AvertD vs. Proove Opioid Risk
AvertD (SOLVD Health, formerly Prescient Medicine/AutoGenomics) and the Proove Opioid Risk Algorithm (Proove Genomics, formerly Proove Biosciences) are both genetically informed risk-assessment tools intended to help clinicians identify patients at elevated risk of opioid use disorder (OUD) before or during opioid therapy. The two tests differ substantially, however, in their intended use population, algorithm design, regulatory pathway, and the volume and setting of clinical evidence supporting them.
Quick Comparison
| Dimension | AvertD | Proove Opioid Risk |
|---|---|---|
| Manufacturer | SOLVD Health (formerly Prescient Medicine / AutoGenomics) | Proove Genomics (formerly Proove Biosciences) |
| Regulatory Status | FDA Premarket Approval (PMA), December 2023; NYS CLEP approval, January 2024 | Laboratory Developed Test (LDT); pursuing NYS CLEP approval and an expedited FDA pathway using AvertD as a predicate, plus ISO 15189 |
| Indicated Use | Single point-in-time screen before a patient's first oral opioid prescription (4–30 days) for acute pain, in opioid-naïve adults | Risk stratification for patients being considered for, or already managed on, opioid therapy — both acute and chronic pain, opioid-naïve and opioid-experienced |
| Algorithm Composition | 15 SNPs (genetic markers only) | 12 SNPs plus 6 non-genetic (phenotypic) risk factors in a combined algorithm |
| Published Clinical Validity Studies | 1 published study (67 patients) | 3 published studies (2,784 patients across pain management, primary care, and addiction-treatment settings) |
| Published Clinical Utility Studies | 0 published studies | 3 published studies (12,534 patients across primary care and pain-management clinics) |
| Real-World Patients Tested | Newly commercialized | >200,000 patients tested |
| Conference Posters (2012–2017) | 2 | >100 |
| Medical-Society Research Awards | 0 | 3 (ASIPP, AAN, ASRA) |
| Reported Accuracy (ROC/AUC) | Up to 0.92 in a single 67-patient study | Up to 0.967 across 3 published studies |
By the Numbers
Reported accuracy (ROC/AUC) side by side, followed by the evidence base behind each test.
Reported accuracy (ROC/AUC) — AvertD: single 67-patient study. Proove: 3 studies, 2,784 patients.
Patients in Validity Studies
AvertD: 67
Proove: 2,784
Clinical Utility Studies
AvertD: 0
Proove: 3 studies · 12,534 patients
Real-World Patients Tested
AvertD: Newly commercialized
Proove: >200,000
Conference Posters
AvertD: 2
Proove: >100
Medical-Society Awards
AvertD: 0
Proove: 3
Detailed, Source-by-Source Comparison
Tap a row to expand the full comparison, including sample sizes and study populations.
AvertD
- Post-acute care, short-term basis of less than 30 days, in patients naïve to opioids.
Proove Opioid Risk
- Post-acute care and chronic care basis, for patients taking opioids for medication management.
Why the Proove Opioid Risk Algorithm Is Different
Different Point in the Prescribing Pathway
AvertD's FDA-approved indication is narrow and time-limited: a single genetic screen administered before a patient's very first oral opioid prescription for acute pain of 4–30 days' duration, in patients who have never previously taken opioids. It is not indicated for patients already receiving opioid therapy, for chronic pain management, or for guiding dose changes, opioid rotation, or discontinuation. The Proove Opioid Risk Algorithm has been validated and used clinically across both acute and chronic pain populations, in patients who are opioid-naïve and in patients already established on opioid therapy — supporting a broader range of prescribing decisions, including initial risk stratification, dose titration, opioid rotation, and discontinuation in favor of non-opioid alternatives. Primary care and pain management physicians manage patients across the full arc of opioid therapy, not only the moment before a first prescription — a population AvertD's label does not address.
Genotype-Only vs. Genotype-Plus-Phenotype Design
AvertD's risk score is derived exclusively from 15 genetic markers. The Proove algorithm combines a partially overlapping panel of 12 genetic markers with 6 additional, non-genetic phenotypic risk factors — including personal and family history of substance use, personal history of mental health disorders, and age — weighted together in a single composite score. Personal and family history of substance use disorder, co-occurring mental health conditions, and age at first opioid exposure are independently established, clinically actionable risk factors that a purely genetic test does not capture. By incorporating these factors directly into its algorithm rather than relying on genotype alone, the Proove test is designed to reflect a more complete risk picture for each patient.
Depth and Breadth of Clinical Evidence
AvertD's FDA approval rests on a single published validation study of 67 patients, together with an unpublished registration trial of 897 patients restricted to a short-term, post-operative setting, and the test itself only recently became commercially available. The Proove algorithm has been evaluated in five independent validity cohorts (>3,700 patients) and three independent clinical utility studies (>12,500 patients) spanning primary care, pain management, orthopedic, psychiatric, and addiction-treatment settings, in addition to a 6,161-patient Medicare beneficiary outcomes study, and has been used clinically in over 200,000 patients to date. That real-world experience is reflected in scientific engagement too: work on the Proove algorithm has been presented in more than 100 posters at major pain-medicine, neurology, and anesthesiology meetings between 2012 and 2017, and has received three research awards (ASIPP, AAN, and ASRA) — compared with two posters and no awards for AvertD to date.
Regulatory Pathway Is Not, By Itself, a Measure of Clinical Performance
FDA approval of AvertD reflects that it met the bar for its specific, narrow acute-care indication; it is not, by itself, evidence that AvertD is more clinically valid or useful than the Proove algorithm across the broader population of patients being considered for or managed on opioid therapy. Premarket approval (PMA) evaluates safety and effectiveness for the indication as labeled, and AvertD's own labeling carries a boxed warning that results must not be used alone to make treatment decisions and must be interpreted only as part of a full clinical evaluation. The FDA also required SOLVD Health to conduct a post-approval study of more than 3,000 additional patients to further characterize performance across demographic subgroups underrepresented in its original validation cohort. Proove Genomics has elected a different regulatory strategy for the relaunched test: initial commercialization as a Laboratory Developed Test (LDT), with concurrent pursuit of New York State CLEP approval, an expedited FDA clearance pathway using AvertD as a predicate device, and ISO 15189 accreditation to support international use — intended to bring the broader-indication, genotype-plus-phenotype algorithm to market while working toward the same regulatory rigor AvertD has already established, not to suggest that regulatory status alone is the right basis for comparing the two tests.
Summary
AvertD and the Proove Opioid Risk Algorithm serve different points in the opioid-prescribing pathway, are built from different inputs, and are backed by different depths of clinical evidence. Clinicians and payers evaluating either test should weigh indicated use, algorithm inputs, and the underlying evidence base — rather than regulatory status alone.
Sources
- [1] U.S. Food and Drug Administration. Summary of Safety and Effectiveness Data (SSED), PMA P230032 (AvertD, SOLVD Health / Prescient Medicine). Approved December 19, 2023.
- [2] AvertD Test product information. avertdtest.com. Accessed August 2026.
- [3] Farah JR, Lee C, Kantorovich S, Smith GA, Meshkin B, et al. (2017) Evaluation of a Predictive Algorithm that Detects Aberrant Use of Opioids in an Addiction Treatment Centre. J Addict Res Ther 8:312. doi:10.4172/2155-6105.1000312
- [4] Brenton A, Richeimer S, Sharma M, Lee C, Kantorovich S, Blanchard J, Meshkin B. Observational study to calculate addictive risk to opioids: a validation study of a predictive algorithm to evaluate opioid use disorder. Pharmgenomics Pers Med. 2017 May 18;10:187-195. doi: 10.2147/PGPM.S123376. PMID: 28572737; PMCID: PMC5441670.
- [5] Sharma M, Lee C, Kantorovich S, Tedtaotao M, Smith GA, Brenton A. Validation Study of a Predictive Algorithm to Evaluate Opioid Use Disorder in a Primary Care Setting. Health Serv Res Manag Epidemiol. 2017 Aug 24;4:2333392817717411. doi: 10.1177/2333392817717411. PMID: 28890908; PMCID: PMC5574481.
- [6] Brenton A, Lee C, Lewis K, Sharma M, Kantorovich S, Smith GA, Meshkin B. A prospective, longitudinal study to evaluate the clinical utility of a predictive algorithm that detects risk of opioid use disorder. J Pain Res. 2018 Jan 5;11:119-131. doi: 10.2147/JPR.S139189. PMID: 29379313; PMCID: PMC5759857.
- [7] Additional figures cited above — the McGill University (Canada Excellence Chair of Pain Genetics) validation cohort, the 6,161-patient Medicare Part B outcomes study, AvertD's 897-patient registration trial, and two of the three Proove clinical utility studies (5,315 patients/76 clinics; 1,822 patients/18 clinics) — are unpublished data on file and not yet available as a citable published source.