Factors associated with a change in smoking habit during the first COVID-19 lockdown: an Italian cross-sectional study among ever-smokers

Table 5 Distribution of the 416^a current smokers before lockdown according to an improvement in smoking habit (multinomial logistic regression analysis)

Characteristics	Current smokers before lockdown^c	Quitting smoking during lockdown		Still current smokers, but decreasing n° of cigarettes per day during lockdown
Characteristics	Current smokers before lockdown^c	%	OR^b (95% CI) quit vs not quitting and no decrease	%	OR^b (95% CI) decrease vs not quitting and no decrease
Total	416^a	10.1		13.5
Sex
Men	160	8.8	1.00^d	9.4	1.00^d
Women	256	10.9	1.68 (0.83–3.40)	16.0	1.75 (0.90–3.40)
Age group (years)
18–39	124	10.5	1.00^d	22.6	1.00^d
40–54	147	13.6	0.70 (0.28–1.73)	12.3	0.63 (0.28–1.41)
55+	145	6.2	0.25 (0.09–0.71)	6.9	0.29 (0.11–0.74)
p for trend			0.005		0.008
Group of participants
SCSs	215	13.5	1.00^d	11.6	1.00^d
Healthcare provider	149	6.1	0.28 (0.12–0.66)	9.4	0.50 (0.24–1.06)
Students	52	7.7	0.31 (0.08–1.18)	32.7	1.53 (0.59–3.96)
Marital status
Married/cohabitants	196	10.2	1.00^d	11.2	1.00^d
Divorced/separated	66	6.1	0.56 (0.18–1.80)	12.1	1.04 (0.42–2.57)
Widowed	18	16.6	1.71 (0.41–7.10)	5.6	0.54 (0.06–4.49)
Single	136	11.0	1.07 (0.47–2.44)	18.4	0.78 (0.35–1.78)
Pack Years
I tertile – low (< 13)	148	9.5	1.00^d	24.4	1.00^d
II tertile	137	13.1	0.83 (0.31–2.17)	12.4	0.37 (0.16–0.85)
III tertile – high (≥28.35)	131	7.6	0.45 (0.14–1.44)	2.3	0.05 (0.01–0.21)
p for trend			0.158		< 0.001

Distribution of the 416^a current smokers before lockdown according to an improvement in smoking habit (i.e., smoking cessation or reduction in number of cigarettes per day) due to the COVID-19 lockdown, overall and by socio-demographic and smoking-related characteristics. Corresponding odds ratios^b (ORs) and 95% confidence intervals (CIs). (northern Italy, 2020)
SCS Smoking Cessation Service
^aOf the total 423 current smokers before lockdown, seven subjects did not report their change in smoking habit during lockdown
^bEstimated by multinomial logistic regression models after adjustment for sex, age, group of participants (SCSs; healthcare provider; students); estimates in bold are those statistically significant at the 0.05 level
^cThe sums does not add up to the total because of a few missing values
^dReference Category

ISSN: 1471-2458