Multivariate linear regression models evaluating the impact of a history of obstructive sleep apnea on predicting direct cost while controlling for a number of perioperative variables.
| β (SE) | 95% CI | P Value | |
| Total Direct Cost | |||
| Model 1 | 1125.66 (1720.14) | −2249.18 to 4500.50 | 0.5130 |
| Model 2 | −1155.54 (1602.16) | −4298.93 to 1987.85 | 0.4709 |
| Model 3 | −297.31 (1431.57) | −3106.04 to 2511.42 | 0.8355 |
| Model 4 | 632.78 (781.29) | −900.12 to 2165.68 | 0.4182 |
Abbreviation: SE, standard error.
Note: Multiple models were constructed to evaluating a variety of clinical factors. Model 1 had only sleep apnea as an independent variable to predict cost as the dependent variable. Model 2 included model 1 plus age, sex, American Society of Anesthesiologists physical status classification system status, Elixhauser Comorbidity Index, number of segments, and obesity. Model 3 included model 2 plus length of surgery, total volume of crystalloids, total volume of colloids, total volume of red blood cells, total volume of platelets, total volume of fresh frozen plasma, and total volume of cryoprecipitate. Model 4 included model 3 plus length of hospital stay and length of intensive care unit stay.