The study by Sonntag et al. (1) is noteworthy as one of the first to provide differential estimates of lifetime costs of overweight and obesity for different levels of socioeconomic status (SES). It also provides a new statistical microsimulation approach, referred to as the Differential Costs (DC)-Obesity model, which can be used to generate estimates of lifetime costs of obesity and overweight, in addition to being employed by public and health policymakers to identify and provide targeted intervention to selected SES groups. The authors have incorporated multiple improvements in the microsimulation from previous studies: for example, the time-to-event modeling to estimate the parameters used in the Markov-based models to simulate and predict outcomes, accounting for competing risks and the weight history within their approach, which is an important factor often not accounted for in studies evaluating the role of BMI with long-term outcomes (2, 3). Study findings highlight that people with obesity and low SES are estimated to incur two times higher costs compared with people with obesity and high SES. In contrast, results for overweight suggest that people with high SES have the highest cumulative excess costs, while those with the lowest SES have the lowest costs. The cost difference between SES groups was of larger magnitude within the obesity category compared with the overweight category. The cumulative overweight category cost increases were moderate compared with those of the obesity category. We agree with the authors that these findings inform the need to prioritize resources and target interventions to specific BMI and SES categories. Despite the authors’ advanced approach, more work needs to be done to ensure that findings from simulations can be used to inform policy. The DC-Obesity model parameters, which are based on a longitudinal data set from a German population, need to be cross- and externally validated, similar to what has been done in the Future Elderly model (4-6); predictive value of the DC-Obesity model must be evaluated and these trends must be confirmed in other countries with different races and ethnicities. In addition, the study uses a multidimensional SES indicator that incorporates aggregated information about education, income, and occupational class. We suggest reporting estimates for each of these components individually in order to determine the unique contribution to the increases in excess costs. Subsequent studies may focus on further understanding the role of weight history and estimating the duration limits of obesity and overweight states when potential exists to reverse excess costs due to obesity and overweight. More specific information about costs of illnesses needs to be provided to understand which are driving excess expenditures. Finally, while the study did attempt to account for unintentional weight loss, a challenging task when dealing with observational data, further sensitivity analyses by incorporating different statistical approaches could shed light on the robustness of these findings (7, 8). We commend Sonntag and colleagues for introducing the DC-Obesity model, a simulation tool with several improvements mentioned earlier. With further refinement and validation, the DC-Obesity model has the potential to be particularly useful in countries where policy planners are dealing with dramatic cuts in health care programs and have an urgency to reduce health care spending.