Unintended consequences of disseminating behavioral health evidence to policymakers: Results from a survey-based experiment

Abstract

Background

Communication research demonstrates that messages often have unintended consequences, but this work has received limited attention in implementation science. This dissemination experiment sought to determine whether state-tailored policy briefs about the behavioral health consequences of adverse childhood experiences (ACEs), compared to national policy briefs on the topic, increased state legislators’/staffers’ perceptions of the policy brief relevance and parental blame for the consequences of ACEs, and whether effects differed between Democrats and Republicans.

Method

A preregistered, web-based survey experiment with U.S. state legislators/staffers was conducted in 2021 (n = 133). Respondents were randomized to view a policy brief about the behavioral health consequences of ACEs that included state-tailored data (intervention condition) or national data (control condition) and then answered survey questions. Dependent variables were perceived policy brief relevance and parental blame for the consequences of ACEs.

Results

The mean policy brief relevance score was 4.1% higher in the intervention than in the control condition (p = .24), but the mean parental blame score was 16.5% higher (p = .02). When outcomes were dichotomized, 61.2% of respondents in the intervention condition rated parents as “very much to blame” for the consequences of ACEs compared to 37.1% in the control condition (p = .01). When the sample was stratified by political affiliation, the effect of the state-tailored policy brief on parental blame was larger in magnitude among Democrats and not significant among Republicans. The intervention policy brief increased the mean parental blame score by 22.8% among Democrats relative to the control policy brief (p = .007) and doubled the proportion rating parents as “very much to blame” (52.2% vs. 26.1%, p = .03).

Conclusions

Despite limited statistical power, state-tailored policy briefs significantly increased state legislators’/staffers’ perceptions of parental blame for the behavioral health consequences of ACEs, relative to a policy brief with national data. Unintended messaging effects warrant greater attention in dissemination research and practice.

Plain Language Summary

Prior studies have tested the effects of dissemination strategies on policymakers’ engagement with research evidence. However, little research has assessed the potential unintended consequences of disseminating evidence to policymakers. This knowledge gap is important because a large body of communication research has found that messages often have negative and unintended effects. This paper adds to the literature on disseminating evidence about behavioral health issues to policymakers. A web-based survey experiment was conducted in which U.S. state legislators/staffers were randomized to view a policy brief about adverse childhood experiences (ACEs) that was tailored to present data from their state (intervention condition) or a policy brief that presented national data and then completed a survey. Legislators/staffers who viewed the state-tailored policy brief perceived the brief as slightly more relevant, but also rated parents as being significantly more to blame for the behavioral health consequence of ACEs. The effect of the state-tailored policy brief on parental blame is an unintended messaging effect that signals cause for caution when disseminating evidence about ACEs to policymakers and other practice audiences. More broadly, the findings highlight a need to consider unintended messaging effects in dissemination research and practice.

Keywords

dissemination policy dissemination evaluation

Introduction

Dissemination research—defined as the study of the packaging and communication of research evidence for public health and clinical practice audiences (National Institutes of Health, 2022)—is underdeveloped in the field of implementation science (Ashcraft et al., 2020; Baumann et al., 2022; Brownson et al., 2018; Purtle, Marzalik, Halfond, et al., 2020). Dissemination research that informs how evidence about mental health and substance use (i.e., behavioral health) issues is communicated to policymaker audiences is important given the influence that policymakers have on service systems and the structural determinants of behavioral health (Dopp & Lantz, 2020; Purtle et al., 2017; Purtle, Nelson, Bruns, et al., 2020; Purtle, Nelson, Counts, et al., 2020). Policymaker-focused dissemination research has historically been understudied in implementation science (Hoagwood et al., 2020; Purtle et al., 2015).

However, recent field experiments have tested the effects of manipulating the framing of research evidence on U.S. policymaker engagement with research evidence (e.g., measured by email view and link click rates) and survey-based experiments have tested framing effects on support for evidence-based policies (Brownson et al., 2011; Levine, 2020, 2021; Long et al., 2021; Niederdeppe et al., 2016, 2021; Purtle et al., 2022; Scott et al., 2022; Winett et al., 2021). Although this work contributes a limited evidence base that can inform dissemination practice, it does not shed light on any potential unintended consequences of dissemination. Both of these areas have received very limited attention in dissemination research, among policymakers or other practice audiences.

Petty and Cacioppo’s (1986) elaboration likelihood model of persuasive communication posits that the perceived relevance of dissemination materials is a key construct that spurs engagement with, and cognitive processing of, research evidence. Increased cognitive engagement with dissemination materials, prompted by heightened perceived relevance, may have the benefit of changing knowledge and attitudes in ways that are aligned with evidence (Purtle, Marzalik, Halfond, et al., 2020). However, such cognitive engagement could also have unintended consequences.

The use of heuristics (i.e., cognitive shortcuts) when processing evidence is common and can lead to overly simplistic conceptualizations of the causes of complex problems and negative attitudes toward social groups related to the problem (Cairney & Kwiatkowski, 2017; Mousavi, 2018). A large body of communication research has documented backlash, or “boomerang,” effects in which messages have unintended consequences among some segments of an audience (Byrne & Hart, 2009; McGinty et al., 2018). For example, a 2019 survey-based experiment with state legislators found that the inclusion of narratives about people who would benefit from evidence-based childcare policies in evidence summaries decreased support for these policies among Republicans, while it increased support among Democrats (Winett et al., 2021).

One domain of behavioral health evidence that could be highly susceptible to unintended messaging effects is evidence about adverse childhood experiences (ACEs, e.g., child abuse and neglect, exposure to violence in the home and community; Ellis & Dietz, 2017; Felitti et al., 1998). ACEs are well-established risk factors for behavioral health problems and suicide in adulthood (Afifi et al., 2008; Hughes et al., 2017). Public-, provider-, and policymakers-targeted communication campaigns have sought to spread evidence about ACEs. However, some researchers and practitioners have expressed concerns that haphazard dissemination of ACEs evidence could have unintended consequences (Bethell et al., 2017; Campbell, 2020; Lacey & Minnis, 2020; Macvarish & Lee, 2019; McEwen & Gregerson, 2019).

Recent public opinion research provides some support for these concerns. A 2019 public opinion survey of U.S. adults found that 62.5% perceived parents as “very much to blame,” for the consequences of ACEs (Purtle, Nelson, and Gollust, 2021)—despite the survey emphasizing that ACEs include experiences outside the home (e.g., community violence). A 2020 public opinion experiment found that evidence about the economic impacts of ACEs increased support for policies to address ACEs, but also increased parental blame for the consequences of ACEs (Gollust et al., 2022). These findings are consistent with public opinion research indicating parental blame for childhood obesity (Wolfson et al., 2015) and maternal blame for developmental issues (Richardson et al., 2014; Sharp et al., 2018).

Study Aims

This Short Report presents results from a survey-based dissemination experiment conducted with U.S. state legislators and staffers about the evidence of ACEs being risk factors for behavioral health conditions. The experiment was the final component of a multiple-year dissemination research project focused on communicating evidence about ACEs (Gollust et al., 2022; Purtle et al., 2023; Purtle, Nelson, Gebrekristos et al., 2022; Purtle, Nelson, & Gollust, 2021; Purtle, Nelson, Srivastav et al., 2021). Among other project findings, a field experiment with state legislators found that state-tailored dissemination materials increased engagement with evidence among Democrats but not Republicans (Purtle, Nelson, Gebrekristos et al., 2022). The current study's aims were to (a) determine whether state-tailored policy briefs increase the perceived relevance of policy briefs compared to national policy briefs, (2) determine whether state-tailored briefs affect perceptions of parental blame for the behavioral health consequences of ACEs compared to national policy briefs, and (c) assess whether the direction and magnitude of these effects differ between Democrats and Republicans.

Methods

A preregistered (https://osf.io/cgh64), clustered randomized survey-based dissemination experiment with U.S. state legislators/staffers was conducted. Complete details about the cluster randomization approach and design are provided elsewhere (Purtle, Nelson, Gebrekristos et al., 2022). In short, each U.S. state was randomized to one-of-three policy brief conditions. The intervention policy brief contained state-tailored data about rates of ACEs and state-tailored economic evidence about the costs of ACEs, the enhanced control policy brief contained state-tailored data about rates of ACEs but no economic evidence, and the control policy brief included national data about rates of ACEs and no economic evidence (see Supplemental File A, e.g., policy briefs). The study was approved by the Drexel University Institutional Review Board.

Procedure

All U.S. state legislators were postmailed a one-page letter providing information about the survey and then three email invitations were sent (one per week). The survey was fielded between April 8, 2021, and May 2, 2021. Recruitment materials stated that the survey could be completed by the legislator or a staffer. The survey was sent to 6,153 state legislative officials with valid email addresses in the KnowWho legislative database and completed by 133 respondents (response rate = 2.16%; CONSORT Diagram is Figure 1) from 40 states.

Figure 1

CONSORT Flow Diagram

After informed consent and introductory survey questions, the survey displayed the one-page policy brief that corresponded with the condition to which the respondent's state was randomized. After viewing the policy brief, respondents answered survey questions about the policy brief and their attitudes related to ACEs.

Outcomes and Measures

The independent variable was a categorical indicator of the policy brief condition to which the respondent was randomized. In the preregistration, we planned to combine the two state-tailored conditions and compare results to the control condition and compare all three conditions separately. Because the response rate was lower than anticipated, and thus sample sizes were small across the three study conditions, the analyses reported here combined the two state-tailored conditions (hereafter referred to as the intervention condition).

The primary dependent variable measured the perceived relevance of the policy brief. This was the sum of two 7-point Likert scale items—adapted from prior work (Brownson et al., 2011; Purtle, Dodson, et al., 2018)—which assessed the extent to which the brief was perceived as relevant to respondents “in [their] work” and “residents in [their] state” (1 = Not relevant, 7 = Very relevant; α = .70, r = .56). The secondary dependent variable measured parental blame for the behavioral health consequences of ACEs, also adapted from prior work (Barry et al., 2013; Wolfson et al., 2015). This 7-point Likert scale item assessed the extent to which “parents of children who have experienced [ACEs] are to blame for mental health and substance use problems their children develop as adults” (1 = Not all to blame, 7 = Completely to blame). Consistent with how similar measures have been used in prior work (Purtle, Nelson, & Gollust, 2021), we also created dichotomous measures of the policy brief being perceived as “very relevant” (summed score ≥12) and parents being perceived as “very much to blame” (score ≥ 6).

Covariates included a range of demographic characteristics, including whether the survey was being completed by a legislator or staffer. Covariates also assessed whether the respondent had heard of ACEs prior to the survey and whether the respondent had seen the policy briefs sent during the field experiment portion of the study (Purtle, Nelson, Gebrekristos et al., 2022). These questions were asked before the policy brief was displayed.

Analysis

Consistent with prior dissemination experiments with state legislators (Brownson et al., 2011; Long et al., 2021; Niederdeppe et al., 2016, 2021; Purtle et al., 2022; Scott et al., 2022; Winett et al., 2021) and methodology standards in the field (De Boer et al., 2015; Mutz et al., 2019), our primary analytic approach was unadjusted and used independent sample two-tailed t tests to compare means and chi-square tests to compare differences in dichotomized outcomes. Our secondary analytic approach adjusted for demographic differences across the study conditions, significant at p < .05, in logistic regression models in which the dichotomized outcomes were the dependent variable. We also stratified the sample by political party affiliation and repeated the analyses to assess differences in the intervention effect between Democrats and Republicans.

Results

The sample was comprised of 98 respondents from the intervention condition (state-tailored policy brief) and 35 from the control condition (national policy brief). Table 1 shows demographic characteristics across the study conditions. The proportion of respondents that were legislators, as opposed to staffers, was significantly higher in the intervention than in the control condition (78.6% vs. 57.1%, p = 0.01) as was the proportion with a postgraduate degree (62.1% vs. 27.3, p < .001). Eighty-three percent of respondents had previously heard of ACEs, with no significant difference in the proportion between the study conditions.

Table 1

Demographic and Baseline Characteristics of Experiment Participants

	All (N = 133)		Intervention (n = 98)		Control (n = 35)
	N	%	n	%	n	%	χ²	p
Role
Legislator	97	72.9	77	78.6	20	57.1	6.00	0.01
Staffer	36	27.1	21	21.4	15	42.9	6.00	0.01
Political party
Democrat	90	68.2	67	69.1	23	65.7	0.13	0.72
Republican	42	31.8	30	30.9	12	34.3	0.13	0.72
Age
18–44	38	31.9	28	32.6	10	30.3	0.09	0.96
45–59	40	33.6	29	33.7	11	33.3
60+	41	34.5	29	33.7	12	36.4
Highest level of education
High school or less	3	2.5	0	0.0	3	9.1	16.94	<.001
Technical school, some college, college degree	54	45.0	33	37.9	21	63.6
Postgraduate	63	52.5	54	62.1	9	27.3
Gender
Female	76	62.8	58	65.9	18	54.5	4.69	0.20
Male	42	34.7	28	31.8	14	42.4
Nonbinary	1	0.8	0	0.0	1	3.0
Race/ethnicity
White	106	87.6	77	87.5	29	87.9	0.003	0.96
Black, African American	5	4.1	3	3.4	2	6.1	0.43	0.51
Asian	3	2.5	3	3.4	0	0.0	1.15	0.28
Native Hawaiian, Pacific Islander	2	1.7	2	2.3	0	0.0	0.76	0.38
American Indian or Alaska Native	1	0.8	1	1.1	0	0.0	0.38	0.54
Hispanic, Latino	6	5.0	5	5.7	1	3.0	1.19	0.55
Saw policy brief prior to the survey
Yes	52	39.1	41	41.8	11	31.4	2.86	0.24
No	49	36.8	37	37.8	12	34.3
Not Sure	32	24.1	20	20.4	12	34.3
Heard of ACEs prior to the survey
Yes	110	82.7	80	81.6	30	85.7	0.86	0.65
No	17	12.8	14	14.3	3	8.6
Not sure	6	4.5	4	4.1	2	5.7

Note. ACE = adverse childhood experience.

Main Effects of State-Tailored Policy Briefs on Perceived Relevance and Parental Blame

The state-tailored policy brief slightly increased the perceived relevance of the brief but also significantly increased parental blame for the behavioral health consequences of ACEs. The mean relevance score was 4.1% higher in the intervention than in the control condition (12.4 vs. 11.9, t = 1.12, p = .24) and the mean parental blame score was 16.5% higher (4.6 vs. 3.9, t = 2.37, p = .02; Figure 2A). The magnitudes of these differences were large when the variables were dichotomized (Figure 2B). Among intervention condition respondents, 76.0% rated the policy brief as “very relevant” compared to 68.6% in the control condition (χ² = 0.75, p = .76) while 61.2% of respondents in the intervention condition rated parents as “very much to blame” compared to only 37.1% in the control condition (χ² = 6.04, p = .01). The effect of the state-tailored policy brief on parents being perceived as “very much to blame” remained significant after adjusting for demographic differences between the study conditions (adjusted odds ratio [AOR] = 3.11; 95% CI = 1.23, 7.90, p = .02; Table 2).

Figure 2

Unadjusted Experiment Results (N = 133)

Table 2

Adjusted Effects of State-Tailored Policy Briefs (Intervention) on the Policy Brief Being Perceived as “Very Relevant” and Parents Being Perceived as “Very Much to Blame” for the Behavioral Health Consequences of ACEs, Relative to an Untailored National Policy Brief (Control)

	Policy brief perceived as “very relevant”				Parents perceived as “very much to blame” for behavioral health consequences of ACEs
	AOR	Lower 95% CI	Upper 95% CI	p	AOR	Lower 95% CI	Upper 95% CI	p
Model 1: all respondents (N = 133)	1.17	0.42	3.22	0.76	3.11	1.23	7.91	0.02
Model 2: only Democrats (n = 90)	0.65	0.15	2.92	0.58	4.71	1.32	16.86	0.02
Model 3: only Republicans (n = 42)	2.63	0.54	12.74	0.23	1.65	0.28	9.66	0.58

Note. ACE = adverse childhood experience; AOR = adjusted odds ratio; CI = confidence interval.

Logistic regression results. Reference group = respondents assigned to the control condition. Models adjusting for legislator/staffer role and the highest level of education, variables that differed significantly between the intervention and control conditions. Models 1–3 run separately. “Very relevant” = score ≥12 on a 14-point scale, “very much to blame” = score ≥ 6 on a 7-point scale.

Effects of Policy Briefs, Stratified by Political Party Affiliation

When analyses were limited to Democrats (n = 90), perceptions of policy brief relevance were nearly identical between the intervention and control conditions (mean = 12.7 vs. 12.6, t = 0.18, p = .86; proportion rating the brief as “very relevant” 78.8% vs. 82.6%, χ² = 0.15, p = .69). However, the intervention increased parental blame among Democrats (Figure 3A and C). The mean parental blame score was 22.8% higher among Democrats in the intervention than in the control condition (4.4 vs. 3.5, t = 2.74, p = .007) and the proportion rating parents as “very much to blame” was also significantly higher (52.2% vs. 26.1%, χ² = 4.72, p = .03). This effect persisted after adjusting for demographic differences between the study conditions (AOR = 4.70; 95% CI = 1.32, 16.86, Table 2).

Figure 3

Unadjusted Experiment Results Stratified by Political Party Affiliation of the Respondent

When analyses were limited to Republicans (n = 42), the intervention increased the perceived relevance of the policy brief, although not significantly. The rating of policy brief relevance was 7.9% higher in the intervention than control condition among Republicans (mean = 11.8 vs. 10.9, t = 1.37, p = .18) and 69.0% rated the brief as “very relevant” in the intervention condition compared to 41.7% in the control condition (χ² = 2.65, p = .10; Figure 3B and D). The mean ratings of parental blame were nearly identical in the intervention and control conditions (mean = 5.0 vs. 4.8, t = 0.40, p = .69). However, the proportion of Republicans rating parents as “very much to blame” was notably higher in the intervention than in the control condition (80.0% vs. 58.3%, χ² = 2.08, p = .15).

Discussion

Despite a small sample size and limited statistical power, we found that state-tailored policy briefs significantly increased legislators’/staffers’ perceptions of parental blame for the behavioral health consequences of ACEs, relative to an untailored policy brief with national data. Among Democrat legislators/staffers, we found that the state-tailored policy brief more than quadrupled the odds of a respondent perceiving parents as “very much to blame” for these consequences of ACEs.

We found some evidence that state-tailored policy briefs are perceived as more relevant than national policy briefs among Republicans, but these differences were not statistically significant (which is not surprising given the small sample size of Republicans, n = 42). While we observed that state-tailored policy briefs had almost no effect on perceived relevance among Democrats, levels of perceived relevance were high in the intervention and control conditions (means 11.6 and 11.7, respectively, on a 14-point scale) and the lack of difference could be due to a ceiling effect. High ratings of policy brief relevance among Democrats are consistent with prior research indicating that Democrat state legislators greatly value research evidence about behavioral health (Purtle, Dodson, et al., 2018; Purtle, Lê-Scherban, et al., 2018).

The finding that certain framings of ACE evidence can produce parental blame is consistent with results from the aforementioned public opinion experiment which found that framing ACE evidence in economic terms generated policy support but also parental blame (Gollust et al., 2022). Taken together, these results signal cause for caution when disseminating evidence about ACEs to help ensure that this evidence does not produce or perpetuate parental blame. In dissemination practice, such caution could be expressed by emphasizing that many ACEs occur outside of the home that structural inequities contribute to ACEs and the intergenerational nature of ACEs and historical trauma (Gehlert & Cuestas, 2021). However, such messages should also be tested to help ensure that they do not produce unintended consequences.

Strengths and Limitations

A strength of the study is its randomized design, which enhances internal validity. A key limitation of the study, however, is its low response rate of 2.16%, which creates uncertainty about external validity. The sample skewed Democrat (68.2% of respondents compared to 44.7% of state legislative offices at the time of the study; National Conference of State Legislatures, 2022) and 83% of respondents had heard of ACEs prior to the survey. This proportion having heard of ACEs is substantially higher than that among state legislators in a 2017 survey (33%; Purtle et al., 2019) and among the general public in a 2020 survey (13%; Gollust et al., 2022). Thus, is likely that the sample overrepresents legislators/staffers who are knowledgeable about ACEs. The fact that the state-tailored policy brief had the effect of increasing parental blame among these policymakers—who presumably rely less on heuristics when engaging with evidence about ACEs, given their prior knowledge of the topic—raises questions about whether the magnitude of this effect would be even larger among legislators/staffers who did not have prior knowledge about ACEs.

An additional limitation relates to parental blame being measured by a single item. Although consistent with prior measures of parental blame (Barry et al., 2013; Gollust et al., 2022; Wolfson et al., 2015), study results would be strengthened if replicated using a multiitem parental blame measure.

Conclusions

State-tailored, opposed to national, policy briefs about the behavioral health consequences of ACEs could produce parental blame for the consequences of ACEs among U.S. state legislators/staffers. This effect may be particularly pronounced among Democrats. Dissemination researchers and practitioners should be cognizant of the fact that packaging certain evidence in specific ways can have unintended consequences among some segments of a target audience. The unintended effects of disseminating research evidence warrant greater attention in the field of implementation science.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Robert Wood Johnson Foundation (grant no 76069).

ORCID iD

Félice Lê-Scherban

References

Afifi

T. O.

Enns

M. W.

Cox

B. J.

Asmundson

G. J.

Stein

M. B.

Sareen

(2008). Population attributable fractions of psychiatric disorders and suicide ideation and attempts associated with adverse childhood experiences. American Journal of Public Health, 98(5), 946–952. https://doi.org/10.2105/AJPH.2007.120253

Ashcraft

L. E.

Quinn

D. A.

Brownson

R. C.

(2020). Strategies for effective dissemination of research to United States policymakers: A systematic review. Implementation Science, 15(1), 1–17. https://doi.org/10.1186/s13012-020-01046-3

Barry

C. L.

Brescoll

V. L.

Gollust

S. E.

(2013). Framing childhood obesity: How individualizing the problem affects public support for prevention. Political Psychology, 34(3), 327–349. https://doi.org/10.1093/pubmed/fdq019

Baumann

A. A.

Hooley

Kryzer

Morshed

A. B.

Gutner

C. A.

Malone

Walsh-Bailey

Pilar

Sandler

Tabak

R. G.

Mazzucca

(2022). A scoping review of frameworks in empirical studies and a review of dissemination frameworks. Implementation Science, 17(1), 1–15. https://doi.org/10.1186/s13012-022-01225-4

Bethell

C. D.

Solloway

M. R.

Guinosso

Hassink

Srivastav

Ford

Simpson

L. A.

(2017). Prioritizing possibilities for child and family health: An agenda to address adverse childhood experiences and foster the social and emotional roots of well-being in pediatrics. Academic Pediatrics, 17(7), S36–S50. https://doi.org/10.1016/j.acap.2017.06.002

Brownson

R. C.

Dodson

E. A.

Stamatakis

K. A.

Casey

C. M.

Elliott

M. B.

Luke

D. A.

Wintrode

C. G.

Kreuter

M. W.

(2011). Communicating evidence-based information on cancer prevention to state-level policy makers. Journal of the National Cancer Institute, 103(4), 306–316. https://doi.org/10.1093/jnci/djq529

Brownson

R. C.

Eyler

A. A.

Harris

J. K.

Moore

J. B.

Tabak

R. G.

(2018). Getting the word out: New approaches for disseminating public health science. Journal of Public Health Management and Practice, 24(2), 102–111. https://doi.org/10.1097/PHH.0000000000000673

Byrne

Hart

P. S.

(2009). The boomerang effect: A synthesis of findings and a preliminary theoretical framework. Annals of the International Communication Association, 33(1), 3–37. https://doi.org/10.1080/23808985.2009.11679083

Cairney

Kwiatkowski

(2017). How to communicate effectively with policymakers: Combine insights from psychology and policy studies. Palgrave Communications, 3(1), 1–8. https://doi.org/10.1057/s41599-017-0046-8

10.

Campbell

T. L.

(2020). Screening for adverse childhood experiences (ACEs) in primary care: A cautionary note. JAMA, 323(23), 2379–2380. https://doi.org/10.1001/jama.2020.4365

11.

De Boer

M. R.

Waterlander

W. E.

Kuijper

L. D.

Steenhuis

I. H.

Twisk

J. W.

(2015). Testing for baseline differences in randomized controlled trials: An unhealthy research behavior that is hard to eradicate. International Journal of Behavioral Nutrition and Physical Activity, 12(1), 1–8. https://doi.org/10.1186/s12966-015-0162-z

12.

Dopp

A. R.

Lantz

P. M.

(2020). Moving upstream to improve children’s mental health through community and policy change. Administration and Policy in Mental Health and Mental Health Services Research, 47(5), 779–787. https://doi.org/10.1007/s10488-019-01001-5

13.

Ellis

W. R.

Dietz

W. H.

(2017). A new framework for addressing adverse childhood and community experiences: The building community resilience model. Academic Pediatrics, 17(7), S86–S93. https://doi.org/10.1016/j.acap.2016.12.011

14.

Felitti

V. J.

Anda

R. F.

Nordenberg

Williamson

D. F.

Spitz

A. M.

Edwards

Marks

J. S.

(1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. American Journal of Preventive Medicine, 14(4), 245–258. https://doi.org/10.1016/s0749-3797(98)00017-8

15.

Gehlert

Cuestas

S. M.

(2021). 5 ways to create compelling messages about childhood trauma using data. http://www.bmsg.org/blog/5-ways-to-create-compelling-messages-about-childhood-trauma-using-data/

16.

Gollust

S. E.

Nelson

K. L.

Purtle

(2022). Selecting evidence to frame the consequences of adverse childhood experiences: Testing effects on public support for policy action, multi-sector responsibility, and stigma. Preventive Medicine, 154, 106912. https://doi.org/10.1016/j.ypmed.2021.106912

17.

Hoagwood

K. E.

Purtle

Spandorfer

Peth-Pierce

Horwitz

S. M.

(2020). Aligning dissemination and implementation science with health policies to improve children’s mental health. American Psychologist, 75(8), 1130–1145. https://doi.org/10.1037/amp0000706

18.

Hughes

Bellis

M. A.

Hardcastle

K. A.

Sethi

Butchart

Mikton

Jones

Dunne

M. P.

(2017). The effect of multiple adverse childhood experiences on health: A systematic review and meta-analysis. The Lancet Public Health, 2(8), e356–e366. https://doi.org/10.1016/S2468-2667(17)30118-4

19.

Lacey

R. E.

Minnis

(2020). Practitioner review: Twenty years of research with adverse childhood experience scores—advantages, disadvantages and applications to practice. Journal of Child Psychology and Psychiatry, 61(2), 116–130. https://doi.org/10.1111/jcpp.13135

20.

Levine

A. S.

(2020). Why do practitioners want to connect with researchers? Evidence from a field experiment. PS: Political Science & Politics, 53(4), 712–717. https://doi.org/10.1017/S1049096520000840

21.

Levine

A. S.

(2021). Single conversations expand practitioners’ use of research: Evidence from a field experiment. PS: Political Science & Politics, 54(3), 432–437. https://doi.org/10.1017/S1049096520002000

22.

Long

E. C.

Pugel

Scott

J. T.

Charlot

Giray

Fernandes

M. A.

Crowley

D. M.

(2021). Rapid-Cycle experimentation with state and federal policymakers for optimizing the reach of racial equity research. American Journal of Public Health, 111(10), 1768–1771. https://doi.org/10.2105/AJPH.2021.306404

23.

Macvarish

Lee

(2019). Constructions of parents in adverse childhood experiences discourse. Social Policy and Society, 18(3), 467–477. https://doi.org/10.1017/S1474746419000083

24.

McEwen

C. A.

Gregerson

S. F.

(2019). A critical assessment of the adverse childhood experiences study at 20 years. American Journal of Preventive Medicine, 56(6), 790–794. https://doi.org/10.1016/j.amepre.2018.10.016

25.

McGinty

Pescosolido

Kennedy-Hendricks

Barry

C. L.

(2018). Communication strategies to counter stigma and improve mental illness and substance use disorder policy. Psychiatric Services, 69(2), 136–146. https://doi.org/10.1176/appi.ps.201700076

26.

Mousavi

(2018). What do heuristics have to do with policymaking? Journal of Behavioral Economics for Policy, 2(1), 69–74. https://sabeconomics.org/journal/RePEc/beh/JBEPv1/articles/JBEP-2-1-10.pdf

27.

Mutz

D. C.

Pemantle

Pham

(2019). The perils of balance testing in experimental design: Messy analyses of clean data. The American Statistician, 73(1), 32–42. https://doi.org/10.1080/00031305.2017.1322143

28.

National Conference of State Legislatures. (2022). State partisan composition. https://www.ncsl.org/research/about-state-legislatures/partisan-composition.aspx

29.

National Institutes of Health. PAR-19-274. https://grants.nih.gov/grants/guide/pa-files/PAR-19-274.html

30.

Niederdeppe

Roh

Dreisbach

(2016). How narrative focus and a statistical map shape health policy support among state legislators. Health Communication, 31(2), 242–255. https://doi.org/10.1080/10410236.2014.998913

31.

Niederdeppe

Winett

L. B.

Fowler

E. F.

Gollust

S. E.

(2021). Evidence-based message strategies to increase public support for state investment in early childhood education: Results from a longitudinal panel experiment. The Milbank Quarterly, 99(4), 1088–1131. https://doi.org/10.1111/1468-0009.12534

32.

Petty

R. E.

Cacioppo

J. T.

(1986). The elaboration likelihood model of persuasion. In Communication and persuasion (pp. 1–24). Springer.

33.

Purtle

Bowler

Boughter-Dornfeld

Nelson

K. L.

Gollust

S. E.

(2023). Newspaper coverage of adverse childhood experiences and toxic stress in the United States, 2014–2020: Consequences, causes, and solutions. Trauma, Violence, & Abuse, 24(1), 313–323. https://doi.org/10.1177/15248380211029407

34.

Purtle

Brownson

R. C.

Proctor

E. K.

(2017). Infusing science into politics and policy: The importance of legislators as an audience in mental health policy dissemination research. Administration and Policy in Mental Health and Mental Health Services Research, 44(2), 160–163. https://doi.org/10.1007/s10488-016-0752-3

35.

Purtle

Dodson

E. A.

Nelson

Meisel

Z. F.

Brownson

R. C.

(2018). Legislators’ sources of behavioral health research and preferences for dissemination: Variations by political party. Psychiatric Services, 69(10), 1105–1108. https://doi.org/10.1176/appi.ps.201800153

36.

Purtle

Lê-Scherban

Wang

Brown

Chilton

(2019). State legislators’ opinions about adverse childhood experiences as risk factors for adult behavioral health conditions. Psychiatric Services, 70(10), 894–900. https://doi.org/10.1176/appi.ps.201900175

37.

Purtle

Lê-Scherban

Wang

Shattuck

P. T.

Proctor

E. K.

Brownson

R. C.

(2018). Audience segmentation to disseminate behavioral health evidence to legislators: An empirical clustering analysis. Implementation Science, 13(1), 1–13. https://doi.org/10.1186/s13012-018-0816-8

38.

Purtle

Marzalik

J. S.

Halfond

R. W.

Bufka

L. F.

Teachman

B. A.

Aarons

G. A.

(2020). Toward the data-driven dissemination of findings from psychological science. American Psychologist, 75(8), 1052–1066. https://doi.org/10.1037/amp0000721

39.

Purtle

Nelson

K. L.

Bruns

E. J.

Hoagwood

K. E.

(2020). Dissemination strategies to accelerate the policy impact of children’s mental health services research. Psychiatric Services, 71(11), 1170–1178. https://doi.org/10.1176/appi.ps.201900527

40.

Purtle

Nelson

K. L.

Counts

N. Z.

Yudell

(2020). Population-based approaches to mental health: History, strategies, and evidence. Annual Review of Public Health, 41, 201–221. https://doi.org/10.1146/annurev-publhealth-040119-094247

41.

Purtle

Nelson

K. L.

Gebrekristos

Lê-Scherban

Gollust

S. E.

(2022). Partisan differences in the effects of economic evidence and local data on legislator engagement with dissemination materials about behavioral health: A dissemination trial. Implementation Science, 17(1), 1–15. https://doi.org/10.1186/s13012-022-01214-7

42.

Purtle

Nelson

K. L.

Gollust

S. E.

(2021). Public opinion about adverse childhood experiences: Social stigma, attribution of blame, and government intervention. Child Maltreatment, 27(3), 344–355. https://doi.org/10.1177/10775595211004783

43.

Purtle

Nelson

K. L.

Srivastav

Gollust

S. E.

(2021). Perceived persuasiveness of evidence about adverse childhood experiences: Results from a national survey. Academic Pediatrics, 21(3), 529–533. https://doi.org/10.1016/j.acap.2020.05.031

44.

Purtle

Peters

Brownson

R. C.

(2015). A review of policy dissemination and implementation research funded by the National Institutes of Health, 2007–2014. Implementation Science, 11(1), 1–8. https://doi.org/10.1186/s13012-015-0367-1

45.

Richardson

S. S.

Daniels

C. R.

Gillman

M. W.

Golden

Kukla

Kuzawa

Rich-Edwards

(2014). Society: Don't blame the mothers. Nature News, 512(7513), 131–132. https://doi.org/10.1038/512131a

46.

Scott

Pugel

Fernandes

Cruz

Long

E. C.

Giray

Storace

Crowley

D. M.

(2022). Cutting through the noise during crisis by enhancing the relevance of research to policymakers. Evidence & Policy, 1–18. https://doi.org/10.1332/174426421X16535828173307

47.

Sharp

G. C.

Lawlor

D. A.

Richardson

S. S.

(2018). It's the mother!: How assumptions about the causal primacy of maternal effects influence research on the developmental origins of health and disease. Social Science & Medicine, 213, 20–27. https://doi.org/10.1016/j.socscimed.2018.07.035

48.

Winett

L. B.

Niederdeppe

Gollust

S. E.

Fowler

E. F.

(2021). When “tried and true” advocacy strategies backfire: Narrative messages can undermine state legislator support for early childcare policies. The Journal of Public Interest Communications, 5(1), 45–45. https://doi.org/10.32473/jpic.v5.i1.p45

49.

Wolfson

J. A.

Gollust

S. E.

Niederdeppe

Barry

C. L.

(2015). The role of parents in public views of strategies to address childhood obesity in the United States. The Milbank Quarterly, 93(1), 73–111. https://doi.org/10.1111/1468-0009.12106