[Skip to Navigation]
Sign In
Research Letter
Impact of Policy on Children
June 1, 2020

Frequency and Specificity of Pediatric Health Policy Discussions in Political Campaigns

Author Affiliations
  • 1Mayo Clinic Alix School of Medicine, Rochester, Minnesota
  • 2Department of Political Science, St Olaf College, Northfield, Minnesota
JAMA Pediatr. 2020;174(8):795-796. doi:10.1001/jamapediatrics.2020.0932

Well-designed public policy can promote health among children,1-3 but there is reason to worry that pediatric health issues may get crowded out of policy discussions and debates. Politicians are incentivized to craft their health care policies around the interests of voting-age citizens, and politicians are unlikely to see children as important drivers of health care spending.1,4 To assess the national political profile of pediatric health issues, we analyzed a database of Congressional candidate campaign websites.


As previously described,5 research assistants identified campaign websites for Democratic and Republican candidates for the US House of Representatives. All issue position pages from each website for election cycles from 2008 to 2018 were collected. Archived websites from 2008 to 2012 were accessed through the Library of Congress and Internet Archive websites. Websites from 2014 to 2018 were contemporaneously collected within 3 weeks of each election. Each issues page was assigned up to 3 issue categories (eg, environment, infrastructure) based on which policy domain(s) it addressed. Generally, page titles clearly indicated specific domains (eg, “On Immigration”).

From this database of 37 948 issues pages, we analyzed health care and education issues pages. Many non–health care policy domains may influence childhood health, and education pages merited inclusion because education policy has widespread impact specifically on children and because education pages were more common than pages about other relevant policy domains (eg, poverty, inequality).

We curated health care and education pages through automated keyword searches (Figure 1). The authors collaboratively generated broad keyword lists in an effort to capture all pages that might plausibly relate to pediatric health. Two authors (T.J.B. and G.B.A.) then reviewed texts to determine whether they (1) discussed children’s health and (2) discussed specific policies or initiatives related to children’s health. A text qualified as a pediatric health text if it discussed any goal, policy, or initiative directly related to children’s health. A text qualified as a specific pediatric health policy text if it discussed (1) a method or mechanism for promoting a pediatric health outcome or (2) a particular pediatric health goal more specific than a general commitment to access, affordability, or quality. Coders jointly reviewed a convenience sample of 50 texts from 2018 and excluded these texts from interrater reliability calculations. All other texts were reviewed independently and in duplicate. Coding disagreements were resolved by consensus. For comparison purposes, all health care pages in the database mentioning Medicare were extracted. Two-sided t tests were performed using R version 3.5.1 (R Core Team), with P < .05 considered significant.

Figure 1.  Flow Diagram of Page Identification Strategy
Flow Diagram of Page Identification Strategy

Relevant variations were used for each search term (eg, teenager, teenagers, and teenage). Listed under each category is a generic example of a text that would be included in that category. There were 9 overlapping texts that were counted as both education pages and health care pages. CHIP indicates Children’s Health Insurance Program; SCHIP, State Children’s Health Insurance Program; WIC, women, infants, and children.


Our classification scheme (Figure 1) demonstrated substantial interrater reliability, with 81.2% overall agreement and a Cohen κ of 0.72. We found 631 of 4574 candidate websites (13.8%) included discussions of pediatric health on health care or education issues pages. By comparison, 1523 of 4574 candidate websites (33.3%) mentioned Medicare on health care pages (difference, 19.5%; 95% CI, 17.8-21.2; P < .001). Specific pediatric health policies were discussed on 508 of 4574 websites (11.1%) and on 508 of the 631 websites that mentioned pediatric health (80.5%) (Figure 2). Compared with Republican candidate websites, Democrat candidate websites were significantly more likely to include specific policies, with 422 of 2276 Democratic websites (18.5%) and 86 of 2298 Republican websites (3.7%) including specific policies (difference, 14.8%; 95% CI, 13.0-16.6; P < .001).

Figure 2.  Discussion of Pediatric Health Policy on House Campaign Websites
Discussion of Pediatric Health Policy on House Campaign Websites


Political campaigns are a crucial route by which candidates communicate policy priorities to the electorate, and candidates have a normative obligation to pursue these priorities once elected.6 Childhood provides an ideal window for implementing policies that promote lifelong health, yet only 508 of 4574 House candidate websites (11.1%) propose specific pediatric health policies. Overall discussion of pediatric health varies widely year to year, suggesting macroscopic political trends (eg, presidential elections, public opinion, major legislative initiatives [such as State Children’s Health Insurance Program reauthorizations in 2009 and 2018]) may influence individual campaigns. These findings suggest there is significant work to do to raise the national political profile of childhood health issues, and future studies should examine state and local political contexts. With children facing new threats, such as vaccine-preventable diseases, e-cigarettes, and the high cost of life-saving medications, advocates must ensure that officeholders develop effective policies that address children’s unique health needs.

Back to top
Article Information

Accepted for Publication: January 24, 2020.

Corresponding Author: Tyler J. Benning, BA, Mayo Clinic Alix School of Medicine, 200 First St SW, Rochester, MN 55905 (benning.tyler@mayo.edu).

Published Online: June 1, 2020. doi:10.1001/jamapediatrics.2020.0932

Author Contributions: Mr Benning had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Benning, Chapp.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Benning.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Benning, Chapp.

Administrative, technical, or material support: Ashby, Chapp.

Study supervision: Chapp.

Conflict of Interest Disclosures: None reported.

Meeting Presentation: Data from this study were reported at a poster session of the Midwest Regional American Physician Scientist Association Conference; October 5, 2019; Rochester, Minnesota.

Forrest  CB, Riley  AW.  Childhood origins of adult health: a basis for life-course health policy.   Health Aff (Millwood). 2004;23(5):155-164. doi:10.1377/hlthaff.23.5.155PubMedGoogle ScholarCrossref
Miller  S, Wherry  L.  The long-term effects of early life Medicaid coverage.   J Hum Resour. 2019;54(3):785-824. doi:10.3368/jhr.54.3.0816.8173R1Google ScholarCrossref
Delaney  L, Smith  JP.  Childhood health: trends and consequences over the life course.   Future Child. 2012;22(1):43-63. doi:10.1353/foc.2012.0003PubMedGoogle ScholarCrossref
Galbraith  AA, Carroll  AE, Christakis  D.  JAMA Pediatrics call for papers on election-year policies and children’s health.   JAMA Pediatr. 2019;173(9):813-814. doi:10.1001/jamapediatrics.2019.2636PubMedGoogle ScholarCrossref
Chapp  C, Roback  P, Johnson-Tesch  K, Rossing  A, Werner  J.  Going vague: ambiguity and avoidance in online political messaging.   Soc Sci Comput Rev. 2018;37(5):591-610. doi:10.1177/0894439318791168Google ScholarCrossref
Mansbridge  J.  Rethinking representation.   Am Polit Sci Rev. 2003;97(4):515-528. doi:10.1017/S0003055403000856Google ScholarCrossref