Introduction

International migration into, and within, Europe continues to increase, with migrants forming a growing proportion of the population in many countries [1]. Population growth, along with other important demographic changes, such as population ageing, can be expected to put new pressures on healthcare systems due to altered service demand [2,3]. Much discussion and political debate in the United Kingdom (UK) on migrants’ use of healthcare has centred on emergency services. It is often argued that migrants place increased pressure on already overstretched services [4]. However, these debates are often unsupported by empirical evidence as there is a paucity of detailed research on migrants’ use of emergency services.

The UK’s National Health Service (NHS) provides health services that are free for users at the point of access for any person who is ‘ordinarily resident’ in the UK [5]. These services are offered on the basis of clinical need rather than ability to pay and include primary care services, urgent and emergency care, as well as hospital services. Importantly, any person, irrespective of migrant status or time since arrival in the UK, is currently entitled to free general practitioner (GP), primary care, and emergency services [5]. Parents, or caregivers, can access urgent and emergency care for their children through their GP; via NHS 111, which provides online or telephone advice to people with urgent medical problems; by making an emergency call; by accessing a walk-in centre or minor injuries unit; or by accessing an emergency department (ED). No referral is needed to access an ED. EDs provide 24-hour urgent and emergency care, while GPs, walk-in centres, and minor injuries units have restricted opening hours, with some services having a provision for out-of-hours urgent care. In non-emergency situations, parents and caregivers are encouraged, where possible, to consult their GP, or to access other urgent care services, before using an ED.

The current demand for NHS emergency care is unprecedented. Adding to this pressure, and of concern, is the increase in ED attendances and admissions of children under 5 years over the last decade [6]. Of these attendances, 60% are non-urgent [7]. Children are often vulnerable users of emergency services, dependent on their caregivers when accessing care [6]. Migrants to the UK are typically young and of childbearing age [8] and contribute to a substantial proportion of the UK birth rate [9]. It is therefore important to understand whether differences in patterns of paediatric ED use by migrant populations, as compared to the UK-born population, may contribute to increased paediatric ED use.

In many other European countries, higher use of EDs by immigrants, compared to people born in-country, has been observed in both children and adults [10]. Evidence suggests migrants tend to access EDs for low-acuity presentations, which indicates that barriers to more appropriate primary healthcare services exist [10]. To ensure the provision of appropriate and accessible emergency services to the demographically changing population, and to manage demand for these services, it is essential to understand paediatric ED utilisation by migrant caregivers for their children [11].

This study aimed to establish whether there are differences in paediatric ED utilisation in the first 5 years of life for children born to migrant, as compared to non-migrant, mothers in the Born in Bradford (BiB) cohort.

Methods

Results

Description of the cohort

The cohort included 10,168 children, of whom 3,620 (35.6%) were born to migrant mothers. The majority of migrant mothers were from South Asia (Table 1), of whom 88.2% were of Pakistani origin. At the time of recruitment, 58.7% of migrant mothers had been living in the UK for at least 5 years, 35.2% had been in the UK for 1 to ≤5 years, and 6.1% had lived in the UK for less than 1 year at time of study recruitment. The relative deprivation in this cohort, particularly among migrant mothers, is highlighted by the large proportion of mothers who lived in the most deprived areas of Bradford (Table 1). Migrant mothers, on average, lived closer to BRI ED, and for a greater proportion of these mothers (99.9% versus 97.9%), BRI was the closest ED to their home at the time of their child’s birth.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic and socio-economic characteristics of mothers in the Born in Bradford (BiB) cohort.

https://doi.org/10.1371/journal.pmed.1003043.t001

ED attendance and volume of utilisation

Overall, 3,104 (30.5%) children had at least 1 ED attendance in the first 5 years of life, with a total of 5,395 ED visits (Table 2). The proportion of children who ever visited the ED was lower for children of migrant mothers (29.4% versus 31.2%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. ED visit frequency by maternal migrant status and migrant sub-group.

https://doi.org/10.1371/journal.pmed.1003043.t002

The greatest proportion of ED attendances for this cohort of children took place in the first year of life (Table 3). Patterns of ED utilisation across weekends and weekdays were similar for both groups. Of the 5,395 ED attendances, 57.3% resulted in a hospital admission of at least 2 days. The most common reason for ED attendance, for both migrant and non-migrant children, was for respiratory conditions, with infectious diseases and injuries being the next most common reasons for presentations. For children who attended the ED, the average distance from home to hospital was 3.56 km, with migrants, on average, living closer to the BRI ED (2.58 versus 4.05 km).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Details of ED attendances (N = 5,395).

https://doi.org/10.1371/journal.pmed.1003043.t003

The average utilisation rate for the whole cohort was 106.12 visits per 1,000 children per year ([5,395/10,168] ÷ 5 × 1,000), and the rate was higher for children of migrant mothers as compared to children of UK/Irish-born mothers (Table 2). The results in Table 2 show that in sub-group analysis, when not adjusting for covariates, the highest rates of ED utilisation were for children whose mothers were from South Asia (112.32 visits per 1,000 children per year), from Europe or Central Asia (111.18 visits per 1,000 children per year), or considered established migrants (115.95 visits per 1,000 children per year).

Among ED users (those who used the ED at least once), the crude utilisation rates show that the highest rates of ED utilisation were by children of mothers from Europe or Central Asia (453.16 visits per 1,000 children per year) and Africa (420.83 visits per 1,000 children per year) and those most recently arrived in the UK (382.76 visits per 1,000 children per year).

Multivariable analyses: Likelihood of any ED use

The multiple logistic regression analyses confirmed that children of migrant mothers were significantly less likely than children of UK/Irish-born mothers to have visited the ED in the first 5 years of life when adjusting for other important covariates of interest (odds ratio [OR] 0.88 [95% CI 0.80 to 0.97], p = 0.012) (Table 4). The findings highlight that, while children of migrant mothers from all regions appear less likely to use the ED than children of UK/Irish-born mothers, this difference is only significant for children of mothers from Europe/Central Asia (OR 0.73 [95% CI 0.55 to 0.95], p = 0.018) and from Africa (OR 0.68 [95% CI 0.49 to 0.95], p = 0.022).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Unadjusted and adjusted ORs and IRRs for ED utilisation.

https://doi.org/10.1371/journal.pmed.1003043.t004

Children of migrant mothers, irrespective of the mother’s time since arrival in the UK, were less likely to have visited the ED. However, the multivariable analyses illustrate that, with increasing time since the mother’s arrival in the UK, children of migrant mothers show increasingly similar patterns of utilisation compared to children of UK/Irish-born mothers.

Discussion

In this study we observed that children born to migrant mothers were less likely than those born to UK/Irish-born mothers to make a first attendance to the ED in the first 5 years of life. Children of migrant and non-migrant mothers attended the ED on similar types of days and with similar conditions, and similar proportions were admitted to hospital. However, among the sub-population of children who attended the ED, those born to migrant mothers had a higher utilisation rate compared to children of UK/Irish-born mothers. This indicates a higher rate of return to the ED for children of migrant mothers once the service has been accessed. Although the rates of use were found to be statistically significantly different, the absolute differences in ED utilisation between children of migrants and non-migrant mothers, both for first use (5.86 additional ED visits per 1,000 person-years) and for repeated use (40.99 additional ED visits per 1,000 person-years), were rather small. These findings highlight the importance of analysing both the likelihood and the volume of service utilisation separately when seeking to understand patterns of ED utilisation.

Our findings show that after adjusting for important covariates, children of migrant mothers had a lower odds of first ED use in the first 5 years of life. To our knowledge, the only other study looking at paediatric ED use by children born to migrant mothers, although conducted in a different context, found contrasting results [23]. Our findings may suggest that either (1) children of migrant mothers in Bradford are generally not an unwell population in need of emergency care or (2) children of migrant mothers did not utilise the ED because they received care elsewhere. Of more concern is the possibility that medical care was not sought when children were in need. Existing evidence suggests that migrants in more vulnerable circumstances, such as undocumented migrants, experience substantial barriers to care [24–27].

Time since mother’s arrival in the UK was found to be an important factor in understanding likelihood of ED use. No previous studies looking at paediatric ED utilisation by children of migrant mothers have accounted for this as an explanatory factor. The odds of ED use were lowest for children of short-term migrants, and, with increasing length of stay in the UK, the difference in likelihood of first use for children of migrant and non-migrant mothers was seen to diminish. These findings may demonstrate that the migrant mothers who most recently arrived in the UK may be unfamiliar with the healthcare system and may not initially seek care for their children in the ED. With increasing length of stay in the UK, migrants’ understanding of the health service may develop, and in turn their children’s likelihood of paediatric ED use becomes more similar to that of children of UK/Irish-born mothers.

Despite all sub-groups of children born to migrant mothers in the cohort being less likely to have a first visit to the ED, once ED services were accessed, migrant mothers were found to be more likely to return to the service. This finding reflects those of other studies [23] and has important service implications for EDs and the wider healthcare system. Frequent paediatric ED visits may be expected if this population has higher levels of underlying chronic illness [28]. However, frequent attendances may also reflect access barriers migrants face when seeking other forms of healthcare [23,29], poor understanding of the host country’s healthcare system, prior positive experiences of the ED, and previous experiences within the healthcare system that result in a preference for seeking care in the ED [29–32]. Parents may also access the ED because this model of care most closely resembles the healthcare service in their home country.

ED utilisation rates were higher, but variable, for children of migrant mothers from all regions as compared to children of UK/Irish-born mothers. In particular, the results show significantly higher rates of utilisation by children of mothers from Europe/Central Asia and from Africa. Higher rates of ED utilisation by migrants by region of origin have been found in other studies [23,33–36]. Recognising the heterogeneity within the migrant population, and identifying differential use by people from different nation states and ethnicities, is important to enable health services to better understand population healthcare needs and to target health policies and interventions to meet these needs. Although, in our study, the effect sizes for some migrant populations did not reach statistical significance and have wide confidence intervals, most likely due the size of the sub-group, the patterns of utilisation by these sub-groups may have clinical and public health relevance and should be explored further.

Children of migrant mothers, irrespective of time since the mother’s arrival in the UK, showed higher frequency of ED utilisation following first access. When adjusted for covariates, ED utilisation rates remained significantly higher for children of established migrants. This contradicts the expectation that, over time, as migrant mothers become more familiar with the healthcare services in the UK, their rate of paediatric ED utilisation will become more similar to that of UK/Irish-born mothers. One possible explanation for these findings is that some children in this cohort born to established migrant mothers may have greater or more complex healthcare needs. This may be a reasonable explanation given that in Bradford the infant mortality rate is higher than the national average, along with high levels of morbidity within the Bradford population [14]. A lack of data meant that health status could not be controlled for. Other explanations again may include a range of demand- and supply-side factors including satisfaction with previous ED encounters, barriers to accessing out-of-hours services, the convenience of out-of-hours services, and long waiting times for primary care appointments [30,37].

This study has several limitations. A lack of clinical information made it impossible to analyse the acuity with which children presented, and children’s underlying morbidity may confound these results. The high proportion of hospital admissions for children of migrant mothers does suggest that for many of these children the utilisation of the ED was for conditions that require ongoing medical care. Therefore, one possible explanation for the higher rate of ED utilisation among children of migrant mothers may be that a child’s underlying health status drives both first ED use and subsequent ED use. If migrant mothers with the sickest children make use of the ED, it would not be unexpected that these children, due to the severity of their condition, are more likely to visit the ED more frequently.

A further limitation to this study is sources of selection bias in the BiB cohort, as well as bias that may have arisen from restricting the analytic cohort to mothers for whom there were complete data on the country of birth and other covariates of interest. Although the participation rate for the BiB study was not determined for migrant as compared to non-migrant mothers, some small differences overall between women recruited and those not recruited are evident in the BiB cohort [14]. Women who were recruited were older and lived in less deprived circumstances. In addition, more South Asian women than women from other regions were recruited to the BiB cohort. Thus, selection bias may be present in this study if migrants were less likely to join the BiB cohort, as might be the case particularly for vulnerable migrants such as those living in the most deprived areas and those most recently arrived in the UK. A further source of potential bias is evident in this study: A larger percentage of migrant mothers and less educated mothers were excluded from the analytical cohort due to missing data (S1 Table). These 2 sources of bias may have selected educated migrant mothers with higher socio-economic status and better language skills into the study, who would be expected to differ less from UK/Irish-born mothers. Thus, the findings of this study might have underestimated migrants’ use of services, or underestimated the use of ED services by some migrant populations.

It is possible that ED attendances for children of both migrant and non-migrant mothers were underestimated if children were taken to EDs outside of Bradford when in need of urgent care. Migrant populations are relatively mobile and may have been more likely to have sought healthcare in other EDs. Although BRI is the only ED facility within Bradford, and for 98.6% of children in the cohort was the closest ED to their home at the time of birth, it is possible that urgent and emergency care was sought in other EDs.

A further limitation is that due to incomplete data, we were unable to account for multiple children per mother within the cohort. However, by adjusting for mother’s parity, the effect of having additional children, and the experience that this brings, will have been adjusted for. Finally, the BiB cohort largely included children with mothers of British and Pakistani origin and thus may be quite different from other populations [14]. However, because we analysed the findings by maternal region of origin sub-groups, the findings may be more generalisable to other populations.

Our findings add to our knowledge and highlight differences in ED utilisation patterns between children of migrant and non-migrant mothers. The challenge for healthcare services is to identify those children accessing EDs most and to work with their parents to better understand their healthcare needs. It is important to understand when, and why, migrant and non-migrant mothers make their first ED visit for their child and, importantly, why some migrant mothers choose to return to the service.

Our study found that children of migrant mothers were less likely to use the ED for the first time in the first 5 years of life than children of non-migrant mothers, but, once accessed, the rate of repeat attendance was higher. Our findings confirm that immigrant groups use the ED differently when the analysis is adjusted for covariates of interest. Further research and better data are needed to understand these patterns of utilisation, the variations in use between people with different origins and backgrounds, and the reasons for these differences in utilisation. Understanding the reasons for frequent or repeated ED use among some migrant groups is important to ensure that the healthcare service is meeting the needs of the demographically changing population, while simultaneously addressing demand in paediatric EDs.

Supporting information

Acknowledgments

BiB is only possible because of the enthusiasm and commitment of the children and parents in BiB. We are grateful to all the participants, health professionals, and researchers who have made BiB happen.