Impact of the pandemic on leisure physical activity and alcohol consumption – who were affected and who were not? Results from a Swedish longitudinal cohort study using two different methods to measure change

doi:10.21203/rs.3.rs-3485945/v1

Download PDF

Research Article

Impact of the pandemic on leisure physical activity and alcohol consumption – who were affected and who were not? Results from a Swedish longitudinal cohort study using two different methods to measure change

https://doi.org/10.21203/rs.3.rs-3485945/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The COVID-19 pandemic was associated with increased morbidity and high mortality due to the Sars-Cov-2 virus, but the pandemic also influenced health behaviors such as physical activity (PA) and alcohol consumption. The aim of this study was to examine potential changes in leisure PA and alcohol consumption in Sweden during the pandemic, and to explore whether identified changes varied between sex, age groups and socioeconomic status (SES).

Methods

Data were retrieved from two waves of the longitudinal cohort study Life conditions, Stress and Health (LSH) (n = 2,523). Two measures of change were used; longitudinal change relative to baseline (2012–2015) and reported change compared to before the pandemic. For these two change measures, differences between sex, age group and SES were analyzed using multinomial logistic regression.

Results

Irrespective of change measure, the proportion with decreased PA was larger among women than men. Also, relative to baseline, women were less likely to having increased PA. However, according to the reported change relative to before the pandemic, women were more likely to have increased PA. Longitudinal change in PA compared to baseline followed a reversed age gradient, while, according to reported change, a decrease in PA during the pandemic was most prevalent in respondents 45 years of age at baseline (OR = 1.8, CI: 1.2–2.5) and respondents 50 years of age at baseline (OR = 1.7, CI: 1.2–2.4). High SES was associated with a greater variability in PA. Alcohol consumption was generally reduced during the pandemic. However, individuals aged 40 or 45 years at baseline were more likely than others to have initiated risky alcohol consumption.

Conclusions

Women were more likely to change their PA levels during the pandemic. The most profound decreases in PA were found in individuals of working ages. Although the pandemic led to a general decrease in alcohol consumption, individuals aged 40 and 45 were more likely than other ages to have initiated risky alcohol consumption. In conclusion, societal restrictions during a pandemic pose a risk for decreased PA among individuals in working ages, but can also open a window of opportunity, not least for women, to increase their PA levels.

COVID-19

leisure physical activity

alcohol consumption

age

sex

socioeconomic status

longitudinal

questionnaire

The outbreak of the COVID-19 pandemic in early 2020 had a tremendous direct effect on population health, with high morbidity and mortality due to Sars-Cov-2. The pandemic was accompanied by confinements and recommendations introduced by governments and authorities to prevent the spreading of infection. Also, to protect themselves many individuals, not the least elderly persons or those in medical risk groups, remained in solitude to a higher extent than prescribed by official regulations. These changed life circumstances posed risks of indirect effects of the COVID-19 pandemic, e.g., social isolation, job loss, financial difficulties, stress and anxiety (1). These indirect effects could also lead to changes in leisure physical activity (PA) and alcohol consumption. During most of the pandemic, working from home, if possible, was mandatory or recommended in most countries. Studies have identified an increased risk of physical inactivity related to the transition to home working (2, 3). Also, many fitness centers and sport centers were placed in lockdown, resulting in reduced opportunities for maintaining existing PA habits. Similarly, patterns of alcohol consumption were affected by the pandemic. A majority of restaurants and event venues which normally serve alcohol were closed, while most private social arrangements were prohibited. Consequently, availability of alcohol beverages in public settings was dramatically reduced, and alcohol consumption was thereby transferred to the privacy of the home environments. While this restricted socially induced drinking, it can be worrisome given that studies have shown that solitary drinking sometimes is fraught with higher alcohol consumption than social drinking (4). Furthermore, stressful life experiences are known to be related to excessive alcohol consumption (5), and societal crisis situations tend to increase alcohol consumption, especially among men, young people and those living in single households (6).

It is urgent to be aware of systematic population level changes in PA or alcohol consumption, as both physical inactivity and excessive alcohol consumption are major risk factors for poor health. Physical inactivity is highlighted as one major cause of chronic diseases (7), and misuse of alcohol is an important contributor to the global disease burden (8). Also, both physical inactivity and alcohol consumption have been shown to account for a substantial preventable premature mortality worldwide (9). Recently, it has been reported that even moderate alcohol consumption is a risk factor for ill-health (10, 11).

Both a systematic review (12) and an international online survey covering countries all over the world consistently found that the pandemic led to an overall decrease in PA (13). However, another review of studies examining the change in PA during to the pandemic reported mixed results (14). Even though the majority of included studies found decreases in PA, about one third reported either an increase or a mixed pattern.

Already before the pandemic, it was well established that PA habits differ among population groups. In Sweden, such examples include the fact that physical inactivity increases by age, and that the proportion of physically inactive individuals is larger in low socioeconomic status (SES) groups (15). It is also possible that the impact of the pandemic on PA differs among population groups. In one UK study, decreases in PA were more common in women (16), whereas other European studies have reported that the overall decrease in PA during the pandemic was especially manifested in men (17–20). A review of studies examining changes in PA during the pandemic found no consistent differences in changes between age groups (14). Several studies have found increasing SES inequalities in PA during the pandemic (21), both regarding level of education (20, 22, 23) and income (23, 24). However, a Finnish study comparing responses from the same individuals before and during the pandemic found no significant differences in change in PA across groups with different educational levels (25). Thus, whether the pandemic affected population groups differently in terms of changes in PA is still not fully known.

Existing studies do not offer a definite picture of the overall change in alcohol consumption during the pandemic. Two systematic reviews both found a mixed pattern with slightly more studies reporting an increased consumption than those reporting a decrease (12, 26). On the other hand, a cross-sectional online survey of adults in 21 European countries suggested an overall decrease in alcohol consumption compared to before the pandemic (27). The same conclusion was drawn in a meta-analysis of European studies examining changes in alcohol use during the pandemic (28). A Norwegian study reported a modest overall decrease in alcohol consumption, but an increase in the proportion of heavy drinkers (29).

Like for PA, patterns of alcohol consumption differed across population groups already before the pandemic. Self-reported risk-consumption of alcohol in Sweden was more common among men and was less common among men and women aged 65 years or above (15).

There are some findings indicating that the impact of the pandemic on alcohol consumption differed across population groups, even though they are not fully consistent. Several studies have observed that men were more likely than women to increase their alcohol consumption during the pandemic (30–32), whereas a Norwegian study found that women were more likely to increase their alcohol consumption (33), and an Australian study found no differences between men and women (34). Several studies have concluded that high educational level (33, 35, 36) and/or high income (34) are associated with an increase in alcohol consumption during the pandemic. On the other hand, one study from the UK observed no associations between SES and change in alcohol consumption (37).

Change can be measured in several ways. The two most common ways are either to make a longitudinal comparison of self-reported assessments from the same individual at two different time points (measured change or indirect measure of change) or to ask the individuals whether any change has occurred from one time point to another (reported change or direct measure of change). Evaluations have shown that results sometimes can differ substantially depending on the type of change measure (38). Some studies on the impact of the pandemic on PA or alcohol consumption mentioned above were based on cross-sectional surveys using reported change, whereas others were based on longitudinal surveys using measured change. Consequently, even beyond the contextual differences among countries, the selection of specific change measures likely explains some of the disparity of results regarding the impact of the pandemic on PA and alcohol consumption.

Sweden handled the pandemic in a different way than most other countries, as there were no total lockdowns. Measures were mostly in the form of recommendations, which opened for a larger extent of individual freedom and interpretation in terms of how strictly to adhere to them. It is therefore possible that the impact of the pandemic on PA and alcohol consumption was different in Sweden compared to many other countries. Some previous Swedish studies have explored how PA habits were affected by the pandemic. One cross-sectional study reported an overall decrease in PA after the onset of the pandemic, with no differences in changes between men and women. The largest decreases were found in the youngest and oldest age groups (39). Another Swedish cross-sectional study identified the largest decreases in PA in the oldest age group (70 + years) (40). A comparison of two independent Swedish cross-sectional surveys (from 2018 and 2021) addressed to the general population 16–84 years old, detected increased SES inequalities in PA in 2021 (during the pandemic) compared to 2018 (before the pandemic) (41). However, corresponding findings based on longitudinal studies are still lacking. Further, we have not found studies on change in alcohol consumption across population groups in Sweden during the pandemic.

Knowledge on how health inequalities are potentially exacerbated by a pandemic in terms of health behavior could play an important role for professionals working with prevention and health promotion. Based on such knowledge, appropriate strategies, designed both to effectively reduce the spread of the infection and to minimize negative indirect effects on health behaviors, can be developed.

Therefore, it is urgent to evaluate the Swedish strategy of tackling the pandemic, and its effects on health behaviors. Further, it is important to know if potential effects vary across population groups. To make this evaluation more robust, and to elucidate possible differences across results, this study analyzed both longitudinal measured change and cross-sectional reported change.

AIM

This study examined changes in leisure PA and alcohol consumption among middle aged and elderly Swedish persons before and during the pandemic, by using both a longitudinal change measure and a reported change measure. In addition, the aim was to study to what extent these changes differ between sex, age or SES.

This study utilized data from two waves of the longitudinal cohort study Life conditions, Stress and Health (LSH). The study population included a random sample of persons aged 40, 45, 50, 55, 60, 65 and 70 years in two counties in south-eastern Sweden (Östergötland and Jönköping). Between 2012 and 2015, the sampled persons were invited by the county councils to their primary health care center for a health dialogue. Participants who attended these health dialogues were invited to participate in a wave of the LSH study called LSH II, which for the present study constitutes our baseline data. In total, 28,702 citizens were invited to health dialogues, of which 12,164 (42%) accepted. Of those, 6,860 (56%) individuals (3,880 women and 2,980 men) also agreed to participate in the LSH II study. In total, LSH II participants responded three questionnaires, two as a preparatory basis for the health dialogue and one additional LSH-specific questionnaire.

During the pandemic, in January 2021, a follow-up questionnaire was distributed to LSH II respondents in one of the two counties (Östergötland) (n = 3,643), covering in part the same questions as the baseline questionnaires, with the addition of several pandemic-related questions. This data collection was completed in April 2021 after one postal reminder. The response rate for the COVID questionnaire was 73%, resulting in a sample of 2,523 persons, 1,493 women and 1,030 men.

Indicators of SES

Educational level and disposable household income were chosen as indicators of SES. This information was retrieved at baseline from registry data.

Educational level was measured in terms of highest achieved education and was divided into five categories: ‘Only elementary school’; ‘Two-year of secondary school’; ‘Three-or-four-year of secondary school’; ‘University studies of maximum three years’; and ‘University degree more than 3 years’.

Household income was divided into quartiles based on total disposable household income after taking the number of household members into account.

Leisure PA

Both in the baseline and in the COVID questionnaires, information on leisure PA was captured from the question: ‘How much do you exercise physically in your leisure time?’, with options: ‘Sedentary leisure time’, ‘Moderate exercise’, ‘Strenuous exercise’ and ‘Intense exercise’. The two last response categories were combined into one category (vigorous exercise). Also, the COVID questionnaire included a question prompting respondents to assess how their PA habits had changed since before the pandemic, with a bipolar response scale: ‘Considerably more now’, ‘Somewhat more now’, ‘unchanged’, Somewhat less now’ and ‘Considerably less now’. In the analyses, the two first and the two last categories were combined, resulting in a three-level variable: ‘More’, ‘Unchanged’ and ‘Less’.

Alcohol consumption

Information on the extent of risky alcohol consumption was assessed by two questions targeting how much alcohol the respondents typically drink in one week and how often the respondents drink a defined large amount on one occasion. If respondents typically have more than 9 drinks a week (women)/14 drinks a week (men) or 4 drinks (women)/5 drinks (men) on one occasion, their alcohol consumption was classified as risky. The questions were identical at baseline and in the COVID questionnaire. In addition, the COVID questionnaire included a question prompting respondents to assess how their alcohol habits had changed compared to before the pandemic, with options: ‘Drink considerably more now’, ‘Drink somewhat more now’, ‘unchanged’, ‘Drink somewhat less now’ and ‘Drink considerably less now’. In the analyses, the two first and the two last response categories were combined, resulting in a three-level variable: ‘More’, ‘Unchanged’ and ‘Less’.

Potential confounders

Severe health problems, change in cohabitation status and impact of the pandemic on the private financial situation were considered as potential confounding variables for the associations under study.

Respondents were assessed as having severe health problems if they in the baseline questionnaire reported severe problems due to any of the following physician diagnosed conditions: myocardial infarction or stroke, angina pectoris, cancer, chronic lung disease, rheumatoid arthritis, asthma or allergy, gastrointestinal disease, musculoskeletal disorders, neurological disease, renal disease or depression.

Change in cohabitation status was assessed by questions at both questionnaires targeting whether the respondents were married/cohabiting with a partner or not. Impact of the pandemic on the private financial situation of respondents was assessed by a question in the COVID questionnaire asking: “What impact has the pandemic in total had on your financial situation?”, with options: ‘No impact’, ‘Very negatively’, Somewhat negatively’, ‘Somewhat positively’, and ‘Very positively’.

Statistical analyses

Longitudinal and reported change of leisure PA and alcohol consumption, respectively, were used as outcome measures and tested for associations with sex, age and the two indicators of SES (educational level and disposable household income). Bivariate associations were tested by chi-two tests. Multinomial logistic regression analysis was used to estimate odds ratios (OR) with 95% confidence intervals (95% CI), to describe the associations adjusted for confounders.

The adjustment was made in two steps. First, the associations were adjusted for status at baseline in terms of extent of severe health problems, age (in models including SES), leisure PA (in models including changes in PA), risky alcohol consumption (in models including changes in risky alcohol consumption) (Model I). In order to examine whether the observed differences in change could be explained by material/structural factors (change in cohabitation status and/or impact of the pandemic on the private financial situation), associations were in a second step also adjusted for these variables (Model II).

All analyses were performed using SPSS version 29 (IBM Corp, Armonk, New York, USA).

Leisure PA

Overall, there were more respondents reporting decreases in leisure PA levels than reporting an increase according to measured change (19.9% vs. 12.8%). About the same ratio was seen according to reported change even though the levels were almost 7 percentage units higher (26.7% vs. 19.5%) (Table 1). More women than men decreased their leisure PA levels during the pandemic, both according to the longitudinal change measure (21.2% vs. 18.0%) and the reported change measure (30.4% vs. 21.3%). The proportion of respondents with a longitudinal decrease in their PA were higher (23.0%) in the oldest group (70 years of age at baseline) than in other age groups (Table 1). For both change measures, the respondents 45 years of age at baseline and respondents 50 years of age at baseline more often than most other age groups decreased their PA (Table 1).

Table 1

Overview of change in leisure PA, longitudinal change and reported change
		Longitudinal change in leisure PA					Reported change in leisure PA
		Unchanged	More PA	Less PA	Total	p-value	Unchanged	More PA	Less PA	Total	p-value
Total n (%)		1576 (67.3)	301 (12.8)	466 (19.9)	2343		1343 (53.8)	486 (19.5)	665 (26.7)	2494
Sex	Men	654 (66.9)	148 (15.1)	176 (18.0)	978	0.007	635 (62.1)	170 (16.6)	218 (21.3)	1023	< 0.001
	Women	922 (67.5)	153 (11.2)	290 (21.2)	1365		708 (48.1)	316 (21.5)	447 (30.4)	1471
Age	40	148 (63.5)	48 (20.6)	37 (15.9)	233	< 0.001	118 (47.4)	62 (24.9)	69 (27.7)	249	< 0.001
	45	168 (59.8)	42 (14.9)	71 (25.3)	281		141 (47.2)	62 (20.7)	96 (32.1)	299
	50	174 (62.1)	39 (13.9)	67 (23.9)	280		150 (51.4)	47 (16.1)	95 (32.5)	292
	55	226 (67.3)	44 (13.1)	66 (19.6)	336		180 (50.0)	69 (19.2)	111 (30.8)	360
	60	282 (70.0)	57 (14.1)	64 (15.9)	403		228 (53.6)	106 (24.9)	91 (21.4)	425
	65	307 (71.1)	51 (11.8)	74 (17.1)	432		277 (60.2)	81 (17.6)	102 (22.2)	460
	70	271 (71.7)	20 (5.3)	87 (23.0)	378		249 (60.9)	59 (14.4)	101 (24.7)	409
Educational level	Compulsory	187 (70.0)	31 (11.6)	49 (18.4)	267	0.18	186 (64.1)	49 (16.9)	55 (19.0)	290	< 0.001
	Secondary school 2 years	446 (68.6)	76 (11.7)	128 (19.7)	650		431 (63.2)	102 (15.0)	149 (21.8)	682
	Secondary school 3 years	249 (70.3)	41 (11.6)	64 (18.1)	354		207 (55.1)	73 (19.4)	96 (25.5)	376
	Post-secondary school 3 years	265 (62.2)	58 (13.6)	103 (24.2)	426		227 (49.2)	94 (20.4)	140 (30.4)	461
	Post-secondary school more than 3 years	428 (66.6)	95 (14.8)	120 (18.7)	643		290 (42.6)	167 (24.6)	223 (32.8)	680
Household income	Q1	395 (67.8)	76 (13.0)	112 (19.2)	583	0.90	369 (59.2)	99 (15.9)	155 (24.9)	623	0.002
	Q2	385 (65.5)	83 (14.1)	120 (20.4)	588		350 (55.7)	131 (20.9)	147 (23.4)	628
	Q3	399 (67.6)	75 (12.7)	116 (19.7)	590		319 (51.6)	118 (19.1)	181 (29.3)	618
	Q4	396 (68.3)	67 (11.6)	117 (20.2)	580		303 (48.8)	138 (22.2)	180 (29.0)	621

For the reported change both more leisure PA (24.6%) and less leisure PA (32.8%) were more common among respondents with the highest educational level. Reports of unchanged leisure PA levels were more likely among respondents in lower educational level groups. When using the longitudinal change measure, the magnitude of differences between SES groups regarding changes in leisure PA was smaller (Table 1).

The proportion of respondents with a longitudinal increase in their leisure PA was lower among women than among men (11.2% vs. 15.1%). However, according to the reported change the proportion with an increase in PA was higher among women (21.5% vs. 16.6%).

(Insert Table 1 here)

Alcohol consumption

Alcohol consumption generally decreased during the pandemic, both in terms of a higher proportion of respondents reporting risky consumption at baseline but not during the pandemic (20.6%) than the other way round (7.6%), and according to reported change in alcohol consumption from before the pandemic (13.4% reporting a decrease vs. 8.7% reporting an increase) (Table 2). The reported change in alcohol consumption did not significantly differ between sexes. The proportion of respondents who increased their alcohol consumption was highest among those 40 years of age at baseline (46–49 years during the pandemic). This applied both the proportion of respondents that had risky alcohol consumption during the pandemic, but not at baseline (9.9%) and the proportion reporting an increase in alcohol consumption from before the pandemic (14.2%) (Table 2). The proportion of respondents with risky alcohol consumption of alcohol during the pandemic, but not at baseline was also high (10.0%) in the oldest group (70 years of age at baseline).

Table 2

Overview of change in alcohol consumption, longitudinal change and reported change
		Longitudinal change in risk consumption of alcohol					Reported change in alcohol consumption
		Unchanged	Change from consumption not considered risky to risky consumption	Change from risky consumption to a consumption not considered risky	Total	p-value	Unchanged	Drinks more	Drinks less	Total	p-value
Total n (%)		1658 (71.8)	175 (7.6)	475 (20.6)	2307		1884 (77.9)	210 (8.7)	325 (13.4)	2419
Sex	Men	654 (69.7)	63 (6.7)	220 (23.6)	937	< 0.001	766 (77.5)	93 (9.4)	130 (13.1)	989	0.56
	Women	1004 (73.3)	112 (8.2)	254 (18.5)	1370		1118 (78.2)	117 (8.2)	195 (13.6)	1430
Age	40	160 (69.0)	23 (9.9)	49 (21.1)	232	< 0.001	163 (66.0)	35 (14.2)	49 (19.8)	247	0.002
	45	213 (75.0)	20 (7.0)	51 (18.0)	284		218 (73.6)	27 (9.1)	51 (17.2)	296
	50	193 (72.6)	17 (6.4)	56 (21.1)	266		228 (80.0)	22 (7.7)	35 (12.3)	285
	55	252 (75.4)	16 (4.8)	66 (19.8)	334		279 (80.2)	26 (7.5)	43 (12.4)	348
	60	282 (71.2)	31 (7.8)	83 (21.0)	396		328 (79.2)	34 (8.2)	52 (12.6)	414
	65	303 (71.3)	31 (7.3)	91 (21.4)	425		355 (80.9)	38 (8.7)	46 (10.5)	439
	70	255 (68.9)	37 (10.0)	78 (21.1)	370		313 (80.3)	28 (7.2)	49 (12.6)	390
Educational level	Compulsory	190 (73.9	23 (8.9)	44 (17.1)	257	0.04	221 (79.8)	22 (7.9)	34 (12.3)	277	< 0.001
	Secondary school 2 years	434 (69.7)	58 (9.3)	131 (21.0)	623		545 (83.3)	39 (6.0)	70 (10.7)	654
	Secondary school 3 years	265 (75.3)	17 (4.8)	70 (19.9)	352		290 (79.7)	34 (9.3)	40 (11.0)	364
	Post-secondary school 3 years	312 (72.1)	25 (5.8)	96 (22.2)	433		343 (76.2)	39 (8.7)	68 (15.1)	450
	Post-secondary school more than 3 years	455 (71.3)	51 (8.0)	132 (20.7)	638		481 (71.9)	76 (11.4)	112 (16.7)	669
Household income	Q1	381 (67.3)	62 (11.0)	123 (21.7)	566	< 0.001	468 (80.0)	42 (7.2)	75 (12.8)	585	0.25
	Q2	426 (74.2)	39 (6.8)	109 (19.0)	574		480 (78.9)	43 (7.1)	85 (14.0)	608
	Q3	422 (73.9)	28 (4.9)	121 (21.2)	571		465 (76.1)	63 (10.3)	83 (13.6)	611
	Q4	427 (72.0)	45 (7.6)	121 (20.4)	593		468 (76.6)	62 (10.1)	81 (13.3)	611

(Insert Table 2 here)

Among SES groups, both the proportion of respondents reporting increased alcohol consumption (11.4%) and the proportion of respondents reporting decreased alcohol consumption (16.7%) were highest in the highest education group. According to the longitudinal change measure, the proportion of respondents changing from alcohol consumption not considered risky at baseline to a risky alcohol consumption during the pandemic was somewhat higher (e.g., 11.0% in household income Q1) in the lowest SES groups.

Multivariate analyses

Leisure PA

Figure 1 & additional files 1 & 3 illustrate that the observation that women were more likely to decrease their leisure PA remained after adjusting for baseline level of leisure PA and prevalence of severe health problems (OR = 1.5, CI: 1.1–1.9 for the longitudinal change measure, OR = 1.8, CI: 1.4–2.2 for the reported change). Further adjustments for change in cohabitation status and impact of the pandemic on the private financial situation similarly did not change these findings.

(Insert Fig. 1 here)

After adjustment for baseline level of leisure PA and severe health problems, the longitudinal change in leisure PA followed an age gradient (Fig. 1 & additional file 1). The older the respondents the smaller the proportion with an increase in their leisure PA levels and the larger the proportion with a decrease. This pattern remained after further adjustment for material/structural factors. The likelihood of a reported decrease in leisure PA was highest among respondents 45 years of age at baseline (OR = 1.8, CI: 1.2–2.5) and respondents 50 years of age at baseline (OR = 1.7, CI: 1.2–2.4).

The pattern that respondents with the highest educational level were more likely to report an increase and a decrease in leisure PA, respectively, compared to respondents with other educational levels was consistent also after adjustment for baseline levels of PA and prevalence of severe health problems. In all other education groups OR was below 1 (additional file 3). This finding was also not explained by change in cohabitation status or impact of the pandemic on the private financial situation, as results did not change after full adjustment. A similar pattern was found when using household income as measure on SES. Respondents in the highest income quartile were most likely to either have increased or decreased their PA levels. However, the associations were not as strong as in the analyses utilizing educational level as measure of SES, and many differences were not statistically significant. In the bivariate analyses of the association between SES and the longitudinal change in leisure PA, no significant differences between SES groups were found (Table 1). However, respondents in the lower SES groups were considerably more likely to be physically inactive in their leisure time at baseline than respondents in higher SES groups (additional file 5). This meant that for a larger proportion of respondents in the lower SES groups a measured decrease was technically impossible, as they already belonged to the lowest category at baseline, compared to respondents in higher SES groups. When adjusting for baseline level of PA, all other education level groups, except for the one with the highest secondary education, were more likely to have had a longitudinal decrease in their PA levels compared to the reference group (post-secondary education longer than 3 years) (Fig. 1 & additional file 1).

In the bivariate analyses, the proportion of respondents with an increase in their leisure PA levels were smaller among women compared to men according to the longitudinal change measure, but larger among women according to the reported change measure. This result remained after the first step of adjustments (longitudinal change: OR = 0.7, CI: 0.5–0.9; reported change: OR = 1.6, CI: 1.3-2.0) (additional files 1 & 3). The proportion of respondents with a reported increase in their leisure PA was highest among those 40 years of age at baseline (OR = 1.8, CI: 1.2–2.7), 45 years of age at baseline (OR = 1.6, CI: 1.0-2.3), and 60 years of age at baseline (OR = 1.6, CI: 1.1–2.3).

Alcohol consumption

Neither the likelihood of changing from an alcohol consumption not considered risky at baseline to a risky alcohol consumption during the pandemic nor the reported change in alcohol consumption from before the onset of the pandemic significantly differed between men and women (Fig. 2).

(insert Fig. 2 here)

When adjusting for baseline alcohol consumption, the increased probability for respondents aged 70 years at baseline of having changed their alcohol consumption into a risky consumption was not statistically significant (OR = 1.4, OR: 0.8–2.4) (Fig. 2 & additional file 2). Thus, the findings in bivariate analyses were partly explained by a lower proportion of respondents with a risky alcohol consumption at baseline among the oldest respondents. On the other hand, respondents 40 years of age at baseline were more likely than the reference group (60 years of age) to have increased their alcohol consumption (Fig. 2). This result was found both in terms of longitudinal change (OR = 2.8, CI: 1.5–5.3) and reported change (OR = 2.0, CI: 1.2–3.3). This higher likelihood of an increase in alcohol consumption was partly explained by material/structural factors, as OR decreased somewhat after full adjustment. According to the longitudinal change measure, respondents 45 years old at baseline also had a higher probability of having changed their drinking habits to a risky alcohol consumption during the pandemic (OR = 1.9, CI: 1.0-3.7) (Fig. 2).

Table 3

Distribution of reported change in alcohol consumption (all response categories)
		Drinks considerably less	Drinks somewhat less	Unchanged	Drinks somewhat more	Drinks considerably more	Total (%)
Educational level	Compulsory	5	7	80	7	1	100
	Secondary school 2 years	5	6	83	6	< 1	100
	Secondary school 3 years	4	7	80	8	1	100
	Post-secondary school 3 years	6	9	76	8	1	100
	Post-secondary school more than 3 years	4	12	72	11	< 1	100
Household income	Q1	6	6	80	6	1	100
	Q2	5	9	79	7	1	100
	Q3	5	9	76	10	1	100
	Q4	3	10	77	10	< 1	100

The differences between SES groups in longitudinal change from baseline in the proportion of respondents having moved to risky alcohol consumption were still small after adjustment for baseline level of alcohol consumption (Fig. 2 and additional file 2). Respondents with the shortest secondary education (2 years) were most likely to have changed to risky alcohol consumption, with a borderline statistical significance (OR = 1.6, CI: 1.0-2.4) compared to the reference group (post-secondary education of more than 3 years). As in the bivariate analyses, a reported increase in alcohol consumption or a reported decrease in alcohol consumption, respectively, were both more common in the highest SES groups (additional file 4). However, as Table 3 shows, this circumstance only applies to the two response categories ‘Drink somewhat more’ and ‘Drink somewhat less’. For the more extreme changes: ‘Drink considerably more’ and ‘Drink considerably less’, the proportions of responses were not larger in the highest SES groups than in other SES groups.

In this study, where we used both a longitudinal change measure and a reported change measure, we showed that the extent of leisure PA was considerably affected by the pandemic across several population groups. Both increases and decreases in PA levels were commonplace. We also observed a general decrease in alcohol consumption, with no major differences across population groups.

Both according to the longitudinal change measure and the reported change measure, women were more likely than men to decrease their leisure PA levels after the onset of the pandemic. Most other studies on changes in PA during the pandemic have reported that a decrease in PA was more common among men (17–20, 42). However, similarly to this study, a British study found that women were more likely to decrease their levels of PA (16). On the other hand, based on reported change, women were also more likely to increase their PA, a result which was in line with results from a study from the UK (42). We also found that both longitudinal change and reported change indicated that younger respondents (40 or 45 years old at baseline) were more likely to increase their leisure PA levels. According to the longitudinal change measure, the likelihood of a reduction in leisure PA increased by age. However, according to the reported change, respondents 55 years or younger at baseline were more likely to report a decrease in leisure PA compared to respondents older than 55 years. As 6–9 years had passed from baseline to the outbreak of the pandemic, the majority of respondents 55 years or younger at baseline were likely to still be in the work force during the pandemic, while the majority of the respondents older than 55 years at baseline were probably retired during the pandemic. That younger respondents are more likely to increase their PA levels and older respondents more likely to decrease their PA can potentially be seen as a natural aging process. However, our opposite finding, based on the reported change, that persons in working ages more frequently decreased their leisure PA aligns well with the result of a Polish study (19). As many employees in Sweden worked from home during the part of the pandemic captured in the present study, the findings also echo previous studies suggesting an association between working from home and physical inactivity (2, 3). When using the reported change measure, a divergent finding in the present study compared to many previous studies (39, 40, 43, 44) was that the oldest respondents did not decrease their leisure PA levels more than respondents of other ages. This might be explained by an inability in some of the other studies to separate the overall trend of larger decrease in PA in older age groups from the specific pandemic-related change in PA.

Several previous studies have reported an increased SES gradient in PA during the pandemic (20–24, 41). Results based on the longitudinal change measure in the present study confirm this tendency, even though the time-period for the measurement in this study also included several years before the pandemic. However, when analyzing the reported change, specifically targeting change from the onset of the pandemic, respondents with the highest SES were on the contrary most likely to decrease their leisure PA.

Regardless of change measure, we found a general reduction in alcohol consumption during the pandemic. This reduction was most profound among respondents 40 and 45 years old at baseline. However, among respondents 40 years of age at baseline, the pattern was heterogeneous, also including a higher likelihood of increased alcohol consumption. Our finding that alcohol consumption more often changed among respondents of middle age compared to older respondents is in line with results from several studies (31, 34–36, 45–47). According to the longitudinal change measure, men were more likely to go from risky alcohol consumption at baseline to alcohol consumption not considered risky during the pandemic. However, the reported change in alcohol consumption during the pandemic did not differ between men and women. An interpretation of this result can be that the more profound reduction in alcohol consumption among men occurred well before the pandemic, and that the changes in alcohol pattern during the pandemic did not differ between men and women. Compared to respondents in other SES groups, respondents in higher SES groups both more often reported an increase in alcohol consumption and more often reported a decrease in alcohol consumption. Several previous studies have found a larger increase in alcohol consumption during the pandemic among those with high educational level (30, 33, 35, 36, 48, 49), and one study reported the same pattern as in the present study with both larger increases and larger decreases in alcohol consumption among respondents in higher education groups (42). However, in the present study these changes were not seen when using the longitudinal change measure, also targeting a period well before the pandemic, and the reported changes were exclusively explained by a higher proportion reporting ‘somewhat more’ or ‘somewhat less’.

An interesting feature of this study was that change in health behaviors was measured in two ways, both using a longitudinal change measure and a reported change measure. Reported change is often biased by both recall bias (50) and present state effect (51). However, recall bias has been shown not to be a major problem when the recall time interval is relatively short and the specified point in time to recall is meaningful to the respondent (38). The time interval to recall in the present study was about a year, but as the outbreak of the pandemic was such an overwhelming shift in life situations for the respondents recall bias was probably only of a minor importance. However, the results might still be biased by the present state effect. This effect implies that respondent´s assessments of change are affected by their current states. For instance, respondents currently physically inactive are more likely to report a decrease in PA even though their PA is unchanged. This means that reported changes often are overestimated (52). On the other hand, a disadvantage using a longitudinal change measure is that floor and ceiling effects sometimes bias assessments (53). This occurred in the present study, as respondents with leisure PA matching the lowest response category already at baseline did not have a response category matching even less leisure PA. Another drawback with the use of a longitudinal change measure is that small changes are sometimes not possible to detect when using only a few response categories. There might be that respondents have decreased or increased their leisure PA level, but that the new level still falls into the same response category.

In addition to the methodological differences discussed above, an important aspect in this study is that the reported change captures a time interval of about a year, whereas the longitudinal change measure captures a time interval of 6–9 years. This means that the disparate findings when using the two types of measures can be explained either by the validation problems discussed above or by the fact that some changes in leisure PA and/or alcohol consumption took place before the onset of the pandemic and was therefore not relevant in the reported change measurement. For instance, as mentioned above, the results for our longitudinal change measure aligned well with findings in previous studies in that SES differences in PA increased over time, including after the outbreak of the pandemic. However, the reported change complemented this picture by showing that specifically after the onset of the pandemic respondents with high SES more often decreased their leisure PA than respondents with lower SES. The divergent findings in relation to most other studies can either be explained by the fact that those studies were not able to disentangle the specific pandemic effect from the long-term trend of increasing SES inequalities in PA, or that our results were influenced by a present state effect resulting in an overestimation of the reported decrease in PA among respondents with high SES. They might well have had an ambition to spend more time with leisure PA during the pandemic than they were actually able to, and therefore had an impression that they had decreased their PA levels even though this was not the case.

Previous studies have shown that increased financial difficulties or economic worries due to the pandemic situation were associated with decreased PA and increased alcohol consumption (54, 55). However, in the present study our results did not change when adjusting for the impact of the pandemic on the private financial situation.

A strength of this study is the longitudinal cohort design complemented with retrospective questions that enables measuring change both using a longitudinal change measure and a reported change measure. Another strength is the comprehensive design of the LSH research program, which includes multiple health risk behaviors and a broad range of variables capturing life conditions and change in life conditions. The population-based sample made it possible to adjust analyses for possible confounders. As the subgroups in the study (defined by sex, age or SES) were all represented by substantial numbers of respondents, our assessment is that the results are not limited by power. A potential shortcoming is that reported change in leisure PA and alcohol consumption, respectively, were measured by single questions with only a limited number of response categories. This means that minor nuances in behavioral patterns and changes might have been missed. It also implies that we cannot determine if, for instance, a decrease in leisure PA poses a risk of health deterioration. Also, an aspect that calls for caution in the interpretation of the results of the longitudinal change measure is that the questionnaires at baseline was accompanied by a health dialogue, whether the COVID questionnaire did not involve any personal interaction with health care representatives. This implies a risk that responses at baseline were more biased towards social desirability. This can probably especially apply to questions targeting alcohol consumption. However, the proportion of respondents that reported a risky alcohol consumption was still higher at baseline. If a social desirability bias existed, the actual decrease in alcohol consumption would likely be even more pronounced than in our findings. Finally, the sample only consists of individuals 40 years and older at baseline, which means that our findings cannot be generalized to the general adult Swedish population.

Based on data from a longitudinal cohort study in Sweden, we found a general decrease in leisure PA, and that women were more likely to change their PA levels during the pandemic. The most profound decreases in leisure PA were observed in individuals of working age. Although the pandemic led to a general decrease in alcohol consumption, individuals aged 40 or 45 at baseline were more likely than other respondents to initiate a risky pattern of alcohol consumption. In conclusion, societal restrictions during a pandemic pose a risk for decreased PA, especially among individuals in working age. However, it can also open a window of opportunity, not least for women, to increase their PA levels. In addition, in future pandemics targeted interventions to prevent vulnerable groups from developing a risky alcohol consumption should be considered.

COVID-19 Coronavirus disease 2019

CES-D Center for Epidemiologic Studies Depression Scale

CI Confidence interval

LSH Life conditions, stress and health study

OR Odds ratio

PA Physical activity

Sars-Cov-2 Severe acute respiratory syndrome coronavirus 2

SES Socioeconomic status

SOC Sense of coherence

WHO the World Health Organization

Ethics approval and consent to participate

Ethical approval for the baseline study has been achieved from the regional ethical review board in Linköping (Dnr 02-324), as well as for the COVID questionnaire (Dnr 2020-07230). The study is conducted in accordance with the guidelines of the Declaration of Helsinki. Informed consent was obtained from all individual participants before enrolment in the study.

Consent for publication

Availability of data and materials

The datasets generated and analyzed during the current study are not publicly available due to privacy reasons but are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was funded by the Swedish Research Council (2004-1881), the Swedish Heart and Lung Foundation (20040530), and the county council of Östergötland. Author Fredrik Granström also had grants from Forte (2022-00842).

Authors' contributions

MW, KF, HFL and ML were involved in planning the data collection. FG, MW, KF and IR planned the conception of the study and the study design. FG analyzed and interpreted the data. FG was the major contributor in writing the manuscript. All authors evaluated and discussed the results. All authors read and approved the final manuscript.

Hossain MM, Sultana A, Purohit N. Mental health outcomes of quarantine and isolation for infection prevention: a systematic umbrella review of the global evidence. Epidemiol Health. 2020;42:e2020038.
Rapisarda V, Loreto C, De Angelis L, Simoncelli G, Lombardo C, Resina R, et al. Home Working and Physical Activity during SARS-CoV-2 Pandemic: A Longitudinal Cohort Study. Int J Environ Res Public Health. 2021;18:13021.
Chandrasekaran B, Ganesan TB. Sedentarism and chronic disease risk in COVID 19 lockdown - a scoping review. Scott Med J. 2021;66:3–10.
Kuendig H, Kuntsche E. Solitary versus social drinking: an experimental study on effects of social exposures on in situ alcohol consumption. Alcohol Clin Exp Res. 2012;36:732–8.
Keyes KM, Hatzenbuehler ML, Hasin DS. Stressful life experiences, alcohol consumption, and alcohol use disorders: the epidemiologic evidence for four main types of stressors. Psychopharmacology. 2011;218:1–17.
Gonçalves PD, Moura HF, do Amaral RA, Castaldelli-Maia JM, Malbergier A. Alcohol Use and COVID-19: Can we Predict the Impact of the Pandemic on Alcohol Use Based on the Previous Crises in the 21st Century? A Brief Review. Front Psychiatry. 2020;11:581113.
Booth FW, Laye MJ, Lees SJ, Rector RS, Thyfault JP. Reduced physical activity and risk of chronic disease: the biology behind the consequences. Eur J Appl Physiol. 2008;102:381–90.
Degenhardt L, Charlson F, Ferrari A, Santomauro D, Erskine H, Mantilla-Herrara A, et al. The global burden of disease attributable to alcohol and drug use in 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Psychiatry. 2018;5:987–1012.
Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380:2095–128.
Griswold MG, Fullman N, Hawley C, Arian N, Zimsen SR, Tymeson HD, et al. Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2018;392:1015–35.
Rovira P, Rehm J. Estimation of cancers caused by light to moderate alcohol consumption in the European Union. Eur J Public Health. 2021;31:591–6.
Nindenshuti PM, Caire-Juvera G. Changes in Diet, Physical Activity, Alcohol Consumption, and Tobacco Use in Adults During the COVID-19 Pandemic: A Systematic Review. Inquiry. 2023;60:469580231175780.
Ammar A, Brach M, Trabelsi K, Chtourou H, Boukhris O, Masmoudi L, et al. Effects of COVID-19 Home Confinement on Eating Behaviour and Physical Activity: Results of the ECLB-COVID19 International Online Survey. Nutrients. 2020;12:1583.
Wunsch K, Kienberger K, Niessner C. Changes in Physical Activity Patterns Due to the Covid-19 Pandemic: A Systematic Review and Meta-Analysis. Int J Environ Res Public Health. 2022;19:2250.
Folkhälsomyndigheten. Folkhälsan i Sverige – Årsrapport 2023; 2023. Available from: https://www.folkhalsomyndigheten.se/contentassets/a448b27d603c44f590fc1aff741b0d5d/folkhalsan-sverige-arsrapport-2023.pdf. Accessed 28 Aug 2023.
Braithwaite VS, Sharp SJ, Koulman A, Wareham NJ, Rennie KL. An investigation of factors affecting changes in health behaviours during the COVID-19 pandemic in a UK population-based cohort study. Public Health. 2022;212:46–54.
Castañeda-Babarro A, Arbillaga-Etxarri A, Gutiérrez-Santamaría B, Coca A. Physical Activity Change during COVID-19 Confinement. Int J Environ Res Public Health. 2020;17:6878.
Rodríguez-Larrad A, Mañas A, Labayen I, González-Gross M, Espin A, Aznar S, et al. Impact of COVID-19 Confinement on Physical Activity and Sedentary Behaviour in Spanish University Students: Role of Gender. Int J Environ Res Public Health. 2021;18:369.
Zaworski K, Kubińska Z, Dziewulska A, Walasek O. Physical Activity of Poles in the Care for Their Health Potential Before and During the COVID-19 Pandemic. Disaster Med Public Health Prep. 2022;16:487–90.
Aragón-Vela J, Delgado-Floody P, Guzmán-Guzmán IP, Salas-Sánchez J, Martínez-Redondo M, Lucena Zurita M, et al. Effect of COVID-19 confinement on physical activity patterns in relation to sociodemographic parameters in Spanish population. J Sports Med Phys Fitness. 2022;62:830–7.
Mitchell JJ, Dicken SJ, Kale D, Herbec A, Beard E, Shahab L. Longitudinal changes and correlates of meeting WHO recommended levels of physical activity in the UK during the COVID-19 pandemic: Findings from the HEBECO study. PLoS ONE. 2022;17:e0273530.
Sher C, Wu C. Who Stays Physically Active during COVID-19? Inequality and Exercise Patterns in the United States. Socius. 2021;7:2378023120987710.
de Boer WIJ, Mierau JO, Schoemaker J, Viluma L, Koning RH. The impact of the Covid-19 crisis on socioeconomic differences in physical activity behavior: Evidence from the Lifelines COVID-19 cohort study. Prev Med. 2021;153:106823.
Fearnbach SN, Flanagan EW, Höchsmann C, Beyl RA, Altazan AD, Martin CK, et al. Factors Protecting against a Decline in Physical Activity during the COVID-19 Pandemic. Med Sci Sports Exerc. 2021;53:1391–9.
Jääskeläinen T, Sääksjärvi K, Pietilä A, Männistö S, Kaartinen NE, Lundqvist A, et al. Poor health status before the COVID-19 pandemic is associated with unfavourable changes in health-related lifestyle. Scand J Public Health. 2023;51:792–6.
Roberts A, Rogers J, Mason R, Siriwardena AN, Hogue T, Whitley GA, et al. Alcohol and other substance use during the COVID-19 pandemic: A systematic review. Drug Alcohol Depend. 2021;229:109150.
Manthey J, Kilian C, Carr S, Bartak M, Bloomfield K, Braddick F, et al. Use of alcohol, tobacco, cannabis, and other substances during the first wave of the SARS-CoV-2 pandemic in Europe: a survey on 36,000 European substance users. Subst Abuse Treat Prev Policy. 2021;16:36.
Kilian C, O'Donnell A, Potapova N, López-Pelayo H, Schulte B, Miquel L, et al. Changes in alcohol use during the COVID-19 pandemic in Europe: A meta-analysis of observational studies. Drug Alcohol Rev. 2022;41:918–31.
Rossow I, Bye EK, Moan IS, Kilian C, Bramness JG. Changes in Alcohol Consumption during the COVID-19 Pandemic-Small Change in Total Consumption, but Increase in Proportion of Heavy Drinkers. Int J Environ Res Public Health. 2021;18:4231.
Ferrante G, Camussi E, Piccinelli C, Senore C, Armaroli P, Ortale A, et al. Did social isolation during the SARS-CoV-2 epidemic have an impact on the lifestyles of citizens? Epidemiol Prev. 2020;44(5–6 Suppl 2):353–62.
Villette PA, Lyonnard O, Trehu C, Barais M, Le Goff D, Le Floch B, et al. Changes in Alcohol Consumption after 1 Year of the COVID-19 Pandemic: A Cross-Sectional Study in a Region of France. Int J Environ Res Public Health. 2022;19:15049.
Levy I, Cohen-Louck K, Bonny-Noach H. Gender, employment, and continuous pandemic as predictors of alcohol and drug consumption during the COVID-19. Drug Alcohol Depend. 2021;228:109029.
Bramness JG, Bye EK, Moan IS, Rossow I. Alcohol Use during the COVID-19 Pandemic: Self-Reported Changes and Motives for Change. Eur Addict Res. 2021;27:257–62.
Neill E, Meyer D, Toh WL, van Rheenen TE, Phillipou A, Tan EJ, et al. Alcohol use in Australia during the early days of the COVID-19 pandemic: Initial results from the COLLATE project. Psychiatry Clin Neurosci. 2020;74:542–9.
Varin M, Hill MacEachern K, Hussain N, Baker MM. Measuring self-reported change in alcohol and cannabis consumption during the second wave of the COVID-19 pandemic in Canada. Health Promot Chronic Dis Prev Can. 2021;41:325–30.
Kyaw Hla Z, Ramalho R, Teunissen L, Cuykx I, Decorte P, Pabian S, et al. Socioeconomic and Environmental Factors Associated With Increased Alcohol Purchase and Consumption in 38 Countries During the Covid-19 Pandemic. Front Psychiatry. 2021;12:802037.
Smith ML, Herbert A, Hughes A, Northstone K, Howe LD. Socioeconomic position and adverse childhood experiences as risk factors for health-related behaviour change and employment adversity during the COVID-19 pandemic: insights from a prospective cohort study in the UK. BMC Public Health. 2022;22:1820.
Meyer T, Richter S, Raspe H. Agreement between pre-post measures of change and transition ratings as well as then-tests. BMC Med Res Methodol. 2013;13:52.
Elvén M, Kerstis B, Stier J, Hellström C, von Heideken Wågert P, Dahlen M, et al. Changes in Physical Activity and Sedentary Behavior before and during the COVID-19 Pandemic: A Swedish Population Study. Int J Environ Res Public Health. 2022;19:2558.
Eek F, Larsson C, Wisén A, Ekvall Hansson E. Self-Perceived Changes in Physical Activity and the Relation to Life Satisfaction and Rated Physical Capacity in Swedish Adults during the COVID-19 Pandemic-A Cross Sectional Study. Int J Environ Res Public Health. 2021;18:671.
Brattlöf F, Gustafsson PE, San Sebastián M. Prevalence and change in social inequalities in physical activity before and during the COVID-19 pandemic in Sweden. Int J Equity Health. 2023;22:21.
Bann D, Villadsen A, Maddock J, Hughes A, Ploubidis GB, Silverwood R, et al. Changes in the behavioural determinants of health during the COVID-19 pandemic: gender, socioeconomic and ethnic inequalities in five British cohort studies. J Epidemiol Community Health. 2021;75:1136–42.
Elisabeth AL, Karlen SB, Magkos F. The Effect of COVID-19-related Lockdowns on Diet and Physical Activity in Older Adults: A Systematic Review. Aging Dis. 2021;12:1935–47.
Oliveira MR, Sudati IP, Konzen VM, de Campos AC, Wibelinger LM, Correa C, et al. Covid-19 and the impact on the physical activity level of elderly people: A systematic review. Exp Gerontol. 2022;159:111675.
Steffen J, Schlichtiger J, Huber BC, Brunner S. Altered alcohol consumption during COVID-19 pandemic lockdown. Nutr J. 2021;20:44.
Koopmann A, Georgiadou E, Reinhard I, Müller A, Lemenager T, Kiefer F, et al. The Effects of the Lockdown during the COVID-19 Pandemic on Alcohol and Tobacco Consumption Behavior in Germany. Eur Addict Res. 2021;27:242–56.
Guignard R, Andler R, Quatremère G, Pasquereau A, du Roscoät E, Arwidson P, et al. Changes in smoking and alcohol consumption during COVID-19-related lockdown: a cross-sectional study in France. Eur J Public Health. 2021;31:1076–83.
Radoš Krnel S, Roškar M, Hovnik Keršmanc M, Rehberger M, Levičnik G, Hočevar Grom A. Changes in Alcohol Consumption among Different Population Groups during the SARS-CoV-2 Pandemic: Outcomes of the Slovenian Cross-Sectional National Survey (SI-PANDA). Int J Environ Res Public Health. 2022;19:13576.
Schmits E, Glowacz F. Changes in Alcohol Use During the COVID-19 Pandemic: Impact of the Lockdown Conditions and Mental Health Factors. Int J Ment Health Addict. 2022;20:1147–58.
Mancuso CA, Charlson ME. Does recollection error threaten the validity of cross-sectional studies of effectiveness? Med Care. 1995;33(4 Suppl):As77–88.
Kamper SJ, Ostelo RW, Knol DL, Maher CG, de Vet HC, Hancock MJ. Global Perceived Effect scales provided reliable assessments of health transition in people with musculoskeletal disorders, but ratings are strongly influenced by current status. J Clin Epidemiol. 2010;63:760–6.
Schmitt J, Di Fabio RP. The validity of prospective and retrospective global change criterion measures. Arch Phys Med Rehabil. 2005;86:2270–6.
Baker DW, Hays RD, Brook RH. Understanding changes in health status. Is the floor phenomenon merely the last step of the staircase? Med Care. 1997;35:1–15.
Sampson L, Ettman CK, Abdalla SM, Colyer E, Dukes K, Lane KJ, et al. Financial hardship and health risk behavior during COVID-19 in a large US national sample of women. SSM Popul Health. 2021;13:100734.
Alpers SE, Skogen JC, Mæland S, Pallesen S, Rabben ÅK, Lunde LH, et al. Alcohol Consumption during a Pandemic Lockdown Period and Change in Alcohol Consumption Related to Worries and Pandemic Measures. Int J Environ Res Public Health. 2021;18:1220.

No competing interests reported.

Additionalfile1.pdf
Additional file 1.pdf. Results of multinomial logistic regression. Longitudinal change in leisure PA by population groups.
Additionalfile2.pdf
Additional file 2.pdf. Results of multinomial logistic regression. Measured change in alcohol consumption by population groups.
Additionalfile3.pdf
Additional file 3.pdf. Results of multinomial logistic regression. Reported change in PA by population groups.
Additionalfile4.pdf
Additional file 4.pdf. Results of multinomial logistic regression. Reported change in alcohol consumption by population groups.
Additionalfile5.pdf
Additional file 5.pdf. Baseline levels of PA by SES

Download PDF

Reviews received at journal
16 Dec, 2023
Reviewers agreed at journal
07 Dec, 2023
Reviewers agreed at journal
10 Nov, 2023
Reviewers invited by journal
07 Nov, 2023
Editor assigned by journal
07 Nov, 2023
Editor invited by journal
30 Oct, 2023
Submission checks completed at journal
30 Oct, 2023
First submitted to journal
24 Oct, 2023

You are reading this latest preprint version

Impact of the pandemic on leisure physical activity and alcohol consumption – who were affected and who were not? Results from a Swedish longitudinal cohort study using two different methods to measure change

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

BACKGROUND

AIM

MATERIALS AND METHODS

RESULTS

Leisure PA

Alcohol consumption

DISCUSSION

CONCLUSION

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1