Factors related to increased alcohol misuse by students compared to non-students during the first Covid-19 lockdown in France: the Confins study

Kinouani, Shérazade; Macalli, Mélissa; Arsandaux, Julie; Montagni, Ilaria; Texier, Nathalie; Schück, Stéphane; Tzourio, Christophe

doi:10.1186/s12889-024-18182-w

Table 6 Factors associated with an increase in binge-drinking frequency^a according to student status

From: Factors related to increased alcohol misuse by students compared to non-students during the first Covid-19 lockdown in France: the Confins study

	Students		Non-students
	Bivariate analysis^b	Model 1^c	Bivariate analysis^b	Model 1^d
Characteristics	OR (95% CI)	AOR (95% CI)	OR (95% CI)	AOR (95% CI)
≥ 25 years (ref. 18–24 years)	1.81 (0.76–4.34)	1.80 (0.71–4.54)	/	/
Men (ref. women)	1.23 (0.48–3.18)	1.31 (0.48–3.55)	2.27 [0.99–5.80]	/
Living with a partner (ref. single or other)	/	/	/	/
Studying in the health field (ref. no studying in the health field)	0.49 (0.20–1.21)	/	/	/
Self-rated economic situation during childhood or adolescence among students
• Very comfortable to comfortable	Ref	/	/	/
• Suitable	0.66 (0.22–2.04)	/	/	/
• Difficult to very difficult	3.15 (0.98–10.08)	/	/	/
Current main source of income among students
• Family	Ref	/	/	/
• Scholarship	0.81 (0.23–2.91)	/	/	/
• Paid activities	0.58 (0.16–2.08)	/	/	/
• Others (student loan, thrift, etc.)	3.03 (0.93–9.87)	/	/	/
Medical diagnosis of mental disorders during lifetime (ref. no medical diagnosis of mental disorders)	1.74 (0.73–4.16)	/	2.49 (1.04–5.94)*	/
Tobacco smoking in the past 12 months (ref. former smoking or no smoking)	3.27 (1.41–7.61)**	2.99 (1.25–7.13)*	3.05 (1.33–7.00)**
• Men			/	13.30 (3.21–55.10) ***
• Women			/	1.58 (0.33–4.08)
Alcohol use disorder^e in the past 12 months (ref. no alcohol use disorder)	/	/	/	/
Having been affected (formerly or currently) by Covid-19, with or without diagnostic test (ref. no having affected by Covid-19)	/	/	3.11 (1.14–8.47)*	/
Lockdown alone (ref. no being alone during lockdown)	2.25 (0.91–5.59)	/	/	/
Being confined in a French zone with an excess of deaths (ref. no confined in risk zone)	/	/	/	/
Physical activity in the last seven days (ref. no physical activity)	0.30 (0.13–0.70)**	0.36 (0.15–0.89)*	0.24 (0.10–0.58)**	0.26 (0.10–0.68)**
Suicidal thoughts in the last seven days (ref. no suicidal thoughts)	4.50 (1.89–10.67)***	4.24 (1.67–10.64)**	/	/
PHQ-9 ≥ 10^f			2.67 (1.12–6.40)*	/

AOR Adjusted odds ratio, OR Odds ratio, Ref Reference, CI Confidence interval
^*p < 0.05, **p ≤ 0.01, ***p ≤ 0.001
^aThe change in binge-drinking during lockdown was explored in those who drank alcohol at least once a month and had previously experienced binge-drinking during the past 12 months
^bBivariate analysis: tested potential factors were categorized age; gender; living with partner; working (or studying) in the health field; medical diagnosis of mental disorders during lifetime; smoking status in the past year; alcohol use disorder in the past year; having been affected (formerly or currently) by Covid-19, with or without a diagnostic test; lockdown alone; being confined in a French zone with an excess of deaths; physical activity in the last 7 days; suicidal thoughts in the last 7 days; PHQ-9 ≥ 10; GAD-7 ≥ 10; self-rated economic situation during childhood or adolescence; and current main income source (collected only among students). Only those associated with the outcome with significance at 0.25 were included into the initial multivariate logistic regression model and are presented in Table 6. PHQ-9 ≥ 10, GAD-7 ≥ 10, and suicidal thoughts were collinear. Even when these variables were associated with outcome in the bivariate analysis at the p < 0.25 threshold, none of the three were introduced into the initial multivariate logistic regression models. The variable “suicidal thoughts” was kept in the initial logistic regression model among students, PHQ-9 ≥ 10 was kept in among non-students
^cModel 1 (n = 593): multivariate logistic regression model with backward stepwise selection by forcing age in classes and gender in the models on complete data (no missing data)
^dModel 1 (n = 307): multivariate logistic regression model with backward stepwise selection by forcing gender into the models on complete data (no missing data). The variable age in classes was not forced in this model; it did not converge in the bivariate analysis due to the small sample size. An interaction between gender and smoking status was significant at the p < 0.25 threshold and introduced into the multivariate logistic regression model
^eAlcohol use disorder was only explored in those who declared having drank alcohol at least once a month in the past 12 months. It was defined according to AUDIT-C score ≥ 3 in women and ≥ 4 in men
^fPHQ-9: Patient Health Questionnaire 9 items scored as 0–4 (none), 5–9 (mild), 10–14 (moderate), 15–19 (moderately severe), 20–27 (severe)

Back to article page

ISSN: 1471-2458

Contact us

Submission enquiries: bmcpublichealth@biomedcentral.com
General enquiries: ORSupport@springernature.com

BMC Public Health

Contact us