Improving Quality of Care Through Fully Empanelled Care Team Optimization and an Introduction to the Dena ina Wellness Model of Care Kenai, AK Dena ina Health Clinic Kenaitze Indian Tribe Kenai, AK Dena ina Health Clinic Kenai, AK I. Time for some Changes II. Development of Care Teams Overview Patient Population: 3,011 (1% monthly growth) III. IV. Dena ina Health Clinic s Innovative Approach to Empanelment Our success with Advanced Access Scheduling Staff: 23 (1 physician & 3 mid levels) EHR implemented since 2008 V. Acknowledging a need for Wellness Program & Behavioral Health Integration VI. Optimizing Fully Empanelled Care Teams in the Dena ina Wellness Model of Care Reasons for Change Walk In: Very long waits Scheduled appointments: several days out Continuity with provider: random Clinical quality measures: not tracked Health Records: some providers transitioning while others continued to use paper Provider centered care Pa ent Centered Care Work processes: random Collaboration with dental, wellness, and behavioral health = non existent Before Care Team integration Brief history Prior to 2010, MAs would alternate rooming patients from the queue when they arrived. This caused issues when various MAs carried more weight than others. MAs took turns throughout the week acting as the triage MA. They would search the chart to find out which provider the patient saw last, then room them for that provider (this does not work when you have a new provider). This process became very frustrating because workloads were constantly unbalanced due to provider schedules and styles. Everyone was supposed to do everything for everybody. 1
Transitioning to the Teamlet Model By the end of 2010, we had 3 full time providers, and 3 full time medical assistants. Provider/MA pairs were established Eventually co located Erika & Penny Kasie & Carol Economies of Scale Shane & Jane "Everyone working on the same task needs to keep in sync, so as more people are added they spend more time trying to find out what everyone else is doing" (Wikipedia Brooks's Law) Jan. 2010 Assembled Care Teams (1 Provider : 1MA) Supported by 2 RN's and a Health Educator Apr. 2010 Provider meetings, nursing meetings, front office meetings discussing Open Access Panel reconciliations PDSA Analyses Feb. 2010 Empanelment started and 100% completed ~3,020 patients divided evenly by last name Pts were then allowed to choose the provider of their choice 2010 Implemented Open Access scheduling Only 4 appointments per provider Same Day appointments Mar. 2010 Sent out brochures explaining our empanelment process New Walk In slip 2010 Decreased Walk In provider utilization by 40%. Patient Empanelment Definition: Empanelment A deliberate attempt to identify the group of patients for whom a physician or team is responsible. (Safety Net Medical Home Initiative 2010.) We decided to empanel our patients by last name. This allowed us to: 1. Avoid searching through every medical record to see which provider the patient saw most frequently or last. 2. Avoid having to develop an elaborate scheme to weight panels (e.g. patients with multiple diagnoses, chronic disease, age, etc.) 3. Avoid having to assign each patient as they came in to be seen. a. When asked who their provider is, most patients would say it didn t matter or that they ll see whoever. Feb. 2010 Empanelment started and 100% completed ~3,020 patients divided evenly by last name Pts were then allowed to choose the provider of their choice Mar. 2010 Sent out brochures explaining our empanelment process New Walk In slip Dr. Scheffel, I ve been working on setting up the empanelment protocols and here s what I ve established: For Pt. & Providers According to our newsletter, we ll be assigning pts to care teams (providers have agreed to retain their patients they ve seen regularly each has a list of 100 300) Pts. will be notified about their provider assignment and will have the option to change based upon their preference at their next visit For the Provider Panels According to Amazing Charts, we have 3,020 active patients. These can be divided evenly by last name (to keep families with same provider) A Hend = 1,008 Seen in the last year = 500 Henk.. Patr. = 1,003 Seen in the last year = 516 Patt Z.. = 1,009 Seen in the last year = 512 I know you mentioned assigning coefficients to patients with chronic conditions, elders, and pts with multiple conditions. In theory this numerical analysis for panel balancing would undoubtedly work; however, that would require extensive health information management. I think using random distribution by last name will provide a fairly even distribution. At most, this method only fills panels up to 75% (after pt swops and DM panels), which allows for plenty of room for adjustments. Start by Communicating with Providers Plan outline was sent to all providers explaining the plan to distribute patients by last name. It was emphasized that after the panel assignments, patients were allowed to override the assignment and change to a provider of their choice (*we ve observed a phenomenon that most patients were satisfied with their assignment). Process of Empanelment This is a continuation of the outline sent to providers: Process of Empanelment the actual process may require us to roll up our sleeves and do the input one by one It s hard to imagine panel hopping would be frequent, so it s likely that we re only going to have to actively manage about 10% of the pt. population. Since it is quite simple to designate a pt panel in their chart, 20 patients charts can indicate their panel in less than 5 minutes (about 12.5hrs for 3K pts). The idea about utilizing a Dr. Ancillary Panel came from an IPC Foundations Series call! We made a 4 th panel that collects all the patients that onetime visitors, inactive, received care elsewhere, etc. 2
Empanelment Sustainability Patients arriving at the clinic are always asked who their provider is and scheduled accordingly. New patients or patients not seen in the past 18 months (after empanelment) were asked to choose a provider. In order to keep panels balanced, we would close a panel and assign patients to the smallest panel if the patient was indifferent. Conclusion of outline sent to providers: Conclusion I believe we should move forward with this empanelment process as soon as possible. We are currently in the process of re designing our scheduling/visiting policy, which may be very confusing for patients. We can pitch both concepts at the same time, which would help us avoid two separate confusing changes. This change may also help alleviate some of the scheduling frustrations. Active management of the panels is inevitable. The proposed method will work to get things started and work very well for patients who are indifferent about who they see. After most patients have been empanelled, we can focus on allowing patients to join available panels based upon their individual preferences. Care Team Brochure Brochures were produced to describe the importance of empanelment and introduce the three Care Team providers. The inside of the brochure displayed all three mid level providers and their biographies. The most important feature of this brochure was for patients to be able to recognize their provider. Patients that do not have a PCP indicated in their chart were given the brochure to decide which provider they want to see. Age Distribution 1 Year Later Panel Sizes Kasie Erika Shane NotSeenin18Months Total 4/7/2011 980 1011 912 639 3542 27.7% 28.5% 25.7% 18.0% Age Distribution Teamlet A Teamlet Teamlet B C NotSeenin18Months 28.0% 28.2% 28.3% 103 15.5% 0 10 664 18.7% 186 187 188 11 18 552 15.6% 163 29.5% 162 29.3% 148 26.8% 79 14.3% 19 45 1496 42.2% 404 27.0% 421 28.1% 365 24.4% 306 20.5% 46 64 638 18.0% 173 27.1% 182 28.5% 165 25.9% 118 18.5% 65 79 152 4.3% 43 28.3% 48 31.6% 38 25.0% 23 15.1% 80+ 40 1.1% 11 27.5% 11 27.5% 8 20.0% 10 25.0% Total: 3542 1600 1400 1200 1000 800 600 400 Age Distribution 200 0 0 10 11 18 19 45 46 64 65 79 80+ Why Primary Care needs to be integrated with Behavioral Health and Wellness Programs "People with serious mental illnesses have higher rates of other illnesses and die earlier, on average, than the general population, largely from treatable conditions associated with modifiable risk factors such as smoking, obesity, substance abuse, and inadequate medical care" Colleen L. Barry, Ph.D., and Haiden A. Huskamp, Ph.D. NEJM 2011 Up to 50% this group of exceptionally high utilizers, may have unmet behavioral health needs. Lefevre, F., Reifler, D., Lee, P., Sbenghe, M., Nwadiaro, N., Verma, S., & Yarnold, P. Journal of General Internal Medicine (1999). Provider Meeting 4/22/10: Open Access Implementation Notes Major Concerns: Lacking medical assistant support throughout the day The notion that MA s will be loaded with more work Too busy Not enough patients being offered Same Day Appointments Appointment blocks are incongruent with designated appointment time Concerns about having to work late Concerns about seeing too many patients in a day Schedule looks full, but it is not Concerns about sending patients a contradicting message Too much of a change Added work for Care Team Advanced Access Responses 1 & 2.) Some MA s are spending a great deal of time calling patients to make appointments. If Mabel schedules half of the appts and the MA schedules the other half, and 1/3 of patients are No Shows, then a significant amount of the work the MA has done is in vain! For further clarification, MA s will not be hindered with appointment making every appointment. This will actually free up some of the MA s availability to work efficiently in other areas of concern. 3.) Having open schedules is the goal. Upon successful implementation and workload efficiencies, we will have an opportunity to get ahead of demand and have ample time for other duties. This design isn t intended to keep providers schedules maxed out. Its intent is for better continuity of care and increased access. The notion of moral hazard is misplaced in this context. Just because we offer more access doesn t necessarily mean we will be seeing more patients. 4.) In the past few weeks, there has been no room to add Same Day pts to the provider schedule. Having pts call in the same day will yield a plethora of empty time slots. Additionally, with revised time designations, providers will have more time to provide the level of care they see fit. The only time we have been able to offer Same Day appointments is when the workload is light, which contradicts #3. 5.) & 8.) As of yesterday, 4/23/2010, it has been communicated to the front that 20 min appointments will no longer be used. Appointments will be either 15, 30, 45, and 60 minutes. This, in conjunction with provider appointment editing will make availability more transparent. 6.) We can implement a policy that limits your time for appointments (e.g. 9am 5pm 4:30pm cutoff). We are not adopting SCF s policies. 7.) In order to regulate patient volume, adjustments to appointment lengths would regulate how many patients can be seen that day. For example, 30min appointment can be scheduled for 45min. in an effort to see less patients if needed. 8hrs 16pts (30min/pt) 8hrs 11pt (45min/pt) 9.) & 10.) It s difficult to believe that we re actually sending any message. Our volume records do not indicate any sort of variance in regards to scheduled appointments vs. Walk Ins/Same Days (due to recent circumstances we have not been able to offer Same Day appts). For the last 4 months (before Care Teams, Empanelment, etc), we have not seen ANY variation in the schedule. The average variation has been minimal. There are two types of providers. Those that love the Open Access Design and those that don t know they love the Open Access Design. Anonymous/J.Huhndorf Solutions that can be immediately implemented: 20 minute visits will no longer be used (only 15, 30, 45, and 60 minute visits will be assigned) Providers can edit their appointments that were shorter than scheduled when they are ready. This will indicate that they are available for Same Day appointments, if needed. 3
Dena ina Health Clinic s High Leverage Changes Empanelment Advanced Access High Continuity Care Teams Improved Continuity of Care Dena ina Wellness Center 100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% 100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% Care Team A Continuity Care Team C Continuity 100.00% 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% 700 600 500 400 300 200 100 0 Care Team B Continuity Patient Volume Jan '10 Mar '11 52,000 square feet Completely integrated medical, dental, behavioral, and wellness all sharing a common Electronic Wellness Record (EWR) You can roll a number of independent players up into a single organization horizontal integration to generate economies of scale. Or you can bring the treatment of chronic disease under one roof vertical integration and make the treatment more effective and convenient. Regina E. Herzlinger (Harvard Business Review) Dena ina Wellness Center The greatest mistake in the treatment of diseases is that there are physicians for the body and physicians for the soul, although the two cannot be separated. Plato Wellness Person Dena ina Health Clinic: Getting national attention for clinic redesign successes 2010 Tribal Best Practices Conference in Minnesota 2011IPC Foundations Series on the topics of Empanelment and Care Teams 2011, In partnership with the Institute for Healthcare Improvement, presented to private sector clinics through Achieving Excellence in Primary Care Series 2011 Alaska Area Improvement Support Team meetings 2012 Alaska Center for Rural Health 2012 North Dakota Center for Rural Health CMS Healthcare Innovation Challenge participant (results pending) 4
Determinants of Health 101 Determinants of Health + DWC Law of the Instrument Maslow, The Psychology of Science, 1966 Physicians and mid level providers may no longer be the quarterbacks Almost all the literature about the Patient Centered Medical Home talks about a physicianled team If all you have is a hammer everything looks like a nail. What if physicians and mid level providers were simply wide receivers? DWC Care Team 5
Medical Dental Wellness Behavioral It s we who move the world and it s we who ll pull it through Henry Reardon (in Ayn Rand s Atlas Shrugged) Recap: The Dena ina Wellness Model of Care D Dynamic Care Teams: Alternating lead roles depending on most appropriate scope of services E Electronic Wellness Record (EWR) shared amongst integrated primary care services N Network of Tribal services incorporated to address the Social Determinants of Health A All patients are empaneled to a Dynamic Care Team ' I Integrated primary care services include Behavioral Health, Wellness Programs, Medical, and Dental N No longer place "Medical Provider" in quarterback role; allows members to work at top of licensure A Advanced Access scheduling model allows for increased access and patient centered care No one has to change. Survival is optional. W. Edwards Deming How wonderful it is that nobody need wait a single moment before starting to improve the world. Anne Frank 6