Morbidity and Mortality Reduction Associated With Polysomnography Testing in Idiopathic Pulmonary Fibrosis

A Population-Based Cohort Study

Nicholas T. Vozoris; Andrew S. Wilton; Peter C. Austin; Tetyana Kendzerska; Clodagh M. Ryan; Andrea S. Gershon

Disclosures

BMC Pulm Med. 2021;21(185) 

In This Article

Results

Derivation and Description of the Cohort

There were 5044 individuals with IPF identified, aged 66 years and older, of whom 201 (4.0%) received PSG during the accrual period (Figure 2). Out of this group, 189 (94.0%) exposed individuals were matched to an equal number of controls. Before propensity score matching, compared to the control group, the exposed group had a younger mean age and consisted of a greater proportion of men and rural residents, had a smaller percentage of low income individuals, and multiple markers of IPF severity were more prevalent (such as, being admitted to hospital or ICU for respiratory-related reasons in the preceding year, having CHF, having other comorbidities, and systemic corticosteroid and respiratory antibiotic receipt). After propensity score matching, exposed and control individuals were adequately balanced on baseline characteristics, with standardized differences being below 10% for all variables, except five (rural residence, diabetes, kidney disease, antiplatelet/anticoagulant drug receipt, and year of cohort entry), where trivial imbalance remained (standardized differences ranged from 11 to 17%) (Table 1 and Additional file 1).

Figure 2.

Flow diagram outlining exposed and control group identification

Main Analysis

In the propensity score matched cohort, compared to controls, individuals undergoing PSG had significantly reduced rates of respiratory-related hospitalization (HR 0.43, 95% CI 0.24–0.75, p = 0.003, NNT 11) and all-cause mortality (HR 0.49, 95% CI 0.30–0.80, p = 0.004, NNT 10) (Table 2). The cumulative incidence of respiratory-related hospitalization was significantly lower and the probability of survival was significantly higher among individuals undergoing PSG versus controls (Figure 3).

Figure 3.

a Cumulative incidence function curves for respiratory-related hospitalization and b Kaplan–Meier curves for all-cause mortality

Sensitivity Analyses

By Respiratory-related Hospitalization. In the subgroup of individuals that experienced no respiratory-related hospitalization in the year prior to the index date, compared to controls, individuals undergoing PSG had significantly reduced rate of all-cause mortality (HR 0.34, 95% CI 0.23–0.83, p = 0.01), but not respiratory-related hospitalization (Table 3). In the subgroup of individuals with ≥ 1 respiratory-related hospitalization in the year prior to the index date, significantly reduced rate of respiratory-related hospitalization (HR 0.38, 95% CI 0.15–0.99, p = 0.05) was observed among individuals undergoing PSG relative to controls, but there was no significant difference all-cause mortality rate.

By CHF. Compared to controls, there were no significant differences in rates of respiratory-related hospitalization or all-cause mortality among those undergoing PSG, in both the subgroup without CHF complication and in the subgroup with CHF complication (Table 4).

By Systemic Corticosteroid Receipt. In the subgroup of individuals with no systemic corticosteroid receipt in the year prior to the index date, there were no significant differences among individuals having undergone PSG versus controls in respiratory-related hospitalization and all-cause mortality rates (Table 5). In the subgroup of individuals that received a systemic corticosteroid in the year prior to the index date, there was significantly lower rate of respiratory-related hospitalization (HR 0.37, 95% CI 0.14–0.94, p = 0.04), but not all-cause mortality, among those undergoing PSG relative to controls.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....