Associations between adverse childhood experiences and trust in health and other information from public services, professionals and wider sources: national cross sectional survey

Introduction

Adverse childhood experiences (ACEs) include child physical, verbal and sexual maltreatment as well as exposure to other potential sources of trauma, such as growing up in domiciles with domestic violence, mental ill health and substance misuse.1 Exposure to ACEs is consistently associated with low educational attendance and attainment, school suspension and expulsion, and reduced employment prospects.2–5 ACEs are also linked to the adoption of health harming behaviours, such as smoking, alcohol misuse, illicit drug use and sexual risk taking6; involvement in antisocial behaviours, crime and violence7 8; and poorer diets and obesity.9 10 Across the life course, higher levels of ACE exposure are predictive of poorer mental health6 11 and associated with earlier development of non-communicable disease, including cancer, cardiovascular disease, and subsequent premature mortality.6 12 13 Increasingly biological and epigenetic markers of ill health and premature ageing are being disproportionately identified in those exposed to ACEs, indicative of causal relationships.14–16 ACEs are a common feature of populations in high, middle and low income countries, and predict poorer life course outcomes across each economic level.17–19 A meta-analysis of data from 206 ACE studies in 22 countries found that 60% of adults had been exposed to at least one ACE while growing up, with 16% exposed to four or more ACEs.17 Consequently, preventing ACEs and mitigating their potential impacts are major public health considerations as well as factors of substantial economic importance, with the health burden attributable to ACEs having been estimated to cost over US$1 trillion per annum across North America and Europe alone.20

Individuals who have suffered ACEs may benefit from health promotion information that, for instance, advises reductions in substance use or adherence to health guidelines and encourages healthier diets and exercise. Equally, health protection advice on safer sex, routine infection, biometric screening and vaccine uptake is likely to be especially pertinent to those with ACEs, and therefore at higher risks of infection or ill health.6 15 21 Early engagement with health services when ill health develops is also a predictor of better outcomes22 23 and likely to be especially important to those with ACEs who may develop chronic disease earlier in life. However, studies suggest that exposure to ACEs can influence individuals’ likelihood of: adopting health advice on behaviour change;24 vaccination uptake;25 engaging in screening programmes which might lead to earlier health and social interventions;26 27 and complying with medication and other clinical treatment regimens.28 Recent pandemic conditions highlighted poorer uptake of both health protection advice and COVID-19 vaccination in those with ACEs,29 increasing risks of infection and transmission.

A major factor associated with adoption of public sector advice and successful engagement with health, social and other support services is levels of trust in sources and providers.30–32 Moreover, with commercial websites, charities, TV, radio, peers and other sources often providing information independent of the public sector sources, relative trust in public sector versus other commercial, charity and social media outlets is an important consideration, especially when trying to reach and support vulnerable groups.33 However, relatively little is known about how ACEs may impact levels of trust in health advice, public and other systems, or general sources of health and other information.

Here, we used data from a national survey to examine levels of trust in different public, private, social and charitable agencies and professionals. We tested the hypothesis that levels of low trust will vary according to the number of ACEs experienced. Based on the findings, we considered which individuals, agencies and communication platforms are best placed to engage and influence those with a history of ACEs.

Materials and methods

A national household survey of residents in Wales (aged ≥18 years) was conducted between November 2022 and March 2023. A methodology using face-to-face surveys at participants’ places of residence was chosen as this has provided higher levels of compliance in previous ACE studies than, for instance, telephone based surveys.34 Moreover, data on location of domicile (used in residence based surveying) provides an ecological measure of deprivation that can be used in stratified sampling. Such data may not be available when sampling through telephone number lists. Sampling and data collection were undertaken by a professional market research company (MRC). To ensure adequate capture of individuals with exposure to multiple ACEs (≥4, n~200), a target sample of 2000 participants was set based on ACE prevalence rates in previous national surveys.35 A stratified quota sampling methodology was used to obtain a sample representative of national sociodemographics. The sampling unit was lower super output areas (LSOAs; geographical units with a population of approximately 1500 people), with sampling stratified by Welsh Health Board and deprivation quintile (using Welsh index of multiple deprivation (WIMD) 2019 scores).36 WIMD is a combination of eight separate domains of deprivation (income, employment, health, education, access to services, community safety, physical environment and housing), with each domain compiled from a range of different indicators.

A proportionate sample of 200 LSOAs was randomly selected from each deprivation quintile in each health board. A target of 10 interviews was set for each selected LSOA, with quota samples by age and sex. Residential addresses within selected LSOAs were identified using the postcode address file. Only one individual per household was eligible to participate, with study inclusion criteria being age ≥18 years, resident in a selected LSOA and cognitively able to participate.

Interviewers were a mix of men and women, all employed by the MRC. Each was trained to ensure they understood the purpose and content of the survey, with training including information on ACEs and the need for sensitivity and objectivity when undertaking interviews. Households were visited by interviewers and, on contact, potential participants were provided with a participant information sheet and a letter of authority from Public Health Wales. Interviewers detailed: the purpose of the survey and how findings would be used; survey content; and the voluntary, confidential and anonymous nature of the survey. All study materials were provided in Welsh and English language, and participants were able to complete the survey in the language of their choice. All participants provided informed consent, recorded electronically within the survey. Face-to-face interviews were conducted at the door via computer assisted personal interviewing with computer assisted self-interviewing for more sensitive questions (eg, ACE questions). Participation took an average of 22 min, following which all participants were provided with a thank you leaflet containing information for appropriate national support services. Just under half (49%) of all individuals contacted agreed to participate in the survey (n=2007).

Results

The sample and, for comparison, national demographics are shown in online supplemental table A2. Just over half of the sample were women (54.5%); 47.1% were aged 18–49 years and 52.9% were aged ≥50 years; 95.6% were of white ethnicity; and proportions in each deprivation quintile ranged from 19.3% (quintile 1, most deprived) to 20.6% (quintile 5, least deprived). Overall, the sample was a good match to general population characteristics for ethnicity and deprivation but included a lower percentage of individuals aged 18–29 years (14.4%) and more women (54.5%) than the national population (19.2% and 51.1%, respectively). Over half of all participants reported 0 ACEs (56.6%), with 17.9% reporting 1 ACE, 14.6% 2–3 ACEs and 10.9% ≥4 ACEs.

For each trust rating, some individuals were categorised as ‘don’t know/not applicable’. In particular, online sources such as websites, health apps and social media were not rated as frequently by older respondents and several health service sources were not rated as frequently by those of other than white ethnicities. Individuals with ACEs were more likely to be able to provide a trust rating for health advice from NHS 111 and for social services. Overall reporting levels for each question are shown in table 1 with breakdowns by demographics and ACE count in online supplemental tables A3–A6.

Trust in health advice

Ability to provide a trust rating in health advice was high for most types of health service staff (>96%) and friends, family or colleagues (90.4%), falling to 69.5% for NHS 111 and 44.2% for virtual health professionals (table 1). Less than 5% of those rating health advice from hospital doctors, nurses and pharmacists, and 7.7% of those rating GPs, reported low trust in this advice (table 1). Low trust increased for remote health advice delivery methods, to 1 in 10 (10.3%) of those rating NHS 111 and almost 1 in 5 (19.1%) of those rating virtual health professionals. Overall, friends, family or colleagues were less trusted than NHS 111 but more trusted than virtual health professionals (table 1).

After accounting for sociodemographic confounders (age, sex, ethnicity and deprivation), ACE count was found to be significantly associated with low trust in health advice from each health professional source, with the likelihood of a low trust response being more than twice as high in those with ≥4 ACEs (vs 0 ACEs, table 2). There was a smaller increase in low trust in health advice from friends, family or colleagues but this was still significantly different between the 0 and ≥4 ACEs categories (table 2). Adjusted means identified an increase in low trust prevalence for health advice from hospital doctors from 3.3% (0 ACEs) to 8.8% (≥4 ACEs), and for health advice from GPs from 5.3% to 10.4%, respectively (figure 1). For health advice from NHS 111, low trust increased from 11.9% (0 ACEs) to 24.1% (≥4 ACEs) and for virtual health professionals from 19.2% to 34.2%, respectively (figure 1).

Table 2

View inline|View popup

Logistic regression analysis of low trust in health advice from different sources by exposure to adverse childhood experiences and sociodemographics

Request permission

Figure 1

Adjusted percentages reporting low trust in different sources of advice and information, by count of adverse childhood experiences (ACE).

Low trust in hospital doctors and nurses was least common in those aged ≥70 years, while low trust in virtual health professionals was highest in those aged 50–69 years (table 2). Sex and ethnicity were not independently associated with trust in any source of health advice, while low trust in GPs was higher in the more deprived quintiles and low trust in friends, family or colleagues highest in the least deprived quintile (table 2 and online supplemental table A7).

Discussion

Positive relationships with parents and other caregivers in childhood are related to stronger attachments,39 40 and help develop the ability to trust others.41 When children have been hurt or neglected by caregivers, their ability to trust others may be diminished along with their ability to form attachment bonds,42–44 even subsequently with their own children.45 Moreover, maltreated children can be more likely to interpret neutral facial expressions as angry or aggressive,46 with immediate implications for developing attachment and trust.43 While such limited findings illustrate the impacts ACEs may have on an individual’s ability to trust others, less empirical information is available on how ACEs may affect trust in information and advice from professionals, and public and private organisations. In a largely separate literature, trust in organisations often considers issues such as their consistency, reliability, transparency, ethics and integrity, and proven competence.47–50 However, such work has generally not considered how individuals’ experiences of ACEs may affect their ability to trust information and advice from organisations and professionals for health and other purposes.

Here, our results identified strong associations between exposure to ACEs and greater prevalence of low trust in different support systems, as well as health and other advice and information from professionals, agencies and information sources. Of particular concern to healthcare and public health systems, individuals with ≥4 ACEs (vs 0 ACEs) were between two and three times more likely to have low trust in GPs, hospital doctors, nurses and pharmacists (table 2). Critically, those with a high ACEs count are more likely to have greater healthcare, health protection and health promotion needs.6 Consequently, low trust may result in poorer uptake of health improving behaviour advice, specifically in those most likely to develop non-communicable diseases, infections and other physical and mental health issues at earlier ages. Furthermore, trust in sources of advice is an important predictor of compliance with such advice.51 Consequently, individuals with higher ACE counts may also be less likely to adopt and maintain fidelity to medication regimens, other treatment plans or vaccination schedules.26 52 Health systems should recognise that those most in need of their support may also be least likely to trust their advice. Recent developments in trauma informed approaches to health and well being have begun considering how all systems may better meet the needs of those who have experienced trauma, including ACEs.53 54 However, further empirical work is urgently needed, especially when exposure to ACEs remains a common experience (here, 43.4% had at least one ACE and 10.9% ≥4 ACEs).

Overall, low trust was much more prevalent in advice and information from virtual (online and apps) or remote (NHS 111) health advice sources than directly from health professionals (table 1), although response levels varied by source (see limitations). Sources carrying the NHS brand or associated with health professions were more trusted than their more generic counterparts (eg, low trust in NHS websites vs general internet sites, table 1), with social media the least trusted source (71.5% low trust; table 1). Regardless, there were strong positive associations between multiple ACE exposure and low trust for all such sources (tables 2–4). With increasing proportions of public health advice and service contact moving to remote platforms (eg, phone and internet), lower trust in these services should be of concern. Moreover, individuals with higher ACE counts are often greater users of health systems and, for instance, adjusted prevalence of low trust in health advice from the NHS 111 service rose from 11.9% with 0 ACEs to 24.1% with ≥4 ACEs (figure 1).

Similar increases in low trust with higher ACEs were seen for all services, including health services (tables 1 and 3). The greatest increase in the likelihood of low trust with ACEs was seen for police (≥4 ACEs vs 0 ACEs; aOR 3.81), although adjusted odds also at least doubled in those with ≥4 ACEs for health and social services, charities and government (table 3). While evidence supports individuals with higher ACEs having more experiences as perpetrators and victims of crime,7 55 results here identified an estimated increase in prevalence of low trust in police services from 17.0% (0 ACEs) to 43.8% (≥4 ACEs; figure 1). Some police forces are already considering how they make services more trauma informed in order to better engage and build trust with those with a history of adversity.56 57 Typically, however, in the UK and elsewhere, such measures are not yet an integrated feature of police training and operations.

Across all services and systems, the highest prevalence of reported low trust was seen for government (table 1). However, public health advice is frequently referred to as government guidance or guidelines and issued by government departments. Examples include advice for healthy eating58 and physical activity,59 while in some countries health warning labels on alcohol products are identified as governmental rather than health service or medical advice (eg, USA;60). Our results suggested labelling information and advice as governmental may lead to lower levels of trust in it and potentially provide commercial organisations wishing to lessen its uptake an opportunity to exploit a less trusted title (eg, governmental guidance). The differences in trust for those with high levels of ACEs were particularly stark. An estimated 23.0% of those with ≥4 ACEs reported low trust in health services, rising to 73.7% with low trust in government (figure 1). Consequently, our results suggest that provision of national health related or other guidance should consider the benefits of an exclusive health service brand rather than a governmental one.

Finally, our results suggested very low levels of trust in social media, with considerably greater trust in more traditional information platforms, such as TV/radio programmes (table 1). Again, for all communication platforms, low trust was significantly increased in those with higher ACEs (table 4, figure 1). However, increasing amounts of public funding are being invested in social media messaging in order to reach different population groups. The value of such investments may need recalibrating if three quarters of those with high ACE counts have low trust in social media and even among those with no ACE exposure most individuals have low trust in advice and information on such platforms.

Conclusions

In the UK and elsewhere, trust in institutions and figures of professional standing or other authority has diminished over recent decades.65 However, public health and healthcare systems rely on trust as a mechanism to ensure advice is followed and for fidelity to treatment. The alternative, seen during the COVID-19 pandemic, is the use of legislation and criminal justice agencies to enforce health measures (eg, restricting movements). The choice of institution or professional with which information is associated may be a critical decision in its credibility. Individuals with ACEs are an important population for health, social, criminal justice and other sectors to influence. Those with ACEs may require disproportionate amounts of public sector or charitable support, and may benefit more from services but be less likely to trust in such services and the advice they provide. Greater trust in services by those with healthier childhoods will only serve to perpetuate intergenerational cycles of adversity and associated inequalities. Thus trauma informed approaches are required that understand who the most credible communicators are and how best to develop trust in essential support services.