When to use the Bonferroni correction

PURPOSE: The Bonferroni correction adjusts probability (p) values because of the increased risk of a type I error when making multiple statistical tests. The routine use of this test has been criticised as deleterious to sound statistical judgment, testing the wrong hypothesis, and reducing the chance of a type I error but at the expense of a type II error; yet it remains popular in ophthalmic research. The purpose of this article was to survey the use of the Bonferroni correction in research articles published in three optometric journals, viz. Ophthalmic & Physiological Optics, Optometry & Vision Science, and Clinical & Experimental Optometry, and to provide advice to authors contemplating multiple testing. RECENT FINDINGS: Some authors ignored the problem of multiple testing while others used the method uncritically with no rationale or discussion. A variety of methods of correcting p values were employed, the Bonferroni method being the single most popular. Bonferroni was used in a variety of circumstances, most commonly to correct the experiment-wise error rate when using multiple 't' tests or as a post-hoc procedure to correct the family-wise error rate following analysis of variance (anova). Some studies quoted adjusted p values incorrectly or gave an erroneous rationale. SUMMARY: Whether or not to use the Bonferroni correction depends on the circumstances of the study. It should not be used routinely and should be considered if: (1) a single test of the 'universal null hypothesis' (Ho ) that all tests are not significant is required, (2) it is imperative to avoid a type I error, and (3) a large number of tests are carried out without preplanned hypotheses.

The development of functional in vitro toxicity tests

In vitro toxicity tests which detect evidence of the formation of reactive metabolites have previously relied upon cell death as a toxicity end point. Therefore these tests determine cytotoxicity in terms of quantitative changes in specified cell functions. In the studies involving the CaC0-2 cell model, there was no significant change in the transport of [3H] L-proline by the cell after eo-incubation with either dapsone or cyclophosphamide (50µM) and rat liver microsomal metabolite generating system. The pre incubation of the cells with N-ethylmalemide to inhibit Phase II sulphotransferase activity, prior to the microsomal incubations, resulted in cytotoxcity in all incubation groups. Studies involving the L6 cell model showed that there was no significant effect in the cell signalling pathway producing the second messenger cAMP, after incubation with dapsone or cyclophosphamide (50µM) and the rat microsomal metabolite generating system. There was also no significant affect on the vasopressin stimulated production of the second messenger IP3, after incubation with the hydroxylamine metabolite of dapsone, although there were some morphological changes observed with the cells at the highest concentration of dapsone hydroxylamine (100µM). With the test involving the NG115-401 L-C3 cell model, there was no significant changes in DNA synthesis in terms of [3H] thymidine incorporation, after eo-incubation with either phenytoin or cyclophosphamide (50µM) and the rat microsomal metabolite generating system. In the one compartment erythrocyte studies, there were significant decreases in glutathione with cyclophosphamide (50µM) (0.44 ± 0.04 mM), sulphamethoxazole (50µM) (0.43 ± 0.08mM) and carbamazepine (50µM) (0.47 ± 0.034 mM), when eoincubated with the rat microsomal system, compared to the control (0.52 ± 0.07mM). There was no significant depletion in glutathione when the erythrocytes were eoincubated with phenytoin and the rat microsomal system. In the two compartment erythrocyte studies, there was a significant decrease in the erythrocyte glutathione with cyclophosphamide (50µM) (0.953 ± 0110mM) when co-incubated the rat microsomal system, compared to the control (1.124 ± 0.032mM). Differences were considered statistically significant for p<0.05, using the Student's two tailed 't' test with Bonferroni's correction. There was no significant depletion of glutathione with phenytoin, carbamazepine and sulphamethoxazole when co-incubated with the rat microsomalsystem, compared to the control.

Statistical guidelines for clinical studies of human vision

Citation information: Armstrong RA, Davies LN, Dunne MCM & Gilmartin B. Statistical guidelines for clinical studies of human vision. Ophthalmic Physiol Opt 2011, 31, 123-136. doi: 10.1111/j.1475-1313.2010.00815.x ABSTRACT: Statistical analysis of data can be complex and different statisticians may disagree as to the correct approach leading to conflict between authors, editors, and reviewers. The objective of this article is to provide some statistical advice for contributors to optometric and ophthalmic journals, to provide advice specifically relevant to clinical studies of human vision, and to recommend statistical analyses that could be used in a variety of circumstances. In submitting an article, in which quantitative data are reported, authors should describe clearly the statistical procedures that they have used and to justify each stage of the analysis. This is especially important if more complex or 'non-standard' analyses have been carried out. The article begins with some general comments relating to data analysis concerning sample size and 'power', hypothesis testing, parametric and non-parametric variables, 'bootstrap methods', one and two-tail testing, and the Bonferroni correction. More specific advice is then given with reference to particular statistical procedures that can be used on a variety of types of data. Where relevant, examples of correct statistical practice are given with reference to recently published articles in the optometric and ophthalmic literature.

Regional variability in visual field sensitivity during hypercapnia

PURPOSE: Previous investigations have demonstrated a relative vascular autoregulatory inefficiency of the inferior compared to the superior retina in healthy subjects breathing increased CO2. The purpose of this study was to determine whether the superior and inferior visual field sensitivities of healthy eyes are similarly affected during mild hypercapnia. DESIGN: Experimental study. METHODS: Visual field analysis (Humphrey Field Analyser; SITA standard 24-2 program) was carried out on one randomly selected eye of 22 subjects (mean age, 27.7 ± 5 years) during normal room air breathing and isoxic hypercapnia. The Student paired t-tests were used to compare the visual field indices mean deviation (MD) and pattern standard deviation (PSD) for each breathing condition. A secondary, sectoral analysis of mean pointwise sensitivity was performed for each condition. In each case a P value of <.01 was considered statistically significant (Bonferroni corrected). RESULTS: Visual field MD was -0.23 ± 0.95dB during room air breathing and -0.49 ± 1.04dB during hypercapnia (P = .034). Sectoral pointwise mean sensitivity deteriorated by 0.46dB (P = .006) in the upper visual hemifield during hypercapnia, whereas no significant difference was observed for the lower hemifield (P = .331). CONCLUSIONS: The upper visual hemifield exhibited a significantly greater degree of deterioration in pointwise visual field mean sensitivity compared to the lower hemifield during hypercapnic conditions. This suggests that the upper visual hemifield and hence inferior retina is more susceptible to insult during hypercapnia than the superior retina in healthy individuals. A regional susceptibility of inferior retinal function to altered vascular or metabolic effects may account for the earlier and more frequent inferior nerve fibre damage associated with glaucomatous optic neuropathy. © 2003 by Elsevier Science Inc. All rights reserved.