Impact of the COVID-19 crisis on work and private life, mental well-being and self-rated health in German and Swiss employees: a cross-sectional online survey

This was not an experiment but an observational survey. The researchers tested whether self-reported changes in work and private life during the COVID-19 crisis (e.g., working from home, short-time work, changes in leisure time, caring duties) were associated with:

**Perceived overall impact** of the crisis on work life and private life (separately rated as "worsened," "stayed the same," or "improved")

**Mental well-being (MWB)** — measured using the WHO-5 Well-Being Index

**Self-rated health (SRH)** — a single-item global health question

They also examined sociodemographic factors (age, gender, living situation, education, income) as potential moderators.

The key comparisons were between people who reported negative, neutral, or positive perceived impacts, and between those who experienced specific work/life changes versus those who did not.

**Sample size:** 2,118 employees

**Population:** German and Swiss workers aged 18–65, working at least 20 hours per week, in various occupations

**Recruitment:** Via an online panel service (respondi AG) that maintains representative panels

**Country breakdown:** 1,059 from Germany, 1,059 from Switzerland

**Gender:** 50.3% female, 49.7% male

**Mean age:** 44.1 years (SD = 12.0)

**Education:** 40% had a university degree; 60% had lower educational levels

**Income:** 30% reported household net income below €2,500/month; 30% above €4,500/month

**Living situation:** 35% lived alone; 65% lived with partner or family

**Work hours:** Mean 37.5 hours/week (SD = 9.5)

**Occupation types:** 27% in health/education/social work, 18% in manufacturing, 14% in trade, 12% in business services, 10% in public administration, 19% other

**Data collection period:** April 22 to May 11, 2020 — during the first wave of COVID-19 lockdowns in Germany and Switzerland

The sample was designed to be broadly representative of the working populations in both countries, though it excluded those working fewer than 20 hours/week (e.g., many part-time workers, students with side jobs).

All measures were self-reported via an online questionnaire. No objective or physiological measurements were taken.

**Perceived impact on work life:** Single item: "How has the COVID-19 crisis affected your work life overall?" Response options: "worsened," "stayed the same," "improved"

**Perceived impact on private life:** Same format: "How has the COVID-19 crisis affected your private life overall?"

**Mental well-being (MWB):** WHO-5 Well-Being Index — a 5-item scale asking about positive mood, vitality, and interest over the past two weeks. Each item scored 0–5, total 0–25, with higher scores indicating better well-being. Scores below 13 indicate poor mental well-being and possible depression.

**Self-rated health (SRH):** Single item: "How would you rate your general health?" on a 5-point scale from "very good" to "very bad." Dichotomized for analysis into "good/very good" vs. "fair/poor/bad."

**Work and life changes:** Self-reported binary (yes/no) items about whether specific changes occurred due to COVID-19: working from home (first time vs. experienced), mandatory short-time work (Kurzarbeit), reduction in work hours, increase in work hours, reduction in leisure time, increase in leisure time, reduction in caring duties, increase in caring duties

**Sociodemographics:** Age, gender, country, education, income, living situation (alone vs. with partner/family), number of children, occupation

**Study design:** Cross-sectional observational survey. This means all data were collected at a single point in time (April–May 2020), with no follow-up, no control group, and no intervention.

**Sampling and recruitment:** Participants were drawn from existing online panels maintained by respondi AG, a market research company. Quota sampling was used to match the German and Swiss working populations on age, gender, and region. Invitations were sent via email, and participants completed the survey online. The response rate was not reported, which is a common limitation of panel-based studies.

**Statistical approach:** Logistic regression models were used to examine associations. For the perceived impact outcomes (worsened vs. not worsened; improved vs. not improved), separate binary logistic regressions were run. For mental well-being (dichotomized as <13 vs. ≥13 on WHO-5) and self-rated health (good/very good vs. fair/poor/bad), logistic regressions included both perceived impact variables and specific work/life changes as predictors, controlling for sociodemographics. Results are reported as odds ratios (OR) with 95% confidence intervals (CI). No corrections for multiple comparisons were mentioned.

**What this design can and cannot prove:**

**Can prove:** Associations and correlations between variables at one point in time. The study can identify which groups of people were more likely to report negative or positive impacts, and which changes were linked to better or worse well-being.

**Cannot prove:** Causality. Because this is cross-sectional, we cannot determine whether the COVID-19 crisis *caused* changes in well-being, or whether people with pre-existing lower well-being were more likely to perceive negative impacts. There is no baseline data from before the pandemic. The direction of causality could go either way, or a third factor (e.g., pre-existing health conditions, personality, job security) could explain both.

**Cannot prove:** Temporal sequence. The survey asks about "changes" retrospectively, but recall bias is a major concern. People who feel worse now may be more likely to remember negative changes.

**Cannot prove:** Generalizability beyond the first wave. The data were collected during a very specific period (April–May 2020) when lockdowns were strict and uncertainty was high. Findings may not apply to later waves, post-vaccine periods, or different countries with different policies.

**Major methodological weaknesses:**

1. **No pre-pandemic baseline** — impossible to know whether reported changes reflect actual change or pre-existing differences

2. **Single-item measures** for perceived impact — no validated scale, unknown reliability

3. **Retrospective self-report** of changes — subject to recall bias and current mood state

4. **No objective measures** of work hours, income, or health

5. **Cross-sectional design** — cannot establish direction of effects

6. **Response rate not reported** — potential for selection bias (people who were more affected may have been more or less likely to respond)

7. **Dichotomization of continuous outcomes** (WHO-5, SRH) reduces statistical power and may obscure meaningful variation

8. **No adjustment for multiple comparisons** — some significant findings may be due to chance

**Primary outcomes (perceived impact):**

**Work life worsened:** 30.1% of employees reported their work life had worsened due to COVID-19

**Work life improved:** 10.3% reported improvement

**Work life unchanged:** 59.6% reported no change

**Private life worsened:** 29.8% reported their private life had worsened

**Private life improved:** 13.0% reported improvement

**Private life unchanged:** 57.2% reported no change

**Associations with perceived negative impact on work life:**

Mandatory short-time work was the strongest predictor: OR = 4.63 (95% CI: 3.53–6.08), p < 0.001 — meaning employees on short-time work were nearly 5 times more likely to report worsened work life

Reduction in work hours: OR = 2.04 (95% CI: 1.52–2.73), p < 0.001

Increase in work hours: OR = 1.57 (95% CI: 1.08–2.28), p = 0.018

Living alone (vs. with partner/family): OR = 1.42 (95% CI: 1.11–1.81), p = 0.005

Younger age (per 10-year increase): OR = 0.85 (95% CI: 0.77–0.94), p = 0.002 — younger employees more likely to report worsening

**Associations with perceived positive impact on work life:**

Working from home for the first time: OR = 2.89 (95% CI: 2.02–4.14), p < 0.001 — nearly 3 times more likely to report improvement

Already experienced with working from home: OR = 1.88 (95% CI: 1.30–2.71), p = 0.001

Increase in leisure time: OR = 1.72 (95% CI: 1.19–2.48), p = 0.004

**Associations with perceived negative impact on private life:**

Reduction in leisure time: OR = 3.32 (95% CI: 2.52–4.37), p < 0.001

Living alone: OR = 2.22 (95% CI: 1.72–2.86), p < 0.001

Younger age (per 10-year increase): OR = 0.74 (95% CI: 0.66–0.82), p < 0.001

Increase in caring duties: OR = 1.90 (95% CI: 1.38–2.62), p < 0.001

Reduction in caring duties: OR = 1.79 (95% CI: 1.09–2.93), p = 0.021

**Associations with perceived positive impact on private life:**

Increase in leisure time: OR = 3.52 (95% CI: 2.60–4.76), p < 0.001

Living with partner/family (vs. alone): OR = 2.24 (95% CI: 1.55–3.24), p < 0.001

Short-time work: OR = 1.69 (95% CI: 1.20–2.38), p = 0.003

Increase in caring duties: OR = 1.65 (95% CI: 1.15–2.37), p = 0.007

**Secondary outcomes (mental well-being and self-rated health):**

**Low mental well-being (WHO-5 < 13):** 33.7% of the sample scored below the cutoff, indicating poor well-being

**Poor/fair self-rated health:** 27.2% rated their health as fair, poor, or bad

**Associations with low mental well-being:**

Perceived negative impact on work life: OR = 1.68 (95% CI: 1.30–2.17), p < 0.001

Perceived negative impact on private life: OR = 2.30 (95% CI: 1.78–2.97), p < 0.001

Mandatory short-time work: OR = 1.56 (95% CI: 1.18–2.06), p = 0.002

Perceived positive impact on private life: OR = 0.53 (95% CI: 0.37–0.76), p < 0.001 — protective effect

Increase in leisure time: OR = 0.57 (95% CI: 0.42–0.78), p < 0.001 — protective effect

**Associations with poor self-rated health:**

Perceived negative impact on work life: OR = 1.57 (95% CI: 1.20–2.05), p = 0.001

Perceived negative impact on private life: OR = 1.67 (95% CI: 1.28–2.18), p < 0.001

Mandatory short-time work: OR = 1.47 (95% CI: 1.10–1.97), p = 0.010

Living alone: OR = 1.32 (95% CI: 1.03–1.69), p = 0.027

No significant associations were found between perceived positive impact on work life and either mental well-being or self-rated health.

**Short-time work** was the single strongest predictor of negative outcomes: employees on short-time work were **4.6 times more likely** to report worsened work life, and **1.6 times more likely** to have low mental well-being. This is a large effect — comparable to the difference between having a chronic illness versus not in many health studies.

**First-time home workers** were **2.9 times more likely** to report improved work life — a substantial positive effect, though it only applied to about 10% of the sample.

**Loss of leisure time** was strongly linked to negative private life impact (OR = 3.3) and **gain in leisure time** was linked to positive private life impact (OR = 3.5) — both large effects, roughly equivalent to the impact of doubling or halving one's weekly free time.

**Living alone** doubled the odds of reporting a negative impact on private life (OR = 2.2) and increased the odds of poor self-rated health by 32% — a moderate but meaningful effect.

**Perceived negative impact on private life** was associated with **2.3 times higher odds** of low mental well-being — a strong association, comparable to the effect of major life stressors like divorce or job loss in other studies.

**Perceived positive impact on private life** was associated with **47% lower odds** of low mental well-being — a substantial protective effect.

In absolute terms: among those who reported their private life had worsened, roughly 50% had low mental well-being, compared to about 25% among those who reported improvement. This is a 25-percentage-point difference — large enough to be clinically meaningful.

**Acknowledged by authors:**

Cross-sectional design prevents causal inference

Self-reported data subject to recall and social desirability bias

Single-item measures for perceived impact (not validated)

Sample limited to German and Swiss employees working ≥20 hours/week — not generalizable to other countries, unemployed, or part-time workers

Data collected during first wave only — may not reflect later phases of the pandemic

No data on pre-pandemic mental health or work conditions

**Additional critical limitations:**

**No objective verification** of work changes (e.g., actual hours worked, income loss) — all self-reported

**Recall bias is severe** — asking people to remember "changes" during a highly stressful period is unreliable; current mood strongly colors retrospective judgments

**No control for personality traits** (e.g., neuroticism, optimism) that could influence both perception of impact and well-being

**No data on COVID-19 infection status** — people who were sick themselves or had sick family members may have had very different experiences

**No data on job type** beyond broad categories — a retail worker and a software engineer both classified as "working from home" had vastly different experiences

**Dichotomization of WHO-5** (cutoff at 13) loses information — a score