This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Showing Showing Health Information Value in Health Information Value in
a Community Networka Community Network
David F. Lobach, MD, PhD, MSDavid F. Lobach, MD, PhD, MS
Division of Clinical Informatics Division of Clinical Informatics Department of Community and Family MedicineDepartment of Community and Family MedicineDuke University Medical Center, Durham, North Duke University Medical Center, Durham, North
AHRQ Health Information Technology ConferenceAHRQ Health Information Technology Conference June 9, 2005June 9, 2005AHRQ Health Information Technology ConferenceAHRQ Health Information Technology Conference June 9, 2005June 9, 2005
Project DescriptionProject Description Community Network (COACH)Community Network (COACH) Decision Support Tool (SEBASTIAN)Decision Support Tool (SEBASTIAN) Data-Driven InterventionsData-Driven Interventions
Selection of Evaluation MetricsSelection of Evaluation Metrics Study DesignStudy Design Obtaining and Using DataObtaining and Using Data Project ChallengesProject Challenges Practical IssuesPractical Issues
Population Management - Population Management - Asynchronous recommendations Asynchronous recommendations generated though a rule-based generated though a rule-based intelligent agent running on a intelligent agent running on a clinical data repositoryclinical data repository
Patient Management - Interactive, Patient Management - Interactive, evidence-based recommendations evidence-based recommendations delivered to the point of caredelivered to the point of care
Collaborators and PartnersCollaborators and Partners Specific Aims and ObjectivesSpecific Aims and Objectives Intervention OverviewIntervention Overview Study MeasuresStudy Measures Study DesignStudy Design Data SourcesData Sources
This work was funded in part by a grant fromThis work was funded in part by a grant from
Hospitals Duke University Hospital (DUH)Durham Regional Hospital (DRH)
Clinics Duke Primary Care ClinicsDuke Affiliated Primary Care PracticesPrivate Primary Care ClinicsLincoln Community Health Center (LCHC)Duke Specialty Care Clinics
Urgent Care Facilities Duke Urgent CareLincoln Community Health Center (LCHC)
Government Agencies Durham County Health Department (DCHD)Durham County Department of Social Services (DSS)NC Medicaid
Community Organizations Durham Community Health Network (DCHN)
To evaluate the clinical, To evaluate the clinical, organizational and financial value of organizational and financial value of HIT in a community network from a HIT in a community network from a societal perspective. societal perspective.
To evaluate the value of HIT in a To evaluate the value of HIT in a community network from the community network from the perspective of specific stakeholder perspective of specific stakeholder groups, including patients, providers, groups, including patients, providers, hospitals, payers, and purchasers. hospitals, payers, and purchasers.
Clinical Outcomes Three ED encounters within 90 daysLow severity ED encounter ED encounter for asthmaED encounter for diabetesHospitalization for asthmaHospitalization for diabetes
Care Quality No mammogram in 2 yearsNo Pap smear in 3 yearsNo Chlamydia screening in 1 yearNo post-partum visit after delivery in 21 to 56 daysNo glycated hemoglobin test in 1 yearNo lipid panel in 1 yearNo urine protein/albumin test in 1 yearNo dilated eye examination in 1 yearOne or fewer well child visits in 1 yearNo well child visits in 1 year
Care Coordination Two missed appointments in 60 daysNo Follow-up appointment after ear infection in 2 wks
Care Quality No staging of asthmaNo asthma action plan in 2 yearsPatient-reported health risk: new onset pregnancyPatient-reported health risk: missed >10 days of school/yearPatient-reported health risk: no blood test for lead poisoning in past yearPatient-reported health risk: no blood test for lead poisoning everPatient-reported health risk: signs or symptoms of vision problemsPatient-reported health risk: signs or symptoms of hearing problemsPatient-reported health risk: high risk of TB exposurePatient-reported health risk: concern about weightPatient-reported health risk: recreational drug usePatient-reported health risk: alcohol abusePatient-reported care barrier: need assistance to get care servicesPatient-reported barrier: need assistance to get medical equipmentPatient-reported care barrier: need assistance to get medications
Care Coordination Pending appointment and h/o missed appointments in last 60 days
Optimal Measures:Optimal Measures: Solid clinical outcomes of Solid clinical outcomes of significance to health care (e.g. mortality)significance to health care (e.g. mortality)
Will any true clinical outcomes be impacted Will any true clinical outcomes be impacted by the intervention that can be detected by the intervention that can be detected during the study period?during the study period?
Can a change in surrogate clinical measures Can a change in surrogate clinical measures be detected during the study period?be detected during the study period?
Will any process measures be impacted by Will any process measures be impacted by the intervention?the intervention?
What are the cost implications of the What are the cost implications of the intervention?intervention?
Will any descriptive/qualitative measures be Will any descriptive/qualitative measures be effected by the study intervention?effected by the study intervention?
How can that endpoint be How can that endpoint be measured*?measured*?
1.1. Relative number of ED visitsRelative number of ED visits2.2. HEDIS quality metricsHEDIS quality metrics3.3. Rates of follow-up appointmentsRates of follow-up appointments4.4. Costs of care providedCosts of care provided5.5. Satisfaction surveysSatisfaction surveys
*Consider what has been measured in similar *Consider what has been measured in similar studies. Don’t re-invent the wheel!studies. Don’t re-invent the wheel!
What type of comparison needs to What type of comparison needs to be made between control and be made between control and intervention groups to detect a intervention groups to detect a significant difference?significant difference?1.-5.1.-5. Compare differences between Compare differences between
groups – Wilcoxon rank sum groups – Wilcoxon rank sum statisticstatistic
ED encounters per 1000 pt yearsHospitalizations per 1000 pt yearsPrimary care encounters per 1000 pt yearsHgbA1c levels
Care Quality.Preventive ServicesChronic Disease
Management for Asthma & Diabetes
HEDIS scores for mammograms, Pap smears, Chlamydia screeningHEDIS scores for asthma staging, recording action plans, prescribing inhaled
corticosteroids HEDIS scores for performing glycated hemoglobin, microalbumin, lipid panel,
dilate eye examHEDIS scores for well child visitsAHRQ PQI for asthmaAHRQ PQI for uncontrolled diabetes, short- & long-term complications
Care Coordination. Missed AppointmentsF/U Appointments
Missed appointments per 1000 patient yearsRate of f/u after ear infection per 1000 patient years
Costs/Revenues. ED costsHospitalization Costs
Costs of HIT interventions including personnel/hardware/software costsCo-pay costsProfessional feesTechnical feesPharmaceutical reimbursement ED costsHospitalization costs
Satisfaction.Patient SatisfactionPatient Quality of LifeProvider Opinions
Comparisons of the six clinical/organizational and three economic endpoints. Strategies 1-3 are compared with strategies 4-6.
ComparisonsAmong HITTreatments
Comparisons of the six clinical/organizational and three economic endpoints. Indicator variables are used to denote the three highlighted strategies: provider alerts, feedback reports, and patient reminders.
Optimal Data:Optimal Data: Coded clinical data Coded clinical data entered electronically by the data entered electronically by the data generating sourcegenerating source
What coded data is available?What coded data is available? What clinical data is available?What clinical data is available? What claims data is available?What claims data is available? Is there a data dictionary for Is there a data dictionary for
interpreting the data?interpreting the data? What is required to make the data What is required to make the data
available?available? Will chart audits be required?Will chart audits be required?
Controlling ExpectationsControlling Expectations Avoiding “Scope Creep”Avoiding “Scope Creep” Start up FundingStart up Funding Operational FundingOperational Funding Obtaining Data from PartnersObtaining Data from Partners HIPAA ImplicationsHIPAA Implications
Open House @ Duke Open House @ Duke UniversityUniversity
Learn more about the AHRQ-funded Duke HIT Learn more about the AHRQ-funded Duke HIT Value Project at Duke University Medical CenterValue Project at Duke University Medical Center
Friday, August 12, 2005Friday, August 12, 2005 9:00 AM to 4:00 PM9:00 AM to 4:00 PM Detailed project presentationsDetailed project presentations Onsite demonstrationsOnsite demonstrations Cost: ~$25 for lunch, snacks and local Cost: ~$25 for lunch, snacks and local
transportationtransportation For more information and registration:For more information and registration:
DCIDCI Jennifer M. Macri, MSJennifer M. Macri, MS Garry SilveyGarry Silvey Allen MayersAllen Mayers Kensaku KawamotoKensaku Kawamoto Jan Willis, MSJan Willis, MS
DCHDCH Susan Yaggy, MPASusan Yaggy, MPA Fred Johnson, MBAFred Johnson, MBA Jessica SimoJessica Simo Pam PhillipsPam Phillips
Other Duke GroupsOther Duke Groups Eric Eisenstein, DBAEric Eisenstein, DBA Kevin Anstrom, PhDKevin Anstrom, PhD Vic Hasselblad, PhD David Crosslin, MS Bob Rezzarday Bob Rezzarday
Outside Collaborators Richard Low (Topsail Richard Low (Topsail