Table 3.

Event study estimates of the effect of relative week according to week delta and omicron variant on GP use from 29 November 2021 to 23 January 2022

Relative week	OR	SE	P value	Lower CI	Upper CI	Test for equal OR (P value)*
Omicron
Week −5	1.02	0.11	0.86	0.83	1.25	0.85
Week −4	1.07	0.08	0.37	0.93	1.23	0.53
Week −3	0.97	0.08	0.75	0.83	1.15	0.13
Week −2	0.91	0.08	0.27	0.77	1.08	0.62
Week −1	1
Week 0	6.67	0.69	0.00	5.45	8.18	0.01
Week 1	5.51	0.72	0.00	4.26	7.12	<0.01
Week 2	1.51	0.12	0.00	1.29	1.77	0.46
Week 3	1.32	0.15	0.01	1.06	1.64	0.70
Week 4	0.86	0.29	0.65	0.44	1.68	0.18
Delta
Week −5	0.97	0.24	0.90	0.60	1.57
Week −4	1.18	0.18	0.27	0.88	1.60
Week −3	1.20	0.11	0.04	1.01	1.44
Week −2	0.96	0.07	0.63	0.83	1.12
Week −1	1				1
Week 0	8.19	0.52	0.00	7.24	9.27
Week 1	7.10	0.93	0.00	5.50	9.17
Week 2	1.39	0.08	0.00	1.24	1.56
Week 3	1.26	0.08	0.00	1.11	1.42
Week 4	1.34	0.08	0.00	1.19	1.50

Regression results from the main specification using a logit model. Number of person-weeks=4 906 952. We included controls for year of birth, calendar week, region of residence, country of birth, sex and parents’ country of birth. SEs were clustered at the municipality level.

*The column shows p values from Wald tests of equal OR for omicron versus delta, based on the regression output.

GP, general practitioner.