Value-based performance analytics built around the things that actually determine the result: who is attributed to you, how the benchmark was set, where cost is genuinely concentrated, and which quality measures are a gate rather than a lever.
Most ACO and Stars programmes are managed as quality programmes with a cost report attached. That gets the emphasis backwards in shared savings and inverted for Stars. CaliberFocus builds the analytics around the specific economics of your arrangement, unifies the care gap work that is currently being run four separate times against the same physicians, and focuses cost effort on the small share of patients and drivers where it can move.
In shared savings, success this year makes next year harder. Any strategy that ignores that arithmetic works once.
Value-based arrangements ask providers to manage total cost and quality for a population whose membership is not fully known during the performance year, against a benchmark derived partly from their own history, judged on quality thresholds that are frequently set retrospectively. Every one of those is manageable. None of them is managed by a quality dashboard.
The benchmark ratchets
Historical spend informs the benchmark. Reducing cost successfully lowers the bar you are measured against next time.
Attribution moves under you
Patients enter and leave the attributed population during the year. You are managing a denominator that is still changing.
Quality is a gate in one programme and a lever in another
Treating shared savings and Stars quality identically misallocates effort in both directions.
Cut points are a moving target
Managing to last period thresholds means aiming at where the target used to be.
Four programmes, four gap lists, one physician
The same patient can receive the same clinical request from multiple teams working from different lists.
Cost is concentrated and effort is not
A small share of patients and a small number of drivers account for most controllable spend, while programme effort is often spread evenly.
A quality dashboard is not a value-based strategy.
Quality performance is necessary and measurable, which is why it absorbs attention. In shared savings, total cost of care against a benchmark is driven by post-acute placement, avoidable utilization, network leakage and a comparatively small group of complex patients.
Quality performance is necessary and measurable, which is why it absorbs attention. In shared savings, total cost of care against a benchmark is driven by post-acute placement, avoidable utilization, network leakage and a comparatively small group of complex patients.
The first question in any engagement is how your arrangement actually pays. Benchmark construction, risk track, attribution method, quality gate mechanics and the treatment of high-cost outliers determine where effort returns anything.
Step 1
Model the arrangement economics
Benchmark construction, trend, risk adjustment treatment, attribution method, quality gate mechanics and outlier handling.
Step 2
Establish the attributed population and volatility
Who is in, how they arrived, who may leave, and what the population looked like in prior periods.
Step 3
Concentrate the cost analysis
Identify the patients and drivers accounting for most controllable spend.
Step 4
Separate the quality gate from the quality lever
Establish what must be passed, what genuinely increases the result, and what is being over-invested in.
Step 5
Unify the gap lists
One prioritized list per patient serving every programme, delivered once to the physician.
Step 6
Stratify and match the intervention
Different segments need care management, outreach, pharmacy intervention, post-acute redirection—or nothing at all.
Step 7
Address leakage and network integrity
Target a major controllable cost lever that is frequently under-managed.
Step 8
Monitor in-year against a moving target
Use scenario ranges rather than one projected
result.
result.
Step 9
Reconcile at settlement
Compare projected to actual, understand every variance, and carry the learning into next-cycle assumptions.
We will tell you which measures to stop working.
Some measures are already comfortably past the threshold, cannot move enough to earn anything, or are too small to affect the result. Continuing to work them consumes clinical goodwill needed elsewhere.
Some measures are already comfortably past the threshold, cannot move enough to earn anything, or are too small to affect the result. Continuing to work them consumes clinical goodwill needed elsewhere.
The technical foundation is a single reconciled view of the attributed population, with cost, quality, risk and utilization on the same patients.
Population and Economics
Attribution Management
Reconstruct how the population is formed, track entry and exit, and reconcile against the payer or CMS view.
Benchmark and Financial Modeling
Model benchmark construction and the effect of current performance on future cycles.
Cost Concentration Analysis
Identify controllable spend by patient segment, driver and setting.
Leakage and Network Analytics
Understand where attributed patients receive care outside the network, what it costs and what is genuinely addressable.
Quality and Gaps
Unified and Multi-Gap Identification
One prioritized gap list per patient across ACO quality, Stars/HEDIS-type measures, MIPS and risk documentation.
Gate Versus Lever Analysis
Distinguish measures that must be passed from measures that actually increase the result.
Cut Point and Threshold Tracking
Monitor moving thresholds and use scenario ranges around where a measure may land.
Medication Adherence Analytics
Support fill-gap identification, pharmacy coordination and targeted outreach.
Operate and Govern
Risk Stratification and Intervention Matching
Segmentation ending in a specific intervention per segment, including the segment where the correct intervention is none.
Provider and Network Performance
Practice and clinician-level cost, quality and leakage views adjusted for panel complexity.
In-Year Projection and Scenarios
Projected settlement position with ranges and assumptions visible.
Intervention Effectiveness
Completion and outcome rates by outreach method, workflow and intervention.
What CaliberFocus does, and does not do?
We are not a care management vendor and we do not run your outreach. We build the analytics and the population view that decide where care management should be pointed. We will also tell you which parts of a Stars rating a provider organization can genuinely influence and which belong to the plan.
We are not a care management vendor and we do not run your outreach. We build the analytics and the population view that decide where care management should be pointed. We will also tell you which parts of a Stars rating a provider organization can genuinely influence and which belong to the plan.
Provider organizations influence some value-based domains directly, some partially, and some not at all. The provider-influence column should drive resourcing.
| Domain | What Moves It | Provider Influence |
|---|---|---|
| Preventive and screening measures | Pre-visit gap identification, standing orders, outreach, and closing the loop on completed care not captured | High. Largely operational and highly winnable |
| Chronic condition control | Panel management, medication titration, follow-up cadence and documentation of results in a readable field | High, though slower than screening |
| Medication adherence | Fill gap identification, pharmacy coordination, synchronization, targeted outreach before a gap becomes a failure | High, and disproportionately weighted in Stars |
| Avoidable utilization | Access, after-hours availability, ED alternatives, post-discharge follow-up within days | High, and the largest controllable cost lever alongside post-acute |
| Post-acute placement and length of stay | Preferred network design, placement decisions at discharge, and active management during the stay | High, and routinely under-managed |
| Network leakage | Referral workflow, specialist availability, patient navigation and visibility into where care is going | Moderate to high, and the least actively managed of the controllable levers |
| Transitions and readmissions | Discharge process, medication reconciliation, follow-up scheduling before discharge | Moderate to high |
| Patient experience | Access, communication, care coordination and continuity, which move slowly and lag the intervention | Moderate, with a long feedback delay |
| Plan operations measures | Call centre performance, appeals handling, enrolment and plan administration | Low to none. These belong to the plan and should be resourced accordingly |
Open and closed is not enough. The two most commonly skipped states are actionable, meaning there is still time and a realistic route to close it, and verified, meaning the source data the measure reads now reflects it.
Eligible
Open
Actionable
Routed
Attempted
Completed
Verified
Reported
One Gap List, Not Four
The same patient frequently appears on MIPS, ACO quality, Stars/HEDIS-type and risk documentation gap lists generated by different teams. The clinical action is often identical. Consolidating them into one prioritized list per patient reduces clinician burden and improves performance at the same time.
The same patient frequently appears on MIPS, ACO quality, Stars/HEDIS-type and risk documentation gap lists generated by different teams. The clinical action is often identical. Consolidating them into one prioritized list per patient reduces clinician burden and improves performance at the same time.
Attribution decides which patients count, which costs count against you and which quality gaps matter. It is not fixed, and its volatility has direct financial consequences.
Reconstruct the logic
Understand precisely how attribution is determined in your arrangement and what service patterns drive it.
Track movement through the year
Patients entering and leaving, with the cost and quality implications of each.
Identify influenceable attribution
Where continuity of primary care drives attribution, that is an operational lever rather than an administrative fact.
Report the movement, not just the total
Starting population, additions, removals, provider movement and denominator change each period.
Attribution decides which patients count, which costs count against you and which quality gaps matter. It is not fixed, and its volatility has direct financial consequences.
CR
Clinical Risk
Complexity or deterioration risk routed to intensive care management for a deliberately small group.
UR
Utilization risk
Likelihood of avoidable acute or high-cost care routed to targeted intervention and access work.
GO
Gap opportunity
Actionable gaps closable through normal practice workflow.
ER
Engagement risk
Hard-to-reach and overdue populations requiring an outreach design response.
TR
Transition risk
Patients requiring support after discharge or movement between settings.
NI
No intervention
The segment where none is warranted, preserving limited care-management capacity for the patients who need it.
Adjust for the panel
Complexity, payer mix and social factors need accounting for before comparison.
Lead with the actionable
Leakage, post-acute patterns and referral behaviour are things a practice can change.
Route through network leadership
Practice performance conversations come from network and clinical leadership, not from an analytics team or finance report.
The largest gap is rarely the largest opportunity. Prioritization needs the arithmetic rather than intuition.
Value and proximity to threshold
What a realistic change is worth under the arrangement, including whether a gate has already been passed.
Population size and reachability
Enough eligible patients to move the rate, with a realistic route to reach them within the period.
Effort, capacity and time remaining
Score opportunities against the clinical capacity actually available and the time left to achieve a documented result.
Cross-programme value
A single action closing gaps in several programmes ranks above one closing a gap in only one.
Act now
High value, time sensitive, actionable, with capacity available.
Plan
Important, with a defined future intervention point such as a scheduled encounter.
Monitor
Worth watching, not worth working yet.
Do not chase
Low value or not actionable in the time remaining. Naming these explicitly frees capacity for the first category.
Benchmarking, With the Caveats Visible
Case mix, payer mix, social risk and market structure can shift results. We show the adjustment basis alongside any benchmark, and where a comparison cannot be adjusted honestly, we say so.
Value-based arrangements are settled on someone else’s calculation. Continuous reconciliation against payer and programme files protects the result before variances become settlement surprises.
Reconciliation
- Attributed population reconciled continuously against payer or programme view
- Quality results reconciled against the calculating entity output
- Settlement compared with projection and variances carried into next-cycle assumptions
Measure and logic control
- Measure logic versioned by programme and performance year
- One governed definition layer serving every programme
Data currency and identity
- Data freshness visible on every worklist
- Patient identity resolved before attribution and cost analysis
- Provider identity consistent across payer, EHR and analytics environments
- Claims completeness monitored
Access and oversight
- Clinician-identifiable performance data access controlled
- A named owner accountable for the result
- Clinical leadership involved in prioritization
- Compliance visibility where quality and risk documentation overlap
The gap between your number and their number is information, not an annoyance.
Variances usually trace to attribution differences, claims lag, exclusion handling or a measure specification applied differently. Knowing that during the year is valuable; discovering it at settlement is a surprise.
Data Owner
Source integrity
Measure Owner
Definition and interpretation
Operational Owner
Acts when performance moves outside range
Clinical Owner
Clinical interpretation or intervention
Reporting Owner
Reconciliation and external reporting
A single-year result can be produced by a favourable benchmark, a mild season or a one-off. What matters is whether the programme built anything that works again next cycle, when the benchmark has moved against you.
| Category | What We Measure | Why It Matters |
|---|---|---|
| Financial Result | Performance against benchmark, and projected effect of this year's result on the next benchmark | A saving that makes next cycle harder is different from a structural one |
| Cost Drivers | Movement in avoidable utilization, post-acute cost and length of stay, leakage rate | Where a repeatable result actually comes from |
| Quality Position | Gate measures passed with margin, lever measures relative to threshold, measures deliberately stopped | Distinguishes required performance from performance that earns |
| Gap Efficiency | Gaps closed per clinician contact, duplication removed across programmes | The clinician-burden measure and a core sustainability indicator |
| Attribution Stability | Population volatility, reconciliation variance against the payer view | The denominator underneath every other number |
| Intervention Effectiveness | Completion and outcome by outreach method and workflow | Shows where capacity should be increased and where withdrawn |
| Actionability | Share of surfaced opportunities reaching an owner with enough time to act | Analytics without operational ownership is reporting |
| Projection Accuracy | In-year projection against settled result, and variance explained versus unexplained | Whether leadership can plan on the reporting |
The first year is usually the easiest.
Early savings often come from correcting obvious leakage, fixing post-acute placement and closing gaps nobody was working. Those are one-time gains that also lower the benchmark for the next cycle. Plan for the second and third year in the first.
Early savings often come from correcting obvious leakage, fixing post-acute placement and closing gaps nobody was working. Those are one-time gains that also lower the benchmark for the next cycle. Plan for the second and third year in the first.
We will model how your arrangement actually pays, show you where controllable cost is concentrated, identify the measures worth working and the ones worth stopping, and quantify the duplication in your current gap lists. That analysis takes weeks and it usually reorders the priority list substantially.
Bring us a specialty or clinical setting where clinicians are spending too much time creating notes. We will assess where ambient documentation fits, what must remain clinician controlled, how it should integrate with your EHR, and how to measure whether it is actually reducing burden.
One conversation with people who have run these deployments, and a written readiness view you can use with or without us.
- AI Agents and Workflow Automation
- Voice and Conversational AI
- Document AI and Intelligent Processing
- Generative AI and Enterprise Copilots
- AI Strategy and Governance
- HCC and Risk Adjustment Analytics
Security & Compliance
