Je. Sackoff et al., Trends in prescriptions for highly active antiretroviral therapy in four New York City HIV clinics, J ACQ IMM D, 23(2), 2000, pp. 178-183
Objective: To describe trends in prescriptions for antiretroviral therapies
and factors associated with prescriptions for highly active antiretroviral
therapy (HAART).
Methods: Medical records of patients at four HIV clinics in New York City w
ere reviewed every 6 months. For the four 6-month periods 1997 to 1998, we
identified patients with a CD4(+) nadir <500 cells/mu l; sample sizes were
434, 432, 503, and 643, respectively. Trends in HAART prescriptions were te
sted by logistic regression using robust variance estimates because some pa
tients contributed more than one time period. Associations between HAART pr
escriptions and patient characteristics were tested by chi(2) and multiple
logistic regression analysis.
Results: Patients were predominantly black or Hispanic (89%-90%) and male (
66%-68%), and injection drug use was the most prevalent HIV risk 38%-49%. F
rom 1997 to 1998, HAART prescriptions increased from 54% to 89% of antiretr
oviral prescriptions, and the proportion that included an nonnucleoside rev
erse transcriptase inhibitors (NNRTI) increased from 3% to 10%. HAART presc
riptions were inversely associated with CD4(+) nadir group during all time
periods, and in the second half of 1998, patients with CD4(+) nadir between
50 and 199 cells/mu l were as likely to be prescribed HAART as the most Im
munosuppressed patients (CD4(+) nadir <50 cells/mu l; 91% versus 92%). HAAR
T prescriptions were associated with clinic, HIV risk, and other patient ch
aracteristics in some time periods but not consistently.
Conclusions: In these four HIV clinics, prescriptions for HAART increased s
ignificantly from 1997 to 1998, leveling off at 89% in the second half of 1
998. Increasingly, HAART was prescribed for healthier patients and included
an NNRTI.