Research Methodology
How We Research Healthcare Cost Containment
Cost containment claims are everywhere in employee benefits — and most of them are inflated. This page documents how we evaluate the evidence base for each strategy, source savings estimates, assess employer fit, and apply AI analysis — so you can distinguish strategies with strong evidence from those with marketing-driven claims.
01 — Editorial Standards
Independence from vendor savings claims
Cost containment vendors routinely publish inflated savings claims. Our editorial policy requires independent corroboration for any savings estimate we present.
No vendor-sponsored savings claims
We do not publish vendor-provided savings estimates without independent corroboration from peer-reviewed research, actuarial studies, or employer-reported data.
Evidence grading
Every strategy is graded on evidence strength: Strong (multiple independent studies), Moderate (growing evidence base), Emerging (limited evidence, promising early data), or Unproven (marketing claims without independent evidence).
Savings range vs. point estimate
We present savings as ranges, not point estimates, to reflect the variation across employer size, industry, and implementation quality. Point estimates from vendor marketing materials are not used.
Implementation risk disclosure
Every strategy profile includes an honest assessment of implementation risks — member disruption, network issues, legal exposure — not just savings potential.
02 — Evidence Standards
How we grade the evidence for each strategy
Not all cost containment strategies have equal evidence. We apply a four-level evidence grading system to every strategy we profile.
Strong
Multiple independent peer-reviewed studies, actuarial analyses, or large-scale employer case studies with consistent results. Savings estimates are well-supported and reproducible.
Moderate
Growing evidence base with some independent studies and employer case studies. Savings estimates are plausible but vary more widely across implementations.
Emerging
Limited independent evidence. Early employer case studies are promising but not yet replicated at scale. Savings estimates are speculative.
Unproven
No independent evidence. Savings claims are based on vendor marketing materials or single-employer case studies that have not been independently verified.
03 — Strategy Evaluation
How we evaluate each cost containment strategy
Each strategy is evaluated on savings potential, evidence strength, implementation risk, and employer fit. This table summarizes our current assessment of the six primary strategies.
| Strategy | Savings range | Evidence | Key risks | Best fit |
|---|---|---|---|---|
| Reference-based pricing (RBP) | 15–40% on facility claims | Strong — multiple employer case studies and actuarial analyses | Member balance billing, network disruption, employee relations | Self-funded employers with high facility utilization and strong HR communication capacity |
| Direct primary care (DPC) | 10–25% on downstream utilization | Moderate — growing employer case study base; limited RCT evidence | Upfront membership cost, member adoption, geographic availability | Employers with high primary care utilization and geographically concentrated workforce |
| PBM reform (pass-through + rebate) | 15–30% on pharmacy spend | Strong — state Medicaid audits and employer case studies | Contract negotiation complexity, formulary disruption, specialty drug management | Self-funded employers with $500K+ annual pharmacy spend |
| Centers of excellence (COE) | 20–50% on targeted high-cost procedures | Strong for specific procedures (spine, cardiac, oncology) | Travel burden on members, network disruption, limited procedure scope | Employers with high volume of elective high-cost procedures |
| Site-of-care optimization | 10–30% on outpatient and imaging spend | Strong — well-documented price variation by site of care | Member inconvenience, steerage program design complexity | Employers with high outpatient imaging, infusion, and ambulatory surgery utilization |
| Data-driven plan management | Varies — enables targeting of other strategies | Foundational — required to measure impact of all other strategies | Data quality, HIPAA compliance, analytical capacity | All self-funded employers; prerequisite for advanced cost containment |
04 — Benchmark Data
Where our cost containment benchmark data comes from
Cost containment benchmarks are drawn from independent research, not vendor marketing materials.
Health Care Cost Institute (HCCI) Employer-Sponsored Insurance Data
Claims-level data on utilization and spending patterns used to benchmark cost containment opportunity by strategy.
RAND Corporation Health Insurance Experiment and follow-on studies
Foundational research on price sensitivity, utilization, and the impact of plan design on healthcare spending.
Johns Hopkins Bloomberg School of Public Health — RBP research
Peer-reviewed research on reference-based pricing outcomes, balance billing rates, and employer savings.
Journal of the American Medical Association (JAMA) — DPC studies
Peer-reviewed research on direct primary care outcomes, utilization impact, and employer savings.
Milliman Medical Index and actuarial research
Annual actuarial analysis of employer healthcare cost trends used to contextualize strategy savings estimates.
Employer case studies (anonymized, employer-reported)
Employer-reported implementation data on savings, member disruption, and implementation challenges. Anonymized to protect employer confidentiality.
05 — AI Analysis
How AI is used in cost containment tools
The Cost Containment Calculator and related tools use GPT-4o to generate employer-specific strategy recommendations. This section documents what the AI does and does not do.
What AI does
- Synthesizes employer profile data with strategy evidence to generate prioritized recommendations
- Identifies which strategies have the strongest fit for the employer's size, industry, and funding type
- Estimates savings ranges based on employer-provided spend data and published benchmarks
- Flags implementation risks specific to the employer's context
What AI does not do
- Access real-time claims data or employer-specific cost drivers
- Provide actuarial certifications of savings estimates
- Replace a benefits consultant or actuary for strategy selection
- Guarantee savings outcomes — all estimates are population-level heuristics
06 — Limitations
What our cost containment research cannot tell you
Cost containment outcomes are highly implementation-dependent. Our methodology has specific limits every employer should understand.
Implementation quality drives outcomes
Published savings ranges assume competent implementation. Poor communication, inadequate member support, or weak vendor management will produce results well below published ranges — or negative results.
Group-specific claims analysis required
Our tools use employer-provided spend data and population-level benchmarks. A group-specific claims analysis by a licensed actuary or benefits consultant will identify cost drivers and strategy fit more accurately.
Member disruption is real
Strategies like RBP and site-of-care optimization create member disruption. Our savings estimates do not account for the cost of member complaints, HR time, or potential legal exposure from balance billing disputes.
Vendor selection matters
The same strategy implemented by different vendors produces very different results. Our strategy analysis does not substitute for rigorous vendor evaluation and RFP processes.
Legal and regulatory exposure
Some cost containment strategies (RBP, certain DPC structures, captive arrangements) have legal and regulatory complexity. Engage qualified ERISA counsel before implementing any strategy with legal exposure.