Impact of COVID-19 pandemic on depression incidence and healthcare service use among patients with depression: an interrupted time-series analysis from a 9-year population-based study

Table 1 Generalized linear regression of pandemic impact on the first-year healthcare resource utilization among cohorts of incident patients

	A&E	Inpatient admission	Inpatient bed-day	Outpatient, all-cause	Outpatient, psychiatric	All visits, all-cause
First-year rates of healthcare resource utilization
2014 incident cohort^a	0.67	0.67	5.04	9.55	4.5	10.89
2015 incident cohort	0.50	0.48	4.84	9.54	4.55	10.52
2016 incident cohort	0.50	0.49	5.23	9.48	4.62	10.48
2017 incident cohort	0.58	0.63	4.80	9.62	4.58	10.83
2018 incident cohort	0.57	0.63	4.56	9.49	4.73	10.69
2019 incident cohort^a	0.41	0.40	2.75	7.42	2.91	8.23
2020 incident cohort (diagnosed in the pandemic period)	0.45	0.50	3.68	8.03	3.45	8.98
2021 incident cohort (diagnosed in the pandemic period)	0.54	0.61	4.12	9.01	4.12	10.15
2022 incident cohort (diagnosed in the pandemic period)	0.46	0.51	3.40	8.33	3.81	9.30
Impact of pandemic on the first-year rate of healthcare resource utilization^b
RR	0.891	0.971	0.769	0.887	0.821	0.892
95% CI	0.792–1.001	0.823–1.147	0.699–0.846	0.846–0.930	0.764–0.883	0.846–0.940
p-value	0.052	0.728	< 0.001^*	< 0.001^*	< 0.001^*	< 0.001^*

Healthcare resource utilization was studied only in the first calendar year of diagnosis in each cohort
Abbreviations: A&E Accident & Emergency, CI confidence interval, RR risk ratio
^*Statistical significance at 0.05 in generalized linear regression using negative binomial log-link function
^aPatients were newly diagnosed in the year of major social movement
^bAnalysis excluded data of the 2014 and 2019 cohorts to adjust for the confounding effect of social movement

ISSN: 1741-7015