Learn How

Learn how AI can help HCOs to promote early identification of frailty, address functional decline in frail patients, and reduce the incidence of adverse events. Discover how AI can help to identify frailty in the absence of standardized measures, distinguish high risk patients in need of enhanced support, and predict adverse events and complications.

Automatically ingest data from dozens of health data sources including...

Medical Claims

Data extracted from health insurance medical claims with details about dates and place of service, diagnosis codes, key procedures, use of medical equipment, and provider specialties.

ADT Records

Data from Admit, Discharge, and Transfer feeds and patient data notification services that synchronize patient demographic, diagnostic, and visit information across healthcare systems.

Social Needs Assessments

Self-reported data that identify an individual's specific needs and the acute social and economic challenges they are experiencing.

Closedloop icon with a line

ClosedLoop generates explainable predictions using

thousands of auto-generated, clinically relevant contributing factors

Jeremy Banks
84-Year-Old Male
Risk of a Serious Fall-Related Injury in the Next 6 Months
Risk Score Percentile
Impact on Risk  |  Contributing Factor
+28% | Frailty Score
+19% | # of Unplanned Admissions (12M)
+13% | # of Units of Durable Medical Equipment (6M)
2 to 5
+10% | Levels of Caregiver Support

What Is Frailty?

Frailty is an aging-related state of decreased physiological reserve that results in increased vulnerability to poor health outcomes, worsening mobility and disability, hospitalizations, and mortality.¹ Long recognized within the field of geriatrics as a clinical syndrome, frailty occurs in approximately 25% of people aged 65 and over.² Frailty also contributes greatly to cost of care. The estimated annual cost directly attributed to frailty is more than $14,000 per patient after controlling for other variables.³

Why It Matters

People who are frail have a decreased ability to maintain or return to homeostasis after stressful events or aggressive interventions.  This often leads to loss of independence.⁴ Compared to non-frail patients, they are 15% more likely to develop inpatient complications, and in the year following treatments for critical illness, have 30% increased disability in daily activities.¹𝄒⁵ For older adults, frailty may be a better predictor of mortality than age. It doubles the risk of in-hospital mortality and increases the risk of one-year mortality by 50%.⁵𝄒⁶  

Even though frailty is a measurable phenotype that can be identified with standardized measures, such as unintentional weight loss of 10 pounds in the past year, these performance measures are not routinely captured in clinical encounters.⁷  Frailty is also not captured in administrative claims databases.  But, validated claims-based algorithms have shown good discrimination of frailty and high predictive ability with adverse health outcomes. These algorithms present a powerful tool to enable provider identification and risk assessment of frail patients.⁷𝄒⁸

AI Presents an Opportunity

Early identification of frailty via predictive analytics is vital, as the evidence shows that frailty can be managed and reduced.  Interventions designed to improve nutrition, stimulate cognition, and promote physical activity can reduce frailty between 35 and 45%.⁹𝄒¹⁰  Frailty also has greater reversibility than disability, and prioritizing screening can provide critical information to identify poor prognosis and reversible risk factors.  

Did You Know…

  • 25% frailty occurs in approximately 25% of people aged 65 and over²
  • Half as likely frail patients are half as likely to be discharged to home¹¹
  • 20% frail patients are roughly 20% more likely to be readmitted within 12 months¹¹
  • $14,000 is the estimated annual cost of care per patient attributed to frailty³
Citations & Footnotes

1. Bellal, Joseph, et al. “Superiority of Frailty Over Age in Predicting Outcomes Among Geriatric Trauma Patients: A Prospective Analysis.” JAMA Surgery, vol. 8, no. 149, Aug. 2014, pp. 766–772, doi:10.1001/jamasurg.2014.296. 

2. Muscedere, John, et al. “The Impact of Frailty on Intensive Care Unit Outcomes: A Systematic Review and Meta-Analysis.” Intensive Care Medicine, vol. 43, no. 8, 2017, pp. 1105–1122, 10.1007/s00134-017-4867-0.

3. Simpson, Kit N., et al. “Effect of frailty on resource use and cost for Medicare patients.” Future Medicine, vol. 7, no. 8, 29 May 2018. doi.org/10.2217/cer-2018-0029.

4. Hubbard, Ruth E., et al. “Frailty Status at Admission to Hospital Predicts Multiple Adverse Outcomes.” Age and Ageing, vol. 46, no. 5, 22 May 2017, pp. 801–806, 10.1093/ageing/afx081.

5. Brummel, Nathan E., et al. “Frailty and Subsequent Disability and Mortality among Patients with Critical Illness.” American Journal of Respiratory and Critical Care Medicine, vol. 196, no. 1, 2 Dec. 2016, pp. 64-72, doi:10.1164/rccm.201605-0939OC.

6. Bagshaw, Sean M., et al. “Association between frailty and short- and long-term outcomes among critically ill patients: a multicentre prospective cohort study.” Canadian Medical Association Journal, vol. 186, no. 2, 25 Nov. 2013, pp.95-102. doi:10.1503/cmaj.130639.

7. Segal, Jodi B, et al. “Development of a Claims-Based Frailty Indicator Anchored to a Well-Established Frailty Phenotype.” Medical Care, vol. 55, no. 7, 2017, pp. 716–722, 10.1097/MLR.0000000000000729.

8. Cuthbertson, Carmen C, et al. “Controlling for Frailty in Pharmacoepidemiologic Studies of Older Adults: Validation of an Existing Medicare Claims-Based Algorithm.” Epidemiology (Cambridge, Mass.), vol. 29, no. 4, 2018, pp. 556–561, 10.1097/EDE.0000000000000833

9. Apóstolo, João, et al. “Predicting Risk and Outcomes for Frail Older Adults: An Umbrella Review of Frailty Screening Tools.” JBI Database of Systematic Reviews and Implementation Reports, vol. 15, no. 4, 2017, pp. 1154–1208, 10.11124/JBISRIR-2016-003018.

10. Ng, Tze Pin, et al. “Nutritional, Physical, Cognitive, and Combination Interventions and Frailty Reversal Among Older Adults: A Randomized Controlled Trial.” The American Journal of Medicine, vol. 128, no. 11, 1 Nov.  2015, pp.1225-1236, doi:10.1016/j.amjmed.2015.06.017.

11. Castillo-Angeles, Manuel, et al. “Association of Frailty With Morbidity and Mortality in Emergency General Surgery By Procedural Risk Level.” JAMA Surgery, 25 Nov. 2020, 10.1001/jamasurg.2020.5397.

ClosedLoop is an exciting and important partner in our strategy to develop timely predictive insights through operationalized AI solutions that will drive member engagement and better health outcomes.

Pat Wang
President & CEO, Healthfirst

Our partnership with ClosedLoop adds the depth of AI to more accurately identify those patients who are most likely to benefit from our primary care interventions as well to evaluate these interventions and make more relevant and timely modifications to improve their effectiveness.

David Klebonis
Chief Operating Officer, PBACO

We are extremely impressed with the predictive modeling capabilities the ClosedLoop platform has delivered. The ClosedLoop team has exceeded all defined goals and benchmarks set to date and we anticipate a substantial return on our investment as these predictions are operationally deployed.

Cheryl Lulias
President & Executive Director, MHN

We’re able to store and operationalize analytics directly from ClosedLoop. That’s driving real value—it’s accelerated the implementation of key insights into clinical workflows and it allows us to more easily account for all of the different factors that influence intervention decisions.

Christer Johnson
Chief Analytics Officer, Healthfirst

At SWHR, our patient-centered network is committed to innovative care models that are value-based, high-quality, and data-driven. Because of our partnership with ClosedLoop, we’re better able to identify and act on the most impactful opportunities for our physician partners to improve patient results and reduce unnecessary costs.

Dr. Jason Fish
Chief Medical Officer & SVP, Southwestern Health Resources

We compared the ClosedLoop predicted spend to the spend predicted by a leading rules-based risk prediction engine and found that 75% of the time ClosedLoop was closer, and in most cases significantly closer, to the actual spend.

Phillip Bruns
Chief Technology Officer, CareATC

Learn What ClosedLoop Can Do for Your Organization

The industry’s best collection of customizable predictive models for common healthcare use cases.

Talk To An Expert