White Paper

Use of Patient-Reported Outcomes in Physical Therapist Practice: Report of the Industry Stakeholder Collaborative Task Force on Outcomes Instrumentation

Background

As health care continues to shift toward value-based payment models, it is incumbent on physical therapy practices to incorporate patient-reported outcomes into their standard process of care delivery. Patient-reported outcome tools (PROs) are survey-based instruments that capture the patient and/or caregiver’s perspective regarding health (or disability) status or quality of life. A large number of PROs are available for use, but they vary widely in terms of their reliability, validity, and administrative effort to implement. Payers are already adopting programs that require use of one or more PROs for quality measurement in value-based payment models (e.g., CMS).

Given the current environment, it is essential for the physical therapy profession to take a leadership role in determining what PROs are most appropriate for a variety of functional and operational measures, scalable to any size practice, practical, affordable, and clinically relevant to inform care decisions and support participation in value-based payment models. A level of PRO standardization also is needed across organizations and practices to facilitate outcomes comparisons, establish benchmarks, and populate registries.

In August of 2015, representatives from the physical therapy industry started a dialogue about the establishing outcomes measures for the industry. This joint group agreed that there was a need to help physical therapists in outpatient practice sort through the large number of PROs in an efficient and informed way. A task force (TF) was formed to explore potential PROs that could be adopted in a standardized fashion to better position the physical therapy profession for participation in value-based payment programs. The TF initially was comprised of participants from the industry stakeholders group. Additional TF members were added as the work evolved to provide specific expertise. This white paper describes the TF’s purpose, scope of work, and recommendations regarding PRO adoption.

Purpose

The purpose was to identify a PRO or a set of PROs that would serve as the industry standard for outpatient physical therapist services. These PROs would be proposed for incorporation into registries to drive payment reform and for value-based payment programs adopted under the Medicare Part B benefit as well as by commercial payers.

Composition of the Task Force

The TF was comprised of physical therapists from several organizations who have expertise managing musculoskeletal and neuromuscular conditions. Additional expertise in research, payment, and/or health policy also was contributed. Appendix A lists those on the TF and their affiliations.

Scope of Work

1. Identify criteria for potential PROs.
2. Determine process for researching PROs.
3. Identify how outcomes data will be integrated with current procedural terminology (CPT) billing codes.
4. Assess clinic ease of use for recommended PROs as well as their impact on clinical reasoning and decision making.
5. Evaluate existing registries and their capacity for integrating PROs.
6. Draft a plan for integrating PROs into a registry.
7. Determine payer readiness for integration with, and use of data from, recommended PROs.
8. Establish recommended PROs and prepare a proposal for APTA and Industry Stakeholder consideration.

Operational Construct for Task Force Work

The TF set forth the following points as an operational framework to help specify and target goals of production:

1. Offer a formulary of tools.
2. Provide a methodology by which to evaluate said tools (i.e., Scorecard). The result is a helpful tool that anyone can use to inform their decision. The TF will filter out 80 percent of the universe of instruments down to what may work best for a broad contingency of potential physical therapy users (e.g., American Physical Therapy Association [APTA] Registry, American Academy of Orthopaedic Surgeons [AAOS] alignment, free, paid for tools, etc.), thus making their decision unbiased, agnostic, and less daunting.

3. Develop guidelines for payers for the purpose of managing providers who use the pay-for-performance process: First, providers must collect outcome data in the context of number of treatment visits in order to make objective clinical management decisions. Without measures, any system becomes subjective, which renders discussions regarding patient care decisions between payers and providers almost useless.
• Second, outcome measures should have published psychometrics of reliability and validity.
• Third, the measures should be comparable between providers, clinics, and geographical locations in logical ways.
• Fourth, the outcome measures and number of treatment visits should be risk adjusted, which facilitates meaningful interpretations based on patient case-mix.
• Fifth, the measure of treatment effectiveness in tandem with the number of treatment visits should be tracked throughout treatment, so clinicians, as well as the payer, can see how the patient is progressing.

Scope of Work Deliverables

To address the scope of work, the TF developed a matrix in an Excel spreadsheet format to assess and compare PROs that could apply to various practice settings paid under the Medicare Part B benefit. In addition, a scorecard was added so that users from any size practice could evaluate PROs by their various qualities identified in the matrix in order to efficiently come to an informed decision as to what best fits their individual practice’s circumstance. The scorecard uses objective criteria to assign a score to each instrument to rank order the instruments for high-level comparison and critique.

Identification and Selection of Patient-Reported Outcomes

One of the primary purposes of the group was to evaluate the literature in search of PRO instruments used in physical therapy and orthopedic rehabilitation. Such a task required a tremendous amount of research. A spreadsheet was constructed to begin organizing what was found (see the initial tool count by category in Table 1).

Aligning with other health care professions engaged in value-based payment models also was a priority for the TF, particularly for positioning physical therapy for bundled payments. The PROs adopted by the American Academy of Orthopaedic Surgeons (AAOS) as well as the PROs endorsed by the American Medical Association (AMA) were reviewed. In addition, review of PROs used in skilled nursing facilities was initiated in anticipation that these tools may be used by physical therapists providing Medicare Part B services in this setting.

The subgroup responsible for this aspect of the work presented their initial findings to the TF members who then developed what has become the model herein. Once a comprehensive list of tools was collected, the TF reduced the list to a more manageable size from which recommendations could be made. The group decided to include the most well-known and utilized PROs in the revised list. They also discussed what information would be helpful to know about each tool in order to determine whether it should be recommended. They decided on the following (asterisks denote which criteria were considered most important):

• Source/URL for the tool
• Number of items on the tool*
• Age range addressed by the tool
• Integration into care
• Software required or not*
• Registries/orthopedic associations that use the tool
• Included in AAOS Recommended Measures or not
• Included in APTA PT Outcome Registry or not
• National Quality Forum (NQF) endorsed or not
• Used in MD/PT outcomes systems or not
• Ease of use by practice:
• Time*
• Cost*
• Clinical utility/decision making
• Ease of use by patient:
• Time
• Number of translations
• Cultural bias
• Ability to understand results
• Instrument psychometrics*
• Reliable/valid*
• Minimal clinically important difference (MCID) established
• Responsive
• Payers that utilize the tool

The final spreadsheet matrix listed the reduced set of PROs with documentation of the status of each characteristic listed here when such information was available. Endorsement of a PRO by the National Quality Forum (NQF) was noted when present because the endorsement process evaluates reliability, validity, and feasibility and has been used as a criterion for adoption by the Centers for Medicare & Medicaid Services (CMS) for inclusion in pay for performance programs. The spreadsheet also indicated if a PRO is used by any existing registry. There are multiple registries that would apply to physical therapy, and many have started collecting outcome data through various instruments.

Physical Therapy Industry Survey on PRO Utilization

The TF determined that a baseline assessment of current PRO use in physical therapist practice was needed to inform recommendations about instrument selection.

To assess current practice, the work group developed a 20-question survey evaluating the outcome tools used, implementation, and ways that the data was purposed. Between June 22 and August 15, 2016, the survey was distributed to all APTA Section membership, Alliance members, Therapy Partners members, among other groups. One thousand and one respondents, 91 percent of whom were APTA members, had complete surveys. Survey respondents were 63 percent clinicians and 27 percent administrators with practices ranging from 2 to 500 plus clinics and representing all 50 states and Puerto Rico. The practice type was primarily outpatient orthopedic and subspecialties such as hand therapy, aquatic, and women’s health (94 percent), with less than 10 percent comprising acute, short- and long-term care facilities. Results showed that 82 percent of PROs were collected at initial evaluation and 64 percent at discharge. Paper-based surveys accounted for 62 percent of all surveys, making it the primary method for collection.

The most striking result of the survey was the discordance across practices for outcome utilization. There was little consensus among practices on use of patient-reported outcome measures. With regard to global or health outcome surveys, less than 13 percent of all users reported using any global health measure. The FOTO (Focus on Therapeutic Outcomes) Global Measure was most used at 14 percent with PROMIS 10, VR-12, and EQ5D all garnering less than 3 percent. The Oswestry Disability Index (ODI), Neck Disability Index (NDI), Disabilities of the Arm, Shoulder and Hand (DASH), and Lower Extremity Functional Scale (LEFS) ranged from 52 percent to 74 percent. These findings were consistent with actual use data in 2015 from 146 outpatient practices distributed across 31 states representing 2,452 therapists (# of unique patient administrations: LEFS = 14,769, Oswestry = 5044, NDI = 4780, DASH = 3650) (source: BMS Practice Solutions, Upland, CA). No other regional or disease-specific outcome measures exceeded 20 percent. (See Figures 1A-C.) Patient outcomes data is primarily used for quality improvement and Medicare reporting (see Figure 2).

Payer Survey on Outcomes Use and Value-Based Payment Programs

The scope of work for this TF included assessing the use and adoption of PROs within the commercial payer industry. The work group developed a survey to gauge the level of interest and participation of payers in using PROs for outpatient physical therapy, as well as their initiation of value-based pilots for payments of outpatient physical therapy. The survey was sent to known contacts in the payer industry, including provider network managers, medical directors, and operations managers. The respondents had an average of 24 years in the industry. Responses from six different health plans were received and compiled.

All of the health plans hold the expectation that physical therapy providers should be measuring and reporting clinical outcomes on the care they provide. Four of the six health plans are currently requiring physical therapy providers to report outcome measures. Two of the six health plans will use outcome data to potentially exclude physical therapy providers from their network. Most of the health plans will use outcome data to incentivize providers in the network.

When asked if they are currently using time-based Current Procedural Terminology (CPT) codes and if this system should remain unchanged, 50 percent of the respondents “somewhat disagree” and 16.7 percent “strongly disagree.” When asked if the current CPT codes for physical therapy should be merged or collapsed, the response was 83.3 percent “somewhat disagree.” When asked if physical therapy should be billed using a single code per visit, the responses were equally weighted between “strongly disagree,” “somewhat disagree,” and “somewhat agree.” Interestingly, when asked if reimbursement should be higher for more complex patients, the payers were indecisive in their response. More than 50 percent of the health plans were executing value-based payment with providers, and 33 percent of the health plans are currently executing value-based payment pilots, specifically with physical therapy providers.

With this information, the group could assume that some payers are interested and moving toward using clinic outcomes to measure quality of physical therapy services. In addition, the value-based reimbursement pilots for payment of physical therapy services have begun and will continue to expand in the future for these health plans. Coding reform alone is of little impact to these payers. For this group of payers, measuring and reporting quality outcomes is expected, and potentially influencing how they manage their networks. It would be beneficial to repeat this survey biannually, and attempt to expand the number of respondents to encompass additional geographical areas and a wider variety of health plans. The work group would consider issuing the survey again at the start of the New Year.

Task Force Recommendations

Despite widespread reported use of PROs, there is a startling lack of use of global measure of health status tools, and no clear consensus on which instruments should be used. The prevalence of paper-based administration challenges aggregation and reporting of these measures on both clinical and national levels. There is an urgent need for identification and agreement on a core set of tools including a global and regional outcome measure. This will set the foundation to generate data needed for development of performance-based measures beyond process-based measures.

The results of the physical therapy industry survey and the collective experience of the TF suggest that this lack of consensus among practicing clinicians places the profession at risk as value-based payment models evolve. In our opinion, adoption of a limited formulary of PROs that provide meaningful assessments of patient function at the patient, clinician, and aggregated clinic levels are critical to demonstrate physical therapy’s value. When considering the limited scope and availability of current outcome-based (as opposed to process-based) performance measures by organizations such as NQF, the cost of implementation and ability of these measures to communicate across physical therapy settings and within the health care arena are important when selecting tools for use.

Thus we suggest a formulary approach to selection of best-in-class PROs (Table 2) that should be considered across the profession. We recommend using (at minimum) a global measure of health and regional outcome measures (i.e., back/neck/knee) at initial evaluation and discharge to assess patient response to treatment. This standardization would provide the opportunity for scaled aggregation and development of PROs as performance measures as well as provide context for risk adjustment and stratification approaches with payers.

Case Studies

To demonstrate how outcome measures may be used in the emerging health care markets, we have provided two case studies (Appendices B, C, and D) that demonstrate approaches for PROs in payer relationships. These case studies can be used as potential models for demonstrating return on investment and provide a framework for organizations to evaluate and implement needed changes to prepare their clinicians, clinics, and the profession for the quickly changing health care market.

Conclusion

The emerging health care arena is rapidly moving to value-based models of payment. The current dynamic environment places significant stress on providers and the physical therapy profession to produce data to drive payment reform beyond time-based billing and process-based quality. The first step toward this goal is alignment of PROs that show promise of developing into performance-based measures. Additionally, global measures of health, along with other demographic and medical information, are critical to the risk adjustment and stratification needed for functional pay-for-performance models. It is urgent that the industry consider ways to nationally aggregate patient demographics, PROs, and physical therapy utilization data to develop best approaches to pay-for-performance models.

Appendix B: Case Study, PT Northwest

Case study on clinical integration of therapeutic outcomes

The players in this case study include a physical rehabilitation group, PT Northwest (PTNW), and the rehabilitation arm of a large health system, Intermountain Healthcare (IH). PTNW is a privately owned, independent physical rehabilitation organization:

• 11 outpatient physical rehabilitation clinics, physical rehabilitation services provided to 2 local hospitals, Athletic Trainer Certified (ATC) coverage for 3 universities, 5 high schools, and 2 minor league sports teams
• Employs 115 individuals consisting of physical therapists, occupational therapists, speech-language pathologists, and athletic trainers
• All locations in Oregon
• Named an Oregon Top Workplace by the Oregonian newspaper

IH is a not-for-profit health system:

• Based in Salt Lake City, Utah
• 22 hospitals and a broad range of clinics and services
• Employs about 1,400 primary care and secondary care physicians at more than 185 clinics in the Intermountain Medical Group and health insurance plan from SelectHealth

In October 2013, after months of collaborative discussion, PTNW launched a beta version of IH Rehabilitation Outcomes Management System (ROMS). IH began development of ROMS in 2000, implementing it into practice in 2002, and with modest updates along the way they have integrated use of ROMS across 54 of their outpatient orthopedic physical therapy practices to date. IH’s partnership with PTNW would be the first attempt to trial commercialization of ROMS. Integral members from IH provided our therapists with training on published treatment-based classifications, and clinical support staff were trained on operation and processes surrounding ROMS. Use of ROMS was implemented across all of our outpatient clinics for PT and OT patients being seen for musculoskeletal-related conditions. Every new patient with neck or low back–related complaints is classified according to established clinical treatment–based guidelines. Classifications for extremity-based conditions are pathoanatomical in nature. Patient-reported outcomes measures (PROMs) are collected at each visit and are regionally specific, including the Modified Oswestry Disability Index (MODI), Neck Disability Index (NDI), Disabilities of the Arm, Shoulder and Hand (DASH), Lower Extremity Functional Scale (LEFS), and the Knee Outcome Survey (KOS). PROMs are aggregated across a patient’s stay and at discharge can be assessed collectively across regional outcome scales, by classification, by clinic, by therapist, and/or by referring provider. All outcomes were open and accessible company-wide.

After the initial training sessions provided by IH employees, open training sessions were held to review routine data processes and site navigation as well as detailed reports to all therapists. Clinic director and support staff were provided with support as needed for site navigation, but initial attention remained solely focused on data collection. Through this we established clearly defined benchmarked metrics for ROMS admission rate, classification rate, and discharge rate. These metrics were a minimum threshold of 100 percent for admission rate, 90 percent for classification rate, and 90 percent for discharge rate. Support staff and clinic directors were held accountable to these metrics, and we soon implemented monthly reports on these metrics. All focus was placed on getting our support staff to admit patients into ROMS, provide and score the PROM at each visit, record the PROM in ROMS, have the therapist complete the regional classification form at initial evaluation, and finally discharge the case from ROMS when their case is discharged from therapy. While yes, this process just seemed like more work at first, over time as processes improved, the value in the information being received and the predominantly positive patient feedback has dampened any initial frustration from the added work. The consistent message remained as resources and training were provided to ensure we got the patient entered in, classified, outcome forms completed and entered, then verifying that their case was discharged. If a case does not have more than one outcome measure, the patient is not classified, or someone fails to discharge the case appropriately from ROMS, the outcome data is not populated into the data pool for analysis and becomes mute.

While emphasis was placed on data collection, we began to see the hopeful consequences of objectifying a patient’s response to treatment and overall progress in care. Many therapists began to use these tools on a more frequent basis not only for goal setting but also for engaging the patient on difficult aspects of their daily life gleaned from the outcome tool displays. The tools have also opened conversations with the patient on how we are looking to objectively monitor their progress and response to therapy while they are under our care. Being able to share with the patient daily outcome reporting in graphical representation has been an incredibly useful aid in particular with conversations on a plateau in treatment response and discharge planning. Our therapists have been empowered when seeing the change before them; they are able to use a patient’s response as a means to stay the course, adjust the treatment plan, or prepare for discharge. As data has been aggregated, the therapists have been able to then visualize their overall clinical outcomes as a group and within specific classifications. This has provided them with firsthand objective knowledge of their own clinical outcomes. Areas of clinical success may push some toward being a champion for others in the selected or group diagnosis/classification. Areas of deficit have provided valuable information to not only the therapist but to PTNW as a whole. We have brought in continuing education courses for the entire group based on these deficiencies and have been able to guide individual therapist in-house training/support and external continuing education based on this. While there may have been some clear concern from treating therapists regarding judgment by superiors and/or peers, the global message is that we are collectively looking at every avenue to improve, and the primary concern is that we are collecting and recording data first. More than anything this has empowered our therapists to improve and be able to see the improvement objectively within ROMS.

Implementation of ROMS has continued to open doors for us with local providers and practice groups. We have been asked to collaborate on a low back pain (LBP) pilot project with a large physician network. Discussions on our use and transparency of clinical outcomes garnered us a seat on an LBP and opioid task force committee for a regional Medicaid Coordinated Care Organization. Not only has the consistent use of clinical outcomes provided us these opportunities, but also the transparency with which we approach them as well as the benchmarked data to a well-respected health care system, IH, has earned us repeated praise as a key differentiator among our group.

1. Increased patient volume each year, providing 10.0 percent more initial evaluations while seeing a reduction of 18.3 percent in follow-up visits per case. PTNW visits per case are now 41.6 percent under the national average.
2. Improved clinical outcome change score for 4 out of the 5 aggregated clinical outcome measures across each year since ROMS implementation (see Figure 3).
3. Reduced overall number of visits (see Figure 4) and length of stay (see Figure 5) trending each year since ROMS implementation.
4. Sustained high levels of patient satisfaction rating at 97.3 percent at discharge while collecting a discharge satisfaction survey return rate of 91.5 percent.
5. Employees have responded positively to our clinical outcomes approach. Oregon’s largest newspaper, The Oregonian, named PTNW an Oregon Top Workplace in 2016.

It was clear from the inception of ROMS at PTNW that we had clearly raised the bar. No longer would we stand by with our clinical finger in the air attempting to monitor the clinical outcomes of our patients, therapists, or clinics. We are all-in as a company, and our therapists are more engaged with their patients on a day-to-day basis. This has led to improved outcomes in less time and visits without affecting our patient satisfaction. We believe these results are from our practical application of the value equation by Michael E. Porter and Thomas H. Lee, MD, from their article in the October 2013 Harvard Business Review, “The Strategy That Will Fix Healthcare,” the value equation being the best outcome for the patient at the lowest cost. We feel that using ROMS as a platform to engage our patients and improve therapist-patient communication has been essential to these improved metrics. Therapists are more informed and better able to make improved clinical decisions regarding treatment provided and discharge planning while also becoming more knowledgeable about their own clinical strengths and weaknesses. Our clinic directors are able to more objectively engage with employed therapists regarding patient care and outcomes to provide therapists with different levels of coaching moments, from in-house clinical in-service skills review to yearly goal setting. Our company has been impacted tremendously and raised the bar in care among our therapists. We have further been able to foster new provider and payer relationships and solidify current relationships through transparency and collaboration as we all work to maximize value for the patients we serve. This approach should well prepare PTNW for the shift from a fee-for-service payment model to an anticipated value-based payment model.

The next step for us, while not losing sight of the established processes and patient engagement, will be to look for additional avenues in which to positively challenge our therapists and implement more quality assurance measures to improve adherence to clinical practice guidelines and minimize the degree of variability among our treatment therapists.

We all know we should be collecting clinical outcomes, but are we? If you are, then what are you doing with it? How are you assessing outcomes across your company, within and between clinic(s) or among therapists? How are you engaging your patients in their functional progression? How are you justifying care and functional gains objectively to your referring provider or payer? Implementation of clinical outcomes and management/tracking software allows you to answer these questions and so many more. While yes, the collection and data entry take time in themselves, the results are powerful and the time is well spent. With clear processes in place designed to meet the needs of your facility/company, anyone can begin the path and raise their level of excellence. This investment is worth it. The results are improved clinical outcomes, improved patient engagement and satisfaction, improved relationships with referral sources and payers, and ability to clearly demonstrate better overall value.

Kennedy Hawkins, MBA, JD, LLM
President of PT Northwest
Past President, National Association of Rehabilitation Agencies & Providers

Joshua Christopherson, DPT, Cert. SMT
Director of Clinical Outcomes & Care Pathways, PT Northwest

Appendix C: Case Study, Intermountain Healthcare

Title: “Reducing the percentage of patients who fail to improve when implementing a standardized care process model in the treatment of patients receiving total knee arthroplasty”

In 2002 a large, integrated, not-for-profit health care system implemented a process for collecting patient-reported outcomes (PROs) on all patients treated in their outpatient orthopedic physical therapy clinics. For patients with knee pain, the Knee Outcome Survey (KOS) was selected as the PRO measurement tool. The minimal clinical important difference (MCID) with the KOS has been determined as 10 points of improvement. Patients who fail to meet the KOS MCID during their episode of physical therapy treatment are deemed patients who “fail to improve.” The percentage of patients who fail to improve was calculated on all patients in postoperative care for a total knee arthroplasty (TKA) for the largest commercial payer in the state. This percentage was tracked annually and a goal set to make a statistical reduction in this metric in 2013. If therapists met the goal, they would receive a financial payment from payer. At the time, there was not a standardized process for treating TKA in this health system. Although some improvement was made, the goal was not met. A standardized care process model (CPM) was implemented in the treatment of TKA patients in 2014. The implementation of this CPM included provider education, support of additional evidence-based treatment options, and a compliance program.

Provider education initially involved a two-day educational session with an expert in the treatment of postoperative TKA and the introduction of the care process model. The TKA CPM included evidence-based treatment guidelines based on impairment limitations such as quadriceps strength and knee range of motion. Treatment options not previously used were introduced, including functional electrical stimulation when a knee extensor lag was present and the use of a “bag hang” producing long duration stretching when full passive knee extension was not obtained. Follow-up educational sessions were scheduled throughout the year to reinforce the initial expert education. Functional electrical stimulators and bags for the “bag hang” exercise were placed in each clinic. Nonuse of passive modalities was also part of the CPM. Compliance was determined by documenting treatment according to the CPM, and chart audits on 100 percent of the patients were performed in the 4th quarter of 2014 to verify the results. A compliance standard of 90 percent was set as the goal.

Failure to improve with the patients treated according to the CPM showed a statistical improvement in 2014 with a further reduction in 2015. Improvements have been maintained in 2016. Although not part of the goal nor an emphasis of the CPM, the number of visits per episode correspondingly decreased, improving the value of the therapy provided. The compliance audit revealed that 93 percent of therapists were following the CPM.

This case study supports implementation of a standard care process model in the postoperative treatment of patients who have received a total knee arthroplasty and the use of an outcomes system to document the results. Providing tools to implement the standardized model and setting up an audit process to verify results are also key components.

References

1. Meier W. et al. Total knee arthroplasty: muscle impairments, functional limitations, and recommended rehabilitation approaches. J Orthop Sports Phys Ther. 2008;38(5):246-256.

2. Mizner R, Petterson S, Snyder-Mackler, L. Quadriceps strength and the time course of functional recovery after total knee arthroplasty. J Orthop Sports Phys Ther. 2005;35(7):424-436.

3. Petterson S, Snyder-Mackler L. The use of neuromuscular electrical stimulation to improve activation deficits in a patient with chronic quadriceps strength impairments following total knee arthroplasty. J Orthop Sports Phys Ther. 2006;36:678-684.

4. Stevens-Lapsley JE1, Balter JE, Wolfe P, Eckhoff DG, Kohrt WM. Early neuromuscular electrical stimulation to improve quadriceps muscle strength after total knee arthroplasty: a randomized controlled trial. Phys Ther. 2012 Feb;92(2):210-26.

5. Piva SR1, Gil AB, Almeida GJ, DiGioia AM 3rd, Levison TJ, Fitzgerald GK. A balance exercise program appears to improve function for patients with total knee arthroplasty: a randomized clinical trial. Phys Ther. 2010 Jun;90(6):880-94.

6. Bade MJ1, Stevens-Lapsley JE. Early high-intensity rehabilitation following total knee arthroplasty improves outcomes. J Orthop Sports Phys Ther. 2011 Dec;41(12):932-41.

7. Brennan GP, Fritz JM, Houck L.T.C. K, Hunter SJ. Outpatient rehabilitation care process factors and clinical outcomes among patients discharged home following unilateral total knee arthroplasty. J Arthroplasty. (2014), doi: 10.1016/j.arth.2014.12.013.

Submitted by:
Stephen Hunter, PT, DPT, OCS

Appendix D: Case Study, HealthPartners

The players in this case study include regional payer HealthPartners (HP) and physical therapy providers of Therapy Partners (TPI). HP is a nonprofit, consumer-governed health plan:

• 1.5 million medical and dental plan members
• Regional network of more than 148,000 doctors and other care providers in Minnesota, western Wisconsin, South Dakota, and North Dakota
• HealthPartners and Cigna’s combined national network offers nearly 950,000 doctors and other care providers, plus 6,000 hospitals in the United States
• Ranked among the top 30 plans in the nation according to NCQA’s Private Health Insurance Plan Rankings 2013-14

TPI is a management service organization:

• 15 independent therapy practices
• Single-source third-party contracting with all state, regional, national, government payers
• Clinics in Minnesota and western Wisconsin
• 104 clinicians providing over 170,000 patient visits annually
• Value-based contract with HealthPartners since 2010
• 2011 Excellence in Innovation Award by HealthPartners

Establishment of Conservative Care Spine Network: In 2003 HP established a Conservative Spine Care Network to run a pilot on providing effective conservative care for patients with low back pain. Guidelines for the program were established, notice and applications were sent to physical therapists, chiropractors, and nonoperative physicians who had payer contracts with HP. Approximately 30 groups were admitted to the network, which included hospital-based, private practice, chiropractic, and physician clinics. Measurement of functional outcomes was required using a Modified Oswestry Low Back Pain Disability Index (ODI), which was captured at the clinic and faxed to HP. Providers were free to market their participation in this special network.

Five-Year Pilot Program: The network met twice a year to review results, which were at first blinded, but then unblinded to the whole network. During the meetings the data capture process was tweaked, best of practice ideas were shared, and feedback by clinicians and patients was given. The meetings were well attended, because practices wanted to appear engaged and learn more about other practices that were their competitors. Compliance with the reporting requirements of the program was satisfactory or high as assessed by HP. HP promoted the network to their own physicians to encourage referrals to the special network providers. A cohort group was also assessed and compared to the network on volume and utilization.

Pilot Program Results:

1. During the five years, HP and the specialty network did not see a significant increase in volume of HP members accessing the specialty network.
2. Utilization of therapy visits was no different between the specialty network and the cohort group.
3. ODI outcomes did not improve over time by the specialty network, at either the practice level or network level.
4. ODI outcomes were similar across all practices in the network, which provided no differentiation between practices. Even though HP knew different models of care were being employed between the practices and clinicians, one model did not stand out superior to another in terms of measurement using ODI.

Summary Statement by HP: “After five years of detailed analysis of Oswestry results, HealthPartners cannot determine any difference in the quality of outcomes of providers of spine care.” The Conservative Spine Care Network was disbanded. HP has since gone on to addressing the high frequency of spinal fusion within the Twin Cities market and established a policy and program titled the Medical Spine Center. The program has elements of the previous spine network but creates an adversarial relationship between surgeons and nonop providers, and patients and conservative care providers, and has not aligned the interests of providers and patients.

New Pilot with TPI in 2010: HP continued to seek solutions to measuring and understanding quality as it relates to conservative care, particularly by physical therapists. They realized that ODI did not provide them any ability to determine who are the better quality providers. Traditional UM programs did not include outcome or quality metrics and were costly to administer. TPI had migrated to reporting outcomes using FOTO in 2008 and introduced HP to FOTO by reporting FOTO scores in place of ODI scores during the years of the specialty network program. Intrigued by the large national database, research, risk adjustment, and independence of FOTO, HP reached out to TPI in 2009 to do a pilot using FOTO outcomes as the quality outcome measure. The actual pilot began in 2010 and ran for 12 months.

HP has per diem contracts with all independent physical therapy providers and increased the base per diem rate of TPI to offset the cost of using FOTO and additional meeting time to work with HP on refining the process. During the pilot, HP, TPI, and FOTO met quarterly to review outcomes and refine benchmarks. HP considered the pilot a success, because they now had outcomes data and not just claims data; plus, they had a provider group that stayed engaged with the pilot and worked with them to refine it. TPI considered the pilot a success, because for the first time in their history, and at that time the first time in physical therapy, a value-based contract, based on outcomes, formed the basis of the payer contract.

Win-Win: HP has now made this model a contract option for independent physical therapy practices. HP is interested in evolving this model into other payment methodologies and will continue to embrace outcomes and reward value with improved payment. TPI has a strong relationship with HP and has negotiated better paying contracts each year.

Conclusions: This case study demonstrates that payers are interested in determining and rewarding the better providers. Outcomes are not a solution to the payer’s questions about who are the better providers in a market or a network. But outcomes with a strategy to align incentives of providers and affect payment can be a successful model between provider and payer. It takes an interested payer and provider to determine the model that works best for both parties. It will not be a successful model if the payer wins and provider loses; in the end the payer does not really achieve its goals. Payers are not interested in solving small cost issues, which is what PT is in their budgets. We as providers need to show the payer a model of care in which they win and we win. We must go to them.

Submitted by:
Craig Johnson