What the current evidence shows

Success is demonstrable but contingent, heterogeneous, and sensitive to organizational form. CMS’s PY2024 results are an important counterweight to simplistic failure narratives, yet per-capita results differed markedly by organizational type. These descriptive differences do not prove causality, but they signal that governance, proximity to primary care, capital structure, and delivery-model design shape performance.

Evidence synthesis approach

This report is a targeted rapid evidence synthesis, not a formal systematic review. It integrates recent systematic and scoping reviews, empirical implementation studies, implementation-science frameworks, and authoritative CMS performance data. Priority was given to peer-reviewed literature on implementation mechanisms rather than commentary on the aspiration toward value-based care. Definitions vary, much implementation evidence is observational, and program results are descriptive; convergence across evidence streams is therefore more decision-relevant than any single study.

The failure architecture

Seven mechanisms interact to produce an operating-model gap. The literature repeatedly describes implementation as partial, interpretively variable, and weakly evaluated: in a scoping review of 62 publications, most hospitals implemented only one or two components of the value agenda (van Staalduinen et al., 2022; Khalil et al., 2025).

Failure is systemic: the reinforcing loop

No single barrier explains performance. Incentive conflict amplifies resistance; poor data increases friction; workflow friction weakens clinician trust; weak trust reduces adoption; and low adoption makes financial performance appear to be a data or contracting problem, restarting the cycle. The traveling pulse below traces that reinforcing loop.

Economic contradiction and model complexity

Organizations cannot consistently optimize volume and value with the same governance logic. Many health systems enter value-based arrangements while the majority of revenue, productivity targets, capital allocation, and physician compensation remain anchored in fee-for-service. This is not merely an incentive-design defect; it is an unresolved enterprise strategy.

The dual operating system

The organization is asked to reduce avoidable utilization while service lines are measured on volume and contribution margin. Leaders may publicly sponsor value while operationally protecting volume, producing rational caution among clinicians and managers. Contractual complexity compounds the problem: value-based purchasing programs differ in attribution rules, risk corridors, quality measures, benchmark methodology, reconciliation timing, and downside exposure (Pandey et al., 2023). When contracts send different signals, clinicians experience value-based care as an administrative overlay rather than a coherent clinical strategy. Attribution lag and retrospective reconciliation further weaken the connection between today’s decision and next year’s result. Risk can also be mispriced or unfairly distributed: financial risk may be transferred to clinical units without the resources or authority to change the determinants of utilization (Kim et al., 2022).

Culture, behavior, and clinical ownership

The decisive implementation unit is the care team, not the contract office. Value-based care asks clinicians to alter established ordering, referral, documentation, follow-up, and team-practice behaviors embedded in professional norms, local routines, specialty identities, and time constraints. A technically correct dashboard does not automatically overcome habit, clinical uncertainty, patient expectations, throughput pressure, or fear of missed diagnosis.

Engagement is not activation

Gray et al. (2020) argue that transformation requires substantive physician leadership and engagement. Daniels et al. (2022), drawing on 43 interviews across eight hospital improvement teams, identified multidisciplinary engagement, medical leadership, practical team organization, organizational structure, and integration with existing improvement work as central determinants of success.

Culture becomes visible in local choices

• Whether variation is discussed as a learning opportunity or a threat to autonomy
• Whether physicians trust risk adjustment, data lineage, and peer comparisons
• Whether care managers can influence physician decisions or merely work around them
• Whether leaders remove low-value tasks when they add new value-based work
• Whether improvement teams can test, learn, and adapt without waiting for annual contract reconciliation

A practical behavioral architecture

Workflow integration and data-to-action failure

A dashboard outside the workflow is often a report about yesterday, not a decision tool for today. Organizations commonly invest in population-health platforms, registries, risk stratification, and quality dashboards, yet the resulting information may be accurate and operationally inert. Health-IT literature shows positive associations with ACO participation but mixed relationships between technical capability and performance, reinforcing the distinction between infrastructure and use (Balio et al., 2019).

The four conversion failures

Value creation requires that data survive four sequential conversions. The progress line traces the chain and flags where each conversion breaks.

Minimum viable operational data

Patient value, access, and equity

Outcome measurement can still miss what patients value and what prevents them from receiving care. In a systematic review of 21 studies, consultation via questionnaires or interviews was the dominant form of involvement; advisory roles, co-design, and collaborative teams were rare, and no examples of patients co-leading improvement committees were found (van der Voorden et al., 2023).

Three ways patient value is diluted

Equity is an implementation property

The updated CFIR explicitly strengthens attention to innovation recipients and equity-related determinants (Damschroder et al., 2022). Equity cannot be relegated to retrospective stratification: transportation, digital access, language, caregiving burden, housing instability, and distrust may alter reach, adherence, and outcomes. Payment designs that ignore these determinants can penalize organizations serving high-need populations or encourage risk selection (Kim et al., 2022). Cross-country evidence identifies infrastructure limitations, data scarcity, fragmented systems, limited delivery capacity, and financing constraints as barriers to value-based implementation (Touchton et al., 2026).

Questions the patient partnership should answer

• What does a good outcome mean to the people receiving this care?
• Where does the pathway create avoidable time, cost, uncertainty, or coordination burden?
• Which implementation strategies are feasible for patients with different resources and constraints?
• What unintended consequences would not be visible in claims or clinical outcome measures?

The radiology operations lens

Radiology shows why value-based care cannot be reduced to utilization control. Imaging is both a cost center and a clinical decision infrastructure. Reducing scans without protecting access, accuracy, and follow-through can shift cost downstream; optimizing scanner throughput alone can preserve inappropriate ordering, delay, protocol variation, and an open diagnostic loop.

The end-to-end radiology value pathway

High-value radiology requires balanced measures

Govern implementation as a clinical capability

Outcomes are lagging indicators; implementation outcomes explain whether the model is taking hold. The common governance error is to monitor cost and quality while assuming implementation is occurring. The updated CFIR organizes determinants across the innovation, outer setting, inner setting, individuals, and implementation process (Damschroder et al., 2022); the implementation-outcomes literature emphasizes acceptability, adoption, appropriateness, feasibility, fidelity, cost, penetration, and sustainability (Proctor et al., 2023).

The executive learning system

• Diagnose determinants before selecting interventions; do not assume every low-performing site has the same barrier
• Define anticipated implementation outcomes before launch and measure actual implementation outcomes during execution
• Use rapid-cycle learning to adapt workflow while protecting the essential clinical function of the model
• Separate implementation failure from innovation failure: a poorly implemented intervention cannot establish that the intervention itself is ineffective
• Scale only after demonstrating adoption, fidelity, operational reliability, clinical benefit, and acceptable burden

A staged operating-model roadmap

An executive implementation roadmap developed for this report.

Seven recommendations

1. Make the economic transition explicit. Map where fee-for-service and value incentives conflict, revise internal scorecards, and define how leaders will manage revenue displacement.
2. Organize around priority patient pathways. Select a small number of clinically and financially material populations; redesign end-to-end work before adding broader contract scope.
3. Build physician and frontline ownership into governance. Give clinical leaders real authority over pathway design, measures, resources, and adaptation, not ceremonial sponsorship.
4. Embed intelligence in workflow. Convert every analytic signal into a timely decision, including owner, standard work, capacity commitment, and escalation rule.
5. Measure implementation as rigorously as outcomes. Track reach, adoption, fidelity, burden, acceptability, feasibility, and sustainability before interpreting lagging cost results.
6. Co-design with patients and stratify for equity. Include patients in workflow redesign and test whether access, burden, and outcomes differ across populations.
7. Scale through learning, not mandate. Expand only when the model demonstrates clinical legitimacy, workflow reliability, outcome improvement, and sustainable economics.

Questions for the executive team and board

1. What care-delivery behaviors must change for each major contract to succeed, and where are those behaviors currently measured?
2. Which legacy financial incentives contradict the value strategy, and who owns the resolution of the conflict?
3. Can frontline clinicians explain the clinical rationale, operational workflow, and feedback loop for the model?
4. What proportion of identified opportunities is converted into completed work, and where does conversion fail?
5. How are patient priorities, navigation burden, and equity incorporated before pathway design is finalized?
6. Which implementation outcomes are reviewed before cost and quality results, and what action follows deterioration?
7. What evidence must be present before the organization expands risk, population scope, or geographic scale?

Executive readiness scorecard

Use the scorecard before contracting, before scaling, and during quarterly governance. Rate each domain from 1 (absent or unreliable) to 5 (institutionalized and demonstrably effective). A high actuarial opportunity should not compensate for weak operating-model readiness. Any score of 1 or 2 in clinical ownership, workflow integration, data reliability, patient/equity design, or governance should trigger a constrained pilot rather than enterprise expansion.

References

Filter the evidence base by domain. All entries link to the publisher of record via DOI.