VitreosHealth unveils exciting new predictive models for ‘hidden, hard to find’ risk factors for population health management.
VitreosHealth unveils exciting new predictive models for ‘hidden, hard to find’ risk factors for population health management
New models are some of the first to be applied in population health for better outcomes.
Sept 16, 2016.Today VitreosHealth, the leader in predictive and prescriptive health insights for population health management,announced the development of new predictive risk models that are among the first to be used for population health. These new models represent VitreosHealth’s commitment towards continuous innovation in delivering high impactsolutions which captures the unique attributes of its‘clients’ populations.
Gaps-in-Care predictive model
Historically, all gaps-in-care were considered equally important. VitreosHealth has analyzed millions of lives over several years, and the intervention analysis reveals that some gaps are more important than others. Addressing those will result in improved outcomes.
The new “Gaps-in-Care” predictive model identifies for each high risk population cohort the 2-3 gaps-in-care which when closed, will result in higher quality outcomes while lowering the high-cost interventions in the shortest possible duration. At a time when most Payviders have limited care management resources, it is crucial for them to target the right patients and the right gaps to maximize the return on investment on their care management programs.
MENTAL HEALTH predictive model
Multiple studies performed over the last ten years have revealed that patients with mental health cost 3-5 times more than patients without mental health. Yet, many mental health conditions are undiagnosed, or diagnosed late. VitreosHealth has developed a new ‘Onset of Mental Illness’ predictive model that predicts which patients have a higher likelihood of the onset of a mental illness (Anxiety, Depression, etc.) that could result in significant non-compliance issues in their chronic care management programs. After analyzing the multitude of variables, the model identifies15 key variables with weight ages to predict high risk mental illness members. Such insights allow VitreosHealth’s clients to develop care management programs that incorporate necessary screening and assistance by a licensed behavioral health counselor.
Patient Motivation Index Model
Changing a patient’s behavior is notoriously complicated. Even after identifying a patient who is likely to have an adverse event soon, it is often very difficult to engage the patient. VitreosHealth has now developed an exciting model ‘The Patient Motivation Index’ predictive model which predicts those patients with a higher propensity to actively participate in closing their Gaps-in-Care. After initially considering hundreds of clinical and non-clinical variables, the model uses 10-15 risk factors to identify the patients with the greatest likelihood for positive engagement.
VitreosHealth has also announced that predictive Palliative care models will be commercially available by December 2016. Palliative and Hospice Care have emerged as the most expensive categories for Medicare & Medicaid ACOs. These categories are not being managed properly and care is being delivered in a highly fragmented manner across many SNFs and Hospices. Being able to predict patients that may need palliative care in the future can enable Payviders to plan care to be provided in a way that can improve outcomes and reduce costs.
“While traditional population health efforts focus on closing the most gaps, the reality is that the impact can be very limited overall. All gaps are not equal,” says Kirit Pandit, co-founder and Chief Risk Model Officer. “By developing these new models, we are able to determine which gaps to close with the greatest overall impact, in terms of cost and population health.”
All of these new predictive risk models will be incorporated into VitreosHealth’s underlying predictive & prescriptive analytics engine available to all clients and their populations.