For cost-containment reasons, French authorities have decided to remove a large set of pharmaceuticals out of the list of reimbursed products. This decision is controversial and the argument is frequently made that dereimbursed products are largely substituted by reimbursed more expensive drugs, so that no savings actually results from that measure. OBJECTIVES: We analyse a preceding case of product dereimbursement to evaluate this argument. METHODS: We retrospectively followed a cohort of patients who received a prescription of a product composed of omega3 fatty acids in the indication of hypertriglyceridemy. This drug is no longer reimbursed since September 1998. We collected prescription data before and after the dereimbursement date. A “substitution list” including all drugs that can be thought to be a substitute was established. Finally, all patients were distributed into three groups, the “continuation group” in which patients were still treated, the “discontinuation group” in which patients discontinued their treatment without receiving any product of the substitution list, and the “substitution group” in which patients discontinued the treatment but received at least one prescription from the substitution list. Data came from a well known and reliable physician’s panel. RESULTS: We received a complete set of data for 1070 patients. The most frequent strategy was the “substitution one” with 41% of patients, 32.5% of patients simply discontinued their treatment and 26.5% continued to take the drug. Statins were the most frequent substitution drugs. Although they were generally more expensive, the monthly average cost per patient was cut by about 20%. The real savings were only 70% of the expected savings because of the substitution strategy. CONCLUSIONS: This dereimbursement decision appeared to be a redistributive game in which the winners were health insurance and the statins manufacturers and the losers the manufacturer of the concerned drug and patients whose out-of-pocket payment increased.