Population Analytics for ACOsThe biggest challenges for ACOs are:
- Population health management across the continuum-of-care
- Patient attribution
- Demand planning for its specialist resources, procedures and facilities
- Keeping patients within the Network with better access to care
Population Health Management:
For years, healthcare insurance companies (payers) have mined claims data for chronic patients and have built predictive models to identify high-risk patients. Armed with historical reports, case managers designed intervention programs that were meant to prevent complications among chronic patients and reduce ER visits and hospitalizations. While this approach has seen some success, limitations far outweigh merits. Most primary care physicians are highly skeptical of claims-based predictive models because they have no latest clinical basis, and give no consideration to the current state of health of the patient. Moreover, there is a complete lack of causation. Why a patient is considered high-risk? What are the clinical reasons for the high risk score? How do we lower the risk score? How does the score measure the effectiveness of my care management program?
Most claims-based analysis use coding from claims as the crux for the predictive models. Providers are skeptical of this approach as the objective of coding is to maximize revenue and not to capture the exact state of health of the patient
“Care management analytics continues to advance in adoption and sophistication throughout the payer market, but several factors have led to diminished expectations. To date, the value-limiting factor in care management programs has not been identifying members with opportunity for improved outcomes. Instead, the difficulty is getting identified members engaged in an appropriate program. As a result, expectations for care management analytics have declined, and technology is approaching the “trough of disillusionment’”…..Gartner Hype Report 2011
PSCI Solution: To address this gap, PSCI has built the Next Generation Population Health Predictive Models in partnership with physicians and clinicians over the last 4 years. These models mine the latest provider EMR data along with claims, financial and quality data. The models captures the patients latest State-of-Health (SOH) by chronic condition, compliance to best practices, prior patient utilization trends and socio-economic factors in determining the Risk of Hospitalization for the patient population. This approach helps systematically peel the factors driving the high risk of hospitalization to determine the drivers to design effective care management programs. To date we have built in excess of 20 models for chronic diseases and conditions covering most of the high frequency and high cost disease conditions.
The Multi-dimensional interactive analysis tool helps analyze each patients use of resources from multiple dimensions across the continuum-of-care to determine the right patient attribution to a physician.
To understand PSCI's EMR-based approach to Next Generation population risk stratification and predictive analytics, please request a demo and read white paper Population Health Management: Real-Time State-of-Health Analysis Solutions.Patient Attribution:
One of the most critical problems ACOs face is how to assign a patient to a physician or a team of physicians who would be responsible for caring for the patient across the continuum of care. This assignment is a very important decision. The most common methodology is to use historical claims and identify the physicians and specialists with whom the patient has had most visits. While this approach is useful, it is also retrospective. It assumes that the patient will need to see the same specialists in the future. This assumption is erroneous. A correct approach is to use predictive risk scores and identify what health issues a patient is likely to have in the future. For example, a 70 year old woman with no history of osteoporosis may be at a high risk for osteoporotic complications in the next 5 years, and may need to have an orthopedic and spinal surgeon in her assigned team.
PSCI Solution: PSCI's predictive risk scores coupled with multi-dimensional analysis allows attribution based on a number of criteria such as (a) past physician use (b) future expected physician use (c) geographical proximity to specialist (d) socio-economic factors and others.Demand Planning:
The success of an ACO depends on optimal design of the ACO Network. ACOs need to have right number of PCPs, specialists at the right locations to ensure access to quality care for their population and at the same time avoid excess capacity. Successful ACOs need to:
- Design an optimal provider network with the right mix of PCPs and specialists, with affiliate specialist centers located according to population risk analysis;
- Develop very focused and successful marketing programs to capture the captive market and ensure there is no leakage in terms of your population seeking care outside the ACO facilities.
- Measure, track and monitor patient population state-of-health scores by chronic condition to proactively reach out to patients for better access to care and to ensure they are within the ACO network.
PSCI Solution: PSCI is the only solution in the market that uses the population State-of-Health (SOH) scores to predict the demand for specialists (cardiologists, Endocrinologists, etc.) for an ACO by specific region or by zip code. PSCI application also helps in demand planning for specialist procedures and potential hospitalizations over the next 1-2 years. All this is based on the current ACO population SOH scores by chronic condition.
Gone are the days where planning to determine the number of cardiologists your ACO needs and where to locate the specialty centers were based on population census data. We have saved our customers millions of dollars by helping them avoid make wrong investment decisions. Our demand planning tool predicts the demand for specialist resources, procedures, facility usage based on the current state of health and the risk of hospitalization data by each chronic condition.Access to care:
Various recent studies have shown that out-of-network visits and ER visits can be reduced if patients have better access to care. A patient encountering a health ailment who cannot reach his PCP/specialist/nurse on phone is much more likely to call 911 or show up at the Emergency Department. Socioeconomic factors also play a big role in determining who needs more access to care. An affluent, highly educated, married male is perhaps less likely to call 911 than a single, unemployed male suffering from depression. Is it important for ACOs to segment their populations into groups that require better access to care.
PSCI Solution: PSCI uses very precise socio-economic data to calculate 'access of care' scores for each patient. By combining access-of-care requirements with clinically calculated risk scores, ACOs can provide better access to those patients who need it most.