Inequalities in self-rated health across European Union countries: a Simple-Means MAIHDA demonstration using the European Social Survey
Using a Simple-Means MAIHDA approach on European Social Survey data, this study reveals that health inequalities in the EU are primarily driven by shared sociodemographic patterns rather than national differences, demonstrating that country-specific averages poorly predict individual health outcomes compared to strata-based analysis.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Health is rarely distributed evenly across a population. In any society, some groups are healthier than others, and these differences often follow predictable patterns based on where people live, how much money they have, their age, and their gender. For decades, public health officials have tracked these inequalities by calculating averages: the average health of a country, or the average health of people with low incomes. While these numbers tell us that a country like Latvia reports more poor health than Greece, or that older people generally report worse health than younger ones, they hide a crucial reality. An average is a single number that smooths over the messy, complex details of real life. It cannot tell us how well a specific group of people is doing, nor can it reveal whether the gap between rich and poor is wider in one country than in another. To truly understand health, we need to look at the specific combinations of these factors together, rather than just at the separate parts.
A team of researchers set out to do exactly this, using a massive dataset covering 24 European Union countries. They analyzed the self-reported health of over 171,000 people, asking them to rate their own health as very good, good, fair, bad, or very bad. Instead of just looking at the average health of each nation, the researchers broke the population down into 384 distinct groups. Each group was defined by a specific mix of three factors: the country they lived in, their sex, their age, and whether their household found it difficult to live on their current income. By examining these 384 specific slices of society, the researchers could see not just who was sick, but how the country itself shaped the relationship between a person's circumstances and their health. They wanted to know if a country simply made everyone slightly healthier or sicker, or if it actually changed the size of the gap between the rich and the poor, the young and the old.
The results revealed a story that challenges the way we usually think about national health statistics. First, the researchers found that the differences between countries were surprisingly small compared to the differences between social groups. While the overall rate of poor health varied from about 4% in Greece to 15% in Latvia, the variation between the specific social groups was much larger. Within any single country, the health of a young man with a comfortable income was vastly different from that of an older woman struggling to pay her bills. In fact, the gap between these social groups was more than twice as large as the gap between the countries themselves. This means that knowing a person's country tells you very little about their health, but knowing their specific social situation—whether they are an older woman with financial difficulties, for instance—tells you a great deal.
Furthermore, the study showed that while every country followed the same general pattern—where older people and those with money troubles were less healthy—the size of these gaps changed depending on where you were. Some countries, like Sweden, managed to narrow the health gap for older women with financial difficulties, making their health much closer to that of the wealthy. Other countries, like Croatia, saw that same gap widen significantly. This proves that national policies and environments do more than just shift the average health up or down; they actively reshape the inequalities within the population. The researchers calculated that about 80% of the inequality followed a shared European pattern, but the remaining 20% was unique to each country, showing that local conditions can either buffer or amplify the disadvantages of being poor or old.
Perhaps the most striking finding concerns how well we can predict a person's health based on these groups. The researchers tested how accurately they could identify people with poor health just by knowing which of the 384 groups they belonged to. They found that this method was quite good at sorting the population, correctly identifying about three-quarters of the people with poor health. However, it also meant that one in four people with poor health lived in groups that were considered "low risk" by the average standards. This highlights a classic public health dilemma: if we only target interventions at the groups with the highest average rates of sickness, we will inevitably miss a large number of sick individuals who happen to live in groups that are generally healthier. Conversely, a strategy focused only on the country level would be almost useless, as the country itself explained less than 1% of the differences in individual health.
Ultimately, this research suggests that the old way of monitoring health—relying on national averages and broad socioeconomic categories—is insufficient. It paints a picture that is too smooth, hiding the sharp edges where inequality actually lives. The study demonstrates that health is deeply personal and contextual, shaped by the specific intersection of a person's identity and their environment. While the broad trends of poverty and age affecting health are consistent across Europe, the severity of these effects is a local matter, determined by the unique conditions of each nation. For policymakers, the lesson is clear: to improve health effectively, interventions must be precise enough to target the specific groups that need them, while also recognizing that a universal approach is necessary because a significant number of sick people will always fall outside the highest-risk categories. The data shows us that the map of health is not a simple gradient of countries, but a complex landscape of specific social positions, and understanding that landscape is the first step toward fixing it.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.