Jonikas, Glover, & Cook, 2014 Jonikas, Glover, & Cook, 2014 Special Session of the Mental.

Jonikas, Glover, & Cook, 2014

www.cmhsrp.uic.edu/health/


www.cmhsrp.uic.edu/health/index.asp

Special Session of the Mental Health DivisionImproving health outcomes among individuals

with serious mental illnessesPresentations

Improving health outcomes through use of disease registries and care coordination for patients with co-occurring diabetes and mental illness. Judith A. Cook, Ph.D & Pamela

Steigman, M.A.

Community health screening and its role in public health and psychiatric epidemiology. Lisa A. Razzano, Ph.D., Jessica A. Jonikas, Ph.D. & Peggy Swarbrick, Ph.D. OT.

A randomized trial to improve health and wellness among a public mental health sample of individuals with severe mental illness. E. Sally Rogers, Sc.D. & Mihoko Maru, M. A., M.S.W.

The Recovery Center: A model designed to promote health and wellness through education and coaching. Dori Hutchinson, Sc.D.

Discussant: Crystal Blyler Ph.D. Mathematic Policy Institute

Authors have no conflicts of interest to declare.



Implementing a Diabetes Registry and Care Coordination in Community Mental and Physical Health Clinics

Click icon to add picture

Judith A. Cook, Ph.D. & Pamela J. Steigman, M.A.

American Public Health Association Annual Meeting

Chicago, ILNovember 2, 2015



U.S. Department of Education, National Institute on Disability & Rehabilitation Research

Substance Abuse & Mental Health Services Administration, Center for Mental Health Services

Cooperative Agreement #H133G100028

With thanks to our funders

Cook & Steigman, 2015


http://www.cmhsrp.uic.edu/health/index.asp





Project Collaborators

UIC Center on Psychiatric Disability & Co-Occurring Medical Conditions

UIC College of Nursing, Integrated Health Care Clinics

Thresholds Psychiatric Rehabilitation Centers UIC Eye & Ear Infirmary & Dr. LaVallee UIC Podiatry Clinic & Dr. Robert Laveau Kennedy-King College’s Dental Hygiene

Department








Key Study Partners

Sue BraunEmily BrigellKathy ChristiansenKristin DavisKaty DobbinsJay Forman

Ann HeesackerAsma JamiSheila O’NeillDeborah PavickJessica JonikasCrystal GloverJoni Weidenaar






A Public Health Crisis

• On average people with SMI die 25 years earlier than general population

• 2/3 mortality due to treatable medical conditions

• Attributed to lifestyle (poor diet, low physical activity), systemic factors (health care access, care fragmentation) & social determinants (poverty, discrimination)

Higher diabetes prevalence increases risk for developing…

• Hypertension• Hyperlipidemia• Heart disease • Kidney disease• Gum disease/loss of

teeth• Nerve damage/loss of

feet • Eye disease/blindness • Costs are 2.4 times

greater; nearly 40% of costs due to long-term complications.Cook & Steigman, 2015



Jonikas, Glover, & Cook, 2014www.cmhsrp.uic.edu/health/index.asp

Use of a Registry to Manage Care for Diabetes in Integrated Health Clinics for Adults with Serious Mental Illnesses* Judith A. Cook, PhD, Principal Investigator

Intervention tested:

1. Electronic Diabetes Registry & Education Materials

Promote adherence to ADA standards of care Develop tailored diabetes education resources**

2. Care Coordinator Link patients to specialty care in accordance

with ADA standards Educate patients about diabetes self-

management

*http://www.cmhsrp.uic.edu/health/medical_home_registry.asp

**



What is a Diabetes Registry? Database with demographics,

illness characteristics, treatment delivered, and specialty care arranged/delivered

Information from electronic and paper records guides care, tracks outcomes, and informs plans for improving care

Supports proactive care by facilitating care planning, sharing of information with other providers, and generating patient reminders

Generates charts and graphs to support illness self-management

Generates reports to monitor team and system performance

Overall goal is to improve adherence to treatment guidelines and self-management


www.cmhsrp.uic.edu/health/index.aspCook & Steigman, 2015








Standard Target

Blood Glucose Control (HbA1c) Less than 7%

Blood Pressure Less Than 140/90 mmHg

LDL cholesterol Less Than 100 mg/dl

Urine Screening for Microalbumin

Annual screening

Dilated eye exam Annual screening

Foot exam for neuropathy Annual screening

Dental exam Annual screening

Vaccinations Lifetime and annual

Sample Standards Tracked in Diabetes Registries for Individual & Population Management






One electronic database contains data from multiple sources to inform complex disease management

Allows immediate focus on managing chronic disease at population level

Can be used by multiple parties (clinicians, patients, administrators) to facilitate care delivery while meeting care standards

Why Registries for Standards of Care?

(Ortiz, 2006)








Population Studies using a Diabetes RegistryImproving Diabetes Care in a Large Health Care System: An Enhanced Primary Care ApproachSperl-Hillen, et al. (2000). Joint Commission Journal on Quality and Patient Safety

Improved glycemic and lipid control among approximately 7,000 adults with diabetes.

The Impact of Planned Care and a Diabetes Electronic Management System on Community-Based Diabetes Care:The Mayo Health System Diabetes Translation Project Montori et al. (2002). Diabetes Care.Registry use augmented the impact of planned care on performance outcomes (increased use of specialty medical care) and certain metabolic outcomes. Did not impact glucose levels.



http://www.ingentaconnect.com/content/jcaho/jcjqs;jsessionid=hb7wlplhu0ne.alice






Diabetes Registries: Across Clinics

Improving Diabetes Outcomes Using a Web-Based Registry and Interactive Education: A Multisite Collaborative Approach Morrow, R. et al. (2013) Journal of Continuing Education in the Health Professions

• Electronic diabetes registry in 7 clinics in NY • With educational module on the registry and patient communication

With each quarter post-Registry, patients were:• 1.4 times more likely to have A1C ≤ 9• Almost twice as likely to have LDL < 100 • 1.3 times more likely to have BP < 140/90

Likelihood of adherence increased over the initial quarters (except for BP; adherence fell over time). There was a drop-off among all indicators after 5 quarters, suggesting ongoing support and training are needed








Registries vs. Electronic Health Records Most EHRs are not built to function as registries,

so can’t support population-based care It can take years for population reporting from an

EHR A registry is relatively easy and inexpensive

• Can have nearly immediate impact on clinic practice and client engagement & outcomes

It can be instructive to learn population-based care parameters prior to implementing an EHR via a registry• Allows you to design EHR processes to support needs

identified by registry use

Content adapted from: www.powershow.com/view/21d14-MzEyZ/Using_Excel_for_a_HgA1c_Registry_powerpoint_ppt_presentation



http://www.powershow.com/view/21d14-MzEyZ/Using_Excel_for_a_HgA1c_Registry_powerpoint_ppt_presentation

http://www.powershow.com/view/21d14-MzEyZ/Using_Excel_for_a_HgA1c_Registry_powerpoint_ppt_presentation






• Combine use of registry with care coordination to improve patient outcomes

• Enhance adherence to ADA standards of care by improving care delivery and coordination

• Link patients to specialty care in accordance with ADA standards

• Teach patients about diabetes & promote self-management

• Develop new treatment/service resources• Monitor health indicators and outcomes over time

The Purpose of Our Project








The Collaboration

UIC College of Nursing operates nurse-managed Integrated Health Clinics (IHCs) where study participants received primary care. At the time the initial study sample was drawn, the IHCs were serving 220 patients with co-occurring diabetes & mental illness.

Thresholds houses one IHC at its program on Chicago’s north-side and one on the south-side. The agency provided mental health care and case management that linked clients to medical care, encouraged attendance at IHC and specialty care appointments, and helped them self-manage co-occurring conditions.

UIC Center on Psychiatric Disability and Co-Occurring Medical Conditions built the registry populated with data on 220 patients; funded the study’s Care Coordinator who liaisoned with IHC and Thresholds staff








Sample Patient Specific Registry Report





SamplePatient ReportCard



Sample Patient Education

Study time frame

Pre-Intervention

Period24 months

April 1, 2010 – March 31, 2012

InterventionImplementation

Period12 months

April 1, 2012-March 31, 2013

Follow-UpPeriod

12 monthsApril 1, 2013 –March 31, 2014

Background Characteristics of Registry Study Participants at Baseline*

Background Features Full Sample (N=179)

North Clinic (N=88)

South Clinic (N=91)

Male 66.5% 68.2% 64.8%

Race: African AmericanWhite

Hispanic/Asian/Other

61.8%28.3% 9.8%

46.5%40.9%11.7%

77.0%**14.9% 8.0%

Age (mean, SD) 51.22 (9.8) 52.46 (9.6) 50.0 (9.8)

Education: < High SchoolHigh SchooL Grad

Some College +

30.9%41.9%27.2%

25.3%50.7%20.9%

35.8%34.5%29.9%

Diagnosis: BipolarDepression

Schizoaff/Schizophrenia

19.0%19.0%62.0%

18.2%21.6%60.2%

19.8%16.5%63.8%

*Baseline = April 1, 2010**At baseline, clinics differed significantly on race with a higher % of African Americans at south clinic (p<.05)

Changes in Clinical Outcomes Over Time

Quality Measure

Nadir Value at Pre-Intervention Mean (SD)

Most Recent Follow-UpMean (SD)

Independent T-Testt, p-value

A1c 8.0 (2.4) 6.9 (1.7) 5.0, p<.001

LDL 107.4 (34.8) 89.4 (29.9) 4.9, p<.001

HDL 43.5 (14.5) 44.2 (13.0) 0.4, p=.682

Triglycerides 173.2 (116.2) 144.2 (84.8) 2.4, p=.015**

Total Cholesterol 182.0 (42.6) 165.4 (37.8) 3.6, p<.001

Cholesterol/HDL 4.5 (1.6) 3.9 (1.3) 3.5, p=.001

Triglycerides/HDL 4.4 (3.6) 3.4 (2.4) 2.7, p<.008***

Trig/HDL < 2.0 (%) 19.9% 30.4% 2.1, p=.042

Trig/HDL > 4.0 (%) 41.8% 29.6% 2.1, p=.035

Comparisons of Pre- & Follow-Up Lab Values (N=179)*

*Pre-Intervention period = April 1, 2010 – March 31, 2012; Follow-Up period = April 1, 2013 – March 31, 2014 **In sub-analysis, statistically significant decrease in triglycerides was present for only the North clinic***In sub-analysis, statistically significant decrease in triglycerides/HDL ratio was present for only the North clinic

Responder Rate: Meeting 3 ADA Standards A1c<7, LDL<100 and BP<140/90*

Nadir Value at Pre-Intervention (%)

Most Recent Follow-Up (%)

Paired T-Testt(DF), p-value

Responder Rate 9.15% 39.22% 7.23(152), p<.001

Patient-Level With-Subjects Comparison of Pre- & Follow-Up Responder Rates (N=128 pairs)

Combined Clinics Comparison of Pre- & Follow-Up Responder Rates (N=179)

Nadir Value at Pre-Intervention (%)


Independent T-testt(DF), p-value

Responder Rate 9.83% 39.49% 6.70 (328), p<.001

*Yu, G. C., & Beresford, R. (2010). Implementation of a Chronic Illness Model for Diabetes Care in a Family Medicine Residency Program. Journal of General Internal Medicine, 25(Suppl 4), 615–619. doi:10.1007/s11606-010-1431-9

Proportion Meeting Cut-Offs of Clinical Standards

Clinical StandardsNadir Value at Pre-Intervention (%)


Paired T-Testt, p-value

A1c<7% (N=170) 46.5% 72.9% 7.18, p<.001 A1c<8% (N=170) 65.9% 86.5% 5.83, p<.001A1c>=10% (N=170) 18.8% 7.6% 3.96, p<.001

LDL<100 (N=134) 44.0% 67.2% 4.66, p<.001

LDL<130 (N=134) 72.4% 92.5% 5.32, p<.001

BP<140/90 (N=179) 38.0% 49.0% 10.11, p<.001

Patient-level Within-Subjects Comparison of Pre- & Follow-Up Lab Values

Proportion Meeting Cut-Offs of Clinical Standards

Clinical StandardsNadir Value at Pre-Intervention (%)


Independent T-Testt, p-value

A1c<7% 47.1% 72.7% 5.01, p<.001

A1c<8% 66.1% 86.6% 4.62, p<.001

A1c>=10% 19.0% 7.6% 3.16, p=.002

LDL<100 44.3% 66.7% 4.04, p<.001*

LDL<130 73.6% 92.4% 4.43, p<.001

BP<140/90 39.3% 82.7% 9.35, p<.001

Combined Clinics Comparison of Pre- & Follow-Up Lab Values (N=179)

*In clinic sub-analysis, statistically significant increase in proportion of patients with LDL<100 was present for only the North clinic

Changes in Clinical Outcomes Over Time Controlling for Site and Race (N=179)*

Outcome Variable Z-Score P-value

A1c -5.52 p<.0001

A1c >= 6.5 -6.09 p<.0001

A1c > 8.0 -5.45 p<.0001

LDL 0.24 p<.0001

LDL > 130 -4.56 p<.0001

Triglycerides -3.78 p<.0001

Total Cholesterol -1.39 p<.0014

Triglycerides/HDL Ratio -4.04 p<.01

Blood Pressure < 140/90 9.74 p<.0001

*Random regression analysis was used to detect changes over 16 study quarters using the nadir value (worst) at pre-intervention, controlling for clinic site and race.

Changes in Specialty Care Outcomes Over Time (N=179)

Type of CareCompleted Referrals

During Pre-Intervention

% (N)

Completed Referrals During Follow-Up

% (N)

Chi-Square p-value

Dental 3.5 (6) 16.8 (29) 16.815, p<.001

Optometry 23.1 (40) 33.3 (58) 4.464, p=.023

Podiatry 17.2 (30) 27.6 (48) 5.354, p=.014

Care coordination was effective in helping to improve the specialty care referral completion rate. Between the pre– and Follow-Up time points…

completed dental referrals increased by 380% completed optometry referrals increased by 44% completed podiatry referrals increased by 60%.



Shifting from reaction to prevention

Moving from individual level to population-based care

Getting multiple partners invested

Key Barriers to Registry Use

Time to load and maintain the spreadsheet or database

Measuring performance can be threatening

Just another fad?

Adapted from:www.powershow.com/view/21d14-MzEyZ/Using_Excel_for_a_HgA1c_Registry_powerpoint_ppt_presentation Cook & Steigman, 2015





Learn about our registry studywww.cmhsrp.uic.edu/health/medical_home_registry.asp

Free Diabetes Toolkit http://www.cmhsrp.uic.edu/health/diabetes-library-home.asp

Jonikas, Glover, & Cook, 2014 Jonikas, Glover, & Cook, 2014 Special Session of the Mental.

Documents

Transcript of Jonikas, Glover, & Cook, 2014 Jonikas, Glover, & Cook, 2014 Special Session of the Mental.