Is a charge a cost if nobody pays it?
CHEST

Approximately one third of patients undergoing coronary artery bypass graft surgery will develop atrial fibrillation (AF). (1) AF is more common in elderly patients and in patients with COPD or hypertension. Its occurrence, and particularly its recurrence, were associated with encephalopathy, strokes, renal dysfunction, infection, in-hospital deaths, more use of CT scans and noncardiac ultrasonography, and longer ICU and hospital stays. (1,2) However, patients with AF were less likely to have myocardial infarctions or congestive heart failure, and they underwent fewer echocardiograms and EEGs. (1,2) Additionally, AF is a risk factor for late mortality. (2)

While a variety of medicines, including amiodatone, have been shown to decrease the occurrence of AF, (3,4) disagreement persists about the cost benefits of prophylaxis with amiodarone. Daoud et al, (5) using the sum of direct variable, fixed direct costs, and indirect costs, found that amiodarone prophylaxis decreased hospital costs by $8,161 per patient. In a mathematical modeling exercise using departmental cost/charge ratios and physician work relative value units, Mahoney et al (6) claimed that the use of IV amiodarone for universal prophylaxis in patients undergoing coronary artery bypass graft surgery would increase in-hospital costs by $24,934 for each episode of AF prevented. Other studies, using a variety of analyses, have found no difference in costs between groups receiving amiodarone and control groups. (7-10) Into this controversy step Kerstein et al (see page 716) with a novel and convenient method for administering amiodarone that decreased the incidence of AF from 26 to 6%. Given the assoeiation between AF and morbidity and mortality, the authors should be encouraged to perform a randomized, double-blinded, placebo-controlled study of amiodarone and AF. The authors also report that universal prophylaxis is very cost-effective, saving $1,242 per treated patient. However, the authors do not determine the cost. They multiplied the average length of hospital stay by a constant charge per day to arrive at a total "cost." There are three main problems with this, as follows: (1) length of stay correlates poorly with direct variable cost after cardiac surgery (11); (2) most of a hospital charge consists of fixed or indirect costs that are not saved by preventing AF; and (3) their calculation does not capture any costs related to the adverse effects of amiodarone. Only if the use of amiodarone increases length of stay is it recognized as a cost. Amiodarone can cause pulmonary infiltrates, thyroid dysfunction, heart block, and ventricular dysrhythmias. In this length of stay-based accounting system, all extra tests and procedures related to the adverse effects of amiodarone have no cost. They are free, which is obviously not correct.

Then how shall we determine the costs or savings of using amiodarone to prevent AF? Cost studies are usually divided into the following four types: cost-minimization; cost-benefit; cost-utility; and cost-effectiveness. Cost-minimization studies compare two or more equally efficacious therapies to determine which is the least expensive. Cost-benefit studies necessitate converting all outcomes (eg, pain, emesis, renal failure, myocardial infarction, and death) to a monetary value. Cost-utility studies determine the cost of a utility metric, such as $10,000, for each quality-adjusted year of survival. Cost-effectiveness studies determine the monetary cost of preventing unwanted outcomes (eg, death, ventilated associated pneumonia, and AF). Recommendations on conducting cost-effectiveness studies have been promulgated by the US Public Health Service (12) and the European Society of Intensive Care Medicine. (13) We would use a cost-effectiveness study to determine the monetary cost for each case of AF prevented. The results can range from negative cost (ie, amiodarone use saves more money than it costs, as found by Kerstein et al), to positive cost (ie, amiodarone use increases costs but at least prevents AF), (6) to infinite cost (ie, amiodarone use costs more and does not prevent AF). Next, we would determine whose perspective is determining cost. The view can be that of the hospital, the insurance company, the patient, or the society. They are not interchangeable. An action that reduces hospital cost, such as early hospital discharge, may increase the cost to the patient or insurance company by, for example, the need to pay for home health care or a stay at an extended-care facility.