Answers are grounded in this paper and cite their sources.
The range of serum inorganic phosphorus was determined (Fiske and SubbaRow method) on a normal population of 908 persons. This was accomplished by fitting complex curvilinear functions of phosphorus on age separately for each sex. These functions were deemed necessary, because it became apparent that serum phosphorus concentration varied with both age and sex. Ninetynine per cent confidence limits were placed on the normal functions, and the lower limit of this confidence belt was used as a discriminant for the diagnosis of familial hypophosphatemia. Similar functions were fitted separately by sex to a population of 70 affected persons from five families segregating for familial hypophosphatemia. It was found by studying the overlap between the confidence limits of the normal and abnormal populations that the lower 99 per cent confidence limit of the normal function was a superior diagnostic tool. The true operating characteristics of this discriminant were then determined. This required a general mathematical solution of the problem of evaluating a discriminant when the populations used for the comparison contain persons with incorrectly diagnosed disease, and there is also an error associated with the chemical determination.
No takes yet. Share an insight, caveat, or question.
Greenberg et al. (1960) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: