Why this exists

The evidence on social determinants is not a background section. Disability carries an adjusted hazard ratio of 2.80 for fatal opioid overdose, recent incarceration 2.70, unemployment 2.46. A quarter of deaths in a large cohort of people experiencing homelessness were drug overdoses, at roughly twelve times the state adult rate.

Those numbers have an operational meaning that gets missed. They mean that the interventions in this program have preconditions. Delivery needs an address. Monitoring needs a phone that stays charged. Safe storage needs a lockable space in a home the person controls. Follow-up needs transport. A program that enrolls someone in an intervention whose preconditions they do not have has not helped them; it has set them up to fail and then recorded the failure against their name.

So the verification here is not about establishing whether a claim is true. It is about establishing what will actually work.

How it works

• Structured assessment with the patient, covering housing stability, transportation, the ability to store medication securely, who else is in the household, income and benefit status, and social support.
• A site visit where one is agreed, arranged in advance and with the patient present — for the practical purpose of confirming that delivery can reach them and that safe storage is possible, and of identifying what is missing.
• Findings that produce referrals. If the assessment identifies a housing problem, a transport problem or a benefits problem, the output is a referral to whoever handles that in the agency’s jurisdiction. An assessment that only produces a record is extraction.
• Revisited, not fixed. Circumstances change faster in this population than anywhere else in medicine.

Questions this raises

Is the site visit optional?

Yes. It is arranged in advance with the patient, and declining it does not remove anyone from the program or from care. Its purpose is practical — confirming that delivery can reach the person and that medication can be stored safely — and that purpose evaporates if the visit is coercive. See consent that means something.

Are findings reported to benefits or housing enforcement?

No. Findings produce referrals to services the patient may need, not reports to enforcement. A social determinants assessment that fed enforcement would be worthless within a month, because nobody would answer honestly. See data governance and our note on why trust is load-bearing.