Getting A Handle On Why Good Drugs Sometimes Don’t Work¹ is a clearly written, if sometimes dense, explanation of the difficulties in and means of improving the effectiveness of medications and the unique importance of compliance in this paradigm.

The article’s acknowledged theoretical groundwork consists of Deming’s quality principles, especially the insight that high quality requires low variance.

This notion is more specifically expressed in the Peck-Harter Model of variability in response to medication. In their work with theophylline, Peck and Harter documented the wide variability of response (from none to maximal) to recommended doses of the medication given to average-sized patients with mild asthma.²

These results demonstrated that the total variability of a drug’s effect, in congruence with the classical quality control axiom, is equal to the square root of the sum of the squares of the variability of each of the factors that affect the final result:

1. The drug’s formulation, release kinetics, & absorption into the bloodstream
2. Patient compliance
3. How the medication is metabolized and eliminated once it enters the body
4. How the medication affects the body

It becomes mathematically clear that improvements to any one of these four aspects can have only a relatively small effect on the result. The key to improvement then is to ultimately reduce the variability of each of these components. Urquhart works through the arithmetic of sample cases to further elucidate this point.

Deming suggests ranking sources of variation by the size of their impact on total variability and by the degree to which each source can pragmatically be ameliorated. Urquhart lists compliance as the most significant source of variation, followed in descending order, by pharmacokinetics and pharmacodynamics.

The conclusive point is practicality. Urquhart writes, “… the improvements that can be made through pharmionics [Urquhart's term for the quantitative assessment of how patients take prescribed medications] and medication management, and those that can be made through application of genetic knowledge, seem about equally likely, but it is probably less costly first to assure that drug intake is adequate before pursuing other possibilities that entail expensive tests.”

Footnotes

1. J Urquhart, J R Coll Physicians Edinb 2004; 34:95-98 [back]
2. Harter JG, Peck CC. Chronobiology: suggestions for integrating it into drug development. Ann NY Acad Sci 1991; 618:563-71 [back]

• Asking About Adherence
• Does Mistrust Of A Pharmaceutical Manufacturer Cause Patient Noncompliance?
• Learning From Patient Noncompliance
• Dispensing Pills In Style
• The How To Use Medication Organizers Introduction and Pillbox Pictorial