Table 2

Fixed effects models testing associations between changes in time spent on leisure activities and changes in mental health and well-being across the follow-up period (6 April 2020–6 September 2020)

	Depressive symptoms		Anxiety symptoms		Life satisfaction
	Coef (95% CI)	P value	Coef (95% CI)	P value	Coef (95% CI)	P value
Reading for pleasure
Low	0.01 (−0.03 to 0.05)	0.548	0.02 (−0.02 to 0.06)	0.283	0.01 (−0.03 to 0.06)	0.531
High	0.02 (−0.02 to 0.06)	0.311	0.01 (−0.03 to 0.05)	0.610	0.01 (−0.04 to 0.06)	0.648
Arts/crafts
Low	0.04 (0.00 to 0.08)	0.083	0.01 (−0.02 to 0.05)	0.416	−0.01 (−0.06 to 0.04)	0.730
High	0.01 (−0.03 to 0.06)	0.579	−0.01 (−0.06 to 0.03)	0.621	0.06 (0.00 to 0.11)	0.034
Digital arts
Low	0.01 (−0.03 to 0.06)	0.582	−0.01 (−0.04 to 0.03)	0.753	−0.04 (−0.09 to 0.02)	0.180
High	0.04 (−0.03 to 0.10)	0.251	0.04 (−0.01 to 0.09)	0.161	−0.05 (−0.12 to 0.02)	0.166
Gardening
Low	−0.07 (−0.11 to 0.02)	0.003	−0.05 (−0.09 to 0.01)	0.015	0.04 (−0.01 to 0.09)	0.107
High	−0.05 (−0.09 to 0.00)	0.046	0.00 (−0.06 to 0.05)	0.882	0.06 (0.00 to 0.12)	0.043
Watching television
Low	0.04 (−0.01 to 0.09)	0.148	−0.02 (−0.07 to 0.03)	0.457	0.00 (−0.07 to 0.06)	0.948
High	0.06 (0.01 to 0.10)	0.010	−0.02 (−0.06 to 0.03)	0.479	−0.01 (−0.06 to 0.04)	0.785
Listening to music
Low	−0.01 (−0.04 to 0.03)	0.725	−0.01 (−0.04 to 0.02)	0.407	0.00 (−0.04 to 0.04)	0.927
High	−0.02 (−0.06 to 0.02)	0.346	−0.01 (−0.05 to 0.02)	0.457	0.00 (−0.04 to 0.04)	0.870
Woodwork/DIY
Low	0.02 (−0.05 to 0.08)	0.604	0.02 (−0.05 to 0.08)	0.619	−0.04 (−0.10 to 0.03)	0.260
High	0.00 (−0.05 to 0.05)	0.984	−0.02 (−0.07 to 0.03)	0.428	0.11 (0.03 to 0.18)	0.004
Other hobbies
Low	0.02 (−0.03 to 0.07)	0.376	0.03 (−0.02 to 0.07)	0.235	−0.04 (−0.09 to 0.02)	0.158
High	0.01 (−0.03 to 0.06)	0.577	0.00 (−0.04 to 0.05)	0.858	−0.02 (−0.09 to 0.04)	0.438

Note. Time spent doing activities measured on the last weekday. Low=less than 30 min doing activity during the day. High=30 min or more spent on activity during the day. Both low and high were compared with doing none of this activity. Outcomes were standardised, so coefficients represent changes in SD units. Bold text indicates p<0.05. Fixed effects analysis does not test the direction of associations, so we cannot conclude that changes in leisure activities lead to subsequent changes in mental health and well-being.