Variables predictive of mortality for T12 model
| Beta-coefficient from lasso | |
| Insurance | |
| Private insurance only | Reference group |
| Medicare only | 0 |
| Combination of any insurance including Medicaid | 0.051 |
| Oxygen therapy | |
| No | Reference group |
| Yes, continuously | 0.459 |
| Yes, nocturnal and/or with exertion | 0 |
| Yes, during exacerbation | 0 |
| Yes, prn | 0 |
| Unknown | 0 |
| Depression | |
| No | Reference group |
| Yes | 0.006 |
| # of pulmonary exacerbations in the year preceding advanced stage CF diagnosis | 0.016 |
| Interim change in FEV1 (T12–T0) | |
| Improvement >10% (+10%) | Reference group |
| Between −10% and 10% | 0 |
| Deterioration >10% (−10%) | 0.445 |
| Interim number of pulmonary exacerbations (T0–T12) | 0.131 |
| Lung transplant in the past 12 months | |
| No | Reference group |
| Yes | −0.494* |
Parameter estimates do not represent the true magnitude of effect. In other words, unlike in Cox regression models that use classical techniques of variable selection, exponentiation of the lasso beta-coefficient is not an estimate of the true HR. However, when taken in context of the model it can determine relative importance in prediction and positive or negative association with the predicted outcome.
*A negative value indicates predictive of lower mortality.
CF, cystic fibrosis; FEV1, forced expiratory volume in 1 s.