Summary
The results show that, although EQ-5D-5L seemed to be steady overall between 2012 and 2017, there is evidence of increasing inequality in HRQoL across population subgroups. The authors identified disparities in EQ-5D-5L across sex. Overall, males reported higher scores than females at each timepoint in the study, and this disparity increased from 2015 onwards. The younger female population (aged 18–24 years) accounted for this increasing inequality, as they reported the largest decline in HRQoL over time. The most deprived quintile of the population had the lowest HRQoL at every timepoint; in particular, deprived females reported the lowest EQ-5D-5L score across the population that worsened over time. The most deprived regions suffered decreases in HRQoL over time, while wealthier regions improved. The key driver of the decline in EQ-5D-5L over time was increasing levels of anxiety and depression.
Strengths and limitations
The authors analysed the well-known and widely used EQ-5D-5L instrument, whose reliability as a tool for measuring and valuing health status has been supported by decades of evidence-based research.20 They used large-scale, nationally representative data that monitored the trajectory of EQ-5D-5L across 3.9 million responders.
The study was unable to include recent GPPS surveys (2018 and 2019) in the analyses as questions on EQ-5D-5L were removed, and previous years of GPPS used alternative sampling methods (before 2012) or measures of HRQoL (early 2012). The study is therefore restricted to survey waves between mid-2012 and 2017.
The 2016 and 2017 GPPS were conducted in the winter period between January and March. Earlier surveys from 2012–2015 included an additional summer data collection between July and September. The current study’s results, however, display a similar trend during 2016 and 2017 as per the previous years.
Comparison with existing literature
The results support the growing evidence base that health inequalities across certain population subgroups are widening. In particular, the authors’ findings echo those reported in the recent follow-up to the Marmot Review, that social determinants of health such as sex, region, and socioeconomic circumstances play a large role in determining health.3
The gender gap in life expectancy has been a common feature of mortality trends for many years, both in England and across the world more generally, with females living longer lives than males.21,22 It is also widely reported that females spend more of these life-years in poor health, and so the gender gap in healthy life expectancy is relatively smaller.23,24 Previous research suggests that this sex differential in health is not only attributed in part to increased female longevity, but also to structural differences in fundamental characteristics between males and females, and their respective roles in society.25,26
Some studies have additionally explored gender inequality in HRQoL, finding that, on average, females tend to report lower scores than males. After adjusting for functional disability, differences in self-reported health persist as a result of differences in sociodemographic and socioeconomic characteristics, such as age, race, education, and income.27–29 The authors’ results show that, although females overall are more likely than males to experience day-to-day health-related limitations that adversely impact their quality of life, there are particular domains — such as selfcare, mobility, and usual activity — for which they report equivalent scores. Conversely, their scores for pain/discomfort and anxiety/depression are lower than for males. These findings are in line with the evidence that males are more likely to experience life-threatening health shocks that adversely impact their ability to perform daily tasks, whereas females are more likely to experience chronic but non-life-threatening disorders that test their pain threshold and mental health.30
The social gradient in health has been a major area of research for many decades.31,32 It is well understood that health inequalities, as a result of socioeconomic factors, can heavily influence life expectancy and healthy life expectancy.33 Recent evidence suggests that these inequalities are widening further, and disproportionately affect females living in deprived communities.3 The authors’ findings show that from 2015 onwards females living in the most deprived areas experienced a worsening HRQoL, while the situation over the same time period was steady for males. The most deprived areas in this study sample were regions located in the North and the Midlands, which experienced the most negative changes in EQ-5D-5L scores over time, thus reinforcing the already well-established North–South divide in inequalities in health.34–36
An alarming finding is that the youngest males and, particularly, females were the main subpopulations to experience a fall in HRQoL between 2012 and 2017. The driving factor was an increase in anxiety and depression. Mental health disorders are a large contributor to the overall health of young people, and a key determinant of disability and mortality, both in youth and later life.37–39 In the UK, the reported prevalence of affective disorders in young people is rising considerably, with young females being the most impacted.40–42 A growing body of research has evaluated potential determinants of mental health disorders, particularly among young adults, finding that, in addition to well-recognised social and economic risk factors, the current generation of young adults is faced with a novel range of problems relating to social media, educational pressures, financial uncertainty, and changing cultural norms.43 Some of these risk factors — for example, the rising psychological distress among students and graduates — may disproportionately affect young females compared with young males.44,45 Further, young people in general are less likely to seek medical help during an emotional crisis, which may explain the worsening trend in anxiety and depression over time if mental health concerns are not addressed.46 Ultimately, poor mental health could also be a mechanism for decreasing life expectancy if it leads to suicide. There has been a significant increase in the suicide rate in recent years for both young males and young females; despite low overall number of deaths, the rate has increased by 83% since 2012 for females aged 10–24 years.47
Implications for research and practice
Slowing improvements in life expectancy and widening health inequalities have prompted concerns about the progress of society as a whole.3 This study suggests that, although policymakers should continue to understand the main drivers behind the trend in longevity and healthy life expectancy, some additional thought should be given to address similar trends in HRQoL.
Several studies have explored the potential reasons behind the 2015 fall in life expectancy and the slowing of improvements in mortality post-2011. Many have linked these changes to the government austerity policies from 2010 onwards that resulted in reductions of health, social care, and other public budgets likely to affect the social determinants of health.1,48,49 Others attribute these changes to the increased prevalence of influenza, or even the possibility that life expectancy has reached its physical limits.1 The drivers behind these trends are still unclear; however, it is likely that some of the factors that affect mortality also affect quality of life or, as previously highlighted, quality of life can be directly linked to mortality.
An additional finding from this study that warrants attention is that of declining HRQoL in young adults, which seems to be linked to rising mental health issues. There are ongoing concerns over the increased prevalence of mental health problems in England, the widening gap in mental health inequalities, and the potential link to welfare policies and austerity measures as the main contributing factors.50,51 Developing interventions to address these worrying trends should be a policy priority.
Future research could further investigate health inequalities by exploring trends in EQ-5D-5L by ethnicity, for example, given the availability of this data in the GPPS.
