Are there differences between the rates of new diagnoses reported pre and during the COVID-19 pandemic?

Problem

Office for National Statistics reports have shown that the COVID-19 pandemic has caused an increase in non-COVID-19 causes of death. One mechanism may be changes in the provision of healthcare. The English Longitudinal Study on Aging (ELSA) is a large scale study collecting data from adults over the age of 50 years. Recently two additional waves of ELSA data were collected in view of the COVID-19 pandemic. Here, ELSA data have been used to investigate differences between the rates of new diagnoses reported pre and during the COVID-19 pandemic, and if there were differences in rates of diagnosis between different health conditions.

Approach

Data in the ELSA wave pre-COVID-19 (wave 9 (W9), June 2018 – June 2019) and in the two waves collected during the pandemic (COVID-19 waves 1 (CW1), June-July 2020 and waves 2 (CW2), November-December 2020) were compared. Results were adjusted for total number of participants per wave and time between waves. Only diseases included in all three waves of questionnaires were included. Logistic regression was conducted to investigate potential confounders.

Findings

W9, CW1 and CW2 included 7289, 5825 and 5092 individuals, respectively. There was minimal difference in the overall rate of new diagnosis reported pre (163/1000 individuals/year) and during the COVID-19 pandemic (159/1000 individuals/year). However, there were variations in diagnosis rates for different conditions. Between W9, CW1 and CW2 there was a decrease in the rates of diagnosis of dementia (10.6 to 4.3 to 1.6/1000/year), cancer (17.1 to 13.2 to 12.6/1000/year), stroke (7.3 to 6.1 to 3.9/1000/year) and angina and myocardial infarction (15.2 to 9.4 to 8.6/1000/year). In contrast there was an increase in the rate of some new diagnoses e.g. arthritis from 30.3 to 41.7 to 44.4/1000 individuals/year.Analysis of W9 data indicated that decreased age, decreased income and increased household size increased the rate of new diagnosis (p=0.2, p<0.01 and p<0.01 respectively). Analysis of CW1 and CW2 both showed a statistically significant effect for age and wealth only, in the direction as described above (p<0.05 for all).

Consequences

Decreased rates in the diagnosis of cancer, cardiovascular disease and dementia could be contributing to the increase in non-COVID-19 deaths. The reasons for these changes in diagnosis rates now requires further investigation to understand the factors from a patient and healthcare perspective that contribute to the lower rates of diagnosis for these important conditions.