What Personalization Can—and Cannot—Promise
This chapter tests the central promise of personalized recurrent kidney-stone prevention: if treatment is selected from a person’s stone composition or urine measurements, do outcomes improve compared with empiric treatment? The question matters because testing can make a prevention plan look more precise, while the evidence may only show that a measurement identifies a possible mechanism. The useful judgment is not whether testing is always good or always unnecessary, but whether the result can change a decision enough to justify its burden and uncertainty.
The attractive logic of selective treatment
The logic behind testing-guided care is straightforward. Recurrent stones arise through different combinations of urine volume, pH, calcium, oxalate, uric acid, citrate, composition, and clinical conditions. If a test identifies a relevant abnormality, treatment can be selected for that abnormality rather than offered to everyone. A person with persistent low urine pH might be considered for a different strategy from someone with persistent hypercalciuria; a person with hyperuricosuria might not be treated as though they had the same problem as someone without it. This reasoning is coherent, but coherence is not the same as proof that the testing pathway improves outcomes.
A testing study has to answer more than whether the test produces information. It has to compare what happens when that information directs care with what would have happened under general or empiric care. The comparison should also make clear what treatment was offered, to whom, for how long, and what outcome counted as success. Otherwise, a lower laboratory value may be mistaken for better prevention even when painful episodes, procedures, or quality of life were not improved.
The AHRQ review found only one randomized trial that explicitly tailored dietary advice using baseline metabolic findings. Kocvara and colleagues studied adults with a first idiopathic calcium stone. The tailored group received advice based on findings such as hypercalciuria, hyperuricosuria, hyperuricemia, hyperoxaluria, or magnesium deficiency; the comparison group received general dietary recommendations. At three years, recurrence was reported in 7 percent of the tailored-advice group and 23 percent of the general-advice group. The calculated risk ratio was 0.30, with a 95 percent confidence interval from 0.14 to 0.65.
That result is encouraging in direction, but the AHRQ review graded the evidence insufficient. It came from one trial with serious limitations and imprecision. The population was adults with a first idiopathic calcium stone, not all recurrent stone formers, and the intervention was dietary advice rather than a complete test-directed medication pathway. The result also does not reveal whether one particular metabolic finding or one particular dietary change accounted for the difference. It shows why tailoring deserves investigation; it does not establish a reliable personalized algorithm.
The pharmacologic comparison was even more observational. A claims-based study compared people who received preventive medication after a prior 24-hour urine test—described as selective treatment—with people who received preventive medication without such testing—described as empiric treatment. The medicines included thiazides, alkali therapy, allopurinol, or combinations.