Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 May 16.
Published in final edited form as: J Health Commun. 2021 May 16;26(4):272–280. doi: 10.1080/10810730.2021.1927259

Development of a scale to measure trust in public health authorities: Prevalence of trust and association with vaccination

Taylor A Holroyd 1,2, Rupali J Limaye 1,2,3,4,5, Jennifer E Gerber 1, Rajiv N Rimal 3, Rashelle J Musci 6, Janesse Brewer 1, Andrea Sutherland 1,8, Madeleine Blunt 1, Gail Geller 3,7,8,9, Daniel A Salmon 1,3,5
PMCID: PMC8225577  NIHMSID: NIHMS1710255  PMID: 33998402

Abstract

Infectious disease outbreaks highlight the importance of trust in public health authorities to avoid fear and improve adherence to recommendations. There is currently no established and validated measure for trust in public health authorities. We aimed to develop and validate an instrument that measures trust in public health authorities and to assess the association between trust in public health authorities and vaccine attitudes. We developed 20 items to measure trust in public health authorities. After implementing a survey in January 2020, we investigated relationships between the items, reduced the number of items, and identified latent constructs of the scale. We assessed variability in trust and how trust was associated with vaccine attitudes, beliefs, and self-reported vaccine acceptance. The pool was reduced to a 14-item trust in public health authorities scale and we found that this trust model was strongly associated with acceptance of vaccines. Our scale can be used to examine the relationship between trust in public health authorities and adherence to public health recommendations. The measure needs to be validated in other settings to determine whether they are associated with other areas where the public question public health authority recommendations.

Keywords: Vaccine hesitancy, Vaccine policy, Survey research, Child and adolescent health, Public health

INTRODUCTION

The current COVID-19 pandemic highlights the importance of trust in public health authorities to improve public adherence to evidence-based recommendations and minimize unnecessary fear. Public trust in the United States government has decreased over the last several decades, with only 17% of Americans trusting the government to do what is right most of the time. This erosion of trust is particularly concerning from a public health perspective, especially given challenges in measuring trust in public health. Measures of trust in government, healthcare providers (Anderson and Dedrick 1990, Stecula, Kuru et al. 2020), health insurers (Zheng, Hall et al. 2002, Goold, Fessler et al. 2006), public health after disasters (Eisenman, Williams et al. 2012), medical researchers (Geller, Bernhardt et al. 2005, Hall, Camacho et al. 2006, Yarborough, Fryer-Edwards et al. 2009), the United States vaccine safety system (Frew, Murden et al. 2019), and the health system overall (Ozawa and Sripad 2013) have been previously developed, and from these we can learn about aspects of public trust and how the public views different authorities. However, there is no established and validated measure for trust in public health authorities specifically, such as local and state health departments, the Centers for Disease Control and Prevention, the Food and Drug Administration, and other government health agencies (U.S. Department of Health & Human Services 2003). Trust in public health authorities can affect public attitudes and behaviors, and ultimately have a substantial impact on health decisions and adherence to health recommendations (Altman and Morgan 1983). Furthermore, the level of trust in public health authorities across the United States population or subpopulations has not previously been characterized, preventing us from fully understanding how trust in public health authorities can impact health behaviors like vaccine uptake.

Public health authorities provide guidance and countermeasures to the public, but this system is only effective if the public trusts the information provided and adheres to recommendations. Studies have demonstrated the importance of trust in government authorities for effective risk communication (Peters, Covello et al. 1997), the importance of trust in healthcare providers for adherence to clinical recommendations (Hevey 2007, Dunbar-Jacob 2012, Orom, Underwood et al. 2018), and the interplay between trust in government and trust in science (Wellcome Trust 2018). Specific trust in public health authorities, however, has not been widely examined.

Vaccines provide an excellent example of why trust in public health authorities and adherence to their recommendations are crucial. Declining confidence in immunization has contributed to vaccine refusals, increased exemptions to school immunization requirements, and outbreaks of vaccine-preventable diseases (Salmon, Haber et al. 1999, Omer, Pan et al. 2006, Omer, Enger et al. 2008, Glanz, McClure et al. 2009, Atwell, Van Otterloo et al. 2013, Delamater, Leslie et al. 2017). In 2019, the World Health Organization declared vaccine hesitancy one of the top ten threats to global health (World Health Organization 2019). Trust in government was associated with vaccination intention during the 2009 H1N1 influenza pandemic (Quinn, Parmer et al. 2013, Freimuth, Musa et al. 2014). Low trust in government is associated with refusal of childhood vaccines (Salmon, Moulton et al. 2005) and opposition to compulsory vaccination policies (Taylor-Clark, Blendon et al. 2005).

Measuring trust in public health authorities in a valid and reliable way is essential to examine and improve adherence to public health authority recommendations. This study aimed to 1) develop and validate an instrument to comprehensively measure trust in public health authorities (TiPHA scale), 2) characterize the level of trust in public health authorities in the United States, and 3) evaluate whether trust in public health authorities is associated with vaccine attitudes and acceptance.

MATERIALS AND METHODS

Item generation

In order to generate the item pool regarding trust in public health authorities, we first reviewed existing literature relevant to trust in public health authorities from the domains of trust in government (Taylor-Clark, Blendon et al. 2005, Lee, Whetten et al. 2016), risk communication (Peters, Covello et al. 1997, Vaughan and Tinker 2009), and vaccine acceptance (National Vaccine Advisory Committee 2011, Quinn, Parmer et al. 2013, Freimuth, Musa et al. 2014, National Vaccine Advisory Committee 2014). Based on existing literature and in-depth interviews (Limaye, Oloko et al.), we initially identified ten crucial content domains of trust: beneficence, efficiency, innovation, objectivity, competence, equity, transparency, responsiveness, accuracy, and integrity. Using these ten domains, we developed a 20-item pool related to trust in public health authorities (Table 1). Six items were modified from studies by Salmon et al. and Lee et al. addressing issues of beneficence, objectivity, equity, and integrity relating to trust in government (Salmon, Moulton et al. 2005, Lee, Whetten et al. 2016). The remaining 14 items were developed de novo based on constructs identified in the literature. The item pool was reviewed by the research team to ensure content validity and confirm that all identified domains were sufficiently covered.

Table 1:

Initial 20-item pool concerning trust in public health authorities and frequency in study population

Content domain Item Participants agreed or strongly agreed, No. (%)
Beneficence They do everything they should to protect the health of the population a 1646 (86)
They are partly responsible for the illegal drug problems in this country a 803 (42)
Efficiency They use resources well b 1576 (82)
They waste money on health problems b 568 (30)
Innovation They keep trying the same things to help the public, even when they don’t work very well c 942 (49)
They come up with new ideas to solve health problems d 1669 (87)
Objectivity They provide the public with complete and accurate information about important health issues a 1545 (80)
They base recommendations on the best available science d 1668 (87)
Competence They are not always able to help the health of the public d 1366 (71)
They ensure the public is protected against diseases c 1652 (86)
Equity They are more concerned about some racial and ethnic groups than other groups a 741 (39)
They are concerned about all people, without caring about who has more or less money a 1475 (77)
Transparency They sometimes hide information from the public c 1120 (58)
They accurately inform the public of both health risks and benefits of medicines c 1579 (82)
Responsiveness They do not respond appropriately to emergencies and disasters b 626 (33)
They quickly help the public with health problems b 1535 (80)
Accuracy They make unhelpful recommendations d 651 (34)
They provide skewed information c 840 (44)
Integrity They were responsible for creating HIV and AIDS a 541 (28)
They believe in what they recommend for the public c 1648 (86)
Open in a new tab
a

Source references[27, 29]

b

Novel item[28, 33]

c

Novel item[25, 26, 33]

d

Novel item[33]

Study design and sampling

We conducted a panel survey among adults aged 18 and older in the United States in January-February 2020. Participant recruitment was conducted by the web-based survey panel company Qualtrics (Provo, UT). 1,925 participants completed the survey from panel volunteers selected to match the demographic profile of the United States. This study was ruled exempt by the Institutional Review Board at the Johns Hopkins School of Public Health.

Survey development

The 20-item pool regarding trust in public health authorities (Table 1) included both positively and negatively worded items to avoid affirmation bias (Devellis R.F. 2003). Participants used a 4-point Likert scale (“Strongly agree”, “Agree”, “Disagree”, “Strongly disagree”) to indicate how they felt “about public health authorities such as local and state health departments, the Centers for Disease Control and Prevention (CDC), and Food and Drug Administration (FDA)”.

Respondents identified their gender, age, education, household income, race or ethnic group, whether they had children, and age of their youngest child. Different versions of 4-point Likert scales were used to assess vaccine hesitancy, attitudes, beliefs and vaccine acceptance among respondents without minor children (non-parents and parents with adult children over age 18) (Gilkey, Magnus et al. 2014), respondents whose youngest child was aged 10 years and under (Opel, Taylor et al. 2011), and respondents whose youngest child was aged 11 to 17 years (Gilkey, Magnus et al. 2014).

The overall survey was pre-tested in 131 individuals from the same panel to ensure question clarity and completeness of response options. A convenience sample of twenty pre-test individuals were interviewed by phone to further ensure survey clarity. Changes made to the overall survey instrument during pre-testing included re-wording of questions, addition of response options, and removal of 20 questions.

Data analysis

We utilized factor analysis to investigate relationships between the trust items and identify which items compose a scale measuring trust in public health authorities. Principal components analysis with promax oblique rotation was conducted on the 20-item pool, assuming correlations between components. Eigenvalues were examined to identify the number of factors accounting for a large proportion of variation in the items. A scree plot was generated to confirm the number of factors. Only factors with a standard eigenvalue cutoff >1.0 were included (Yeomans and Golder 1982). Exploratory factor analysis with maximum likelihood estimation was conducted to examine the factor loadings and identify whether any items with consistently low factor loadings or high cross-loadings could be removed from the scale and subsequent analyses. We used a standard but stringent factor loading cutoff of >0.7 (Opel, Taylor et al. 2011). The Cronbach α coefficient was calculated to evaluate the internal consistency of the reduced scale and identified factors (Devellis R.F. 2003).

To assess the validity of the TiPHA scale, we utilized confirmatory factor analysis (CFA) to statistically evaluate the number of factors that represented the best and most parsimonious fit for the overall data (Boelen, van den Hout et al. 2008, Atkinson, Rosenfeld et al. 2011, Yen, Sousa et al. 2014). Several fit statistics were used to evaluate which model best represents the data: root-mean-squared error of approximation (RMSEA), comparative fit index (CFI), Tucker Lewis index (TLI), standardized root mean square residual (SRMR), and chi-square. Models that represent a good fit of the data have RMSEA and SRMR values ≤0.08, CFI and TLI values >0.90, and a lower chi-square value compared to other models(Hu 1999). Stata version 14 (College Station, TX) and Mplus Version 8 (Los Angeles, CA) were used for all data analysis.

The prevalence of trust across the population and among subpopulations was evaluated using descriptive statistics and Pearson’s chi-square test for independence. The sum of the 14 TiPHA scale items was calculated for each individual and dichotomized above and below the median value to create a binary variable for reporting high or low trust in public health authorities. We conducted multiple logistic regression to assess whether vaccine attitudes, beliefs, and acceptance were associated with reported trust in public health authorities (Stata version 14, College Station, TX), adjusting for participant demographic information, including gender, income, education, race, age, region, and parent status. Respondents who selected the “don’t know” option regarding specific vaccines were not included in analysis for vaccine acceptance. Respondents who indicated they had accepted or planned to accept vaccines were both counted as accepting vaccines.

RESULTS

Participants

1,925 adults completed the survey. The majority of participants (64%) were the parent of at least one child under age 18. We observed broad variability in participant sociodemographic characteristics (Table 2); these characteristics are similar to the demographic profile of the United States as enrollment quotas were used.

Table 2:

Demographic characteristics of survey respondents and reported level of trust in public health authorities

Characteristic Entire sample (n=1,925), No. (%) Participants with low trust (n = 1037), No. (%) Participants with high trust (n = 888), No. (%)
Age
18–24 years 157 (8) 101 (64) 56 (36)
25–34 years 534 (28) 328 (61) 206 (39)
35–44 years 358 (19) 188 (52) 170 (48)
45–54 years 303 (16) 140 (46) 163 (54)
55–64 years 259 (14) 78 (30) 181 (70)
65 years or older 310 (16) 89 (25) 231 (75)
Race
White 1403 (73) 796 (57) 607 (43)
Black 278 (14) 170 (61) 108 (39)
American Indian / Native American 89 (5) 51 (57) 38 (43)
Asian 120 (6) 67 (56) 53 (44)
Native Hawaiian or Pacific Islander 25 (1) 15 (60) 10 (40)
Not reported 87 (5) 52 (60) 35 (40)
Ethnicity
Hispanic 354 (18) 197 (56) 157 (44)
Non-Hispanic 1564 (81) 713 (46) 851 (54)
Household income
Under $49,999 855 (44) 427 (50) 428 (50)
$50,000 – $99,999 508 (27) 233 (46) 275 (54)
$100,000 – $149,999 236 (12) 118 (50) 118 (50)
$150,000 or more 290 (15) 116 (40) 174 (60)
Education
Some high school or graduate 893 (46) 459 (51) 434 (49)
Some college or college graduate 728 (38) 322 (44) 406 (56)
Post-graduate 277 (14) 114 (41) 163 (59)
Parent status
Not a parent 481 (25) 235 (49) 246 (51)
At least one child ≤10 years of age 724 (38) 387 (53) 337 (47)
At least one child 11–17 years of age 515 (27) 219 (42) 296 (58)
At least one child aged 18 or older 181 (9) 60 (33) 121 (67)
Not reported 24 (1) 14 (58) 10 (42)
Open in a new tab

Participants with low and high trust were compared using Pearson’s chi-square test for independence. All characteristics had p-values <0.05 comparing low and high trust groups.

Survey item reduction

Principal components and exploratory factor analyses identified two factors that emerged with eigenvalues much greater than one (6.80 and 4.39, respectively). A scree plot was generated to confirm that these two factors explained a significant proportion (56%) of variability in the items. Upon examining the two-factor solution, we concluded that six items could be removed from the scale based on low factor loadings <0.7. After removing these items, we re-conducted the analyses and determined that the 14-item two-factor solution most parsimoniously summarized the remaining data, accounting for 62% of variability in trust in public health authorities (Table 3).

Table 3:

Reduced 14-item Trust in Public Health Authorities scale and factor loadings for two-factor model

Factors and items Factor loadings
Factor 1: Beneficence
They do everything they should to protect the health of the population 0.74
They keep trying the same things to help the public, even when they don’t work very well* 0.77
They base recommendations on the best available science 0.77
They are more concerned about some racial and ethnic groups than other groups* 0.78
They are concerned about all people, without caring about who has more or less money 0.74
They accurately inform the public of both health risks and benefits of medicines 0.77
They make unhelpful recommendations 0.73
They believe in what they recommend for the public 0.80
Factor 2: Competence
They are partly responsible for the illegal drug problems in this country* 0.78
They use resources well 0.68
They waste money on health problems* 0.71
They come up with new ideas to solve health problems 0.73
They ensure the public is protected against diseases 0.71
They quickly help the public with health problems 0.74
Open in a new tab
*

Negatively worded items were reverse-coded in data analysis.

Trust in public health authorities was best conceptualized as two factors, beneficence and competence. Factor 1, beneficence (eigenvalue 5.41) accounted for 64% of the variance and contained eight items related to public health authorities helping the public and utilizing resources well (Table 3). Factor 2, competence (eigenvalue 3.30) accounted for 36% of the variance and contained six items related to public health authorities sharing information and solving problems. The items contained in the two factors are distinctive enough that they comprehensively capture the spectrum of trust in public health authorities (Devellis R.F. 2003).

Reliability

We computed the Cronbach α coefficient to assess the internal consistency of the original 20-item pool, the shortened 14-item TiPHA scale, and each of the factor scales. The Cronbach α coefficient for the original 20-item pool was 0.89. The Cronbach α coefficient for the shortened 14-item scale was 0.86. The α coefficients for the two factors individually were 0.92 for beneficence and 0.87 for competence.

Validity

The fit statistics indicate that two-factor model was a significant improvement over one-, three, and four-factor models (Appendix A1). The two-factor model had a lower RMSEA value (0.000), higher CFI and TLI values (0.987 and 0.985 respectively), lower SRMR value (0.019), and a significant change in chi-square compared to one-, three-, and four-factor models. Based on these results and in the interest of parsimony, the two-factor model was selected as the best fit for the data. We found this two-factor structure was invariant across age, education, and race characteristics. These findings provided strong evidence to support a two-factor representation of trust in public health authorities and to establish construct validity of the 14-item TiPHA scale (Appendix A2).

Prevalence of trust in public health authorities

We found moderate variability in trust in public health authorities (Table 1). Most participants agreed or strongly agreed with statements indicating trust, such as that public health authorities base recommendations on the best available science (87%). A substantial minority of participants agreed or strongly agreed with statements indicating distrust, like that public health authorities waste money on health problems (30%). Concerningly, 48% of participants agreed or strongly agreed that public health authorities sometimes hide information from the public.

915 (48%) participants had low trust and 1,010 (52%) had high trust in public health authorities. High trust participants were older, more often white, had higher educational levels and higher household incomes, while low trust participants were younger, more often black or Hispanic, and reported lower education and lower household incomes (Table 2). Among the participants who were parents, those with high trust were more likely to have an adolescent or adult child, while those with low trust were more likely to have a child aged 10 and under.

Trust in public health authorities was associated with the odds of participants agreeing with statements used to assess vaccine hesitancy (Table 4). Participants with high trust were more likely to agree with positive statements about vaccines and less likely to agree with negative statements. Among parents of children aged 10 and under, high trust in public health authorities was associated with decreased odds of participants reporting that children get more vaccines than are good for them (odds ratio [OR] 0.35, 95% confidence interval [CI] 0.23–0.51), it is better for children to develop immunity by getting sick (OR 0.22, 95% CI 0.14–0.33), and it is better for children to get fewer vaccines simultaneously (OR 0.46, 95% CI 0.32–0.68). Among parents of adolescents aged 11 to 17, high trust in public health authorities was associated with increased odds of participants reporting that vaccines are necessary to protect adolescent health (OR 12.88, 95% CI 4.71–35.20), vaccines do a good job in preventing diseases (OR 14.01, 95% CI 5.19–37.85), and that unvaccinated teenagers may contract a disease such as pertussis or human papillomavirus (OR 3.36, 95% CI 1.90–5.94). Among respondents without minority children, high trust in public health authorities was associated with increased odds of participants reporting that vaccines are necessary to protect adult health (OR 4.24, 95% CI 2.60–6.91), vaccines do a good job in preventing diseases (OR 8.10, 95% CI 4.60–14.25), and unvaccinated adults may contract and spread influenza (OR 3.33, 95% CI 1.49–3.30).

Table 4:

Unadjusted and adjusted odds ratios of participants reporting vaccine hesitancy associated with high reported trust in public health authorities

Vaccine hesitancy statement Participants agreed or strongly agreed, No. (%) Unadjusted OR (95% CI) Adjusted OR (95% CI)
Parents of children aged 10 and under (n = 724†)
Children get more vaccines than are good for them. 264 (37) 0.34 (0.24–0.50)* 0.35 (0.23–0.51)*
It is better for children to develop immunity by getting sick than by getting a vaccine. 199 (28) 0.21 (0.14–0.32)* 0.22 (0.14–0.33)*
It is better for children to get fewer vaccines at the same time. 313 (43) 0.48 (0.34–0.69)* 0.46 (0.32–0.68)*
Parents of adolescents aged 11 to 17 (n = 515)
Vaccines are necessary to protect the health of teenagers 471 (92) 12.61 (4.88–32.56)* 12.88 (4.71–35.20)*
Vaccines do a good job in preventing the diseases they are intended to prevent 468 (91) 13.81 (5.36–35.56)* 14.01 (5.19–37.85)*
If I do not vaccinate my child, he/she may get a disease such as pertussis or human papillomavirus (HPV) and cause other people to get sick 440 (85) 3.44 (2.04–5.80)* 3.36 (1.90–5.94)*
Respondents without minority children (n = 686)
Vaccines are necessary to protect the health of adults. 568 (83) 3.96 (2.56–6.12)* 4.24 (2.60–6.91)*
Vaccines do a good job in preventing the diseases they are intended to prevent 572 (83) 7.80 (4.68–13.02)* 8.10 (4.60–14.25)*
If I do not get vaccinated, I may get influenza or the flu and cause other people to get sick. 512 (75) 2.98 (1.62–3.27)* 2.22 (1.49–3.30)*
Open in a new tab

OR, odds ratio. CI, confidence interval. Odds ratios adjusted for age, education, income, race, region, and parent status.

†

While 724 parents of children aged 10 and under participated in the survey, 215 did not receive these vaccine hesitancy questions due to a survey error.

*

p < 0.01

In all sub-populations, participants with high trust in public health authorities were more likely to agree with positive statements about vaccines (Table 5). Among parents of children aged 10 and under, high trust was associated with increased odds of participants reporting that they trust the information they receive from doctors about vaccines (OR 6.34, 95% CI 3.28–12.26), can openly discuss vaccine questions with their doctor or their child’s doctor (OR 4.89, 95% CI 2.29–10.43), vaccines are very safe (OR 3.94, 95% CI 2.36–6.58), and trust pharmaceutical companies to make safe and effective vaccines (OR 5.64, 95% CI 3.69–8.60). Similar associations were observed among parents of adolescents aged 11 to 17 and among respondents without minor children (Table 5). Participants with high trust were also more likely to report they trust vaccines that have been around for longer, but to a lesser extent than the other key beliefs in all sub-populations.

Table 5:

Unadjusted and adjusted odds ratios of participants agreeing with vaccine-related beliefs associated with high reported trust in public health authorities

Key immunization belief Participants agreed or strongly agreed, No. (%) Unadjusted OR (95% CI) Adjusted OR (95% CI)
Parents of children aged 10 and under (n = 724)
I trust the information I receive from doctors about vaccines 621 (86) 6.29 (3.35–11.83)* 6.34 (3.28–12.26)*
I can openly discuss my questions about vaccines with my doctor. 656 (91) 4.94 (2.38–10.27)* 4.89 (2.29–10.43)*
Vaccines are very safe 611 (84) 3.67 (2.29–5.89)* 3.94 (2.36–6.58)*
I trust pharmaceutical companies to make very safe and effective vaccines. 535 (74) 5.95 (3.96–8.96)* 5.64 (3.69–8.60)*
I am more likely to trust vaccines that have been around for a while than newer vaccines. 522 (72) 1.66 (1.19–2.31)* 1.70 (1.20–2.42)*
Parents of adolescents aged 11 to 17 (n = 515)
I trust the information I receive from doctors about vaccines 454 (88) 13.83 (5.36–35.68)* 14.48 (5.36–39.10)*
I can openly discuss my questions about vaccines with my doctor. 471 (92) 5.87 (2.35–14.70)* 6.03 (2.21–16.47)*
Vaccines are very safe 456 (89) 6.47 (3.34–12.53)* 6.83 (3.30–14.11)*
I trust pharmaceutical companies to make very safe and effective vaccines. 403 (78) 8.69 (5.21–14.48)* 10.25 (5.81–18.09)*
I am more likely to trust vaccines that have been around for a while than newer vaccines. 368 (72) 2.05 (1.39–3.01)* 2.17 (1.43–3.29)*
Respondents without minority children (n = 686)
I trust the information I receive from doctors about vaccines 561 (82) 5.01 (3.01–8.32)* 5.54 (3.14–9.77)*
I can openly discuss my questions about vaccines with my doctor. 595 (87) 3.75 (1.98–7.08)* 3.91 (1.85–8.27)*
Vaccines are very safe 551 (80) 4.27 (2.82–6.47)* 4.97 (3.10–7.97)*
I trust pharmaceutical companies to make very safe and effective vaccines. 492 (72) 4.96 (3.44–7.14)* 6.00 (3.95–9.12)*
I am more likely to trust vaccines that have been around for a while than newer vaccines. 439 (64) 1.10 (0.80–1.51) 1.14 (0.80–1.62)
Open in a new tab

OR, odds ratio. CI, confidence interval. Odds ratios adjusted for age, education, income, race, region, and parent status.

*

p < 0.01

Trust in public health authorities was highly associated with adherence to vaccine recommendations. High trust was associated with influenza vaccine acceptance in the 2019–20 flu season among respondents without minor children (OR 1.75, 95% CI 1.22–2.52), and for children of respondents whose youngest child was under age 18 (OR 1.76, 95% CI 1.28–2.42). Among all parents, high trust was associated with acceptance of diphtheria-tetanus-acellular pertussis (OR 3.14, 95% CI 1.45–6.81) and varicella (OR 6.30, 95% CI 2.51–15.83) vaccines for their child. High trust was also associated with acceptance of measles-mumps-rubella vaccine among parents with children aged 10 years and under (OR 9.85, 95% CI 2.20–44.00) and acceptance of meningococcal vaccine among parents with adolescents aged 11 to 17 (OR 2.65, 95% CI 1.08–6.49).

DISCUSSION

Declining trust in government in general and the growing importance of maintaining trust in public health authorities have emphasized the need for an instrument to systematically measure trust in public health authorities and identify how to improve public adherence to recommendations. We demonstrated that the 14-item TiPHA scale has good reliability and validity in the nationally representative population. Two dimensions of trust in public health authorities emerged (beneficence and competence) with good internal consistency. While trust in public health authorities was generally high, a concerning proportion of participants agreed with statements questioning the credibility of public health authorities. Trust in public health authorities was highly associated with vaccine attitudes, beliefs, and acceptance.

Our scale is an important contribution to the literature on trust in public health authorities. While previous trust literature has focused on disciplines outside of public health, little has been explored about how to measure trust in public health authorities. The TiPHA scale should be validated in additional populations, particularly among healthcare providers as they are important users of recommendations from public health authorities. These results provide very useful baseline data against which to measure the degree to which trust in public health authorities changes or has changed since this survey, such as within the COVID-19 context. Future studies should also explore the relationship between trust in public health authorities and adherence to recommendations other than vaccines, such as social distancing and handwashing during COVID-19.

This study has several limitations. It is possible that other survey questions inquiring about vaccine intentions and behaviors may have impacted participant perceptions of trust in public health authorities. The sample was comprised of participants with internet access who were willing and able to complete an online panel survey, creating selection bias and limiting generalizability. These data from a self-selected sample within a non-probability-base panel cannot be used to make inferences about the US population. We must implement the scale in additional populations to further establish scale reliability and validity. We may not have captured people with lower trust or interest in science and, subsequently lower trust in public health authorities, and as such we may have underestimated the measured trust constructs. The TiPHA scale does not identify varying trust in different public health authorities; grouping public health authorities together may make it challenging for participants to respond accurately, for instance if they feel differently about their local health department versus the CDC.

This study establishes our understanding of the two-factor structure of trust in public health authorities. The importance of beneficence and competence are particularly apparent in the context of COVID-19. With the ongoing development and approval of COVID-19 vaccines, this measure will allow us to assess whether trust or distrust in communication from public health authorities have an impact on COVID-19 vaccine uptake. Further validation of our TiPHA scale in additional surveys will facilitate assessment of trust in public health authorities over time. This innovative scale will help policymakers, providers, and public health authorities to better understand public trust, effectively communicate with the public, and improve adherence to recommendations. Further exploration of trust in public health authorities among sub-populations with varying levels of trust will enable the development of targeted interventions to address low trust and improve compliance with public health authorities during both routine and emergency public health activities.

Acknowledgements

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by National Human Genome Research Institute [grant number RM1HG009038].

Footnotes

Declaration of Conflicting Interests

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

Ethical Statement

All authors have agreed on authorship, read and approved the manuscript, and given consent for submission and subsequent publication of the manuscript. This study was ruled exempt by the Institutional Review Board at the Johns Hopkins Bloomberg School of Public Health.

REFERENCES

  1. Altman DE and Morgan DH (1983). “The role of state and local government in health.” Health Aff (Millwood) 2(4): 7–31. [DOI] [PubMed] [Google Scholar]
  2. Anderson LA and Dedrick RF (1990). “Development of the Trust in Physician scale: a measure to assess interpersonal trust in patient-physician relationships.” Psychol Rep 67(3 Pt 2): 1091–1100. [DOI] [PubMed] [Google Scholar]
  3. Atkinson TM, Rosenfeld BD, Sit L, Mendoza TR, Fruscione M, Lavene D, Shaw M, Li Y, Hay J, Cleeland CS, Scher HI, Breitbart WS and Basch E. (2011). “Using confirmatory factor analysis to evaluate construct validity of the Brief Pain Inventory (BPI).” J Pain Symptom Manage 41(3): 558–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Atwell JE, Van Otterloo J, Zipprich J, Winter K, Harriman K, Salmon DA, Halsey NA and Omer SB (2013). “Nonmedical vaccine exemptions and pertussis in California, 2010.” Pediatrics 132(4): 624–630. [DOI] [PubMed] [Google Scholar]
  5. Boelen PA, van den Hout MA and van den Bout J. (2008). “The factor structure of Posttraumatic Stress Disorder symptoms among bereaved individuals: a confirmatory factor analysis study.” J Anxiety Disord 22(8): 1377–1383. [DOI] [PubMed] [Google Scholar]
  6. Delamater PL, Leslie TF and Yang YT (2017). “Change in Medical Exemptions From Immunization in California After Elimination of Personal Belief Exemptions.” JAMA 318(9): 863–864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Devellis RF (2003). Scale Development: Theory and Applications. Thousand Oaks, Sage Publications, Inc. [Google Scholar]
  8. Dunbar-Jacob J, Schlenk E, McCall M. (2012). Patient Adherence to Treatment Outcomes. Handbook of Health Psychology. Baum A, Revenson TA, and Singer JE New York, Psychology Press. [Google Scholar]
  9. Eisenman DP, Williams MV, Glik D, Long A, Plough AL and Ong M. (2012). “The public health disaster trust scale: validation of a brief measure.” J Public Health Manag Pract 18(4): E11–18. [DOI] [PubMed] [Google Scholar]
  10. Freimuth VS, Musa D, Hilyard K, Quinn SC and Kim K. (2014). “Trust during the early stages of the 2009 H1N1 pandemic.” J Health Commun 19(3): 321–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Frew PM, Murden R, Mehta CC, Chamberlain AT, Hinman AR, Nowak G, Mendel J, Aikin A, Randall LA, Hargreaves AL, Omer SB, Orenstein WA and Bednarczyk RA (2019). “Development of a US trust measure to assess and monitor parental confidence in the vaccine system.” Vaccine 37(2): 325–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Geller G, Bernhardt BA, Gardner M, Rodgers J. and Holtzman NA (2005). “Scientists’ and science writers’ experiences reporting genetic discoveries: toward an ethic of trust in science journalism.” Genet Med 7(3): 198–205. [DOI] [PubMed] [Google Scholar]
  13. Gilkey MB, Magnus BE, Reiter PL, McRee AL, Dempsey AF and Brewer NT (2014). “The Vaccination Confidence Scale: a brief measure of parents’ vaccination beliefs.” Vaccine 32(47): 6259–6265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Glanz JM, McClure DL, Magid DJ, Daley MF, France EK, Salmon DA and Hambidge SJ (2009). “Parental refusal of pertussis vaccination is associated with an increased risk of pertussis infection in children.” Pediatrics 123(6): 1446–1451. [DOI] [PubMed] [Google Scholar]
  15. Goold SD, Fessler D. and Moyer CA (2006). “A measure of trust in insurers.” Health Serv Res 41(1): 58–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hall MA, Camacho F, Lawlor JS, Depuy V, Sugarman J. and Weinfurt K. (2006). “Measuring trust in medical researchers.” Med Care 44(11): 1048–1053. [DOI] [PubMed] [Google Scholar]
  17. Hevey D. (2007). Adherence to Health Recommendations. Cardiovascular Prevention and Rehabilitation. Perk J, Gohlke H, Hellemans I, et al. London, Springer London: 293–300. [Google Scholar]
  18. Hu L, Bentler PM. (1999). “Cutoff criteria for fit indices in covariance structure analysis: conventional criteria versus new alternatives.” Struct Equ Modeling. 6: 1–55. [Google Scholar]
  19. Lee C, Whetten K, Omer S, Pan W. and Salmon D. (2016). “Hurdles to herd immunity: Distrust of government and vaccine refusal in the US, 2002–2003.” Vaccine 34(34): 3972–3978. [DOI] [PubMed] [Google Scholar]
  20. Limaye RJ, Oloko OK, Holroyd TA, Omer SB and Salmon DA “Characterizing trust in public health authorities.” Manuscript under development.. [Google Scholar]
  21. National Vaccine Advisory Committee (2011). Vaccine Safety White Paper Final Report. [Google Scholar]
  22. National Vaccine Advisory Committee (2014). “Recommendations from the National Vaccine Advisory committee: Standards for adult immunization practice.” Public Health Rep 129(2): 115–123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Omer SB, Enger KS, Moulton LH, Halsey NA, Stokley S. and Salmon DA (2008). “Geographic clustering of nonmedical exemptions to school immunization requirements and associations with geographic clustering of pertussis.” Am J Epidemiol 168(12): 1389–1396. [DOI] [PubMed] [Google Scholar]
  24. Omer SB, Pan WK, Halsey NA, Stokley S, Moulton LH, Navar AM, Pierce M. and Salmon DA (2006). “Nonmedical exemptions to school immunization requirements: secular trends and association of state policies with pertussis incidence.” JAMA 296(14): 1757–1763. [DOI] [PubMed] [Google Scholar]
  25. Opel DJ, Taylor JA, Mangione-Smith R, Solomon C, Zhao C, Catz S. and Martin D. (2011). “Validity and reliability of a survey to identify vaccine-hesitant parents.” Vaccine 29(38): 6598–6605. [DOI] [PubMed] [Google Scholar]
  26. Orom H, Underwood W 3rd, Cheng Z, Homish DL and Scott I. (2018). “Relationships as Medicine: Quality of the Physician-Patient Relationship Determines Physician Influence on Treatment Recommendation Adherence.” Health Serv Res 53(1): 580–596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ozawa S. and Sripad P. (2013). “How do you measure trust in the health system? A systematic review of the literature.” Soc Sci Med 91: 10–14. [DOI] [PubMed] [Google Scholar]
  28. Peters RG, Covello VT and McCallum DB (1997). “The determinants of trust and credibility in environmental risk communication: an empirical study.” Risk Anal 17(1): 43–54. [DOI] [PubMed] [Google Scholar]
  29. Quinn SC, Parmer J, Freimuth VS, Hilyard KM, Musa D. and Kim KH (2013). “Exploring communication, trust in government, and vaccination intention later in the 2009 H1N1 pandemic: results of a national survey.” Biosecur Bioterror 11(2): 96–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Salmon DA, Haber M, Gangarosa EJ, Phillips L, Smith NJ and Chen RT (1999). “Health consequences of religious and philosophical exemptions from immunization laws: individual and societal risk of measles.” JAMA 282(1): 47–53. [DOI] [PubMed] [Google Scholar]
  31. Salmon DA, Moulton LH, Omer SB, DeHart MP, Stokley S. and Halsey NA (2005). “Factors associated with refusal of childhood vaccines among parents of school-aged children: a case-control study.” Arch Pediatr Adolesc Med 159(5): 470–476. [DOI] [PubMed] [Google Scholar]
  32. Stecula DA, Kuru O. and Jamieson KH (2020). “How trust in experts and media use affect acceptance of common anti-vaccination claims.” The Harvard Kennedy School (HKS) Misinformation Review. [Google Scholar]
  33. Taylor-Clark K, Blendon RJ, Zaslavsky A. and Benson J. (2005). “Confidence in crisis? Understanding trust in government and public attitudes toward mandatory state health powers.” Biosecur Bioterror 3(2): 138–147. [DOI] [PubMed] [Google Scholar]
  34. U.S. Department of Health & Human Services. (2003). “Public Health.” Retrieved June 20, 2019, from https://www.hhs.gov/hipaa/for-professionals/special-topics/public-health/index.html.
  35. Vaughan E. and Tinker T. (2009). “Effective health risk communication about pandemic influenza for vulnerable populations.” Am J Public Health 99 Suppl 2: S324–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Wellcome Trust (2018). Wellcome Global Monitor: How does the world feel about science and health? [Google Scholar]
  37. World Health Organization. (2019). “Ten threats to global health in 2019.” Retrieved September 3, 2019, from https://www.who.int/emergencies/ten-threats-to-global-health-in-2019.
  38. Yarborough M, Fryer-Edwards K, Geller G. and Sharp RR (2009). “Transforming the culture of biomedical research from compliance to trustworthiness: insights from nonmedical sectors.” Acad Med 84(4): 472–477. [DOI] [PubMed] [Google Scholar]
  39. Yen PY, Sousa KH and Bakken S. (2014). “Examining construct and predictive validity of the Health-IT Usability Evaluation Scale: confirmatory factor analysis and structural equation modeling results.” J Am Med Inform Assoc 21(e2): e241–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Yeomans KA and Golder PA (1982). “The Guttman-Kaiser Criterion as a Predictor of the Number of Common Factors.” Journal of the Royal Statistical Society 31(3): 221–229. [Google Scholar]
  41. Zheng B, Hall MA, Dugan E, Kidd KE and Levine D. (2002). “Development of a scale to measure patients’ trust in health insurers.” Health Serv Res 37(1): 187–202. [PubMed] [Google Scholar]

RESOURCES