Impact of lung function decline on time to hospitalisation events in systemic sclerosis-associated interstitial lung disease (SSc-ILD): a joint model analysis

Table 2 Association between slope of FVC% predicted and risk of first hospitalisation endpoints over 52 weeks

	Time to first all-cause hospitalisation or death (n = 568)	Time to first SSc-related hospitalisation or death (n = 570)	Time to first admission to ER or admission to hospital followed by admission to ICU or death (n = 572)
Longitudinal sub-model^a
Estimated slope difference nintedanib vs. placebo (95% CI)	1.16 (0.00, 2.32)	1.44 (0.33, 2.55)	1.33 (0.18, 2.48)
P value	0.0497	0.01	0.02
Time to event sub-model^b
Number of patients with event, n (%)	78 (13.7)	42 (7.4)	75 (13.1)
Difference in estimated slope of FVC% predicted, HR (95% CI)
1-unit decrease	1.13 (1.07, 1.18)	1.14 (1.07, 1.21)	1.05 (0.98, 1.12)
3-unit decrease	1.43 (1.24, 1.65)	1.48 (1.23, 1.77)	1.15 (0.95, 1.41)
5-unit decrease	1.81 (1.42, 2.30)	1.91 (1.41, 2.60)	1.27 (0.91, 1.76)
P value	< 0.0001	< 0.0001	0.15

Data collected during the treatment period
^aRandom effects normal linear model of FVC% predicted with predictor variables ATA status and FVC% predicted at baseline, a separate slope for patients on treatment, trajectories modelled by a linear trend, and an unstructured variance–covariance matrix
^bPiecewise exponential baseline hazard, stratified by ATA status, and endogenous time-dependent covariate FVC% predicted as estimated slope of the longitudinal response
ATA, anti-topoisomerase antibody; CI, confidence interval; ER, emergency room; FVC, forced vital capacity; HR, hazard ratio; ICU, intensive care unit; SSc, systemic sclerosis

ISSN: 1478-6362