outcomes were reported in three studies, which showed suboptimal participants’ engagement in receiving the implementation or follow-up. The home-based education by Taylor et al., in 2009 only reached 63.0% of the participants in the strategy arm, and only as low as 34.0% of them were eventually exposed to one of the education materials.47 A similar implementation by Taylor et al., in 2013 on Cambodian Americans delivered complete home-based education to 79.2% of the participants.44 The other participants refused the education completely (9.6%) or partially (4.0%) or did not commit to follow-up (4.0%).44 Group-based CHB education at Vietnamese American community-based organizations reached 100% of participants; however, only 52.0% received the testing navigation component.37

One study further reported secondary outcomes. A multicomponent strategy by Ma et al., in 2018 effectively increased both CHB testing uptake and hepatitis B vaccination.38 92.8% and 84.0% of people who tested negative for CHB initiated and completed vaccination series in the implementation strategy arm, while the figures in the control arm were 29.4% and 17.6%, respectively.38

Meta-regression

Mixed-effect meta-regression analyses were conducted to investigate the large heterogeneity and to associate the number of implementation strategy mechanisms with effectiveness in promoting CHB testing uptake. This analysis was performed only for the LHWs-driven strategy thanks to the acceptable number of ten studies.23 The number of targeted sources of behavior showed a dose–response relationship with CHB testing uptake, with a significant increase in the effectiveness of 23.8% (95% CI 14.6–33.0, R2 = 84.6%) per additional source of behavior targeted. Likewise, for every additional function included in the implementation strategy, effectiveness significantly increased by 17.9% (95% CI 1.6–34.1%, R2 = 59.2). For better visualization of the incremental change, the random-effect risk differences by sources of behavior or strategy functions were presented in Fig. 3.Fig. 3 A whisker plot showing random-effect risk difference and its confidence interval sub-grouped by the number of strategy mechanisms (sources of behavior or strategy functions), with number of studies in each subgroup. Meta-regression shows that a significant increase in the effectiveness of 23.8% (95% CI 14.6–33.0, R2 = 84.6%) was observed per additional source of behavior targeted. Likewise, for every additional function included in the implementation strategy, effectiveness significantly increased by 17.9% (95% CI 1.6–34.1%, R2 = 59.2).

Sensitivity analyses

Subgroup analyses based on study characteristics were conducted where possible, as shown in Supplemental Table S3. No significant differences were detected between subgroups. The I2 value for heterogeneity remained over 50% in most subgroups, except for studies implementing only educational activities, where the I2 value was 0.0%.

Training modalities for LHWs

One study with some concerns for bias investigated different training modalities for LHWs to conduct community-based health education. Shireman et al. found that in-person training significantly increased CHB testing uptake among church-goers (RD = 27.9% [95% CI 18.9–36.8%]) than online training with similar contents.43 Also, the proportion of hepatitis B vaccination in the in-person training arm was higher than in the online training arm (17.0% vs. 5.9%).43 Despite the difference in effectiveness in increasing CHB testing uptake and hepatitis B vaccination, the total training costs were comparable between the two arms.43

CHB education by HCPs coupled with decentralized testing

A total of 1912 participants were included in two trials with some concerns of bias applying the model in which HCPs provided CHB education and decentralized testing.40,41 The meta-analysis shows the pooled RD of 62.5% (95% CI 53.1–71.9%) with small heterogeneity of I2 being 27.5% (Fig. 2). An outreach strategy for underprivileged people living in shelters in France included individual consultation and group information regarding CHB and hepatitis C, followed by decentralized testing.41 This model of patient group information helps to enable patients’ psychological capabilities and restructure the social norms regarding CHB testing (Table 5). On the other hand, Rosenberg et al. conducted a bundle of care at primary mental health clinics, including individual education and pretest counseling about infectious diseases in general, including hepatitis B, followed by decentralized testing and further incentivized by immunization and risk reduction education.40 The care bundle triggered patients' automatic motivation, making them desire to obtain CHB testing (Table 5).

Digitalized CHB education

The effectiveness of health education delivered through digital applications was investigated in two studies with some concerns for bias, adding up to 1852 participants.36,48 However, these two studies’ effect estimates were not pooled due to inconsistent settings and comparators. VanderVeen et al., 2014 deployed CHB education websites to educate people in the community and observed an RD of −1.5% (95% CI −9.0 to 5.8%).48 Two arms received the web-based education; the contrast of interest was between the culture- and behavior-adapted CHB contents in one arm and generic CHB-related information in the other arm.48

On the contrary, Khalli et al., in 2022 tested the implementation of an iPad-based mobile application to educate patients and facilitate the patient-provider discussion regarding hepatitis testing in primary care clinics.36 Both patients and HCPs were beneficiaries. This implementation model engaged up to six strategy functions and four behavior targets, as shown in Table 5. The CHB testing uptake was increased by 30.7% (95% CI 15.5–48.6%) compared to usual care.36 Other implementation outcomes were also reported. 70.4% of patients started discussing with providers in the mobile app arm, while only 16.5% did in usual care. Additionally, 51.1% of providers recommended CHB testing, while only 13.2% recommended it in usual care.36

Crowdsourced CHB education on social media

Two studies with 1308 participants and a high risk of bias tested an approach in which CHB educational content was crowdsourced and delivered on social media.33,49 Meta-analysis in Fig. 2 shows a minimal increase in CHB testing proportion by 3.1% (95% CI −2.2 to 8.4%, I2 = 0.0%). This innovative virtual community-based outreach enhanced the social opportunities for CHB knowledge and testing. Implementation-wise, challenges included the fidelity and contamination of crowdsourced education on social media. In Wong's study in 2022, 61.4% of participants received all educational materials, and 26.9% did not see any of them during the study period.49 Fitzpatrick et al., in 2019 noted that 27.9% of participants in the crowdsourcing arm saw no educational materials, and 52.9% shared crowdsourced materials with others, while 9.0% of men in control were exposed to the crowdsourced materials.

CHB testing electronic reminder

Four studies were conducted on 24,254 patients to assess the effectiveness of electronic reminders in promoting CHB testing uptake in primary care clinic settings.30,31,34,42 One study demonstrates some concerns for bias, while others at low risk. The pooled RD was 8.4% (95% CI 3.7–13.1%) with large heterogeneity (I2 = 95.0%) in Fig. 2. In these studies, electronic reminder pop-ups identified patients at higher risk of hepatitis B (i.e., Asian and Pacific Islanders, and people from countries with CHB prevalence >2%). The implementation by Chak et al., 2018 and 2020 only included electronic reminders,30,31 while Hsu et al. and Sequeira-Aymar 2022 added an education persuasion component, which improved the psychological capabilities and reflective motivation of HCPs (Table 5).34,42 Of significant note, the tool used in the study by Sequeira-Aymar et al. was for multiple infections, one of which is hepatitis B.42

Concerning implementation outcomes, only Hsu et al. reported on CHB test orders by healthcare providers.34 In the electronic reminder arm, where all encounters were prompted with a CHB alert, 53.7% (36/67) of patients were ordered CHB tests, and 83.3% (30/36) completed the testing order. In the usual care arm, 1.6% (1/63) were ordered, and none completed the order.

Financial incentivization

Two studies with a low risk of bias implemented financial incentivization to increase CHB testing. Meta-analysis was not done due to the difference in the nature of the implementations. Flanagan et al. offered pay-for-performance financial incentives to primary care providers to encourage as many testing uptakes as possible in their clinics.25 This strategy led to an increase of 4.6% (95% CI 0.5–16.3%) in CHB testing uptake and was deemed cost-effective at a willingness-to-pay thresholds above £8540 per QALY. In contrast, Zhang et al. targeted directly on the community with a one-off community-driven and pay-it-forward incentives program at community-based organizations.50 Every person was offered a free test with community-generated messages and then asked if they would like to donate money to support others to receive free testing. The RD was 35.2% (95% CI 24.1–46.3%). The financial cost is $69 per case of viral hepatitis identified.

POC testing

One study with a low risk of bias compared the effectiveness of fingerstick POC HBsAg testing with a 30-min turnaround time to standard venipuncture-based testing with a one-week turnaround time.29 This strategy, which was conducted in the setting of primary care clinics, increased the CHB testing uptake by 27.2% (95% CI 19.9–34.5%) compared with the standard care. Furthermore, linkage-to-care rates were also improved, with 90.0% of infected patients in the POC arm compared to 83.3% in the standard arm.

Computer-assisted interviewing of high-risk people

One study with some concerns for bias implemented computer-assisted self-interview (CASI) to promote CHB testing through mitigating social desirability in reporting activities at risk for CHB infection.39 Patients who visited sexual health clinics would self-report their behaviors to a computer before moving on to consultation with care providers. Richens et al. found a minimal and insignificant increase in CHB testing (RD = 0.6% [95% −3.1 to 4.2%]) compared to the traditional approach in which patients reported their behaviors to providers face-to-face.39

Discussion

We have systematically synthesized evidence on the effectiveness and implementation of diverse strategies, uniformly characterized using BCW framework, to promote CHB testing in primary care and community settings. Among the evaluated strategies, CHB education delivered by HCPs, together with decentralized testing, demonstrated a prominent effect. Importantly, implementation strategies enriched with a larger number of BCW mechanisms, as exemplified in the case of LHWs-driven CHB education, showed larger effect on CHB testing. However, evidence from LMICs is absent. Additionally, less than half of the studies provided clear evaluations of the implementation outcomes of the strategies.

Interestingly, we found that multicomponent CHB education programs led by LHWs, if incorporating more BCW's targeted sources of behavior or strategy functions, demonstrated significantly higher effectiveness. This may suggest that combining multiple theory-driven components when developing an implementation strategy enhances its effectiveness in CHB testing promotion.14 Due to the limited number of available studies in the literature, we could not identify independent components or optimal combinations of components or apply similar meta-regression models to other implementation strategies. Therefore, future studies are encouraged to validate our hypothesis and improve our analysis.

HCPs involvement in CHB education and testing recommendations was the most effective strategy in increasing CHB testing.40,41 With the highest effect size and moderate heterogeneity, this strategy showed great potentials for application in different settings. However, as we consider expanding this strategy, we should appreciate that its success may rely on additional tools to drive behavior change among HCPs. For instance, reminders or best practice alerts for HCPs in electronic health records have been proven effectively in prompting testing orders.30,31,34,42 Yet, these alerts may lead unintended side effects like alert fatigue.62 Hsu et al. reported that only about half of the physicians complied with the alert in their study.34 Another potential tool is financial incentivization for physicians. That said, this approach showed the minimal effectiveness and questionable cost-effectiveness of LMICs.25

Implementation strategies utilizing LHWs for community-based CHB education were extensively studied.27,28,32,35,37,38,43,44,46,47,61 LHWs understand the community, require no clinical training, are cost-effective, and so are considered suitable for resource-limited areas.63 This approach has also proven effective in other health behaviors, such as the uptake of childhood immunization or the initiation of breastfeeding.63 Additionally, the community can be extended to online platforms. The educational materials can be developed, distributed, and benefited by social media users per se in a crowdsourced approach.33,49 However, challenges like low compliance and high contamination, especially on social media, may undermine the effectiveness of these community-based strategies.33,37,44,47,49 Future implementations should proactively address these considerations.

On the other hand, none of the included studies explored the inherent interplay of system and societal factors with the implementation strategies. Understanding these critical contextual factors would help implementation practitioners decide if the findings are likely to generalize to given settings.12,64 These contextual factors could be characterized by policy categories in the BCW framework, such as the availability of related fiscal measures or guideline.14 For example, implementation strategies may interact with financial barriers in some countries, such as Viet Nam and the Philippines, where CHB tests are not covered for the general population.65 Conversely, the release of CHB universal screening guideline by the Center of Disease Control and Prevention in the United States could accelerate testing progress.66

Furthermore, the evidence gap persists, particularly concerning geography and race. None of the trials included in our review were conducted in LMICs, where the majority of the CHB burden exists.2 The lack of studies from Africa and among African immigrants living in North America and Europe represents a substantial gap that future studies should address. Additionally, screening recommendations are evolving, and studies conducted in select populations (e.g., Asian immigrants) may have limited applicability in a setting where universal testing is now recommended, as is the case in the US.66 Other critical contextual factors, such as culture, resources, and infrastructure, vary greatly by setting and profoundly impact the accessibility, acceptability and affordability of any implementation strategies.12 The absence of evidence may impede the most affected countries from making informed policy decisions regarding CHB testing.

The limitations of this review are noted. Firstly, although we conducted comprehensive searches on four major medical and social literature databases, omitting LMIC-specific databases is inevitable due to the language barrier. To mitigate this, we traced references through reference lists and citations of included studies and relevant reviews and reviewed similar articles on large databases. Given the lack of published studies from LMICs, generalization of the findings to such settings should be done with consideration of variation in contextual factors that may mediate effectiveness. Secondly, caution is also warranted in generalizing the estimates to future programs or trials due to large heterogeneity in some strategies. Still, we stress that the consistency in the direction of effects suggests a degree of generalizability across varying population groups (at least those represented in the studies). The fact that strategies inevitably operate differently across contexts is a well-recognized reality in implementation.12,64 This fact echoes the call for more studies focusing on both effectiveness and implementation on diverse races and in geographical and economic areas.

In conclusion, during our evaluation of implementation strategies, some have demonstrated high effectiveness in some settings. However, understanding critical contextual factors and relevance in LMICs contexts is paramount for optimizing real-world effectiveness across settings. Additionally, bundling strategies with policy and health system-level factors can amplify impact and sustainability, facilitating progress toward hepatitis B elimination targets.

Contributors

Conceptualization, TVK, TNDP, CJH, DYD; Methodology, TVK, TNDP, CJH, DYD; Validation, TVK, ML, PN, HN, TNDP, CJH, and DYD; Formal Analysis, TVK; Screening, Reviewing and Data Collection, TVK, PP, ML, QL, PN, HN, DN, BT, CC; Writing—Original Draft Preparation, TVK; Writing—Review & Editing, TVK TNDP, PP, ML, QL, PN, HN, DN, BT, CC; AG, CJH, DYD; Visualization, TVK; Supervision, TNDP, CJH, DYD; Project Administration, TVK; Access to and verification of data, TVK, CJH, and DYD; Responsibility for the decision to submit the manuscript: TVK, CJH, and DYD.

Data sharing statement

Datasets and R codes are available to readers upon request on GitHub.

Declaration of interests

DYD has received financial support from Mai Dolch for the Center of Excellence for Liver Disease in Vietnam at Johns Hopkins School of Medicine; research grants from the Ludwig Institute for Cancer Research, Gilead Sciences, Fujifilm Medical Systems, and Roche Diagnostics International Ltd.; honoraria and travel support from BMJ Best Practices, Fujifilm Medical Systems, Roche Diagnostics International Ltd., and Techno Orbits; and equipment and materials from Fujifilm Medical Systems and Roche Diagnostics International Ltd. DYD has also served on Data Safety Monitoring Boards (DSMBs) or advisory boards for IQVIA. HN has received payment from A. Menarini Singapore Pte. Ltd. for a presentation on the economic evaluation of tenofovir alafenamide in chronic hepatitis B in Vietnam; financial support from VinHealth for the economic evaluation of tenofovir alafenamide in Vietnam; and financial support from Mahidol University for evaluating strategies to prevent mother-to-child transmission of hepatitis B virus (HBV). PN has received financial support from VinHealth for the economic evaluation of tenofovir alafenamide in chronic hepatitis B in Vietnam. All other authors declare no competing interests.

Appendix A Supplementary data

Supplementary Tables and Figures

Supplemental Material 1

Supplemental Material 2

Supplemental Material 3

Supplemental Material 4

Acknowledgements

Sincere thanks to Dr Jay Vaidya, MD PhD (BEADCore Team, Johns Hopkins Medicine) for his consultation on epidemiological and biostatistical methods and to Jacob White, MLS (Welch Medical Library, Johns Hopkins Medicine) for developing and conducting the search strategies.

TVK is the inaugural recipient of the Tran Dolch Post-Doctoral Fellowship in Hepatology at Johns Hopkins School of Medicine, Baltimore MD, USA.

Appendix A Supplementary data related to this article can be found at https://doi.org/10.1016/j.eclinm.2024.102818.
==== Refs
References

1 Cui F. Blach S. Mingiedi C.M. Global reporting of progress towards elimination of hepatitis B and hepatitis C Lancet Gastroenterol Hepatol 8 4 2023 332 342 36764320
2 Thomas D.L. Global elimination of chronic hepatitis N Engl J Med 380 21 2019 2041 2050 31116920
3 Tan M. Bhadoria A.S. Cui F. Estimating the proportion of people with chronic hepatitis B virus infection eligible for hepatitis B antiviral treatment worldwide: a systematic review and meta-analysis Lancet Gastroenterol Hepatol 6 2 2021 106 119 33197397
4 Oru E. Trickey A. Shirali R. Kanters S. Easterbrook P. Decentralisation, integration, and task-shifting in hepatitis C virus infection testing and treatment: a global systematic review and meta-analysis Lancet Global Health 9 4 2021 e431 e445 33639097
5 Sharma M. Ying R. Tarr G. Barnabas R. Systematic review and meta-analysis of community and facility-based HIV testing to address linkage to care gaps in sub-Saharan Africa Nature 528 7580 2015 S77 S85 26633769
6 Shah H.A. Abu–Amara M. Education provides significant benefits to patients with hepatitis B virus or hepatitis C virus infection: a systematic review Clin Gastroenterol Hepatol 11 8 2013 922 933 23639601
7 Zhou K. Fitzpatrick T. Walsh N. Interventions to optimise the care continuum for chronic viral hepatitis: a systematic review and meta-analyses Lancet Infect Dis 16 12 2016 1409 1422 27615026
8 Ortiz E. Scanlon B. Mullens A. Durham J. Effectiveness of interventions for hepatitis B and C: a systematic review of vaccination, screening, health promotion and linkage to care within higher income countries J Community Health 45 2020 201 218 31332639
9 Haridy J. Iyngkaran G. Nicoll A. Hebbard G. Tse E. Fazio T. eHealth technologies for screening, diagnosis, and management of viral hepatitis: a systematic review Clin Gastroenterol Hepatol 19 6 2021 1139 1150.e30 32896632
10 Rajkumar V. McCausland K. Lobo R. A rapid review of interventions to increase hepatitis B testing, treatment, and monitoring among migrants living in Australia Int J Environ Res Publ Health 19 10 2022 5947
11 Moonen C.P. den Heijer C.D. Dukers-Muijrers N.H. van Dreumel R. Steins S.C. Hoebe C.J. A systematic review of barriers and facilitators for hepatitis B and C screening among migrants in the EU/EEA region Front Public Health 11 2023 1118227
12 Eshun-Wilson I. Ford N. Mody A. Strengthening implementation guidelines for HIV service delivery: considerations for future evidence generation and synthesis PLoS Med 20 3 2023 e1004168
13 Michie S. van Stralen M.M. West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions Implement Sci 6 1 2011 42 21513547
14 Michie S. Atkins L. West R. The behaviour change wheel: a guide to designing interventions 1st ed. 2014 Silverback Publishing Great Britain
15 Curran G.M. Bauer M. Mittman B. Pyne J.M. Stetler C. Effectiveness-implementation hybrid designs: combining elements of clinical effectiveness and implementation research to enhance public health impact Med Care 50 3 2012 217 226 22310560
16 Proctor E. Silmere H. Raghavan R. Outcomes for implementation research: conceptual distinctions, measurement challenges, and research agenda Adm Pol Ment Health 38 2 2011 65 76
17 Grossetta Nardini H.K. Wang L. The Yale MeSH analyzer 2024 http://mesh.med.yale.edu/
18 Hoffmann T.C. Glasziou P.P. Boutron I. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide BMJ Br Med J 348 2014 g1687 24609605
19 Cumpston M. Li T. Page M.J. Updated guidance for trusted systematic reviews: a new edition of the Cochrane Handbook for Systematic Reviews of Interventions Cochrane Database Syst Rev 10 10 2019 Ed000142 31643080
20 Sterne J.A.C. Savović J. Page M.J. RoB 2: a revised tool for assessing risk of bias in randomised trials BMJ 366 2019 l4898
21 Higgins J.P. Thompson S.G. Quantifying heterogeneity in a meta-analysis Stat Med 21 11 2002 1539 1558 12111919
22 DerSimonian R. Laird N. Meta-analysis in clinical trials Control Clin Trials 7 3 1986 177 188 3802833
23 Harrer M. Cuijpers P. Furukawa T. Ebert D. Doing meta-analysis with R: a hands-on guide 2021 Chapman and Hall/CRC
24 van Aert R.C.M. Wicherts J.M. van Assen M.A.L.M. Conducting meta-analyses based on p values:reservations and recommendations for applying p-uniform and p-curve Perspect Psychol Sci 11 5 2016 713 729 27694466
25 Flanagan S. Kunkel J. Appleby V. Case finding and therapy for chronic viral hepatitis in primary care (HepFREE): a cluster-randomised controlled trial Lancet Gastroenterol Hepatol 4 1 2019 32 44 30477810
26 Rao J.N. Scott A.J. A simple method for the analysis of clustered binary data Biometrics 48 2 1992 577 585 1637980
27 Ahmadi S. Jamshidimanesh M. Zafarghandi M.B.S. Ranjbar H. The effectiveness of peer education interventions on HIV- and HBV-preventive behaviours in women with substance-related disorders: a cluster randomised control trial HIV AIDS Rev 18 3 2019 183 192
28 Bastani R. Glenn B.A. Maxwell A.E. Cluster-randomized trial to increase hepatitis B testing among Koreans in Los Angeles Cancer Epidemiol Biomarkers Prev 24 9 2015 1341 1349 26104909
29 Bottero J. Boyd A. Gozlan J. Simultaneous human immunodeficiency virus-hepatitis B-hepatitis C point-of-care tests improve outcomes in linkage-to-care: results of a randomized control trial in persons without healthcare coverage Open Forum Infect Dis 2 4 2015
30 Chak E. Li C.-S. Chen M.S. MacDonald S. Bowlus C. Electronic health record alerts enhance mass screening for chronic hepatitis B Sci Rep 10 1 2020 19153
31 Chak E. Taefi A. Li C.S. Research article electronic medical alerts increase screening for chronic hepatitis B: a randomized, double-blind, controlled trial Cancer Epidemiol Biomarkers Prev 27 11 2018 1352 1357 30089680
32 Chen M.S. Jr. Fang D.M. Stewart S.L. Increasing hepatitis B screening for Hmong adults: results from a randomized controlled community-based study Cancer Epidemiol Biomarkers Prev 22 5 2013 782 791 23613027
33 Fitzpatrick T. Tang W. Mollan K. A crowdsourced intervention to promote hepatitis B and C testing among men who have sex with men in China: a nationwide online randomized controlled trial eClinicalMedicine 16 2019 64 73 (Fitzpatrick T., tsfitz@uw.edu) Department of Medicine, University of Washington, Seattle, WA, United States(Tang W.; Pan X.; Tucker J.D.) UNC Project China, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States(Mollan K.) School of) 31832621
34 Hsu L. Bowlus C.L. Stewart S.L. Electronic messages increase hepatitis B screening in At-risk asian american patients: a randomized, controlled trial Dig Dis Sci 58 3 2013 807 814 23073671
35 Juon H.S. Lee S. Strong C. Rimal R. Kirk G.D. Bowie J. Effect of a liver cancer education program on hepatitis B screening among Asian Americans in the Baltimore-Washington metropolitan area, 2009-2010 Prev Chronic Dis 11 2014 130258 (Juon H.-S.) Department of Health, Behavior, and Society, Johns Hopkins Bloomberg School of Public Health, 624 N. Broadway, Room 704, Baltimore, MD 21205. E-mail: hjuon@jhsph.edu(Lee S.) University of Maryland School of Public Health, College Park, Maryla)
36 Khalili M. Kim N.J. Tsoh J.Y. Health within reach—a patient-centered intervention to increase hepatitis B screening among Asian Americans: a randomized clinical trial J Gen Intern Med 37 13 2022 3242 3250 34993863
37 Ma G.X. Fang C.Y. Seals B. A community-based randomized trial of hepatitis B screening among high-risk Vietnamese Americans Am J Publ Health 107 3 2017 433 440
38 Ma G.X. Lee M.M. Tan Y. Efficacy of a community-based participatory and multilevel intervention to enhance hepatitis B virus screening and vaccination in underserved Korean Americans Cancer 124 5 2018 973 982 29131316
39 Richens J. Copas A. Sadiq S.T. A randomised controlled trial of computer-assisted interviewing in sexual health clinics Sex Transm Infect 86 4 2010 310 314 20551234
40 Rosenberg S.D. Goldberg R.W. Dixon L.B. Assessing the STIRR model of best practices for blood-borne infections of clients with severe mental illness Psychiatr Serv 61 9 2010 885 891 20810586
41 Sahajian F. Bailly F. Vanhems P. A randomized trial of viral hepatitis prevention among underprivileged people in the Lyon area of France J Public Health 33 2 2011 182 192
42 Sequeira-Aymar E. Cruz A. Serra-Burriel M. Improving the detection of infectious diseases in at-risk migrants with an innovative integrated multi-infection screening digital decision support tool (IS-MiHealth) in primary care: a pilot cluster-randomized-controlled trial J Trav Med 29 7 2022
43 Shireman T.I. Adia A.C. Tan Y. Online versus in-person training of community health workers to enhance hepatitis B virus screening among Korean Americans: evaluating cost & outcomes Prev Med Rep 19 2020 (Shireman T.I.; Adia A.C.) Department of Health Services, Policy and practice, School of public Health, Brown University, Providence, RI, United States(Tan Y.; Zhu L.; Rhee J.; Ma G.X., grace.ma@temple.edu) Center for Asian Health, Lewis Katz School of Me)
44 Taylor V.M. Bastani R. Burke N. Evaluation of a hepatitis B lay health worker intervention for Cambodian Americans J Community Health 38 3 2013 546 553 23299978
45 Taylor V.M. Burke N.J. Sos C. Do H.H. Liu Q. Yasui Y. Community health worker hepatitis B education for Cambodian American men and women Asian Pac J Cancer Prev 14 8 2013 4705 4709 24083730
46 Taylor V.M. Gregory Hislop T. Bajdik C. Hepatitis B ESL education for Asian immigrants J Community Health 36 1 2011 35 41 20559696
47 Taylor V.M. Hislop T.G. Tu S.P. Evaluation of a hepatitis B lay health worker intervention for Chinese Americans and Canadians J Community Health 34 3 2009 165 172 19127416
48 van der Veen Y.J. van Empelen P. de Zwart O. Visser H. Mackenbach J.P. Richardus J.H. Cultural tailoring to promote hepatitis B screening in Turkish Dutch: a randomized control study Health Promot Int 29 4 2014 692 704 23574695
49 Wong W.C.W. Marley G. Li J. A randomized controlled trial enhancing viral hepatitis testing in primary care via digital crowdsourced intervention NPJ Digit Med 5 1 2022
50 Zhang Y. Li J. Xie Y. Pay-it-forward incentives for hepatitis virus testing in men who have sex with men: a cluster randomized trial Nat Med 29 9 2023 2241 2247 37640859
51 Fung L.C. Nguyen K.H. Stewart S.L. Chen M.S. Tong E.K. Impact of a cancer education seminar on knowledge and screening intent among Chinese Americans: results from a randomized, controlled, community-based trial Cancer 124 2018 1622 1630 29578592
52 Zhang C. Huang Y. Ji L. Zhu Q. Wang L. Wang J. Effectiveness of health management team program to enhance prevention of mother-to-child transmission of hepatitis B virus in Ningxia, China BMC Publ Health 24 1 2024 67
53 Yin X. Wang W. Chen H. Real-world implementation of a multilevel interventions program to prevent mother-to-child transmission of HBV in China Nat Med 30 2 2024 455 462 38297093
54 Luo Y.-L. Gao S. Zhang Q.-F. Eliminating mother-to-child transmission of hepatitis B virus: practice and progress in Baoan, a national pilot district of China BMC Publ Health 24 1 2024 58
55 Thahir S. Muhindo E. Turigye B. Implementation of hepatitis B screening into routine antenatal care to prevent mother-to-child transmission in rural Western Uganda Open Forum Infect Dis 10 9 2023
56 Johnson J. Carr C. McDonald S.J. Flood M.M. Talking testing: impact of a training intervention on midwives' antenatal HIV, hepatitis B and hepatitis C screening practice Women Birth 34 5 2021 e520 e525 33158791
57 Gbadamosi S.O. Eze C. Olawepo J.O. A patient-held smartcard with a unique identifier and an mHealth platform to improve the availability of prenatal test results in rural Nigeria: demonstration study J Med Internet Res 20 1 2018 e18
58 Keel P. Edwards G. Flood J. Assessing the impact of a nurse-delivered home dried blood spot service on uptake of testing for household contacts of hepatitis B-infected pregnant women across two London trusts Epidemiol Infect 144 10 2016 2087 2097 26833270
59 Koruk I. Koruk S. Çopur A. Simsek Z. A intervention study to improve HBsAg testing and preventive practices for hepatitis B in an obstetrics hospital 2011
60 Villacis-Nunez D.S. Orenstein E. Selvaggio P. Rouster-Stevens K. Wang C-s Thakral A. Improving hepatitis B screening prior to rituximab: a quality improvement Project Children 10 7 2023 1142 37508639
61 Xiao Y. Wallace J. Ahad M. Assessing the feasibility, acceptability and impacts of an education program on hepatitis B testing uptake among ethnic Chinese in Australia: results of a randomised controlled pilot study BMC Publ Health 21 1 2021 1861
62 McGreevey J.D. 3rd Mallozzi C.P. Perkins R.M. Shelov E. Schreiber R. Reducing alert burden in electronic health records: state of the art recommendations from four health systems Appl Clin Inform 11 1 2020 1 12 31893559
63 Lewin S. Munabi-Babigumira S. Glenton C. Lay health workers in primary and community health care for maternal and child health and the management of infectious diseases Cochrane Database Syst Rev 3 2010
64 Mehrotra M.L. Petersen M.L. Geng E.H. Understanding HIV program effects: a structural approach to context using the transportability framework J Acquir Immune Defic Syndr 82 2019 S199 S205 31764255
65 Holt B. Fernandez M. Nguyen D. Embedding viral hepatitis into primary healthcare: results of a strategic landscape analysis in Vietnam and the Philippines Lancet Reg Health West Pac 44 2024
66 Conners E.E. Panagiotakopoulos L. Hofmeister M.G. Screening and testing for hepatitis B virus infection: CDC recommendations - United States, 2023 MMWR Recomm Rep 72 1 2023 1 25
