For Pacific children in Aotearoa New Zealand under 20 years of age, this study investigates whether parental education is protective in terms of hospitalisations. For 139,686 Pacific and 659,055 other (non-Māori/non-Pacific) children, a population data extract for them and their parents was used from linked 2013 Census, health and demographic data. Binary indicators of incidence of hospitalisations between 1 March 2013 and 28 February 2018 were analysed using modified Poisson regression models with robust standard errors. The highest parental educational level of either parent ranged from 0 indicating no qualification and increased incrementally to 10 signifying a PhD. Over the study period, 1% of Pacific children were hospitalised in private hospitals and 24% with a potentially avoidable hospitalisation. Highest parental education level was protective for Pacific children, with a single level in parental qualification associated with a small but significantly lower risk of potentially avoidable hospitalisation (RR = 0.97, p < 0.0001), but a higher risk of private hospitalisation (RR = 1.25, p < 0.0001). This finding remained significant, independent from the contribution of increased socioeconomic benefits that accompanied improved education. These findings support ongoing Pacific focused initiatives for promoting continuing education as an investment for the future health of Pacific families.

What we already know

• Pacific children have higher rates of publicly funded and lower rates of private hospitalisations than other children.

• Studies have shown that parental education is protective for many of their children’s health conditions. Any education advantage depends upon the type of condition, locality, and the status of external physical, social, economic, and political environment.

• Evidence of the extent to which other factors mediate the advantage from parental education is inconclusive.

What this article adds

• For Pacific research, this study uses a novel approach that includes public and private hospitalisations.

• Our key finding is that highest parental education level was protective in terms of fewer potentially avoidable hospitalisations but increased preventive private hospitalisations.

• Parental education advantage for Pacific and other children remained beyond the increased social or economic status expected to accompany improved educational status.

Introduction

Parental education has been found to be protective for many kinds of mental and physical conditions experienced by children (Cochrane et al. 1982; Zill 1996; Desai and Alva 1998; Dowd et al. 2009; McLaughlin et al. 2011; Teranishi et al. 2011; Rammohan et al. 2012; Berchick 2016; van der Tas et al. 2017). International studies have shown that higher parental education is protective for many health conditions, but this effect is dependent upon the type of condition, country of residence, and exposure to risk or protective factors (Zill 1996; Berchick 2016). The definition of educational ‘advantage’ is varied in international studies, yet most show a reduction in morbidity or other related outcomes (Zill 1996; Desai and Alva 1998; Hatt and Waters 2006; Dowd et al. 2009). Increased parental education has been found to be positively associated with higher levels of diagnoses verified by a health professional, assumed to be an advantage as it led to early intervention and prevention of more extreme conditions (Zill 1996; Berchick 2016). One study suggested that educated mothers had greater health literacy than other mothers in that they may be more able to detect symptoms of developing conditions (Berchick 2016). In some cases, the level of maternal education was only advantageous for detecting readily identifiable conditions (Cochrane et al. 1982; Hatt and Waters 2006; Dowd et al. 2009; van der Tas et al. 2017) such as autism, ADHD, asthma and allergies (Li et al. 2017). In addition, the commonly identified symptoms were easier to identify, preventable and less difficult to treat than other conditions (Cochrane et al. 1982; Hatt and Waters 2006; Dowd et al. 2009; van der Tas et al. 2017).

The evidence of links between parental education and their children’s health is abstruse. Several studies have reported the influence of education was either mediated by other factors or not found to be causal (Dom et al. 2009; Kvaavik et al. 2012; Litzelman et al. 2013). There are mixed findings, with some studies suggesting income or poverty has a stronger influence than parental education (Zill 1996). Other studies have found that parental education remained a stronger predictor of a child’s health than income, or other socio-economic factors (McLaughlin et al. 2011; Statistics New Zealand 2021c). Relatively few studies have sought to look at ethnic differences in the influence of parental education on outcomes associated with their children’s health, and none in the Aotearoa New Zealand context. Some international studies, highlighting racial or ethnic inequalities or reporting a specific ethnic group’s prevalence of health outcomes, included parental education as a control variable in regression models for adjusted rates (Flores et al. 1999; Gao et al. 2006; Teranishi et al. 2011; Berchick 2016).

The New Zealand Ministry of Education’s (2020) Action Plan for Pacific Education 2020–2030 characterises success as the creation of vibrant, dynamic and successful Pacific learners, secure and confident in their identities, languages and cultures while being able to navigate and succeed across all curriculum areas. Therefore it stands to reason that within this expansive vision of educational success, the benefits of education could extend beyond career aspirations to include better physical health and well-being outcomes for individuals and their families. The Ministry of Health’s (2020b) Ola Manuia: Pacific Health and Wellbeing Action Plan 2020–2025 also acknowledges the benefits of education in terms of health literacy is a priority to improved Pacific health status. It is known that approximately 90% of Pacific people in Aotearoa/NZ have low health literacy (Statistics New Zealand and Ministry of Pacific Island Affairs 2010) which may present barriers for accessing timely and effective healthcare. In a sample of Pacific mothers from the Pacific Islands Families cohort, it was found that having a tertiary qualification increased the likelihood of greater health literacy (Sa’uLilo et al. 2018). Increases in education may facilitate Pacific mothers’ engagement and communication with health professionals as common barriers highlighted by Pio and Nosa (2020) included the perceived stigma of language barriers and misunderstanding of medical jargon; it is possible that increased education may assist with this communication and more effective treatment.

At present, Pacific peoples make up 8% of the total population of Aotearoa New Zealand (Statistics New Zealand 2021a) and the effects of long-term socio-economic disparity persist in the form of inequities in both education (Ministry of Education 2020) and health (Ministry of Health 2020b) outcomes. As Pacific communities continue to plan and invest in education and health, it will be increasingly important for decision-makers is to understand the intergenerational impact of education on health. Specifically, how pathways to realising greater health benefits might be optimised and the further exploration of alternative pathways to accessing these benefits among those with reduced access to higher educational outcomes.

There are few studies that look specifically at educational advantage to health in Pacific communities. In studies that have included parents’ level of education in the analysis, it has typically been included as a control variable in regression analyses rather than a key explanatory influence. One study in Hawai’i found parental education had a beneficial impact upon obesity among children (Teranishi et al. 2011) and the Pacific Islands Families Study, in Aotearoa New Zealand, showed a strong association between mother’s education and depression among their children (Paterson et al. 2014). Conversely, the Pacific Islands Families Study, in a separate publication, also reported no significant maternal educational advantage in low birth weight and other birth outcomes (Gao et al. 2006).

While the associations between parental education and childhood health outcomes, including hospitalisations, have been acknowledged, little research has investigated the relationship between increased parental education and the incidence of Pacific child hospitalisations in Aotearoa New Zealand, particularly with potentially avoidable conditions. Potentially avoidable hospitalisations include diagnosed conditions that are considered preventable, largely comprised of ambulatory sensitive hospitalisation and avoidable injury (Ministry of Health 2020a; Statistics New Zealand 2021c). International studies suggest that low levels of income and education were associated with public over private hospital use (Meleddu et al. 2020) and the preference toward private use were related to immediate access (Owusu-Frimpong et al. 2010) and worth of treatment (Exley et al. 2012) as well as perceived greater care (Meleddu et al. 2020). Patients seen in private practice in Aotearoa New Zealand tended to be female, older in age and non-Māori (Penno et al. 2021).

An opportunity to analyse parental educational qualifications and Pacific children seen in hospital presents itself through the use of Stats NZ – Tatauranga Aotearoa New Zealand’s unique database, the Integrated Data Infrastructure (IDI; Statistics New Zealand 2021b). The IDI houses data pertaining to hospitalisations through Aotearoa New Zealand’s National Minimum Dataset linked with whole of population demographic data and census data, which includes some education measures.

This paper seeks to confirm evidence of parental educational advantage to Pacific children hospitalised in Aotearoa New Zealand, particularly with a potentially avoidable hospitalisation or seen in a private hospital. Furthermore, it explores the extent to which other factors mediate any parental educational advantage. We hypothesise that there is an association between highest parental education levels and the incidence of children hospitalised in Aotearoa New Zealand.

Methods

Results

shows the exclusions from the total resident population under 20 years of age for the cohort in this study. In the IDI, approximately 1.2 million children and young people under 20 years of age were resident in Aotearoa, New Zealand in 2013. Of those, just over one million were identified with census data and at least one linked parent. From that group, 139,686 Pacific and 659,055 NMNP children and their parents were included in this study.

describes the hospitalisations of Pacific and NMNP children in the study period. It shows that just over a quarter (26.5%) of Pacific children and over one in five (21.4.%) NMNP children were hospitalised in the study period. Few children were seen in private hospitals, the rate among Pacific children (0.9%) was a quarter that for NMNP children (3.7%). Conversely, Pacific children were more likely to be seen with a PAH condition than NMNP children, 23.5% and 16.9%, respectively. Most children hospitalised had at least one PAH, 88% of Pacific and 77% of NMNP children hospitalised in the study period (not shown in ).

describes the demographic variables in the two cohorts of children. Parents of Pacific children in the study attained, on average, a highest education qualification level of 2.7, just below a level 3 qualification typically obtained at year 13 from school. That is less than 4.7 for parents of NMNP, equivalent to a work or other undergrad post school qualification. Pacific children were slightly younger than NMNP children with fewer girls than boys. They were also more likely to have: a single parent; younger parents; two parents born in New Zealand; two Pacific and English speaking parents; and two parents affiliated with a Christian denomination. They were less likely to: live in homes owned by their parents; have incomes in the upper quartiles of household earnings; and live in areas of low deprivation.

shows the bivariate relationship between parents’ education level and their child’s risk of any hospitalisation. With each additional level of highest parental educational level, the unadjusted risk of any hospitalisation decreased by 1.5% for a Pacific child (RR = 0.99, p < 0.0001) and by 3.0% in NMNP children (RR = 0.97, p < 0.0001). For Pacific children, their unadjusted likelihood of a private hospitalisation increased by 25% (RR = 1.25, p < 0.0001) with a single unit of highest parental educational level and 13% (RR = 1.13, p < 0.0001) for NMNP children. The adjusted risk ratios for private hospitalisations of 1.08 for Pacific children and 1.05 for NMNP children (p < 0.0001 for both) suggested a strong effect size from highest parental educational level. For Pacific and NMNP children, respectively, from the percentage reduction in risk ratio changes, other covariate factors accounted for 45% and 27% of reduced risk to their child’s any hospitalisation from a single level of highest parental education, and 69% and 63% of private hospitalisations.

Unadjusted parental educational advantage measured around 3% reduction in children’s risk of PAH for a single level of parental education with risk ratios of 0.97 (p < 0.0001) for Pacific and NMNP children (). After adjustment for other covariates, risk ratios of PAH for a single level of parental qualification were 0.99 (p < 0.0001) and 0.97 (p < 0.0001) for Pacific and NMNP, respectively. Overall, the adjustment for other predictor variables accounted for 44% of the reduced risk from highest parental educational level to Pacific children’s PAH and 28% for NMNP children.

shows predicted rates for Pacific children hospitalised over the study period by highest parental educational level, using coefficients from adjusted models for Pacific and NMNP children. Specifically, the results labelled NMNP, show the predicted numbers and proportions of hospitalised Pacific children calculated from the coefficients of the adjusted regression models for NMNP children. Similarly, Pacific results were calculated using coefficients of the adjusted regression models for Pacific children. For parents without any qualification, 27.7% (95% CI: 26.5%–28.9%) and 28.6% (95% CI: 27.2% to 30.2%) of Pacific children were predicted to have any hospitalisation using respective Pacific and NMNP child regression estimates. Pacific children of parents with the highest qualification of level 9 or more, a Master’s degree or PhD, were less likely to be hospitalised with 21.8% (95% CI: 20.1%–23.1%) using Pacific and 17.4% (95% CI: 16.4%–18.6%) using NMNP estimates. shows, while there is an ethnic difference, the gradient of the slope for any hospitalisations in Pacific children across parental qualification levels is steeper for NMNP estimates than Pacific, but not so different for PAH, reiterating what is shown in . Few Pacific children, of parents with no qualifications, were seen in private hospitals with rates increasing as their highest parental educational level increased to 3.6% (95% CI = 2.7%–3.7%) using Pacific and 5.0% (95% CI = 4.4%–6.1%) using NMNP estimates.

From , taking the sum of the difference between the predicted numbers of Pacific children hospitalised, using Pacific and NMNP regression estimates, resulted in 686 (2%) fewer children hospitalised. Though comparatively small, that decrease includes a 1250 (70%) increase in Pacific children seen in private hospitals. As reported in , for any hospitalisation, the slope for the predicted incidence rates with increasing education qualification levels, though steeper for NMNP models than Pacific predictions of Pacific children hospitalised, run parallel for the respective models of PAH. This suggests that NMNP children gain more from their parents’ education than Pacific for any hospitalisations but not so for PAH. Taking the sum of the difference between the two medium shaded blue lines (PAH) in , the adjusted regression model for NMNP children predict 6431 (20%) fewer Pacific children with PAH. However, if all parents progressed by a single level of qualification, 1141 (4%) fewer children would have a PAH. Ultimately, if all Pacific children had the same opportunities as parents with the highest education levels, that is if 18% of all children with PAH, 7324 (24%) fewer children would be hospitalised.

Discussion

This study sought to understand the influence of parental education on health in Pacific children. While we included findings for NMNP children, the aim of this study was to produce findings that can influence effective strategies to improve the health of Pacific families. This study is novel for Pacific research in the context of Aotearoa New Zealand as, first, the analyses include data from public and privately funded hospitalisations. Secondly, it is a novel exploration of education advantage of private and potentially avoidable hospitalisations in Pacific families, adjusted for other influences. These analyses identified use of public and privately funded hospitals by Pacific and NMNP children, as well as the changes in use as a function of increased parental education. A notable difference in admission to public and private hospitals was observed between Pacific and NMNP children, with NMNP more likely to be seen in private hospitals and, by deduction, Pacific children in publicly funded hospitals. The findings show a clear advantage from highest parental educational level on hospitalisation rates in Pacific children, with more accessing privately funded health care which is consistent with the growing trend of use in Aotearoa/NZ due to longer waiting lists and perceived lower quality of care in the public system (Penno et al. 2021).

This study showed that Pacific children have higher rates of children hospitalised than NMNP children, consistent with studies of healthcare in indigenous and migrant communities internationally (Flores et al. 1999; Gao et al. 2006; Teranishi et al. 2011; Berchick 2016). It should be noted that NMNP in Aotearoa New Zealand is comprised largely of children with European ethnicity, the largest ethnic group, but includes other ethnic groups that have recent migrants. Fewer Pacific children were hospitalised as their parents’ level of education increased. For children of parents who had not reported any formal education qualifications there were no ethnic inequalities in hospitalisations. While sharing the same direction as Pacific children, hospitalisations decreased more steeply for NMNP children and the inequality between Pacific and NMNP children increased with the qualification level of parents. This evidence suggests that while Pacific children benefit from increased parental educational level, for reasons not fully explained by the available data, NMNP children display greater parental education advantage.

Overall findings from this study are aligned with international evidence that parental education is protective for their children’s health and well-being (Cochrane et al. 1982; Zill 1996; Desai and Alva 1998; Hatt and Waters 2006; Dowd et al. 2009; McLaughlin et al. 2011; Teranishi et al. 2011; Rammohan et al. 2012; Berchick 2016; Li et al. 2017; van der Tas et al. 2017). In this study, for each additional level of their parents’ education, the risk that a Pacific child was hospitalised with an avoidable condition decreased by a modest but statistically significant 3%. Other covariates mediated between two-fifths and a quarter of the parental advantage to children’s PAH rates in Pacific and NMNP children, respectively. Importantly, they suggest significant benefits, in terms of increased quality of private care and reduced preventable hospitalisations among Pacific and NMNP children, could be attained from continued education of their parents. Those benefits extend beyond the economic advantage that also accompanies families with educated parents. While it is beyond the data captured in the IDI to assess the influence of many barriers to care or factors such as health literacy, other studies suggest that significant benefits of increased parental education level to Pacific and NMNP children’s health may arise as a function of greater health literacy (Sa’uLilo et al. 2018; Pio and Nosa 2020). Such evidence would justify the Ministry of Health prioritising education in terms of health literacy to improved Pacific health (Ministry of Health 2020b).

A longstanding disparity in health outcomes between Pacific and NMNP has been highlighted within this study; however, this project provides insights into future pathways that could lead to positive health outcomes in Pacific families. The positive finding is that highest parental educational level is a protective mechanism for their children’s health. The findings also estimate the potential burden of PAH that could be avoided through lifestyle as educated parents may be more confident with or acculturated to the health system and are perhaps more likely to use primary care providers and have the resources to afford private healthcare (Cochrane et al. 1982; Desai and Alva 1998; Rammohan et al. 2012; Berchick 2016). Additionally, recent Pacific migrants may prefer to use public hospitals, particularly emergency departments, in place of primary care providers, as this might be more akin to what they would do in their Island home (Southwick et al. 2012). Regardless of how affordable primary care options are, any cost is a barrier to families who are financially stretched.

An overarching goal for Pacific peoples in Aotearoa New Zealand is thriving communities, which includes priority areas of young people, education and health (Ministry of Education 2020). This paper focuses on the potential for reducing preventable illness by means of lifting educational attainment as a modifiable element of a wider array of socioeconomic determinants of health which impact on Pacific communities. There is recognition within the Aotearoa New Zealand education sector that current programmes do not meet the needs of Pacific learners and there is a need for the development of new and innovative approaches to support Pacific learners. Thus, while education initiatives by the Ministries of Education and Health are positive, culturally sensitive and targeted responses from both the education and health systems are required to better equip the health workforce and Pacific families with the necessary support and this study provides further rationale for resource to be catered toward this solution. Further support for investment in these approaches is provided here as an upstream mechanism for helping address longstanding health inequities.

Conclusion

This study found that each level of parental education had a modest but significant protective influence on hospitalisation of Pacific children in Aotearoa New Zealand, that extended beyond the contribution of increased social and economic benefits that accompanied improved education. The cross-generational legacy of this education, for health outcomes, adds further weight to the need for and national benefits from investing in educational programmes that meet the needs of Pacific learners. These findings support initiatives for promoting continued education as an investment to future health of Pacific families.

Statistics New Zealand disclaimer

The results are not official statistics They have been created for research purposes from the Integrated Data Infrastructure (IDI) which is carefully managed by Stats NZ. For more information about the IDI please visit https://www.stats.govt.nz./integrated-data/. Access to the data used in this study was provided by Stats NZ under conditions designed to give effect to the security and confidentiality provisions of the Statistics Act 1975. The results presented in this study are the work of the author, not Stats NZ or individual data suppliers.

Disclosure statement

No potential conflict of interest was reported by the author(s).

This work was supported by the Health Research Council of New Zealand under Grant HRC 20/115 and HRC 20/116.