Methods for Health Care Quality Improvement Research III

Methods for QI Research Work at the QI/Research Interface Lisa V. Rubenstein, MD, MSPH Academy Health, 6/2004 The Quality Chasm Practice Research Theory and Practice of QI Research QUALITY IMPROVEMENT STRUCTURE •Providers •Organization •Environment •Patient Pop PROCESS •Clinical actions OUTCOME status •Economy •Satisfaction • Health  providers  systems Provider Behavior Quality Improvement Research  Systematic investigation to identify methods for improving quality that generalize to a recognizable type of site  Aims to change structure  Aims to improve care across all patients in need in the site Cardiovascular Prevention Project (Baggett)  Goals  Improve secondary cardiovascular prevention  Setting  Urban Gen Med inpatients  Questions  Intervention design?  Generalizable, fundable project? How could the intervention be strengthened? Intervention Design Approaches Quality Improvement Research Stem and Leaf diagrams Flow charts Asking front-line workers* PDSA cycles to improve the intervention Empirical research* Theory Asking experts Pilot tests for intervention feasibility *Already carried out Cause-and-Effect Diagram (stem/leaf, fishbone, Ishikawa) Note: Exclude “Not Indicated” Patients Staff Education Clinicians Cost Communication Education Preferences Priority-setting Re. guidelines Decision support Financial Responsibility Poor Prevention Education Decision Support Communication Policy Materials List of Needed Components  Clinician education and decision support  Include assessment  Patient self-management support  Include preference assessment  Staffing policies re. responsibility and communication  Priority-setting process  Decisions about how to handle costs Community Resources and Policies Health System Organization of Health Care Selfmanagement Support Decision Support Clinical Information Systems Delivery System Design Informed, activated patient Productive Interactions Prepared, proactive practice team Functional and Clinical Outcomes Flow Chart Patient admitted to (list units, clinicians) When should each prevention activity occur? (e.g., ACE inhibitors earlier) Who should carry out each prevention activity? (e.g., exercise prescription by Rehab Service?) Discovery: Subgoals Need to be Considered Separately  NHLBI Guidelines  Aspirin & antiplatelet agents, beta blockers, ACE inhibitors/ARBs, lipid assessment and treatment, exercise counseling and prescriptions, diabetes control, blood pressure control, dietary counseling and prescriptions, smoking cessation counseling and referral  Each needs goals, priority-setting, flow Intervention Design Approaches Quality Improvement Research Stem and Leaf diagrams* Flow charts* Asking front-line workers* PDSA cycles to improve the intervention Empirical research* Theory Asking experts Pilot tests for intervention feasibility *Already carried out Is this project ready to go?  Investigators asked, “how can we craft a nationally generalizable and fundable research project?” Where Is This Project on the Path from Research into Practice? VA QUERI Model            Identify target problem  Define best practices  Assess care variations Assess determinants of care Develop intervention models  Evaluate effectiveness  QI Identify successful model characteristics Intervention Develop dissemination models Design Evaluate dissemination Implement as routine policies and procedures Evaluate performance measures Depression Example •Effectiveness of telephone CM •Cost effectiveness •Web tools •QI Methods •Seamless with usual care Katon: Collaborative Care for Depression •Anxiety •Minor depression •Somaticizers •Elderly Model Dissemination Full Translation VA TIDES/WAVES RWJ ICICE, demonstrations Partners in Care and QID •Fee for service, network, staff model, practice size •Rural vs urban Theories as signposts Prior Research  “In one institution, using educational interventions and nurse care managers to remind house staff was moderately effective”  “Moderately effective” may imply a large population impact—what are our goals?  What tasks did the nurse care manager do, and can we take care of them cheaper?  Was the education optimal?  Which specific prevention activities improved? Ways to Choose an Approach  Consult with experts  Rate each potential intervention component on feasibility and potential impact  Do the above, and involve stakeholders, including senior organizational leaders and patients, in identifying promising approaches ……Producing real change is hard …and it probably paid off. Evaluation Design Formative Evaluation  Formative: For improving the intervention, and for replicating it  QI projects are not generalizable in the absence of a formative evaluation and documentation of intervention components  Partners in Care—HA, website  QI approaches with good formative evaluations are publishable  Breakthrough Series  PDSA and “how to” frameworks  Literature review and theory Intervention Design Approaches Quality Improvement Research Stem and Leaf* diagrams Flow charts* Asking front-line workers* PDSA cycles to improve the intervention* Empirical research* Theory* Asking experts* Pilot tests for intervention feasibility* QI Research Designs  PDSA (early or late)  Pre/Post, time series for exposed individuals in the site  Cluster non-randomized experiments  Pre/post, time series across site or providers  Matched usual care  Cluster randomized trials Are We Having Fun Yet…