Effect of Receiving Preoperative PET on Chances of Unnecessary Surgery
| Conventional multivariate analysis* | Instrumental variable analysis† | ||
| Odds ratio | Adjusted risk difference | Odds ratio | Adjusted risk difference |
| 0.87 (0.66, 1.16), P = 0.351 | −3.0%(−9.1%, 3.1%), P = 0.335 | 0.53 (0.34, 0.82), P = 0.004 | −10.2% (−20.2%, −0.3%), P = 0.044 |
↵* Additional covariates include age; marital status; race; preexisting comorbidity including chronic obstructive pulmonary disease, congestive heart failure, and chronic renal or liver function; and whether mediastinoscopy was performed.
↵† Same covariates as in conventional multivariate analysis were included in instrumental variable analysis in addition to second-stage residuals.
Data in parentheses are 95% CIs.