Depression Symptoms and Antiretroviral Adherence in HIVPositive Clinic Patients

1
Depression Symptoms and Antiretroviral Adherence
in HIV-Positive Clinic Patients

• Mark V. Bradley, M.D.
• Research Fellow, HIV Center for Clinical and
  Behavioral Studies, New York State Psychiatric
  Institute and Columbia University
• HIV Center for Clinical and Behavioral Studies
• Grand Rounds June 26, 2008

2
Background HIV and antiretroviral adherence

• The effectiveness of antiretroviral regimens
  depends upon high levels of patient adherence.
• Treatment failure is predicted by poor adherence
• High levels of adherence are required to ensure
  virologic suppression and prevent resistant
  strains (varies by regimen class type).
• Most studies show that 40-60 of patients are
  less than 90 adherent

3
Determinants of Adherence

• Structural
• Housing
• Access to care
• Financial resources
• Transportation
• Medication Regimen Characteristics
• Complexity/Pill burden
• Side effects
• Individual-level factors
• Education and health literacy
• Physical symptoms
• Use of avoidant coping strategies
• Health beliefs
• Psychiatric symptoms/disorders

4
Psychiatric disorders and ARV adherence

• Substance use disorders
• Intravenous drug use
• Marijuana
• Cocaine use including crack
• Problem alcohol use
• Methamphetamine
• Serious mental illness psychotic illnesses and
  bipolar disorder
• Anxiety disorders including PTSD
• Depressive symptoms / disorders

5
Background Depression and HIV disease

• High prevalence of depressive disorders in HIV
  samples
• Depression predicts poorer medical outcomes in
  HIV (Clinical progression, mortality), even after
  controlling for adherence

6
Depression and adherence

• Wagner et al, J Clin Epidemiol, 2001. 54 Suppl 1
  p. S91-8.
• Palepu et al, substance abuse treatment.
  Addiction, 2004. 99(3) p. 361-8.
• Barfod et al AIDS Patient Care STDS, 2005. 19(5)
  p. 317-25.
• Ammassari A., et al., Psychosomatics, 2004.
  45(5) p. 394-402.
• Arnsten et al, J Gen Intern Med, 2002. 17(5) p.
  377-81.
• Blanco et al, AIDS Res Hum Retroviruses, 2005.
  21(8) p. 683-8.
• Boarts et al, AIDS Behav, 2006.
• Carrieri et al., Int J Behav Med, 2003. 10(1) p.
  1-14.
• Catz et al., Health Psychol, 2000. 19(2) p.
  124-33.
• Gonzalez et al, Health Psychol, 2004. 23(4) p.
  413-8.
• Gordillo, et al Aids, 1999. 13(13) p. 1763-9.
• Murphy et al., Arch Pediatr Adolesc Med, 2005.
  159(8) p. 764-70.
• Holzemer et al., AIDS Patient Care STDS, 1999.
  13(3) p. 185-97.
• Reynolds et al., AIDS Behav, 2004. 8(2) p.
  141-50.
• Tucker et al., Am J Med, 2003. 114(7) p. 573-80.
• Waldrop-Valverde et al, Patient Care STDS, 2005.
  19(5) p. 326-34.

• Depression is a robust predictor of nonadherence
  across a range of studies and methodologies
• Most of these studies have examined depression
  symptoms rather than categorical diagnoses.

7
Depression and adherence

• Cardiac disease and diabetes research has also
  found that depression predicts poor medication
  adherence

• Gehi, A., et al., Depression and medication
  adherence in outpatients with coronary heart
  disease findings from the Heart and Soul Study.
  Arch Intern Med, 2005. 165(21) p. 2508-13.
• Kalsekar, I.D., et al., Depression in patients
  with type 2 diabetes impact on adherence to oral
  hypoglycemic agents. Ann Pharmacother, 2006.
  40(4) p. 605-11.
• Lustman, P.J. and R.E. Clouse, Depression in
  diabetic patients the relationship between mood
  and glycemic control. J Diabetes Complications,
  2005. 19(2) p. 113-22.
• Barth, J., M. Schumacher, and C. Herrmann-Lingen,
  Depression as a risk factor for mortality in
  patients with coronary heart disease a
  meta-analysis. Psychosom Med, 2004. 66(6) p.
  802-13.

8
Background Depression and adherence

• Two studies provide retrospective evidence that
  treatment of depression improves adherence in
  HIV populations (Yun et al, JAIDS 2005 Cook et
  al, AIDS Care 2006)
• Research in other medical illnesses (diabetes,
  cardiocascular disease) have suggested
  prospectively and retrospectively that treating
  depression may improve adherence (Lustman, Arch
  Gen Psychiatry 2006 Katon et al, Arch Intern Med
  2005 )

9
Background Study rationale

• To date, no published prospective research has
  demonstrated that treating depression improves
  adherence in HIV-positive depressed, nonadherent
  medical patients.
• The symptom threshold for adherence problems is
  not known.
• The time from depression response to adherence
  improvement is not known.
• The specific components of depression
  symptomatology responsible for adherence failures
  are not known.

10
Pilot study Methods

• Naturalistic design
• Following depressed, antiretroviral nonadherent
  HIV clinic patients who have recently started or
  optimized treatment for depression
• Monitoring their depressive symptoms and
  antiretroviral adherence as they continue
  antidepressant treatment.

11
Methods Sample

• HIV adult patients
• Referred to study based on history of depression
  and/or nonadherence
• Recent initiation or change in antidepressant
  treatment (medication switch, titration, or
  augmentation) or initiation of psychotherapy
• Followed in one of three HIV medical or mental
  health clinics at Columbia Med Ctr., or the
  Center for Special Studies at Cornell.

12
Methods Eligibility criteria

• Currently on antiretrovirals
• Meet the criteria for Major Depressive Disorder,
  Minor Depressive Disorder, or Dysthymic Disorder
  (SCID)
• Demonstrates lt80 adherence at baseline
• Does not meet criteria for substance use disorder
  in the past month
• Fluent in English
• No h/o bipolar disorder

13
Methods Measures

• Adherence
• Chesneys ACTG Follow-Up Questionnaire for
  Adherence to Antiretroviral Medications
• Visual Analog Scale
• Pill Count
• Viral load
• Depression
• Hamilton Depression Scale
• Depression Module of the SCID

14
Methods Measures/co-variates

• Substance use HIV Center Substance Use
  Questionnaire (potential depression-nonadherence
  mediator)
• Cognition
• Rey Verbal Learning Test
• WAIS Letter-Number Sequence
• Stroop
• Color Trails A and B
• Controlled Oral Word Association
• WAIS Test of Adult Reading

15
Assessment timeline
16
Methods Analytic Plan

• Linear regression models to examining
  associations between changes in adherence scores
  and changes in HAM-D scores, controlling for
  substance use at each time point.
• Generalized estimating equations will be used to
  account for within-subject correlation across the
  three time points.
• In secondary analyses, we will examine
  relationships between specific depression
  symptoms (such as depressed mood, insomnia, and
  anergia) and adherence

17
Recruitment feasibility

• Recruitment procedures commenced in November,
  2007
• Recruitment represented a major challenge to this
  study
• The intersection of specific eligibility criteria
  in several domains resulted in many patients
  being screened out of the study
• Depressive disorder
• lt80 adherent in past 4 days - 1 week
• Recent onset/change in depression treatment
• Fluent in English
• Not actively using substances
• No history of bipolar disorder
• No psychotic symptoms

18
Recruitment feasibility

• Many patients identified and treated for
  depression demonstrate good adherence
• Many patients systematically identified as
  nonadherent by their clinicians also demonstrate
  other exlusionary features, especially active
  substance use and comorbid psychopathology

19
Preliminary findings Sample

• 9 participants recruited to date

20
Preliminary findings Baseline Depression and
Adherence Scores
21
Preliminary Findings

• 4 participants have completed to date.
• These subjects have overall demonstrated some
  evidence of improvement in adherence which
  occurred alongside improvements in depression
  scores
• 2 participants have not followed up after
  baseline due to re-emergent, severe substance use
  problems
• 3 participants remain in the process of data
  collection

22
Preliminary resultsCompleters
23
Preliminary conclusions (1)

• Depressed, nonadherent HIV-positive patients
  demonstrate a degree of psychosocial complexity
  and comorbidity that makes recruitment
  challenging.
• Studies designed to examine this population may
  require a degree of tolerance for this
  complexity, rather than highly restrictive
  eligibility criteria

24
Preliminary conclusions (2)

• When substance use disorders are not an active
  issue, individual cases suggest that treating
  depressive disorders may be one method for
  improving adherence in depressed patients
• Future research will require larger samples and
  longer follow-up periods in order to elucidate
  relationships between depression treatment and
  adherence changes.

25
Acknowledgements

• This study has been funded by the HIV Centers
  Pilot Studies Program and by the Columbia
  Department of Psychiatry Frontier Fund.
• Dr. Bradley is supported by a training grant from
  NIMH (T32 MH19139 Behavioral Sciences Research
  in HIV Infection Principal Investigator, Anke A.
  Ehrhardt Ph.D. Training Director Theo Sandfort,
  Ph.D.).
• The HIV Center for Clinical and Behavioral
  Studies at the New York State Psychiatric
  Institute and Columbia University is supported by
  a grant from NIMH (P30-MH43520 Principal
  Investigator Anke A. Ehrhardt Ph.D.).

26
Acknowledgements

• Mentor
• Robert H. Remien, PhD
• Study Advisors
• Judith G. Rabkin, PhD
• Milton Wainberg, MD
• Cheng-Shiun Leu, PhD
• HIV Center Expertise
• Patricia Warne, PhD
• Katherine Elkington, PhD
• Research Assistant
• Elizabeth Arias, MA

• Harkness-6
• Karen Brudney, MD
• Noga Shalev, MD
• Anne Skomorowsky, MD
• Lucy Ann Wicks Clinic
• Joan Storey, PhD
• Vera Smith, PhD
• Alexandra Bloom, PhD
• Elizabeth Wade, PhD
• Center for Special Studies
• Todd P. Loftus, MD
• Joseph F. Murray, MD