Theoretical framework

Patients’ perceptions of marketing directed at doctors can be explored further by examining how selective attention influences what information is processed and what is ignored, particularly in the context of the doctor-patient interaction. The concept of selective attention, rooted in cognitive psychology, suggests that individuals can only process a fraction of the stimuli they are exposed to [38–40]. This means that despite being surrounded by multiple forms of information, including marketing messages, many individuals – both doctors and patients – filter out what they consider irrelevant, allowing them to focus on information that aligns with their immediate needs or interests.

The selective attention theories differ in explaining how information is prioritised [41–43]. However, the central concept of selection helps explain why some individuals notice marketing efforts while others do not, even when exposed to them. Notably, for many patients, their primary focus is not on the presence of the pharmaceutical industry in a doctor’s environment but on their health and the information provided by the medical professional. As a result, the presence of pharmaceutical marketing may not register at all. The question can be posed of whether specific socio-demographic characteristics predispose individuals to notice pharmaceutical marketing directed at doctors.

Also, the phenomenon of loss of trust in physicians or healthcare, which may be associated with persuasive pharmaceutical marketing, can be understood through another concept connected with theories of selective attention – specifically, cognitive dissonance [44]. Cognitive dissonance posits that individuals experience psychological discomfort when confronted with two conflicting beliefs or when their behaviour contradicts their beliefs. In the context of selective attention, this discomfort can influence how individuals perceive and filter information. When exposed to information that challenges their existing beliefs or behaviours, individuals are prone to selectively attend to information that reinforces their current attitudes or actions while disregarding or dismissing information that creates dissonance. This process is often referred to as confirmation bias [45].

Thus, when a patient notices a pharmaceutical company logo or other marketing elements in the doctor’s office, they may experience a conflict between this observation and their expectation that the doctor should always act in their best interest. To resolve this cognitive dissonance, the patient might adjust their attitudes or beliefs. One potential response is to reduce their trust in the doctor to reconcile their perception of the doctor’s integrity with the presence of commercial marketing influences. This process could also trigger confirmation bias, where the patient selectively focuses on information that supports their altered view, such as instances where the doctor’s actions appear commercially motivated, while disregarding evidence that might contradict this view. This reinforcement of scepticism about the doctor’s objectivity can be particularly concerning in areas like vaccine distrust. Since trust is a foundational element of the patient-physician relationship, essential for effective healthcare, perceived conflicts of interest between the pharmaceutical industry and physician interactions can lead patients to question the motivations behind treatment recommendations, ultimately undermining care.

Research suggests that age is a significant factor in selective attention, with studies indicating a decline in selective attention capabilities as individuals age [39,46]. Gender has also been shown to play a role in selective attention mechanisms [47,48]. Moreover, the multifaceted responsibilities associated with having children necessitate a high degree of selective attention focused on their needs [49]. Additionally, research from China indicates that young adults who were born and raised in rural areas may process attended and unattended information differently from their urban counterparts [50]. Health status also influences selective attention; studies demonstrate that health anxiety heightens selective attention toward both internal and external health-related threats [51].

The specificity of the Polish pharmaceutical market in the cultural context

This is the first study exploring examining systematically the effects of pharmaceutical marketing on patient trust in the context of Poland, the sixth-largest pharmaceutical market in Europe [52]. The Polish market is characterised by high levels of pharmaceutical marketing activity towards physicians, including detailing, sponsorships of medical conferences, printing educational materials, providing gifts and drug samples, research funding, advertisements in medical journals and employing doctors as speakers or consultants, etc. [15,53,54]. According to the Central Statistical Office [55], 2024 marked a record year for prescription drug sales in Poland, both in volume and value. The number of prescriptions issued rose by 3.7% compared to the previous year. Furthermore, the total value of prescription drugs purchased reached 29.6 billion PLN (≈ 6,96bn EUR; exchange rate from 3 November 2025: 1 EUR = 4.25 PLN) reflecting a year-on-year growth of 3.5 billion PLN (≈ 0,82bn EUR). This positive trend underscores the market’s potential, providing strong support for the ongoing use of aggressive pharmaceutical marketing strategies.

The cultural context of pharmaceutical marketing in Poland is different from that of many Western countries, which have been the focus of most research to date. Poland has been described as a ‘culture of distrust’ [56], a sentiment rooted in historical experiences of the nation’s complex and turbulent history. During the Communist era, citizens were subjected to widespread state propaganda and systematic manipulation of information, eroding trust in institutions. After the collapse of communism, the shift to a market economy in 1989 brought significant challenges, including social inequalities, widespread anomie, and frustration with the inefficiency of political elites [56–58]. These historical and socio-political dynamics have left a lasting impact, fostering deep scepticism towards established institutions. This context is relevant when considering the public’s perceptions of the pharmaceutical industry and healthcare. Based on the IPSOS Trustworthiness Monitor [59], only 28% of Poles trust pharmaceutical companies, compared to a global trust level of 32%. The latest IPSOS survey on trust in professions [60] reveals that this distrust extends to medical professionals, with Poland recording one of the lowest levels of trust in doctors among the 32 countries surveyed. Romania reported the lowest trust level at 40%, followed by Poland at 41% and Japan at a similar trust level. The global average was 59%.

The low levels of trust have not been improved by the restrictions on physician-industry interactions imposed by the Pharmaceutical Law [61], amended in 2007 to harmonise Polish law with Directive 2001/83/EC and Directive 2004/27/EC. This change transformed the chaotic nature of pharmaceutical marketing into a strict law (even stricter than those required by the European Union), which, among other things, prohibits giving gifts to doctors worth more than 100 PLN (≈ 23,51 EUR). It also requires PSRs to include information consistent with the Product Information Leaflet. Only authorised individuals to issue prescriptions (e.g., physicians, some nurses) can accept drug samples after submitting a written request. The PSR must keep a record of free drugs distributed, and a doctor can receive no more than five packages of a given drug per year. According to Polish law, a pharmaceutical company may sponsor a doctor’s conference fee, travel, accommodation, and meals. However, if the total value exceeds 100 PLN, they must issue a tax document for the doctor. Subsequent secondary legislation has, for example, banned PSRs from entering doctors’ offices during working hours [62]. However, their enforcement is inconsistent, as evidenced in research by Makowska et al. [63].

Similarly, self-regulation by the medical profession, the Code of Medical Ethics [64,65], aims to establish ethical guidelines for the pharmaceutical industry’s interactions with physicians’, promoting transparency and preventing conflicts of interest. Similar initiatives exist within hospitals [66]. Additionally, the pharmaceutical industry has implemented self-regulation through the introduction of transparency codes [67,68]. Nevertheless, challenges remain in fully implementing and enforcing ethical standards in cooperation with the pharmaceutical industry across the medical community [63].

Study sample

A cross-sectional online survey was conducted using an internet panel. As there are around 20 million Internet users in Poland, G*Power 3.1 suggested a sample size of 968 people (alpha of 0.05, effect size of d = 0.2, allocation ratio of 0.25 for the t-test). Our actual sample was slightly higher at 1,057. The sample was selected from Adriadna Panel – one of the biggest Internet research panels in Poland, using cross-quotas, considering age, gender, and place of residence, resulting in 50 layers. Additionally, educational level and region were marginally added.

Data collection procedure

Invitations to take part in the research were distributed and supplemented by the Ariadna panel provider to fill missing quota cells. For this reason, a conventional response rate cannot be defined here – we do not observe a denominator covering all individuals eligible to participate. Instead, we report process metrics: eligibility rate – 90.6% (1357/1497; share of eligible cases among all starts); completion rate among eligible participants (raw) – 80.5% (1093/1357; completion before cleaning among eligible cases); usable rate – 70.6% (1057/1497; completions after cleaning among all starts). Respondents received 15 points from the Ariadna panel for completing the survey; these points are accumulated across different surveys and can later be redeemed for rewards (e.g., books) The incentive was independent of response content. Data collection took place between May 16 and May 19, 2023. On average, it took around 12 minutes to complete the online survey.

Survey instrument

The questionnaire, available through Figshare (https://doi.org/10.6084/m9.figshare.24598488), covered a range of topics related to pharmaceutical industry transparency and marketing, organised into six main sections along with 16 demographic questions. Each section included between 4 and 22 statements for respondents to evaluate. Before fielding the study in the Ariadna panel, we conducted a preliminary pilot with 12 participants. Their feedback was used to refine item wording, response options, and the questionnaire flow. In parallel, the instrument was reviewed by three experts in survey methodology, thereby supporting its content validity (adequate coverage of the construct domain) and face validity (the apparent appropriateness of the items for measuring the target construct) [69]. These steps were undertaken to ensure a sound questionnaire structure and plain, comprehensible language for respondents.

This article analysed only a few statements from sections one and three. Statements regarding trust in physicians and the pharmaceutical industry were selected from section one, each assessed on a 10-point Likert-type scale (1–10), consistent with commonly used trust measures [70,71]. From section three, we used six yes/no items that specifically examine the perception of pharmaceutical marketing presence in doctors’ offices. These items were developed by the authors based on their domain expertise. To limit recall bias, we used a yes/no response format, and deliberately did not anchor items to a recent time window, as our aim was to capture whether an event had ever occurred and was memorable enough to be recalled. Accordingly, we avoided asking for frequencies or salience, which would likely have prompted reconstructive responding – something we sought to minimize. The internal consistency of this six-item measure, assessed with Cronbach’s alpha, was.67, which is acceptable given the small number of items on the scale. To contextualize α, the mean inter-item correlation was.25 (range = .10–.69), which lies within the commonly recommended [72].

As the questionnaire was lengthy, we reduced respondent fatigue by using varied question formats, avoiding fully open-ended items and, where necessary, replacing them with semi-open questions. Items were grouped into logical blocks, and respondents could track their progress and proximity to completion via a progress bar.

Ethical Approval, Informed Consent, and Reporting Guidelines

The project received approval from the Research Ethics Committee of Kozminski University no. 04/05/2023. All participants provided written informed consent. We used the STROBE cross sectional checklist when writing our report [73].

Statistical analysis

SPSS 29 was used for statistical analysis of the data. First, the frequency of encountering different marketing activities directed at doctors was calculated. Next, the mean and standard deviation of trust in physicians and pharmaceutical companies were computed. Subsequently, t-tests were conducted to examine whether there was a difference in trust towards doctors and separately towards pharmaceutical companies among individuals who had encountered any form of marketing directed at doctors. The effect size of Cohen’s d was calculated. We also calculated the number of situations each respondent encountered and examined how these situations were interconnected by creating a Pearson’s bivariate correlation matrix.

As the two statements regarding meeting PSRs and waiting longer because of PSRs were correlated (close to 0.7), we decided to delete one of them. Then, we conducted linear regression analyses with the number of situations as the dependent variable. The independent variables in regression included age, gender, the presence of children in the household, the highest level of education, size of place of residence, and self-assessed health. Variables were entered into the initial model simultaneously, and their selection was guided by prior literature and the relevance to the presented topic. The model meets the basic assumptions for multiple regression [74].

Policy and practice implications and recommendations

This erosion of trust may have far-reaching implications for public health. Patients may become reluctant to take prescribed medications, refuse recommended or mandatory vaccinations, and disregard medical advice [87–89]. This distrust can lead to the proliferation of conspiracy theories about healthcare and increased reliance on unproven alternative medicine [90], ultimately resulting in worse health outcomes and undermining patient safety. Previous research aligns with these findings, showing that trust in the healthcare system is fragile and can easily be undermined by perceptions of commercial influence [22,37]. However, the observed contextual factors highlight the need for country-specific interventions. Therefore, policy approaches developed in Western European or US contexts may not yield the same outcomes in Poland without accounting for cultural nuances and historical distrust in institutions. Similar patterns have been observed in other fields throughout Poland’s economic transition to capitalism and liberal democracy, including international aid [91].

Several practical recommendations emerge from this study to address the challenges of physician-pharmaceutical company interactions. First, healthcare providers should implement comprehensive training programs and continuing medical education courses to raise physicians’ awareness of the potential consequences of pharmaceutical collaborations, particularly focusing on recognising, resisting and challenging potentially manipulative marketing techniques [92]. Second, there should be increased emphasis on promoting and enforcing transparency legislation regarding physician-pharmaceutical company interactions. Third, existing Polish regulations prohibiting meetings between PSRs and physicians during working hours should be more strictly enforced. Finally, efforts should be made to minimise the presence of pharmaceutical promotional materials in doctors’ offices, although this will be challenging to implement in resource-constrained settings, particularly in underfunded public clinics. These recommendations aim to maintain professional independence while acknowledging the practical constraints faced by healthcare institutions.

Limitations

Despite using cross-quotas based on age, gender, and place of residence to select the sample, the study relies on an internet panel, which may lead to sampling bias. Panel participants may have specific characteristics that motivate them to complete surveys in exchange for rewards. This self-selection into the panel, together with heterogeneous response propensities among panellists, can introduce systematic biases. Not all individuals in Poland have internet access [93], and certain groups – such as those from lower socio-economic backgrounds, older individuals, or rural residents – may be underrepresented in online surveys. Taken together, these factors could limit the ability to generalise the findings to the broader population. Furthermore, since the study focuses on the experiences of participants from Poland, the results may not apply to other countries with different cultural, social, or healthcare contexts. Therefore, the findings should be considered within the study’s specific context. Finally, because we asked about ever-observed events (rather than a specific recent visit), some respondents may have failed to recall their experiences; in such cases, ‘no’ responses would underestimate prevalence and bias associations toward zero. In addition, dichotomous items do not capture frequency or salience (e.g., a prominent logo or a persistent conversation with a PSR with the door open), further reducing information.

Future research

This study illuminates important aspects of pharmaceutical marketing and its impact on trust in healthcare systems, but several critical areas warrant further investigation. Future research should examine the long-term and diverse consequences of pharmaceutical marketing encounters on patient health outcomes and treatment adherence, utilising longitudinal studies to track these relationships over time.

Methodologically, future studies could anchor questions to a bounded recall window and replace binary indicators of marketing visibility with simple frequency or salience scales. Alternatively, experimental designs could be used to determine whether marketing cues were in fact noticed.

Subsequent analyses should also incorporate richer psychological and relational constructs (e.g., attitudes toward the industry, health literacy, health anxiety, locus of control, exposure to information sources, and the patient–physician relationship). By contrast, conventional sociodemographic variables, which explained little in our data, appear to offer limited explanatory value on their own.

Additionally, researchers should investigate perceptions of pharmaceutical marketing directed at physicians, incorporating qualitative research methods such as in-depth interviews and focus groups to provide a richer contextual understanding of patient experiences and perspectives. Finally, empirical studies are needed to evaluate the effectiveness of specific regulatory interventions in enhancing patient trust, particularly focusing on measurable outcomes and implementation challenges in various healthcare settings. Such research would provide valuable insights for developing evidence-based strategies to optimise the balance between pharmaceutical industry engagement and public trust in healthcare systems.