How well do covariates perform when adjusting for sampling bias in online COVID-19 research? Insights from multiverse analyses

COVID-19 research has relied heavily on convenience-based samples, which-though often necessary-are susceptible to important sampling biases. We begin with a theoretical overview and introduction to the dynamics that underlie sampling bias. We then empirically examine sampling bias in online COVID-19 surveys and evaluate the degree to which common statistical adjustments for demographic covariates successfully attenuate such bias. This registered study analysed responses to identical questions from three convenience and three largely representative samples (total N = 13,731) collected online in Canada within the International COVID-19 Awareness and Responses Evaluation Study ( www.icarestudy.com ). We compared samples on 11 behavioural and psychological outcomes (e.g., adherence to COVID-19 prevention measures, vaccine intentions) across three time points and employed multiverse-style analyses to examine how 512 combinations of demographic covariates (e.g., sex, age, education, income, ethnicity) impacted sampling discrepancies on these outcomes. Significant discrepancies emerged between samples on 73% of outcomes. Participants in the convenience samples held more positive thoughts towards and engaged in more COVID-19 prevention behaviours. Covariates attenuated sampling differences in only 55% of cases and increased differences in 45%. No covariate performed reliably well. Our results suggest that online convenience samples may display more positive dispositions towards COVID-19 prevention behaviours being studied than would samples drawn using more representative means. Adjusting results for demographic covariates frequently increased rather than decreased bias, suggesting that researchers should be cautious when interpreting adjusted findings. Using multiverse-style analyses as extended sensitivity analyses is recommended.

Comparison of pH and motility of the small intestine of healthy subjects and patients with symptomatic constipation using the wireless motility capsule

Gastrointestinal luminal pH shows a rise from the duodenum to the terminal ileum in healthy individuals. Our objectives were to compare the pH in the proximal small intestine (SI) (first 60 min of small intestinal transit) lumen of human volunteers and patients with symptomatic constipation; to quantify contractile pressure profiles of the proximal SI, and to assess the relationship between luminally-recorded contractile pressure and small intestinal transit times (SITT) of a non-disintegrating capsule that measures pH and pressure activity (wireless motility capsule). We used previously acquired records from 39 healthy subjects and 41 patients with symptomatic constipation. Mean pH (±SD) of the proximal SI was similar in healthy subjects and patients with constipation at 6.2 (±0.6) and 6.3 (±0.4), respectively. In 13 of the healthy subjects, pH did not rise uniformly in the proximal SI though the pH_median was 6.0 (5th, 95th percentiles 3.09, 7.06) and the pH fluctuated over a mean period of 28 min. Large inter-individual variability in frequency of pressure activity (Ct) and area under pressure curve (AUC) were observed in the proximal SI of healthy subjects and patients with constipation. Median AUC was 3996 mmHg s^-1 (5th, 95th percentiles 948, 16866 mmHg s^-1) in these two populations combined. Ct and AUC showed a strong direct linear correlation at r = 0.91, p < 1 × 10^-6. An inverse correlation (suggesting longer SITT with lower pressure activity) was observed between Ct/AUC and SITT in both healthy subjects and patients with symptomatic constipation. The pooled results for both groups showed: AUC and SITT correlation at r = -0.49, p < 1 × 10^-6. We concluded that both the frequency and amplitude of contractions in the proximal SI are important for the propagation of non-disintegrating capsules. The observed pH fluctuations in the proximal SI may impact supersaturation and precipitation of weakly basic drugs.