Historical review: comparison with other methods.

Each technique used to measure compliance has limitations. A simple and commonly used technique is to ask the patient whether the regimen has been followed. When the patient states that the regimen has been followed exactly, it usually has not. If only minor deviation is reported, the deviation is usually understated. However, if the patient states that the regimen was not followed, this is almost always corroborated. In another approach, the attending physician or medical staff estimates the patients' level of compliance. They usually overestimate, but not by as much as the patients. Physicians vary in their ability to accurately estimate their patients' compliance, and some physicians' estimates are better for one type of patient than another. Another measure of compliance that can be easily obtained is faithfulness in keeping clinic or office appointments. Although useful in itself, it is usually not a good indication of how well patients are following their medication regimen. A fourth approach is to determine the number of tablets for which patients obtain prescriptions. This is a poor indicator of medicine intake. Further, the number of pills that the patient actually obtains from the pharmacy is not a good indicator. Probably the best of the simple measures is a pill count. To measure this, the pills remaining in the patient's prescription bottle are counted. The difference between the number remaining and the number given in the prescription is presumed to have been taken. Although the pill count is useful, when this count is compared to blood or urine levels of the medication or of a marker contained in the tablets, some patients show discrepancies. Thus, none of the simple measures is entirely satisfactory.