An economic evaluation of hospital-based hemodialysis and home-based peritoneal dialysis for pediatric patients.

The purpose of this study was to assess the relative health system cost of pediatric ambulatory hospital-based hemodialysis and home-based peritoneal dialysis, including both continuous ambulatory peritoneal dialysis and continuous cycling peritoneal dialysis when either treatment is equally appropriate. A cost analysis was performed from the viewpoint of the "study hospital" and service providers (physicians) using treatment protocols, based on current clinical practice, which incorporate procedures to establish dialysis access sites, ongoing dialysis maintenance, and possible complications. Cost estimates used information from the period between April 1, 1993, to March 31, 1994, including fully allocated inpatient and outpatient costs. A sensitivity analysis was conducted to analyze the effect of complications on treatment costs. Total annual costs (in 1994 Canadian dollars, $1.00 CDN approximately $0.75. US) of a typical and uncomplicated continuous ambulatory peritoneal dialysis, continuous cycling peritoneal dialysis, and hemodialysis patient were $47,569, $48,658, and $76,023, respectively. Differences in cost between peritoneal dialysis and hemodialysis patients were due to hemodialysis maintenance costs, which were attributed to larger physician fees (25.8 percent), greater direct treatment costs incurred by the study hospital (14.2 percent), and higher overhead costs (60.0 percent). The expected total cost of hemodialysis complicated by an arteriovenous fistula clot and central venous line blockages, or peritoneal dialysis complicated by hernia repair and peritonitis was $78,568 and $50,438 for hemodialysis and peritoneal dialysis, respectively. For the range of complication probabilities considered, expected total costs were always lower with peritoneal dialysis than with hemodialysis. The cost analysis demonstrates that peritoneal dialysis is less costly than hemodialysis for pediatric patients. Such analyses are but one component of the treatment decision, and as such, should not be viewed as the sole means to yield a treatment decision, but rather as a device for systematically evaluating the alternative treatment options.