Publications

Clinical predictive algorithms are increasingly being used to form the basis for optimal treatment policies--that is, to enable interventions to be targeted to the patients who will presumably benefit most. Despite taking advantage of recent advances in supervised machine learning, these algorithms remain, in a sense, blunt instruments--often being developed and deployed without a full accounting of the causal aspects of the prediction problems they are intended to solve. Indeed, in many settings, including among patients at risk of readmission, the riskiest patients may derive less benefit from a preventative intervention compared to those at lower risk. Moreover, targeting an intervention to a population, rather than limiting it to a small group of high-risk patients, may lead to far greater overall utility if the patients with the most modifiable (or preventable) outcomes across the population could be identified. Based on these insights, we introduce a causal machine learning framework that decouples this prediction problem into causal and predictive parts, which clearly delineates the complementary roles of causal inference and prediction in this problem. We estimate treatment effects using causal forests, and characterize treatment effect heterogeneity across levels of predicted risk using these estimates. Furthermore, we show how these effect estimates could be used in concert with the modeled "payoffs" associated with successful prevention of individual readmissions to maximize overall utility. Based on data taken from before and after the implementation of a readmissions prevention intervention at Kaiser Permanente Northern California, our results suggest that nearly four times as many readmissions could be prevented annually with this approach compared to targeting this intervention using predicted risk.

Introduction: Chest physiotherapy (CPT) is commonly used in acutely ill patients, although evidence for its effectiveness is limited and controversial. Electrical Impedance Tomography (EIT) is a recent technique that can offer new information on lung function during ventilation. Hypothesis: This study used EIT (Dr?r Pulmovista 500) to evaluate the effects of CPT in spontaneously breathing hypoxemic patients hospitalized in our Dept of Intensive Care. Methods: We enrolled 51 adult patients in February and March 2012 who required CPT based on clinical status, gas exchange and chest X-ray findings. Patients were evaluated before, immediately after and one hour after a 15 min session of CPT in a semi-recumbent position. Treatments (including FiO2) were not changed during the study period. We studied the evolution in SpO2, variations in regional distribution of ventilation (TV) and changes in end-expiratory lung impedance (?EELI). Secondary outcomes included changes in hemodynamic status and respiratory rate. Results: CPT resulted in significant increases in SpO2 (p < 0.02) and?EELI also increased by 48% (p<0.001) immediately after CPT. There was a positive correlation between?SpO2 (maximal SpO2 reached during CPT – SpO2 before CPT) and?EELI in the dorsal region (r= 0.322, p=0.021). There was also a positive relationship between the PaO2/FiO2 (P/F) ratio at baseline and the?EELI in the dorsal region: patients with a lower P/F ratio had a lower or a negative?EELI. After CPT there was a significant decrease in the dorsal regional distribution of ventilation (p=0.043), together with a trend towards an increase in the ventral regional distribution of ventilation over time (p=0.053). Hemodynamic variables and respiratory rate remained stable over time. Conclusions: These results provide evidence of effectiveness of a 15 min session of CPT in improving gas exchange and lung recruitment (positive?EELI) in spontaneously breathing hypoxemic patients. Improvement in gas exchange was sustained one hour after CPT.