Hypertension Self-management The Use of Telemedicine as an

Hypertension Self-management: The use of Telemedicine as an Intervention Tool Hayden Bosworth, Ph.D. Center for Health Services Research in Primary Care, Durham VA Medical Center Departments of Medicine, Psychiatry and School of Nursing Duke University Medical Center Outline    Prevalence and Impact of Hypertension Chronic Disease Self-management: Barriers and Facilitators Translation Studies – – Veterans Study to Improve The Control of Hypertension (V-STITCH) Hypertension Intervention Telemedicine Study (HINTS)   Take Home Messages Future Directions Prevalence of Hypertension   1 in 3 adult Americans ~ 65 million Americans (JAMA 2003) – ~45 million prehypertensive   ~ 8 million veterans (37%) (Med Care Res Rev 2003) Lifetime risk for normotensive 55 year old: 90% (JAMA 2003) Hypertension Treatment Facts  Life Style Matters – – – – – Weight Loss (any means) DASH Diet Low Na Diet Exercise Limited Alcohol  Medications Work – Nothing Better Than Thiazide Diuretics – Most Patients Require > 2 medications Goal of Hypertension Self-management Hypertension as a Model for Self-management Complex, long-term, chronic disease  Requires initiation and maintenance of multiple behaviors  Requires provider/patient communication  Significance of Self-management Adherence    >80% of adults took at least 1 medication in the last week; 25% took at least five Cost of medication non-adherence >$100 billion/year ~50% of patients non-adherent with medication   Rates of non-adherence higher in lifestyle recommendations ~50% of treatment failures are due to unrecognized patient non-adherence Traditional Paradigms Fail clinical trial information alone does not result in adequate BP control  Specialist-based care not solution  Primary care clinic based management is not sufficient  – Frequent contact with doctors in clinics does not lead to BP control Disease Management Hypertension: Evidence  Cochrane review (2006) – 59 trials • Reduces SBP (8-10 mm Hg)) • Reduces DBP (4-7 mm Hg) • Improves all cause mortality – Self-monitoring alone (17 trials) • Reduces DBP by 2 mm Hg Health Decision Model Patient Characteristics Perceived Risks Coping & Stress Comorbidities Cognition Policy Provider Characteristics Communication Style Medication Regimen Literacy Intensity of Therapy Side Effects Treatment Guidelines Adherence Medical Environment Depression Mental Health Social Environment TREATMENT ADHERENCE Bosworth HB, Olsen MK, Oddone EZ. (2006). Am Heart J 149:795-803.l Bosworth HB & Oddone EZ. (2002). J Nat Med Ass. 94; 236-248 BLOOD PRESSURE CONTROL Patient Characteristics Related to Self-management 1. 2. Risk Perceptions / Knowledge Cognition • • • Memory Inductive Reasoning Verbal Comprehension 3. 4. Literacy / Numeracy Coping / Stress • • • Avoidance Daily hassle Stigma Patient Characteristics Related to Self-management 5. 6. 7. Comorbidities Medication Side Effects Depression/mental health Social Characteristics Related to Self-management 1. 2. Social Network Social Support • • Tangible/instrumental Emotional 3. Culture Medical Environment Related to Self-management Access and Barriers 1. Insurance (i.e., co-payments, deductibles) 2. Transportation 3. Organization and staffing Provider Characteristics Related to Self-management 1. Evidence-based Guideline Compliance • Medication Complexity • Medication Intensity 2. Provider Communication Provider Factors: Clinical Inertia Failure of providers to initiate or intensify therapy when indicated  Reasons:  – Overestimation of care provided – “Soft” reasons to avoid intensification – Lack of education, training or practice organization – Lack of belief of efficacy Phillips, et al. Ann Intern Med; 2001 Issues in Patient-Provider Communication  Poor patient-physician communication is common – Physicians do >60% of talking during a visit • Instrumental and biomedically focused • Rarely address psychosocial issues – ~50% of the time physicians do not name the medicine or give dosing instructions – Many patients reluctant to express • Expectations or medication preferences • Misunderstandings about the regimen  Poor patient-provider communication may contribute to health disparities in minority populations  How do you translate this information into an intervention? Veterans Study to Improve The Control of Hypertension The V-STITCH Study VA Health Services Research Investigator Initiated Award, 2001-06 The V-STITCH Study  A randomized controlled trial testing two interventions designed to improve BP control – – Patient Intervention: Self-Management Provider Intervention: Decision Support    Durham VAMC General Medicine Clinics Patients with hypertension on medications 24 month intervention and follow-up The V-STITCH Study Design Providers Randomly Assigned (clusters) R1 Provider Intervention R2 Provider Reminder R2 Patient Intervention N = 150 Patient Usual Care N = 151 Patient Intervention N = 144 Patient Usual Care N = 143 Patient Intervention Patient Intervention Tailored Behavioral Delivered via Telephone 1. Hypertension Knowledge • African American • Diabetes • Family history Literacy Memory Patient’s Relationship with Primary Care Provider Social Support Side Effects Lifestyle Factors (smoking, alcohol, exercise, diet, stress) 8. Missed Appointments 9. Pill Refill 2. 3. 4. 5. 6. 7. Patient Intervention Frequency of Nurse-base calls 1st & 13th months 3rd & 15th months 5th & 17th months 7th & 19th months Physician Interaction, Memory, Literacy, Side effects Hypertension knowledge, Memory, Literacy, Side effects, Missed appointments, Pill refills Lifestyle, Memory, Literacy, Side effects, Missed appointments, Pill refills Social Support, Memory, Literacy, Side effects, Missed appointments, Pill Refills 9th & 21st months 11th & 23rd months Any month Physician Interaction, Memory, Literacy, Side effects, Missed appointments, Pill Refills Memory, Literacy, Side effects, Missed appointments, Pill Refills Patient initiated Patient Intervention Mode of Administration Use of Telephone  Telephone contact has been shown to be effective in changing patient behavior (Am J Hypertens 1996, Am J Prev Med 2002)   Allow reaching more patients Tend to be more acceptable and convenient than in-person interventions.  Most U.S. homes have phones (>97%) – useful tool to deliver an intervention (U.S. Bureau of Census, 2003)  May enhance the interventions’ costeffectiveness, due to reduced intervention costs and reduced visit rates. Provider Intervention Automated Treatment for Hypertension: EvideNce-based Advice (ATHENA)   Displayed at point-of-care Summarized the hypertension-relevant information from clinical record   Individualized for the patient Educated as well as reminded  Displayed reasons / explanations  Provided continuous quality improvement quarterly ATHENA: BP - Prescription Graphs Provider Control Group   Displayed patient's most recent BP Displayed patient’s current antihypertensive drug regimen   Provided opportunity to update BP Offered no advisories or recommendations for medication management Simply a reminder for hypertension  Primary Care Providers 24 Attending Physicians 6 Physician Assistants 2 Registered Nurse Practioners  17 intervention providers received full decision support tailored to specific patient 15 control providers received display with most recent BP  Patient Identification 4017 ICD code for Hypertension Hypertensive Medication Enrolled Durham VA 816 Contacted either by Telephone or In-person 588 Enrolled • 76% participation rate • 85% 24 month retention rate 190 Refused 38 Excluded • Hospitalized last 3 months • Dementia diagnosis • Resident in nursing home • Severely impaired hearing or speech Patient Characteristics (N=588) Male Mean age Married Live alone White African American High school or less Inadequate income 98% 63 years (21-87) 68% 22% 57% 40% 51% 23% Patient Characteristics Taking BP meds for > 5 years Close relative with hypertension No exercise Smoke Diabetic BP in control at Baseline: < 130 / 85 mm/Hg diabetic < 140 / 90 mm/Hg non-diabetic Mean Systolic BP 138.4 (SD=18) Mean Diastolic BP 75.5 (SD=11) 64% 65% 44% 30% 40% 42% Primary Outcome Blood pressure control at every primary care provider clinic visit over 24 months patient 1 yes bp control bp control yes patient 2 no 0 10 20 30 40 time in weeks 50 60 no 0 10 20 30 40 time in weeks 50 60 patient 3 yes bp control bp control yes patient 4 no 0 10 20 30 40 time in weeks 50 60 no 0 10 20 30 40 time in weeks 50 60 Blood Pressure Control Rates Primary Analysis 0.7 Behavioral N=144 0.6 BP Control Combined N=150 0.5 Reminder N=143 0.4 0.3 Decision Support N=151 0.2 Time Effect: P=.01 0 6 Group*Time Effect: P=.11 12 18 24 Time in Months Nurse Behavioral Intervention vs. None Secondary Analysis 0.70 BP Control RN Behavioral N=294 P=0.03 No RN N=294 0.40 0 0.50 0.60 6 12 18 24 Time in Months Compliance with Nurse Telephone Intervention  Patients completing all 12 scheduled study calls: 85% Average length of call: 3 minutes (SD 2.5 min)  Primary Care Visits During Study (24 Months) Mean Usual Care 7.7 SD 4.7 Behavior only Decision Support Combined 7.3 7.4 3.6 3.6 7.1 3.5 Two-Year Outpatient Costs Cost Category Average cost per Subject $2,863,775 $9,741 Total cost Behavioral intervention (n=294) Non-Behavioral (N=294) $2,822,215 $9,599 Average Behavioral Intervention Costs Per Patient over 24 months Cost Category Number of Patients Patients Overseen by Nurse 1120 840 560 Direct Costs/per patient Average min/per patient $70 $94 ($61-81) ($82-$109) 15 20 $141 ($123-$163) 30 Provider Intervention Results ATHENA displayed at 68% of visits (929/1370) • Among displayed, providers interacted with • 54% BP control when provider interacted intervention 57% of time (38.5% overall) versus 45% when provider did not interact Provider Intervention Results Most common reasons for disregarding recommendations • 68% inadequate BP control due to med • 68% concern that an inaccurate BP • 46% insufficient time reading was used to generate recommendations non-adherence Summary  Brief telephone intervention improved BP control by 21% at 24 months • 12.6% improvement compared to the non-behavioral group   No increase in clinic utilization Cost effective  Computer Decision Support did not significantly improve BP control rates at 24 months Next Study  How can we overcome provider inertia with a stronger medication management intervention? Focus intensive interventions on those at greater risk (i.e., out of control) Can we monitor and treat blood pressure outside of clinic?   Hypertension Intervention Telemedicine Study (HINTS) Department of Veterans Affairs, Grant IIR 04-426 (2005-2008) Established Investigator Award, American Heart Association (2006-2011) Hypertension Intervention Nurse Telemedicine Study (HINTS)  600 primary care veterans with poor BP control Home BP tele-monitoring used to activate interventions Nurse-administered via telephone for 18 months   HINTS Study: Design Four Group Design Usual Care  PCP drive management, no special program Tailored Behavioral Phone Intervention  Home BP monitoring evaluated by nurse  Tailored behavioral modules Medication Management (ATHENA) Phone Intervention  Home BP monitoring evaluated by nurse  Medication management implemented by study MD/RN Combined Intervention  Home BP monitoring evaluated by nurse  Medication management/tailored behavioral modules Why BP Monitors as Interventions? • Improve BP control • Self-monitoring programs are used in • clinical practice to assist PCP in treating their patients Encourage patients to monitor their disease Provide objective information to motivate patients to control their health condition • Case for Telemedicine  Effective use of home BP monitoring improve hypertension outcomes Treating at home may avert visits and result in better BP control Alternative way to integrate home BP monitoring into primary care   HINTS Study: Telemedicine Baseline Patient Characteristics 546 subjects enrolled Minority Low Literate Diabetic Males 51% 38% 44% 98% Summary of Intervention • Safety alert activated (2 values within 12 hours >175 systolic, >105 diastolic, pulse <40 or >110) • • 144 times, 51 unique pts Intervention activated 687 times, 241 unique pts Praise alert activated 74 times, 68 unique pts Summary of Intervention • Technicals activated 634 times, 220 unique pts 7% Did not understand how to set up or use equipment 66% nonadherence 27% technical problems with equipment Home Readings: Console View RN:MD Dialogue for Medication Change Outcomes  BP control – 0, 6, 12, 18 months     Health-related quality of life (SF-12) Hypertension knowledge Adherence to hypertension regimen Cost-effectiveness of both interventions Summary Need to consider • • • • • • Alternative methods of implementing interventions Telemedicine not panacea for all Cost of implementing interventions Methods of reimbursement Not just initiating, but maintaining multiple health behaviors Both patient/provider and possibly system Recommendations • Self-management adherence-enhancing strategies need to occur: Introduction of treatment Later in the course (remediation) Maintenance (less attention) • Strategies include: • Social Support • Educational Interventions (written and/or verbal • instructions delivered individually, group, telephone, or audiovisually) Behavioral Strategies (self-monitoring, positive reinforcement, goal setting, cueing, chaining) Recommendations Educational Interventions • Knowledge alone will not change behavior • Establish what is known before offering new knowledge • Use concrete examples Recommendations Ways of Presenting Written Information • Instructions should be clear and structured • Picture charts, color-coded medication schedules and large print Recommendations Behavioral Interventions Strategies include: • Developing prompts and reminder systems • Identifying a potential relapse into old behavior • Setting appropriate and realistic goals • Simplifying regimens to once or twice daily •Use opportunities to model behavior •Reinforce positive behaviors Recommendations Clinical Issues • Key validated question “Have you missed any pills in the past week?”  Sensitivity > 50% of those with low adherence  Specificity of 87% • Common misperceptions should be anticipated and avoided  i.e., medication can be stopped when the prescription runs out or symptoms are guides to when to take the medication Recommendations Clinical Issues (continued) • Missing appointments is correlated with lower adherence rates - first sign of dropping out of care entirely, the most severe form of nonadherence. • Telephone or appointment reminders provide relatively easy method to overcome nonadherence. Recommendations  Effective, collaborative providerpatient communication should be the foundation of all clinical interventions designed to improve patient selfmanagement. Future Directions & Conclusions  Examining tailoring of intervention mode to needs and intensity of intervention (Stepped level care) Disseminating and sustaining interventions in the community Expanding behavioral interventions to multiple chronic diseases Translating evidence into practice    Acknowledgements Research Team Alice Neary Mike Harrelson Pam Gentry Rowena Dolor Martha Adams Laurie Leeson Courtney Van Houtven Sharon Hooker Janet Grubber Melinda Orr Felicia McCant Laura Svetkey Tara Dudley Shelby Reed Anthony Goodin Ben Powers Tina Hong Maren Olsen Kelly Deal Mary Goldstein Laurie Marbrey Santanu Datta Gwen McKoy Cindy Rose David Simel Relevant Publications 1. Bosworth HB, Olsen MK, McCant F, et al. Hypertension Intervention Nurse Telemedicine Study (HINTS). Am Heart J 2007;153(6):918-24. 2. Bosworth HB, Olsen MK, Goldstein MK, et al. The veterans' study to improve the control of hypertension (V-STITCH): design and methodology. Contemp Clin Trials 2005;26:155-68. 3. Chan AS, Coleman RW, Martins SB, et al. Evaluating provider adherence in a trial of a guideline-based decision support system for hypertension. Medinfo 2004;11(Pt 1):125-9. 4. Goldstein MK, Coleman RW, Tu SW, et al. Translating research into practice: organizational issues in implementing automated decision support for hypertension in three medical centers. J Am Med Inform Assoc 2004;11(5):368-76. 5. Goldstein MK, Hoffman BB, Coleman RW, et al. Implementing clinical practice guidelines while taking account of changing evidence. Proc AMIA Symp 2000:300-4. 6. Goldstein MK, Hoffman BB, Coleman RW, et al. Patient safety in guidelinebased decision support for hypertension management: ATHENA DSS. Proc AMIA Symp 2001:214-8. Relevant Publications 7. Lin ND, Martins SB, Chan AS, et al. Identifying barriers to hypertension guideline adherence using clinician feedback at the point of care. AMIA Annu Symp Proc 2006:494-8. 8. Bosworth HB, Oddone EZ. Telemedicine and Hypertension. J Clin Outcomes Management 2004;11(8):517-522. 9. Bosworth HB, Oddone EZ, Weinberger M. Patient treatment adherence: Concepts interventions, and measurement. Mahwah, NJ: Lawrence Erlbaum Associates, 2006. 10. Bosworth HB, Dudley T, Olsen MK, et al. Racial differences in blood pressure control: potential explanatory factors. Am J Med 2006;119(1):70. 11. Bosworth HB, Oddone EZ. A model of psychosocial and cultural antecedents of blood pressure control. Journal of the National Medical Association 2002;94:236-248. 12. Bosworth HB, Olsen MK, Gentry P, et al. Nurse administered telephone intervention for blood pressure control. Patient Educ Couns 2005;57(1):5-14. 13. Bosworth HB, Olsen MK, Oddone EZ. Improving blood pressure control by tailored feedback to patients and clinicians. Am Heart J 2005;149(5):795-803.