How Spectra Health works, and why you can trust the results

We use publicly validated instruments — not a proprietary black box

Spectra does not ask you to trust a single vendor's secret algorithm. When a patient's brief adaptive screen indicates a possible condition, we administer the individually validated, publicly available instruments clinicians already know, each with decades of peer-reviewed psychometric data you can look up independently.

Because these are established, non-proprietary tools, your clinicians, your payers, and your compliance team can all evaluate them on their published merits. (Full citations at the bottom of this page.)

The scoring is deterministic — there is no AI making clinical judgments

Spectra's scoring, severity banding, and note drafting are rule-based and deterministic. The same answers always produce the same result. We do not use a large language model (LLM) or generative AI to score patients, generate clinical content, or make any clinical decision. This is a deliberate design and compliance choice: results are reproducible and auditable, and no patient data is sent to a third-party AI service.

Screening is a prompt for clinical judgment — never a diagnosis

A positive screen tells a clinician where to look; it is not a diagnosis. Every result is reviewed by the licensed clinician, who makes all clinical decisions. Spectra supports the clinician; it does not replace them.

We minimize the patient data we hold

Spectra is built to require no patient name, date of birth, or Social Security number. Each screening is tied to a random case number the practice reconciles in its own chart. An optional de-identified mode hard-blocks any patient reference. Less identifying data means lower breach exposure for your practice.

The billing support is designed to be defended, not just claimed

Spectra doesn't only tell your biller what to bill. It enforces the documentation rules (for example, CPT 96130 requires a signed evaluation note, sufficient time, and documented medical necessity), drafts the evaluation note, and assembles a payer-ready audit file so the code holds up under review. You bill what you can document and keep.

What we don't do yet — stated plainly

Some instruments (for example, certain ADHD and bipolar tools) are not enabled in Spectra today because we won't ship an instrument until its licensing and clinical review are complete. We'd rather tell you what's live than imply coverage we don't have.

Why screening for behavioral health in primary care matters

Behavioral health conditions are common in primary care and frequently go unrecognized, and primary care is often the last clinical contact before a crisis — for example, research has found that a large share of people who die by suicide had a primary-care visit in the weeks beforehand (Luoma, Martin & Pearson, American Journal of Psychiatry, 2002). National guidance supports routine screening: the U.S. Preventive Services Task Force (USPSTF) recommends screening adults for depression and for anxiety. Spectra operationalizes that guidance efficiently.

Sources (verify independently)

1. USPSTF, Screening for Depression in Adults.
2. USPSTF, Screening for Anxiety Disorders in Adults (2023).
3. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J Gen Intern Med. 2001.
4. Spitzer RL, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006.
5. Bush K, et al. The AUDIT-C. Arch Intern Med. 1998.
6. Gierk B, et al. The Somatic Symptom Scale-8 (SSS-8). JAMA Intern Med. 2014.
7. NIDA. NIDA Quick Screen / NM-ASSIST.
8. Luoma JB, Martin CE, Pearson JL. Contact with mental health and primary care providers before suicide. Am J Psychiatry. 2002.