Table 4.

Agreement between CPET-VO_2peak and Questionnaire-VO_2peak for changes (∆) from T₀ to T₁

Variable	n	ICC	95%-CI	SEE	AUC	95%-CI	Cut-off value	Sens (%)	Spec (%)	PPV (%)	NPV (%)
∆CPET-VO2peak	58	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a
∆ FitMáx-VO2peak	51	0.43	0.18‒0.63*	2.07	0.77	0.63–0.91	1.0	71	75	81	63
∆FitMáx-VO2peak without cycling	56	0.27	0.00‒0.49*	2.23	0.72	0.59–0.86	1.8	61	78	80	58
∆VSAQ-VO2peak	58	0.19	-0.06‒0.42	2.25	0.66	0.52–0.80	3.4◊	62	58	68	52
∆DASI-VO2peak	45	0.18	-0.10‒0.44	2.40	0.64	0.48–0.81	2.7	62	63	70	55

Number of subjects per variable (n) and ICC with corresponding 95%-CI are reported for the relation between ΔCPET-VO_2peak and Δquestionnaire-VO_2peak from T₀ to T₁. * Statistically significant ^◊The cut-off value for VSAQ-VO_2peak is also the smallest improvement (which is ~ equal to 1.0 MET) that could be measured with VSAQ [12].

Abbreviations: AUC, area under the curve; CPET, cardiopulmonary exercise test; DASI, duke activity status index; ICC, intraclass correlation; n, number of subjects; n/a, not applicable; NPV, negative predictive value; PPV, positive predictive value; SEE, standard error of the estimate; Spec, specificity; Sens, sensitivity; VO_2peak, peak oxygen uptake; VSAQ, veterans specific activity questionnaire; 95%-CI, 95% confidence interval