Skip to main content
. 2022 Jun 15;14:1137–1148. doi: 10.2147/NSS.S369316

Table 3.

Factors Associated with CPAP Titration Success During the Therapeutic Night. CPAP Acceptance Using Univariable and Multivariable Logistic Regression Analyses

Variables in the Equation p-value OR [95% C.I.]
Univariable Model
Age (Year) 0.24 1.04 [0.97–1.12]
Body mass index (kg/m2) 0.012 1.334 [1.07–1.67]
PaCO2 (mmHg) 0.25 1.21 [0.87–1.68]
HCO3 (mmol/L) 0.14 1.39 [0.90–2.15]
FVC (% predicted) 0.99 1.00 [0.95–1.05]
FEV1 (% predicted) 0.91 0.99 [0.95–1.047]
Apnea hypopnea index (events/hr) 0.049 1.14 [1.001–1.30]
Desaturation index (desaturations/hr) 0.034 1.22 [1.02–1.46]
Time with SpO2 <90% (mins) 0.24 1.754 [0.69–4.43]
Lowest recorded SpO2 (%) 0.05 0.74 [0.55–1.003]
Mean nocturnal SpO2 (%) 0.10 0.71 [0.474–1.07]
Sex (Female) 0.23 3 [0.50–17.95]
Diabetes mellitus 0.54 1.667 [0.32–8.59]
Hypertension 0.08 4.72 [0.86–26.04]
Multivariable logistic model (Forward Wald method)*
Desaturation index (desaturations/hr) 0.027 1.33 [1.033–1.712]

Notes: *Multicollinearity: none. Overall accuracy: 80.6%, Sensitivity: 91.7%, Specificity: 42.9%, Area under the curve (ROC): 87.2%, Omnibus tests of model: p=0.002, Hosmer-Lemeshow goodness of fit: p=0.609, Nagelkerke R square: 42.2%.