Self-reported frequency of handwashing among pet and non-pet owners in different situations: results of four surveys of the general adult population in Germany.

Background: Zoonotic diseases are partly associated with pets. However, data is sparse on pet owners' compliance with preventive recommendations. Also, research focuses on self-reports, which are subject to overestimation biases, i.e., assessing one's actual performance to be better than it is. One reason is task difficulty: people tend to overestimate their performance on hard tasks. Regarding handwashing, compliance after touching animals should be harder for pet vs. non-pet owners due to the number of opportunities. This study tests for differences in self-reported handwashing between pet and non-pet owners, and explores reasons for non-compliance. Thus, it aims to provide insights on how to improve self-report behavioral assessment methods in public health and One Health research.

Methods: Data from cross-sectional computer-assisted telephone surveys of the general population in Germany aged 16-85 years in 2012, 2014, 2017 and 2019 were analyzed (N = 15,559; response rate: 45.9%). Handwashing frequency was operationalized for nine indications using the item "How often do you wash your hands in each of the following situations: 'never/almost never', 'seldom', 'mostly', 'always/almost always'?", with the latter defining compliance. In 2017 and 2019, those reporting to 'never/almost never' or 'rarely' wash hands were questioned regarding possible reasons. Chi²-tests, Cohen's d's and multiple logistic regressions were used.

Results: Pet and non-pet owners differed in self-reported handwashing compliance primarily in the indication "After touching animals" (35.5% vs. 55.7%, effect size: d = 0.45). For other indications (e.g., "After using the toilet"), differences were insignificant (≤|3.6%|, d ≤ 0.11). Additionally, 79% of pet owners who rarely or almost never washed their hands after touching animals felt it is not necessary (non-pet owners: 67.1%; d = 0.34). Reporting to not have an appropriate washing facility available was rarer among pet owners (44.5% vs. 63%, d = 0.41). Differences regarding other reasons were trivial (d ≤ 0.16), including "It takes too long" (16.9 vs. 13.3%; p = .138 in multiple regression).

Conclusions: Study limitations include that due to unknown true compliance, over- and underestimations have to be inferred. Yet, that the only substantial difference between pet and non-pet owners pertained to "After touching animals" suggests such effects. While pet owners obviously adjust for task difficulty, the likely residual overestimation should be reduced by measures using script-based covert recall or survey items with response categories constructed to better resemble subjective compliance ratios.