In 1987 and 1990 in Massachusetts, surveys were conducted to determine the size, pattern of distribution, and trends in the population of children assisted by medical technology. The authors obtained an unduplicated count of all Massachusetts children from 3 months to 18 years of age who used one or more of the following: tracheostomy, respirator, oxygen, suctioning, gastrostomy, jejunal or nasogastric feedings, ostomies, urethral catheterization, ureteral diversion, intravenous access, or dialysis. By comparing counts obtained from medical and educational sources, the authors were able to perform a capture-recapture analysis to estimate the overall number of children dependent upon these technologies. The number of children identified in our surveys increased from 1,085 in 1987 to 1,540 in 1990. However, the capture-recapture analysis yielded estimates of 2,147 plus or minus 230 for 1987 and 2,237 plus or minus 131 for 1990. This suggests that the population of children dependent upon medical technology was essentially stable during this period, and that the 42 percent increase in the number of children identified in our survey reflected improved sampling techniques. During the 3 years, shifts in the pattern of technology use were noted, however. Use of oxygen and gastrostomy increased, and urostomy use declined. A change in the age distribution of the children was also documented, with a shift in the preponderence of technology use from 12 to 24 months in 1987 to children in the first year of life in 1990. Using the 1990 estimate and the 1990 U.S. census figures, an overall prevalence estimate of 0.16 percent was calculated. Applying this to the U.S.child population yields an estimate of 101,800 children assisted by medical technology nationwide(assuming comparable technology use in other States). This information will facilitate policy analysis and program planning on regional and national levels for this medically complex group of children.