A patient mentions she has been splitting her metformin doses to make the bottle last until payday. It goes in the chart. The Z-code goes on the claim. She leaves with a refill she still cannot afford.

Nothing failed there, exactly. The screen worked. It surfaced a real barrier, in her own words, in under two minutes. What did not happen is the half that makes screening worth doing — the part where an answer becomes a referral before she walks out the door.

The Measure Was Never Just a Screen

NCQA named it plainly. SNS-E, Social Need Screening and Intervention, covers three domains — food, housing, and transportation — and reports each one twice: once for the screening, once for the intervention that follows a positive result [1]. Screening a full panel and referring no one does not produce a partial score. It produces a strong score on half a measure and a zero on the other half.

Screening Found a Home in the Visit. Referral Usually Did Not.

Most practices are not indifferent to the referral half. They are structurally unequipped for it, and the reason is mundane: screening got scheduled, and referral did not.

Screening has a slot. Discuss it yearly when patients come in for their well exam, or fold it into the annual forms they fill out in the waiting room [3]. Adult comprehensive visits are already built to go beyond basic check-ups by creating space for meaningful conversations about a patient’s physical, emotional, and social well-being [4]. The screen has a minute, an owner, a form field, and a code.

The referral usually has none of those. It has a good intention and a photocopied resource list. Ask a practice what happens to a positive food-insecurity answer at 10:40 a.m. on a Tuesday, and the answer describes a person, not a process.

Make the Referral a Step, Not a Project

While the tool you choose matters, the follow up after its results will make the difference to the patient’s health outcomes. Use PRAPARE, the CMS Accountable Health Communities tool, SCRA, or Equality Health’s Health Access Survey — any of them will find the need [5].

Then route it in the same encounter. In CareEmpower®, the Referral button sits in the top-right of the main patient panel, and it goes two places [6]:

• Out to the community. The button opens the FindHelp referral tool and refers patients directly to social service organizations and appropriate support services [6].
• In to the care team. The same button refers patients to our Care Specialists for extended assistance [6]. Community Health Workers take it from there — they live in the communities they serve, meet members in person or virtually, and connect them to food, housing, transportation, and emotional support [7].

Then code it. Submit Z-codes and CPT II codes with the claim, every time [5]. Z59.41 for food insecurity, Z59.82 for transportation insecurity, Z59.811 for housing instability with risk of homelessness [8]. These are supplemental reporting codes, not primary diagnoses [8] — which is why they get skipped, and why the data vanishes when they are.

Write the Referral So Someone Can Act on It

One step is worth slowing down for. The referral form asks for services needed, urgency, and notes, and the notes field is where referrals either land or stall. Include contact numbers, preferred language, the best time of day to call, and whether a caregiver or power of attorney is involved [9]. Urgency is a real choice, not a formality: Priority draws a response in four to seven days, Routine in eight to 14 [9]. Anything emergent or same-day goes by phone instead of through the form [9].

A referral that says “food insecurity” hands a Care Specialist a fact. A referral that says “food insecurity, Spanish-preferred, works nights, daughter manages appointments, reachable before noon” hands them a plan.

Screening Data Is Only Worth the Loop You Close

This is the part worth keeping after the workflow details fade.

A program that tracks non-medical drivers of health (NMDOH, or SDOH in some markets) by counting screenings completed is a documentation burden wearing a program’s clothes. Surveys completed measure staff compliance with a form. It says nothing about whether one patient got a ride, a food box, or help with a utility bill.

The only number that means anything is closed loops: positive screens that produced a referral, and referrals that produced a service. That is measurable today. Referrals appear per patient in the Referrals tab of the Manage Visit window, and across the whole TIN from the main Referrals tab [10], and feedback on referral activity comes back through the same platform rather than disappearing into a void [11].

Pull that view once a month and put two numbers beside each other: positive screens, and referrals placed. The distance between them is your actual program. Everything above that line is documentation.

The Question Is Already Asked

Nobody needs convincing that food, housing, and transportation shape outcomes. That argument was won years ago, and screening rates show it — practices ask.

What is left is a workflow problem with an unglamorous answer. The referral has to be a step inside the encounter that is already happening, taken by the person already in the room, on the platform already open. Not a program to stand up. Not a committee. A button, a note with enough detail to act on, and a number you look at once a month.

You are already asking the question. Finish the sentence.

References

[1] NCQA. “HEDIS Measure Crosswalk, MY 2025” — SNS-E (Social Need Screening and Intervention): food, housing, and transportation domains, each reported for screening and for intervention.

[2] “HEDIS MY 2026 Provider Guide” — SNS-E revision: added codes for screening numerator events and intervention denominator/numerator events.

[3] Equality Health. “SDOH — Quality Guide.” March 2025, p. 1 (TIP: discuss yearly at the well exam or fold into annual forms).

[4] Equality Health. “Annual Comprehensive Visits for Adults 18+ (ACV) — Pathways to Better.” June 2025, p. 1.

[5] Equality Health. “SDOH — Quality Guide.” March 2025, p. 1 (four-tool screening list; submit Z-codes and CPT II codes with claim submission); corroborated in “Screening for Social Determinants of Health (SDOH) — Pathways to Better,” June 2025, p. 2.

[6] Equality Health. “Screening for Social Determinants of Health (SDOH) — Pathways to Better.” June 2025, p. 2 (Referral button, both destinations).

[7] Equality Health. “Meet the Equality Health Care Team.” Member flyer series, November 2025 (CHW role description).

[8] Equality Health. “SDOH — Quality Guide.” March 2025, p. 2, Exhibit 1 (Z-code categories); same exhibit in the June 2025 P2B guide, p. 3.

[9] Equality Health. “CareEmpower User Guide.” February 2026, p. 39 (Care Coordination Referral form: services needed; urgency — Priority 4–7 days, Routine 8–14 days; notes guidance; emergent/same-day by phone).

[10] Equality Health. “CareEmpower User Guide.” February 2026, pp. 40–41 (Referrals tab, per-patient and TIN-wide).

[11] Equality Health. “CareEmpower User Guide.” February 2026, p. 38 (EH shares feedback on referral activities back through CareEmpower directly).