Refer to the weight loss tab. When used together, weight and session both are statistically significant predictors at the 0.10 level. Rerun the regression, excluding "Client" as an x variable. By its p-value (.78 rounded), and the fact it is simply a numbering of clients with no specific value other than order, it obviously does not contribute to the model. When used together, both "weight" and "# sessions" p-values are