Inequalities in traditional health metrics, such as life expectancy, mortality and morbidity are well documented in the literature, but less is known about disparities in outcomes and experiences as reported by patients. This chapter describes how health outcomes and experiences of people vary by factors such as education, income and gender. A better understanding of these disparities is fundamental for designing health policies that deliver equitable results. The results show that inequalities between men and women are persistent; men report better health outcomes and well-being even when other relevant factors such as multimorbidity, age and socio‑economic factors are considered. Men also trust the healthcare system more than women. People with a lower socio‑economic status face a double disadvantage: not only do they fall ill earlier in their lives, but once sick, they also experience worse outcomes compared to their higher earning or higher educated counterparts. International comparisons show an opportunity for international learning as gender and socio‑economic gaps are much larger in some countries than in others.
Does Healthcare Deliver?
Abstract
What PaRIS tell us about gender- and socio‑economic inequalities
The PaRIS data confirm the gender health paradox: women live longer than men but consistently report poorer physical and mental health. This pattern is evident in most of the countries surveyed, with the gap consistently in favour of men. In addition, women report lower levels of well-being in almost all countries, with scores typically between 3% and 5% lower than men, and in some cases the gap is as high as 9%. The gender gaps are persistent and remain after controlling for socio‑economic factors, age and multimorbidity.
Among the PaRIS population with one or more chronic condition, gender differences in disease prevalence highlight the need for gender-sensitive health policies. Men are significantly more likely to report high blood pressure, cardiovascular conditions, and diabetes, while, women more often report arthritis, anxiety and depression, and neurological conditions such as epilepsy or migraine.
Men have more confidence in the healthcare system than women. Gender differences in access to services and gender bias in treatment contribute to the trust gap between men and women. In a third of countries, the gender gap was more than 10%, and in all but two countries it was more than 5%.
People with lower education and incomes face a double disadvantage: not only do they fall ill earlier, but once sick, they also experience worse outcomes compared to their higher earning or higher educated counterparts.
All countries show health inequalities between socio‑economic groups. Mental health, for example, shows a clear income gradient: the bottom third of the income distribution scores between 5 and 10 percentage points lower on the mental health scale than those in the top third. In addition, well-being and social functioning – key dimensions of health – are consistently better for those with higher levels of education and income. Although these patterns are generally consistent across countries, there are notable international differences. Switzerland, for example, shows very small or no gaps in all health indicators.
In most countries, people with higher incomes have more trust in the healthcare system than people with lower incomes. This gap is considerable with a difference of 10 percentage points or more in the proportion of people who say they trust the healthcare system. There are some countries like Spain and Romania with no or minor trust gaps.
Despite socio‑economic disparities in health outcomes, PaRIS reveals surprisingly small differences in how different socio‑economic groups experience healthcare. Experiences of the quality, co‑ordination and person-centredness of care were broadly consistent across income and education levels. The notable exception was confidence in managing one’s own health, where people with higher incomes and better education felt significantly more confident. The minimal differences in healthcare experiences may be partly explained by the fact that PaRIS patients report on their experiences of primary care, which is relatively accessible and affordable.
To address gender and socio‑economic health disparities, healthcare systems should collect and report disaggregated data across diverse demographics, including underrepresented groups such as non-binary individuals, people facing financial hardship and those facing overlapping disadvantages. Standardising tools and integrating these metrics into performance monitoring frameworks will reveal inequities, guide targeted interventions, and enhance accountability. International initiatives like PaRIS can support shared learning and measurable progress toward equity for all.
Inequalities in healthcare outcomes, experiences, and trust in healthcare systems persist as a significant challenge. This chapter explores how healthcare outcomes and experiences as reported by people with chronic conditions vary by factors such as education, income and gender, due to their direct relevance to shaping effective health policy.
While disparities in traditional health metrics, such as life expectancy, mortality and morbidity are well documented, less is known about disparities in outcomes and experiences as reported by patients. A better understanding of these disparities is fundamental for designing health policies that deliver equitable results. This chapter focuses mainly on gender and socio-economic inequalities.
When interpreting the results, it should be considered that PaRIS focuses specifically on primary care users – those who had contact with their primary care practices within six months prior to sampling. This means that any observed inequalities between groups could reflect differences in access to care. As PaRIS is a survey of primary care users rather than a population-based survey, it is possible that people who face significant barriers to accessing primary care may be under-represented in some countries. However, it seems plausible that many people with access problems are still included, as one contact with primary care within six months is enough for them to be part of the sample, and even those with access barriers may manage to have at least this level of contact in most countries.
To address this potential limitation, the survey included questions on access barriers, allowing for comparisons between PaRIS results and data from population-based surveys at the country level. This comparison shows a strong correlation between PaRIS results and those of population-based surveys, suggesting that a sizeable proportion of people experiencing access problems are, in fact, covered by the PaRIS sample. Although the potential impact of “access bias” cannot be completely ruled out, the analysis suggests that it is likely to be minimal. A more detailed explanation of this analysis can be found in Annex 5.A and Chapter 7.
Across the OECD, gender gaps persist in all spheres of public and social life (OECD, 2023[1]). Understanding gender differences is also paramount in dissecting inequalities in health outcomes and experiences. Gender, extending beyond biological distinctions, plays a significant role in shaping health disparities. Discrepancies in disease prevalence, healthcare access, and treatment experiences underscore the relevance of scrutinizing gender dynamics in healthcare (Alcalde-Rubio et al., 2020[2]) (Crespí-Lloréns, Hernández-Aguado and Chilet-Rosell, 2021[3]).
Health outcomes as well as experiences with healthcare are affected by biological factors, social roles, attributes, behaviours and expectations related to gender.
Women generally live longer than men. This gender gap averaged 5.4 years across OECD countries in 2022: life expectancy at birth for women was 83 years, compared to 77.6 years for men (OECD, 2023[4]).
Despite women’s longevity, women also report worse health and higher morbidity than men, a phenomenon known as the gender-health paradox (Phillips, O’Connor and Vafaei, 2023[5]). Multiple factors may contribute to this paradox, including biological, behavioural, social, lifestyle and socio-economic factors. This section describes how outcomes and experiences with care as reported by patients differ between men and women with chronic conditions.
In PaRIS, gender was measured with the question “which of the following best describes you?“ with answering categories “female”, “male”, “other”, “prefer not to say”.
People who do not identify as male or female, intersex individuals and transgender people face significant challenges in healthcare (Zeeman and Aranda, 2020[6]; Hsieh and Shuster, 2021[7]; Allory et al., 2020[8]). Despite progress in understanding gender diversity, disparities persist, and require attention within healthcare policy. However, the numbers within the “other” category, are often too small to allow meaningful analyses or reliable conclusions when it comes to international comparisons. This limits our ability to capture and address the unique healthcare outcomes of these populations. For this reason, analyses referring to people of other or undisclosed genders were carried out for the total sample rather than by country.
While the overall number of chronic conditions was comparable between men and women, there were significant gender differences for certain conditions. Men were more likely to report high blood pressure (58.5% of men versus 49.1% of women), as were cardiovascular or heart conditions (26.9% versus 16.8% of women) and diabetes (23.0% versus 15.2% of women). On the other hand, women were more likely to report arthritis or ongoing joint problems (43.3% versus 29.1% of men) and depression or other mental health conditions (19.5% versus 11.0% of men). Neurological conditions such as epilepsy or migraine were also more prevalent among women (8.6% versus 4.6% of men). Other long-term health problems, not included in the list, were reported more frequently by women (27.5%) than by men (21.9%) (Figure 5.1).
Percentage of people per chronic conditions (only people with one or more chronic conditions included)
Note: * Difference is statistically significant (p<0.05). Conditions are sorted by the relative difference calculated by dividing the prevalence in men by the prevalence in women.
Source: OECD PaRIS 2024 Database.
5.2.1. There are considerable gender gaps in health and well-being favouring men across all countries
For general health, a slight gender gap was observed in the total PaRIS population with chronic conditions: on average, 68% of men and 64% of women rated their health as good, very good, or excellent. Although there was a slight advantage for men in all but one country, the differences were not statistically significant. Nevertheless, breaking down health into its individual dimensions helps to shed more light on gender inequalities.
Physical and mental health gaps
Gender gaps become more visible when focusing on specific dimensions of health. Figure 5.2 shows the differences in aggregated country scores for physical and mental health between men and women. Women report less favourable scores than men in all countries. Women report a score 2 points lower on physical health and 1.5 points lower for mental health. While these gaps appear relatively small (10 points correspond with around 1 standard deviation in a standard population), there is considerable variation in the size of these inequalities between countries. For example, gaps in physical health of less than 1.5 points are found in the Czechia, Wales and Canada, while Saudi Arabia has a gap of more than 4 points. Gaps in mental health were less than 1 point in Australia, Iceland, the United States and Norway, but more than 2 points in, for example, Portugal and Greece.
Absolute differences in average scores per country for men and women with at least one chronic condition on PROMIS scales for physical and mental health
Note: People with one or more chronic conditions. Physical health: PROMIS® Scale v1.2 – Global Health component for physical health and mental are T-score metrics, higher values represent better physical health. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample: only includes people of 65 years and older. Physical health: all gaps statistically significant (p<0.05) except for Canada and the United States. Mental health: all gaps statistically significant (p<0.05) except for Australia, Iceland, Luxembourg, Romania and the United States.
Source: OECD PaRIS 2024 Database.
Men report better well-being and social functioning
Figure 5.3 illustrates the differences between men and women per country in terms of well-being and social functioning. There are significant differences in well-being, in all cases in favour of men. On average, the difference on the WHO‑5 well-being scale is 4.7 points (on a scale of 0‑100). In all countries, the gap is greater than 3 points, except in Luxembourg and the United States, where there are no statistically significant differences.
While gender differences are also found for social functioning in most countries, the difference is only statistically significant in Portugal, where almost 10% more men than women are positive about their social activities and roles.
Absolute differences in average scores for men and women on the WHO‑5 well-being index and differences in percentages of men and women who are positive about how they carry out usual social activities and roles
Note: People with one or more chronic conditions. Well-being: WHO‑5 well-being index. Response to five questions measuring well-being. Raw scale 0‑25 converted to 0‑100 scale with 0 being the lowest possible well-being and 100 the highest. Social functioning: PROMIS® Scale v1.2 – Global Health. Answer to the question: “In general, please rate how well you carry out your usual social activities and roles [further specified in questionnaire]”, “good, very good or excellent” versus “fair or poor”. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample: only includes people of 65 years and older. Well-being: gaps not statistically significant (p>0.05) for Luxembourg and the United States. Social functioning: gap only statistically significant in Portugal (p<0.05).
Source: OECD PaRIS 2024 Database.
5.2.2. Gender differences in experiences with care: Fewer women than men trust the healthcare system
Men and women also differ in how they experience healthcare.
Figure 5.4 shows gender gaps in the overall experienced quality of care and trust in the healthcare system. In most countries, men are more likely to report positive experiences of the quality of care received, although the differences are not statistically significant, except in Slovenia.
There are much larger gender differences in trust in the healthcare system, with men more likely to express trust. In a third of countries, the gender gap is over 10% and in all but two countries it is over 5%. Chapter 6 looks in more detail at the differences in trust in the healthcare system.
Absolute differences in percentage of women and men who reported positive experiences regarding quality of primary care, and trust in the healthcare system
Note: People with one or more chronic conditions. Experienced quality: Response to question: “When taking all things into consideration in relation to the care you have received, overall, how do you rate the medical care that you have received in the past 12 months from your primary care centre?”, “good, very good or excellent” versus “fair or poor”. Trust in healthcare system: Response to question: “How strongly do you agree or disagree that the healthcare system can be trusted?”, “strongly agree, agree”, “neither agree nor disagree, disagree, strongly disagree”.
* Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample: only includes people of 65 years and older. Experienced quality: gaps not statistically significant, except for Slovenia (p<0.05). Trust: gaps statistically significant except for Portugal, Romania, Spain and the United states (p<0.05).
Source: OECD PaRIS 2024 Database.
I believe my gender plays a role; as a Latina woman, I’ve felt my concerns are sometimes not taken seriously due to stereotypes about women’s health. While my education empowers me to navigate the system, it also highlights gaps in care for those who lack similar knowledge
Betsy, 68 years old, female, living with multiple chronic conditions, including obesity, dedicated caregiver for her husband, who lives with cancer, and for her daughter, who has Type 1 diabetes and chronic kidney disease
Gender inequalities are prevalent across many sectors, including healthcare, where disparities manifest in access to services and gender biases in treatment. This imbalance may contribute to the trust gap we identified in our data, where women express less confidence in healthcare systems. Although we cannot draw definitive conclusions about the factors driving this trust gap, OECD studies suggests that gender biases – such as limited access to gender-specific healthcare – are likely contributing factors (OECD, 2023[9]). Women’s concerns, particularly around chronic conditions, are often under-represented in research and decision-making, which can lead to experiences of neglect or inadequate care (Foo, Sundram and Legido-Quigley, 2020[10]).
Although our data are inconclusive, investing in gender-sensitive policies could be a valuable step towards closing this trust gap. Measures such as increasing women’s representation in health leadership positions, ensuring equitable access to services, and training health professionals to address gender-specific needs can all help build trust in healthcare systems. Such initiatives would not only improve trust but also lead to more inclusive, responsive and effective healthcare systems.
5.2.3. Other experiences with care
Gender gaps for several other PREMs were analysed: confidence in managing own health and well-being, co‑ordination and person-centredness. These gaps are not shown in this chapter, as they are not statistically significant. For co‑ordination and person-centredness the gaps are favouring men in almost all cases. Confidence in self-management is the only PREM that provides a more mixed picture, with gender differences in both directions.
5.2.4. Other or non-disclosed genders
The proportion of respondents who identified as a non-binary (other) gender, or chose to not disclose their gender, was 0.7% of the total survey population, with notable variations between countries.1 These small numbers make country-level comparisons challenging, therefore only results for the total PaRIS population can be reported. The reasons why people may choose not to report their gender are unknown, and some within this group may still identify as either a man or a woman.
Although these limitations prevent drawing firm conclusions, the results, as shown in Figure 5.5, indicate that this group scores significantly lower on all PaRIS key indicators compared to men and women. This suggests that gender-related issues extend beyond the binary framework of men versus women.
Physical health is reported as approximately 11.4% lower than that of men and 7.4% lower than that of women. Mental health scores are around 10.6% lower than men and 7.6% lower than women. Well-being is rated 15.3% lower than men and 8.9% lower than women. Social functioning follows a similar pattern, with scores approximately 8.8% lower than men and 6.8% lower than women. Confidence in self-management shows a slight difference of 6.4% lower than men and 6.3% lower than women. People with other or non-disclosed genders are also less likely to trust the healthcare system: 24.7% lower than men and 13.0% lower than women. Co‑ordination is 14.1% lower than men and 2.9% lower than women. Person-centredness is 23.1% lower than men and 15.5% lower than women.
Particularly the low trust levels likely reflect the challenges that non-binary people face in navigating healthcare systems that are predominantly designed around binary gender models (Teti et al., 2021[11]). The disparities experienced by those of other or non-disclosed genders may be compounded by a lack of visibility and tailored support within healthcare policies. Often, healthcare systems are structured around binary gender models, neglecting the specific needs of non-binary or gender-nonconforming individuals. This invisibility can exacerbate existing barriers, contributing to feelings of exclusion and mistrust towards healthcare services.
Literature shows that people who do not identify as man or woman report feeling invisible or marginalised within the healthcare system and feel that they “did not exist for the health system” due to the absence of inclusive documentation and practices (Gómez-Ibáñez et al., 2024[12]). These limitations in data collection and healthcare engagement underscore the need for more inclusive strategies and education for healthcare workers to address the specific health and support needs of non-binary and gender-nonconforming populations, thereby fostering greater equity and trust in healthcare services.
Scores on nine indicators of people with non-binary or non-disclosed genders as a percentage of the scores or men and women
Note: Percentages were calculated by dividing scores of other or non-disclosed gender by men and other or non-disclosed gender by women.
Source: OECD PaRIS 2024 Database. See Chapter 2 for a detailed description of the indicators.
Socio‑economic factors, such as income, education, and employment status, play a pivotal role in shaping classic health indicators like life expectancy and healthy life expectancy. These disparities are well-documented, revealing pronounced gaps in access to healthcare services and overall health outcomes.
Socio‑economic disparities in health outcomes are influenced by various interconnected factors. Firstly, social determinants of health, including housing conditions, labour circumstances, and employment opportunities, significantly impact peoples’ overall well-being (Belloni, Carrino and Meschi, 2022[13]). Poor housing conditions, precarious employment, and limited access to resources exacerbate health inequities among socio‑economically disadvantaged populations (Rolfe et al., 2020[14]). Secondly, socio‑economically disadvantaged groups may face barriers in accessing good quality healthcare services, including affordability, geographical proximity, and cultural competence (Khatri and Assefa, 2022[15]). Additionally, disparities in healthcare capabilities and health literacy further compound these challenges, with education levels correlating with people’s understanding of health information and their ability to navigate the healthcare system effectively. Finally, health behaviours such as smoking, alcohol consumption, diet, and physical activity patterns often correlate with socio‑economic factors (de Boer et al., 2020[16]).
Countries’ policies can influence such inequalities. Some countries’ policies may better address the needs of deprived groups, leading to more equitable outcomes. This may include supporting people in accessing good quality care and reducing complexity in the healthcare system, and targeted programmes to promote better health outcomes for socio‑economically disadvantaged populations.
While the provision of an extensive explanatory model goes beyond the scope of this report, this section describes socio‑economic gaps in reported outcomes, experiences and trust and on how this varies across different countries.
Indicators for socio‑economic status focus on education and income.
Income is measured in three categories, high, median and low household income. The categories are relative to the income distributions in specific countries.
Educational level was measured with the ISCED‑11 scale, which distinguishes 9 levels, varying from early childhood education to doctoral or equivalent. In some figures, the classification will be simplified to “low, middle and high”).
A special category has been defined of people who face financial hardship. These people answered “always or often” to at least one of the following questions: How often in the past 12 months would you say you were worried or stressed about the following things? 1) having enough money to buy healthy meals? 2) Having enough money to pay your rent or mortgage? 3) Having enough money to pay for other monthly bills, like electricity, heat, and your telephone? The size of this group varied from 4% to 30% of the population in a country.
Source: PaRIS Patient Questionnaire; International Labour Organization (ISCED); The Commonwealth Fund (2017), International Health Policy Survey.
5.3.1. Income and educational gaps
Income and education are key social determinants of health outcomes, often closely linked but also having distinct and interconnected effects on people’s health and well-being.
Higher income generally provides better access to healthcare, healthier living conditions, and improved nutrition (McMaughan, Oloruntoba and Smith, 2020[17]). In many countries, those with higher incomes also benefit from resources that contribute to overall better health. Additionally, financial stability reduces stress, positively impacting mental health.
Education plays a crucial role in equipping people with the knowledge and skills necessary for healthy behaviours, managing health conditions and effective navigation of healthcare systems (Conti, Heckman and Urzua, 2010[18]). Higher education levels correlate with better health literacy, enabling individuals to make informed health decisions (Chapter 4).
Therefore, it is also plausible that education and income may influence how people experience healthcare. For instance, those with higher education and income levels may be better equipped to navigate within healthcare settings, potentially leading to different healthcare experiences and outcomes (Hahn and Truman, 2015[19]; Andermann, 2016[20]).
Other social determinants, such as access to safe housing, employment opportunities, and community resources, further influence health outcomes and often intersect with income and education levels. Understanding these complex interactions is essential for developing strategies to improve people‑centredness of healthcare systems and reduce health inequalities.
Because of the close relationship between education and income, outcomes and experiences in this section are presented by both factors, with comparisons of income groups controlled for education.
People with a lower socio‑economic status get sick earlier in their lives
While results in this section are age and sex-standardised it is important to realise that people with lower education and income on average get chronic conditions at a younger age. This has been shown extensively in the literature and was also confirmed by the PaRIS data. Figure 5.6 shows the number of chronic conditions by age groups and broken down by education within age groups.
This figure illustrates that both the likelihood of having any chronic condition and the number of chronic conditions increase with age. There are notable differences between education levels: among the youngest group, 34% of higher-educated people have no chronic conditions, compared to only 23% among those with the lowest education levels. However, these differences diminish with age, and in the oldest age group, nearly the same high percentage of people have chronic conditions, around 92% for higher-educated and 94% for lower educated.
A similar pattern is observed for multimorbidity (having two or more chronic conditions). In the youngest age group, 32% of higher-educated people live with multiple chronic conditions, compared to 46% of those with lower education – a gap of 14 percentage points. As age increases, this difference narrows: among those aged 55‑64, the gap reduces to 11 percentage points; for people aged 65‑74, it further shrinks to 7 percentage points; and by age 75 and older, the difference is only 3 percentage points.
The absence of differences between education groups in the oldest age category may for an important part be explained by a “survival effect”; people in the highest education categories can expect to live around six years longer than those in the lowest education groups (OECD, 2017[21]). Being lower educated and having (multiple) chronic conditions may therefore be a cumulation of risk factors for dying at a younger age.
Percentages of all PaRIS respondents with 0, 1 or 2 or more chronic conditions, broken down by age group and education
Note: Self-reported chronic conditions.
Source: OECD PaRIS 2024 Database.
5.3.2. People with a high income and higher education report better physical and mental health
PaRIS data show that also among people who live with chronic conditions, socio‑economic factors matter. Although there are exceptions on some indicators and some countries, in general, people with higher education and income show considerably more favourable scores on patient-reported outcomes.
Figure 5.7 shows the self-reported physical and mental health disparities among different education and income groups. Statistically significant gaps are observed for both indicators across all countries. This suggests that while education and income are closely linked, each has an independent effect on health outcomes.
In most countries, the size of these gaps is similar between education and income groups, typically ranging between 2 and 5 points. Notably, countries with larger gaps in health outcomes between education groups tend to also exhibit larger gaps between income groups. However, there are exceptions. For instance, in the Netherlands and Norway, the mental health disparities between education groups are relatively small, whereas the gaps between income groups are more pronounced.
Mental and physical health, high and low education and high- and low-income groups compared across countries
Note: People with one or more chronic conditions. PROMIS® Scale v1.2 – Global Health component for physical health and mental are T-score metrics, higher values represent better physical health. *Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available. The comparison between income groups was controlled for education. All differences are statistically significant (p<0.05), except for Canada by education in mental health.
Source: OECD PaRIS 2024 Database.
Figure 5.8 shows that countries where lower-income groups report better physical health tend to have smaller disparities between income groups. While this pattern is not entirely consistent, countries with relatively small gaps – such as Switzerland and the Netherlands – show physical health scores for the lowest income group that exceed those of the middle‑income group in countries like Romania and Portugal.
PROMIS physical health scores for low-, middle‑ and high-income groups
Note: People with one or more chronic conditions. Physical health: PROMIS® Scale v1.2 – Global Health component for physical health is a T-score metric with a range of 16‑68, and a good-fair cutoff of 42, higher values represent better physical health. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. The comparison between income groups was controlled for education.
Source: OECD PaRIS 2024 Database.
5.3.3. Income gaps in social functioning vary by a factor of four between countries
As shown in Figure 5.9, also in well-being and social functioning there are clear differences between education and income groups, in all cases in favour of the higher income and higher educated groups. Also here, both education and income have independent effects. For social functioning, the inequalities by income groups are somewhat larger in most countries.
For social functioning the differences between countries are remarkable, from almost no gap between education groups in Switzerland to gaps between 15% and 20% in e.g. Romania, Greece and Iceland. Even after controlling for education, this gap is also apparent between income groups, with a four‑fold variation between countries.
For well-being, the differences between countries are somewhat smaller but also here, countries differ considerably, varying from almost no gap of more than 10 percentage points between education and income groups.
Absolute differences in average scores for income and education groups on the WHO‑5 well-being index and differences in percentages of low and high educated and people with low and high income who are positive about how they carry out usual social activities and roles
Note: People with one or more chronic conditions. Social functioning: Answer to the question: “In general, please rate how well you carry out your usual social activities and roles [further specified in questionnaire]”, “good, very good or excellent” versus “fair or poor”. Well-being: WHO‑5 well-being index. Response to five questions measuring well-being. Raw scale 0‑25 converted to 0‑100 scale with 0 being the lowest possible well-being and 100 the highest. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available. The comparison between income groups was controlled for education. All differences are statistically significant (p<0.05), except for education gap in well-being in Canada and Switzerland and education gap in social functioning in Canada, France and Switzerland.
Source: OECD PaRIS 2024 Database.
“As a single man with a university degree and a high income, I don’t experience anxiety about healthcare or related costs. When attending a peer group with other people living with HIV, I feel privileged – I don’t have to worry about my living conditions, have access to quality information about my chronic health conditions, and can take time off work if needed.”
Robert, 54 years old, living with HIV and chronic heart disease
As shown in Figure 5.10, countries where fewer people report good social functioning show the largest gaps between income groups. A similar pattern (not shown) exists between education groups.
Percentage of people reporting good, very good or excellent social functioning among people with chronic conditions in low, middle and high-income groups
Notes: People with one or more chronic conditions. Social functioning: Answer to the question: “In general, please rate how well you carry out your usual social activities and roles [further specified in questionnaire]”, “good, very good or excellent” versus “fair or poor”. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions.
Source: OECD PaRIS 2024 Database.
Well-being is not only related to income and education, there is also a clear negative correlation with financial hardship. In most countries, people who say they have problems paying their bills report lower levels of well-being as shown in Figure 5.11. Also here, there is a relation between the size of the gap and the overall level, meaning that countries with lower levels of well-being also have wider gaps. However, this relationship is less straightforward.
Scores on WHO-well-being scale of people with and without financial hardship
Note: WHO‑5 well-being index. Response to five questions measuring well-being. Raw scale 0‑25 converted to 0‑100 scale with 0 being the lowest possible well-being and 100 the highest. Countries are sorted from highest score on WHO5‑ well-being index for financial hardship group (left) to lowest score (right). * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available. Gaps are statistically significant for: Australia, Czechia, France, Greece, Iceland, Romania, Saudi Arabia, Slovenia, Spain and Wales (p<0.05). No valid data were available for Norway and Canada.
Source: OECD PaRIS 2024 Database.
5.3.4. People with higher incomes and education have more positive experiences with healthcare and have more trust in the healthcare system
Quality of care
Figure 5.12 shows gaps between education and income groups for experienced quality of care and confidence to manage own health. People with higher education are more often positive about the quality of the care that they receive, albeit that the differences are small and, in most countries, not significant. Only for Portugal a significant difference of 8 percentage points was found in the share of people who rated the quality of care good, very good or excellent.
The differences between income groups are somewhat larger, with statistically significant gaps between 2 and 6 percentage points in Czechia, the Netherlands, Luxembourg, Norway and Slovenia and 11 percentage points in Portugal.
People with higher education and those who earn higher incomes are more often confident in their ability to manage their own health and well-being. Beyond the complexity of their health situation, this confidence is likely tied to the skills and health literacy that tend to be more prevalent among highly educated people. However, even after controlling for education, the gaps between income groups remain remarkable. In half of the countries studied, the difference in confidence between education groups is 10 percentage points or more, and similar disparities exist between income groups. Nonetheless, some countries, such as Luxembourg, Spain and Saudi Arabia, show little to no significant gaps between these groups.
Gaps between education and income groups in general experienced quality of care and confidence in managing own health
Note: People with one or more chronic conditions. L.E: Low education; H.E: High education; L.I: Low income; H.I: High income. Confidence to self-manage: P3CEQ. Response to question: “How confident are you that you can manage your own health and well-being?”, “confident or very confident” versus “somewhat confident or not confident at all”. Experienced quality: Response to question: “When taking all things into consideration in relation to the care you have received, overall, how do you rate the medical care that you have received in the past 12 months from your primary care centre?”, “good, very good or excellent” versus “fair or poor”. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available.
The comparison between income groups was controlled for education. Gaps between education groups for experienced quality only statistically significant for Romania (p<0.05). Gaps between income groups for experienced quality statistically significant for Czechia (p<0.05). Gaps between education groups for self-confidence statistically significant for Australia, Belgium, Czechia, France, the Netherlands, Portugal, Romania and the United States (p<0.05). Gaps between income groups for self-confidence statistically significant for Belgium, Czechia, France, the Netherlands and Wales (p<0.05).
Source: OECD PaRIS 2024 Database.
Other experiences with care show minor gaps
Analyses on indicators of co‑ordination and person-centredness showed few education and income gaps. For person-centredness differences between 1 and 5 percentage points were found between income groups in the Netherlands, Norway, Portugal and Wales.
Figure 5.13 shows, for each country, how people experiencing financial hardship and those who do not experience person-centred care. As can be seen in the figure, there is no significant difference between these groups in any of the countries. This suggests that primary care professionals generally adapt to people’s individual needs, regardless of their economic position.
Scores on PC3Q person-centredness scale for people facing financial hardship and those not facing financial hardship
Note: P3CEQ Questionnaire. Response to eight questions measuring if care is person-centred. Scale ranges from 0 (lowest experienced person-centredness to 24 (best experienced person-centredness). Countries are sorted from highest score on person-centredness for financial hardship group (left) to lowest score (right). * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available. None of the gaps are statistically significant (p<0.05). No valid data were available for Norway and Canada.
Source: OECD PaRIS 2024 Database.
Trust in healthcare systems: Both education and income matters
In most countries higher educated people have more trust in the healthcare system. Differences between countries are considerable, varying from no or minor differences in Spain, Luxembourg, Greece and France to gaps of over 10 percentage points in Czechia, Australia, Portugal and Belgium. Gaps between income groups are even larger, with half of the countries showing gaps of 10 percentage points or more.
Gaps between education and income groups in trust in the healthcare system
Note: People with one or more chronic conditions. L.E: Low education; H.E: High education; L.I: Low income; H.I: High income. Trust in healthcare system: Response to question: “How strongly do you agree or disagree that the healthcare system can be trusted?”, “strongly agree, agree” versus “neither agree nor disagree, disagree, strongly disagree”. * Data for Italy refer to patients enrolled in outpatient settings for specialist visits in selected regions. **United States sample only includes people of 65 years and older; data for income groups not available.
The comparison between income groups was controlled for education. Gaps between education groups statistically significant for Australia, Belgium, Canada, Czechia, Iceland, Italy, Norway and Portugal. (p<0.05). Gaps between income groups statistically significant for Belgium, Canada, Czechia, Iceland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Saudi Arabia, Slovenia, Spain and Switzerland (p<0.05).
Source: OECD PaRIS 2024 Database.
This chapter examines disparities in health outcomes, highlighting gaps related to specific factors such as gender and socio‑economic status. However, these factors are not isolated in real life. For example, a person may be a woman with a lower income and live with multiple chronic conditions. These intersections of identity and circumstance mean that health inequalities are often the result of multiple, interrelated factors. Figure 5.15 presents results from regression analyses that account for several factors simultaneously – including gender, income, education, multimorbidity, and place of birth – to provide a more comprehensive understanding of how these elements relate independently with self-reported health outcomes.
As expected, relatively strong associations were found between the three outcome measures – physical health, mental health, and well-being – and multimorbidity. In this analysis, multimorbidity primarily served as a control variable, given that it is more prevalent among lower socio‑economic groups. Chapter 3 delves deeper into the specific aspects related to multimorbidity.
The results indicate persistent gender and socio‑economic disparities, even after accounting for factors such as age, multimorbidity, and place of birth. The strongest effects were observed for income, particularly between the highest and lowest thirds of the income distribution. The gender effect was most pronounced for well-being, with women scoring 4 points lower on a 0‑100 scale.
People born abroad reported slightly poorer health and well-being, although these effects were relatively small. For well-being, the relationship was largely explained by other factors. These small effects should be interpreted cautiously, as immigrants who do not speak the national language well are notoriously hard to reach in surveys, potentially making this group underrepresented.
Results presented in Figure 5.15 represent the overall PaRIS population, but results may differ by country. Additional analyses have been done to detect interaction effects between gender and income, gender and education and gender and place of birth. These analyses did not yield any statistically significant results.
Note: Uncontrolled: relations only controlled for age; Controlled: unique regression effect controlling for all other variables shown and age. Multimorbidity effect for people with 1 chronic condition compared to people with multiple chronic conditions; Born in country: compared to people born outside the survey country; Income high and median compared to people with low income; Education high and average compared to people with low education; Male compared to female. Random intercept models with patient, practice and country level. Physical health: PROMIS® Scale v1.2 – Global Health component for physical health is a T-score metric with a range of 16‑68, and a good-fair cutoff of 42, higher values represent better physical health. Mental health: PROMIS® Scale v1.2 – Global Health component for mental health is a T-score metric with a range of 21‑68, and a good-fair cutoff of 40, higher values represent better mental health. Well-being: WHO‑5 well-being index. Response to five questions measuring well-being. raw scale 0‑25 converted to 0‑100 scale with 0 being the lowest possible well-being and 100 the highest. All effects are statistically significant (p<0.05) except controlled effects for born in country and mental health and well-being.
Source: OECD PaRIS database 2024.
5.5.1. Gender inequalities in health are persistent and ask for a targeted policy approach including explicit target setting
Gender inequalities persist across various domains of life, a fact widely acknowledged by experts and policy makers. In response, the OECD has been encouraging countries to adopt comprehensive, whole‑of-government strategies for gender equality and mainstreaming (OECD, 2023[9]). The findings from PaRIS reaffirm the persistence of gender gaps while providing more nuanced insights, particularly in the area of health outcomes. While women live longer, they also report worse health outcomes. Moreover, these differences remain after considering age, socio‑economic status and number of chronic conditions, indicating a “real” gender inequality.
The PaRIS data reveal that gender gaps are particularly pronounced in healthcare outcomes like well-being and physical health, while they are less evident in healthcare experiences. While measures of healthcare experience, such as perceived quality and person-centredness, are largely influenced by the care received, broader factors play a significant role in shaping individuals’ well-being and physical and mental health. While the data do not allow for firm conclusions about causality, it seems that factors affecting gender inequalities lie largely outside primary care.
As shown in previous OECD work on this topic, gender inequalities are not exclusive to healthcare, but are part of a much broader problem that should be addressed with comprehensive, gender-sensitive strategies that integrate gender perspectives throughout planning, implementation, and evaluation. Policies should adopt an intersectional approach, recognising how factors like age, ethnicity, and socio-economic status intersect with gender to influence health outcomes. Strengthening the collection and use of gender-disaggregated data is crucial for identifying gaps and tailoring interventions. Healthcare systems could work on capacity building through training on gender sensitivity to reduce biases and foster equitable care.
5.5.2. Targeted policy interventions should address socio‑economic inequalities in outcomes and experiences
People with lower education face a dual disadvantage: they tend to fall ill earlier in their lives, and once sick, they experience worse outcomes compared to their higher-educated counterparts. While income and education are correlated, each exerts an independent influence on health outcomes and experiences. For income in particular, the relationship is bidirectional: people with lower incomes may have fewer resources to improve their health, while poor health conditions can, in turn, limit their opportunities to earn a higher income.
A similar pattern emerges as to that observed in the analysis of gender-based inequalities. While disparities in health outcomes are substantial, differences in healthcare experiences are relatively small. Primary care professionals, as the first point of contact in the healthcare system, play a critical role in addressing inequalities between socio‑economic groups. The relatively small gaps in healthcare experiences suggest that primary care professionals often succeed in tailoring care to the needs of diverse socio‑economic groups.
However, disparities in outcomes, such as physical and mental health and overall well-being, are shaped by a complex interplay of factors that extend beyond the healthcare system. This indicates that policy interventions focusing solely on healthcare systems are unlikely to be sufficient. Health inequalities are deeply interwoven with broader socio‑economic disparities. In the domain of healthcare, policies may focus on investing in healthcare services in deprived neighbourhoods, community-based programmes to improve health literacy and training of healthcare workers to recognise and mitigate socio-economic barriers.
Trust in the healthcare system deserves particular attention. The differences in trust levels between genders, as well as between education and income groups, are significant. Healthcare systems and policy makers have a crucial role in building and maintaining trust among people with chronic conditions. This topic is further explored in Chapter 6.
5.5.3. Integrating health equity metrics into health system performance monitoring
To effectively address gender and socio‑economic disparities in health, healthcare systems must commit to routinely collecting and monitoring disaggregated data across a broader range of demographic dimensions that expose disparities in health outcomes and experiences. Many healthcare systems may not be accustomed to systematically gathering such detailed data, but doing so is crucial to accurately identifying and addressing inequalities. Without this level of detail, the specific challenges faced by particularly vulnerable groups risk being overlooked (OECD, 2024[22]).
Building the capacity for routine disaggregated data collection requires a fundamental shift in how health systems operate. Standardising tools and processes for data collection is essential, alongside expanding data gathering efforts to include groups often marginalised or underrepresented in traditional health statistics – such as immigrants, non-binary individuals, and those with complex or overlapping disadvantages. This broader scope will provide a clearer understanding of health inequities and the structural barriers that perpetuate them. Such insights can then inform policy development and resource allocation, ensuring interventions are focused and effective in meeting the needs of underserved populations.
Finally, integrating this expanded data into performance monitoring frameworks also enhances accountability and transparency. Regular public reporting on disaggregated metrics, such as health outcomes and experiences by genders, would demonstrate a commitment to equity and foster trust among the populations served. Initiatives like PaRIS, which provide international benchmarking of equity metrics, can facilitate the sharing of best practices, enabling countries to learn from one another and improve outcomes for all demographic groups. Monitoring and accountability mechanisms are vital to oversee progress and enforce compliance with gender and socio-economic equality objectives.
This chapter highlights the persistence of structural inequalities in health outcomes and experiences across PaRIS countries, with significant gender and socio‑economic disparities. While primary care systems deliver relatively equitable patient-reported experiences, gaps in outcomes – particularly for women and lower-income or less-educated groups – reflect the influence of broader societal determinants. Tackling these disparities requires a whole‑of-government approach that integrates healthcare and social policies to address root causes, such as poverty, education gaps, and gender biases. Systematically measuring and reporting disparities will enable better-targeted interventions, resource allocation, and accountability, addressing these inequalities to not only improve individual well-being but also strengthen the equity and resilience of healthcare systems, fostering a fairer and more inclusive society.
References
[2] Alcalde-Rubio, L. et al. (2020), “Gender disparities in clinical practice: are there any solutions? Scoping review of interventions to overcome or reduce gender bias in clinical practice”, International Journal for Equity in Health, Vol. 19/1, https://doi.org/10.1186/s12939-020-01283-4.
[8] Allory, E. et al. (2020), “The expectations of transgender people in the face of their health-care access difficulties and how they can be overcome. A qualitative study in France”, Primary Health Care Research & Development, Vol. 21, https://doi.org/10.1017/s1463423620000638.
[20] Andermann, A. (2016), “Taking action on the social determinants of health in clinical practice: a framework for health professionals”, Canadian Medical Association Journal, Vol. 188/17-18, pp. E474-E483, https://doi.org/10.1503/cmaj.160177.
[13] Belloni, M., L. Carrino and E. Meschi (2022), “The impact of working conditions on mental health: Novel evidence from the UK”, Labour Economics, Vol. 76, p. 102176, https://doi.org/10.1016/j.labeco.2022.102176.
[18] Conti, G., J. Heckman and S. Urzua (2010), “The Education-Health Gradient”, American Economic Review, Vol. 100/2, pp. 234-238, https://doi.org/10.1257/aer.100.2.234.
[3] Crespí-Lloréns, N., I. Hernández-Aguado and E. Chilet-Rosell (2021), “Have Policies Tackled Gender Inequalities in Health? A Scoping Review”, International Journal of Environmental Research and Public Health, Vol. 18/1, p. 327, https://doi.org/10.3390/ijerph18010327.
[16] de Boer, W. et al. (2020), “How are lifestyle factors associated with socioeconomic differences in health care costs? Evidence from full population data in the Netherlands”, Preventive Medicine, Vol. 130, p. 105929, https://doi.org/10.1016/j.ypmed.2019.105929.
[10] Foo, K., M. Sundram and H. Legido-Quigley (2020), “Facilitators and barriers of managing patients with multiple chronic conditions in the community: a qualitative study”, BMC Public Health, Vol. 20/1, p. 273, https://doi.org/10.1186/s12889-020-8375-8.
[12] Gómez-Ibáñez, R. et al. (2024), “Describing Non-Binary People’s Encounters with the Healthcare System: A Qualitative Study in Catalonia (Spain)”, Archives of Sexual Behavior, Vol. 53/5, pp. 1957-1967, https://doi.org/10.1007/s10508-024-02849-x.
[19] Hahn, R. and B. Truman (2015), “Education Improves Public Health and Promotes Health Equity”, International Journal of Health Services, Vol. 45/4, pp. 657-678, https://doi.org/10.1177/0020731415585986.
[7] Hsieh, N. and S. Shuster (2021), “Health and Health Care of Sexual and Gender Minorities”, Journal of Health and Social Behavior, Vol. 62/3, pp. 318-333, https://doi.org/10.1177/00221465211016436.
[15] Khatri, R. and Y. Assefa (2022), “Access to health services among culturally and linguistically diverse populations in the Australian universal health care system: issues and challenges”, BMC Public Health, Vol. 22/1, p. 880, https://doi.org/10.1186/s12889-022-13256-z.
[17] McMaughan, D., O. Oloruntoba and M. Smith (2020), “Socioeconomic Status and Access to Healthcare: Interrelated Drivers for Healthy Aging”, Frontiers in Public Health, Vol. 8, https://doi.org/10.3389/fpubh.2020.00231.
[22] OECD (2024), Rethinking Health System Performance Assessment: A Renewed Framework, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/107182c8-en.
[4] OECD (2023), Health at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/7a7afb35-en.
[1] OECD (2023), Joining Forces for Gender Equality: What is Holding us Back?, OECD Publishing, Paris, https://doi.org/10.1787/67d48024-en.
[9] OECD (2023), Toolkit for Mainstreaming and Implementing Gender Equality 2023, OECD Publishing, Paris, https://doi.org/10.1787/3ddef555-en.
[21] OECD (2017), “Life expectancy by sex and education level”, in Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/health_glance-2017-7-en.
[5] Phillips, S., M. O’Connor and A. Vafaei (2023), “Women suffer but men die: survey data exploring whether this self-reported health paradox is real or an artefact of gender stereotypes”, BMC Public Health, Vol. 23/1, https://doi.org/10.1186/s12889-023-15011-4.
[14] Rolfe, S. et al. (2020), “Housing as a social determinant of health and wellbeing: developing an empirically-informed realist theoretical framework”, BMC Public Health, Vol. 20/1, p. 1138, https://doi.org/10.1186/s12889-020-09224-0.
[11] Teti, M. et al. (2021), “A Qualitative Scoping Review of Transgender and Gender Non-conforming People’s Physical Healthcare Experiences and Needs”, Frontiers in Public Health, Vol. 9, https://doi.org/10.3389/fpubh.2021.598455.
[6] Zeeman, L. and K. Aranda (2020), “A Systematic Review of the Health and Healthcare Inequalities for People with Intersex Variance”, International Journal of Environmental Research and Public Health, Vol. 17/18, p. 6533, https://doi.org/10.3390/ijerph17186533.
To explore the relation between access problems reported in population-based surveys and in PaRIS, figures from PaRIS were compared with figures from Eurostat. An important difference is that the Eurostat data refer to people 16 years and over and the PaRIS data to people 45 years and older. The percentages in the Eurostat data are consistently higher, which could partly be explained by the different age category. There is, however, a strong correlation, indicating that countries with bigger access problems also have more patients with access problems in PaRIS.
Percentage of people reporting access problems
Note: To calculate the percentage for PaRIS patients, two items were combined: the percentage indicate the share of people answering ‘always or often” to the following questions: How often did you have a health problem but did not seek care because of difficulties in travelling to your primary care centre? How often did you have a health problem but did not seek care, or did not take a prescription medicine because of the cost? The Eurostat item is called: self-reported unmet needs for medical examination and care with as reasons: too expensive, too far to travel or waiting list. For Wales, Eurostat data from the United Kingdom were used.
Source: OECD PaRIS 2024 Database; Eurostat (2023 or latest available year), https://ec.europa.eu/eurostat/databrowser/view/sdg_03_60__custom_13475331/default/table?lang=en.
← 1. People were included who answered “other” or “prefer not to say” to the question “which of the following best describes you” (female, male, other, prefer not to say). People who skipped the question were excluded.