StudyBackedHealth: Product & Provider Scoring Rubric

The rule: Score every product fully before looking up its affiliate program or commission rate. The score is final once written. Attach links afterward.

Final score = weighted average, shown as X.X / 10.

---

1. Clinical Evidence: 30%

What the published research says about the product or the medications/ingredients it provides.

Score	Standard
9–10	Multiple peer-reviewed RCTs directly support efficacy claims; effect sizes clearly reported
7–8	Solid trial evidence, some gaps (short duration, narrow populations)
5–6	Limited or mixed studies; relies partly on extrapolation
3–4	Mostly observational data or studies on different doses/forms than what’s sold
0–2	No credible evidence, or evidence contradicts marketing claims

Checklist:

• Found the actual studies (PubMed), not just the company’s summary of them
• Checked whether studied dose/form matches what’s sold
• Noted who funded the key studies
• For GLP-1 providers: brand-name vs. compounded sourcing documented

2. Consumer Sentiment: 25%

Aggregated from Reddit, Trustpilot, BBB, app stores. Minimum 100 reviews across 2+ platforms before scoring.

Score	Standard
9–10	Consistently positive across platforms; complaints are minor and acknowledged by company
7–8	Mostly positive; recurring but non-serious complaints (shipping, app bugs)
5–6	Mixed; recurring complaints about core service (billing, access to clinicians)
3–4	Frequent serious complaints: surprise charges, cancellation traps, ghosting
0–2	Pattern of harm reports, fake-review evidence, or regulatory action

Checklist:

• Read negative reviews first
• Separated complaints about the medication (side effects) from the company
• Checked BBB complaint resolution rate
• Searched “[brand] + scam / lawsuit / refund” on Reddit

3. True Cost & Transparency: 20%

Score	Standard
9–10	All-in monthly price clear before signup; no step-up pricing; easy cancellation
7–8	Pricing clear but extras (labs, shipping) add up; cancellation straightforward
5–6	Intro pricing that steps up after month 1; cancellation requires contact
3–4	Pricing hidden until after intake quiz; auto-renewal traps
0–2	Bait pricing, undisclosed fees, documented cancellation obstruction

Checklist:

• Calculated real cost over 6 months, not month 1
• Documented what’s NOT included (medication itself? labs? visits?)
• Tested how many clicks to find the cancellation policy

4. Care Quality & Safety: 15%

Score	Standard
9–10	Board-certified prescribers, real intake screening, ongoing monitoring, easy clinician access
7–8	Licensed clinicians, decent screening, slower follow-up
5–6	Minimal screening; asynchronous-only; prescriber credentials vague
3–4	Rubber-stamp intake; no follow-up care
0–2	Prescribes without meaningful evaluation; safety red flags

For non-prescription products: third-party testing, manufacturing standards (cGMP), ingredient disclosure.

5. Support & Experience: 10%

Customer service responsiveness, app/site usability (weight this for older users: phone support available? readable interface?), coaching or education included.

---

Disqualifiers (automatic “Not Recommended” regardless of score)

• Sells prescription drugs without any clinician involvement
• Health claims that contradict FDA position
• Documented fake reviews or undisclosed paid rankings
• Active regulatory enforcement for patient harm

After scoring

1. Lock the score.
2. THEN check affiliate program availability and terms.
3. If your #1 has no program: publish it as #1 anyway. State plainly: “We earn nothing from this recommendation.”
4. Log the date and re-score quarterly.

How we classify providers

Once a provider is scored, it lands in one of three tiers (updated June 2026):

• Recommended: scored at or above our threshold, with no disqualifier. These are the programs we'll point you to.
• Others: providers we reviewed and scored below the ones we recommend, but that carry no FDA action, no documented fake reviews, and no harm pattern. We list them in the open, with their scores and the caveats, so you can compare, but we don't steer you to them.
• Avoid: an automatic disqualifier (an FDA warning letter, documented fake reviews, or prescribing with no clinician) or a documented harm pattern. These are on our do-not-recommend list.

We added the middle tier in June 2026 because "not recommended" had been doing two different jobs, lumping a mediocre-but-transparent provider in with one that holds an FDA warning letter. Those aren't the same thing, and you deserve to see the difference. The scoring didn't change; only the labels did. The independence rule still applies: no provider pays us anything, and a provider's tier is never for sale.

Two rubrics: FDA-approved vs compounded

FDA-approved brand programs and compounded providers are different products for different reasons, so we judge them on two separate rubrics, and a score on one is not comparable to a score on the other.

The FDA-approved programs are scored on the 31-point methodology above, which leans on clinical evidence: the RCTs behind the labeled drug. Compounded GLP-1 is made to order and hasn't been through FDA premarket review, so scoring it on "clinical evidence" would be measuring something it can't have. Instead, the compounded rubric judges what protects a compounded buyer in practice:

• Pricing transparency & value (30%): whether the real ongoing all-in cost is disclosed up front. We don't penalize an above-board pricing structure: a commitment or volume discount, dose-based tiering, or a clearly-flagged intro-month promo is fine as long as the real ongoing price is posted. We do penalize pricing that's hidden until after an intake quiz, undisclosed or mandatory fees, and cancellation or refund traps.
• Medication quality & pharmacy standards (25%): use of FDA-regulated 503A/503B compounding pharmacies, third-party testing, PCAB accreditation, and transparent pharmacy disclosure.
• Provider credentials & medical oversight (20%): board certification, the quality of the initial evaluation, ongoing monitoring, and contraindication screening.
• Patient support & communication (15%): response times, care-team access, side-effect management, dose-adjustment responsiveness.
• Shipping speed & reliability (10%): consistent delivery, supply-chain reliability, and how shortages are handled.

When we can't independently verify a dimension, we score it a neutral 5, not a low mark. If a provider doesn't name its pharmacy, or there's no public track record for its support or shipping, we don't hold against it something we couldn't check, but we don't credit it either, so an unverified dimension can't lift a provider above one that does disclose what it's doing. A documented problem is different: a refund or cancellation trap, an FDA action, or a clear complaint pattern is verified, and it scores low on its own merits.

The compounded scores rank those providers against each other on the compounded page, never against the FDA-approved programs. The same independence rule applies (we take no money from any provider), and the same disqualifiers send a provider to Avoid regardless of rubric: an FDA warning letter, fabricated testimonials or fake results imagery, prescribing with no clinician, or a documented harm pattern.

How we source community reports

Our articles cite patterns from patient communities: "65 of 184 comments reported minimal or no side effects." Here is exactly how those numbers are made, and the rules we follow.

The method

• We read high-engagement threads in GLP-1 patient communities (primarily Reddit), analyze the loaded comments, and count how many match each recurring theme.
• Every count is stored in a dated internal dataset with the thread list and method, and each article's sources section links the threads we analyzed, so you can open them and check our characterization yourself.
• We re-collect quarterly. Sample sizes and collection dates are stated wherever a number appears.

Why we paraphrase, and never name users

• Privacy: people in patient communities share weights, medical histories, and personal struggles under pseudonyms, to peers. Reprinting that with a username on a commercial site re-amplifies it to an audience they never chose. We summarize patterns instead.
• Advertising law: even on a non-commercial site, an individual's dramatic success story quoted next to a provider recommendation can function as an atypical-results testimonial, something weight-loss advertising rules rightly restrict. Aggregated, hedged patterns keep our community reporting analysis, not endorsement.
• Discipline: quotes invite cherry-picking; counts don't. If 65 of 184 comments say the opposite of the dramatic story, we have to report that.

The limits, stated plainly

• Communities self-select: success stories and engaged patients are over-represented; quiet quitters are invisible.
• Theme counts are directional, not a survey. They tell you what a community talks about, not population rates. For rates, we cite the FDA labels and trials.
• Community patterns never override clinical evidence in our scoring. They inform the consumer-sentiment portion of the rubric above.