)
The advent of integrated care systems (ICSs) has made addressing health inequalities everyone’s business. So it’s increasingly important for leaders across health and social care to understand health inequalities statistics.
Health inequalities statistics are often not as simple as they look. There are many different ways to measure health inequalities, and different methods can sometimes produce contradictory results. These different methods are already well understood by academics and analytical teams, but it’s equally valuable for everyone to be able to interrogate the numbers – whether it’s an ICS lead who is presented with statistics at a board meeting, an operational manager reviewing their monthly data, or a medical director of a trust reading about the latest research in the media.
Learning about statistics can be daunting, so as a starting point here are three questions to ask next time you see health inequalities statistics, along with some hypothetical examples that demonstrate why it’s important to question how inequalities are measured. These examples use inequalities in elective waiting lists – a topic many leaders are thinking about as part of the national ask to recover services inclusively.
1. Does this show absolute or relative inequality, or both?
Measuring inequalities is about measuring differences between groups. Relative measures of inequality look at the ratio between groups, while absolute measures of inequality look at the magnitude of the difference between groups.
In the example below, in relative terms, people living in the most deprived areas waited twice as long for elective care as the least deprived in both trust A and trust B. However, in absolute terms people in the most deprived areas in trust A were waiting 18 weeks longer for care than in the least deprived areas, whereas in trust B they were only waiting nine weeks longer.
So are the inequalities in trust B smaller than trust A, or are they the same? Both can be true at the same time, which is why it’s often helpful to see both measures to make sure you don’t miss any trends. If you are only presented with one measure, ask why, as there are advantages and disadvantages to using different methods in different situations.
)
2. Does this show inequalities in a positive or negative outcome?
Lots of measures of health and health care are binary – you are either infected or not infected, vaccinated or unvaccinated, attending an appointment or not attending an appointment. It makes a difference, though, whether you choose to measure the positive or negative outcome.
The example below shows the percentage of people who attended their elective care appointment, split by deprivation. Here you can measure inequalities based on either attendance rates or did not attend rates. If you measure attendance rates, the people in the least deprived areas are 1.5 times more likely to attend their appointment than the most deprived. If you measure did not attend rates, the people in the most deprived areas are twice as likely to not attend than the least deprived. The latter sounds worse than the former, but the two figures are based on the same data.
Again, it’s often helpful to see both measures to get a more comprehensive understanding of heath inequalities, especially if there are comparisons across time or between different areas. At the very least, a consistent method should be used.
)
3. Would a more complex measure of health inequalities be better?
In the examples above, health inequalities are measured in ratios and absolute differences, which are simple ways of measuring inequalities that tend to be well understood by many audiences. However, in some cases it might be better to use more complex methods to measure inequalities. For example, what if you wanted to compare more than two trusts or to look at a range of deprivation levels?
The World Health Organization describes complex measures of inequality as a ‘single number that is an expression of the amount of inequality existing across all subgroups of a population’ – examples include the concentration index and the slope index of inequality (for example, Marmot’s social gradient in health).
As measuring health inequalities becomes more important, it could be beneficial for more systems and organisations to use these complex measures of inequality. They take slightly longer for analytical teams to calculate, but they are useful for leaders who want a more comprehensive view of how inequalities are changing over time or how their inequalities compare with others.
The main purpose of these questions is to add depth to conversations around addressing health inequalities. Asking these questions will help ensure leaders obtain the data they need to set objectives and priorities for addressing health inequalities – and that they have the correct data to accurately assess progress and make meaningful comparisons between different areas.
These are just a few examples based on elective waiting lists; there are many more questions that could be asked around elective waiting list data and other types of inequalities. Not everyone needs to be an expert in health inequalities statistics, but being able to ask the right questions can help leaders make better decisions on addressing health inequalities.