pmc10089060931704Health Promot PractHealth Promot PractHealth promotion practice1524-83991552-6372407311381298997010.1177/15248399251354966HHSPA2153220ArticleAssessing Early Impact of the Hear Her® Communication Campaignhttp://orcid.org/0000-0001-9570-8789SpencerPatriciaPhD1http://orcid.org/0000-0002-9309-7627OuyangLijingPhD1CoxShannaMSPH1http://orcid.org/0000-0002-4308-4369TevendaleHeatherPhD1http://orcid.org/0000-0002-6746-3515CarriganSarahMPH1http://orcid.org/0000-0003-1004-4245BehmBrittanyMPH1http://orcid.org/0000-0002-3613-0929RosenthalJacquelineMA2http://orcid.org/0000-0003-1837-2358ThompsonAshley E.PhD34http://orcid.org/0009-0008-1083-3379LakustaMikeBS3http://orcid.org/0009-0001-2171-5449StrunzEricMPH5http://orcid.org/0000-0002-5875-0475BarfieldWanda D.MD, MPH, FAAP, RADM USPHS (ret.)1National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA, USAGlobal Health Center, Centers for Disease Control and Prevention, Atlanta, GA, USAEthniFacts LLC, Irving, TX, USAUniversity of Minnesota Duluth, Duluth, MN, USANational Foundation for the Centers for Disease Control and Prevention, Atlanta, GA, USAAddress correspondence to Patricia Spencer, Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, 4770 Buford Highway, MS S107-2, Atlanta, GA 30341, USA; rup9@cdc.gov.3320267202629720251732026274724737Article reuse guidelines: sagepub.com/journals-permissions

The Hear Her communication campaign was launched nationally in August 2020 to raise awareness of urgent maternal warning signs during and after pregnancy and improve communication between patients and their health care providers. To assess aided campaign awareness during early implementation, an internet panel survey was conducted a month after the campaign launch (Wave 1) with a follow-up survey 7 months later (Wave 2). The survey was fielded nationwide with enhanced recruitment for respondents in the campaign’s priority markets: Atlanta, Dallas, New Orleans, New York, and St. Louis. Aided campaign awareness increased 14.2 percentage points overall among pregnant and postpartum survey participants between waves, with the highest increase in awareness among non-Hispanic Black pregnant and postpartum survey participants (19.8 percentage points). Among Wave 2 survey respondents, those aware of the campaign were more likely to search for information when something did not feel right (adjusted prevalence ratio [aPR] = 1.05), discuss concerns with someone close (aPR = 1.14), or desire more information from their health care provider on pregnancy-related complications or warning signs and symptoms (aPR = 1.62)—after adjustment for age, race/ethnicity, education, geographical location (residence in campaign’s priority markets for media buys), and household income. These findings suggest that aided awareness of the Hear Her campaign increased among pregnant and postpartum survey participants in the first 7 months of the campaign and was associated with outcomes the campaign may influence. This study provides considerations for measuring outcomes associated with implementation of Hear Her and similar campaigns.

Hear Herpregnancy-related complicationspostpartummedia campaignscommunication campaignsmaternal mortality

Too many women die each year in the U.S. from pregnancy-related complications. In 2021, the pregnancy-related death rate in the United States was 33.2 deaths per 100,000 live births, with stark differences by race and ethnicity and geography (Centers for Disease Control and Prevention, 2023, December 15). Data from Maternal Mortality Review Committees (MMRCs) provide a deeper understanding of the circumstances around these deaths and have determined most pregnancy-related deaths are preventable (Petersen et al., 2019; Trost et al., 2022). MMRCs also determine contributing factors to pregnancy-related deaths at the community, health facility, patient, provider, and system levels to inform prevention strategies (Petersen et al., 2019).

Factors contributing to pregnancy-related deaths at the patient and health care provider level include lack of knowledge of warning signs of serious complications, missed or delayed diagnosis, and inappropriate or delayed treatment (Petersen et al., 2019). To address these contributing factors for pregnancy-related deaths, the Hear Her communication campaign was launched in August 2020 to (1) increase awareness of urgent maternal warning signs during and after pregnancy, (2) empower pregnant or postpartum women to speak up if they have any health concerns, (3) encourage support people to carefully listen and help pregnant or postpartum women get the care they need, and (4) improve communication between health care providers and their patients (Behm et al., 2022).

Past health communication campaigns have been successful in advancing public health initiatives through the implementation of organized activities using various forms of media that seek to reach priority audiences. A campaign that encourages the cessation of smoking among U.S. adults aged 18 years or older who smoke was correlated with increased odds of a quit attempt and an estimated 1,005,419 sustained quits lasting at least 1 year (Murphy-Hoefer et al., 2020). A mass media campaign raising awareness of breast cancer screening was effective at reaching low-income African American women (Wallington et al., 2018). Public stroke communication campaigns have been found to have a significant impact on stroke symptom recognition and intention to call emergency medical services (Tan et al., 2022). Through a digital campaign focused on increasing knowledge about low birthweight and pregnancy, significant improvements in positive pregnancy-related intentions in terms of accessing prenatal care and a higher level of local resource awareness were observed among Black women in a priority location (Bonnevie et al., 2022). Anti e-cigarette mass media campaigns have been associated with higher levels of campaign-targeted knowledge, attitudes, and beliefs, and lower odds of intentions to use and current use of e-cigarettes among youth and young adults (Hair, Kreslake, Rath, et al., 2023; Hair, Kreslake, Tulsiani, et al., 2023).

Interventions that use digital storytelling appear to positively affect the health promotion of participants across a range of health issues and diverse communities (Lohr et al., 2022). The Hear Her campaign’s implementation is consistent with public health campaign best practices that include tailoring messages to specific populations and including members from the priority population in message development to ensure cultural sensitivity and trust-building (de Vere Hunt & Linos, 2022; Eppes et al., 2023).

The Hear Her campaign launched with a website, digital media, public service announcements, and additional campaign resources designed to raise awareness of key messages. The campaign centers on stories of women’s experiences with pregnancy-related complications, the symptoms they experienced, and how they got the care they needed. These stories are complemented by posters, palm cards, graphics, social media and other resources designed to raise awareness of the urgent maternal warning signs and support timely, accurate diagnoses. These resources are shared on the Hear Her campaign website and promoted actively through digital and social media channels.

The campaign seeks to reach communities disproportionately impacted by pregnancy-related death (Centers for Disease Control and Prevention, 2024e, April 8). Black women are disproportionately impacted by pregnancy-related deaths (Centers for Disease Control and Prevention, 2023, December 15). Given this disproportionate burden, when conducting formative work to inform the campaign concept and materials, there was a focus on gathering feedback from Black women to test creative concepts (Behm et al., 2022). Priority markets with higher proportions of Black pregnant or postpartum women were identified for media buys (Behm et al., 2022).

The objectives of this study are to assess if aided awareness (the ability of respondents to recognize Hear Her campaign materials after being prompted) (1) increased after campaign implementation and (2) was associated with outcomes the campaign aimed to influence among women who are pregnant or postpartum.

CONCEPTUAL FRAMEWORK

The Audience-Channel-Message-Evaluation (ACME) framework emphasizes the importance of formative, process, and outcome evaluations in health communication campaigns (Noar, 2012). Formative research is integral to making several campaign-related decisions (e.g., pre-testing initial messages); process evaluation assesses if the campaign has gained adequate exposure (e.g., social media metrics such as impressions and engagement); and outcome evaluation assesses the effectiveness of the campaign in changing beliefs or behavior. A key aspect of the ACME framework is selecting audience segments, which influences all other campaign decisions, including messaging and channels. The framework emphasizes the use of behavioral and communication theories to guide the design of effective campaign messages to persuade individuals to think about and act on the message (Noar, 2012). It also highlights the need for ongoing evaluation throughout a campaign’s development and implementation to gather insights and assess effectiveness. Overall, the framework provides an integrated set of principles for designing effective campaigns, increasing the likelihood of success in health communication efforts.

The Hear Her campaign models this approach by prioritizing pregnant or postpartum women as an audience segment to encourage them to discuss pregnancy-related concerns with health care providers. The campaign also aims to influence information-seeking behaviors among this audience.

To refine its messaging, the Hear Her campaign conducted focus groups with priority audiences early in development, integrating evaluation into the creative process (Behm et al., 2022; Noar, 2012). It employed a multichannel strategy and utilized storytelling, consistent with the social cognitive theory, to share women’s experiences with pregnancy complications (Bandura, 2004; Behm et al., 2022). This narrative approach fosters connection and encourages behavior change when the audience identifies with the messages (Briant et al., 2016; Cueva et al., 2015; Hinyard & Kreuter, 2007).

The campaign also aimed to conduct ongoing monitoring of campaign reach for observing performance and identifying areas for improvement (Behm et al., 2022). To evaluate audience exposure and early outcomes of the campaign, internet panel surveys were conducted. This component of evaluation activities aimed to better understand the impact of the campaign by examining changes in intentions and behaviors among pregnant and postpartum women and members of their support systems who have been exposed to campaign messages. For this analysis, process and outcome indicators were selected based on available data that could be influenced by the campaign.

METHODSInternet Panel Survey

An internet panel survey was conducted by EthniFacts, a market research firm contracted by National Foundation for the Centers for Disease Control and Prevention (CDC Foundation) to assess the reach of the Hear Her campaign. This activity was categorized as a non-research, public health practice activity designed to monitor and improve the Hear Her campaign; therefore, Institutional Review Board approval was not needed. The activity aimed to measure increases in aided and unaided awareness and frequency of reported behaviors that are campaign outcomes of interest. It was fielded in two waves—at the beginning of campaign implementation (Wave 1 [August 31, 2020–September 21, 2020]) and 7 months after the start of campaign implementation (Wave 2 [March 21, 2021–April 19, 2021)].

Wave 1 of the data collection occurred at the start of the campaign, coinciding with limited paid media buys to promote the release of the new resources on Facebook, Instagram, Google, YouTube, and other digital channels. There were about 17.2 million impressions from Hear Her campaign ads by the end of Wave 1 data collection. Wave 2 of the data collection was intentionally conducted to coincide with peak media buys. By the end of Wave 2, paid media efforts had resulted in about 94 million impressions of the Hear Her campaign. In addition to paid media buys on digital channels, high impact paid media activities were deployed between waves 1 and 2, such as engaging social media influencers and a celebrity ambassador. The Centers for Disease Control and Prevention (CDC) engaged six influencers in December 2020 and another six influencers in March 2021. The second round of influencers were in market during Wave 2 of the survey. In March 2021, Allyson Felix, an elite track and field athlete and U.S. Olympian, joined the campaign as a spokesperson and ambassador. Engaging a celebrity ambassador boosted visibility of the campaign’s key messages and resources.

Participants were recruited via a programmatic panel exchange of pre-identified mothers, mothers to be, and caregivers, who were then verified through the completion of a pre-screener. Participants were also recruited through social media and affinity group invitations that solicited survey participation. Participants were instructed to click a link to the online survey, then compensated for their time. Compensation during both waves varied based on specific demographic and geographic parameters, with a range of $3 to $15.

The survey was fielded to three segments of survey participants. The three segments were (1) pregnant women defined as those currently pregnant, (2) postpartum women defined as those who had given birth within the last year, and (3) support people defined as those who act as a “go-to” person for pregnant or postpartum women (e.g., spouse, partner, friend, parent). To ensure that survey participants reflected priority markets for media buys, the firm enhanced recruitment and had higher survey quota thresholds for participants living in the campaign’s priority markets for media buys (i.e., Atlanta, Dallas, New Orleans, New York, and St. Louis) and Black pregnant and postpartum women.

As participants completed the survey, daily data cleaning helped preclude data anomalies (e.g., straight-line responses, mislabelled geographies). This resulted in the retention of 1053 respondents in Wave 1 (Pregnant women, n = 241; Postpartum women, n = 521; Support people, n = 291) and 1079 respondents in Wave 2 (Pregnant women, n = 310; Postpartum women, n = 554; Support people, n = 215).

In Wave 1, 947 of the 1053 total respondents were recruited via an opt-in internet panel survey. The remaining 106 were recruited via social media affinity groups and social media ads in the five priority markets. Wave 1 data collection in New Orleans was impacted by a hurricane; therefore, we had fewer respondents than expected (16 vs. the intended 80 completed surveys).

In Wave 2, 762 of the 1079 total respondents were recruited via an opt-in internet panel survey. Of the remaining 317 participants, 134 were screened and recruited through a social media affinity group called “Black Moms Connection,” and 183 were recruited through social media ads in the five priority markets.

Study analyses were limited to responses from pregnant and postpartum women given interest in understanding outcomes among this campaign segment. Demographic data from pregnant and postpartum respondents included race and ethnicity, household income, age, education, and geographical locations. Respondents were asked to identify their ethnicity and race categories; ethnicity included Hispanic or non-Hispanic, and race categories included White, Black, Asian, American Indian or Alaska Native, and Other race. Those that selected more than one race, Asian, American Indian or Alaska Native, and Other race were collapsed into another race/multiple race because of small sample sizes. Categories created for analysis included race/ethnicity (Hispanic; non-Hispanic White; non-Hispanic Black; and non-Hispanic another race/multiple race), household income (<$50,000; $50,000–$74,999; ≥$75,000), mother’s age (18–34; ≥35), mother’s education (high school or less; some college or more), and geographical location (priority markets; rest of the U.S.).

Aided Campaign Awareness

To assess aided awareness of the Hear Her campaign as a process-evaluation indicator, participants during both waves were shown Hear Her assets (Figure 1), thereby establishing aided recall, then asked, “Before now, have you seen these messages or similar messages that use the phrase ‘Hear Her’ and are about women experiencing pregnancy-related complications?” The response options provided were “Yes,” “No,” and “Not Sure.” For this analysis, “No” and “Not Sure” responses were coded as “No.”

Campaign Outcomes of Interest

The survey asked a series of questions on seeking information about pregnancy-related concerns, discussions about pregnancy-related concerns, and interactions with health care providers and members of their support systems. Questions were tailored based on whether respondents were currently pregnant or postpartum (if they had given birth within the last year). Four questions from the survey, as outcome evaluation indicators, were selected to reflect outcomes that may be influenced by campaign awareness: (1) talk to their doctor when something did not feel right; (2) search for information when something did not feel right; (3) discuss concerns with someone close; and (4) desire to receive more information from their health care provider on pregnancy-related complications or warning signs and symptoms.

To define talking to a doctor when something did not feel right, pregnant and postpartum women were asked, “When something just didn’t feel right during your current pregnancy (or during your most recent pregnancy or up to a year after pregnancy), did you ask your doctor (health care provider) about the problem?” Response choices were categorized as: Yes (Yes, every time; Yes, some of the times) and No (No, I didn’t). Those responding that “there was never a time when something didn’t feel right” were excluded from the analysis of this variable such that only those who experienced a concern were included.

The question used to define searching for more information when something did not feel right was only asked among those who responded “Yes, every time,” “Yes, some of the times,” and “No, I didn’t,” to the question described above (excludes those responding that “there was never a time when something didn’t feel right”). To define searching for more information, pregnant and postpartum women were asked, “When something just doesn’t feel right during your current pregnancy (or didn’t feel right during or up to a year after pregnancy), do you search (or did you search) for more information about the problem?” Response choices were categorized as: Yes (Yes, every time; Yes, some of the times) and No (No, I don’t; No, I didn’t). Those responding that “there was never a time when something didn’t feel right” were excluded from the analysis of this variable; only those who expressed a concern were included.

To define discussing concerns with someone close, pregnant and postpartum women were asked, “When you have (or had) pregnancy-related concerns (or concerns up to a year after pregnancy), how often do (or would) you discuss them with someone that you’re close to (or with someone you were close to)?” Response choices: Would Often Discuss (Every time; Most times) and Would Occasionally or Not Discuss (Sometimes; Occasionally; Never). Those responding that they “did not have pregnancy-related concerns” were excluded from the analysis.

To define the desire to receive more information from a health care provider on pregnancy-related complications or warning signs and symptoms, all responding pregnant and postpartum women were asked, “Would you like to receive (or to have received) more information about pregnancy-related complications or warning signs and symptoms from your health care provider?” Response choices: Yes (Definitely yes; Maybe yes) and No (Not sure; Probably not; Definitely not).

Statistical Analysis

To improve comparability between the two samples in Wave 1 and Wave 2, propensity score matching was conducted (Austin, 2014). The Wave variable was used as the binary treatment indicator variable. We used logistic regression modeling to calculate the propensity score (predicted probability) of participating in the survey for each participant, controlling for geographical location (priority markets vs. rest of the U.S.), household income, education, age, and race/ethnicity. Participants in Wave 1 and Wave 2 were then matched based on propensity score using nearest neighbor matching. The matched dataset included a total of 1,372 (Wave 1, N = 686; Wave 2, N = 686) pregnant and postpartum women. The standardized difference between the propensity score means and the variance ratio of the propensity score was used to examine the quality of matching. The standardized mean was reduced in the matched observations, and the largest of these differences is 0.1511, which is less than the recommended upper limit of 0.25 (Rubin, 2001; Stuart, 2010). All treated-to-control variance ratios in the matched observation were also within the recommended range of 0.5 to 2.

Matched data from the propensity score matching process was used to examine if there was an increase in aided awareness of the Hear Her campaign among pregnant and postpartum women during the first 7 months since campaign implementation. Chi-square tests were conducted to examine if the proportion that reported aided awareness was different between waves overall (pregnant and postpartum women combined); by segment (pregnant and postpartum women separately); and by participant characteristics (race/ethnicity, age, education, household income and geographical location [priority market vs. the rest of U.S.]). The percentage point increase in the proportion reporting aided awareness between waves were reported overall and stratified by segment and participant characteristics.

The original (unmatched) data from Wave 2 for 864 pregnant and postpartum women combined (pregnant, N = 310; postpartum, N = 554) were used to assess the association between aided awareness of the Hear Her campaign and campaign outcomes of interest. Logistic regression models were used to estimate the prevalence ratio and to examine the association between aided campaign awareness and the four outcomes of interest (SAS Institute Inc. Usage Note 23003, 2022). Results are presented unadjusted and adjusted for geographical location, household income, race/ethnicity, age, and education. For all analyses, p-values < .05 were considered statistically significant. Analysis was performed using SAS (version 9.4).

RESULTSRespondents’ Characteristics at Wave 1 and Wave 2

Table 1 shows the respondent characteristics across Wave 1 and Wave 2 of the internet panel surveys before propensity score matching. Across both waves, about 40% of respondents were non-Hispanic White and about 40% were non-Hispanic Black and most were between the ages 18 and 34 years and had some college education or more. Supplemental Appendix shows respondent characteristics after propensity score matching.

Change in Aided Awareness of Hear Her Campaign Between Wave 1 and Wave 2

Aided awareness of the Hear Her campaign increased between waves overall and when stratified by most respondent characteristics (Table 2). Significant differences (p < .05) in aided campaign awareness were observed between waves overall, and among: both pregnant and postpartum segments, priority markets and the rest of the U.S., non-Hispanic White and non-Hispanic Black women, and all education, age, and household income groups. Overall, aided campaign awareness increased by 14.2 percentage points between Wave 1 and Wave 2. Aided campaign awareness increased by 19.8 percentage points for non-Hispanic Black women and by 12.3 percentage points for non-Hispanic White women. Aided campaign awareness increased by 10.1 percentage points for those residing in priority markets and 15.6 percentage points for those residing in the rest of the U.S. Aided campaign awareness increased by 16.2 percentage points among postpartum women and 9.0 percentage points among pregnant women.

Association Between Hear Her–Aided Awareness and Campaign Outcomes of Interest

After controlling for age, education, household income, race/ethnicity, and geographical location, the association between aided awareness of the campaign and three campaign outcomes of interest remained statistically significant: Searching for information when something did not feel right (adjusted prevalence ratio [aPR] = 1.05, 95% CI [1.01, 1.09], p = .027); discuss concerns with someone close (aPR = 1.14, 95% CI [1.02, 1.25], p = .019); and desire to receive more information from their health care provider on pregnancy-related complications or warning signs and symptoms (aPR = 1.62, 95% CI [1.36, 1.87], p < .001) (Table 3).

DISCUSSION

Our findings suggest that aided awareness of the Hear Her campaign increased between the launch of the campaign in August 2020 and March 2021. We observed a relatively larger magnitude of increase in aided awareness among non-Hispanic Black pregnant and postpartum women. Our findings also indicate that aided campaign awareness was associated with select campaign outcomes of interest among pregnant and postpartum women. A greater proportion of those with campaign awareness reported searching for information when something did not feel right, discussing concerns with someone close, and desiring to receive more information from their health care provider on pregnancy-related complications or warning signs and symptoms. Altogether, our findings assessed outcomes associated with the early implementation of the Hear Her campaign and can inform subsequent evaluation activities.

Findings suggest that pregnant and postpartum women who were aware of the Hear Her campaign were more likely to search for information when something did not feel right; however, it is unclear the direction of that association. For example, exposure to the Hear Her campaign may prompt searches for more information about pregnancy-related warning signs. Alternatively, searches for more information about pregnancy concerns may have led to exposure to the Hear Her campaign website.

Our analysis also found that pregnant and postpartum women who were aware of the Hear Her campaign were more likely to discuss their problem with someone close to them than those not aware of the campaign. Previous research has shown that pregnant and postpartum women often turn to their support systems to help them cope with difficulties such as loneliness, postpartum depression, and managing a disease such as diabetes (Edwards et al., 2012; Kent-Marvick et al., 2022; Negron et al., 2013; Renbarger et al., 2021).

Another campaign goal is encouraging pregnant and postpartum women to share their concerns with their health care provider. This study found that those with aided campaign awareness were more likely to desire information about pregnancy-related complications or warning signs and symptoms from their health care provider. However, there was no statistically significant association between Hear Her campaign awareness and talking to a doctor when something did not feel right in adjusted analyses. The Hear Her campaign continues to aim to improve communication between health care providers and their pregnant and postpartum patients (Centers for Disease Control and Prevention, 2024b, February 1, 2024d, February 20).

LIMITATIONS

The study has some limitations. First, the study used an internet panel survey based on a non-probability convenience sample. A disadvantage of convenience sampling is the inability to generalize the survey results to the general population. Second, given the use of self-report data, this study may also be subject to response and recall bias. Third, the direction of association cannot be determined. Fourth, because of differences in the way data was collected between waves, this study did not control for experiences of complications of pregnancy which may have impacted information-seeking behaviors. Fifth, Wave 1 data collection (August 31, 2020–September 21, 2020) was not a true baseline as it began almost a month after the launch of the campaign (August 10, 2020). Sixth, differences in some recruitment sources for internet panel participation between baseline (Wave 1) and follow-up (Wave 2) may have contributed to differences between the two waves; however, propensity score matching was used to improve the comparability between both waves. Seventh, those pregnant at the time of the survey may have experienced a concern or exhibited a campaign related outcome of interest later in their pregnancy after the time of survey administration. Eighth, data collection in New Orleans during Wave 1 was impacted by a hurricane that resulted in less completed surveys than intended. Ninth, we collapsed multiple race, Asian, American Indian or Alaska Native, Other race into the heterogenous category “another race/multiple race” for analysis because of small sample size and we were unable to assess campaign awareness among separate race/ethnicity groups that may have different experiences. Tenth, the higher proportion of those responding in the affirmative to outcomes of interest regardless of campaign awareness in Wave 2 may indicate exposure to similar messages. Finally, this study only assessed aided awareness; assessment of unaided awareness and awareness of the urgent maternal warning signs is also important to understand uptake of messages unprompted and other key campaign outcomes of interest.

IMPLICATIONS FOR PRACTICE

Programs that aim to improve maternal health have utilized Hear Her resources as part of their health communication and health promotion efforts. For example, Hear Her campaign materials were used in the development of strategies and tools that highlight urgent maternal warning signs for home visiting programs in Maryland (Stierman et al., 2024). As part of the Birth Equity initiative, the Illinois Perinatal Quality Collaborative includes Hear Her campaign materials as part of standardized postpartum safety education in hospital discharge materials (Illinois Perinatal Quality Collaborative, 2019). Clinical and public health practitioners at various organizational levels (e.g., state, community, health care facility) can learn from these efforts when considering inclusion of Hear Her resources as part of their own activities.

Health-promotion practitioners may also consider tailoring Hear Her resources for local implementation. For example, with funding from CDC, the CDC Foundation supported six community-based organizations to customize their strategies for incorporating Hear Her resources into their current maternal health programs (CDC Foundation, 2024a) and the National Indian Health Board funded six Tribes or Tribal serving organizations to build capacity, implement the campaign and improve maternal outcomes (National Indian Health Board, 2023). Examples of community-based activities include hosting community forums, integrating Hear Her resources into local perinatal care toolkits, and disseminating Hear Her campaign resources through local media channels.

IMPLICATIONS FOR RESEARCH

This study also provides public health researchers with considerations for measuring outcomes associated with implementation of Hear Her and similar campaigns to better understand the potential changes in intentions and behaviors among pregnant and postpartum women and their support systems, with the goal of improving maternal health outcomes across communities.

There is an opportunity for researchers to implement rigorous methods when conducting process and outcome evaluations of activities that leverage Hear Her resources. Developing activity specific logic models can guide what short term and intermediate outcomes to assess (Centers for Disease Control and Prevention, 2024a, August 18, 2024c, May 16). The data limitations of this study provide practice considerations for future data collection and evaluation. For example, if survey respondents can be followed longitudinally, this approach can better control for both observed and unobserved individual characteristics, capture dynamic changes between waves, and reduce estimation bias (Hsiao, 2007). Oversampling can be used to increase representation of smaller populations in evaluation studies (Chen et al., 2021).

CONCLUSIONS

In the first 7 months of implementation, there was increased aided awareness of the Hear Her campaign. Aided awareness of the campaign was associated with outcomes that may have been influenced by the campaign, including wanting more information from providers, discussing concerns with support people, and information-seeking behaviors. Since the initial implementation of the campaign, CDC continued to develop audience-centered resources for pregnant and postpartum women, support people, and health care professionals. CDC also developed culturally and linguistically appropriate resources for different populations, including resources designed with and for American Indian and Alaska Native communities and resources translated into over 21 languages (CDC Foundation, 2024b). Additional evaluations can inform implementation strategies for the Hear Her campaign and contribute to the evidence base on the impact.

Supplementary Material

Supplemental material for this article is available online at https://journals.sagepub.com/home/hpp.

The Hear Her campaign is supported through a partnership with the National Foundation for the Centers for Disease Control and Prevention and funding from Merck through its Merck for Mothers Program (MFM-20-109044).

Disclaimer

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the National Foundation for the Centers for Disease Control and Prevention.

REFERENCESAustinPC (2014). A comparison of 12 algorithms for matching on the propensity score. Statistics in Medicine, 33(6), 1057–1069. https://onlinelibrary.wiley.com/doi/10.1002/sim.600424123228 BanduraA (2004). Health promotion by social cognitive means. Health Education Behavior, 31(2), 143–164. 10.1177/109019810426366015090118 BehmB, TevendaleH, CarriganS, StoneC, MorrisK, & RosenthalJ (2022). A national communication effort addressing maternal mortality in the United States: Implementation of the hear her campaign. Journal of Women’s Health, 31(12), 1677–1685. 10.1089/jwh.2022.0428BonnevieE, BarthC, MayJ, CareyT, KnellSB, WartellaE, & SmyserJ (2022). Growing and glowing: A digital media campaign to increase access to pregnancy-related health information for black women during the COVID-19 pandemic. Health Promotion Practice, 24(3), 444–454. 10.1177/1524839922108384435392694 BriantKJ, HalterA, MarchelloN, EscareñoM, & ThompsonB (2016). The power of digital storytelling as a culturally relevant health promotion tool. Health Promotion Practice, 17(6), 793–801. 10.1177/152483991665802327402721 CDC Foundation. (2024a). Community-based organizations join the hear her® campaign to share maternal health messaging. https://www.cdcfoundation.org/programs/hear-her-evaluation-and-community-based-organization-implementationCDC Foundation. (2024b). Sharing lifesaving messages to help prevent pregnancy-related deaths. https://www.cdcfoundation.org/stories/sharing-lifesaving-messages-help-prevent-pregnancy-related-deathsCenters for Disease Control and Prevention. (2023, December 15). Pregnancy mortality surveillance system. https://www.cdc.gov/maternal-mortality/php/pregnancy-mortality-surveillance/?CDC_AAref_Val=https://www.cdc.gov/reproductivehealth/maternal-mortality/pregnancy-mortality-surveillance-system.htmCenters for Disease Control and Prevention. (2024a, August 18). CDC Approach to Program Evaluation (Step 2- Describe the Program). https://www.cdc.gov/evaluation/php/evaluation-frame-work-action-guide/step-2-describe-the-program.htmlCenters for Disease Control and Prevention. (2024b, February 1). Health Care Professionals. https://www.cdc.gov/hearher/hcp/overview/index.htmlCenters for Disease Control and Prevention. (2024c, May 16). Logic Models. https://www.cdc.gov/library/research-guides/logic-models.html#cdc_report_pub_study_section_5-logic-models-for-program-evaluationCenters for Disease Control and Prevention. (2024d, February 20). Pregnant and postpartum women. https://www.cdc.gov/hearher/pregnant-postpartum/index.htmlCenters for Disease Control and Prevention. (2024e, April 8). Working Together to Reduce Black Maternal Mortality. https://www.cdc.gov/womens-health/features/maternal-mortality.html?CDC_AAref_Val=https://www.cdc.gov/healthequity/features/maternal-mortality/index.htmlChenS, StubblefieldA, & StonerJA (2021). Oversampling of minority populations through dual-frame surveys. Journal of Survey Statistics and Methodology, 9(3), 626–649. 10.1093/jssam/smz05434322557 CuevaM, KuhnleyR, RevelsL, SchoenbergNE, & DignanM (2015). Digital storytelling: A tool for health promotion and cancer awareness in rural Alaskan communities. International Journal of Circumpolar Health, 74(1). 10.3402/ijch.v74.28781de Vere HuntI, & LinosE (2022). Social media for public health: Framework for social media-based public health campaigns. Journal of Medical Internet Research, 24(12), Article e42179. 10.2196/42179EdwardsRC, ThullenMJ, IsarowongN, ShiuCS, HensonL, & HansSL (2012). Supportive relationships and the trajectory of depressive symptoms among young, African American mothers. Journal of Family Psychology, 26(4), 585–594. 10.1037/a002905322888784 EppesEV, AugustynM, GrossSM, VernonP, CaulfieldLE, & PaigeDM (2023). Engagement with and acceptability of digital media platforms for use in improving health behaviors among vulnerable families: Systematic review. Journal of Medical Internet Research, 25, Article e40934. 10.2196/40934HairEC, KreslakeJM, RathJM, PitzerL, BennettM, & ValloneD (2023). Early evidence of the associations between an anti-e-cigarette mass media campaign and e-cigarette knowledge and attitudes: Results from a cross-sectional study of youth and young adults. Tobacco Control, 32(2), 179. 10.1136/tobaccocontrol-2020-05604734290134 HairEC, KreslakeJM, TulsianiS, McKayT, & ValloneD (2023). Reducing e-cigarette use among youth and young adults: Evidence of the truth campaign’s impact. Tobacco Control, tc-2023–057992. 10.1136/tc-2023-057992HinyardLJ, & KreuterMW (2007). Using narrative communication as a tool for health behavior change: A conceptual, theoretical, and empirical overview. Health Education & Behavior, 34(5), 777–792. http://www.jstor.org/stable/4505595717200094 HsiaoC (2007). Panel data analysis—advantages and challenges. TEST, 16(1), 1–22. 10.1007/s11749-007-0046-xIllinois Perinatal Quality Collaborative. (2019). Birth equity—BE. https://ilpqc.org/initiative/birth-equity-be/Kent-MarvickJ, SimonsenS, PentecostR, TaylorE, & McFarlandMM (2022). Loneliness in pregnant and postpartum people and parents of children aged 5 years or younger: A scoping review. Systematic Reviews, 11(1), 196. 10.1186/s13643-022-02065-536071448 LohrAM, Raygoza TapiaJP, ValdezES, HassettLC, GubriumAC, Fiddian-GreenA, LarkeyL, SiaIG, & WielandML (2022). The use of digital stories as a health promotion intervention: A scoping review. BMC Public Health, 22(1), 1180. 10.1186/s12889-022-13595-x35698097 Murphy-HoeferR, DavisKC, KingBA, BeistleD, RodesR, & GraffunderC (2020). Association between the tips from former smokers campaign and smoking cessation among adults, United States, 2012–2018. Prevention Chronic Disease, 17(200052). https://www.cdc.gov/pcd/issues/2020/20_0052.htmNational Indian Health Board. (2023). Hear her campaign: Maternal health promotion for American Indian and Alaska native communities 2023. Request for Applications (RFA). https://www.nihb.org/resources/maternal-health-hear-her_funding-opportunity-overview.pdfNegronR, MartinA, AlmogM, BalbierzA, & HowellEA (2013). Social support during the postpartum period: Mothers’ views on needs, expectations, and mobilization of support. Maternal and Child Health Journal, 17(4), 616–623. 10.1007/s10995-012-1037-422581378 NoarSM (2012). An Audience–Channel–Message–Evaluation (ACME) framework for health communication campaigns. Health Promotion Practice, 13(4), 481–488. 10.1177/152483991038690121441207 PetersenEE, DavisNL, GoodmanD, CoxS, MayesN, JohnstonE, SyversonC, SeedK, Shapiro-MendozaCK, CallaghanWM, & BarfieldW (2019). Vital signs: Pregnancy-related deaths, United States, 2011–2015, and strategies for prevention, 13 States, 2013–2017. MMWR Morbidity and Mortality Weekly Report, 68(18), 423–429. 10.15585/mmwr.mm6818e131071074 RenbargerKM, PlaceJM, & SchreinerM (2021). The Influence of four constructs of social support on pregnancy experiences in group prenatal care. Women’s Health Report, 2(1), 154–162. 10.1089/whr.2020.0113RubinDB (2001). Using propensity scores to help design observational studies: Application to the tobacco litigation. Health Services and outcomes Research Methodology, 2(3), 169–188. 10.1023/A:1020363010465SAS Institute Inc. Usage Note 23003. (2022). Estimating a relative risk (also called risk ratio, prevalence ratio). https://support.sas.com/kb/23/003.htmlStiermanEK, WattsT, LawsonSM, BruceTM, Reed-VanceM, BowerKM, CreangaAA, & Callaghan-KoruJA (2024). Educating families on urgent maternal warning signs: Learning from a pilot to improve training and tools for maternal and child home visiting programs. Health Promotion Practice, 0(0), 15248399241256691. 10.1177/15248399241256691StuartEA (2010). Matching methods for causal inference: A review and a look forward. Statistical Science, 25(1), 1–21. 10.1214/09-sts31320871802 TanJ, RamazanuS, LiawSY, & ChuaWL (2022). Effectiveness of public education campaigns for stroke symptom recognition and response in non-elderly adults: A systematic review and meta-analysis. Journal of Stroke and Cerebrovascular Disease, 31(2), 106207. 10.1016/j.jstrokecerebrovasdis.2021.106207TrostS, BeauregardJ, ChandraG, NjieF, BerryJ, HarveyA, & GoodmanDA (2022). Pregnancy-related deaths: Data from maternal mortality review committees in 36 US States, 2017–2019. MMRIA: Maternal Mortality Review Information App, CDC Reports Using MMRIA Data. https://www.cdc.gov/maternal-mortality/php/data-research/?CDC_AAref_Val=https://www.cdc.gov/reproductivehealth/maternal-mortality/erase-mm/data-mmrc.htmlWallingtonSF, OppongB, IddirisuM, & Adams-CampbellLL (2018). Developing a mass media campaign to promote mammography awareness in African American women in the nation’s capital. Journal of Community Health, 43(4), 633–638. 10.1007/s10900-017-0461-129280089

Images of Hear Her assets used in the surveys to determine if participants recognized previous exposure to the Hear Her campaign.

Characteristics of pregnant and postpartum women participating in Wave 1 and Wave 2 of internet panel surveys used to assess the Hear Her campaign using unmatched data.a

Respondent CharacteristicsWave 1 (N = 762)n%Wave 2 (N = 864)n%
Segmentb
Postpartum521(68.4)554(64.1)
Pregnant241(31.6)310(35.9)
Geographical locationc
Priority markets262(34.4)344(39.8)
Rest of the U.S.500(65.6)520(60.2)
Race/ethnicity
Hispanic84(11.0)150(17.4)
Non-Hispanic White294(38.6)328(38.0)
Non-Hispanic Black314(41.2)327(37.9)
Non-Hispanic another race/multiple raced70(9.2)59(6.8)
Household income
< $50,000382(50.1)278(32.2)
$50,000–$74,999162(21.3)187(21.6)
≥ $75,000218(28.6)399(46.2)
Age
18–34602(79.0)623(72.1)
≥ 35160(21.0)241(27.9)
Education
High school or less184(24.2)150(17.4)
Some college or more578(75.9)714(82.6)

Note: Percentages may not add to 100% due to rounding

Wave 1 was fielded between 8/31/2020–9/21/2020; Wave 2 was fielded between 3/21/2021–4/19/2021.

Pregnant women defined as those currently pregnant. Postpartum women defined as those who had given birth within the last year.

To ensure that survey participants reflected priority audiences, oversampling of participants living in the campaign’s priority markets for media buys (Atlanta, New York, Dallas/Ft. Worth, St. Louis/Kansas City, and New Orleans) were conducted. Other areas outside of these priority markets are defined as Rest of the U.S.

Race and ethnicity included Hispanic, White, Black, Asian, American Indian or Alaska Native, and Other; however, Asian, American Indian or Alaska Native and Other were collapsed into non-Hispanic another race/multiple race because of the small sample size.

Percent awarea of the Hear Her campaign at Wave 1 and Wave 2b overall, and stratified by respondent characteristics using propensity score matched datac

Respondent characteristicsWave 1 (N = 686)%Wave 2 (N = 686)%Percentage point difference between wavesdp-value*
Overall
Aware23.037.2 14.2 <.001
Segmente
 Postpartum
Aware17.934.1 16.2 <.001
 Pregnant
Aware33.842.8 9.0 0.046
Geographical locationf
 Priority markets
Aware30.740.8 10.1 0.016
 Rest of the U.S.
Aware18.834.4 15.6 <.001
Race/ethnicity
 Hispanic
Aware37.449.612.20.089
 Non-Hispanic White
Aware21.834.1 12.3 0.001
 Non-Hispanic Black
Aware18.938.7 19.8 <.001
 Non-Hispanic another race/multiple raceg
Aware26.817.8−9.00.264
Education
 High school or less
Aware12.926.0 13.1 0.006
 Some college or more
Aware25.639.7 14.1 <.001
Age
 18–34
Aware21.034.0 13.0 <.001
 ≥ 35
Aware29.847.0 17.2 0.001
Household income
 < $50,000
Aware17.027.6 10.6 0.002
 $50,000–$74,999
Aware22.839.2 16.4 0.001
 ≥ $75,000
Aware31.746.2 14.5 0.001

Chi-square test was used to test difference in proportion aware of Hear Her between Wave 1 and Wave 2.

Aided campaign awareness.

Wave 1 was fielded between 8/31/2020–9/21/2020; Wave 2 was fielded between 3/21/2021–4/19/2021.

To ensure comparability between the two samples in Wave 1 and Wave 2, propensity score matching was conducted. The Wave variable was used as the binary treatment indicator variable. We used logistic regression modeling to calculate the propensity score (predicted probability) of participating in the survey for each participant, controlling for geographical location (priority markets vs. rest of the U.S.), household income, education, age, and race/ethnicity. Participants in Wave 1 and Wave 2 were then matched based on propensity score using nearest neighbor matching.

Percent increases that correspond with a significant chi-square are noted in bold.

Pregnant women defined as those currently pregnant. Postpartum women defined as those who had given birth within the last year.

To ensure that survey participants reflected priority audiences, oversampling of participants living in the campaign’s priority markets for media buys (Atlanta, New York, Dallas/Ft. Worth, St. Louis/Kansas City, and New Orleans) were conducted. Other areas outside of these priority markets are defined as Rest of the U.S.

Race and ethnicity included Hispanic, White, Black, Asian, American Indian or Alaska Native and Other; however, Asian, American Indian or Alaska Native and Other were collapsed into non-Hispanic another race/multiple race because of the small sample size.

Association between awarenessa of the Hear Her campaign and campaign outcomes of interest among pregnant and postpartum women in Wave 2b of internet panel survey

Awareness of Hear Her campaignCampaign outcomes of interestn (%)
PRc(95% CI)p-valueaPRd (95% CI)p-value
Talk to their doctor when something did not feel right e
YesNo
Overall751(94.6%)43(5.4%)
Aware301(97.1%)9(2.9%)1.04(1.01–1.08)0.0081.04(1.00–1.08)0.055
Unaware450(93.0%)34(7.0%)(ref)--(ref)--
Search for information when something did not feel right f
YesNo
Overall749(95.3%)37(4.7%)
Aware301(97.4%)8(2.6%)1.04(1.01–1.07)0.0161.05(1.01–1.09)0.027
Unaware448(93.9%)29(6.1%)(ref)--(ref)--
Discuss concerns with someone close g
Would Often DiscussWould Occasionally or Not Discuss
Overall544(64.5%)300(35.6%)
Aware222(69.8%)96(30.2%)1.14(1.03–1.25)0.0151.14(1.02–1.25)0.019
Unaware322(61.2%)204(38.8%)(ref)--(ref)--
Desire to receive more information from their health care provider on pregnancy-related complications or warning signs and symptoms h
YesNo
Overall396(45.8%)468(54.2%)
Aware193(60.3%)127(39.7%)1.62(1.39–1.84)<.0011.62(1.36–1.87)<.001
Unaware203(37.3%)341(62.7%)(ref)--(ref)--

Note: Percentages may not add up to 100% due to rounding. PR = Prevalence ratio; CI = Confidence Interval; aPR = Adjusted prevalence ratio

Aided campaign awareness.

Wave 2 was fielded between 3/21/202–4/19/2021

Unadjusted models of aided awareness of the Hear Her campaign messages among pregnant and postpartum women combined, on different outcome variables. Aided campaign awareness of Hear Her was determined by first showing pregnant and postpartum women Hear Her campaign assets followed by the question, “Before now, have you seen these messages or similar messages that use the phrase ‘Hear Her’ and are about women experiencing pregnancy-related complications?” Response choices: Yes; No (No; Not Sure)

Adjusted models were controlled for age, education, household income, race and ethnicity, and geographical location (priority locations vs. rest of the US).

Pregnant and postpartum women were asked, “When something just didn’t feel right during your current pregnancy (or during your most recent pregnancy or up to a year after pregnancy), did you ask your doctor (health care provider) about the problem?” Response choices were categorized as Yes (Yes, every time; Yes, some of the times) and No (No, I didn’t). Those responding that ‘there was never a time when something didn’t feel right’ (n = 70) were excluded from analyses.

Pregnant and postpartum women who responded “Yes, every time,” “Yes, some of the times,” and “No, I didn’t,” to the previous question (talking to doctor) (n = 794) were then asked, “When something just doesn’t feel right during your current pregnancy (or didn’t feel right during or up to year after pregnancy), do you search (or did you search) for more information about the problem?” Response choices were categorized as Yes (Yes, every time; Yes, some of the times) and No (No, I don’t/didn’t). Those responding that ‘there was never a time when something didn’t feel right’ (n = 78) were excluded from analyses.

Pregnant and postpartum women were asked, “When you have (or had) pregnancy-related concerns (or concerns up to a year after pregnancy), how often do (or would) you discuss them with someone that you’re close to (or with someone you were close to)?” Response choices were categorized as Would Often Discuss (Every time; Most times) and Would Occasionally or Not Discuss (Sometimes; Occasionally; Never). Those responding that they ‘did not have pregnancy related concerns’ (n = 20) were excluded from analyses.

Pregnant and postpartum women were asked, “Would you like to receive [or to have received] more information about pregnancy-related complications or warning signs and symptoms from your health care provider?” Response choices: Yes (Definitely yes; Maybe yes) and No (Not sure; Probably not; Definitely not)