Data story · collective restoration

When Sweden vacations together, the country gets happier

The Swedish researcher Terry Hartig tracked how many antidepressants pharmacists dispensed, month by month, for twelve years. His finding is quietly radical: when more Swedes were on vacation at the same time, the whole population was measurably better off — fewer prescriptions went out that month. And the benefit reached people who weren’t even working. Retirees got happier too when the workforce was on holiday.

Jul 8, 20262 sources

At the average above-expected month — about 37,584 extra workers on vacation — SSRI dispensing dropped 1.13% among working-age men and 1.17% among retired women, the largest drop of any group. Small percentages, but spread across millions of people, and repeating every summer.

The more workers on vacation, the fewer prescriptions

Each line is the study’s fitted relationship (ARIMA coefficients, Tables 1–2) between the number of workers on vacation above the expected level and the change in SSRI dispensing that same month. Every line slopes down — including the retirees, in green.

Working men 20–64Working women 20–64Retired men 65+Retired women 65+

The drop at the average month, group by group

The same-month reduction in SSRI daily doses per 1,000 at 37,584 extra vacationers. The two green bars are retirees — people with no job to take a break from.

The raw series it came from

Reconstruction

Dispensing climbed steeply through the 1990s, then plateaued — and every year it dipped in summer and peaked in winter. The vertical marker is the January 1997 co-payment change. This panel redraws Figure 1’s shape (trend, seasonality, co-pay spikes) from the printed plot; it is illustrative, not the paper’s raw values.

Working men 20–64Working women 20–64Retired men 65+Retired women 65+

Data: Hartig, Catalano, Ong & Syme (2013), “Vacation, Collective Restoration, and Mental Health in a Population”, Society and Mental Health 3(3):221–236. Dose-response and effect sizes computed from the paper’s published coefficients (Tables 1–2) and stated percentages; the series panel is a reconstruction of Figure 1. Antidepressant data originally from Sweden’s national pharmacy corporation; vacationing counts from Statistics Sweden.