1 INTRODUCTION

It is widely recognized that health inequalities are experienced by Aboriginal and Torres Strait Islander people in Australia.¹ Broad structural determinants such as racism, play a dominant role in such health outcomes,² including social determinants, the conditions where people learn, live, play and grow.³ Cultural determinants that embody Aboriginal and Torres Strait Islander worldviews are a beneficial influence.⁴ Strategies to improve health and reduce inequality must holistically address these determinants for there to be an improvement in Aboriginal and Torres Strait Islander health.^{5, 6} In the Aboriginal community-controlled health service (ACCHS) model, medical management is holistic and based on cultural safety principles where language and spiritual practices are embedded into care.⁷ ACCHSs, including Aboriginal Medical Services, recognise that health outcomes are determined by socioeconomic and political factors that influence living and working conditions, lifestyle factors and emotional wellbeing.²

Ear health is a significant health issue among Aboriginal and Torres Strait Islander children, who experience some of the highest rates of otitis media, or middle ear infection, in the world.^{8, 9} Although reported ear and hearing problems (0-14 years) decreased from 11% to 2001 to 7% in 2018 to 2019, hearing loss is higher in remote (59%) than non-remote areas (39%).¹⁰ Further, Aboriginal and Torres Strait Islander children are more likely to experience the effects of unmanaged ear health and associated hearing problems on speech, language and education.⁸ The causes of otitis media are multiple and complex and it is a public health challenge.¹¹ However, an examination of the determinants of ear health among Aboriginal and Torres Strait Islander people found health service access to be one of the key predictors of poor ear health, in addition to social factors that influence child health including household overcrowding, passive smoking, low income, and poverty.¹² Aboriginal and Torres Strait Islander children experience a higher incidence and more severe otitis media such as Chronic Suppurative Otitis Media and associated complications, and mild to moderate conductive hearing loss, particularly in remote areas.¹³

The Longitudinal Study of Australian Children (LSAC), a nationally representative cohort of children living in urban and regional areas in Australia, examined the social determinants of healthy ears and health service usage among Australian children with otitis media.¹⁴ In two cross sectional analyses, at age 4 to 5 years and at age 10 to 11 years, 7.9% and 3.3% of Australian children were reported via parent questionnaires to have otitis media, respectively. Sociodemographic and child health factors associated with good ear health included breastfeeding, parental smoking less than once per day and fathers having completed year 10 or higher in their schooling.¹⁴ Approximately 5% of the LSAC baseline sample are Aboriginal and Torres Strait Islander children who were less likely to have good ear health than other (non-indigenous) Australian children in the sample.¹⁴

Few studies have examined health service access and early childhood ear health. There are long waiting times for diagnosis and specialist treatment, including audiology and surgery, regardless of location.^{15, 16} As well, there is a discrepancy in the burden of otitis media in Aboriginal and Torres Strait Islander communities and the provision of specialist services resulting in significant underservicing.¹⁵

The Longitudinal Study of Indigenous Children (LSIC) is a cohort study that is following 1671 Aboriginal and Torres Strait Islander children from 11 urban, regional and remote communities and was established in 2008 (Wave 1).¹⁷ Parent-reported data from Wave 1 (aged 0-5 years) indicated that ear health symptoms were the most common health problem among Aboriginal and Torres Strait Islander children (experienced by 20% of children),¹⁸ with a similar proportion experienced in Wave 6¹⁹ when the cohort was aged from 5 to 10 years suggesting that ear health symptoms are common and persistent.

Health service utilisation was examined in Wave 6 data collected in 2013, where 32% of children had accessed Aboriginal Medical Services in the previous year, a higher proportion than healthcare settings such as general practitioners (GPs) and community health centres; 18% had accessed hospital care.¹⁹ Despite Aboriginal and Torres Strait Islander children experiencing health care access challenges and the high rates of ear health problems being well known, the relationship between health service access and ear health have not previously been studied. Understanding this relationship is an important first step to help inform strategies to improve Aboriginal and Torres Strait Islander child ear health. The current study aims to use LSIC data to examine relationships between parent-reported sociodemographic, child health, health service access factors and ear symptoms.

2 METHODS

3 RESULTS

Parent-reported data on ear symptoms was available for 1309 children in both Waves 2 and 3 (Table 1). Fifty-four (4.1%) children were reported by parents to have ear symptoms in both waves and 225 (17.2%) children had reported ear symptoms in either Wave 2 or 3. 1030 (78.7%) children were reported to have no ear symptoms in either Wave 2 or 3.

Among children whose Parent 1 were partnered, 78% had no reported ear symptoms in Wave 2 and 3 compared to those whose Parent 1 was single (81%; P = .03). A higher proportion of children whose parent's main source of income was a government pension/allowance in Wave 1 reported no ear symptoms in Wave 2 and 3 (81%) compared to those with a different main source of income (75%; P = .02). Among children whose Parent 1 reported a history of ear symptoms themselves, 65% had no ear symptoms in Wave 2 and 3, compared to 79% of children whose parent reported no such history (P < .01). Among children with reported ear symptoms in Wave 1, 60% had no reported ear symptoms in Wave 2 and 3. Among children with no reported ear symptoms in Wave 1, a higher proportion (83%) had no reported ear symptoms in Wave 2 and 3 (P < .01). Among children whose Parent 1 currently smoked, 82% had no reported ear symptoms in Wave 2 and 3. Among children of non-smoking parents a lower proportion (76%) had no reported ear symptoms in Wave 2 and 3 (P = .05). A lower proportion of children with difficulty sleeping had no ear symptoms in Waves 2 and 3 (78%), compared to those with no difficulty sleeping (81%; P = .01). A lower proportion of children who were ever breastfed reported no ear symptoms in Wave 2 and 3 (77%) compared to children who were not breastfed (85%; P = .02). The proportion of children who had been hospitalised in the past year and did not have ear symptoms in Wave 2 and 3 was higher (80%) than those who had not been hospitalised (76%; P < .01). A lower proportion of children who had received one or more home visits from a nurse in the past year at Wave 1 had no ear symptoms in Wave 2 and 3 (76%) compared to those who had not had nurse home visits (79%; P = .03).

In the analysis adjusted for all significant variables and study site clustering (Table 2), children with reported ear symptoms in Wave 1 were more than twice as likely to have reported ear symptoms in Wave 2 and 3 (aOR = 2.91; 95% CI, 1.57-5.40). Children who had been hospitalised in the past year at Wave 1 were also more than twice as likely to have no ear symptoms reported Wave 2 and 3 ([aOR] = 2.16; 95% CI, 1.19-3.94) or in Wave 2 and 3 ([aOR] = 2.42; 95% CI, 1.24-4.73) compared with children who had not been hospitalised.

Table 3 presents the additional analysis of relationships between Wave 1 child health and health service access variables and ear symptoms in both Waves 2 and 3, adjusted each sociodemographic variable and study site clustering. Children whose Parent 1 were single were almost twice as likely to have no ear symptoms reported in Wave 2 and 3 compared to those whose Parent 1 was partnered after adjusting for Parent 1 ear symptoms only ([aOR] = 1.94; 95% CI, 1.06-3.56). Children whose parent's main source of income was a government pension/allowance in Wave 1 were almost twice as likely to have no reported ear symptoms in Wave 2 and 3 after adjusting for Parent 1 ear symptoms ([aOR] = 1.95; 95% CI, 1.11-3.43) or Study child ear symptoms ([aOR] = 1.82; 95% CI, 1.04-3.19) compared to those with a different main source. Children whose Parent 1 reported no history of their own ear symptoms were almost four times as likely to have no reported ear symptoms in Wave 2 and 3 after adjusting for Parent 1 is partnered ([aOR] = 3.90; 95% CI, 1.43-10.62) or main source of income–government pension or allowance ([aOR] = 3.98; 95% CI, 1.45-10.86) compared to children whose parent reported a history of their own ear symptoms. Children with no difficulty sleeping in Wave 1 were more likely to not have ear symptoms in Waves 2 and 3 after adjusting for Parent 1 ear symptoms only ([aOR] = 1.84; 95% CI, 1.02-3.34) compared to children with difficulty sleeping. Children who had received one or more home visits from a nurse in the past year at Wave 1 were over twice as likely to not have ear symptoms in Waves 2 and 3 after adjusting for Main source of income – Government pension or allowance than those who had not received a home visit from a nurse.

4 DISCUSSION

This study presents new evidence of relationships between sociodemographic, child health, and health service factors and ear health among Aboriginal and Torres Strait Islander children. We found that children who had no history of ear symptoms and those who had been hospitalised in the past year, at age 0 to 5 years, had better ear health at age 2 to 7 years, using parent-reported data. We also found some evidence of better ear health among children whose responding parent was single, where their main source of income was a government pension/allowance, where the parent had no history of ear symptoms themselves, where the child had no difficulty sleeping or where the child received at least one home visit from a nurse at age 0 to 5 years but only in analyses adjusted for specific potential confounding factors. However, these effect sizes were small and these variables are subject to multiple complex inter-relationships.

The proportion of parent-reported ear symptoms in the present study of 4.1% in both Waves 2 and 3 and 17.2% in either Wave 2 or 3 is consistent with data from Wave 6,¹⁹ and is broadly consistent with another study³⁰ and Aboriginal and Torres Strait Islander experiences.⁹ This study adds value to prevalence estimates in identifying longitudinal predictors of the prevalence of parent-reported ear symptoms in this large, national sample of Aboriginal and Torres Strait Islander children.¹⁷

In the analyses partially adjusted for parent ear symptoms or child prior ear symptoms only, we found that children whose parent's main source of income was a government pension or allowance,⁶ at Wave 1 were more likely to have better ear health in the 2 years subsequently. Income is a key socioeconomic indicator,²³ and the relationship between both income and socioeconomic status and ear health appears to be complex. A 2009 review of otitis media risk factors acknowledged great variability in the association between otitis media and low socioeconomic status. Limited access to health care may be a mediating factor.¹³ A more recent scoping review of the determinants of otitis media found social and environmental factors such as overcrowded housing, poverty and limited access to services were otitis media risk factors¹² Forty five percent of Aboriginal and Torres Strait Islander people receive government cash pension or allowance as their main source of income,¹⁰ that can be considered a proxy for low socioeconomic status. The direction of the finding in this analysis is a little surprising. It may indicate that LSIC families with this source of income are receiving access to health care through social welfare provisions such as an Australian Government Health Care Card. The Health Care Card reduces health care expenses such as prescription costs,³¹ and may be contributing towards preventing the development, sustainability, or recurrence of ear symptoms. Regardless of source, having adequate income is a key pillar of good health,³ and such income support type needs to be sufficiently resourced to cover costs of living, including those that relate to healthcare.

Previous studies have identified that remote areas have a higher prevalence of otitis media due to socioeconomic status, educational factors, and inaccessibility of services, with health service accessibility and provision affected by remoteness.¹⁵ However, the present study did not find that remoteness or area-level socioeconomic status in Wave 1, measured through an Aboriginal and Torres Strait Islander specific index,²³ predicted parent-reported ear symptoms in Waves 2 and 3. While socioeconomic status and remoteness are intrinsically linked for Aboriginal and Torres Strait Islander people,²⁴ the ongoing effects of settler colonisation and social exclusion have led to complex health inequalities that cannot purely be explained by socioeconomic status.³² Broader, historical and cultural determinants of health exist where Indigenous people still experience comparatively poorer health outcomes when socioeconomic status, educational factors and health behaviours are equal to other populations.³³ Through a cultural lens that not only addresses the health concern, but also underlying cultural determinants,³⁴ service access needs to be viewed not only in terms of availability,³⁵ but also in terms of cultural safety for Aboriginal and Torres Strait Islander people as well as the historical and sociopolitical context.³³ Services must address the biomedical, social and cultural determinants of health in order to improve health equity for all Australians.² Aboriginal and Torres Strait Islander peoples have always understood that health and wellbeing is intrinsically linked to a holistic approach to healthcare, knowledge of culture, connection to land and community, recognising the positive effects that cultural connection has to self-esteem, collective identity and resilience, and the subsequent benefit of these factors to health and wellbeing.³⁴ These values are fundamental to improving ear health.

There was a strong, yet somewhat unexpected, finding in the present study of a relationship between child hospitalisation, for any reason, and better ear health. This may indicate children who have been hospitalised had better access to care and is consistent with the finding of a relationships between home visits from a nurse and better ear health, after adjusting for income source. Healthcare access could include increased proximity to services, more culturally appropriate care, or private health insurance to support access however such variables are not available in LSIC. As access to health services is a known determinant of ear health,¹² it may be that hospitalisation within a supportive, reliable, coordinated and trusted health system,³⁵ can lead to better ear health outcomes. Although it is acknowledged that preventing hospitalisations through quality preventive, primary and community care is optimal,³⁶ and this is reflected in the benefits of timely nurse home visits during the early years. Further, LSIC has already identified that equitable access to health service facilities, including tertiary care, is pivotal for the detection and treatment of health conditions,¹⁹ and we recommend strategies to strengthen access to health care services.

The finding that children without ear symptoms in Wave 1 were more likely to be without ear symptoms subsequently in both Waves 2 and 3 is not surprising given known recurrence but further highlights the impact on future ear health.^{8, 13, 37} We also found that children whose parents had a history of ear symptoms had higher odds of subsequent parent-reported ear symptoms, after adjusting for the potential sociodemographic confounding variables of parent relationship status and income source which may be explained by genetic predisposition.^{11, 13} Some previous Australian,^{13, 14, 38} and international,³⁹ studies have found that breastfeeding and smoke free homes predict better ear health but no associations were found in an epidemiological study in an urban location,³⁷ and this was not apparent in our analyses. We did find that children with no difficulty sleeping had better ear health after adjusting for parent ear symptoms but not other potentially confounding factors; this could be due to bidirectional relationships between sleep and ear symptoms. While a link between ear health and sleep has been found,¹³ good sleep habits may contribute to future ear health.

5 CONCLUSION

This analysis of data from LSIC found new evidence of relationships between sociodemographic, child health, and health service access predictors factors and parent-reported ear health among Aboriginal and Torres Strait Islander children aged 0 to 5 years. These were receiving a government pension or allowance, hospitalisation in the past year and having no prior ear symptoms, which were related to good ear health among children at age 2 to 7 years. These findings contribute to the evidence base that can inform future culturally safe, holistic strategies to improve health equity such as ensuring Aboriginal and Torres Strait Islander families have adequate support and improving healthcare access.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENT

The Longitudinal Study of Indigenous Children (LSIC) has approval from the Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS) Ethics Committee. Additionally, state and territory and/or regional ethics clearance and support have been obtained for all Footprints in Time sites through state and territory Human Research Ethics Committees or their equivalents. Approval to access the LSIC data was obtained from the Australian Department of Social Services.