JMIR Publications

ABSTRACT

Background:

Psychological stress is known to exacerbate dermatologic conditions such as acne, eczema, and compulsive skin behaviors. The COVID-19 pandemic introduced a global stressor with psychosocial effects on skin health. This study analyzes U.S. public interest in stress-induced dermatologic conditions to assess pandemic-associated shifts in six psychodermatologic search terms.

Objective:

To assess whether the onset and resolution of the COVID-19 pandemic were associated with changes in U.S. Google search interest for psychodermatologic conditions and stress-related skin disorders.

Methods:

Relative search volume (RSV, 0-100 scale) for six terms: "skin picking", "trichotillomania", "stress rash", "eczema flare up", "dermatillomania", and "anxiety skin picking", was obtained from Google Trends from January 2018 through December 2024 (n=84 monthly observations per term). A segmented regression model with breakpoints at the World Health Organization’s pandemic timeline (March 2020, pandemic declaration and May 2023, ending the concern) tested for level and slope changes at pandemic onset and afterward. Newey-West (HAC) standard errors corrected for residual autocorrelation. Joint F-tests assessed whether each period differed from the prior trend with Bonferroni and Benjamini-Hochberg (FDR) correction applied across all six terms. A sensitivity analysis applied the identical model to five synonymic terms: “excoriation disorder”, “compulsive skin picking”, “hair pulling”, “stress eczema”, “itching anxiety”. Each synonymic term was paired with the corresponding primary term. Four exploratory terms “acne”, “stress acne”, “stress hives”, and “hair loss from stress”, were also tested.

Results:

Two of the six terms showed significant pandemic-onset effects after multiple-testing correction: “dermatillomania” (level shift +15.20 RSV points, Bonferroni p=0.007, FDR p=0.007) and “anxiety skin picking” (slope change +0.70/month, Bonferroni p=0.023, FDR p=0.012). “Stress rash”, “skin picking”, and “eczema flare up” reached significance under the FDR correction (p=0.041, p=0.041, and p=0.026) but not under Bonferroni. “Trichotillomania” was not significant (FDR p=0.075). “Anxiety skin picking”, “trichotillomania”, and “stress rash” showed post-pandemic reversals robust to both corrections. A sensitivity analysis using 5 synonymic terms produced weaker and inconsistent support. Only 1 pair (“dermatillomania” and “excoriation disorder”) showed partial agreement under FDR. The other four pairs diverged from their primary match term, indicating the pandemic-onset findings are sensitive to both breakpoint placement and search-term phrasing. A secondary exploratory analysis of four additional terms found no effects in either correction.

Conclusions:

Public search interest in psychodermatologic conditions: “dermatillomania” and “anxiety skin picking”, showed statistically significant shifts during the COVID-19 pandemic. A sensitivity analysis using synonym search terms did not consistently reproduce the primary findings, indicating that these results are sensitive to both analytic choices and search-term phrasing. These findings support additional research into psychodermatology and illustrate how public search behavior reflects societal stress. However, a causal interpretation is unlikely with search-volume data.