Question / Intervention Goal / Objectives

Alight Solutions’ Healthcare Navigation directs insured plan members to lower-cost providers who meet quality standards. The program’s goal is to reduce medical costs while maintaining high quality of care.

 

Method/ Calculation

The analysis included 75,613 cases in which a plan member had a medical service provider selected when he/she consulted with the program. The allowed amount for the member’s originally selected provider was then compared to the allowed amount for the provider he/she ultimately chose. (See Table 1.) The analysis includes cases where the member did not change providers after consulting with the program.

The allowed amount comes from the participating employer and thus represents what the plan would have paid. The amount ultimately paid came from medical claims data.

 

Findings / Metric/ Outcome/ Savings

The allowed amount for the providers used by 75,613 members who consulted with the program was $102 million less than the provider they originally selected. This is a 20% savings.

 

Limitation

The savings are gross, not net of Alight’s program fees. Quality of care is not addressed in this analysis.

Table 1

Cases	Allowed Amount For Original Selected Provider	Confirmed Payment to Provider Ultimately Selected	Savings / %
75,613	$513,840,645	$410,947,602	$ 102,893,043/ 20%

 

Question / Intervention Goal / Objectives

AllergyChoices offers disease-modifying treatment of allergies.  The course of sublingual immunotherapy (SLIT) can eliminate the need for future allergy treatment.  Patients who complete the treatment have improved quality of life.

 

Method/ Calculation

This study was a prospective analysis that compiled and compared collected Juniper Rhino-conjunctivitis Quality of Life survey data from patients undergoing sublingual immunotherapy. Collected data for each patient were compared to that particular patient’s baseline data and the same data from two subsequent patient visits.

 

Findings / Metric/ Outcome/ Savings

Two hundred and ninety-nine patients who had an average of 11 months of treatment completed the Juniper questionnaire at baseline, and at two follow-up visits.  They showed statistically significant improvement in the six of the seven domains of quality of life scores from baseline to the end of treatment.  Improvements were seen in the activities, non-nose/eye symptoms, practical problems, nasal symptoms, eye symptoms, and emotional categories; improvement was not statistically significant for sleep problems.

 

Limitations

Adjustments were not made to weigh the self-selection bias among respondents.

 

Question / Intervention Goal / Objectives

Apta Health brings care coordination services to groups with fewer than 1000 lives. Care Coordinators interact with members when they are making choices, guiding them to proven programs and efficient providers. The real-time support strives to eliminate duplication of services and re-admissions. Apta clients had annual cost growth that is lower than national benchmarks.

 

Method/ Calculation

In a case study of a 300-employee group, the analysis measured the employer’s payment per employee per month costs (PEPM) in 2015 and 2016, the year before the intervention and the year after.  The PEPM cost reflected pharmacy benefit rebates; cost for high-dollar claimants was segregated.

Additional analysis was done on 22 Apta clients’ medical and pharmacy cost data for four years.  The analysis combined the cost data for the years 2016 and 2017 and compared it to similarly combined data from 2017 and 2018.

 

Findings / Metric/ Outcome/ Savings

The 300-employee group had an 11.25% decrease in PEPM costs paid by the employer.  After taking into account payments by a reinsurer for catastrophic claims, the plan in total had a 4.04% decrease is costs.  (One claim was over $500,000.)

The multiple employer analysis showed a negative 3.9% change in medical costs from the 2016/2017 to the 2017/2018 years; this compares to a national trend of 10.3%.  Pharmacy costs had a 13.8% increase for the cited time periods, compared to a national trend of 15.9%.

 

Limitations

The statistical significance of the rate decrease for the 300-employee group was not calculated.

The benchmark cited is the actuary’s own book of business; because there was not a description of this population, it is not clear whether this is a reasonable comparison. Benchmarks from Kaiser Family Foundation 2017 employer Health Benefits Survey[1] show slightly lower medical with pharmacy cost trends than the benchmark cited: 3% for single coverage, 4 – 5% for family coverage for 2017 and 2018.

 

DISCUSSION

Intervention Link to Outcome

• State the outcome being measured.

The outcome is the Per Employee Per Month medical and pharmacy cost for a two-year period of time for the case study.

The annual medical and pharmacy rate increase is the outcome for the multi-employer analysis.

• Detail the intervention.

Care Coordinators interact with members when they are making choices, guiding them to proven programs and efficient providers.

• Does the applicant discuss published literature or other credible source demonstrating correlation between intervention and outcome? If yes, describe the correlation and the source cited by applicant. If no, does the literature exist to demonstrate a correlation between intervention and outcome?  The applicant provides no published literature, though there is a body of literature supporting the efficacy of care coordination.

 

Data Source

• Describe the data source.

The data came from the group’s medical and pharmacy claims.  Data on pharmacy rebates presumably came from other records.

• Did the applicant have adequate data from a credible, reliable source?  Yes.
• How is the data source appropriate for the outcome being measured?  The impact of the care coordination program should be apparent in the group’s total medical and pharmacy costs.  Therefore, using the total costs is appropriate.

 

Methodology

• Describe the evaluation methodology, i.e. trend from pre- to post-, comparing similar groups, etc.

The case-study evaluation compared a pre-intervention to a post-intervention period, blending year one of the intervention into both the pre- and the post- periods.

The multi-employer analysis compared the annual rate increases to a benchmark.

• Did the applicant collect and manage data in accord with standard evaluation methodology? Comment on any issues with compiling the measure, such as missing or incomplete data or lack of data on non-participants.   Yes.

 

• Describe the findings/results.

The 300-employee group had an 11.25% decrease in PEPM costs paid by the employer.  After taking into account payments by a reinsurer for catastrophic claims, the plan in total had a 4.04% decrease is costs.  (One claim was over $500,000.)

The multiple employer analysis showed a negative 3.9% change in medical costs from the 2016/2017 to the 2017/2018 years; this compares to a national trend of 10.3%.  Pharmacy costs had a 13.8% increase for the cited time periods, compared to a national trend of 15.9%.

 

Additional Comments from Reviewer

Blending the first intervention year into the pre- and the post-periods may be understating the impact. 

Score

 

Intervention to Outcome SCORE (0-5)	Data Source SCORE (0-5)	Methodology SCORE (0-5)	Quality of Application SCORE (0-3)	Overall SCORE (0-18)
3	5	4	2	14

 

[1] https://www.kff.org/health-costs/report/2017-employer-health-benefits-survey/

**Winner of the 2019 Health Value Award for Program Provider – Specialty Fertility**

Question / Intervention Goal / Objectives

ARC Fertility provides affordable, high-quality fertility care through its program of predictably priced, bundled, evidence-based treatment packages, refund guarantee and financing options. For employers, ARC Fertility’s fertility benefits incur no Per Employee Per Month or administrative fees, require payment only when services are utilized and deliver extensive educational value and concierge services.

The ARC program results in fewer multiple births than the national average, thereby increasing the number of healthy babies and decreasing the cost of health care for pregnancies and babies. 

 

Method/ Calculation

The applicant compared the percentage of treatment cycles resulting in twin births reported to the Centers for Disease Control and Prevention (CDC) Society for Assisted Reproductive Technology (SART) by all clinics nationally with those reported by clinics within the ARC Network over a three-year time period.

 

Findings / Metric/ Outcome/ Savings

The ARC Network’s cycles consistently resulted in fewer births of twins over three years, compared to the national average. The ARC member multiple birth rate as a proportion of all births declined 3% from 2014 to 2015 (ARC: 12.4% to 10.1%; National: 12.8% to 11.2%) and a further 2.5% from 2015 to 2016 (ARC: 10.1% to 7.6%; National: 11.2% to 8.1%), an overall decline of 5.5%. As a network, ARC Members had reduced multiple births by 5%, which is ½% to 1% better than the national average.

 

Limitations

No adjustments were made to reflect differences between ARC Fertility patients and the SART benchmark population. 

 

Arche Healthcare is a medical device and healthcare service company addressing diabetic foot care, and the $30 billion cost of diabetes-related wounds and amputations. In an increasingly outcomes driven healthcare environment, Arche’s proprietary IT platforms offer healthcare providers, health plans and other healthcare stakeholders, seamless and compliant healthcare delivery tools to help prevent the 1st wound.

The model, informed by the latest evidenced-based research on diabetic foot health, demonstrates the results of a comprehensive Lower Extremity Amputation Prevention (LEAP) initiative and the clinical and economic impact that early risk identification and stratification can have on a diabetes population.

As is the case with “financial spreadsheet validation” according to the rules of the Validation Institute, users may change assumptions in order to model different impacts on prospective results.

 

Question / Intervention Goal / Objectives

Ault International Medical Management (AIMM) provides a team of nurses and doctors who work closely with patients to navigate, select, and coordinate their care.  AIMM helps patients choose high-quality, cost-effective care through its programs for utilization management, care management, maternity, transplants, tele-triage, and acute care.  Members are also identified through claims data mining and predictive risk modeling, so interventions can be deployed before high-intensity services are needed.

Employer clients select the level of care management they desire.  The level of care management is based upon MCG and vary according to the stringency of the controls.  The overall goal is to reduce their Per Employee Per Year (PEPY) cost while improving the patients’ quality of care.

 

Method/ Calculation

The PEPY cost in the second or later year of the AIMM program was calculated, including all employees covered by the plan.

 

Findings / Metric/ Outcome/ Savings

The PEPY figures correlate to the level of care management selected by the client. Of the 18 employers included:

• Three applied the well managed guidelines and had PEPY lower than $3,884;
• Seven applied the moderately managed level and had PEPY higher than the well managed level, but lower than $9,142;
• Three selected the loosely managed level and had PEPY higher than the $9,142 but lower than $13,040; and
• Five had PEPY higher than $13,040.

The PEPY for 13 of the 18 employers was lower than the national average $12,612 in 2018, from a Willis Towers Watson Survey.  Employer size ranged from fewer than 30 to more than 1,500.

 

Limitations

As with all observational studies, other factors that are not measured may be influencing results.  For example, certain types of employers may be attracted to the higher levels of care management.

 

Question/ Intervention Goal/ Objectives

BizMed Blue-Ribbon Direct strives to address employers’ short- and long-term health cost and quality issues by strengthening primary care. Primary care practices that apply the Blue-Ribbon Direct tenets may reduce overall expenditures for medical care. The Blue-Ribbon tenets address access, continuity, safety and comprehensiveness of patient care.

For example, training and processes for referrals to specialists are part of the program; with this in place, primary care practices can make specialist referrals less frequent and more effective.  https://www.bizmedsolutions.com/

The Blue-Ribbon primary care practices receive training, systems, and resources to improve the management of patient care. Specifically, the program goals are to increase use of preventive care, improve management of chronic diseases, reduce referrals to specialists, increase medication effectiveness (e.g. compliance), and reduce the need for urgent, emergency, and hospital care.

 

Method/ Calculation

BizMed offers an illustrative business case, demonstrating how Blue-Ribbon practices may reduce medical expenses. The business case makes reasonable assumptions about what percentage of plan members would use Blue-Ribbon practices. It applies national sources to estimate annual increases in health care costs. It uses a credible source to estimate what portion of total health costs are spent on primary, specialist, and hospital care as well as prescription drugs.

 

Findings/ Metric/ Outcome/ Savings

The business case estimates an overall impact of Blue-Ribbon Direct at 5 to 15% of employer total healthcare costs based on four cost components shown in Table I. An employer group whose members use Blue-Ribbon physician practices would spend 20 to 30% less than it previously did on specialty physician services. Similarly, a group using Blue-Ribbon practices could reduce its prescription drug expenses by 10 to 20%. Use of and expense for primary care would increase.

Limitations

The estimated savings from using Blue-Ribbon practices is based upon literature showing that primary care influences certain costs. For example, Cole et al. found that Patient Centered Medical Home patients had lower inpatient admissions than patients of non-PCMH practices. Literature overall on PCMH and other primary care focused initiatives have shown mixed results. The business case’s estimated impact is higher than studies of similar programs have found, but the Blue-Ribbon program requirements may differ or are more focused. Similar programs have not been shown to reduce the cost of specialist care by 20%, for example.

No formal definitions for the cost items were offered; these would be required in order to measure BizMed’s results from an actual case and compare those to the literature.

 

Question / Intervention Goal / Objectives 

Blossom Bariatrics is a weight loss clinic, along with Warm Springs Surgical Center, an ambulatory surgery center that specializes in bariatric surgery.  The Fast Track Program offers a bundled price that covers its patented Safe Sleeve surgery and coordination of care. The program assists patients traveling from other states with transportation, lodging, and follow-up care.  

The goal of the program is to deliver high quality care at a competitive price with few surgical complications.

  

Method/ Calculation 

For the analysis, Blossom reviewed medical records of all patients (n = 2,233) served between January 1, 2017 and December 31, 2019 to determine how many experienced post-surgical complications. The measure then calculated the ratio of patients having complications to all patients receiving the surgery through the Fast Track Program.  

Complications were defined using the American Society for Metabolic and Bariatric Surgery outcome reporting standards, as published by Brethauer et al in 2015. 

 

Findings / Metric/ Outcome/ Savings 

Of the 2,233 patients, 20 had a post-surgical complication. (See Table I.) The complication rate is estimated at 0.9%. This compares favorably with complication rates published in the literature. In the systematic review and meta-analysis by Chang et al, the complication rate was 17% and the re-operation rate was 7%.  (Note: the rate calculated by Blossom includes both complications and re-operations.) 

Limitations 

Since some patients traveled to receive care, complications that occurred after they returned home may be under-reported. 

 

---

[1] Brethauer et al, Standardized outcomes reporting in metabolic and bariatric surgery, Surgery for Obesity and Related Diseases 11, (2015) 489 – 506

[1] Chang S., et al, Bariatric Surgery: an updated systematic review and meta-analysis 2003 – 2012, JAMA Surg. 2014 March 1; 149(3): 275–287. doi:10.1001/jamasurg.2013.3654

BlueCross BlueShield of South Carolina offers large employer groups a comprehensive population health management program. This program combines case management, utilization management, disease management, wellness efforts, patient centered medical homes and value based incentives. At the end of 2015, there were approximately 379,000 members eligible to participate in some aspect of population health management. Using this integrated approach to population health, BlueCross BlueShield of South Carolina achieved a chronic disease event rate for the large employer group that is below the National average and the State Government average.

BlueCross is the only health plan operating in South Carolina that has achieved event rates this favorable, measured validly. Because they reflect cost and health status, event rates are the most important outcomes for both the large employer as well as for its employees.

Further, in 2017, they validated all their major lines of business for 2016, making BlueCross of South Carolina the most carefully vetted health plan by the Validation Institute, with the most favorable validly measured outcomes, in the Southeast.

Question / Intervention Goal / Objectives

The Bockmann Technique strives to improve patients’ mobility and stability by focusing on streamlining the diagnosis and teaching self-treatment techniques.  The improvement is reflected in the change in the Patient Specific Functional Scale (PSFS), which is administered at baseline and at treatment discharge. The Bockmann Technique achieves significant improvement in functional scores in fewer visits than benchmarks.

 

Method/ Calculation

The PSFS score was taken at the initial office visit and at treatment discharge.  The discharge score was taken at either the final office visit or by phone, if no office visit was needed.  Each patient’s net improvement on the PSFS was calculated and compared to the minimum detectable change.  The minimum detectable change is 3 points if the patient is working on only one activity; 2 points if more than one activity.  An average of all patients’ score change was calculated.

The number of visits between start and discharge was also noted for each patient. The average number of visits was calculated, as well as the mode for all patients who had at least one visit, regardless of whether a follow-up PSFS score was available.

The Patient Visit Average (PVA) was calculated by taking the ratio of the visits each month by regular, ongoing patients to visits by new patients.  This measure was done at the level of the chiropractic practice, and then compared to a national benchmark[1].

 

Findings / Metric/ Outcome/ Savings

The average PSFS score improvement was 4.51 points for the 82 patients who had follow-up scores; this average includes 4 patients who had less than the minimum detectable change in score.  (See chart below.) The average number of visits was 2.5 among all 160 patients; the most frequent (mode) number of visits was 2.   This compares favorably to benchmarks from literature, such as Nicholas et al 2012[2], which had an average of 5.8 visits to achieve a 5.1 PSFS score improvement.  Other studies[3] assess the PSFS at the sixth or eighth treatment visit or refer to the amount of time between first and final visits.

Of the 82 patients with follow-up scores, approximately 70 percent achieved a 75% or more improvement in PSFS scores over the course of treatment.  Another 14% had between a 50 and 75% improvement.

The practice’s PVA was 2.0 for 10 months; thus, on average nearly half of the patients served each month were new.  The national benchmark PVA was 51.3, which would reflect close to two percent of patients being new each month.  A lower PVA points to shorter treatment times, fewer office visits, and thereby, lower costs.

 

Limitations

As with all observational studies, the measure is limited to patients who provided a follow-up PSFS score.  In Nicholas et al, 54.8 percent of patients completed the PSFS at discharge; this is somewhat higher than the completion rate in this sample, which was 48.75 percent including all patients, and 51.6 percent excluding patients who were referred elsewhere.

---

[1] Chiropractic Economics, 22^nd Annual Salary & Expense Survey, Issue 8, May 2019

[2] Nicholas et al, The use of the Patient-Specific Functional Scale to measure rehabilitative progress in a physiotherapy setting J Man Manip Ther. 2012 Aug;20(3):147-52. doi: 10.1179/2042618612Y.0000000006

[3] Horn et al, The Patient- Specific Functional Scale: Psychometrics, Clinimetric, and Application as a Clinical Outcome Measure, Appendix D, Journal of Orthopaedic & Sports Physical Therapy 2012 January 42 (1)

Question/ Intervention Goal/ Objectives

Breast and prostate cancer patients who take statins and/or metformin tend to have lower rates of death. Since cancer patients’ expenses are highest at initial diagnosis and at end-of-life1, a lower mortality rate can reduce medical costs for payers. In addition, statins have been shown to reduce the five-year breast cancer recurrence rate; having fewer patients experience recurrence would also lower medical costs.
Care Oncology Clinics use a patented combination of low-cost, low side-effect drugs, which have been shown to improve cancer treatment outcomes. The medications are in addition to and do not alter the patient’s cancer treatment.

This validation uses previously published research and focuses on two cancers that are common in working-age populations: breast and prostate.

Method/ Calculation

A nationwide cohort study of women with breast cancer stratified women by whether they were taking a statin drug. The breast-cancer-specific mortality rate was compared between the two groups.

A meta-analysis combined data from three studies of recurrence free survival among breast cancer patients, stratified by whether and what type of statin medication they were taking.

A population-based retrospective cohort study merged data from public health data sets on prostate cancer patients. The mortality rate for patients who had taken a statin alone or in combination with metformin was compared to patients who had neither drug.

Findings/ Metric/ Outcome/ Savings

The breast cancer cohort study found that patients taking a statin had a 16% lower rate of breast-cancer-specific death than patients not taking a statin.

The meta-analysis found a 12% lower five-year breast cancer recurrence rate for patients taking lipophilic statins.

The prostate cancer patients who took a statin alone or in combination with metformin had 11% lower rate of death overall, and a 20% lower rate of death from prostate cancer than patients who took neither drug.

Limitations

Though the impact of the intervention on cancer mortality rate is well established in literature, the impact on a payer’s medical costs is not yet demonstrated.

Question/ Intervention Goal/Objectives

Cleo is a family benefit platform that helps parents with work/life balance before and after a child is born. Cleo Baby assists the member with pre-natal care, choosing a hospital for delivery, and planning for return to work. Cleo Kids helps parents manage childcare, preventive care, and offers career coaching. Additional supports for special-needs children are also offered.

The goal of the program is to help employees have a complication-free birthing and return to work process.

Method/ Calculation

The following metrics were reviewed.

Focal Point	Valid Metrics: Numerator	Valid Metrics: Denominator
90% of Cleo families return to work on time and stay with their employer 12 months post-baby	People who respond actual return date within 5 days of their projected return date AND People employed by same employer after 12 months	All responses to these questions
80% of Cleo Families experience clinical improvement in mood within 2 weeks	# of people who showed a score improvement of at least 4 points	# of total responses to this question
80% of Cleo families deliver at a high-quality birth facility (defined by lower c-section rate published by credible source such as Leapfrog)	# of Cleo families who had their birth at high-quality hospitals	# of total Cleo family births
Percentage of members who meet lactation goals	# of respondents who report at follow-up that they met lactation goals set at a baseline interview	# of total respondents to this survey question
80% experience clinical improvement in mood symptoms in 1 month	# of people who showed a score improvement of at least 5 points	# of total responses to this question
Productivity of parents of special-needs children	# of Participants with children with special needs who reported no loss in productivity (subset of WPAI)	Total # of productivity responses of all families with special-needs children
High-risk member engagement	# of participants identified as high risk who interact with the program at least monthly	Total # of participants identified as high risk

Findings/ Metric/ Outcome/ Savings

The measures are valid and reasonable methods to assess the impact of the program. The data gathering is consistent and uses validated instruments to assess changes over time in mood or productivity, for
example. Each measure reflects a direct or indirect benefit to the program participant and/or the program sponsor. For example, engagement by members who have been identified as high risk has been shown in research to reduce the frequency of extreme prematurity and newborn respiratory distress. (Alexander & Mackey, 1999) and (Little & al, 2002). The monthly engagement by these members is a proxy measure for how much opportunity the program has to influence health and cost outcomes.

Limitations

The data collected only reflects those who respond to the surveys, which may not represent all Cleo users or be generalizable to other working parents.

Question / Intervention Goal / Objectives

Connect Healthcare Collaboration guides employers to choose health and wellness programs that have demonstrated measurable positive impact on costs and quality.  By implementing proven programs, employers should experience lower overall costs.  

 

Method / Calculation

The analysis compared Per Employee Per Year costs from a pre-implementation year to a post-implementation year.  Costs included medical benefits for all employees covered by the employer’s plan.  Costs for implementing the programs were netted against the savings.  

 

Findings / Outcome/ Savings

The post-implementation year’s medical costs (net of intervention costs) were statistically significantly lower than the pre-implementation year’s.  

 

Limitations 

Savings may be the result of multiple interventions, or combination of interventions; this analysis cannot differentiate at the program level.  

Having more than 12 months of cost data would provide a broader picture of the intervention’s long-term savings.  

 

 

Question/ Intervention Goal/Objectives

Connecticut Joint Replacement Institute puts patient data at the
physician’s fingertips, so that treatment is more responsive and outcomes
more successful. Clinical decision-making is guided at each step by
comprehensive relevant and actionable data. CJRI’s hip and knee
arthroplasty patients have fewer post-discharge emergency department
visits and hospital readmissions 30 days after their surgery. ED and
hospital readmissions are both recognized as quality indicators for this
type of surgery.

Method/ Calculation

Using several data sources, CJRI calculated the percent of hip arthroplasty
and knee arthroplasty patients who has an emergency department
(without a hospital readmission) within 30 days of surgery; and the percent
who has a hospital readmission within 30 days of surgery. All 3,198 patients
who underwent primary total hip and knee arthroplasty in 2019 were
included in the analysis.

These rates were compared to a benchmark from published literature.

Findings/ Metric/ Outcome/ Savings

Among 3,198 CJRI patients for primary hip and knee arthroplasty during
2019, 2.72% had an emergency department visit within 30 days of surgery
but were not readmitted to the hospital. This compares to 3.6% (for total
hip arthroplasty), and 5.3% (for combined total hip and knee arthroplasty).

In 2019, 1.45% of CJRI hip and knee arthroplasty patients were readmitted
to the hospital within 30 days of their surgery. This compares to 4.2% (knee
replacement) and 5.6% (hip replacement).

Limitations

The CJRI patients may differ from the patient groups in the cited literature;
however, the large number of patients and the period of time analyzed
indicate a high consistency in CJRI’s performance.

Question / Intervention Goal / Objectives

Drexi is a full-service Pharmacy Benefit Manager (PBM) that gives plans direct contract pricing with pharmacies and eliminates hidden fees from the process. Drexi enables payors, employers, and consumers to reduce their prescription drug spending.

 

Method/ Calculation

Drexi provided data on 3,448 pharmacy claims, showing the employer’s net cost for the drug and days’ supply with its previous PBM and its net cost with Drexi.  The analysis included only claims for which identical data was available from both the previous PBM and from Drexi.  Data from 1,223 claims was not analyzed since the identical drug had not been filled using the Drexi contract.

 

Findings / Metric/ Outcome/ Savings

The employer’s expense for the 2,226 pharmacy claims with its previous PBM was $185,724.  For the same drugs and days’ supply, the employer paid $168,335 under the Drexi contract.  Overall, the Drexi contract cost the employer 9.36% less.  Savings on individual claims ranged from 95% to -670% (additional cost).  More than 75% of the claims that incurred additional costs were $25 or less.

 

Limitations

Savings for any individual employer would vary based upon their PBM contract provisions.

Question / Intervention Goal / Objectives 

The Eden Health Primary Care Platform offers employees greater access to primary care, using near site clinics, pop-up worksite clinics, as well as virtual visits.   One goal of the program is to reduce the use of specialists, and in turn, reduce expenses for care.  

 

Method/ Calculation 

Their calculation is the proportion of patients whose visit resulted in a referral to a specialist.  

 

Findings / Metric/ Outcome/ Savings 

Of 1,205 patients who had an office visit in a 12-month period, 145 (12.03%) were referred to a specialist.  Eden’s rate for specialist referral is significantly lower than a national benchmark of 33% of patients.  

 

Limitations 

No adjustments were made to reflect characteristics of Eden’s population, which may vary from the population used for the benchmark. 

 

**Winner of the 2019 Health Value Award for Program Provider – Specialty Destination Center of Excellence (U.S.)**

Question / Intervention Goal / Objectives  

Edison’s program send patients to high-quality providers, with the goal of correct diagnosis and efficient treatment.  Travel costs and arrangements to see the designated providers are handled by the program so that employees do not have barriers to care.   

 

Method/ Calculation  

Edison provides an online calculator that estimates, based upon an employer’s number of lives, the frequency of certain conditions and the use of the travel program.  Costs for the travel program and the care delivered by the designated providers is compared to costs for the same services from other providers. Credible sources were used for the non-program providers; in addition, the user can supply his/her own plan’s data in the calculator.  

Savings are stated net of the program fees, except for the initial set-up charges.  

 

Findings / Metric/ Outcome/ Savings  

Using credible estimates, the costs for the program’s designated providers are lower than comparable providers.  Since the fees are triggered by a member’s using the program, the employer’s risk of having higher costs is minimized.   

 

Limitations  

Participation in the program may lead to a member’s diagnosis getting changed, and therefore the care plan.  This could lead to additional savings or additional costs. This is not addressed in the savings estimates.  

Question / Intervention Goal / Objectives 

EmpiRx Health’s approach to managing pharmacy benefits applies stringent evidence-based clinical care, concierge service, and proactive oversight. EmpiRx Health employs a risk-bearing model and assumes accountability to lower pharmacy costs while improving health. Its pharmacists monitor prescriptions and contact prescribers when there is an alternative medication that has equal effectiveness, better safety, and lower price for the patients and payors.  EmpiRx Health proactively manages members’ clinical care using an evidence-based approach backed by real-time data analysis. This approach strives to achieve a lower cost trend by following clinical care guidelines more closely than traditional pharmacy benefit management and by having pharmacists work directly with physicians.  

 

Method / Calculation 

For a 6,000-member group, the analysis calculated the per member per month (PMPM) cost for one baseline year and two post implementation years. The PMPM costs included the dispensing and administration fees and were net of discount for all three years shown. The baseline year had no catastrophic claimants and no claimants were excluded from the two post implementation years for having catastrophic costs. 

 

Findings / Metric / Outcome / Savings 

The PMPM cost in 2016 was $87.24; in 2017, $83.26; in 2018, $89.39. This represents a 2.5% increase over the three-year period where most PBMs were trending in high single digits or double digits year over year. Over the three-year period, the frequency of generic drug prescriptions went up 4.9% (generic dispensing rate); the non-specialty ingredient cost went down 1.8% per claim; and the specialty drug ingredient cost went down 1% per claim. 

 

Limitations 

This assessment did not include rebates.

 

**Winner of the 2019 Health Value Award for Program Provider – Specialty Cardiometabolic**

Question / Intervention Goal / Objectives 

Does the use of web-based, patient-centric health information exchange tools lower the medical costs for patients with chronic conditions?  This analysis focused on patients who had hypertension, diabetes, high cholesterol, ischemic vascular disease, chronic kidney disease, and heart failure.  Published studies have shown that care coordination does lead to lower costs for chronic condition patients; the HIE tools could support better care coordination and therefore lower costs.  

 

Method/ Calculation 

The allowed amount for hospital admissions with a chronic condition diagnosis and the total allowed amount for all medical services was compared between program participants and propensity-score matched non-participants.  The allowed amounts came from medical and pharmacy claims from 89,034 adults age 18 and older who were covered continuously by the insurer for the 12-month period. Allowed amounts were expressed as a rate per 1,000 member months for each group. 

 

Findings / Metric/ Outcome/ Savings 

The intervention group’s PMPM total allowed medical costs were eight percent lower than the control group.  The difference between the two groups was statistically significant; thus, it would not reasonably be due to random variation.  The intervention group also had a reduction in hospital admissions for chronic conditions while the control group had an increase.  These differences were also large enough to be statistically significant.  

 

Limitations 

Any observational study has the potential to have missed an unmeasured difference between the intervention and control group, despite having adjustments for various known factors.  

 

Question/ Intervention Goal/ Objectives

Using analysis of medical claims and the Methods/Calculations below, EquityHealth works with Employers and their Employees to understand their opportunities to improve Quality and lower the costs of their healthcare services. This is accomplished in an iterative, three step process:

• Initial Baseline Audit of the Employer’s Healthcare Supply Chain (remittance/claims data) for at least one full year.
• Based on the Audit results, an Employer-specific Action Plan with interventions that may include any combination of Employee Engagement, revisions to the Transparent Medical Marketplace, Provider Collaboration, Direct Contracting, and Plan/Network design.
• Iterative cycles of Audit/Plan/Interventions until savings opportunities are exhausted.

Method/ Calculation

The following metrics were reviewed:

Findings/ Metric/ Outcome/ Savings

The metrics listed are reasonable and valid methods to gauge the frequency and cost of the events described. The metrics can be used to assess how much opportunity a plan has to improve; the reliability of this estimate varies somewhat with each measure. For example, reducing the percentage of claims that were paid with errors is a simple administrative process and savings will be directly realized. Reducing the frequency of low-value emergency room visits is complex and savings would be more difficult to achieve than the correction of claims payment errors.

Limitations

The extent to which savings are realized depends upon many factors, such as employee communication and incentives to use the alternative medical providers. Thus, the valid and accurate metric is likely to vary from the savings achieved.

**Winner of the 2019 Health Value Award for Program Provider – Value-Based Provider Care Facilitation**

Question / Intervention Goal / Objectives  

By identifying patients’ risk for avoidable hospitalization and giving providers timely and relevant information to guide the care plan, Evolent’s predictive analytics tool helps practices to reduce all-cause hospital admissions and total spending.  A key factor of the system is its capability to put the latest available evidence at the provider’s disposal, so that it can be applied to care planning. 

   

Method/ Calculation  

Data collected by the risk identification tool itself includes medical and pharmacy claims, laboratory orders and results, and social/behavioral assessments.  This data guided selection of an intervention and a propensity-score matched control group. Twelve months of pre-intervention data and six months of post-intervention data were included. All-cause   admissions were calculated per 1,000 members; costs were calculated per member per month.  

 

Findings / Metric/ Outcome/ Savings  

The intervention group had a significantly lower rate of hospital admissions, and of total costs.  

 

Limitations  

Six months of post-intervention data is relatively little; analyzing 12 months of data may lead to a different, perhaps stronger, result for the intervention group. 

 

**Winner of the 2019 Health Value Award for Program Provider – General**

Question / Intervention Goal / Objectives 

Freshbenies offers employer health plan members several programs designed to encourage use of lower-cost medical services.  Freshbenies programs include virtual access to specialists, advocacy services designed for price transparency and cost efficiency, and a pharmacy savings program.  

One offering, the Teladoc program, gives members access to primary care, dermatology, and behavioral health visits by phone and video.    By substituting the lower-cost virtual visit for a face-to-face visit, the program reduces the total costs for a plan. An engaging and unique communication campaign strives to increase use of the lower-cost services.  (Note: the validation review focused only on the Teladoc program.) 

 

Method/ Calculation 

An independent analysis performed by Veracity Healthcare Analytics in 2015 measured 12 months of per member per month spending (including all plan members) for the period before the Teladoc program was implemented, and 18 months from the period after the program was implemented. The population included 80,000 freshbenies’ participants.  The calculation aggregated a reduction in per member per month spending and the change in spending trend (slope).  

 

Findings / Metric/ Outcome/ Savings 

The difference between the observed versus expected spending per member per month was $21.30.  The change in spending trend was -$1.16 PMPM and was statistically significant. The change in PMPM spending was -$9.68 and was not statistically significant.  The two factors together, however, reached statistical significance.  

 

Limitations 

No data was offered and therefore no adjustments were made to reflect other changes the employer or plan may have made at the same time that Teladoc was introduced. Other changes may have had an influence on the PMPM spending and the trend, aside from the Teladoc program.  

The analysis was of four different employer groups in the United States; other employers may have different results from the same intervention.  

 

 

Question / Intervention Goal / Objectives

Green Imaging is owned and operated by board-certified radiologists who manage a virtual medical imaging network (CT scans. MRIs, x-rays, etc.) with locations nationwide. The company offers transparent pricing, centralized scheduling, software that keeps providers up to date on status of referrals, and centralized viewing and reporting software. The objective is to provide quality medical imaging services for a lower cost.

 

Method/ Calculation

Green Imaging compared its fees with the Healthcare Bluebook Regional Fair Price (RFP) for each medical imaging service offered.

 

Findings / Metric/ Outcome/ Savings

A table documented cost differences between Green Imaging’s price and the regional fair price for over 40 procedures, including CT scans, MRIs, x-rays, and mammograms. All examples demonstrated Green Imaging fees were 0.70% to 170% below RFP. 

 

Limitations

The data offered does not address the quality of care.   

 

Question / Intervention Goal / Objectives

Happy Whole Human provides a self-assessment tool and, based upon the results, targeted support for 28 different wellness dimensions.  Dimensions range from engagement at work to emotional wellness.  Participants do re-assessments at set intervals.  Exposure to the program strives to boost participants’ scores on self-perceived health status, productivity, and turnover intention.

 

Method/ Calculation

The wellness assessment included three validated questions for self-perceived health status, work productivity, and turnover intention.  Thirty participants’ scores were taken at baseline and at 12 weeks.  Each participant’s change in score was then calculated (correlated samples).

 

Findings / Metric/ Outcome/ Savings

After 12 weeks, more participants strongly agreed that their health status was excellent though the change was not statistically significant.  There was no change in workdays lost due to illness.  There was a statistically significant increase in participants’ intention to stay with their current employer.

 

Limitations

Different employers may have different results.

 

BROADENED AND DEEPER ENGAGEMENT
Health Partners’ Health Advocacy approach enables the delivery of comprehensive disease and case management solutions to manage care, optimize health and generate cost savings.
Through innovative identification, stratification and engagement techniques, more members receive the support and services they need for successful condition management.
Member support and services improve engagement through:

• Member activation
• Condition literacy
• Care and treatment decisions
• Self-management skills

IMPROVED CLINICAL OUTCOMES
By aligning the intensity and type of support with a member’s needs, overall health and clinical outcomes are improved.

LOWER COST OF CARE
Improved clinical outcomes translate to reduced adverse events and associated costs.

**Winner of the 2019 Health Value Award for Direct Care Provider – General Primary Care**

Question / Intervention Goal / Objectives

The intervention equips primary care practices to work more effectively with patients suffering muscle/bone/ joint related conditions.  By assigning these patients to the appropriate care pathway early in the disease process, the intervention seeks to reduce costs. 

 

Method/ Calculation

The analysis used matched cohorts within a single 5,000-life employer group – one receiving community care and the other receiving care at a Super Clinic.  Per patient per year costs including pharmacy, hospital and professional services were compared between the two cohorts.  The analysis also addressed the care received in the six months following the musculoskeletal care episode. 

 

Findings / Metric/ Outcome/ Savings

After risk adjustment, the average cost per intervention group patient was 51.48% lower than the community care patients’ average cost (p < .0279).  The utilization of MRIs, injections, and surgeries was lower in the intervention group by 49.75%, 39.44%, 78.38% respectively.  The community care patients’ relative risk of seeking follow-up care in the six months after the initial episode was 4.49 times higher than the intervention patients.  All three of these differences between the two groups were statistically significant. 

 

Limitations

Quality and outcome data were not provided for the two groups studied.  The six-month post-treatment follow-up period is a proxy for quality, but patient pain and functional assessments would be superior. 

 

 

It Starts With Me Health endeavors to support overall employee health for clients and our employees/health plan members. We heartily endorse the Employee Health Vendor Code of Conduct, which states that any health and wellness programming should do no harm to employee health, employee morale, corporate integrity or employee/employer finances.

 

Employee Benefits and Harm Avoidance

It Starts With Me Health will recommend health & wellness efforts that are done “with and for” employees & health plan members, not “to” them.

Our focus will be on the promotion of personal well-being of employees, identification and/or avoidance of potentially negative health outcomes for employees/health plan members, analyzing data, and assistance in improving the financial stability of the health plans of our clients.

Our choices and frequencies of suggested health interventions, testing, and educational materials will be as generally consistent as possible with expert guidelines such as Harvard Medical School, the United States Preventive Services Task Force (USPSTF), the CDC, and Choosing Wisely (an initiative of the American Board of Internal Medicine). This is done in an effort to minimize over-programming, over-spending, overwhelming, over-screening, over-diagnosis, and over-treatment. As a specific example, we will discourage our clients from implementing universal financial incentives for annual checkups.

We will recommend to our clients that potential programs and services be evaluated to ensure that employees will not be singled out, fined, or embarrassed for any reason due to our health status.

 

Respect for Corporate Integrity and Employee Privacy

When analyzing our data or other client vendor data, we will provide unfiltered, unbiased, & unspun data, good or bad, so that our clients can make sound business decisions about their mix of services and vendors.

We will not share employee-identifiable health data with employers.

 

Commitment to Valid Outcomes Measurement

Our relevant staff will understand USPSTF guidelines, employee harm avoidance, wellness-sensitive medical event measurement, and outcomes analysis.

If clients wish to engage in health improvement programs internally or via other vendors, we will recommend that we include warnings of any adverse effects that participants might encounter (e.g. with the popular “Biggest Loser” competitions, the programs should be accompanied by warnings regarding potential unhealthy eating behaviors, weight-cycling, awareness of issues with other health conditions and/or medications, and poor long-term results that can result).

In addition, we will advise our clients that for true measurement of any type of cessation or reduction program (weight loss, tobacco, etc….), vendors should be required to provide aggregate reports that outline both immediate improvement and most importantly, the true measure of success – the continued maintenance of that improvement by each individual for at least 12 months.

Our contractual language and data reports will be valid. If we must measure outcomes using other than one of the valid techniques, we will disclose the invalidity (such as “healthy volunteer bias” or the “natural flow of risk” or ignoring dropouts) that is likely to cause savings to be overstated. If we use a participants-vs-non-participants methodology, we will disclose that in the three instances in which we were tested, we were shown to be invalid.

 

Question/ Intervention Goal/ Objectives

The Leapfrog Value-Based Purchasing Program creates composite scores reflecting each hospital’s performance on safety, quality, and resource use standards, benchmarked nationally. Payers and purchasers can use the scores to compare how individual hospitals deliver value compared with peers nationally. From each composite score, payment can be calibrated to reward the highest performers. The ultimate goal of this program is for higher performing hospitals to be recognized and rewarded through various payment strategies, and for payers to have objective, consistent information upon which to gauge performance. Purchasers such as self-insured entities can also utilize The Leapfrog Value Based Purchasing Program data to either tier or build a narrow network to either steer or concentrate volumes on better performing providers.

Method/ Calculation
The Value Based-Purchasing Program organizes the measures used in Leapfrog’s national, annual hospital survey into six “domains:”

• Medication safety
• Inpatient Care Management
• Infections
• Maternity Care
• Inpatient Surgery, and
• Pediatric Care

Each hospital that participates in the Leapfrog Hospital Survey is automatically included in the Value-Based Purchasing Program so that individual hospitals can be compared on:

• VBP Measure Scores – A score ranging from 0 (worst performance) to 100 (best performance) assigned to each measure in the Leapfrog Hospital Survey.
• VBP Domain Composite Scores – A composite score for each domain, calculated by applying a relative weight to each measure score within the domain.
• VBP Value Composite Score – An overall composite score combining all the weighted measure scores in all the domains.

The relative weights assigned to each measure are determined based on the number of patients affected by the measure, the severity of harm being measured or the potential harm when not following clinical guidelines, and resource use (defined as excess cost or cost avoidance) related to the measure. These criteria align with value-based purchasing goals of achieving high quality at the right cost. For each measure in the composite, highest achievement nationally requires both highest quality and lowest use of resources.

For example, the composite includes several measures of healthcare-associated infections, which are weighted heavily. In term of excess cost, infections add excess cost by resulting in additional days in the hospital, additional medications to treat the infections, and readmission risks. In terms of severity of harm, some infections, like central-line associated blood stream infections, can result in severe disability or death. Thus, the Value Score reflects hospital performance on quality and safety measures known to cause significant harm and excess costs. The program’s focus on safety and quality in hospitals is similar to Medicare’s Hospital VBP program, which financially rewards hospitals for meeting benchmarks. The data collection method is a nationally standardized methodology and follows established protocols to ensure reliability. The scoring methodology is the product of published literature and experts such as health economists and represents the priorities of employers and purchasers.

Findings/ Metric/ Outcome/ Savings

This validation is focused on the reliability of the data collected and its connection to value-based purchasing. As such, there are not findings.

Limitations

Responding to the Leapfrog Hospital Survey is voluntary. It is possible that non-responding hospitals, approximately half the hospitals nationally, have differing levels of quality, safety, and value. The composite scores in the Leapfrog Value-Based Purchasing Program compare hospitals against a national benchmark to identify the lowest to highest performance, but the benchmarks cannot incorporate the performance of the non-responding hospitals.

Question / Intervention Goal / Objectives

By selecting a higher-deductible group health plan and reimbursing a portion of the employee’s deductible, Med-A-Vision clients can reduce their total health benefit costs.  The deductible reimbursement is handled through a Health Reimbursement Arrangement (HRA).  Though the HRA’s payments are variable depending upon the group’s claims, the total of the HRA payments and the lower premiums to the insurer is generally less than the cost of a lower-deductible health plan.

The graph below illustrates the three components of the Med-A-Vision program for a 13-employee group: insurance premiums, third-party administrator fees, and deductible reimbursement (HRA benefits).  The graph includes an estimate of what the same group would have paid for a $2,500 deductible plan from an insurer.  The Med-A-Vision program purchases a higher-deductible plan from the insurer ($5,700 in this example); however, the effective deductible to the employee is 30% of $5,700 because the HRA reimburses 70%.

 

Method/ Calculation

Under the Med-A-Vision program, the HRA payments are a variable expense and thus determine whether and how much the employer saves.  Therefore, the analysis focused on these payments.

The analysis calculated what percentage of covered members received an HRA payment, indicating that they had had an expense to which the deductible applied. The analysis included 18,421 person-years from 206 different employers. Company size ranged from one to 217 employees.

This was calculated for the entire group of 206 employers for the three-year period; for employers with 30 or more employees; and for employers with fewer than 30 employees. These were also calculated for each of the three years (2016, 2017, and 2018) separately.

The average HRA payment per claimant was calculated for the entire group for each of the three years. In addition, the average HRA payment per eligible member was calculated.

 

Findings / Metric/ Outcome/ Savings

Over the three-year period, the percentage of covered members who had a claim paid by the HRA varied from 7% to 100% for companies with fewer than 30 employees. It ranged from 16% to 84% for companies with 30 or more employees.  Overall, across the entire 18,421 person-years and all company sizes, approximately 50% of the covered members received payment from the HRA. The rate was nearly the same for each of the three years.

The percentage of members who had a claim is lower than the national average among privately insured people age 18 – 64 of 80.7%.^[1]  Two-sample tests indicated that the Med-A-Vision’s rate was significantly different from the national average. The gap between the actual and expected rates cannot be fully explained, though two factors likely contribute to it:

• The percentage of members who had a claim is understated by an unknown amount because members who had claims not subject to the deductible are not included. For example, plans cannot impose a copay or deductible on preventive care.
• In addition, the presence of the deductible, even though it is subsidized, has a dampening effect on the frequency of claims. It is well-documented that people enrolled in higher-deductible health plans use fewer medical services.

The average per claimant HRA payment was $1,039 in 2016, $1,019 in 2017, and $1,017 in 2018.  Assuming that the HRA payments are 20% of the total medical bill, then claimants had an average of $5,196 (2016), $5,096 (2017), and $5,087 (2018).  The estimated average total medical cost per claimant is 16% lower than the national average of $5,908 (2016)^[2]. The span between actual and the national average likely reflects that group’s members’ rural location in a single state.

The average HRA payment per eligible member was $546 in 2016, $516 in 2017, and $480 in 2018.  The average HRA payment per eligible member includes approximately members who claimed zero HRA benefits (approximately half of total members in each year).  Of members who did have an HRA benefit paid, approximately half claimed less than $500.

 

Limitations

New employers adopting the program will have different experience than the group’s average experience in any one particular year.  Savings cannot be guaranteed.

 

---

^[1] Medical Expenditure Panel Survey Household Components 2016 https://meps.ahrq.gov/mepstrends/hc_use/

^[2] Ibid.

Question / Intervention Goal / Objectives

The MedEncentive Mutual Accountability and Information Therapy (MAIT) program strives to nudge patient and doctor behavior in a manner that improves health and healthcare, lowers costs, and provides fulfillment to doctors and patients (i.e., Triple/Quadruple Aim).  This patented web-based system accomplishes these objectives by employing “aligned financial incentives,” which are offered by health insurers to both doctors and patients for engaging one another in an educational process called “information therapy.”

The program uses a unique patient-doctor aligned-incentive mechanism because it has been found to be more effective at improving compliance than traditional unilateral incentive methods.[1]  The program incorporates information therapy to mitigate the ill effects of inadequate patient health literacy, which studies have shown is associated with higher rates of hospitalizations[2] and preventable emergency department visits,[3] as well as greater medical costs.[4]

A study of the program’s effectiveness, entitled “Reduced Hospitalizations, Emergency Room Visits, and Costs Associated with a Web-Based Health Literacy, Aligned-Incentive Intervention: Mixed Methods Study,” was published in the peer-reviewed Journal of Medical Internet Research [2019; 21(10)].[5]  Validation Institute’s examination of the MedEncentive program was based on this study.

 

Method/ Calculation

The study involved a 5-year pre- and post-, mixed-method test of the program in an employer health plan, comprised of over 1,800 plan members. The study’s quantitative measures compared emergency room visits, hospital admissions, and overall per capita expenditures two years before and three years after program implementation. The pre- and post-comparison included all plan members, regardless of whether the member participated in the MedEncentive program, and included all disease states and medical conditions.  To compare pre- and post-expenditures on an equivalent basis, the post-implementation expenditures were adjusted to the 2013-14 baseline level using a cost-per-unit-of-service conversion method.

Doctor and patient program participation (engagement) rates, patient responses to standard program questionnaires, and voluntary comments posted by doctors and patients were analyzed to test the plausibility of attributing the outcomes to the program.

 

Findings / Metric/ Outcome/ Savings

Emergency room visits per 1000 decreased 14%, or 31.3 visits per 1000, from the 2014 baseline period to the 2015 – 2017 post-implementation annual average, while hospital admissions per 1000 decreased 32%, or 26.5 visits per 1000.  The plan’s net average annual per member expenditures decreased 11%, or $675 dollars, from the 2013-14 pre-implementation period to the 2015-17 intervention period, which was 7.5 times the $90 annual per capita cost of the program.

The patient and doctor participation rates in the program were over 70% and 40%, respectively.  Patients rated the helpfulness of the educational content 4.4 out of 5 (n=15,260) and reported their level of adherence to health recommendations 4.7 out of 5 (n=15,186). These results could be interpreted as supporting the impact of the program.

 

Limitations

The study focused on a health system employer who was able to embed the MedEncentive program into the electronic health records system that served a majority of the members.  The providers’ ready access to the program may have created an advantage that other employers would not be able to replicate. However, other validated and non-validated studies have shown similar results in other types of employers without electronic health record integration.[6]^{, [7]}

Conflicts of interest for two study authors were properly disclosed. 

 

[1] Asch, DA et al. Effect of financial incentives to physicians, patients or both on lipid levels: a randomized clinical trial. JAMA 2015 nov 10; 314(28):1926 – 1935.

[2] Baker DW et al. Functional health literacy the the risk of hospital admission among Medicare managed care enrollees. American Journal of Public Health. 2002 Aug; 92(8):1278-83.

[3] Balkakrishnan et al, The Association of Health Literacy with Preventable Emergency Department Visits: A Cross-Sectional Study, Journal of Academy of Emergency Medicine, 2017 Sep;24(9):1042-1050.

[4] Haun et al, Association between health literacy and medical care costs in an integrated healthcare system: a regional population-based study. BMC Health Services Research 2015; 15:249

[5] Greene 1C, Haun JN, French DD, Chambers SL Roswell RH. Reduced Hospitalizations, Emergency Room Visits, and Costs Associated with a Web-Based Health Literacy, Aligned-Incentive Intervention: Mixed Methods Study. J Med Internet Res 2019;21(10):e14772

[6] Keene N, Chesser A, Hart TA, Twumasi-Ankrah P, Bradham DD. Preliminary benefits of information therapy. J Prim Care Community Health 2011 Jan 01;2(1):45-48.

[7] Parke DW. Impact of a pay-for-performance intervention: financial analysis of a pilot program implementation and implications for ophthalmology (an American Ophthalmological Society thesis). Trans Am Ophthalmol Soc 2007;105:448-460

Nova Healthcare Administrators has developed an approach that applies data about the employer, the benefit plan, and medical claims to find opportunities for lowering cost and improving health.  Nova then follows through with customized interventions and programs that directly address the opportunity areas.

On the measure of Per Member Per Year medical costs, excluding pharmacy and Emergency Room visits per 1,000 member months, Nova demonstrated performance superior to Milliman Medical Index (MMI).

Question / Intervention Goal / Objectives 

Proactive MD works with self-funded employers to offer onsite Advanced Primary Care.  Proactive MD’s Health Centers focus on assessing and responding to patients’ risk levels and managing chronic conditions.  They offer clients a guarantee of their performance to reduce medical spending.  

The on-site clinic engages patients through a strong doctor-patient relationship, a Patient Advocate, and care coordination support.  The program seeks to ensure that specialty care is high quality and effective. This goal is supported by partnerships with other organizations that assess specialty care quality.  In addition, the program strives to reduce spending on secondary and tertiary care by using high-performance networks and applying reference-based pricing.  

There is evidence that patients receiving Direct Primary Care services have lower medical costs than comparable patients receiving care in traditional primary care settings. [1] 

 

Method/ Calculation 

 The analysis matched a group of 86 patients receiving care through the program to an 86-person group receiving care elsewhere.  Matching criteria included selected chronic conditions, age, gender, and member eligibility months. Per member Per month allowed amounts were calculated for both groups for a period of 18 months before the program began, and for 24 months afterward. 

 

Findings / Metric/ Outcome/ Savings 

The treated group had a $42 PMPM decrease from the baseline period to the second year of the program, compared to a $129 PMPM increase in the matched control group.  PMPM dollar amounts for both matched groups are inclusive of Proactive MD’s program cost normalized to a PMPM over the 24 months after the program began. (See Table below.)

 

Limitations 

As with all observational studies where participation is voluntary, an unmeasured variable may have influenced the outcomes. 

 

---

1 Eskew and Klink, Direct Primary Care: Practice Distribution and Cost Across the Nation The Journal of the American Board of Family Medicine November 2015, 28 (6) 793-801;

Question / Intervention Goal / Objectives 

Comprehensive health navigation/care coordination.

 

Method/ Calculation 

Multiple years’ actuarial reports have been reviewed and critiqued. Quantum’s actuarial reports are highly credible. 

 

Findings / Metric/ Outcome/ Savings 

Overall savings vs. trend for the entire population managed.  This approach avoids all regression to the mean and participation bias by including the entire population for all periods.

While they “average” a higher reduction, for validation purposes we are looking for a figure that most companies should be able to achieve. Therefore, the figure is lower than the figure in their actuarial reports.

 

Limitations 

Conflated average results with results that can be expected 90% of the time. The validation percentage is lower than the actuarial report describes for that reason.

The trend is sourced and reasonable. However, it is not a pure cost reduction across the population. Changing the trend assumption would change the savings.

Results over time have weakened noticeably, though still the best in the field.

 

Question / Intervention Goal / Objectives

Quizzify provides a health literacy promotion program, which engages employees in a game using short, multiple-choice and true-false quizzes.  Participants learn about various topics, such as opioids, Humira, sleep, CT scans, stents, health hazards, and hidden sugars.  Studies have shown a link between health literacy and more judicious use of medical services[1], especially emergency department visits[2] and inpatient hospital stays[3].

 

Method/ Calculation

Though Quizzify is a content tool, the guarantee is engagement: using the Validation Institute’s own Benefits Engagement Survey Tool (BEST), Quizzify’s score will be farther “above the line” (drawn at 45 degrees) than any other health benefit measured, as indicated in this example.

The Y-axis of the plot is based upon responses to the two-question BEST survey. Respondents give the number of times they used a program and rate how useful they found it. An optional third question asks whether having the program offered enhances the respondent’s perception of the employer.

In addition, the employer supplies the cost for each program.  This data sets the X axis of the plot.

 

The context of this graph is described in this explainer.

The survey needs to include at least three benefits and ideally, up to seven, and to get responses from a minimum of the lower of 100, or 10% of employees.

 

Findings / Metric/ Outcome/ Savings

Quizzify has achieved this outcome in every instance.

Therefore, Quizzify will place 100% of its fees at risk for being the farthest above the 45-degree line. Since the formula for the Y-axis is uses times usefulness, if Quizzify fails to achieve this goal, the monthly pricing will be reset to zero until Quizzify is indeed farther above the line than any other health benefit.

 

Limitations

While the same question set is asked of all the benefits measured, the survey is not a validated instrument. Subjectivity, and self-selection amongst survey respondents, introduce biases. The biases should be the same for all the benefits measured, however, because the same subset of the employee population completes the survey for all the benefits measured.

 

---

[1] Haun et al, Association between health literacy and medical care costs in an integrated healthcare system: a regional population-based study. BMC Health Services Research 2015; 15:249

[2] Balkakrishnan et al, The Association of Health Literacy with Preventable Emergency Department Visits: A Cross-Sectional Study, Journal of Academy of Emergency Medicine, 2017 Sep;24(9):1042-1050

[3] Baker DW et al. Functional health literacy the the risk of hospital admission among Medicare managed care enrollees. American Journal of Public Health. 2002 Aug; 92(8):1278-83.

 

 

 

**Winner of the 2019 Health Value Award for Direct Care Provider – Specialty Musculoskeletal**

Question / Intervention Goal / Objectives 

Reflexion provides a Virtual Exercise Rehabilitation Assistant for patients who have had total knee arthroplasty, as a substitute for in-person physical therapy. The program uses an avatar, motion capture technology, and oversight by physical therapists to deliver physical therapy services.  The research question is whether Reflexion’s program achieves patient outcomes comparable to conventional physical therapy. 

 

Method/ Calculation 

A randomized controlled trial assigned non-traumatic total knee arthroplasty patients to the intervention or to conventional care.  Measures of patient outcomes included: total healthcare costs for the 12 weeks following the surgery, home health visits, outpatient physical therapy visits, and rehospitalizations.  In addition, the study measured Knee Injury and Osteoarthritis Outcome Score (KOOS) at six and 12 weeks, knee extension, knee flexion, and 10-meter gait speed.  

The service utilization data came from medical claims.  The KOOS is a self-administered survey. The other measures — extension, flexion, gait speed, pain, and falls — were administered by the medical professionals at the six and 12-week point. 

 

Findings / Metric/ Outcome/ Savings 

The intervention patients had 60% lower billable costs than usual care patients for the 12 weeks after surgery.  The $2,745 savings (mean per patient) stems partly from fewer rehospitalizations on average; fewer home health and outpatient physical therapy visits compared to the usual care group.  

The 6-week knee extension, knee flexion, gait speed and the 6- and 12-week KOOS scores for the intervention and the usual care groups did not significantly differ.  

(Note: patients’ data was assigned to their original assigned group, even if they changed groups during the study period.  An “intent to treat” analysis structure was used.) 

 

Limitations 

Patients were selected based upon having a Risk Assessment and Prediction Tool score of six or greater prior to surgery, indicating they were expected to be discharged to home after the surgery.  These results may not be generalizable to all types of total knee arthroplasty patients.  

 

NOTE: Individual states may allow importing prescription drugs, in conflict with federal rules. Enforcement of rules and liability of employers is an evolving issue. See the FDA’s personal drug importation summary.

 

Question / Intervention Goal / Objectives

Rx Manage, a voluntary mail-order pharmacy benefit, offers brand name maintenance drugs at a lower cost.  The program imports drugs that are available from the same manufacturer or license holder at a lower cost than the payer would spend for a U.S. supplier.  When lower-cost or generic drugs become available in the U.S., Rx Manage excludes them from the program.  Employees pay no co-pay when they use the program and employers pay only if the services are used.

 

Method/ Calculation

The applicant compared its final net cost for a sample employer client to a reference price set at 90% of GoodRx’s lowest price (including available coupons) in the state of Florida. GoodRx is a consumer search tool that lists pharmacy prices nationwide. Ninety-two prescriptions went through the program during the eight-month period of time.

 

Findings / Metric/ Outcome/ Savings

The employer paid 65% less for drugs purchased through Rx Manage than the reference price. Estimated savings for the eight-month period were approximately $106,000 which constituted 9% of their total drug spending for this time period.  In addition, employees’ costs were reduced since no co-pay was required. These avoided costs were not estimated.

 

Limitations

The program may have an impact on the purchasing volume for a plan’s Pharmacy Benefit Manager (PBM). This, in turn, could affect rebates and net costs.  These effects are not addressed in the analysis.

 

Predict & Prevent Primary Claims
“The Coaching Choice”

1. Splashlight’s predictive analytics will accurately identify and stratify the high risk and rising risk individuals using three years of historical health claims and workers’ compensation claims. They use the first two years of data to build the model and the third year to prove validity. While many vendors promise a reduction in cost of the highest-cost or highest-risk segment, they recognize that this cohort regresses to the mean, and a demonstration of effectiveness requires showing savings after making a “dummy year adjustment” to account for that regression effect. Splashlight will contract to show savings validly, taking that adjustment into account.

2. Prior to beginning their Prevent services, Splashlight provides a benchmark of predicted future medical costs for health claims and workers’ compensation medical claims. Using their validated predictive analytics methodology they then adjust for those ICD10 codes that are mutually agreed with their clients to be impactable through their prevention services. They will apply measures that are attributable to The Coaching Choice solutions which they expect to result in a reduction of overall costs by at least 5%.

3. Splashlight’s contractual language and outcomes reporting will be transparent and plausible. All research limitations (e.g., “participants vs. non-participants” or the “natural flow of risk” or ignoring dropouts) and methodology will be fully disclosed, sourced, and readily available.

4. Splashlight will measure the actual savings realized each quarter and annually and adjust their savings totals accordingly to show the total costs and savings for the client. Further they will guarantee in their contracts that their costs will not exceed more than 40% of the actual savings achieved over a three year period. If the actual annual savings have not exceeded annual costs after a six month period to adjust for delays in claims received, then Splashlight will continue to provide services at no charge for up to an additional year or until the actual savings are realized through validated measures.

5. The participant experience will be tracked for all participants with the following expectations:
80 percent of the identified population will participate in The Coaching Choice program.
80 percent of participants in The Coaching Choice program will report their willingness to recommend the program to others.
80 percent of participants in The Coaching Choice program will report their improved confidence in their ability to manage their healthcare.
80 percent of the participants in The Coaching Choice program will report that it improved their quality of life.
80 percent of the participants in The Coaching Choice Program will report positive personal experience as a result of participation.

 

Question / Intervention Goal / Objectives

Sustainable Health Index offers a calculator for groups to estimate the financial impact of its program which seeks to interrupt the natural flow of risk (i.e. moderate to high). The program offers participants education and interaction via an app, customized to their risk level and readiness to change.  SHI’s program applies principles and structures that have been proven successful for reducing risk level.[1]  Similar programs delivered digitally have also been shown effective.[2]

 

Method/ Calculation

The calculator estimates a prospective group’s risk levels, using peer-reviewed literature.  SHI uses a peer-reviewed model of the natural flow of risk.

SHI can apply the calculator’s model to a group.  The intervention uses valid methods to assess each person’s health risk and health activation.  Note: not all survey questions are from validated instruments, but the majority is.  For validated questions, SHI uses the recommended scoring methods.  SHI creates a composite score for each participant.  Each participant is then assigned a risk level, and a change in risk level is determined at baseline and at post intervention.  All participants’ changes are included, even those whose risk level increased during the intervention period.

 

Findings / Metric/ Outcome/ Savings

With a basis in peer-reviewed risk flow models and valid measurement of health risk, SHI can credibly estimate the overall impact of its program.

 

Limitations

The calculator’s estimate is not a guarantee of savings or risk reduction for any particular employer.  Rather, it is a demonstration of valid measurement and reasonable estimates.

 

[1] White, JC, Hartley S, Ozminkowsky, RJ Association Between Corporate Wellness Program Participation and Changes in Health Risks J Occup Environ Med. 2015 Oct;57(10):1119-26

[2] Loeppke et al The association of technology in a workplace wellness program with health risk factor reduction J Occup Environ Med. 2013 Mar;55(3):259-6

 

 

 

**Winner of the 2019 Health Value Award for Program Provider – Specialty Wellness**

Question / Intervention Goal / Objectives 

By assessing health risk and providing specific supports and interventions to individuals, U.S. Preventive Medicine’s Preventive Plan strives to interrupt the natural flow of people from lower to higher risk levels.  The natural changes in a population’s mix of health risks is described by Dee Eddington in The Impact of the Prevention Plan on Employee Health Risk Reduction, Population Health Management, Vol 13, No 5, 2010.  

The intervention includes high-tech/high-touch support and encouragement that participants need to adopt healthy behaviors.  Support services include 24/7 nurse hotlines, one-on-one health coaching, contests, group events, and employer incentives.

 

Method/ Calculation 

The assessment of each individual’s risk level is determined by the program’s Health Risk Assessment; criteria for scoring and determining risk levels are described in The Impact of the Prevention Plan on Employee Health Risk Reduction, Population Health Management, Vol 13, No 5, 2010.  

The 2,606 participants’ baseline scores were compared to their score one year after program implementation.  Each risk score was assigned to low, moderate, and high based upon the scoring rubric. The mix of the population’s risk scores was then assessed at baseline and at one-year post implementation.  

 

Findings / Metric/ Outcome/ Savings 

There was a net increase of 9.40% of people in the low-risk category, a decrease of 3.61% in the moderate-risk category, and a 5.79% decrease in the high-risk category. 

 

Limitations 

As the participants came from several different employers, we cannot determine what other factors may have influenced the results.  

 

**Winner of the 2019 Health Value Award for Program Provider – Care Navigation**

Question / Intervention Goal / Objectives

PredictiMed’s intervention identifies people who are at higher risk for colon cancer with the goal of ensuring appropriate testing. Through appropriate testing, cases of colon cancer can be reduced and life years saved.   

 

Method / Calculation 

The applicant identified patients as at-risk based upon their Health Risk Assessment scores ; then, the results of their colonoscopy test was compared to assess the accuracy of the HRA score.  At-risk patients were compared to the general population.  Costs for colonoscopy tests and life-year saved came from credible sources.  Results for the group were compared to a matched control group.  

Return on investment was expressed in two ways: savings from reducing cases of colon cancer, and per life year saved over a three-year period.    

 

Findings / Outcome/ Savings

The savings per avoided colon cancer case was $55,000; the savings per life year saved over a three-year period was $1,995,000.  

 

Limitations 

The study was observational, and therefore, the intervention group may be affected by selection bias.  

Estimates for the control group came from other studies and national statistics.  

 

US Preventive Medicine (USPM) is the first and as of now only wellness company to achieve a sustained and significant reduction in wellness-sensitive medical events (asthma, cardiac events, COPD, congestive heart failure and diabetes) across that portion of its entire book of business for which comprehensive claims data was available (see chart below). The reduction achieved by USPM significantly outpaced the much smaller national decline in these events, as evidenced by both the database maintained by the Disease Management Purchasing Consortium (through 2014) and the federal Healthcare Cost and Utilization Project (HCUP) database (through 2013). USPM is further the only wellness company willing to contractually commit to being measured on this metric in future contracts. This makes USPM the only wellness company to measure validly, achieve significant event reduction when measured validly, and offer valid contractual metrics.

Question / Intervention Goal / Objectives

Vera Whole Health’s primary care model offers on-site clinics, health coaching, longer patient visits, and coordination of care.  By working closely with patients on setting and achieving health goals and providing convenient access to primary care, Vera Whole Health demonstrated lower than expected costs for a 1,192-member group.

 

Method/ Calculation

Two baseline years and two intervention years were used in a pre- post- comparison.  Two linear models were calculated: one with two baseline years, and another with the most recent three years.  The average of these two linear models produces the expected trend.  The actual cost trend is then compared to this expected trend.  The data is risk adjusted via one normalized predictive risk score.  Data is further adjusted for any changes in the health benefits.

The difference between the actual and the predicted trend is the gross savings.  Net savings is calculated by subtracting the operating costs of the program, which includes the startup costs of building the onsite clinic.

 

Findings / Metric/ Outcome/ Savings

The program had net savings $232,458, which is $195 per eligible member per year, after investing approximately $1.0 million in the onsite clinic.

 

Limitations

Savings will vary by employer, and may not be positive in all cases.

 

 

Question / Intervention Goal / Objectives

Virta Health offers a continuous remote care intervention to people living with type 2 diabetes.  Participants adhere to a carbohydrate-restricted nutritional protocol, and, via a mobile app, receive daily monitoring and supervision by a health coach as well as a board-certified physician.  The interaction with the program – daily or more often in the early stages – ensures that medical management is continuous, and as participants’ metabolism changes, the physicians quickly deprescribe medications.

Two non-randomized clinical trials were reviewed: Hallberg et al, Effectiveness and Safety of a Novel Care Model for the Management of Type 2 Diabetes at 1 Year: An Open-Label, Non-Randomized, Controlled Study, Diabetes Therapy (2018) 9:583 – 612; and Athinarayanan et al, Long-Term Effects of a Novel Continuous Remote Care Intervention Including Nutritional Ketosis for the Management of Type 2 Diabetes: A 2-Year Non-randomized Clinical Trial, Frontiers in Endocrinology (2019) 10:348.  These studies focused upon changes in various laboratory tests, ranging from weight and blood glucose to blood cholesterol and liver enzymes.

 

Method/ Calculation

Participants were recruited and voluntarily enrolled in the Virta Health program or the comparison group, which consisted of patients referred to a diabetes education program by their physician.  Patients’ data were assigned to their original group, even if they dropped out.  Both groups had lab values assessed at baseline, at one year and at two years. Intent-to-treat analyses with methods for handling missing data, as well as completers analyses, were performed.  Statistical adjustments were made to account for differences between the two groups’ baseline characteristics.

 

Findings / Metric/ Outcome/ Savings

The table below lists selected measures and the amount of improvement that program participants achieved.  Data is not shown for the Usual Care comparison group because no significant change was observed.

Limitations

The statistical methods used in both studies sought to adjust for baseline differences between Virta Health program participants and usual care participants, but the groups may vary in ways that cannot be measured.

***Winner of the 2019 Health Value Award for Program Provider – Specialty Musculoskeletal**

Question / Intervention Goal / Objectives

Virtual Physical Therapists offers a tele-rehabilitation service, with the goal of delivering high quality care.  Patients get video visits and assessments from physical therapists via smart phone.  A 2018 systematic review of studies found that research on physical therapy telehealth were of moderate quality and found positive impact on health outcomes. One study included in the review showed cost savings for telehealth physical therapy compared to usual care.  The review called for more robust research. 

 

Method/ Calculation

The analysis measured scores for three patients on validated instruments at day 1, 2, 5, 8, 15, and 90 of treatment.  The instruments were the Numerical Rating Scale, Body Pain Diagram, and Patient Specific Functional Scale. 

 

Findings / Metric/ Outcome/ Savings

The patients all showed improving scores over the course of treatment, and each achieved the score reflecting remission of pain and regain of function. 

 

Limitations

The analysis did not compare the three patient case studies to usual care patients.   

 

Validation Institute reviewed Predilytics’ Analytic Design Plan (ADP) for a health plan, which outlined:

· the objectives of the Medication Adherence project,
· the method for achieving those objectives, and
· the statistical and quantitative metrics that are used to document model validation.

Validation Institute reviewed Predilytics’ program evaluation which described how the models were used in a health plan’s operational call center to improve medication adherence for their members.

Predilytics’ evaluation of the predictive models and the operational recommendations reflect a well-grounded understanding of how predictive models perform and accurate assessment of the value of the insights created by these models to inform resource deployment within healthcare operations.

Within that program evaluation, Predilytics used additional statistical analysis to isolate the incremental performance of the predictive model above and beyond call center efficiency gains. Overstatement of attribution is commonplace in predictive modeling in population health, but the model was very carefully designed not to overstate attribution of the Predilytics’ intervention to the outcomes. The one aspect of this study in which overstatement might have been taking place is that the customer service representatives making the outbound calls got better at calling members over the course of the study. As a result, calls towards the end of the period were more successful. Predilytics pointed out this potential confounder to the validator and estimated the impact of it, rather than wait for the validator to point this out to them.

Consequently, the Validation Institute is able to provide validation for Predilytics, as they are well above the standard of using a methodology that is equivalent to or better than the most valid standard commonly used in their segment of the industry.

 

Question / Intervention Goal / Objectives

The Wildflower Maternity digital front door offering delivers a consumer-centric experience for expectant mothers through a mobile application that personalizes education and facilitates calls to action within the provider, payer, and community ecosystems.  The app educates users on personalized health issues, reminds them to take certain action steps through tools and trackers, and connects them to local providers through integrated provider appointment portals, and touch-to-call features for provider, payer, and community resources,  such as ride services.  The goal is to improve maternal care and reduce the cost of pregnancies and deliveries, for both the user and the healthcare system, by:

Educating users on what’s normal and what’s not while engaging them early and often through personalized information and consumer-focused design and features

Identifying risk through clinical content and decision support tools for evolving needs

Promoting actions that improve outcomes by connecting patients with available resources and personalized interventions, resulting in:

• • promoting early prenatal care;
  • reducing the neonatal intensive care admissions;
  • reducing C-sections; and
  • lowering the frequency of low birthweight babies.

Results of the program were published by Bush et al, Impact of a Mobile Health Application on User Engagement and Pregnancy Outcomes Among Wyoming Medicaid Members in the Journal of Telemedicine and E-Health, 2017 Nov 1; 23(11): 891–898.

Method/ Calculation

Data was gathered for users (N=85) and non-users (n=5,158) of the app in the Medicaid population.  The two groups were analyzed to ensure a similar distribution of ages. Chi-Square tests were performed on the association between app use and the four targeted outcomes.  Odds ratios, 95^th percentile confidence intervals, and p values were also calculated for each.

Findings / Metric/ Outcome/ Savings

See Table 3 from the Bush et al article for summary of results.  The app users were significantly more likely to have started prenatal care six months or more before delivery.  The difference between app users and non-users on low birthweight babies was almost significant. For the other outcomes, the difference between users and non-users was not significant.

Limitations

As the program was voluntary, self-selection by participants may have colored the results.  Random assignment to the program or matching the participants to a control group might show different outcomes.