Rapid review of virus risk communication interventions: Directions for COVID-19

Abstract

Objective

In response to COVID-19, we conducted a rapid review of risk communication interventions to mitigate risk from viruses to determine if such interventions are efficacious.

Methods

We searched for risk communication interventions in four databases: Medline, PsycInfo, the ProQuest Coronavirus Research Database, and CENTRAL. The search produced 1572 articles. Thirty-one articles were included in the final review.

Results

Results showed risk communication interventions can produce cognitive and behavior changes around viruses. Results were more consistently positive for interventions focused on HIV/AIDS as compared to influenza. There was no consistent best intervention approach when comparing peer health, audio/visual, and intensive multi-media interventions. Tailoring risk communication toward a target population, in comparison to not tailoring, was related to better outcomes.

Conclusion

The results suggest that risk communication interventions can be efficacious at reducing risk from viruses. They also highlight the complexity of risk communication interventions. Additional research is needed to understand the mechanisms that lead risk communication to reduce risk from viruses.

Practical value

Results support risk communication interventions to reduce risk from viruses.

1. Introduction

Novel viruses are one of the greatest threats to humanity. In the last half-century, several major viral epidemics including but not limited to human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), severe acute respiratory syndrome (SARS), middle east respiratory syndrome (MERS), ebola virus disease (EVD), and swine flu (H1N1) have resulted in loss of life, as well as widespread fear and significant economic impacts. Most recently, the world is facing the global pandemic of coronavirus disease 2019 (COVID-19) which is threatening lives, causing personal distress [[1], [2], [3]], public outcry [4], and economic disarray [5].

There is evidence that risk communication, defined as “the exchange of real-time information, advice, and opinions between experts and people facing threats to their health, economic or social well-being” [6, para. 1], can be effective in controlling pandemic events [[7], [8], [9]]. The Common-Sense Model of Self-Regulation (CSM), consistent with other parallel processing models, proposes the most effective risk communication conveys both the risk (e.g., severity, likelihood) and behavioral strategies to reduce risk [10], such as social distancing, wearing a mask, getting vaccination, and washing hands. Previous research finds risk communication generally improves perceptions of risk severity, vulnerability, and efficacy in virus management [7,11]. It can also improve behaviors to mitigate the threat posed in health crises [12].

Research also suggests that risk communication is complex. Risk communication messages may need to be framed and delivered in multiple and specific ways depending on the risk to maximize the effectiveness of the message to improve perceptions and behaviors [13]. Approaches that are effective for one risk (e.g., cancer-related risks), may not be relevant or functional for another (e.g., HIV) [14]. For example, presenting numerical risk estimates increases perceptions of risk and increases preventive behavior (e.g., obtaining a mammogram to rule out breast cancer) for cancer risks [15], but does not increase preventative behavior for genetic risks [16]. The content of messages also needs to vary depending on the risk (e.g., flood, virus) and the behavioral strategy to reduce risk (e.g., wash hands, have an evacuation strategy). Past reviews have also demonstrated that the effectiveness of risk messaging depends on the audience of interest and other situational factors [17].

According to the CSM, one reason risk communication is complex is that individuals develop mental models of health risks and behaviors to reduce risk [10]. While experts rely on scientific data and statistical modeling to determine the level of risk and appropriate behaviors to mitigate risk [18], oftentimes, the publics’ perceptions of risk are influenced by multiple factors across emotional, social, and other domains, such as personal relevance [18]. These factors also influence the publics’ perceptions of the efficacy of behavioral recommendations to reduce risk and self-efficacy for performing these behaviors. Considering that not all risk communication approaches are effective, and some even backfire, it is important to understand whether risk communication messages are effective to reduce risk from viral risks and if certain risk communication approaches are more effective to reduce risk from viruses.

Previous systematic reviews have examined the efficacy of risk communication for a variety of health risks (e.g., diabetes, stroke, colon cancer, heart disease) [19], or have focused on specific situations (e.g., floods) [20,21], health-related disaster communication [22], genetic risks [16], sexual behaviors [23], pain [24], and vaccinations [17,25]. However, a systematic review on risk communication strategies for viruses does not exist in the literature. A better understanding of risk communication for viruses will allow public health experts to effectively use risk communication to reduce risk from COVID-19 and other pandemics.

The objective of the current review was to determine if there is evidence for the efficacy of risk communication to reduce the spread of viruses. We examined if risk communication can change people’s cognitions and behaviors to reduce the spread of viruses [10]. Our secondary aim was to explore whether some communication approaches are more effective than others for changing people’s cognitions and behaviors to reduce the spread of viruses.

2. Methods

In response to the rapidly evolving COVID-19 pandemic, we conducted a rapid review to determine the most effective communication/dissemination strategies for reducing the spread of viruses by changing people’s cognitions or behavior. A rapid review is a synthesis of knowledge which simplifies aspects of the systemic review process for the aim of producing information within a limited timeframe [26]. Risk communication was operationalized according to the World Health Organization (WHO) definition, as “the exchange of real-time information, advice, and opinions between experts and people facing threats to their health, economic or social well-being” so these affected populations may “take informed decisions to protect themselves and their loved ones” [27, pg.1].

2.1. Search strategy

After consulting with informational experts to pilot the search methodology and syntax, three authors (EP, YCT, and DMW) applied the final syntax through the ProQuest interface to three databases: PsycInfo, Medline, and the ProQuest Coronavirus Research Database. The Cochrane CENTRAL database was also searched. The search included (1) illness, (2) viral transmission, (3) communication, and (4) health behavior. Searches were limited to publications in English, that used a human sample (not an animal sample/study), and that were peer-reviewed (see Appendix A for complete syntax).

2.2. Inclusion and exclusion criteria

To meet the inclusion criteria, articles had to: (1) be empirical, (2) be about a viral illness that can be transmitted to humans, (3) focus on transmission among humans, (4) employ some form of risk/health communication as a predictor, (5) and have a cognitive or behavioral outcome. Articles were excluded from the current review if they: (1) were not in English, (2) not peer-reviewed (e.g., editorial, comment, letter, or newspaper article), (3) used a sample with individuals younger than 18, (4) did not evaluate a clear risk communication/dissemination intervention, (5) conducted an intervention not designed to assist individuals reduce their risk of infection (e.g., designed exclusively for those living with a virus), or (6) did not include quantitative data.

2.3. Article selection and coding

A total of 1572 articles were initially gathered for review (865 from PsycInfo; 684 from Medline; 23 from the ProQuest Coronavirus Research Database). Thirty-five were duplicates and removed. Eight authors each evaluated 195 article titles/abstracts for inclusion, with decisions double-checked by a second author. Authors reached initial agreement on inclusion decisions in 95% of cases, and 100% after discussion of discrepancies. During the title and abstract check, a total of 74 articles were excluded based on the length of intervention and use of qualitative data. In addition to the full text examination on the articles that met the initial pass, these 74 articles were also re-reviewed, with double checks again conducted to ensure accuracy in decision making. The re-review of these 74 initially excluded articles was done to determine the appropriateness of excluding interventions that were long (i.e., 5 h at a time or over 2 weeks in length) and qualitative. Ultimately, lengthy interventions were included and wholly qualitative studies were excluded. Additionally, after further reviewing inclusion criteria, we decided to remove interventions designed exclusively for those diagnosed with a virus. This resulted in the removal of two originally included articles. Initial agreement on inclusion decisions was again high, achieved in 84% of cases and in 100% of cases post discussion (see Fig. 1 for decision flow chart). Thirty-one articles were included in the review.

Fig. 1.

Article Screening and Selection Summary.

2.4. Data extraction and synthesis

Extracted data included authors, year of publication, study design, virus of focus, number of participants, and sample demographics (see Fig. 1). Extracted data on interventions included communication strategy, messaging content, length of intervention, setting, provider, and mode of transmission. Extracted data on analyses included outcome variables, measures, timeframe between comparisons, and achieved results (e.g., primary results, effect sizes). Outcomes were coded as: (1) cognitive risk perception, or how one views the risk, severity, or certainty of infection, (2) cognitions about behaviors, or how one views protective behaviors designed to mitigate risk, (3) behavioral intentions, or intentions to engage in protective behaviors, and (4) behaviors, or one’s engagement in protective behaviors. Some articles examined a change in virus knowledge outcome. These data, while not included in the primary analyses, were collected and synthesized separately (see Table B1, Table B2 in Appendix B).

The heterogeneity of interventions and outcomes precluded meta-analysis. When meta-analytic procedures are not possible, synthesis without meta-analysis (SWiM) guidelines for the synthesis of quantitative data, suggest tallying the number of studies with positive, negative, and no effect [28]. In this review, all articles that included cognitive risk perceptions as an outcome sought to increase participant’s perceptions of risk, therefore increasing cognitions of risk was coded as positive. Similarly, all articles that included cognitions about behaviors, behavioral intentions, or behaviors as outcomes sought to protect individuals from viral infection by reducing risky behaviors and/or increasing protective behavior, therefore changing cognitions about behaviors, behavioral intentions or behaviors to protect from viral infection (reduce risky behavior and/or increase protective behavior) was coded as positive.

Studies were tallied once for each type of outcome. Studies which reported multiple measures for one type of outcome (e.g., multiple measures of cognitive risk perception) could be tallied as having mixed results (e.g., mixture of positive and no effect). Therefore, studies were coded as: (1) positive (pos), (2) negative (neg), (3) no effect (NE), (4) mixture of both positive and no effect results (mixed pos&NE), and (5) mixture of both negative and no effect results (mixed neg&NE). No studies had a mixture of positive and negative outcomes. For ease of interpretation, result tallies for total positive (i.e., combination of pos and mixed pos&NE) and total negative (i.e., combination of neg and mixed neg&NE) are reported.

Initial data synthesis examined the data across all studies and then around virus type and intervention approach, with data arranged to examine patterns in results. Patterns were interpreted if there were at least three studies for a given outcome within the group (e.g., a specific virus or intervention approach). After examining the data, a secondary post-hoc synthesis was conducted to examine the efficacy of intervention message tailoring in producing cognitive or behavior changes for a virus. Tailored messaging was defined in accordance with the WHO definition as any effort to customize risk messaging for a specific target audience to make the messaging more accessible to said audience [29]. An evaluation of target audience tailoring was done because the initial data syntheses showed there was significant heterogeneity and tailoring messaging to a target audience may be efficacious for virus risk communication interventions [29].

2.5. Quality assessment of articles

The Grading of Recommendations Assessment, Development and Evaluation approach (GRADE), a methodology to examine the scientific rigor of articles, was employed to rate the quality of included articles [30]. The GRADE methodology is a widely adopted and supported method for evaluating studies included in reviews. Per the GRADE methodology, articles were provided an apriori score of high for randomized control trials and low for non-randomized control trials. Scores were then downgraded in cases of within-study or publication biases, poor directedness (i.e., examination studied intended population, treatment, or phenomenon), poor precision (e.g., wide confidence interval margins, poor scale reliability), and result inconsistencies amongst studies/trials. Scores were also upgraded if the effects achieved in an article were likely smaller than the true effect, if effects were so large confounds likely did little to obscure true findings, or if the effect appeared proportional to intervention exposure. Scores achieved by included articles ranged from very low (n = 23) to low (n = 8).

3. Results

Among the 31 studies included in the analysis, there was significant variability in design, demographic characteristics, and sample size (see Table 1 ). Most studies (n = 17) utilized a between-group design, comparing an intervention to some form of control group(s). The remaining 14 studies relied on within-group designs via pre-test to post-test comparisons. Results were relatively similar across within and between-group designs (see Table 1).

Table 1.

Demographics of included articles (n = 31).

Authors	Virus	Intervention Type	Treatment content (provider; length)	Tailored: intervention	Comparison group/secondary messaging	Tailored: comparison	Randomized control trial	Sample size	Demographics (setting)	Quality
de Wit JB, Das E, Vet R (2008)	HBV	Audio/visual communication	Written text messaging presented online to participants. Messaging was either statistical or narrative which communicated men who have sex with men are at greater risk for HBV, and then presented an individual who had been infected, introduced in a way to positively connect readers to the fictitious person. All further information related to the message character was presented as first-person quotes. (media; single read-through).	Yes	(a) Communication that men who have sex with men are at risk for HBV and prevalence statistics of HBV-infections (b) Communication that men who have sex with men are at risk for HBV (c) No risk information (media; single read-through).	Yes	Yes	118	All participants were native Dutch men who have sex with men; age: M = 38.3; 51.7% in a stable relationship; 47% had at least some college education (online messaging and surveys).	Low
Vet R, de Wit JB, Das E, (2011)	HBV	Audio/visual communication	Written text messaging presented online to participants. Messaging centered around social norm communication, where a non-infected male who has sex with males (MSM) (MSM) communicate their barriers to, and ultimate acceptance of, the HBV vaccine (MSM without HBV; single read-through).	Yes	Written text messaging presented online to participants. Messaging was a previously validated risk communication script, where the first-hand experience of a MSM diagnosed with HBV was presented who communicated a wish for having perceived greater risk of infection and knowledge of vaccination (MSM with HBV; single read-through).	Yes	Yes	168	All participants were men who has sex with men (MSM); age: M(SD) = 33.8(11.2); 5% were Dutch; 44% had at least a Bachelor’s degree; 37.5% were in a stable partnership (online messaging and surveys).	Low
Coppola V, Camus O (2007)	HIV/AIDS	Audio/visual communication	Participants were given written text epidemiological information of HIV incidence rates, manipulated by orientation: some messages provided an exact number (subdued) or stating the number of cases was over a given number (highly stressed) (N/A; single read-through)	No	Communication of HIV incidence rates, manipulated by framing: messages communicated either the number of incidences per day or per year (N/A; single read-through).	No	No	103	Age ranged from 18–24; all male; all university students; 36% systematically used a condom, 53% occasionally, 11% never (unclear)	Very low
Govender K, Beckett S, Masebo W, Braga C, Zambezi P, Manhique M, George G, Durevall D (2019)	HIV	Audio/visual communication	Participants received SMS messages promoting safe sex practices, specifically condom use. Overall, 35 messages were sent over 29 weeks, one daily in the first week and then once a week (unclear; single read-through)	Yes	Study recruiter provided in-person HIV prevention information once (study recruiter; unclear).	No	Yes	949	Most participants were between 36–49 years old; 76.7% male; 610.0% had not completed high school; 74.8% were truck drivers (electronic phone text messages).	Low
Horn PA, Brigham TA (1996)	AIDS	Audio/visual communication	Participants were provided 3 sessions of education and behavioral rehearsal/modeling. The topic of the sessions was around participant’s sexual behavior, communication, and environmental antecedents to risky behavior (unclear; 2 h sessions).	Yes	N/A	N/A	No	46	All participants were college students with an age range of 18–28 (in-person education sessions).	Very low
Kelly JA, Murphy DA, Washington CD, Wilson TS, Koob JJ, Davis DR, Ledezma G, Davates B (1994)	HIV/ AIDS	Audio/visual communication	Women attended in-person seminars on HIV/AIDS. Information provided in the seminar included HIV risk, protective behaviors, and basic HIV facts (2 female group leaders; 4 weekly 90 min sessions).	No	Women attended in-person seminars on nutrition and other topics relevant to low-income women (unclear; 3 weekly 90 min sessions).	No	Yes	187	All participants were women (in-person health clinic).	Low
Montano NP, Cianelli R, Villegas N, Gonzalez-Guarda R, Williams WO, Tantillo LD (2019)	HIV	Audio/visual communication	The SEPA intervention is a 3 session education program tailored to Hispanic women. Education focused on role playing skills related to safe sex practices (Bi-cultural department of health employees; 2.5 h sessions).	Yes	Delayed implementation of the intervention (Bi-cultural department of health employees; 2.5 h sessions).	No	No	259	All participants were Hispanic women; age: M = 31.6 (in-person education sessions).	Very low
Oswalt SB, Wyatt T (2015)	HIV	Audio/visual communication	The Somos Fuertes program consists of 6 education sessions. These sessions focus on connecting sexual practices/beliefs with culture, conducting female empowerment, along with facts around HIV transmission/protection (program facilitators; 2 h sessions).	Yes	N/A	N/A	No	175	All participants were college attending women; age ranged from 18 to 52 (M = 22.06); 54.3% were Hispanic; 17.2% had not engaged in sex prior to intervention; 93.7% had exclusively male partners (in-person education sessions).	Very low
Turk T, Ewing MT, Newton FJ (2006)	HIV/AIDS	Audio/visual communication	A poster intervention providing information about methods if infection and viral spread. Additional information provided regarding protective behaviors like condom use. Posters were designed via piloted focus groups taking participant opinions into account for the second study phase (unclear; the amount of time taken in the bathroom).	Yes	No messaging presented (N/A; N/A)	No	No	332	Intervention group: age: M = 25.14; 84 male, 82 female. Control group: age: M = 24.66; 79 male, 87 female; more likely to: have received a university education; be currently enrolled as a student, in top income bracket; a total of 85 participants were considered high risk (i.e., multiple sexual partners; clients of commercial sex workers; men who have sex with men; and/or individuals positively predisposed to drug use) (in-person at bars/cafes).	Very low
DeMarco RF, Kendricks M, Dolmo Y, Dolan Looby SE, Rinne K (2009)	HIV	Intensive multimedia communication	The “Women’s Voices Women’s Lives” film designed around communicating HIV risk/consequences to women. The film presented 4 African American women describing the impact of HIV on their lives and addressed topics of: HIV facts, HIV disclosure, and health care needs. The intervention also included exercises and group meetings to further explore topics from the film (film starring HIV positive women; single viewing).	Yes	N/A	N/A	No	131	All participants were women; age: M(SD) = 35(9.67); 47% white; 26% Lantinx; 10% Caribbean Black; 55% single/never married; 29% high school graduate/GEG; number of STDs: M(SD) = 1(2.26) (in-person film viewing and group meetings).	Very low
Wang AL, Lowen SB, Shi, Z, Bissey B, Metzger, DS, Langleben DD (2016)	HIV	Audio/visual communication	Participants viewed television commercials in a lab setting which featured homosexual scenarios and African American actors promoting condom use to preventative behaviors around HIV (commercial about condom usage from various media campaigns; M(SD) = 0.37 min (60.0)).	Yes	Commercials featuring heterosexual scenarios and Caucasian actors promoting condom use to preventative behaviors around HIV (government-sponsored campaigns or commercials produced by condom manufacturers; M(SD) = 0.37 min (60.0)).	No	No	45	M age = 26.84; all were African American men who have sex with men; M years of education = 1.65; 21 HIV positive (in-person in lab/MRI scanner).	Very low
Fogel CI, Crandell JL, Neevel AM, Parker SD, Carry M, White BL, Fasula AM, Herbst JH, Gelaude DJ (2015)	HIV	Intensive multimedia communication	The POWER intervention is an adoption of the SAFE program. Participants engaged in 8 in-person meetings which consisted of education on: the purpose of the program, importance of HIV/STI protection, sexuality, male-female relationships, and other factors related to risky sex. After sessions participants received booster phone calls on the topics covered in the program (trained nurse and social worker; 1.5 h sessions).	No	Control participants received a single STI prevention education session (trained nurse; 1 h).	No	Yes	521	All participants were incarcerated women. Age ranged from 18-60 (M = 33.8). 57.8% were white; 61.0% were high school graduates; 53.2% were employed prior to incarceration; 62.4% were not incarcerated prior to intervention; 52.8% ever had a STI (in-person and over the phone education sessions).	Low
Kaufman MR, Rimal RN, Carrasco M, Fajobi O, Soko A, Limaye R, Mkandawire G. (2014)	HIV	Intensive multimedia communication	Participants were exposed to messaging through multiple mediums (i.e., radio, in-person meetings) regarding HIV as a follow-up to the BRIDGE program. Messaging was designed to produce social/behavior change (program leaders; unclear)	No	N/A	N/A	No	594	All in southern Malawi; Age: M = 29.09; 323 female, 271 male; M education = 5.95 years; 75.3% in a relationship/cohabitating; (mass media radio messaging and in-person meetings).	Very low
Wenger NS, Greenberg JM, Hiborne LH, Kusseling F, Mangotich M, Shapiro MF (1992)	HIV/AIDS	Intensive multimedia communication	Participants were shown either (a) educational multimedia modules which covered areas such as transmission, protective behaviors, condom use, and communication with sexual partners via various outlets (e.g., videotape presentation, lecture, role-play) OR (b) educational multimedia modules plus additionally received HIV testing (physicians familiar with HIV counseling; ∼1 h).	No	No messaging presented (N/A; N/A)	No	Yes	435	Age: M = 23; 72% female; 61% white; 96% unmarried; all students; 63% had had sex without a condom; 23% has had at least 1 STD in the past (in-person at student outpatient health clinic).	Very low
Mustanski B, Parsons JT, Sullivan PS, Madkins K, Rosenberg E, Swann G (2018)	HIV	Intensive multimedia communication	Participants engaged with multimedia eHealth activities (e.g., games, videos) tailored to young men who have sex with men, including videos, interactive animation, and games aimed to increase HIV knowledge, motivate and teach safer behaviors, and instill self-efficacy for HIV prevention strategies (unclear; ∼1 h).	Yes	eHealth control content was the same but not tailored to young men who have sex with men (unclear; ∼1 h).	No	Yes	901	Intervention (n = 445): 53.5% 18–24 years old; 86.5% gay, 11.9% bisexual, 1.6% straight/other; 37.1% White, 23.8% African American, 30.3%, Latino/a, 8.8% other; 45.6% college grad Control (n = 456): 52.5% 18–24 years old; 860.0% gay, 11.2% bisexual, 2.9% straight/other; White, 24.8 African American, 27.4%, Latino/a, 11.6% other; 47.2% college grad (online educational multimedia modules).	Low
Peragallo N, DeForge B, O’Campo P, Lee SM, Kim YJ, Cianelli R, Ferrer L (2005)	HIV	Peer health communication	Project SEPA is a HIV risk-reduction intervention designed to be culturally tailored/sensitive to Latina women. The intervention followed social cognitive theory by integrating skills training and facilitating greater self-efficacy. Additionally, information giving, group discussions, and role playing were all integrated into the intervention (Red Cross trained HIV Latina counselors who were bilingual; unclear)	Yes	Unclear	Unclear	Yes	454	Intervention (n = 263): All were either Mexican or Puerto Rican Latina women; 33.1% 31–29 years old; 84.8% spoke Spanish; 36.9% were in the U.S for between 6-10 years; 40.8% had between 7–11 years of education; Control (n = 191): All were either Mexican or Puerto Rican Latina women; 30.9% 31–29 years old; 74.9% spoke Spanish; 32.3% were in the U.S for between 6–10 years; 43.5% had between 7–11 years of education (in-person education sessions).	Very low
Wyatt TJ, Oswalt SB (2011)	HIV/AIDS	Peer health communication	Five student in-person events were planned: (a) Dramatization/play covering condom use and date-rape, (b) Jeopardy themed event dispelling HIV/STI myths, (c) Author reading covering firs-hand experience with HIV/AIDS from the perspective of underdeveloped countries, (d) First-hand information from an HIV-positive male covering the need for communication and testing, (e) and a presentation covering the history of HIV, comorbidities, national statistics, and prevention information (fellow college students in student-lead organizations; 2 h per meeting)	Yes	N/A	N/A	No	84 for pretest and 89 for post-test	Age range was 18–29 (M(SD) = 21.31(2.75)); 55.4% Hispanic/Latino, 21.7% White, 7.2% Black, 6.0% Asian/Pacific Islander; 75.8% reported one or more sexual partners (in-person at student-led campus events).	Very low
Kelly JA, Lawrence JS, Stevenson LY, Hauth AC, Kalichman SC, Diaz YE, Brasfield TL, Koob JJ, Morgan MG (1992)	HIV/AIDS	Peer Health Education Intervention	Opinion leaders, who were identified by gay bar owners as being popular among gay men, were trained in how to: correct AIDS misconceptions, recommend protective strategies, and endorse protective strategies. These opinion leaders then engaged with bar attendees to provide these messages to those they interacted with. (peer educator; unclear)	Yes	N/A	N/A	No	924	Location 1: Age: M = 31.5; 87% white; location 2: Age: M = 27.1; 80% white; location 3: Age: M = 26.9; 89% white (in-person conversations at regional bars)	Very low
Kocken P, Voorham T, Brandsma J, Swart W (2001)	HIV/AIDS	Peer health communication	Participants engaged in in-person peer education around transmission, the risk for infection, benefits of condom usage, along with how to buy and use condoms (peer educator; 105 min).	Yes	No messaging presented (N/A; N/A)	No	Yes	589	Intervention (n = 293): 43% 20–29 Control (n = 296): 37% 20–29 Overall: 59.8% married, 35.8% had received primary education or less, 370.0% had received former AIDS education (in-person at cafe and mosque settings).	Very low
Probandari A, Setyani RA, Pamungkasari EP, Widyaningsih V, Demartoto A (2020)	HIV	Peer Health Education Intervention	Female sex worker peer educators ran a female condom use education sessions. This included education along with demonstrations of how to use female condoms. Sessions were given twice to each participant and 15-16 participants were present in each session (peer educator; unclear).	Yes	A single routine education in sexual health and HIV prevention (peer educator; unclear).	No	No	230	All participants were female sex Intervention: Age: M = 36.14; 44.5% completed some high school 85.5% had not heard of female condoms; 38.2% had 2 clients (in-person educations sessions).	Very low
Terui S, Huang J, Goldsmith JV. Blackard D, Yang Y, Miller C (2020)	HIV	Peer health communication	Peer educators were trained by professionals in HIV history, impact in the U.S., prevention, and medication (PrEP), and scientific findings. Participants then engaged in in-person educational messaging (undergraduate students enrolled in health communication classes; 3 h).	Yes	N/A	N/A	No	220	Median age = 22; Female, 32.7% Male, 1.4% preferred not to answer; 42.7% African American, 43.2% Caucasion, 5.9% Asian, 4.1% Hispanic; 65.9%; median household income $30,001-40,000 (in-person, peer-to-peer in university classrooms).	Very low
Bourgeois, FT, Simons WW, Olson K, Brownstein JS, Mandl KD (2008)	Influenza	Audio/visual communication	Participants experienced influenza messaging through their personally controlled health record. Influenza messaging occurred in five forms: vaccine reminders, respiratory illness advice, influenza alerts, weekly influenza risk maps, monthly influenza bulletins (Personally controlled health record (PCHR) system PING; unclear)	Yes	Information covering the same areas for cardiovascular health and sun protection (Personally controlled health record (PCHR) system PING; unclear)	No	Yes	99	Intervention (N = 71): age: M(SD) = 46.4(8.6); 58% female; 14% at risk for complications if diagnosed with influenza; 27% received immunization last flu season; 20% received immunization prior to study start. Control (N = 54): age: M = 46.9(9.4); 37%; female; 17% at risk for complications if diagnosed with influenza; 24% received immunization last flu season; 17% received immunization prior to study start (online messaging and surveys).	Very low
Miller S, Yardley L, Little P (2012)	Influenza	Audio/visual communication	Participants received online theory-based messages varying in the level of perceived threat associated with infection. Theory-based messages on information about the medical team, need for preventative behaviors, the connection between hand-washing and flu infection, recommendations for hand-washing from experts, and practical guidelines for hand-washing (medical and Social Science researchers/providers; unclear).	No	Same information with no coping messages (medical and Social Science researchers/providers; unclear).	No	Yes	84	Age: M(SD) = 32.7(11.82); 76.2% women, 19%, men, and 4 did not give their gender; 57.1% reported living in a household with children under the age of 16 (online messaging and surveys).	Very low
Prati G, Pietrantoni L, Zani B (2012)	Influenza: H1N1	Audio/visual communication	Participants read online narrative messages featuring stories from seniors (65 and older) impacted by influenza who subsequently got vaccinated (unclear; single read-through before completing a questionnaire about their intentions to receive the vaccine, social trust (trust in science, medicine), risk perception, efficacy perception of the vaccine, previous flu shot vaccinations, comprehension and believability, and demographics.	Yes	(a) Didactic messages were designed around results from a focus group on African American seniors conducted by Cameron et al. (2009). Messaging was designed around the focus group’s identified perceptions and beliefs around influenza and vaccination, with the Extended Parallel Process Model (EPPM) theory used to identify thematic categories in responses (unclear; single read-through) (b) no messaging presented (N/A; N/A)	No	Yes	311	All were residents of Italy; age ranged from 65 to 84 years (M = 69.74, SD = 5.29); 62.4% were male; 550.0% completed high school, 24.8% some completed university, 15.4% completed middle school/8th grade, and 4.8% primary school level/5th grade (online messaging and surveys).	Low
Chan DK, Yang SX, Mullan B, Du X, Zhang X, Chatzisarantis NL, Hagger MS (2015)	Influenza	Audio/visual communication	Participants attended a lecture where they were advised about wearing facemasks. Messaging was autonomy supportive. Professors asked students to wear a facemask in their lecture hall to prevent H1N1 spread in a hypothetical H1N1 pandemic (hypothetical university class professor; unclear).	No	Advice about wearing facemasks; same request in a hypothetical pandemic, using messaging that was controlling (hypothetical university class professor; unclear).	No	No	705	Age: M = 20.30; 38.16% male; All undergraduate students in China (in-person at university campus).	Very low
Davis OL, Fante RM, Jacobi LL (2013)	Influenza	Audio/visual communication	Two different poster types were hung in the restrooms to prompt hand washing. One poster was the hand washing prompt alone: simple, nonspecific, instructions providing a bulleted list of the procedure needed to thoroughly wash hands The other poster included health information and a prompt including information regarding washing hands as a mean to avoid contracting influenza, as well as steps for thoroughly washing hands At the end of the day, researchers recorded the change in hand soap.	No	No messaging presented (N/A; N/A)	No	No	Unkn-own	Students, faculty, staff, and visitors at the University of varying gender, age, and cultural belonging (in-person at college campus restrooms).	Very low
Wray RJ, Buskirk TD, Jupka K, Lapka C, Jacobsen H, Pakpahan R, Gary E, Wortley P (2009)	Influenza	Audio/visual communication	Participants were exposed to in-person written text with vaccine information. Participants were randomly assigned to either the VSM (treatment condition) or the VIS (control condition). Participants answered a questionnaire with vaccine-related beliefs and intentions. Participants completed questionnaires before and after exposure to their experimental condition. The vaccination safety and mechanisms (VSM), were designed for the purposes of this study. They sought to integrate information from the vaccine information statement (VIS), while also expanding on its purpose. Much like the VIS, vaccine safety and effectiveness is a key piece of communication, communicated via 3 topics: how vaccination works, why vaccination is safe, and that vaccination does not lead to influenza. The VSM also highlights vaccine self-efficacy through communicating its risks and benefits and reassuring readers of their ability to choose to vaccinate or not (CDC/public health communication; single read-through).	No	The VIS is created and provided by the CDC. To be given a vaccine of any kind it is required by federal law to also provide a VIS, designed to inform recipients of risks and benefits associated with vaccination. It is not worded or intended to persuade or reduce fears regarding vaccination (CDC/public health communication; single read-through).	No	Yes	108	Age ranged from 50–60 years; 83% women; 100% were Black/African American; 76% had health insurance (in-person at either participants’ residences, community settings, or university conference rooms).	Very low
Yardley L, Miller S, Scholtz W, Little P (2011)	Influenza	Intensive multimedia communication	Four online seminars presenting on Influenza prevention information. Information included: the need for protective behaviors, importance of hand washing, methods of hand washing, and misconceptions about hand washing/influenza. Online sessions were housed on a website with additional resources provided to participants, along with handouts (medical team/professionals; 4 sessions of unknown duration).	No	No messaging presented (N/A; N/A)	No	Yes	517	Age: M(SD) = 49.76(11.4); 63.83% women; SES deprivation score (SD) = 9.17(6.41) (online messaging and surveys).	Low
Yoo W, Choi D, Park K (2016)	MERS	Established media outlet communication	Participants were surveyed about their expression and reception of MERS-related information through posting, sharing comments, questions, pictures or other information about MERS through Social Networking Sites (SNS) (Facebook, Twitter, Instagram, Pinterest, Kakao Story, Kakao Group, Naver Band, or Between users; unclear), in addition to their Self-efficacy for MERS, perceived susceptibility, perceived severity, handwashing intention, and cough etiquette intention.	No	N/A	N/A	No	1000	Age ranged from 21–69 (M = 45.24(13.46)); 50.2% were male; 52.5% had a bachelor’s, 19.5% had a high school diploma, 15.9% had an associate degree, 11.3% had a graduate degree; the median monthly household income was between $3501–4500; 49.9% had good health, 34.1% had moderate health, and 7% had poor health (online messaging and surveys).	Very low
Johnson BB (2018)	Zika	Audio/visual communication	Participants were randomly assigned to various conditions. In study 1 half the sample saw Zika-prevalence information and the other half saw the same information then viewed CDC maps. In study 2 there were eight manipulations. Condition 1 was similar to the original study although included updated numbers and minor phrasing changes. Conditions 2-4 broke apart prevalence information through geographic distributions, total cases, and transmission routes. Conditions 5 and 6 used CDC’s maps, Condition 7 removed the maps’ caveats, and Condition 8 included birth defects. Maps were targeted to the potentially least prepared consumer; clearly specifying map purpose; using gray tones to convey high and low levels to colorblind readers; and not obscuring higher local risks by averaging data over a large (CDC information; 19.6 min).	Yes	Information included: state case information, total Zika cases in the US, and transmission modalities; study 2 also included the author’s summary of a CDC study on the impact of Zika on birth defects (CDC information; 19.6 min).	No	No	743	Study 2: age: M(SD) = 43.7(13.7); 60.6% Women; 49.2% Bachelors degree holders; 37.8% Liberal, 29.1% Conservative 2 Studies were conducted, data extraction was done on study 2 only to collect data on more outcomes explored in the second examination (online messaging and surveys).	Very low
Chan MS, Winneg K, Hawkins L, Farhadloo M, Jamieson KH, AlbarracínD (2018)	Zika	Established media outlet communication	Participants were surveyed about their risk perceptions and protective behaviors of Zika in relation to posted Zika media posts. Information was disseminated through news websites and legacy media databases, ultimately using sources from the United States (Wall Street Journal, The New York Times, USA Today, The Washington Post, The Miami Herald, The Orlando Sentinel, The Sun-Sentinel, The Tampa Bay Times, ABC, CBS, NBC, CNN, Fox News, and MSNBC).	No	Any written online communication (i.e., media news sights and databases) or broadcasts created by mass media sources including the word Zika or other keywords (Twitter users; unclear)	No	No	29062	Age: M = 54(20.52); 51% women (over-phone survey).	Very low

There was evidence that risk communication interventions for viruses can improve cognitive and behavior outcomes. Across studies, risk communication was shown to positively impact cognitive risk perceptions (e.g., greater perception of viral risk), cognitions about behaviors (e.g., greater efficacy beliefs in protective behaviors), behavioral intentions (e.g., greater intention to engage in protective behaviors) and behaviors to reduce risk (see Table 2 ).

Table 2.

Primary results of included articles (n = 31).

					Cognitive risk perception change outcome			Cognitions about behaviors change outcome			Behavioral intentions change outcome			Behavioral change outcome
Authors	Virus	Interventions	Comparison group/secondary messaging	Comparison type	Outcome variable	Primary result	Statistics	Outcome variable	Primary result	Statistics	Outcome variable	Primary result	Statistics	Outcome variable	Primary result	Statistics
de Wit JB, Das E, Vet R (2008)	HBV	Audio/visual communication: narrative stories sharing personal impact of infection	Audio/visual communication: (a) Statistical evidence of increased risk for infection (b) Statement of risk alone (c) No messaging presented	Inactive and active control group comparison	Cognitions around the perception of HBV: (1) HBV risk (2) HBV severity	Narrative messaging lead only to statistically increased (1) perceived risk to HBV when compared only to both controls (b&c).	(1) F = 3.23 p < 0.05 (2) F = 0.70 p > 0.05				Behavioral intention to receive HBV vaccine	Health risk messaging type did not have a significant effect on intention to obtain the HBV vaccine.	F = 2.19 p = 0.094
Vet R, de Wit JB, Das E, (2011)	HBV	Audio/visual communication: social norm messaging	Audio/visual communication: risk messaging	Active control group comparison	Perceived HBV risk	Exposure to social norm or risk messaging related to significantly greater perceived risk when either were presented alone/not paired.	Social norm message: F = 5.41 p < 0.021 Risk message: F = 12.16 p < 0.001	Vaccination norm	Exposure to social norm messaging only related to significantly greater perceived vaccination norms when not paired with risk messaging	F = 13.43 p < 0.000	Intent to vaccinate	Exposure to social norm or risk messaging only related to significantly greater intent to vaccination norms when either were presented alone/not paired.	Social norm message: F = 6.27 p < 0.013 Risk message: F = 9.17 p < 0.003
Coppola V, Camus O (2007)	HIV/AIDS	Audio/visual communication: incidence rates communicated as either (a) subdued, or the exact incidence rate (b) highly stressed, or the incidence rates being above a number	Audio/visual communication: incidence rates communicated as either (c) yearly number of incidence (d) daily number of incidence	Active control group comparison				Assertion of compulsive beliefs: (1) HIV testing should be compulsory for every sexual partner (2) being tested positive for HIV should be registered (3) a bill circumscribing medical secrecy should be brought to Parliament (4) having unprotected sex while knowing one is HIV positive should be brought to court	Messaging that was stressed and daily framed related to significantly increased agreement with compulsory beliefs (1-4).	Subdued vs stressed: (1) F = 19.66 p < 0.0001 (2) F = 14.42 p < 0.001 (3) F = 13.21 p < 0.001 (4) F = 7.35 p < 0.01 Daily vs yearly: (1) F = 7.09 p < 0.01 (2) F = 10.37 p < 0.01 (3) F = 7.35 p < 0.01 (4) F = 10.45 p < 0.01	Assertion of intentions to: (1) use a condom in their next sexual encounter (2) engage in unsafe sex with an occasional partner (3) take an HIV test in the next six months	Messaging that was stressed and daily framed related to significantly greater intent to (1) use a condom in their next sexual encounter and less intent to (2) engage in unsafe sex with an occasional partner; While a stressed orientation significantly reduced (3) reluctance for HIV testing, framing had no effect on HIV testing intention.	Subdued vs stressed: (1) F = 7.56 p < 0.01 (2) F = 9.04 p < 0.01 (3) F =17.56 p < 0.001 Daily vs yearly: (1) F = 5.48 p < 0.03 (2) F = 10.58 p < 0.01 (3) F < 1.0 p = N.S.
Govender K, Beckett S, Masebo W, Braga C, Zambezi P, Manhique M, George G, Durevall D (2019)	HIV	Audio/visual communication: SMS texts	Audio/visual communication: basic verbal HIV information	Active control group comparison	HIV risk perception	Intervention exposure did not relate to significantly increased HIV risk perception.	OR/B = 0.02 p = 0.37	Condom use self-efficacy	Intervention exposure did not relate to significant increased condom use self-efficacy.	OR/B = −0.02 p = 0.85				Safe sex behavioral engagement: (1) inconsistent condom use (2) didn’t use condoms in last sexual encounter (3) ever tested for HIV (4) HIV testing in last 6 months	Intervention exposure related only to significantly greater (3) rates of ever being tested, or (4) rates of being tested in the last 6 months, for HIV.	(1) OR/B = 0.91 p = 0.60 (2) OR/B = 0.74 p = 0.08 (3) OR/B = 5.17 p = 0.01 (4) OR/B = 1.72 p = 0.02
Horn PA, Brigham TA (1996)	AIDS	Audio/visual communication: in-person education on sexual behavior	N/A	Within-group comparison	Perceived vulnerability to AIDS	The intervention related to a significant increase in perceived vulnerability to AIDS.	Pre: M(SD)= 19.87 (4.52); Post: 22.98 (3.49) p < 0.005	Condom use self-efficacy	Intervention exposure related to significant increases in condom use self-efficacy.	Pre: M(SD)= 21.24 (3.21); Post: M(SD)= 22.43 (2.05) p < 0.007	Intention to use condoms	Intervention exposure related to significant increases in intentions to use a condom.	Pre: M(SD)= 13.78 (3.48); Post: 15.7 (3.1) p < 0.005	Safe sex behaviors: (1) use of condoms in the past week (2) number of partners in the past week (3) discussion of condoms prior to sex (4) acquiring condoms in the past week (5) use of condoms across the study	Intervention exposure related to significant increases to condom use (1) in the past week and (5) across the study, along with (3) condom discussion, and (4) condom acquisition.	(1) t = −3.788 p = 0.000 (2) t = −0.708 p = 0.483 (3) t = −1.689 p = 0.049 (4) t = −3.287 p = 0.001 (5) t = −3.762 p = 0.000
Kelly JA, Murphy DA, Washington CD, Wilson TS, Koob JJ, Davis DR, Ledezma G, Davates B (1994)	HIV/AIDS	Audio/visual communication: HIV/AIDS information	Audio/visual communication: nutrition information	Active control group comparison	Perceived personal risk to HIV/AIDS	Intervention exposure related to significantly increased perceived risk toward HIV/AIDS.	F = 3.67 p < 0.06	Perceived self-efficacy for: (1) initiating condom discussion (2) postponing sex until getting a condom (3) refusing sex without a condom (4) refuting the assertion that using a condom means there is a lack of trust	Intervention exposure related only to significantly increased self-efficacy to (2) postpone sex until getting a condom and (3) refusing sex without a condom.	(1) F = 2.56 p = N.S. (2) F = 10.02 p < 0.002 (3) F = 3.01 p < 0.05 (4) F = 0.66 p = N.S.				(1) Mean number or frequency of: (1a) male sexual partners (1b) unprotected vaginal sex (1c) unprotected vaginal sex partners; (2) Percent of: (2a) intercourse occasions with condom (2b) women using condoms (2c) male partners using condoms	Intervention exposure related only to significantly lower frequencies of (1b) unprotected vaginal sex and greater percentages of condom usage outcomes (2a-c).	(1) F or t = 1.10 p = N.S. (1b) F or t = 4.40 p < 0.04 (1c) F or t = 2.71 p = N.S. (2a) F or t = 13.33 p < 0.001 (2b) F or t = 5.78 p < 0.001 (2c) F or t = 3.58 p < 0.001
Montano NP, Cianelli R, Villegas N, Gonzalez-Guarda R, Williams WO, Tantillo LD (2019)	HIV	Audio/visual communication: SEPA plus HIV testing/counseling	Audio/visual communication: HIV testing/counseling	Within-group comparison				Condom use self-efficacy at: (1) 6-months (2) 12-months	Intervention exposure related to significantly increased condom use self-efficacy at 6 and 12 months.	(1) aPR(95%CI) = 1.84 (1.40-2.43) p < 0.001 (2) aPR(95%CI) = 1.96 (1.50-2.56) p < 0.001				Safe sex behavioral engagement: (1) 6 months - percent of sex encounters with condom, last 30 days (2) 6 months - any condom use (3) 6 months - number of condmless sex events (4) 12 months - percent of sex encounters with condom, last 30 days (5) 12 months - Any condom use (6) 12 months - number of condmless sex events	Intervention exposure related to significantly increased (2) condom use at 6 months, and significantly increases to all outcomes at 12 months (4-6).	(1) aRR(95%CI) = 1.08 (0.84-1.37) p > 0.05 (2) aPR(95%CI) = 1.30 (1.03-1.63) p < 0.05 (3) aRR(95%CI) = 0.96 (0.77-1.20) p > 0.05 (4) aRR(95%CI) = 1.27 (1.01-1.59) p < 0.05 (5) aPR(95%CI) = 1.37 (1.10-1.70) p < 0.01 (6) aRR(95%CI) = 0.80 (0.66-0.97) p < 0.05
Oswalt SB, Wyatt T (2015)	HIV	Audio/visual communication: Somos Fuertes HIV prevention program	N/A	Within-group comparison	Perceived infection risk: (1) perceived STD/STI risk (2) perceived HIV risk	Post intervention participants showed significantly higher perceptions of only (2) HIV risk.	(1) t = 1.88 p > 0.01 (2) t = 3.05 p < 0.01	Safe sex self-efficacy: (1) self-efficacy to engage in STD/HIV protective behaviors (2) self-efficacy to communicate about safe sex	Intervention exposure related to significantly higher efficacy for (1) engaging in protective behaviors and (2) talking about safe-sex.	(1) t = 9.14 p < 0.001 (2) t = 10.11 p < 0.001	Intent to use condoms	Intervention exposure related to significantly higher intentions to use condoms.	t = 4.06 p < 0.01
Turk T, Ewing MT, Newton FJ (2006)	HIV/AIDS	Audio/visual communication: methods of transmission and protection	No messaging presented	Inactive control group comparison	Agreement with the following statements: “I now see that even I could be at risk of AIDS”	The intervention related to a significant increase in agreement with personal risk toward AIDS.	(2) B = 1.170 p = 0.000				(1) Personal intentions to change behavior: (1a) abstain from sex (1b) be faithful to one partner (1c) wear a condom (1d) talk to my family/relatives/friends about AIDS (1e) seek out further AIDS information (1f) Warn people who may be at risk of AIDS (1g) change sexual behavior; (2) Agreement with the following statement: “I intend to use condoms every time I have sex to prevent getting AIDS”	Intervention exposure related only to significantly increased (1c & 2) personal intention to wear a condom.	(1a) χ2 = 1.735 p = 0.188 (1b) χ2 = 3.088 p = 0.079 (1c) χ2 = 5.086 p = 0.024 (1d) χ2 = 0.114 p = 1.0 (1e) χ2 = .841 p = 0.359 (1f) χ2 = 1.313 p = 0.448 (1g) χ2 = 2.496 p = 0.114 (2) B = 0.443 p = 0.009
DeMarco RF, Kendricks M, Dolmo Y, Dolan Looby SE, Rinne K (2009)	HIV	Intensive multimedia communication: narrative stories sharing personal impact of infection	N/A	Within-group comparison							Intentions to engage in safe-sex	Intervention exposure related to significantly greater intentions to engage in safe sex	Pre: M(SD) = 12(2.94) Post: M(SD) = 13(3.07) p < 0.001	Engaging in Safe sex	Intervention exposure related to significantly greater engagement in safe sex	Pre: M(SD) = 6(N/A) Post: M(SD) = 9(N/A) p < 0.001
Wang AL, Lowen SB, Shi, Z, Bissey B, Metzger, DS, Langleben DD (2016)	HIV	Audio/visual communication: gender/race targeted ads	Audio/visual communication: gender/race untargeted ads	Within-group comparison				Attitudes Towards Condom Use	Intervention exposure related to significantly improved attitudes toward condoms.	F = 14.43 p < 0.00001
Fogel CI, Crandell JL, Neevel AM, Parker SD, Carry M, White BL, Fasula AM, Herbst JH, Gelaude DJ (2015)	HIV	Intensive multimedia communication: POWER intervention	Intensive multimedia communication: standard of care STI prevention session	Active control group comparison										Safe sex behavioral engagement: (1) 3 months- Unprotected vaginal intercourse outside of monogamous relationship (2) 3 months- condom use with main partner (3) 3 months- condom use with non-main partner (4) 3 months- STI diagnosis (5) 6 months- unprotected vaginal intercourse outside of monogamous relationship (6) 6 months- condom use with main partner (7) 6 months- condom use with non-main partner (8) 6 months- STI diagnosis	Intervention exposure related only to significantly less (5) unprotected sex and significantly more (6) monogamous condom use 6 months post incarceration when compared to those in the control.	(1) AOR = 0.67 p > 0.05 (2) AOR = 1.60 p > 0.05 (3) AOR = 1.75 p > 0.05 (4) AOR = 1.38 p > 0.05 (5) AOR = 0.57 p < 0.05 (6) AOR = 2.06 p < 0.05 (7) AOR = 0.38 p > 0.05 (8) AOR = 1.34 p > 0.05
Kaufman MR, Rimal RN, Carrasco M, Fajobi O, Soko A, Limaye R, Mkandaw-ire G. (2014)	HIV	Intensive multimedia communication: information around protective behaviors	N/A	Within-group comparison	HIV risk perception	Intervention exposure did not relate to significantly greater HIV risk perception	B = 0.17 p > 0.05	Self-efficacy to protect oneself from exposure	Intervention exposure related to significantly greater HIV protection self-efficacy	B = 0.35 p < 0.01				Engagement in HIV protective behaviors: (1) HIV testing in the past year (2) condom use in last sexual encounter	Intervention exposure related to significantly greater (1) HIV testing in the past year and (2) condom use in last sexual encounter	(1) OR = 1.40 p < 0.001 (2) OR = 1.26 p < 0.05
Wenger NS, Greenberg JM, Hiborne LH, Kusseling F, Mangotich M, Shapiro MF (1992)	HIV/AIDS	Intensive multimedia communication: (a) education covering AIDS transmission and protective behaviors (b) education plus HIV testing	No messaging presented	Inactive control group comparison										HIV protective behavior change: (1) mean number of sexual partners (2) vaginal or anal sex without a condom (3) asking partner about HIV status (4) asking partners about their previous number of partners	Those in the education plus testing group showed only a significant increase in (3) asking about partner HIV status.	(1) Control: M(SD) = 0.72(0.58) Education: M(SD) = 0.70(0.57) Education and testing: M(SD) = 0.84(0.76) p > 0.15 (2) Control: N = 61 Education: N = 68 Education and testing: Nb = 63 p > 0.15 (3) Control: N = 42 Education: N = 41 Education and testing: N = 56 p < 0.05 (4) Control: N = 72 Education: N = 82 Education plus testing: N = 69 p > 0.05
Mustanski B, Parsons JT, Sulliv-an PS, Madkins K, Rosenberg E, Swann G (2018)	HIV	Intensive multimedia communication: MSM targeted ads	Audio/visual communication without a clear speaker: MSM untargeted ads	Active control group comparison										(1) Condomless anal sex (CAS) at: (1a) 12 months (1b) average across follow-up measures; (2) STI infection	Intervention exposure related to significantly reduced (2) STI infection risk and condomless anal sex prevalence rates at (1a) 12 month follow-up.	(1a) PR = 0.83 p=0.04 (1b) PR = 0.89 p=0.07 (2) RR = 0.60 p = 0.01
Peragallo N, DeForge B, O’Campo P, Lee SM, Kim YJ, Cianelli R, Ferrer L (2005)	HIV	Peer health communication: education on HIV and protective behaviors	Unclear	Control group comparison				HIV protective behaviors perceptions: (1) condom-use barriers (2) safe-sex norms	Intervention exposure related only to significantly reduced (1) condom-use barrier perceptions.	(1) χ2 = 16.81, p < 0.001 (2) χ2 = 0.78 p = 0.376	Risk-reduction intentions	Intervention exposure related to significantly greater intentions to reduce risk from HIV.	χ2 = 12.10 p = 0.0005	Engagement in HIV protective behaviors: (1) condom use (2) safe-sex communication	Intervention exposure related to significantly greater (1&2) engagement with HIV protective behaviors.	(1) χ2 = 7.46 p = 0.006 (2) χ2 = 15.01 p = 0.0001
Wyatt TJ, Oswalt SB (2011)	HIV/AIDS	Peer Health Education Intervention: five intervention groups covering varied topics (e.g., prevention, transmission)	N/A	Within group	HIV/STD risk	Intervention exposure related to significantly increased perceptions of risk for HIV/STDs	t = 2.33 p < 0.05	Self-efficacy in effectively convincing a partner to use a condom during anal sex	Intervention exposure related to significantly greater self-efficacy to convince a partner to use a condom.	t = 2.18 p < 0.05	Intentions to use condoms more during oral sex	Intervention exposure related to significantly greater intentions to use condoms during oral sex.	t = 2.26 p < 0.05
Kelly JA, Lawrence JS, Stevenson LY, Hauth AC, Kalichman SC, Diaz YE, Brasfield TL, Koob JJ, Morgan MG (1992)	HIV/AIDS	Peer Health Education Intervention: conversations with trained peer “opinion leaders”	N/A	Within group										Sexual risk-taking across 3 locations: (1) location 1- insertive unprotected sex (2) location 1- receptive unprotected sex (3) location 2- insertive unprotected sex (4) location 2- receptive unprotected sex (5) location 3- insertive unprotected sex (6) location 3- receptive unprotected sex	In location 1 only, intervention exposure related to significantly fewer incidences of unprotected (1) insertive and (2) receptive sex.	(1) z = 2.50 p < 0.01 (2) z = 2.08 p < 0.02 (3) z = 1.79 p < 0.04 (4) z = 2.11 p < 0.02 (5) z = 0.66 p = N/A (6) z = 1.38 p < 0.08
Kocken P, Voorham T, Brandsma J, Swart W (2001)	HIV/AIDS	Peer Health Education Intervention: information on transmission, risk, and prevention	No messaging presented	Within-group comparison	HIV infection risk appraisal	Intervention exposure related to significantly greater perceptions of risk toward HIV infection.	OR = 2.9 p < 0.05	Cognitions around condom usage: (1) condom use self-efficacy (2) belief in the protective effect of condom use (3) perception of condom diminishing satisfaction with sex (4) perception of condom purchase barrier	No difference between groups occurred for cognitions around condom use outcomes.	(1) OR = 1.8 p > 0.05 (2) OR = 1.6 p > 0.05 (3) OR = 1.0 p > 0.05 (4) O R = 0.8 p > 0.05	The intention of condom use in the future	No difference occurred between groups on intentions to use condoms in the future.	OR = 1.2 p > 0.05
Probandari A, Setyani RA, Pamungkasari EP, Widyaningsih V, Demartoto A (2020)	HIV	Peer Health Education Intervention: education specific to female condom use	Peer Health Education Intervention: routine education	Within-group comparison										Safe sex behavioral engagement: (1) use of female condom in last sexual encounter (2) acceptance of female condoms above median	Intervention exposure related to significantly greater (1) use and (2) acceptance of female condoms.	(1) aOR = 17.0 p = S. (2) aOR = 6.1 p = S.
Terui S, Huang J, Goldsmith JV. Blackard D, Yang Y, Miller C (2020)	HIV	Peer Health Education Intervention: information on impact, prevention, and treatment	N/A	Within-group comparison	Cognitions around risk perception: (1) certainty of infection (2) immediacy of HIV consequences (3) HIV threat salience (4) threat severity of infection	Intervention exposure related to significant increases in all outcomes.	(1) t = −3.20 p < 0.01 (2) t = −2.34 p < 0.05 (3) t = −5.79 p < 0.001 (4) t = 4.97 p < 0.001	Efficacy beliefs that: (1) one can prevent/manage infection (2) coping behavior will protect against infection	Intervention exposure related only to significant increases in (1) self-efficacy to prevent/manage HIV infection.	(1) t = −10.98 p < 0.001 (2) t = 0.06 p > 0.05
Bourgeois, FT, Simons WW, Olson K, Brownste-in JS, Mandl KD (2008)	Influenza	Audio/visual communication: Influenza information	Audio/visual communication: Non-influenza information	Active control group	Belief that influenza is serious	Intervention exposure did not relate to significantly higher beliefs that influenza is serious.	OR = 1.2 p = 0.8	Cognitions around influenza and vaccination: (1) vaccine effectiveness beliefs (2) vaccine eligibility beliefs (3) influenza prevention beliefs (4) vaccine benefit beliefs (5) vaccine reaction beliefs	Intervention exposure related to significantly higher beliefs that (1) the influenza vaccine was effective, (3) that there were actions they could take to prevent the flu, and (5) that vaccination was unlikely to cause a severe reaction.	(1) OR = 5.6 p = 0.003 (2) OR = 1.7 p = 0.41 (3) OR = 3.2 p = 0.03 (4) OR = 1.1 p = 0.89 (5) OR = 4.4 p = 0.02				Engagement in protective behaviors: (1) hand hygiene (2) cough etiquette	Intervention exposure did not relate to any significant increase on participant’s engagement in protective behaviors.	(1) OR = 0.9 - 1.9, p = 0.36 - 0.88 (2) OR = 0.7 - 5.7 p = 0.13 - 0.93
Miller S, Yardley L, Little P (2012)	Influenza	Audio/visual communication: (a) low threat of infection (b) high threat of infection	Audio/visual communication: (c) messaging including coping behaviors (d) no coping behaviors provided	Active control group comparison	Threat of infection	A high threat level (b) increased perceptions of infection threat.	Threat level: Partial eta2 = 0.07 p = N/A Coping: Partial eta2 = 0.05 p = N/A	Cognitions around infection: (1) hand washing attitudes (2) view of handwashing as normative (3) behavioral control to wash hands	The combination of threat and coping messages (b & c) related only to significantly higher (1) positive attitudes to hand-washing.	Threat level: (1) Partial eta2 = 0.03 p = N/A (2) Partial eta2 = 0.04 p = N/A (3) Partial eta2 = 0.08 p = N/A Coping: (1) Partial eta2 = .15 p = N/A (2) Partial eta2 = 0.07 p = N/A (3) Partial eta2 = 0.08 p = N/A	Behavioral intent to engage in protective behaviors: (1) intend to wash hands at least 10 times a day (2) intend to wash hands more often (3) intend to wash hands as often as possible (4) intended frequency of hand-washing	The high threat and coping conditions (b & c) related to statistically higher intentions to increase hand-washing (1-4).	Threat level: (1) Partial eta2 = 0.02 p = N/A (2) Partial eta2 = 0.04 p = N/A (3) Partial eta2 = 0.06 p = N/A (4) Partial eta2 = 0.00 p = N/A Coping: (1) Partial eta2 = 0.03 p = N/A (2) Partial eta2 = 0.00 p = N/A (3) Partial eta2 = 0.01 p = N/A (4) Partial eta2 = .11 p = N/A
Prati G, Pietrantoni L, Zani B (2012)	Influenza	Audio/visual communication: narrative stories sharing personal impact of infection	Audio/visual communication: (a) no messaging presented (b) didactic messaging derived from beliefs about infection among the elderly Black community	Inactive and active control group comparison	Influenza risk perception	Intervention exposure related to significantly higher risk perception of influenza in comparison to the no message control (a); no difference between the narrative and didactic conditions (b) occurred.	Narrative: M(SE) = 6.49 (0.18) Control: M(SE)a = 5.83 (0.19) p < 0.05 Didactic: M(SE)b = 6.36 (0.19) p > 0.05	Efficacy of vaccination	Intervention exposure related to greater vaccination self-efficacy in comparison to the no message control (a); no difference between the narrative and didactic conditions (b) occurred.	Narrative: M(SE) = 7.43 (0.15) Control: M(SE)a = 6.93 (0.16) p < 0.05 Didactic: M(SE)b = 7.20(0.16) p > 0.05	Intention to receive Influenza vaccination	Intervention exposure did not relate to significantly greater intention to vaccinate.	N/A
Chan DK, Yang SX, Mullan B, Du X, Zhang X, Chatzisarantis NL, Hagger MS (2015)	Influenza	Audio/visual communication: facemask information and request with autonomy-supportive language	Audio/visual communication: facemask information and request with controlling language	Active control group comparison				Cognitions around wearing a facemask: (1) attitudes about wearing a facemask in their lecture hall in the forthcoming month (2) subjective norm of wearing a facemask in their lecture hall in the forthcoming month (3) perceived behavioral control/ability to wear a facemask in their lecture hall in the forthcoming month	Intervention exposure only related to significantly greater (3) perceived facemask behavioral control.	(1) B = 0.01 p > 0.05 (2) B = 0.04 p > 0.05 (3) B = 0.07 p < 0.05	Intention to wear a facemask in their lecture hall in the forthcoming month	Intervention exposure did not relate to significantly greater intention to wear a facemask.	B = 0.03 p > 0.05
Davis OL, Fante RM, Jacobi LL (2013)	Influenza	Audio/visual communication: instructions for thoroughly washing hands	Audio/visual communication: (a) no messaging presented (b) information on washing hands preventing infection plus instructions	Inactive and active control group comparison										Daily amount of soap used	The prompt alone condition was not different from (a) the no poster condition; a significant decline in average hand soap usage occurred between the prompt alone and (b) health information plus prompt conditions.	Control: M(SD) = 1.05(1.17) Prompt alone: M(SD) = 0.88(0.61) p > 0.05 Health information: M(SD) = 0.75(0.64) p < 0.05
Wray RJ, Buskirk TD, Jupka K, Lapka C, Jacobsen H, Pakpahan R, Gary E, Wortley P (2009)	Influenza	Audio/visual communication: VIS plus vaccination safety and mechanisms (VSM)	Audio/visual communication: vaccine safety and effectiveness (VIS)	Active control group	Cognitions around influenza: (1) belief of susceptibility to the flu (2) belief of severity of the flu	Intervention exposure did not relate to significantly greater perceptions of susceptibility or severity of infection.	(1) M(SD) = 23.5 (3.1); M(SD) = 24.6 (2.9) p = 0.102 (2) M(SD) = 23.6 (30.0); M(SD) = 24.4 (3.1) p = 0.516	Cognitions around influenza and vaccination: (1) self-efficacy in making a vaccination decision (2) belief in the benefit/safety of the flu shot (3) belief in recommenda-tions for the flu shot (4) belief of flu shot efficacy (5) agreement with the following statement: “I worry about side effects from the flu shot”	Intervention exposure related only to significantly higher beliefs in (4) vaccine efficacy.	(1) VSM: M(SD) = 22.7(3.8) VIS: M(SD) = 22.4(3.6) p = 0.555 (2) VSM: M(SD) = 24.9(6.2) VIS: M(SD) = 23.9(5.7) p = 0.878 (3) VSM: M(SD) = 34.6(4.9) VIS: M(SD) = 34.6(4.7) p = 0.511 (4) VSM: M(SD) = 36.5(6.4) VIS: M(SD) = 33.6(6.6) p < 0.001 (5) VSM: M(SD) = 3.0(2.1) VIS: M(SD) = 2.4(1.7) p = 0.558	Intention to vaccinate	Intervention exposure did not relate to significantly higher vaccination intent.	VSM: M(SD) = 36.7(18) VIS: M(SD) = 30.5(18) p = 0.211
Yardley L, Miller S, Scholtz W, Little P (2011)	Influenza	Intensive multimedia communication: Influenza workshop and handouts/website access	No messaging presented	Inactive control group comparison				Hand washing: (1) behavioral norm (2) behavioral control (3) attitudes	Intervention exposure related only to significantly greater (3) hand washing attitudes when compared to the control group	(1) F = 2.23 p = 0.14 (2) F = 0.99 p = .32 (3) F = N/A p = S.	Behavioral intention to wash hands at least 10 times a day	Intervention exposure related to significantly greater behavioral intentions when compared to the control group	F = 14.91 p < 0.001	Hand-washing rates	Intervention exposure related to significantly greater hand-washing rates when compared to the control group	F = 11.71 p = 0.001
Yoo W, Choi D, Park K (2016)	MERS	Established media outlet communication: social media communication	N/A	Within-group comparison	Threat of MERS	Intervention exposure related to significantly increased perceived threat of MERS.	B = 0.18 p < 0.001	MERS protection self-efficacy	Intervention exposure did not significantly predict MERS protection self-efficacy.	B = 0.03 p > 0.05	Intentions to engage in protective behaviors: (1) handwash-ing (2) cough etiquette	Intervention exposure related to significantly higher intention to engage in (1) handwash-ing and (2) cough etiquette.	(1) B = 0.11 p < 0.01 (2) B = 0.12 p < 0.01
Johnson BB (2018)	Zika	Audio/visual communication: (a) case-prevalence messaging alone (b) case-prevalence messaging and mosquito vector maps	N/A	Within-group comparison	Perceptions of Zika: (1) personal risk of Zika (2) concern for Zika	Prevalence statement lead only to significantly greater (1) perceptions of personal risk to Zika; the map condition had no impact on outcomes.	(1a) Cohen’s d = 0.18 p < 0.001 (1b) Cohen’s d = 0.21 p > 0.05 (2a) Cohen’s d = 0.24 p > 0.05 (2b) Cohen’s d = 0.32 p > 0.05				Protective behavior intentions: (1) removing mosquito breeding areas (2) spot spray pesticides (3) avoid travel to infected areas (4) practice safe sex	Prevalence statement lead only to significant increases in behavioral intentions for (1) removing breeding areas and (4) engaging in safer sex; the map condition had no impact on outcomes.	(1a) Cohen’s d = 0.16 p < 0.001 (1b) Cohen’s d(b) = 0.00 p = N.S. (2a) Cohen’s d = 0.02 p = N.S. (2b) Cohen’s d(b) = 0.08 p = N.S. (3a) Cohen’s d = 0.02 p = N.S. (3b) Cohen’s d = 0.04 p = N.S. (4a) Cohen’s d = 0.13 p < 0.05 (4b) Cohen’s d = 0.05 p = N.S.
Chan MS, Winneg K, Hawkins L, Farhadloo M, Jamieson KH, Albarracín D (2018)	Zika	Established media outlet communication: (a) mass media communication (b) social media communication	N/A	Within-group comparison	Risk perception of Zika, at a lag length of: (1) one week (2) two weeks (3) three weeks	(a) Mass media was not associated with risk perceptions across the 3 weeks. (b) Social media was associated with risk perceptions across the 3 weeks (1-3).	(1a) F = 3.87 p = 0.071 (2a) F = 1.52 p = 0.264 (3a) F = 0.81 p = 0.526 (1b) F = 33.38 p < 0.001 (2b) F = 14.02 p < 0.001 (3b) F = 12.90 p = 0.003							Protective behaviors engaged in, at a lag length of: (1) one week (2) two weeks (3) three weeks	(a) Mass media was associated with protective behaviors only (2) two weeks later. (b) Social media was not associated with protective behaviors across the 3 weeks.	(1a) F = 3.75 p = 0.066 (2a) F = 4.00 p = 0.037 (3a) F = 3.10 p = 0.059 (1b) F = 2.37 p = 0.139 (2b) F = 2.28 p = 0.131 (3b) F = 1.74 p = 0.202

Notes: HBV = Hepatitis B; IPC = interpersonal counseling; aOR = adjusted odds ratio; OR = odds ratio; SC = score change; PR = prevalence rate; RR = risk ratio; aPR = adjusted prevalence rate; aRR = adjusted risk ratio B = beta coefficient; N.S. = not significant; S. = significant.

Risk communication interventions focused on HIV/AIDS (n = 19) showed consistent positive findings for changing all outcomes. In contrast, risk communication interventions focused on influenza (n = 7) showed consistent positive findings for improving cognitions about behaviors, but little evidence for improving the other three outcome categories. Moreover, the single negative outcome within this review was for an intervention targeting influenza. There were not enough studies on HBV (n = 2), MERS (n = 1), or Zika (n = 2) to interpret the results.

There was not strong evidence that one type of risk communication approach is more effective at improving virus outcomes. Of the four-risk communication intervention approaches, peer health communication (n = 6) showed the most consistent positive findings for changing cognitive risk perception (see Table 3 ). While peer health communication showed efficacy for improving behavior outcomes as well, intensive multimedia communication (n = 6) showed the most consistent efficacy for producing positive behavior changes. Audio/visual communication (n = 17) showed consistent positive findings for improving cognitive risk perception and cognitions about behaviors, with mixed results for other outcomes. There were not enough studies on established media outlet communication (n = 2) for meaningful conclusions to be drawn.

Table 3.

Summarized results of included articles (n = 31).

	Cognitive risk perception change outcome (pos/neg/no effect)	Cognitions about behaviors change outcome (pos/neg/no effect)	Behavioral intentions change outcome (pos/neg/no effect)	Behavior change outcome (pos/neg/no effect)
Total (n = 31)	Total Pos (pos/pos&NE): 14 (8/6)	Total Pos (pos/pos&NE): 16 (6/10)	Total Pos (pos/pos&NE): 12 (8/4)	Total Pos (pos/pos&NE): 14 (5/9)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 1 (0/1)
	No effect: 4	No effect: 3	No effect: 5	No effect: 1
Between (n = 17)	Total Pos (pos/pos&NE): 6 (3/3)	Total Pos (pos/pos&NE): 10 (1/9)	Total Pos (pos/pos&NE): 6 (3/3)	Total Pos (pos/pos&NE): 7 (2/5)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 1 (0/1)
	No effect: 3	No effect: 1	No effect: 4	No effect: 1
Within (n = 14)	Total Pos (pos/pos&NE): 8 (5/3)	Total Pos (pos/pos&NE): 7 (6/1)	Total Pos (pos/pos&NE): 6 (5/1)	Total Pos (pos/pos&NE): 7 (3/4)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 1	No effect: 2	No effect: 1	No effect: 0
Audio/visual communication (n = 17)	Total Pos (pos/pos&NE): 9 (4/5)	Total Pos (pos/pos&NE): 12 (5/7)	Total Pos (pos/pos&NE): 7 (3/4)	Total Pos (pos/pos&NE): 4 (0/4)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 1 (0/1)
	No effect: 3	No effect: 1	No effect: 4	No effect: 1
Intensive multimedia communication (n = 6)	Total Pos (pos/pos&NE): 0	Total Pos (pos/pos&NE): 2 (1/1)	Total Pos (pos/pos&NE): 2 (2/0)	Total Pos (pos/pos&NE): 6 (3/3)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 1	No effect: 0	No effect: 0	No effect: 0
Peer health communication (n = 6)	Total Pos (pos/pos&NE): 3 (3/0)	Total Pos (pos/pos&NE): 3 (1/2)	Total Pos (pos/pos&NE): 2 (2/0)	Total Pos (pos/pos&NE): 3 (2/1)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 0	No effect: 1	No effect: 1	No effect: 0
Established media outlet communication (n = 2)	Total Pos (pos/pos&NE): 2 (1/1)	Total Pos (pos/pos&NE): 0	Total Pos (pos/pos&NE): 1(1/0)	Total Pos (pos/pos&NE): 1 (0/1)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 0	No effect: 1	No effect: 0	No effect: 0
HBV (n = 2)	Total Pos (pos/pos&NE): 2 (0/2)	Total Pos (pos/pos&NE): 1 (0/1)	Total Pos (pos/pos&NE): 1 (0/1)	Total Pos (pos/pos&NE): 0
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 0	No effect: 0	No effect: 1	No effect: 0
HIV/AIDS (n = 19)	Total Pos (pos/pos&NE): 7 (6/1)	Total Pos (pos/pos&NE): 9 (6/3)	Total Pos (pos/pos&NE): 8 (6/2)	Total Pos (pos/pos&NE): 12 (4/8)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 2	No effect: 1	No effect: 2	No effect: 0
Influenza (n = 7)	Total Pos (pos/pos&NE): 2 (1/1)	Total Pos (pos/pos&NE): 6 (0/6)	Total Pos (pos/pos&NE): 2 (2/0)	Total Pos (pos/pos&NE): 1 (1/0)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 1 (0/1)
	No effect: 2	No effect: 0	No effect: 3	No effect: 1
MERS (n = 1)	Total Pos (pos/pos&NE): 1 (1/0)	Total Pos (pos/pos&NE): 0	Total Pos (pos/pos&NE): 1 (1/0)	Total Pos (pos/pos&NE): 0
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 0	No effect: 1	No effect: 0	No effect: 0
Zika (n = 2)	Total Pos (pos/pos&NE): 2 (0/2)	Total Pos (pos/pos&NE): 0	Total Pos (pos/pos&NE): 1 (0/1)	Total Pos (pos/pos&NE): 1 (0/1)
	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0	Total Neg (neg/neg&NE): 0
	No effect: 0	No effect: 0	No effect: 0	No effect: 0

*Pos = positive effect of risk communication intervention changing cognition/behavior in the intended way (i.e., increasing perceptions of risk, increasing risk mitigation behaviors, reducing risk behaviors), neg = negative effect of risk communication intervention changing cognition/behavior in unplanned direction (i.e., reducing perceptions of risk, decreasing risk mitigation behaviors, increasing risk behaviors, no effect/NE = no effect of risk perception intervention; mixed (pos&NE)=mixture of both positive and no effect results; mixed (neg&NE)=mixture of both negative and no effect results.

Because our initial analysis did not suggest that one type of risk communication approach is more effective, a post hoc data synthesis was done to evaluate the impact of tailoring on the efficacy of risk communication interventions. Nineteen studies tested tailored interventions, including: (1) narrative tailoring, or communicating the impact of viruses on individuals similar to the target population, (2) focus group tailoring, or designing an intervention based on a focus group pilot with the target population, (3) peer communication tailoring, or target audience peers providing risk information to make the information more accessible, and (4) more general efforts to make risk messaging more accessible to the target audience (e.g., constructing an intervention based on literature on the target audience). Twelve studies tested non-tailored interventions. Tailored risk communication interventions, as compared to non-tailored interventions, were consistently related to positive changes in cognitive risk perceptions and behavioral intentions (see Table 4 ).

Table 4.

Summarized results of included articles around tailoring of messaging (n = 30).

	Cognitive risk perception change outcome (pos/neg/no effect)	Cognitions about behaviors change outcome (pos/neg/no effect)	Behavioral intentions change outcome (pos/neg/no effect)	Behavior change outcome (pos/neg/no effect)
Tailored (n = 19)	Total Pos (pos/pos&NE): 10 (5/5) Total Neg (neg/neg&NE): 0 No effect: 2	Total Pos (pos/pos&NE): 9 (4/5) Total Neg (neg/neg&NE): 0 No effect: 2	Total Pos (pos/pos&NE): 9 (6/3) Total Neg (neg/neg&NE): 0 No effect: 3	Total Pos (pos/pos&NE): 8 Total Neg (neg/neg&NE): 0 (3/5) No effect: 1
Non-tailored (n = 12)	Total Pos (pos/pos&NE): 4 (3/1) Total Neg (neg/neg&NE): 0 No effect: 2	Total Pos (pos/pos&NE): 7 (2/5) Total Neg (neg/neg&NE): 0 No effect: 1	Total Pos (pos/pos&NE): 4 (3/1) Total Neg (neg/neg&NE): 0 No effect: 2	Total Pos (pos/pos&NE): 6 (2/4) Total Neg (neg/neg&NE): 1 (0/1) No effect: 0

Note: One article (de witt et al, 2008) compared types of tailored messaging (narrative vs statistical messaging); within the article this comparison accounted for a mixed (pos&NE) tally in cognitive risk perception and a no effect tally in behavioral intentions.

*Pos = positive effect of risk communication intervention changing cognition/behavior in the intended way (i.e., increasing perceptions of risk, increasing risk mitigation behaviors, reducing risk behaviors), neg = negative effect of risk communication intervention changing cognition/behavior in unplanned direction (i.e., reducing perceptions of risk, decreasing risk mitigation behaviors, increasing risk behaviors, no effect/NE = no effect of risk perception intervention; mixed (pos&NE)=mixture of both positive and no effect results; mixed (neg&NE)=mixture of both negative and no effect results.

4. Discussion and conclusion

4.1. Discussion

This review, in response to the rapidly unfolding COVID-19 pandemic, sought to evaluate if and how virus risk information can be effectively communicated to promote cognitive or behavior changes to mitigate infection risk. This review evaluated the efficacy of risk communication interventions to reduce risk from viruses by improving (1) cognitive risk perception, (2) cognitions about behaviors, (3) behavioral intentions, and (4) behaviors. Results showed that risk communication can be efficacious and also highlights the complexities of risk communication to reduce risks from viruses.

Overall, this review suggests risk communication can be efficacious in improving cognitive and behavior outcomes. Intervention efficacy around changing cognitions about a virus and associated protective behaviors is encouraging because risk perceptions can predict engagement in protective behaviors during pandemic events [31], including the current COVID-19 crisis [32].

The results also showed risk communication interventions can directly change behaviors, which is particularly promising. There is growing recognition that while changing cognitions about risk can lead to behavioral change, it does not always. Called a risk perception paradox, it is known that knowledge of risk is not always enough to consistently change behaviors to mitigate risk [33]. Instead, communication must directly target changing behaviors.

Risk communication interventions to reduce risk from HIV/AIDS were most consistently related to improved cognitions and behaviors. In comparison, there was less evidence that interventions focused on reducing risk from influenza were efficacious. A possible explanation for this discrepancy is that influenza is acute and less severe while HIV/AIDS is chronic and more severe. It may be more difficult to change cognitions and behaviors for less severe conditions. The one negative finding, where the intervention resulted in lower perceived risk and fewer protective behaviors, were produced by a study on influenza. Another potential explanation for the distinct results around HIV/AIDS and influenza interventions may be how common influenza is. Common health conditions are generally viewed as less risky [34,35]. Thus, as the COVID-19 pandemic continues, and as COVID-19 becomes more common, extra effort may be needed to ensure risk communication is effective.

There was not strong evidence that one type of risk communication approach is more consistently effective at improving virus outcomes, in part because there were too few studies for most approaches to interpret outcomes. Audio/visual media, which was the least intensive and often included things like posters, was shown to change cognitive risk perception and cognitions about behaviors, but had mixed findings for other outcomes. It is promising that low resource interventions, like posters, can sometimes improve outcomes. However, this review suggests that they are not sufficient on their own. Intensive multi-media interventions, which included online risk communication interventions, were most likely to change behaviors, but there was not enough data to interpret other outcomes. As behaviors are considered the most difficult to change, these interventions may be a promising avenue to change behaviors and should be studied more in the future.

Peer health communication also demonstrated initial positive findings for producing change in cognitive risk perceptions, cognitions about behaviors, and behaviors. This is consistent with previous research which has shown efficacy for peer communication interventions for other health threats [[36], [37], [38]]. Conceptually, peer health communication may provide social support for, or normalization of, protective health behaviors which in turn promotes their employment [39]. Peer health communication is also inherently tailored to the culture of the audience receiving the education.

The positive outcomes from peer health communication led us to hypothesize that tailoring of messaging to specific populations may be particularly beneficial. This is also consistent with the Common-Sense Model (CSM) which proposes that the public develops lay understanding of health threats and behaviors to reduce risk, and as a result, risk communication needs to be tailored to a target audience [10]. In response, we conducted a post-hoc analysis comparing tailored as compared to not tailored interventions. We found that interventions were tailored toward a target audience in multiple different ways including using focus groups, knowledge of the target population, providing narrative messages and using peer educators. Tailored interventions were consistently related to improvements in cognitive risk perception and behavioral intentions. This support for target audience tailoring is in line with the extant evidence of tailoring as particularly efficacious for producing cognitive and behavior changes for the mitigation of health threats [14]. However, less consistent results for the positive impact of tailoring on cognition about behaviors and behavior outcomes supports the need for further study on target audience tailoring on outcomes. Additional research is also needed on the best approaches to deliver tailored interventions, what aspects of communication need to be tailored, and if there is additional benefit to tailoring for the individual as compared to the target group. There were not enough studies to examine these questions in this review.

The results of this review were limited by the heterogeneity of the interventions which precluded meta-analytic procedures. There were also relatively few studies, on only a small number of viruses, which makes it difficult to conclude that any one type of approach is more efficacious than another. Moreover, the viruses included in this review are not perfect analogues for COVID-19. HIV/AIDS, which was the focus of the majority of interventions included in this review, is unique given its methods of transmission and the stigmatization of HIV/AIDS infection. Further, the politicalizing of COVID-19 is unique in comparison to other viruses and pandemic events. As such, the degree to which results apply to COVID-19 risk communication is unclear and in need of further study. Overall, the quality of included studies was generally low. Further, there may be publication bias towards efficacious interventions, limiting the review. However, while conclusions must be viewed as tentative due to these limitations, the results may help guide COVID-19 risk communication.

4.2. Practice implications

The results suggest risk communication interventions may be an effective method for improving cognitions and behaviors to mitigate the risk of COVID-19 infection. We found several different types of risk communication can potentially be efficacious, including simple approaches such as posters. The results also suggest that there may be potential value in tailoring risk communication for specific audiences. Current research on COVID-19 suggests there are a wide-range of beliefs about COVID-19 and behaviors to reduce risk from COVID-19. This includes the increasing politicization of mask wearing, social distancing and vaccines. There may be benefit to tailoring risk communication to address these beliefs.

4.3. Conclusion

The results highlight the complexities inherent to risk communication about viral transmission. This review largely supports risk communication as efficacious in producing positive changes in individuals for the mitigation of viral risk. Results were more consistently positive for interventions focused on HIV/AIDS as compared to influenza. There was no consistent best intervention approach when comparing peer health, audio/visual, and intensive multi-media interventions, with results suggesting that a variety of modalities can be efficacious. There was evidence that interventions tailored to a population can be efficacious when compared to non-tailored interventions.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Dr. Breland is supported by a Department of Veterans Affairs HSR&D Career Development award at the VA Palo Alto (CDA 15-257).

CRediT authorship contribution statement

Darren M. Winograd: Data curation, Formal analysis, Visualization, Writing - original draft, Writing - review & editing, Project administration. Cara L. Fresquez: Formal analysis, Visualization, Writing - original draft, Writing - review & editing. Madison Egli: Formal analysis, Writing - original draft, Writing - review & editing. Emily K. Peterson: Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Alyssa R. Lombardi: Writing - original draft. Allison Megale: Writing - original draft. Yajaira A. Cabrera Tineo: Writing - original draft. Michael G. Verile: Writing - original draft. Alison L. Phillips: Writing - review & editing, Supervision. Jessica Y. Breland: Writing - review & editing, Supervision. Susan Santos: Writing - review & editing, Supervision. Lisa M. McAndrew: Conceptualization, Writing - review & editing, Supervision, Project administration.

Declaration of Competing Interest

All authors declare no competing interests.

Appendix A.

Finalized syntax:

((mesh((Influenza, Human OR SARS Virus OR Virus Diseases)) OR tiab((HIV OR influenza OR flu OR SARS OR virus OR viral OR Zika OR Ebola OR Coronavirus OR MERS OR COVID-19))) AND (mesh((Disease Transmission, Infectious)) OR tiab((transmit OR transmission OR infection OR infectious OR infect OR contagious OR communicable))) AND (mesh((Health Communication OR Risk Communication)) OR tiab((eHealth OR mHealth OR communicat* OR messag*))) AND (mesh((Health Behavior)) OR tiab((adopt OR behavior* OR behavior* OR change OR cognition* OR belief* OR perception OR "risk assessment"))) NOT (pt((editorial OR comment OR letter OR newspaper article))) NOT (mesh((animals)))) AND la.exact("English") AND PEER(yes)

Appendix B.

Table B1.

Primary results of articles which evaluated knowledge change outcomes (n = 11).

					Knowledge change outcome
Authors	Virus	Interventions	Comparison group/secondary messaging	Comparison type	Outcome variable	Primary result	Statistics
Govender K, Beckett S, Masebo W, Braga C, Zambezi P, Manhique M, George G, Durevall D (2019)	HIV	Audio/visual communication: SMS texts	Audio/visual communication: basic verbal HIV information	Active control group comparison	HIV knowledge	Intervention exposure related to significantly higher HIV knowledge.	OR/B = 0.07 p = 0.04
Kelly JA, Murphy DA, Washington CD, Wilson TS, Koob JJ, Davis DR, Ledezma G, Davates B (1994)	HIV/AIDS	Audio/visual communication: HIV/AIDS information	Audio/visual communication: nutrition information	Active control group comparison	AIDS risk behavior knowledge	Intervention exposure related to significantly higher AIDS risk behavior knowledge.	F = 3.47 p < 0.06
Montano NP, Cianelli R, Villegas N, Gonzalez-Guarda R, Williams WO, Tantillo LD (2019)	HIV	Audio/visual communication: SEPA plus HIV testing/counseling	Audio/visual communication: HIV testing/counseling	Within group comparison	HIV knowledge at: (1) 6 months (2) 12 months	Intervention exposure related to significantly higher HIV knowledge at 6 and 12 months.	(1) PR(95%CI) = 1.57 (1.33-1.86) p < 0.001 (2) aPR(95%CI) = 1.63 (1.37-1.95) p < 0.001
Turk T, Ewing MT, Newton FJ (2006)	HIV/AIDS	Audio/visual communication: methods of transmission and protection	No messaging presented	Inactive control group comparison	(1) Knowledge that HIV/AIDS: (1a) spreads via unprotected sex (1b) spreads via needles/drug use (1c) spreads via blood transfusion (1d) spreads via breastfeeding (1e) does not spread via kissing (1f) does not spreads via toilet seats; (2) Agreement with the following statement: “I can reduce my chances of AIDS infection by not injecting drugs”	Intervention exposure related only to significantly higher knowledge that (1a) unprotected sex and (1c) blood transfusions are vectors in HIV/AIDS transmission, along with agreement with (2) “I can reduce my chances of AIDS infection by not injecting drugs”	(1a) χ2 = 3.277 p = 0.07 (1b) χ2 = 2.632 p = 0.105 (1c) χ2 = 7.325 p = 0.007 (1d) χ2 = 1.650 p = 0.199 (1e) χ2 = 1.028 p = 0.502 (1f) χ2 = 0.146 p = 1.0 (2) B = −0.955 p = 0.000
Kaufman MR, Rimal RN, Carrasco M, Fajobi O, Soko A, Limaye R, Mkandaw-ire G. (2014)	HIV	Intensive multimedia communication: information around protective behaviors	N/A	Within group comparison	Knowledge about HIV transmission	Intervention exposure related to significantly higher HIV knowledge	B = 0.20 p < 0.01
Wenger NS, Greenberg JM, Hiborne LH, Kusseling F, Mangotich M, Shapiro MF (1992)	HIV/AIDS	Intensive multimedia communication: (a) education covering AIDS transmission and protective behaviors (b) education plus HIV testing	No messaging presented	Inactive control group comparison	Change in AIDS knowledge	Intervention exposure did not relate to significantly higher knowledge about AIDS	Control: M(SD) = 6.1(0.6) Education: M(SD) = 6.3(0.6) Education plus testing: M(SD) = 6.2(0.7) p > 0.05
Kocken P, Voorham T, Brandsma J, Swart W (2001)	HIV/AIDS	Peer health communication: information on transmission, risk, and prevention	No messaging presented	Inactive control group comparison	Misunderstandings regarding HIV transmission	Intervention exposure related to significantly higher correct answers about HIV risk misconceptions	OR = 5.9 p < 0.05
Probandari A, Setyani RA, Pamungkasari EP, Widyaningsih V, Demartoto A (2020)	HIV	Peer health communication: education specific to female condom use	Peer Health communication: routine education	Within group comparison	HIV knowledge above median level	Intervention exposure related to significantly higher odds for participants to have HIV knowledge above the median.	aOR = 6.6 p < 0.05
Peragallo N, DeForge B, O’Campo P, Lee SM, Kim YJ, Cianelli R, Ferrer L (2005)	HIV	Peer health communication: education on HIV and protective behaviors	Unclear	Control group comparison	HIV knowledge	Intervention exposure related to significantly higher HIV knowledge.	χ2 = 83.10 p < 0.001
Bourgeois, FT, Simons WW, Olson K, Brownste-in JS, Mandl KD (2008)	Influe-nza	Audio/visual communication: influenza information	Audio/visual communication: non-influenza information	Active control group	Knowledge about: (1) hand hygiene (2) couch etiquette (3) injection contacts (4) infection unhealthy behaviors (5) Injection untreated illness (6) infection conditions (6) Influenza vaccine (7) hand cleaners (8) work attendance despite infection	Intervention exposure did not relate to significantly higher protective behavior knowledge	(1) OR = 4.1 p = 0.23 (2) OR = 0.7 p = 0.56 (3) OR = 1.3 p = 0.78 (4) OR = 0.9 p = 0.81 (5) OR = 1.0 p = 0.91 (6) OR = 0.6 p = 0.38 (7) OR = 1.6 p = 0.42 (8) OR = 2.3 p = 0.14
Wray RJ, Buskirk TD, Jupka K, Lapka C, Jacobsen H, Pakpahan R, Gary E, Wortley P (2009)	Influenza	Audio/visual communication: VIS plus vaccination safety and mechanisms (VSM)	Audio/visual communication: vaccine safety and effectiveness (VIS)	Active control group	Agreement with the following statements: (1a) “Common side effects of the flu shot are a sore arm where the shot is given and body aches” (1b) “You can give the flu to others even before you have symptoms” (1c) “The only way to catch the flu is to come in contact with someone who has the flu” (1d) “You get the flu from others who cough and sneeze while they are ill, or by touching something that has the flu virus on it” (1e) “I think the flu shot causes the flu” (1f) “The flu shot is not a cure for the flu and will not help you if you are already ick with the flu”	Intervention exposure related to significantly higher agreement with (1c) "you can give the flu to others even before you have symptoms," and decreased agreement with statement (1f) "I think the flu shot causes the flu"	(1a) VSM: M(SD) = 5.5(1.3) VIS: M(SD) = 5.3(1.4) p = 0.102 (1b) VSM: M(SD) = 5.3(1.7) VIS: M(SD) = 4.6(1.8) p = 0.003 (1c) VSM: M(SD) = 4.1(2.0) VIS: M(SD) = 3.5(1.9) p = 0.580 (1d) VSM: M(SD) = 6.2(0.7) VIS: M(SD) = 5.9(1.2) p = 0.236 (1e) VSM: M(SD) = 4.8(2.0) VIS: M(SD) = 3.9(1.9) p = 0.041 (1f) VSM: M(SD) = 5.5(1.5) VIS: M(SD) = 5.4(1.5) p = 0.928

Notes: SC = score change; OR = odds ratio; aPR = adjusted prevalence rate; B = beta.

Table B2.

Summarized results of articles which evaluated knowledge change outcomes (n = 11).

	Total Pos (pos/pos&NE):	Total Neg (neg/neg&NE)	No effect
Total (n = 11)	9 (7/2)	0	2
Between (n = 8)	6 (4/2)	0	2
Within (n = 3)	3 (3/0)	0	0
Audio/visual communication (n = 6)	5 (3/2)	0	1
Intensive multimedia communication (n = 2)	1 (1/0)	0	1
Peer health communication (n = 3)	3 (3/0)	0	0
Established media outlet communication (n = 0)	0	0	0
HBV (n = 0)	0	0	0
HIV/AIDS (n = 9)	8 (7/1)	0	1
Influenza (n = 2)	1 (0/1)	0	1
MERS (n = 0)	0	0	0
Zika (n = 0)	0	0	0
Tailored (n = 6)	5 (4/1)	0	1
Non-tailored (n = 4)	3 (2/1)	0	1

*Pos = positive effect of risk communication intervention changing cognition/behavior in the intended way (i.e., increasing perceptions of risk, increasing risk mitigation behaviors, reducing risk behaviors), neg = negative effect of risk communication intervention changing cognition/behavior in unplanned direction (i.e., reducing perceptions of risk, decreasing risk mitigation behaviors, increasing risk behaviors, no effect/NE = no effect of risk perception intervention; mixed (pos&NE)=mixture of both positive and no effect results; mixed (neg&NE)=mixture of both negative and no effect results.