Pharmacistmanaged adherence support and its impact on viral load suppression and CD4 count

1
Pharmacistmanaged adherence support and its
impact on viral load suppression and CD4 count

• P. Jagannathan1, A. McElfresh2, I.R. McNicholl2,
  C.B. Hare1,2
• 1 Department of Medicine, San Francisco General
  Hospital,
• University of California, San Francisco, USA
• 2 Positive Health Program, San Francisco General
  Hospital,
• University of California, San Francisco, USA

2
Background

• Antiretroviral (ARV) adherence is key to
  virologic success
• A recent meta-analysis found that ARV adherence
  interventions targeting individuals, practical
  medication management skills, and delivered over
  12 weeks were associated with better adherence
  outcomes1
• Two RCTs of pharmacist-led interventions have
  shown improvements in adherence2,3
• Few studies have successfully demonstrated the
  efficacy of ARV adherence interventions on
  improving clinical and virologic outcomes

• Rueda S et al. Cochrane Database Syst Rev 2006,
  3CD001442.
• Rathbun RC et al. Clin Ther 2005, 27(2)199-209.
• Tuldra A et al. J Acquir Immune Defic Syndr 2000,
  25(3)221-228.

3
HIV Medication Adherence Program HIV-MAP

• Setting Large, urban, public health HIV clinic
  in San Francisco, CA, U.S.
• Pharmacist-run, patient-tailored ARV adherence
  clinic
• Initiated January 2007
• Patients referred at providers discretion
• Both treatment-naïve and treatment experienced
• Providers often referred patients at risk for
  poor outcomes

4
HIV-MAP Intervention

• Trained pharmacists conduct standardized
  assessments at initial encounter
• Assess for substance use, health literacy, social
  support, depression
• Individualized adherence support provided
• Tailored to patient based on initial assessment
• Education and counseling about appropriate ARV
  administration
• Adverse-event management strategies
• Adherence supports including pill
  boxes/medi-sets, reminder devices
• Follow-up visits and telephone calls assess
  progress
• Intervention planned for 24 weeks or longer

5
Programmatic Evaluation Methods

• Retrospective analysis of all clinic patients
• Initiating/changing ARV regimens
• Baseline viral load (VL) gt1,000 copies/mL
• Patients lacking 24 week follow-up data excluded
• Intervention group
• HIV-MAP patients enrolled from 1/1/07 to 12/31/07
  
• Control groups
• Contemporary Control Patients seen in clinic
  during the same time period, but not enrolled in
  HIV-MAP
• Historical Control Patients seen between 9/1/05
  and 8/31/06, prior to HIV-MAP

6
Programmatic Analysis Effect of
HIV-MAP

• Primary outcome Proportion of patients achieving
  VL lt75 copies/ml within 24 weeks following ARV
  initiation/change
• Secondary outcomes Mean change in VL and CD4
  count at 24 weeks
• Between groups comparisons
• Baseline continuous variables analyzed by
  Wilcoxon rank-sum categorical variables analyzed
  by Fishers exact test
• Primary outcome analyzed by logistic regression
  secondary continuous outcomes analyzed by linear
  regression. Outcomes adjusted for baseline VL,
  CD4 count, and ARV status

7
Programmatic Analysis Predictors of
lack of virologic suppression within HIV-MAP

• Baseline characteristics of patients in HIV-MAP
  were used to assess for predictors of lack of VL
  suppression at 24 weeks
• Logistic regression used to analyze predictors of
  lack of VL suppression

8
Baseline characteristics
9
Primary and Secondary Outcomes at 24 weeks
Controlled for Baseline VL, Baseline CD4, and
ARV status at enrollment
10
Predictors of lack of viral load suppression at
24 weeks in HIV-MAP patients
11
Limitations

• Small sample-size
• Non-randomized
• Retrospective analysis
• Limited data on self-reported adherence markers
• Missed appointments
• Did not allow for assessment of relationship of
  adherence to virologic and clinical outcomes
• Not available for control groups

12
Conclusions

• Despite having lower baseline CD4 counts and more
  treatment experience, patients receiving the
  HIV-MAP intervention were equally likely as
  contemporary or historical controls to achieve VL
  suppression within 24 weeks they also had a
  greater decrease in VL over 24 weeks than
  historical controls
• Current tobacco use and, to a marginal extent,
  baseline VL were associated with a lack of VL
  suppression in multivariable models

13
Implications

• Pharmacist-managed adherence programs are
  feasible in large, urban clinics
• These programs may help patients who have more
  advanced HIV disease and/or adherence challenges
  achieve virologic outcomes comparable, and
  possibly better, than the general clinic
  population
• Monitoring and evaluation of adherence
  initiatives is vital in improving interventions,
  and can identify new targeted interventions (i.e.
  smoking cessation) that may impact clinical
  outcomes

14
Acknowledgements

• San Francisco General Hospital Positive Health
  Program providers and staff
• Grant Dorsey for statistical assistance
• Patients