“borderline significant”

While building the multivariable predictive model for SD, we found that diabetes was borderline significant with a negative association ( p = 0.05), even though it was not significant in univariable comparisons ( p = 0.94).

Partly to counterbalance the possible slight liberalism of borderline significant Friedman and/or Nemenyi results, a Bonferroni correction was applied to all tests (whose original significance threshold was set at P < 0.05), leading to a Cochran-Friedman-Nemenyi experiment-wide significance threshold of P < 0.016.

One study reported that IPSS-R outperformed IPSS in predicting OS [ 44 ], with a C-index of 0.74 ( p < 0.001), and showed borderline significant superiority over WPSS ( p = 0.05), although no significant differences were found for PFS prediction.

Age was a borderline significant predictor ( p = 0.05).

A borderline significant difference was observed between subgroups based on stimulant type ( p = 0.05).

Lastly, thrombocytopenia risk was non-significant for cabazitaxel versus other taxanes (RR = 1.47, 95% CI 0.76–2.83; I 2 = 0%) but borderline significant for C25 versus C20 (RR = 1.70, 95% CI 1.00–2.89; p = 0.05) (Fig. 9 ).

For Logical memory, effects of anxiety were only borderline significant ( p = 0.05) but followed the same trend as for the other cognitive tasks, with between-person effects being most important.

After adjusting for these three variables, the odds of live birth omission were 2.0 times greater for birth histories than for pregnancy histories and this was borderline significant ( p = 0.050).

An unexpected, but important finding was a borderline significant ( p = 0.05) excess mortality (all-cause death 2.0% in prolonged DAPT vs. 1.5% in placebo-treated patients) due to more non-cardiovascular deaths [ 25 , 26 ].

When two groups were defined a posteriori based on the insulin values of LDIGIT (120–150 min) , there was a borderline significant group by time interaction for PAI-1 ( p = 0.05; n = 49) and the change in PAI-1 was greater in the insulin-resistant group (177, 61 for HMD vs. 128, 52 ng/mL for HWD; p < 0.001) than in the insulin-sensitive group (137, 80 for HMD vs. 113, 39 ng/mL for HWD; p = 0.02).

Despite a borderline significant reduction in Gls2 relative abundance ( p < 0.05)—which may be attributed to a slight elevation in one Δ pan2 Δ pan3 replicate—the markedly lower abundance of the fused proteins in Δ pan2 Δ pan3 versus WT is consistent with expected regulation by the Pan complex (Figure 7G–I ).

The treatment group effect remained borderline significant ( p = 0.050).

When assessing potential differences in the site of onset between clusters, we found that in KCL BrainBank, there was a borderline significant difference in the proportion of people with limb-onset SALS assigned to the clusters (X 2 = 6.05, p-value = 0.05).

Although the between-group difference in the Lombard effect measured by the sound pressure level was borderline significant [ t (22) = 2.05, p = 0.05], it became significant [ t (20) = 3.19, p = 0.01] after the removal of the outliers (one ASD participant and one NT participant whose Lombard effect measured by sound pressure level was larger than the cross-subject average Lombard effect measured by sound pressure level plus two standard deviations in each group).

There was a small and borderline significant main effect of Direction, F (1,35) = 4.3, p = .05, r 2 = .004, with difficulty ratings in INC blocks generally lower than in DEC blocks.

For agility, the effect size was borderline significant (SMD = 0.65, 95% CI [0.00, 1.30], p = 0.05), suggesting only a possible positive trend.

Post hoc analysis was thus undertaken in those with limiting angina (n = 1507; 13.8%) 47 which reported ivabradine to be associated with a borderline significant 24% reduction in the primary endpoint (hazard ratio, 0.76 [95% CI, 0.58–1.00]; P = 0.05) and a 42% reduction in hospitalization for MI (hazard ratio, 0.58 [95% CI, 0.37–0.92]).

History of HTN had a borderline significant effect on the occurrence of MACE (HR: 1.295, 95% CI: 0.999, 1.677, p = 0.05).

For the MTRR A66G polymorphism, we detected a borderline significant interaction between this polymorphism and overweight/obesity on serum HDL-C levels ( P = 0.050, Table 3 ).

In contrast, SUST scores showed a borderline significant difference in Cluster B ( H = 5.99, p = 0.05) and a significant difference in Cluster C ( H = 6.28, p = 0.043), where Indian grocerants achieved higher compliance levels compared with Ghanaian and Polish counterparts.

Higher dietary OBS was associated with a lower prevalence of stroke ( p for trend = 0.014), although there was a borderline significant negative correlation between dietary OBS at Q4 and stroke (OR 0.568, p = 0.05).

When adjusted for age and the number of comorbidities, the continuous cHIS score showed a borderline significant association with ICU mortality (OR 1.41, 95% CI 1.00–2.09, P = 0.050).

Individuals randomized to receive azathioprine did not see a reduction in the need for lid surgery (OR = 0.60; 95% CI .12, 3.04; P = .54); however, a borderline significant result was observed for radiotherapy (OR = 0.17; 95% CI .03, 1.03; P = .05) with the 95% CI crossing equality (Supplementary Table 3) ( 27 ).

Our results identify the severity of renal dysfunction as a risk factor for severe infectious events by bivariate analysis and it remains borderline significant by multivariate analysis (P = 0.05).

Accordingly, the ferritin/age ratio was higher in dual‐diagnosis patients compared to CDA II patients, with a borderline significant P‐value (P = 0.05, Mann–Whitney test) (Table 2 ).