Patient Engagement

To meet our mission of addressing the needs of vulnerable populations, we also work with Medicaid payers. In collaboration with payer partners, including the Parkland Community Health Plan, we leverage modern machine learning and artificial intelligence (ML+AI), digital tools, and social determinants of health (SDOH) data to impact quality, improve outcomes, and reduce inappropriate costs.

We direct health plans to members with the greatest needs and provide technology-enabled strategies for effective engagement. Our results demonstrate that our delivery of risk-stratified engagement programs has a positive impact on clinical, financial, and operational measures.


  • Our Risk-Stratified Patient Engagement Offerings identify at-risk individuals (earlier in the process) and provide technology-enabled outreach and patient engagement strategies. Our risk prediction models work better when compared to other methods for identifying at-risk individuals because we factor in SDOH variables in addition to analyzing clinical and claims data.

    • Pediatric Asthma – A program utilizing a real-time predictive model that proactively― and dynamically― identifies very high-, high-, and medium-risk pediatric asthma patients for targeted, direct decision support interventions. Our program incorporates multiple touch points and messaging modes to improve medication compliance, reduce school absenteeism, and lower ED/hospital utilization.

    • Pre-term Birth Prevention A program utilizing a real-time predictive model that identifies pregnant women who present at high risk for pre-term delivery. The physician, case manager, and patient are connected via various messaging modes (e.g., EHR alerts, texts, apps) that work to collect and share patient-specific information to reduce the probability of pre-term birth or post-delivery complications.

  • Our Dynamic Management Dashboards offer insights and tools that are necessary for effectively managing value-based care contracts. We have the requisite technology infrastructure to ingest diverse clinical and non-clinical datasets containing either structured or unstructured data. Data outputs are synthesized, simplified, and presented in user-friendly dashboards. Key dashboard features include:

    • Geolocation to the county, zip code, or block level
    • Multiple visualization
    • Collaboration features
    • Nimble, rapid refresh


Goal: Decrease preventable asthma-related ED visits. Hospitalizations, and costs through risk stratification and care re-design.


  • The predictive model risk-stratified the children into different groups based on the likelihood that their asthma would exasperate over the next 3 months and require an ED visits or hospitalization.
  • Proactive outreach to physician practices and care managers.
  • Risk-driven, text-messaging-based patient engagement.

4 Year Impact

  • ~30,000 children with asthma impacted/year
  • 800 high-risk children with asthma on text messaging
  • 50% reduction in annual asthma costs (PMPM)
  • 30% reduction in annual asthma ED visit rates
  • 30-40% reduction in asthma hospitalization rates
  • $24,000,000 direct cost savings


Goal: Decrease pre-term rates & costs through risk stratification and care re-design.


  • The predictive model gathers data from diverse sources to drive early identification of at-risk pregnant women. This model then risk-stratify these women as they moved through their pregnancies on their chances of having an early delivery.
  • Risk-driven, text-messaging-based patient engagement to drive pre-natal visit attendance and offer pre-natal education.

One Year Impact

  • 23,584 pregnancies risk-stratified/year
  • >800 members on text messaging
  • 54% reduction in PMPM baby costs
  • 27% reduction in PTB <35 weeks
  • 24% increase in prenatal care rates
  • $1,000,000 direct cost savings