Around two thirds of the participants (68.5%) had high trust in the health system. Younger participants, women, those born outside Sweden, living in smaller municipalities, having lower education, experiencing economic stress or having lower social capital were associated with a lower trust in the health system. However, lower income was associated with higher trust.
The study showed a moderately high level of trust in the health system in northern Sweden. This can be compared to the national average of 61% reported in 2019 and to the 73.3% reported in southern Sweden in 2007 [4, 20]. An international comparison of 31 countries conducted between 2011 and 2013 showed that Belgium had the highest level with 72% reporting high trust, followed by Spain and Scandinavian countries, with percentages ranging from 56–59%. Large differences between the countries were reported: Germany (42%), France and the United Kingdom (around 30%) and the US (19%) [21]. Comparing countries is however a challenge due to the differences in health systems and cultures and to the lack of a standard method to measure trust [22].
The relation between age and trust is in line with previous research, which has shown that older populations tend to have more trust in the health system [12, 23, 24]. This might be explained by the modernisation theory, which postulates that the economic, political and cultural changes in the post-industrial societies result in a rejection of traditional social institutions. Younger generations are supposed to have greater shifts in cultural values, which in turn might lead to greater mistrust in the institutions; that is, these structural changes make it harder for young people to aspire to the same things to which the previous generations had aspired [25].
This study observed lower trust among women compared to men. Even though these differences were small, this finding was surprising since an earlier study in 2009 from northern Sweden showed an opposite relationship [12], and the literature often shows lower trust among men [9]. Women are usually more exposed to health services through their own experiences or through accompanying their spouse or children. This higher exposure to services together with the continuous threat of closing down healthcare services, including maternity units, in the rural parts of northern Sweden [26], might have led to different experiences and perceptions of healthcare among women, and consequently to lower trust in the health system [27].
According to our results, lower education and experiencing economic stress were also associated with lower trust, while lower income was associated with higher trust. The literature points in different directions regarding the relation between trust and income and education. Some studies have shown a relation between low income and lower trust in healthcare providers and health information [9, 10], while other studies have observed that higher education and income were associated with lower trust in the health system [21, 23, 24]. The different direction of the association between economic stress and annual disposable individual income and trust in this study suggests that these two variables are capturing distinct aspects of socioeconomic status which need further exploration.
Participants born outside Sweden reported lower trust in the health system compared to those born in Sweden. Earlier research in Sweden indicated similarly low trust in Swedish healthcare among refugees and immigrants [13, 14]. These differences could be explained by the perceived discrimination of the system towards participants born outside Sweden [11]. The different health system expectations of immigrants could further lead to worse experiences and thus to lower trust in the health system [13, 14].
According to our results, participants living in smaller rural municipalities had lower trust in the health system compared to those living in larger urban ones. However, a multinational study with 31 countries showed opposite results; people living in urban areas showed lower trust in the health system [21]. Results have been observed in China that are similar to the case of northern Sweden, while no differences between urban and rural areas were reported in the UK [23, 28]. Our results are probably context specific to northern Sweden where there are smaller rural municipalities with no hospitals and relatively long distances from healthcare facilities, and where rural citizens experiencing a policy of abandonment by central authorities. For example, closing some services in rural areas led to a mistrust in institutions in general and in healthcare in particular [17].
Lastly, our study showed that those with better social capital had more trust in the health system. A complex and bidirectional relation between horizontal trust and social or public trust in the health system has been discussed in the literature [29]. Studies have shown that social capital may improve trust in, quality of and access to the health system by altering users’ perception of healthcare [9]. This reinforces the social capital theory, indicating that social experiences and engagement in social activities have an important role in building social and horizontal trust, which in turn helps to build trustworthy organisations [9].
Strengths and limitations
HET survey offers a unique set of data that contains several sociodemographic variables, the trust in health system variable and a representative large sample of the region. Moreover, supplementing the HET with register-based sociodemographic data decreased reporting bias, especially for sensitive data such as income and place of birth.
Several issues should however be considered when interpreting the findings. First, while the 48% response rate is comparable to the response rate of other national surveys, we could not exclude selection bias, and the extent and direction of such bias could not be assessed. Second, given the study design, a reverse causality cannot be excluded. For example, horizontal trust has been associated with high trust in the health system, while trust in government and social institutions might reinforce horizontal trust [30]. Third, some relevant factors such as healthcare needs, experiences in healthcare, health care workers’ behaviours, continuity of care and availability of services could not be measured in this study. Since these factors are important in understanding trust, their inclusion might have altered some of our findings. Additionally, since ethnicity was not measured in the survey, we were not able to assess the Sámi Indigenous population’s trust in the health system. Finally, it is unclear whether our findings can be generalised to other settings in or outside of Sweden.