To assess interpersonal payments in health care financing, often only cross-sectional distributions of individual health care expenditures are available. However, for a proper understanding of solidarity issues, policy makers are often much more interested in lifetime differences because health care expenditures tend to fluctuate strongly each year within a lifetime. Actual lifetime data, however, do not exist, and even will not be available at all in the coming decades. A modeling approach is required to achieve some insight in lifetime health expenditures and their distribution. Using a longitudinal dataset, we propose to use nearest neighbor resampling to extrapolate observed panels into full life cycles. We validate this method and demonstrate how this approach can be used to obtain the distribution of lifetime health care expenditures.