Concerns over pharmaceutical costs and appropriateness of medication use have led state Medicaid programs to restrict drug reimbursement. This article critically reviews 20 years of research on cost sharing, drug reimbursement limits, and administrative limitations on access to particular drugs via formularies, category exclusions, or prior authorization requirements; evaluates their methodological rigor; summarizes the state of current knowledge; and proposes future research directions. Drug reimbursement caps and modest cost sharing can reduce the use of both essential and less important drugs in Medicaid populations; severe reimbursement caps may precipitate serious unintended effects. Limitations on access to particular drugs can cause both rational and irrational drug substitution effects; it is unclear whether such limits reduce expenditures either for drugs or for overall health care.

Triathlons (races involving consecutive swimming, bi cycling, and running) have become commonplace in the United States. These races may involve from 30 min utes to 36 hours of continuous exercise, usually in warm or hot environments. Little has been published regarding the medical and physiological aspects of these events. This paper represents the first large study to date on the subject, including both an analysis of medical complications at six triathlons as well as a prospective electrolyte study conducted at two of these races. Medical records were kept and examined for all ath letes requiring treatment during a typical United States Triathlon Series (USTS) race in 1986 (1,000 starters; finish times, 2 to 4 hours), a typical Ironman Qualifier (IQ) race in 1986 (622 starters; finish times, 4 to 8 hours), and the 1982 through 1985 Hawaii lronman World Championships (4,583 starters; finish times, 9 to 17 hours). At the USTS race, fewer than 2% (17/1,000) of the starters required aid, at the IQ, approximately 10% (61 /622) of the starters were treated, and at the Ironman, an average of 17% (794/4,583) received med ical attention. The most common diagnoses at the USTS and IQ were dehydration and heat exhaustion. At the lronman, dehydration and heat problems were complicated by hyponatremia. Because hyponatremia has been reported as a com plication of ultraendurance events, a prospective study was performed on 36 athletes during a USTS race and 64 athletes at the 1984 lronman race. Prerace and postrace blood samples showed that no athletes were hyponatremic following the shorter USTS race, but 27% (17/64) of the athletes studied were hyponatremic fol lowing the lronman race. Medical personnel should be prepared to treat a minimum of 2% and up to 10% of the athletes in races lasting up to 4 hours, 10% to 20% of those in races lasting 4 to 8 hours, and at least 20% of starters in races lasting between 9 and 17 hours. For races less than 4 hours, the IV fluid of choice should be D5 1/2 NSS (normal saline solution). For races longer than 4 hours, D5NSS should be used for IV resuscitation.

Patients most at risk of reduced access to essential medications because of a reimbursement cap include those with multiple chronic illnesses requiring drug therapy, especially illnesses with a mental health component. Physicians in clinics or large group practices may have maintained patient medication regimens more effectively.