Tablet splitting: much ado about nothing?

Abstract

Pharmacists have been splitting scored tablets to individualize and titrate dosages since the end of the pill-rolling era. We have generally accepted that scored tablets may be evenly divided, resulting in 2 half-tablets containing one-half of the whole tablet strength. Slight powdering is unavoidable, but it is generally accepted that loss of a few molecules of active drug is not likely to be clinically significant. However, even when tablets are split by pharmacists, 1 study found significant weight deviations in almost 10% of half-tablets.1 We do not know if this weight variation correlates with active drug in each tablet half, or more importantly, if this variation will jeopardize clinical outcomes or safety. In some instances, there may be no other way to satisfy the prescribed dose other than tablet splitting. In any case, we rationalize that the patient receives the necessary dose during the course of chronic therapy because the patient eventually consumes all tablet halves as prescribed—1 half-tablet at a time. Assuming a 10% weight deviation, if one half-tablet provides 90%, and the next contributes 110%, on average, the patient receives 100% of the dose over 2 doses. Presumably pharmacists rely on pharmacokinetics and pharmacodynamics when selecting opportunities for tablet splitting and pass on those drugs for which fluctuations may be a concern; one wonders if physicians share such concerns when nonstandard doses are prescribed. Pharmacists must be involved and vigilant when advising individual patients and participating in population-based tablet-splitting programs. We have made assumptions and selected appropriate drugs to be halved based on our understanding of pathophysiology, pharmacology, and pharmacodynamics. We assume homogenous distribution of active drug in whole tablets, and thus, an equal distribution of active drug in the half-tablets. Some variation is expected, and the U.S. Food and Drug Administration (FDA) bioequivalence standards permit variance of plus or minus 20%.2 So, even with whole tablets, the actual dose ingested may fluctuate from whole tablet to whole tablet. Past tablet-splitting research has assessed the half-tablet weight, uniformity, and even the clinical outcome of split tablets,3 including the work of Gee et al. (2002), which assessed clinical, service, and cost outcomes associated with tablet splitting.4 In this issue of JMCP, Hill et al. offer an unprecedented analysis of the accuracy and precision of tablet splitting by measuring the active drug component in tablet halves.5 This use of assay is an important step in assessing half-tablet weight variations.