Background

The 2012 Health and Social Care Act stated that each Clinical Commissioning Group ‘must have regard’ to the need to reduce health inequalities.1 This duty was extended in the 2019 NHS Long Term Plan which requires ‘each local area to set out specific measurable goals and mechanisms by which they will contribute to narrowing health inequalities over the next 5 and 10 years’ and promotes integrated, publicly accountable care.2 Previous studies have investigated socioeconomic inequalities in paediatric inpatient admissions and emergency department attendance. To the authors’ knowledge, no such studies have been conducted for paediatric outpatient services.

The overall objective of this study was to investigate the socioeconomic distribution of new referrals to outpatient clinics undertaken by paediatricians at University College London Hospital (UCLH), hence assessing inequalities. The specific aims were to evaluate inequality in the socioeconomic distribution of new referrals between: (1) different clinics; (2) patients from the UCLH local area and other areas; (3) primary and secondary referral source.

Methods

Thirty-two thousand three hundred and sixty-nine consecutive, accepted, new referrals from April 2007 to June 2018 to UCLH paediatric clinics were reviewed. These were categorised into eight clinic types: allergy, asthma, chronic fatigue syndrome (CFS), diabetes, endocrine, epilepsy, general paediatrics and rapid access. The referral sources for each of the eight clinics were analysed and the sources were categorised into primary or secondary care.

The postcode of each new referral was matched to a lower layer super output area (LSOA) and the index of multiple deprivation (IMD), the official measure of relative deprivation for each LSOA, recorded. Two hundred and twenty-nine postcodes were unmatched and excluded from data analysis. The referrals were grouped into those living in Camden/Islington (~70% of the hospital general clinic referrals) and those from all other areas based on postcode.

The referrals to each clinic were divided into quintiles using the IMD, the proportion of referrals lying in each quintile calculated, and the ratio of most to least deprived quintiles derived.

Results

Figure 1 shows that the majority of referrals to allergy, asthma, epilepsy, diabetes, general paediatrics and rapid access came from a primary care source (general practitioner referral, following an Accident and Emergency (A&E)attendance) whereas the majority of referrals to CFS and endocrine came from a secondary care source (referral from consultant or following an emergency admission).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Referral sources (categorised into primary care, secondary care and other/unknown) by clinic. CFS, chronic fatigue syndrome. GP, general practice. A&E, accident and emergency.

Table 1 presents the proportion of patients referred to each clinic from each deprivation quintile, stratified into those living in Camden and Islington versus those living elsewhere.

Overall, 58.1% of all patients came from Camden/Islington (our local population), ranging from 11.6% of diabetes referrals to 78.4% of rapid access referrals. CFS, endocrine and diabetes had a substantially lower proportion of patients from Camden/Islington.

Figure 2 illustrates that the proportion of patients from each deprivation quintile differs significantly between the local and wider population. Among our local population, the socioeconomic distribution was similar in each of the eight clinic categories, with high proportions in the most deprived quintile (43.5%–48.4% in most deprived quintile vs 0.16%–2.53% in least deprived quintile). Among the CFS, diabetes and endocrine clinics, 17.5%–11.6% patients came from the local area. The socioeconomic distribution of patients attending these three clinics (ratio of most vs least deprived=1.797 for CFS, 1.798 for diabetes, 0.718 for endocrine) was markedly different to those attending the other five clinics (allergy=0.097, asthma=0.052, epilepsy=0.134, general paediatrics=0.078, rapid access=0.057) with a much greater proportion of patients coming from less deprived backgrounds.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

The proportion of patient referrals in each deprivation quintile by clinic from: (A) all areas; (B) Camden and Islington; (C) all other areas. Ordered left to right by increasing proportion of patients from UCLH catchment area. CFS, chronic fatigue syndrome.

Discussion

In this study, we illustrate how inequalities in outpatient referrals can be conveniently studied, analysed and used to inform the development of services in the future. We found that analysis of referrals from our local population provided reassurance that there are no obvious barriers to local patients accessing any of the clinics, as the clinic population broadly matched both the Camden and Islington IMD quintiles (online supplementary appendix 1), expected population prevalence of the conditions (online supplementary appendix 2) and our inpatient paediatric population.3 Although these findings indicate that there are no obvious barriers to access, it could be argued that a fully equitable system would have disproportionately higher activity among the more deprived population.

We chose to consider asthma, allergy, epilepsy, general paediatrics and rapid access as providing secondary care, as GPs provided the major referral route. However, the sources of referral varied between conditions (see figure 1); for example, a high proportion of asthma referrals followed emergency admission. In addition, these five clinics were united by having a large proportion of referrals from our local area (78.4%–47.2%).

CFS and endocrine were considered to provide tertiary care as the majority of referrals came from secondary care sources. We also considered diabetes to provide tertiary care as although the clinic accepted primary care referrals it provides specialist tertiary care services. In addition, these three clinics were united by having a small proportion of referrals from our local area (17.5%–11.6%).

Among the three tertiary care clinics (CFS, endocrine, diabetes), a much greater proportion of the patients was from a less deprived socioeconomic background, highlighting the need for referral inequalities to be audited across networks or regions. This finding cannot obviously be explained by greater prevalence in more affluent groups and may reflect greater barriers to accessing specialist care among more deprived groups. In contrast to diabetes and endocrine, some children with CFS may never come to medical attention.

Adult studies have shown that barriers to accessing specialist services are often lower among more affluent populations, although the National Health Service performs well for equitable care compared with other countries.4 There has been little research into outpatient inequalities in paediatrics in England, although a recent study into a paediatric tertiary chronic pain service similarly found that there were fewer children attending from the most deprived areas than might be expected.5

Conclusions and implications

As we move towards more integrated care systems, clinicians are being increasingly asked to consider the populations served by their services, as well as the individual patients they care for. Therefore, it is vitally important to be aware of barriers that may exist to access a service. While paediatricians often consider access issues around rapid access clinics, telephone hotlines and advice and guidance support, it is equally important to consider whether more specialist clinics provide equal opportunity of access to families, especially those from more deprived backgrounds who may have equal or greater clinical need but less ability to negotiate the system and/or travel further distances. Understanding the accessibility of the services we offer is a key part of being able to advocate for the patients we care for, a key part of being a paediatrician.

Overall, our analyses raise concern that access to tertiary services across the wider population we serve may not be equitable and may reflect barriers to access rather than true need. Further analyses at regional/network and national levels are needed to investigate these concerns in more detail.