Background

Health promotion and disease prevention are essential for improving public health and alleviating the burden of illness in the population. They aim to reduce risk factors and increase healthy behaviours, like healthy lifestyles, vaccinations, early detection screenings, and timely interventions [1]. Equally crucial for effective prevention and health promotion is health literacy, which refers to the ability to access, understand, evaluate, and use health-related information. Higher health literacy enables individuals to make informed decisions about their health, and to adopt healthy behaviours leading to better health outcomes and public health [2]. Conversely, low health literacy is associated with poorer health outcomes and lower use of health care services [3]. In the digital age, health literacy also involves navigating online sources and assessing their reliability [4–6].

Myriad sources and communication formats provide health information, and thus convey the same knowledge in different ways, leading to different levels of comprehension and engagement. Most importantly different sources of information influence behaviour differently [7]. Therefore, it is relevant to assess which are the preferred sources used by the population to inform the best practices to tailor public health education, however considering that individuals encounter different sources of health-related information simultaneously, and thus there is an interlay of influences from all sources.

Primary healthcare providers, such as general practitioners (GPs), specialist physicians (e.g. cardiologists), and pharmacists, are the first point of contact for patients in the healthcare system and play a key role in managing various health conditions. A study in Switzerland showed that 16.4% of the population consulted their general practitioner within two months, while 8% engaged with a private care specialist, and 2% each visited emergency departments or pharmacies [8]. Trust in primary care is a key determinant of patient satisfaction, health outcomes, and adherence to medical advice. A study in the United States showed that a longer patient-physician relationship in primary care improved communication, knowledge, and trust [9]. In turn, trust associated with higher rates of preventive services, such as influenza vaccination, mammography, and eye examinations for diabetics [9]. However, another study in rural areas of the United States found no significant associations between physician continuity and use of recommended preventive services [10].

Family and friends are important sources of health-related information and support for individuals, particularly older adults [11]. They help to manage personal health information and shape health behaviours through trust and influence [12].

The internet and social media are important tools for health promotion and disease prevention. They offer a space for sharing experiences, seeking advice, and finding support, as well as providing health-related information. However, the quality and reliability of online information vary widely, and users need to evaluate the sources carefully.

Traditional media, including television (TV), radio, and newspapers have long been a trusted source of information for many people. These platforms often disseminate health-related information to the public, ranging from preventive measures to updates on public health crises. The trust in traditional media stems from the perception that these media outlets are regulated and adhere to professional standards, although their trust for health information has been declining over the past decade [13,14] while it has increased for interpersonal sources, like physicians [15].

Worldwide and in Switzerland, chronic diseases are mainly caused by alcohol, smoking, physical inactivity, and unhealthy diet. Switzerland aligns its efforts to prevent these issues with the World Health Organization (WHO) action plans both globally and in Europe [16]. However, other health issues also require attention, such as long COVID, mental health, cardiovascular disease, lower respiratory infections, diabetes, dementia, and population ageing [17].

Objectives

Health issues vary in salience and trust among different demographic groups, which implies the need for tailored health information to reach diverse populations effectively. To produce such information, it is necessary to assess the population’s actual behaviour to learn about health. Thus, this study aims to investigate the sources of information, the level of trust, and the health topics that a representative sample of the Swiss population used to learn about health promotion and disease prevention in the previous 12 months.

Study design

A nationwide cross-sectional online survey conducted in a representative sample of the adult Swiss general population.

Setting, participants’ selection and data acquisition

In November 2022 (from 30.10.2022 to 12.11.2022) the LINK Institute, Zürich, Switzerland (https://www.link.ch/) invited 14’501 adult persons to participate in the web-based survey in order to collect 2000 complete and valid questionnaires. The participants were selected from an online panel of 115’000 active recruited members, based on quotas (e.g., quota 1 = German- speaking region, male, between 18 and 29 years old). No contacts were excluded, and Rim weighting [18] was used to adjust for quotas overflow and to ensure that the drawn sample was representative of the Swiss population, according to age, sex, and language region [19].

Survey and variables

The survey utilised a fully structured questionnaire consisting of both semi-open and closed questions, with an average completion time of approximately five minutes. The survey was developed in our Institute considering the most common different sources of information to learn about health promotion and disease prevention. It was internally revised for clarity and comprehension by the authors of the study and by the Link Institute staff responsible for the project. After approval, the Link Institute translated it into German, French and Italian.

Participants reported their actual behaviours regarding the acquisition of health-related information. They had a randomized list of media and interpersonal information sources, and for each item they had to indicate if they had used it or not (binary response yes/no) during the previous 12 months.

They also indicated the level of trust they attributed to each source, on a four-point scale, spanning from one (no trust at all) to four (a significant degree of trust). The specific media sources were Tv, radio, print (magazines and newspapers), social media (Facebook, X (Twitter), Instagram, TikTok, Reddit, WhatsApp/Telegram groups), Swiss federal websites (i.e. Ministry of Health website), foreign authorities websites (i.e. US Food and Drug Administration), news websites, books, and scientific journals. The specific interpersonal sources included general practitioner (GP), specialized physician, pharmacy, health insurance, family and friends, and community organizations (i.e. clubs). Furthermore, they wrote which specific health topics they had sought information on, excluding COVID-19.

The survey also asked if participants were registered with a (GP), the frequency of preventive healthcare discussions with their GP, and a self-assessment of their own health status on a four-point scale (ranging from one, poor health, to four, very good). Essential demographic information was collected including: sex, age (and age group 18–29, 30–44, 45–59, 60+ years old), language-speaking-region (German, French, Italian), residential environment (urban, rural), household size (one to two people, three or more people), education level (primary-medium education, higher education), household income (< 6000 Swiss francs (CHF,) 6001–10000 CHF > 10000 CHF, No answer), profession (owner/executive office, employee, learning, at home, not working), and, employment status (yes part-/full-time, no). Detailed survey questions are accessible in the (S1 Survey in S1 File).

Analysis

The analysis was done on weighted data, with R software [20] and primarily with the R package "survey" [21]. Descriptive analysis was employed to examine the frequency of use of each information source and their associated trust rating as a binary response (no/low trust, responses 1 and 2 vs high trust, responses 3 and 4). Social media sources were merged into a “social media” category including YouTube, Facebook, X (Twitter), Instagram, TikTok, Reddit, and WhatsApp/Telegram groups.

Health topics were assessed with an open question. The responses to this open-ended question were categorised by the first author. Each answer was looked individually and assigned as many codes as necessary from the International Classification of Primary Care, 2^nd edition (ICPC2 [22]). ICPC2 is an accepted and commonly used coding system to record health problems and medical measures in primary care settings. It provides a standardised way to code information form primary care, facilitating data collection and analysis. Additional categories were created (not present in ICPC2), namely “none, “COVID”, “alternative remedies”, and “insurance/registry”. Although individuals were asked to exclude to report information seeking for COVID-19, a considerable number of participants mentioned searching for COVID-19. Therefore, we were obliged to add an additional category labelled “COVID-19”. Questions about alternative remedies and insurance issues are not covered by the ICPC2 coding system. When necessary particular cases were discussed with the last author until an agreement on the most appropriate classification category was found. The frequency of appearance of each health topic category was computed, in total, per age group, and per sex.

Multivariable logistic regression was used to assess the differential reported use of health-information sources per sex and age group controlling for language-speaking-region, residential environment, household size, education, household income, and employment status.

Ethical statement

Guidance was sought from the Ethics Committee of Northwest and Central Switzerland (EKNZ), which advised that formal ethical approval was not required since the survey complies with the general ethical and scientific standards for research with humans (Project-ID: Req-2022-01203).

Participants gave written consent at the beginning of the online survey, and data was collected anonymously.

Acquisition of health-related information: Sources’ use and trust per sex and age group

In the context of the past 12 months, the prevailing sources of health-related information most accessed by participants were the GP, government websites (e.g., Ministry of Health website), family and friends, pharmacy, and television. Fig 1 visually portrays the use percentages of all studied sources, stratified by both sex and age group.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Use and trust of health information sources per age and sex.

web. = websites, Spec. physician = specialised physician, FAW = Foreign Authority Website.

https://doi.org/10.1371/journal.pone.0312120.g001

Fig 1 also shows the degree of trust attributed to each source of information. The most trusted sources were the specialised physician, the GP, the pharmacy, and government websites.

Multivariable logistic regression controlling for demographic variables (language speaking region, residential environment, household size, education, income, and employment status), indicated the differential use and trust of each information source per sex and age (five-year increments). Results are shown in Fig 2.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Association between age and sex and the use and trust of sources of health information (presented as ORs).

Multivariable regressions modeling the association between age and sex, and the use and trust of each health information source studied, controlling for language region, residential environment, household size, education, income, and employment status. Regressions assessing the use of sources used 1997 observations, and regressions assessing the trust of sources used from 1210 observations (social media) to 1963 observations (general practitioner). Results are presented as Odds Ratios (OR) inside the squares. web. = websites, Spec. physician = specialised physician, FAW = Foreign Authority Website.

https://doi.org/10.1371/journal.pone.0312120.g002

Health-related topics searched

All ICPC2 codes except “Z- Social problems” were present in the respondents’ answers. Table 2 displays the frequency of health topics per category overall, and per sex, and age group.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Frequency of health topics searched per sex and age group.

https://doi.org/10.1371/journal.pone.0312120.t002

We created a summary table accessible in the (S2 Summary in S1 File), including the most searched topics per sex and age group, along with the corresponding most used and trusted sources in the same population segments.

Limitations and future directions

The cross-sectional online nature of this survey study has several inherent limitations. First, the data collected represents a specific point in time and does not account for changes over time. This restricts our ability to infer causality or temporal relationships between variables. Second, the reliance on self-reported data may introduce response bias, as participants might provide socially desirable responses or misinterpret questions. Third, despite efforts to ensure a diverse sample, the online nature of the survey may exclude individuals without internet access or digital literacy skills, potentially limiting the generalizability of our findings. Lastly, non-response bias could be present if those who chose to participate in the survey differ significantly from those who did not.

The recent COVID-19 pandemic may have influenced our results. Notably, 4% of participants reported searching for COVID-19, despite explicit instructions to exclude it. This instruction could have been both advantageous and disadvantageous. On one hand, it encouraged participants to consider other health-related topics of interest, potentially limiting responses solely focused on COVID-19 due to its prominence. On the other hand, the mention of COVID-19 in the question might have inadvertently prompted participants to think about it. This highlights the complexity of conducting research during a global pandemic and the potential influence of such significant events on participant responses.

Health literacy is associated with the trust placed in an information source, and in turn, trust is related to behavioural change [15]. Unfortunately, we did not measure health literacy in our sample and we suggest future studies to assess it to gain a more comprehensive understanding of health information-seeking behaviours to improve health education strategies.

Our study has some limitations. First, the topics were assessed with an open question, and primarily the first author coded the responses, discussing unclear cases with the last author. Second, some health topics reported were too unspecific to be assigned to an ICPC2 category. For example, for issues affecting the genital apparatus, if not specified whether it was for men or women, both categories (X and Y) were assigned. This approach ensures a comprehensive analysis despite the lack of specificity. Furthermore, using an internationally recognized classification system in general practice such as ICPC2 can enhance the clarity and credibility of the results. Further, it ensures that the findings align with established standards, and are comparable and usable for an international audience.

Implications for health promotion and prevention

This study elucidates the sources used by the adult Swiss population to learn about health topics and evaluates the sources’ trustworthiness. It also identifies the most frequently searched health topics, indicating areas that need increased educational efforts. The use of these findings by policymakers could potentially enhance the effectiveness of their health-related campaigns. Tailoring communication channels and formats to align with source preferences, collaborating with healthcare providers to bolster patient education, and leveraging digital platforms for targeted outreach, could accomplish this. In terms of content development, aligning health content with community interests, would make it relevant to the population, and thus would contribute to optimize educational initiatives. Moreover, it is imperative to monitor regularly health-related information against scientific evidence, and intervene when necessary. Concurrently, promoting health literacy and critical thinking skills targeted at health-related information across various sources would be highly beneficial.