Regional patterns and trends of hearing loss in England: evidence from the English longitudinal study of ageing (ELSA) and implications for health policy

The study utilised data from the English Longitudinal Study of Ageing (ELSA). The ELSA is a longitudinal prospective cohort study that collects multidisciplinary data from a nationally representative sample of community-dwelling middle-aged and older (aged 50 and above) adults in England [12]. The study started in 2002 and is collecting responses every 2 years on participants’ health, social, wellbeing and economic circumstances. The current sample contains data from eight Waves, covering the period 2002–2017 [13]. As the ELSA follows a longitudinal design, the sample is comprised of a sequence of observations on the same individuals across Waves and the refreshment samples (Cohorts 3, 4, 6 and 7) [13]. Proxy interviews were carried out in case an ELSA panel member refused to further participate [14]. In our analyses, we used the full dataset (74,699 person-years) of self-reported hearing data from all eight Waves of the ELSA.

The ELSA follows the sampling strategy of the Health Survey for England (HSE), which ensures that every address on the small users’ Postcode Address File (PAF) in England has an equal chance of inclusion. Field household contact rates of over 96% were achieved. The study excluded cases not belonging to the target population through ‘terminating events’, such as deaths, institutional moves and moves out of England since taking part in the HSE [15].

In this study, we examined the regional patterns and trends of HL prevalence in England in the ELSA over 15 years (2002–2017). We found that among samples with equal means of age, there was a 15-year increasing trend in HL prevalence in all five classes. The mean HL prevalence increased from 38.50 (95%CI 37.37–39.14) in Wave 1 to 48.66 (95%CI 47.11–49.54) in Wave 8. We identified three critical patterns of findings concerning regional trends: the highest HL prevalence among samples with equal means of age was observed in GORs with the highest prevalence of participants (a) in the most deprived (IMD) quintile (fifth), (b) in routine or manual occupations and (c) that misused alcohol, irrespective of SEP. The APMs for HL showed marked regional variability and evidence of a North–South divide.

Previous research has utilised geographical indices representing social and material disadvantages for identifying health inequalities [7]. Our study provided evidence for the existence of sociospatial inequalities in HL, adding to our previous work that challenged the existing conceptualisation of HL as an inevitable accompaniment of growing old [8]. Globally, there is a dramatic increase in HL cases, going from 42 million people in 1985 to about 360 million in 2011 and over 466 million in 2019 [28]. Our study presented a similar increase pattern but also showed that the increase in HL prevalence is not related to the ageing of the population, as widely believed [29, 30], but could potentially be due to social and lifestyle changes in the population [31]. Supporting our assumption, a previous study found a decline in HL prevalence among US adults aged 20–69 from the 2011–2012 cycle of the US National Health and Nutrition Examination Survey when compared to participants from the previous decade [32]. The explanation given by the authors for the declining prevalence was a reduction in exposure to occupational noise and the beneficial lifestyle changes of the participants, though that population study is not comparable to the ELSA cohort.

In our study, a North-South divide was revealed in hearing health inequalities that was previously unknown. The North-South gap is not surprising, as there is a significant history of socio-economic and health disparities between Northern and Southern England [33, 34]. The higher rates of unemployment and no qualifications in the North than in the South are in line with previous research in England [35]. We also found that alcohol misuse was high in areas with a high prevalence of HL, such as the South West, which over time developed one of the highest prevalence rates of alcohol misuse despite its higher socio-economic status compared to other GORs. This finding supports a previous study on the ELSA that found that alcohol intake above the low-risk-level guidelines [23] was significantly associated with HL among older adults in England, along with socio-economic factors [8]. However, the findings from this study indicate that the relationship between SEP and drinking habits is rather complicated; the last statistical release on adult drinking habits in Great Britain showed that those in managerial and professional occupations drink alcohol in higher proportions compared to those in routine and manual occupations. In addition, similarly to our study, it was found that the South East GOR, when compared to other GORs in England, had a higher proportion of adults drinking alcohol the week before the interview [36].

However, there are also important limitations. First, the unit of our analyses (in GORs) had a low geographic resolution, which introduces uncertainty in the observed relations and may fail to reveal geographic details that we could notice with smaller geographic units. Moreover, it was not possible to perform geographically weighted regression analyses; a minimum of 30 input features is required (instead of nine GORs) to explore the relationships between the areas’ socio-economic characteristics and HL prevalence. Furthermore, the ELSA’s size is regarded as too small to conduct geographic analysis on a larger scale, as numerous participants would be required in each unit.

Future research should build on this analysis using small area statistics (such as Lower Layer Super Output Areas) and investigate more localised patterns and determinants of place-to-place HL differences in England [35]. Such research would help to quantify potential ‘area effects’ on hearing health outcomes, allowing for generalisable results of spatial associations with HL rates. Moreover, the research could help to separate the role of proxies of areas (such as area deprivation) to individual-level determinants of HL (such as lifestyle behavioural choices), as individual choices are rooted in the broader social and economic structural contexts [31].

We were aware that the self-reported measures of HL in the ELSA might underestimate the real HL outcomes; for this reason, we conducted additional work to examine the validity of self-reported data through comparisons with the findings of objective HL measures available only in Wave 7 of the ELSA. We found that the self-reported measures correctly classified seven in every ten people with objectively assessed HL [16]. However, for the scope of our analyses, we assumed the available hearing measure as a suitable indicator of HL.

Another limitation is that the ELSA concentrates on individuals living in private households, so individuals living in institutions (e.g. residential and nursing homes) are not included in the samples [15]. Furthermore, ELSA does not capture the type of HL; future analyses examining types of HL would add important value.

Finally, the domains of IMD are not provided with the ELSA geography file, thereby not allowing further exploration. There was a small number of respondents moving to a different area between Waves, which resulted in an associated change in the IMD quintile [14]. However, a similar number of respondents experienced an increase or a decline in their IMD quintile, and the total numbers of movers did not exceed 1% for any Wave [14]; thus, we concluded that this would be unlikely to affect the validity of our findings.

According to the Global Burden of Disease Study, HL is the third leading cause of years lived with disability in England [37], and accurate prevalence estimates are needed to inform the strategic planning of hearing health policy and health services. To date, the prevalence of HL estimates in the UK is still based on the Medical Research Council National Hearing Study [3]. In addition, the NHS England has recently published the NHS Hearing Loss data tool [38], which provides estimates of the number of people with HL between 2015 and 2035 in order to help organisations plan services on local authority (LA) and CCG levels. However, according to our study, the above tool is inappropriate for estimating the number of people with HL; this study showed that in a representative cohort, there were important differences across different regions in England, which contradicts the Hearing in Adults study that did not find differences across the only four British cities that it was based on (Cardiff, Glasgow, Nottingham and Southampton) [3].

HL has affected a markedly larger proportion of the UK population in 2002–2017. The high levels of spatial clustering for hearing-related outcomes have significant implications for the planning of health services, including the availability of access to hearing aids. The high-risk regions in England must be expansively recognised based on their spatiotemporal HL profiles [39]. This kind of spatial evidence could provide commissioners with robust data based on actual needs, rather than inaccurate estimates of HL prevalence. Such prior knowledge could potentially have altered the North Staffordshire CCG’s decision in 2015 to end the routine free provision of hearing aids for people with mild or moderate HL in their area of duty [40], where according to our analyses, the burden of HL is greatest. This study revealed, therefore, the potential risks from the paucity of robust epidemiological hearing data, which are needed now as much as ever to increase understanding of the impact of social, financial and personal health advantages on HL across the life course [1].

The findings from the time-series analyses in this manuscript might encourage HL preventive strategies, including interventions to promote ‘healthier lifestyles’ and targeted interventions in areas where there are high levels of deprivation clustering. Future research should also explore spatiotemporal diffusion patterns in the ELSA’s international sister studies to acquire a global perspective of socio-spatial inequalities in hearing health.