Type-1 diabetes, an autoimmune disease characterized by damage to pancreatic insulin-producing beta cells, is associated with adverse renal, retinal, cardiovascular, and cognitive outcomes, possibly including dementia. Moreover, the protozoal parasite Toxoplasma gondii has been associated with type-1 diabetes. To better characterize the association between type-1 diabetes and Toxoplasma gondii infection, we conducted a systematic review and meta-analysis of published studies that evaluated the relationship between type-1 diabetes and Toxoplasma gondii infection. A random-effects model based on nine primary studies (total number of participants = 2655) that met our inclusion criteria demonstrated a pooled odds ratio of 2.45 (95% confidence interval, 0.91–6.61). Removing one outlying study increased the pooled odds ratio to 3.38 (95% confidence interval, 2.09–5.48). These findings suggest that Toxoplasma gondii infection might be positively associated with type-1 diabetes, although more research is needed to better characterize this association. Additional research is required to determine whether changes in immune function due to type-1 diabetes increase the risk of infection with Toxoplasma gondii, infection with Toxoplasma gondii increases the risk of type-1 diabetes, or both processes occur.
A two-tailed test comparing the means of two independent populations is perhaps the most commonly used hypothesis test in quantitative research, featured centrally in medical research, A/B testing, and throughout the sciences. When data are skewed, the standard two-tailed t test is not appropriate and the permutation test comparing the two means (or medians) has been a widely recommended alternative, with statistical authors and statistical software packages touting the permutation test’s utility, particularly for small samples. In this presentation, we illustrate that when the two samples are skewed and the sample sizes are unequal, the two-tailed permutation test (as traditionally implemented) can in some cases have power equal to zero, even when the k highest values in the combined data are all found in the group with k observations. Further, in many cases the standard permutation test exhibits decreasing power as the total sample size increases! We illustrate the causes of these perverse properties via both simulation and real-world examples, and we recommend approaches for ameliorating or avoiding these potential problems.
A metabolic disease resulting in elevated blood glucose levels, type-2 diabetes affects approximately 462 million people globally. Although its prevalence appears to be increasing, type-2 diabetes has been associated with various potentially preventable risk factors, including infectious diseases. The protozoal infection with Toxoplasma gondii (Nicolle et Manceaux, 1908) has been associated with type-2 diabetes in two previous meta-analyses. Since the publication of the last meta-analysis supporting an association between type-2 diabetes and T. gondii, several new primary studies have investigated this association. In this meta-analysis, we sought to further characterise the association between type-2 diabetes and T. gondii. We identified primary studies using PubMed, Embase, Scopus and Web of Science. Twenty-five studies met our inclusion criteria for a total of 4,639 patients with type-2 diabetes and 3,492 controls. Eighteen primary studies found a positive association between type-2 diabetes and T. gondii, whereas seven did not. Using a frequentist random-effects meta-analysis model, we found an overall summary odds ratio of 2.77 (95-percent confidence interval: 2.03-3.76), suggesting that the odds people will have type-2 diabetes is 2.7 times higher for people seropositive for T. gondii. Future studies should investigate this association in additional geographical regions and explore whether this association is due to the immunosuppressive effects of type-2 diabetes or whether T. gondii directly or indirectly affects glucose metabolism, or both.