Lessons Learned

LESSONS LEARNED

A first lesson is that continuity of exposure to most pharmaceutical products results in a substantially greater effectiveness, relative to what can be achieved in the setting of ‘usual’ or ‘typical’ care. The conven-tional statistical analysis of drug trials – known as ‘intention to treat analysis’ – provides an all-patient average of drug effectiveness, which is diluted by prevalent under-exposure or, in some instances, no exposure at all. This dilutional effect is particularly stark in oral contraceptive trials, where the concep-tion rate is 0.1% per year among women whose use of the oral contraceptive is, to use the CDC’s term, ‘perfect’, and 5% per year – 50-fold higher – among women whom the CDC terms ‘typical’ compliers. The conception rate in women who are seeking to conceive is about 80% per year, with most conceptions occurring within the first three months after the deci-sion to seek to conceive. Presumably what happens in some of the ‘typical’ compliers is that they have brief periods during which, because of dosing lapses, they are running somewhere near the 80% per year rate. It takes only a few conceptions in a contraceptive trial to raise the evident conception rate from its full-compliance value of 0.1% to some intermediate value greater than 0.1% and less than 80%. That interme-diate value, which appears to be about 5%, probably describes no one, as it is too high for those who use the ‘pill’ punctually and far too low to be descriptive of the patients whose usage of the ‘pill’ is so marginal as to allow them to run at or near the physiological conception rate of 80% in non-contracepting, sexually active women.

A second lesson is that unforgiving pharmaceu-ticals can provide full effectiveness only for the 15–20% of patients who are strictly punctual in their remedication.

A third lesson is that implants or depot injec-tions, if properly designed and developed, can provide continuity of drug exposure throughout the inter-val between placement and replacement of the implant, or during the interval between successive depot injections. How much residual drug should be left in the implant at the scheduled time of replacement depends on how much forgiveness one should design into the implant and its replacement regimen. Analogous considerations apply to depot injections.

A fourth lesson is that ‘professionally administered’ medicines, for example replacement of a long-term implant or administration of a depot injection, is basi-cally a form of DOT.

A fifth lesson is that DOT is labor-intensive, the costs of which should be included in any compari-son of the costs of case-management by other modes and the reckoning of overall cost, including the cost of treatment failures, plus the costs created by events among patients who drop out of treatment before the recommended time. Some aspects, for exam-ple the prospect of preventing vs. not preventing community-wide spread of MDR TB bacilli, cannot be effectively costed.

A sixth lesson is the need to have reliable, quan-titative pharmionic data so that it is clear what role under-usage of prescription drugs plays in failed ther-apy, thus also clearly distinguishing failures of phar-macological origin from failures of pharmionic origin.

A seventh lesson is that it appears to be possible for certain patterns of on–off–on dosing to create hazardous rebound effects or recurrent first-dose effects.

An eighth lesson is the crucial role that erratic dosing appears to play in the emergence of drug resistance in the treatment of infectious and parasitic diseases. On a worldwide basis, this lesson is proba-bly the most important of all because of the leading role that infectious diseases play in morbidity and mortality, measured on a worldwide basis, instead of just in the developed countries, where infectious diseases, though hardly eliminated, have nevertheless been greatly curtailed, and in some cases virtually or completely eliminated, for example syphilis, acute rheumatic fever as a sequel to streptococcal infec-tions, trachoma, malaria and others. Prevalent under-use creates conditions that nullify the effectiveness of anti-infective or anti-parasitic drugs and open the door to emergent drug-resistant micro-organisms, leaving as the only alternative to unchecked disease the often uncertain odds of drug discovery and successful devel-opment into effective pharmaceutical products that pose acceptable risk. Here the key word is ‘accept-able’, because what is acceptable is conditioned upon therapeutic need and what is already available. Were we to have, for example, only one drug of dwindling effectiveness for the treatment of malaria or tuber-culosis or typhoid, the magnitude of acceptable risk for a new agent would necessarily rise, because the alternative in each case is unchecked lethal disease. It is hardly a welcome outcome, but rather the best of a bad bargain.