Mental Health Access and SDOH Intervention: Closing the Gaps Before They Become Claims

The cost management strategies in this series share a common logic: identify where the plan is overpaying and redirect spend to lower-cost alternatives. Domestic steering saves on procedures. Pharmacy optimization saves on drugs. Maternity management saves on birth complications. Mental health access and social determinants of health intervention work differently. They do not reduce the price of a service the member is already consuming. They prevent the service from becoming necessary. The intervention is upstream. The cost reduction is downstream.

Series 06 documented the mental health access failures in small group level funded plans (LFP-06.06). Series 09 quantified the cost amplification that untreated mental health conditions produce across the medical spend (LFP-09.08). This article connects those two threads to a cost management strategy: expand access to evidence-based mental health services, screen for social determinants that predict avoidable utilization, and route members to community resources before unaddressed social needs become medical claims. The TPA that builds this capability does not reduce a line item. It changes the conditions that produce the line item.

The Mental Health Access Strategy

#

Traditional Employee Assistance Programs deliver utilization rates between 2% and 10% of eligible employees. The Employee Assistance Professional Association reported in 2023 that the typical range falls between 6% and 10%, with many traditional programs hovering closer to the low end. The Bureau of Labor Statistics reported that 61% of workers had access to an EAP in 2024. The problem is not availability. The problem is that traditional EAPs route members through 1-800 numbers, impose session limits (typically three to six sessions), restrict provider networks, and create friction that discourages engagement precisely when members need help most.

The cost consequence of this low utilization is substantial. Untreated mental health conditions amplify medical spending across every category. Members with comorbid depression and diabetes, for example, generate medical costs two to three times higher than members with diabetes alone. The amplification is not speculative. It is visible in claims data: higher emergency department utilization, more inpatient admissions, lower medication adherence for chronic conditions, and increased MSK and cardiovascular claims. The connection between untreated anxiety, sleep disruption, musculoskeletal tension, and cardiovascular stress response is well documented in clinical literature.

A new generation of employer-sponsored mental health platforms has demonstrated that the access problem is solvable and that solving it reduces total medical spend. Lyra Health reported in a 2024 Aon study that employers offering its services saw an average 26% annual reduction in health plan spending sustained over four years. That analysis, conducted across employers in the technology, manufacturing, and consumer goods sectors with eligible populations exceeding 30,000 employees, compared costs between members who used Lyra services and matched controls who did not. A separate 2024 Aon analysis found $4,138 in net savings per member in the first year, with a $3.04 return for every $1 invested. Members with alcohol and substance use disorders showed a 60% reduction in employer health plan spending.

Spring Health published results in JAMA Network Open showing a 1.9x return on investment in the first year: $1,070 in net savings per participant after deducting care costs. That study evaluated 13,990 participants from seven employers over a period from November 2019 to May 2023. The findings showed a 30% gross cost reduction and 14% net reduction, with particularly strong savings in high-cost chronic conditions including diabetes, chronic pain, and hypertension. The Validation Institute independently confirmed that Spring Health customers lowered total health plan spend by $2,430 per participant within the first six months of engagement.

These platforms achieve higher utilization than traditional EAPs through digital-first access (members can book appointments within days rather than weeks), AI-powered provider matching, expanded provider networks that include virtual therapy, and elimination of the friction that traditional EAPs impose. Spring Health reported that at one national healthcare system, switching from a traditional EAP produced a 600% increase in utilization and a 4.8x increase in therapy engagement within ten months.

For a level funded TPA serving small groups, the operational translation is clear. Replacing or supplementing the standard EAP with an evidence-based mental health platform produces measurable medical cost reductions that show up in the claims fund. Those reductions improve stop loss performance, strengthen renewal pricing, and differentiate the TPA in a market where mental health access has become a meaningful employer concern.

SDOH Signal Data and Risk Stratification

#

Social determinants of health, the conditions in which people live, work, and age, account for an estimated 30% to 55% of health outcomes according to research cited by the World Health Organization and reiterated in clinical literature. Food insecurity, housing instability, transportation barriers, utility difficulties, and social isolation predict healthcare utilization more strongly than many clinical variables. A member who cannot afford to fill a prescription, who lacks reliable transportation to a follow-up appointment, or who is rationing food because of financial stress will generate avoidable emergency department visits and preventable chronic disease complications.

CMS recognized this relationship by mandating SDOH screening for all inpatient hospital admissions beginning January 1, 2024, covering five domains: food insecurity, housing instability, transportation needs, utility difficulties, and interpersonal safety. The HCPCS code G0136 was created for SDOH risk assessment administration in outpatient settings, reimbursable during annual wellness visits. CMS proposed expanding mandatory SDOH screening to hospital outpatient departments, rural emergency hospitals, and ambulatory surgical centers, with voluntary reporting starting in 2025 and mandatory reporting in 2026.

For employer-sponsored plans, the SDOH signal is available through commercial data vendors. Socially Determined provides zip-code-level and individual-level social risk data that enables risk stratification. Findhelp (formerly Aunt Bertha) connects members to community resources through a platform covering over 500,000 programs nationwide. Unite Us operates a closed-loop referral network that tracks whether members actually receive the services they were referred to. These vendors operate in the Medicaid managed care market primarily, but the underlying data and referral infrastructure is applicable to any payer, including self-funded employer plans.

A study conducted by Adobe Health Systems, published in Patient Safety & Quality Healthcare, reviewed two years of claims data from a Medicare population in the Southwestern United States covering 2,355 individuals. After SDOH risks were identified and community resources deployed, the annual average healthcare cost per member dropped from $13,500 to approximately $9,500, a 31% reduction. Blood pressure monitoring compliance improved by 95%. Diabetic screening compliance improved by 47%. The intervention mechanism was not clinical treatment. It was connecting members to food assistance, housing support, transportation services, and mental health resources personalized to their zip code and circumstances.

Community Resource Routing

#

The operational model has three steps. First, identify SDOH risk through data enrichment (claims-based signals like zip code, pharmacy refill gaps, missed appointments, and emergency department patterns) and direct member screening using validated instruments. Second, route the member to community resources: food assistance programs (SNAP enrollment support, food banks, commodity supplemental food programs), housing support (rental assistance, utility payment programs, weatherization assistance), transportation services (Medicaid non-emergency medical transportation where applicable, ride-share partnerships, volunteer driver networks), and social connection programs (senior centers, community health worker outreach, peer support groups). Third, track whether the referral resulted in the member actually receiving the resource.

The closed-loop referral model matters because open-loop referrals (giving a member a phone number and hoping they call) produce low completion rates. Unite Us reported that its closed-loop network, which tracks referral outcomes, achieves substantially higher resource connection rates than open referrals. The distinction is operationally significant for a TPA: the TPA needs to know whether the member’s food insecurity was actually addressed, not merely whether a referral was generated.

For small group level funded plans, the SDOH intervention is most relevant to the populations Series 06 identified as underserved. Low-wage workers (LFP-06.04) who earn $28,000 to $35,000 face food insecurity, housing cost burden, and transportation barriers at rates far exceeding the general insured population. Workers with chronic conditions (LFP-06.05) whose medication adherence suffers because they cannot afford copays or cannot get to the pharmacy. The 55-to-64 cohort (LFP-06.02) where social isolation after a career transition compounds cardiovascular and mental health risk.

The TPA’s role in this model is coordination, not clinical care. The TPA enriches its claims data with SDOH signals, flags members at highest risk for avoidable utilization, and connects them with a community resource navigation vendor or internal care coordinator. The cost of screening and referral is low: $3 to $10 per member per month for data enrichment and resource navigation, depending on vendor and volume. The avoided medical cost is disproportionately high. A single prevented emergency department visit saves $2,000 to $3,000. A single prevented behavioral health crisis hospitalization saves $10,000 to $15,000.

Implementation and Net Impact

#

Program cost for a 25-person plan includes three components. Mental health access expansion: replacing or supplementing the existing EAP with an evidence-based platform runs approximately $3 to $8 per employee per month, depending on vendor and session limits. For 25 employees, that is $900 to $2,400 annually. SDOH data enrichment and screening: $3 to $10 per member per month for data vendor and screening infrastructure, or $900 to $3,000 annually for 25 members. Community resource navigation: $1 to $5 per member per month for navigation services, or $300 to $1,500 annually. Total implementation cost: approximately $2,100 to $6,900 per year for a 25-person plan.

Savings are harder to isolate because the mechanism is preventive. The savings do not appear as a discrete line item reduction. They appear as claims that did not happen: the emergency department visit that did not occur because the member’s food insecurity was addressed and medication adherence improved, the behavioral health hospitalization that did not happen because the member accessed therapy before a crisis developed, the MSK surgery that was deferred because the member’s depression was treated and they engaged with conservative care. At representative utilization rates, if three members out of 25 avoid a significant medical event (one emergency department visit, one behavioral health crisis, one chronic disease complication) through improved mental health access and SDOH intervention, the avoided claims range from $5,000 to $15,000.

The net impact is moderate on a per-plan basis. This is not geographic arbitrage, where a single procedure generates $20,000 in savings. Mental health access and SDOH intervention produce smaller, diffuse savings distributed across the population. The cumulative effect across a TPA’s book of business is where the strategy becomes meaningful. A TPA managing 200 small groups with an average of 25 members each (5,000 total covered lives) that reduces avoidable utilization by even $200 per member annually generates $1,000,000 in total avoided claims across the book. That reduction improves aggregate stop loss performance, reduces renewal pressure, and creates a competitive differentiation that brokers and employers increasingly value.

The measurement challenge is real. Unlike domestic steering, where the savings can be calculated by comparing the procedure cost at the two facilities, mental health and SDOH savings require counterfactual analysis: comparing what happened to what would have happened without the intervention. The actuarial methods exist (propensity-matched cohort comparison, pre-post analysis with risk adjustment), but they require data infrastructure that most small-group TPAs have not built. Building that infrastructure is part of the cost management investment this series describes (LFP-10.01).