Notably, of these variables, being male and parental psychological distress exhibited different directions of odds ratios for treatment initiation in these regional subgroup analyses, and their interactions with region appeared to be statistically significant (0.034 for being male and <0.001 for parental psychological distress).
The direction of the odds ratios (in terms of showing negative or positive association with the dependent variable) remains identical and although the magnitude and/or significance of the odds indicated some variation, our main results remain unchanged.
While not statistically significant, the direction of the odds ratio for frailty in the group of all CHS participants identifying as white was consistent with pilot data for mt204: in a logistic regression model adjusted for age and sex the odds of frailty were 1.75 (95% CI = 0.74 3.61, p = 0.164) times greater among the group with the C allele than those with the T allele.
A closer examination of the data reveals many nonsignificant trends in which the direction of the odds ratios is in the same direction across all five diseases.
Furthermore, the direction of the odds ratios and linear regression estimates was inconsistent for smoking status, exercise, and fruit and vegetable intake.
We combined the conventional criterion for statistical significance with an evaluation of the extent to which the direction of the odds ratios and linear regression estimates were consistent across all the health conditions.
Stressors showing a significant interaction with single motherhood that point in the opposite direction of decreased odds of smoking in single mothers ('Table 5) were not included in the analysis.
The model was rerun without the 33 divergent observations, producing a model with two of the three explanatory factors (whether a diagnostic code was written in the VCR and system diagnosis) remaining statistically significant and with same direction of the odds ratio point estimates (the model retained similar level of the Hosmer-Lemeshow goodness-of-fit test; p = 0.93, 6 df).
Ikegawa and colleagues also call our attention to the coincidence in direction of the odds ratio from the different studies in relation to knee OA, giving its probability as 1/16 = 0.0625, and that this is unlikely to have occurred by chance alone.
This change was explored in additional analyses; Spearman correlations showed that "supportive teacher" was correlated with both "school work enjoyment" and "necessary help" (rho, 0.27, and rho, 0.40, respectively, p-values < 0.001), and we consider the change in the direction of the odds ratio to be a likely artefact that may be explained by co-linearity between variables.
