1. Disclosure: Wayne Katon, MD LillyWyethForestPfizer IIII II Company Employment Management Independent Contractor Consulting Speaking & Teaching Board, Panel or Committee Membership

2. Enhancing Treatment for Patients with Comorbid Depression, Diabetes and Heart Disease Wayne Katon, MD 1 Mike VonKorff, ScD 2 Elizabeth Lin, MD, MPH 2 Paul Ciechanowski, MD, MPH 1 Greg Simon, MD, MPH 2 Evette Ludman, PhD 2 Joan Russo, PhD 1 Carolyn Rutter, PhD 2 Bessie Young, MD, MPH 1 1 University of Washington School of Medicine 2 Center for Health Studies, Group Health Cooperative NIMH Grants MH 4-1739 and MH 01643 (Dr. Katon)

3. Mrs. K is a 45-year-old female computer programmer with a 5-year history of type 2 diabetes. She started the study in Sept. 2007 based on the following eligibility criteria: PHQ-9 of 20, HbA 1c 9.6. Patient has a history of childhood sexual abuse, has had recurrent depressive episodes and obesity with a BMI of 51 (>30 meets obesity criteria). Prior history of smoking and has sleep apnea Rxed with CPAP.

4. Adverse Bidirectional Interaction Major Depression Smoking Sedentary lifestyle Obesity Lack of adherence to medical regimens Psychophysiologic  Insulin sensitivity  Autonomic NS  Inflammatory markers Medical illness at earlier age Poor symptom control  functional impairment  complications of medical illness  mortality Katon et al. Biol Psychiatry 2003

5. Premature Mortality and Chronic Mental Illness Schizophrenia: 20-25 years Bipolar: 10-15 years Major Depression: 5 to 10 years

6. Etiology of Premature Mortality Suicide, accidents Medical morbidity

7. Medical Morbidity Chronic stress: effects on HPA axis, autonomic nervous system, immune system Health risk behaviors: smoking, sedentary lifestyle, diet/obesity, alcohol/drugs Lack of self care: adherence to medication, diet, exercise, cessation of smoking Psychiatric medications: obesity, metabolic syndrome, diabetes, CAD

8. Behavioral Risk Factors: Depression Behavioral risk factors (smoking, obesity, sedentary lifestyle) account for approximately 40% of all deaths in the U.S. Depression is linked to all 3 Wassertheil-Smoller (2004) have shown in 98,000 women that depression was associated with higher rates of smoking, lack of exercise, obesity, diabetes, high cholesterol levels and rates of hypertension compared to non-depressed populations

9. Meta-Analysis of the Effect of Depression on Patient Adherence Compared to nondepressed patients, the odds are 3 times greater that depressed patients would be nonadherent with medical treatment recommendations DiMatteo MR et al. Arch Intern Med 2000

10. % Smoking by Depression Level Adjusted for demographics, medical comorbidity, diabetes severity, diabetes type and duration, treatment type, HbA 1c and clinic N = 4,225 p None Katon et al. Diabetes Care 2004

11. % BMI > 30 kg/m 2 by Depression N = 4,225 p None Adjusted for demographics, medical comorbidity, diabetes severity, diabetes type and duration, treatment type, Hb A1c and clinic Katon et al. Diabetes Care 2004

12. HbA 1c > 8% by Depression Level Adjusted for demographics, medical comorbidity, diabetes severity, diabetes type and duration, treatment type and clinic N = 4,225 p None Katon et al. Diabetes Care 2004

13. Medication Adherence in Patients with Diabetes Oral Hypoglycemic Lipid Lowering Meds ACE Inhibitors Nonadherent Days (%) Lin et al. Diabetes Care 2004

14. Pathways Epidemiology Study Baseline Mail Survey 1234 5-Year Telephone Survey Disease control (HbA 1c, LDLs, blood pressure) Pharmacy refills (adherence) ICD-9 diagnosis Macrovascular/microvascular complications (chart review) Mortality (Washington State mortality data)

15. Depression: Association with Complications and Mortality Minor Depression Major Depression Microvascular Complications 1.05 (0.83, 1.33)1.33 (1.08, 1.65) Macrovascular Complications 1.32 (0.99, 1.75)1.38 (1.08, 1.78) Mortality (All cause) 1.23 (0.94, 1.61)1.53 (1.19, 196) Foot Ulcers1.50 (0.82, 2.60)2.30 (1.50, 3.70)

16. Pathways Randomized Controlled Trial Participants randomly assigned to Pathways nurse collaborative care intervention (N = 165) vs. usual care (N = 164) Usual Care Primary care or referral to specialty MH care as available Pathways Care Collaborative/stepped care disease management program for depression in primary care Katon et al. Arch Gen Psych 2004

17. Treatment Protocol Behavioral activation/pleasant events scheduling Antidepressant medication Usually an SSRI or other newer antidepressant OR Problem Solving Treatment in Primary Care (PST-PC) 6-8 individual sessions followed by monthly group maintenance sessions Maintenance and Relapse Prevention Plan For patients in remission Katon et al. Arch Gen Psych 2004

18. Intervention vs Control Differences on Mean SCL Depression Scores (Range 0 – 4) I UC Baseline3 mos6 mos12 mos Mean SCL-20 Depression Score Katon et al. Arch Gen Psych 2004

19. Intervention vs Control Differences on Mean HbA 1c I UC Baseline 6 mos12 mos Mean HbA 1C % Katon et al. Arch Gen Psych 2004

20. Intervention vs. Usual Care Differences in Health Risk Behaviors No significant I vs. UC differences in exercise, diet, smoking or checking blood glucose Intervention patients had a significantly lower mean BMI level compared to UC at 12 months Lin et al. Arch Fam Med 2006

21. Depression: Diabetes Lower Total Health Care Costs Over 2 Years $22,258 $21,148 $18,932 $18,035 $1,110 $897

22. Treating depression and other mental Illness is a necessary first step, but not sufficient alone to improve health risk behaviors and chronic medical disease control

23. Health Services Models TeamCare Approaches have been shown to improve quality of care and outcomes of patients with depression, diabetes, asthma and CHF The most complex and medical costly patients often have multiple comorbidities including at least one mental health diagnosis

24. Medicare Patients Depression, diabetes and heart disease are among the most common illnesses in aging populations but fewer than 4% of Medicare beneficiaries with any of these three illnesses have no other chronic medical conditions 80% of those with CHF, 71% with depression and 56% with diabetes have 4 or more chronic conditions Partnership for Solutions 2001

25. Diabetes: Achieve Recommended Risk Factor Targets Less than 10% of diabetes patients attain recommended goals for: HbA 1c < 7.0%, Systolic BP < 130 and LDL < 100mg Poor Adherence found in 20% of patients No evidence of poor adherence but lack of Rx intensification found in 30% of hyperglycemia patients, 47% of hyperlipidemia patients and 36% of hypertensive patients Schmittdiel J et al. JGIM 23:588-94, 2008

26. Challenge: Development of Health Services Models for “Natural” Clusters of Illness Examples: Diabetes, CAD, depression Depression, chronic pain, substance abuse Definition: Illnesses with high prevalence, high comorbidity and bidirectional adverse interactions

27. New NIMH-Funded Study: TeamCare Inclusion Criteria Evidence via automated date (ICD-9) of having diabetes and/or coronary artery disease (CAD) Evidence of poor disease control (HbA 1c > 8.5, blood pressure >140/90, LDL >130) PHQ-9 > 10

28. 10,000 Group Health patients with diabetes and/or CAD & poor disease control Screen 1: PHQ-2 (response rate 82.6%) 14.8% positive (>3 on PHQ-2) Screen 2: 1066 eligible for SQ-2 with PHQ-9 268 with PHQ-9 >10 completed baseline >200 randomized

29. TeamCare Intervention Goals Improve depression care: behavioral activation and antidepressants Improve medical disease control: HbA 1c, HTN, LDL Improve self-care (diet, exercise, cessation of smoking, glucose checks)

30. TeamCare Interventionists 3 diabetes nurse educators Caseload supervision Depression: 2 psychiatrists Diabetes and CAD: nephrologist, family doctor E-Mail to diabetologist for complex cases

31. Nurse Training Motivational interviewing Problem solving Behavioral activation Antidepressants TREAT-to-TARGET: blood glucose, HTN, LDLS

32. Initially, the case manager increased the patient’s Celexa from 20 to 60 mg and also began working with the patient on monitoring blood sugars more frequently and increasing NPH insulin. Trazadone was also added to help with sleep. Her HbA 1c decreased by December to 8.4%. PHQ score initially decreased from 20 to 12 on Celexa 60 mg. and Trazodone 50 mg and Wellbutrin was added at 100 SR with gradually increasing dosages. By mid- November, her PHQ had decreased to a 5 on Celexa 40 mg, Wellbutrin SR 200 mg BID, Trazodone 50 mg.

33. TeamCare Summary Report Initial ClinicEnroll Date PHQ BL Now BP BL Now HbA 1c BL Now LDL BL Now NSH5/19/0819 141/ 69 127/ 77 7.3 6.8168 138 NSH1/9/08152 118/ 80 130/ 80 9.2 8.3138 124 EVM11/12/07149 160/ 98 150/ 85 6.4 6.8108 67 EVM10/30/07132 209/ 119 126/ 76 7.3 7.7119 103 LYN8/23/07143 149/ 71 111/ 58 8.1 7.785 82

34. Improving Adherence Patient self-care materials: book and video on depression, patient manual (Tools for Managing Your Chronic Disease) Nurse support/education/motivational interviewing Medisets Simplifying medication regimen $4 generics to avoid $10 co-pays

35. Self-Care Enhancements Glucometers: Group Health provides Home blood pressure monitors Pedometers to increase exercise Medisets to improve adherence

36. Phases of Treatment Intervene on depression initially Behavioral activation Antidepressant medication

37. Medical Disease Control Is patient adhering to medication regimen? If adhering and in poor control, is patient on optimal dosage? If maximum dosage has been reached should a new medication be tried instead or augmentation of initial medication? Team recommendations of medication changes are reviewed with primary care physician for approval

38. Behavioral Goals Behavioral activation/exercise Dietary changes Checking blood glucose/altering insulin Cessation of smoking

39. The nurse worked with the patient in January/February 2008 on increasing exercise and weight reduction. Patient also began to gather information about gastric bypass surgery. She began to watch food proportion sizes, worked out on a treadmill and joined a pregastric bypass group. Her PHQ-9 in June was a 7, HbA 1c 7.4%, blood pressure 113/82 (had decreased from 132/80) and LDL was 77 (had decreased from 101)..

41. Conclusions Patients with common psychiatric illnesses have significantly shorter life spans due to premature development of medical illnesses. Economies of scale: New health services interventions are needed for patients with multiple comorbidities (one of which is a psychiatric disorder). Integration of evidence-based mental health interventions into primary care and preventative medical interventions into community mental health care are needed to enhance outcomes of patients with comorbidities.