Integration of Depression and Hypertension Treatment: A Pilot, Randomized Controlled Trial

To me, this investigation can be looked at in 2 ways: either as the successful integration of mental health care into primary care, or the successful integrative care of 2 chronic conditions within primary care. Both are important. In either event this investigation is a great step towards illustrating a primary care practice in which a) depression and anxiety are recognized as common and important comorbidities in patients with chronic illnesses; b) care of multiple chronic conditions requires attention to all of them, as well as to psychosocial needs and patient priorities; and c) appropriate care can and should address more than one condition at a time with resulting improved outcomes for both (or even all). Of course we all know this, but it is hard to escape from the disease-specific models we have been trained in.

Based on my own interest in the care of patients with multiple chronic conditions, I was interested in the two pieces of the intervention and how they might translate to the care of multimorbid patients: provision of an individualized program, and integration of depression and hypertension treatment through emphasizing improved adherence. After listening to patients with multiple morbidities explain their need for a personal primary care contact in a recent qualitative investigation,(1) I am becoming increasingly convinced that there is a role for providing patients individualized (non-MD) contact as part of the ongoing management of multiple conditions within primary care.

The challenge of course is figuring out which patients need such contact at which times. Therefore, one next step beyond integrated treatment for two specific diseases is to develop the practice methods to systematically assess and reassess priorities in persons with chronic illness (including screening for depression) so that they can be offered appropriate integrative care as illustrated by this pilot study.

1. Processes of care desired by elderly patients with multimorbidities. Bayliss, EA; Edwards, AE; Steiner, JF; Main, DS. Family Practice 2008; doi: 10.1093/fampra/cmn040

Competing interests:   None declared

Bogner and de Vries’ paper is both well-done and thought provoking.1 They have put teeth into Engel’s idea that a biopsychosocial model was needed to better understand and manage disease states.2 At a large primary care practice, the authors identified a cohort of older adults suffering both depression and hypertension. Cohort members were randomly assigned to integrated care v. usual care. Integrated care used a midlevel provider as a liaison between the physician and patient. Guided by a conceptual framework, the midlevel provider encouraged adherence to anti-hypertensive and anti-depression treatments. Over 6 weeks, it was determined that the integrated care patients experienced superior outcomes (both hypertension and depression) when contrasted to usual care patients (differences with statistical and clinical meaning).

So what’s the big deal, you might ask? Why the reference to the Holy Grail? Do findings from one study really provide access to a source of profound power? Is there something mystical in these results? In brief, the study does illustrate that incredible power may lie in efforts to jointly address both psychosocial and biological aspects of illness. We’ll elaborate this point shortly. As for the mystical, we don’t think it necessary to conjure up magic to understand these results. Instead, we believe these results are best understood as part of a broader primary care knowledge base that supports practicing integrative medicine as illustrated by this study.

We will first address the issue of power. Primary care physicians mainly seek the power to reduce common causes of morbidity and mortality.3 As such, our quest for the “Holy Grail” involves developing the knowledge base and skill set necessary to effectively reduce common causes for morbidity and mortality. Primary care physicians refer to our quest with terms such as evidence based medicine and patient oriented evidence that matters (or POEMS).3 These terms are useful to organize our scholarly and clinical efforts towards the common cause of how to most effectively reduce morbidity and mortality.

So, does addressing our patient’s mental health problems have something to do with reducing morbidity and mortality? Bogner and de Vries findings, while preliminary, seem to suggest so.1 Put another way, many primary care patients struggle with depression and/or anxiety (often chronically).4 There is a substantial body of literature to suggest that chronic depression and anxiety states are associated with factors directly related to patient morbidity and morality (e.g., chronic health issues such as hypertension, diabetes, smoking, alcohol abuse, sedentary lifestyle, obesity, non-adherence with medical treatment).5 Addressing mental health within primary care offers the opportunity to reduce factors that adversely impact both morbidity and mortality (and therein lies the power that primary care doctors value most).

As for the mystical, we would argue that things appear mystical only when one doesn’t have a better understanding of the phenomenon under consideration. In this instance, there is a broad and growing primary care literature to support the idea that mental health and physical illness are interdependent. We’ll offer select examples to illustrate the point. Several studies document that SSRIs (Fluoxetine or Sertraline) effectively treat depression and diabetes in patients with both conditions (e.g., may lower HbA1C up to 2.5%)6 Likewise, treating with anti- depressants (Fluoxetine or Nortriptyline) post-stroke, reduced depression but also reduced long-term mortality regardless of depression status.7 Treatment of post-MI depression appears to reduce depression symptoms, but does not seem to improve morbidity and mortality. However, the depression -CAD link remains an active area of research inquiry.8

In closing, we appreciate the study by Bogner and de Vries. In addition to being technically well-done, it stimulated us to think. We’ll end by summarizing our main points:

1. Power in primary care means the ability to reliably reduce morbidity and mortality; 2. Chronic depression and anxiety (often found in primary care patients) are consistently associated with factors (e.g., chronic disease, smoking, alcohol abuse, etc.) that adversely impact morbidity and mortality; 3. The magic in primary care lies in understanding that by addressing chronic mental health issues, we may change factors that adversely impact morbidity and mortality (and thus improve the patient outcomes that we care most about).

References

1-Bogner, HR. Integration of depression and hypertension treatment: A pilot, randomized control trial. Annals of Family Medicine 2008; 6: 295-301. 2-Engel, G. The need for a new medical model: A challenge for biomedicine. Science 1977; 196: 129-136. 3-Rakel, RE. Textbook of Family Medicine (7th ed.). Philadelphia: Saunders; 2007. 4-Freedy, JR, Magruder, KM, Zoller, JS, Hueston, WJ, & Carek, PJ. Traumatic events and mental health in civilian primary care: Implications for training and practice. Manuscript submitted for publication; 2008. 5-Goodnick, PJ. Use of antidepressants in treatment of comorbid diabetes mellitus and depression as well as in diabetic neuropathy. Annals of Clinical Psychiatry 2001; 13: 31-41. 6- Jorge, RE, Robinson, RG, Arndt, S, and Starkstein, S. Mortality and poststroke depression: A placebo-controlled trial of antidepressants. American Journal of Psychiatry 2003; 160: 1823-1829. 7-von Kanel, R., & Begre, S. Depression after myocardial infarction: Unraveling the mystery of poor cardiovascular prognosis and role of beta- blocker therapy (editorial comment). Journal of the American College of Cardiology 2006; 48: 2215-2217.

Competing interests:   None declared

We appreciate the interest in our work. Below, we have tried to answer the questions raised that we could not address in the paper due to space limitations. Our goal was to pilot test an integrated approach to improving treatment for depression and hypertension that might facilitate its deployment in real world practices with limited resources and competing demands.

The participants’ adherence was monitored for 2 weeks. At the end of the two-week run-in period, participants were randomized. Therefore, the baseline medication adherence was determined from the two-week run-in period. The research coordinator did perform the intervention and collect the outcome data. This absence of blinding of the assessors could potentially bias the assessment. However, we believe this was minimal because the major outcomes of the study were based on objective data that were unlikely to be influenced by raters’ bias. Medication Event Monitoring System (MEMS) caps to measure adherence and the BPTru to measure blood pressure are both automated devices. The Center for Epidemiologic Studies Depression Scale (CES-D) employs standard questions to measure depression. No participants had any problems using the MEMS caps. MEMS caps have a low failure rate and no malfunction of the MEMS caps arose during the pilot study.

Adherence to antidepressants at baseline for participants who were nonadherent (less than 80% adherent) ranged from 0% to 70% with a mean of 28% and a standard deviation of 23%. Adherence to antihypertensives at baseline for participants who were nonadherent (less than 80% adherent) ranged from 0% to 77% with a mean of 31% and a standard deviation of 25%. The mean change in adherence to antidepressants for patients in intervention was 27%. The mean change in adherence to antidepressants for patients in usual care was -7%. The mean change in adherence to antihypertensives for patients in intervention was 25%. The mean change in adherence to antihypertensives for patients in usual care was 0%. We did not have any participants with surprising outcomes, such as markedly worse adherence.

We assessed attendance at the primary care clinic. We found no statistically significant change in attendance at the primary care clinic between intervention condition and usual care participants.

We agree that a good choice for the control group for the follow-up would be enhanced usual care which equalizes the total time spent with the patient between treatment and control conditions. Enhanced usual care will tell us if the integrated package rather than the education alone is the crucial element. A usual care group would be a pragmatic comparison showing the difference between the integrated intervention and no intervention, but not why our intervention works.

Thank you again for your interest.

Hillary R. Bogner MD MSCE Heather F. de Vries MSPH Department of Family Medicine and Community Health Hospital of the University of Pennsylvania Philadelphia, PA

Competing interests:   None declared