RT Journal
A1 Rose DN
T1 BEnefits of isoniazid preventive therapy
JF Archives of Internal Medicine
JO Archives of Internal Medicine
YR 1996
FD March 25
VO 156
IS 6
SP 684
OP 684
DO 10.1001/archinte.1996.00440060114015
UL http://dx.doi.org/10.1001/archinte.1996.00440060114015
AB Sterling and colleagues' studied isoniazid preventive therapy in areas with high isoniazid resistance and concluded that the net benefit of isoniazid is minimal and therefore not worthy of the effort. Their chosen outcome measure, life expectancy extension, however, is calculated incorrectly and furthermore is difficult to understand. These factors minimize the benefits of isoniazid and therefore bias the analysis against choosing preventive therapy.How did the authors calculate life expectancy if they analyzed only 20 years of follow-up? Perhaps the authors mean the average amount of time alive over 20 years. My colleagues and I have found that limiting the observation time to 20 years nearly eliminates the advantages of tuberculosis prevention interventions.2 Longer observation times, by contrast, increase the benefits of prevention programs. To illustrate this, I constructed a simple Markov model of two states: alive and dead. Using vital statistics for the annual probability of dying, I