Health inequalities worsen with the drop in hospital referrals

The article by Rachel Mulholland et al. documents the impact of COVID-19 on hospital-based care across the whole of Scotland up to the end of June 2020. ¹ In common with other areas of the UK, they show the drop in hospital activity from the date of the Work Health Organization pandemic announcement, through the first lockdown in March and April 2020, to the gradual recovery of activity. The authors state there is a need to investigate the impact of this disruption on preventable morbidity and mortality.

We suggest there is also a need to investigate whether this disruption has contributed to an increase in avoidable health inequalities. Patients from older age groups and ethnic minority populations are at higher risk for COVID-19 and may be at higher risk for non-referral during the recovery phase.

In east London, a multi-ethnic area which has high levels of deprivation, we examined non-emergency hospital referrals from primary care during the period 11 March to 30 September 2020 and compared them with the same period in 2019. Referral rates were examined by age, gender, ethnicity and number of co-morbidities in order to investigate referral deficits.

The multivariate model in Table 1 shows that increasing age and male gender show a significant referral deficit and that South Asian groups show a small drop in the odds of referral. However, those with the greatest number of long-term conditions were more likely to be referred during the 2020 period.

Table 1.

Multivariate model for predicting odds of referral March–September 2020 compared to referrals during the same period in 2019. ^a

	Referrals March 2019–September 2019 (%)	Referrals March 2020 –Sept 2020 (%)	Multivariate OR (95% CI), p value
Total referrals	162,227	84,182
Age (years)
18–49 (reference)	89,751 (55.3)	49,287 (58.6)
50–69	49,149 (30.3)	23,966 (28.5)	0.90 (0.88–0.92), p < 0.001
≥70	23,327 (14.4)	10,929 (13.0)	0.86 (0.83–0.88), p < 0.001
Sex
Male (reference)	64,240 (39.6)	32,625 (38.7)
Female	97,987 (60.4)	51,557 (61.2)	1.03 (1.02–1.05), p < 0.001
Ethnicity b
White (reference)	56,659 (34.9)	27,977 (33.2)
South Asian	59,161 (36.5)	28,782 (34.2)	0.97 (0.95–0.99), p = 0.008
Black	21,166 (13.1)	10,364 (12.3)	0.99 (0.97–1.02), p = 0.695
Other	7731 (4.8)	3490 (4.2)	0.90 (0.87–0.94), p < 0.001
Long-term conditions c
0 (reference)	66,981 (41.3)	35,933 (42.7)
1–2	66,339 (40.9)	33,943 (40.3)	1.01 (0.99–1.03), p = 0.417
3 +	28,907 (17.8)	14,306 (17.0)	1.04 (1.01–1.07), p = 0.004

a

Adjusted by Index of Multiple Deprivation based on patient postcode.

b

>90% self-reported ethnicity recording.

c

Based on the Quality and Outcomes Framework.

A major legacy of this period will be lengthening waits for hospital care and a reduction in referrals for those with suspected cancer. Using data from GP records suggests that there are inequalities in this deficit of hospital referrals. This worsens existing inequalities and is most marked for those aged 50 and over.