NoBSmedExamplesAudit your plan →

Clinical-evidence audit

PCOS · insulin resistance · ADHD

This is not medical advice — it's a clinical-evidence audit to help you prepare for a second opinion with a doctor. Don't start, stop, or change any prescription based on it alone.

Input · what you shared
ConditionsPCOS · insulin resistance · ADHD · vitamin-D deficiency
The stackAM: atomoxetine (Strattera — ADHD med) 40 mg · CoQ10 ubiquinol 200 mg · vitamin D3 50,000 IU weekly · B12 1,000 mcg · K2 100 mg · omega-3 4,100 mg (EPA-dominant) · oral minoxidil 1.5 mg + hair complex · copper 2 mg
PM: spironolactone 100 mg · magnesium bisglycinate · K2 (guanfacine — not currently active)
Recently removedMetformin (500→1,000 mg, removed Jan 2026 — doctor stopped it over a rising ALT) · NAC (removed, not tolerated)
You assumedThe stack covers what needs covering · the rising ALT was caused by the metformin
Your askRate my medication & supplement stack — a second opinion on the whole regimen
See exactly what you pasted — your original message, verbatim

This is the unedited stack you sent us — the table above is our cleaned-up reading of it.

AM

  • Strattera — generic (Atomoxetine) — 40 mg · Rx · 1 capsule — ADHD; anxiety.
  • CoQ10 Ubiquinol — 200 mg · Nordic Naturals · 2 softgels — insulin sensitivity, fatigue, endothelial function; AM timing avoids nighttime sympathetic activation.
  • Vitamin D3 — 50,000 IU · Rx · 1 gelcap · Tuesdays onlydeficiency on a 5,000 IU 3×/wk maintenance, so back to 50k; magnesium maintained to support D; recheck in February.
  • Vitamin B12 (Methylcobalamin) — 1,000 mcg · Pure Encapsulations · 1 capsule — bring serum to higher-normal.
  • Vitamin K2 — 100 mg · NOW · 1 capsule — daily even with weekly high-dose D; calcium & vascular safety during D repletion.
  • Omega-3 — 4,100 mg (2,650 EPA + 650 DHA) · Nordic Naturals EPA Extra · 4 softgels (reduced from 5, Jan 2026)EPA-dominant for PCOS/IR/inflammation; target 3:1–4:1; still low-normal after 90 days; may improve lipids.
  • Oral Minoxidil + hair complex — Minoxidil 1.5 mg (Rx) + Biotin 2.5 mg, Vit C 100 mg, Zinc 50 mg, Pantothenic Acid 10 mg, Vit B6 2 mg · Rx · 1 tablet — hair loss/density from PCOS.
  • Copper glycinate — 2 mg · Pure Encapsulations · 1 tablet — offset zinc in hair supplement.

PM

  • Spironolactone — 100 mg · Rx · 2 tablets — hirsutism; potassium monitored quarterly.
  • Guanfacine — 1 mg · Rx (not currently active)ADHD.
  • Vitamin K2 — 100 mg · NOW · 1 capsule — (as AM).
  • Magnesium bisglycinate (intermittent) — 200 mg · Thorne · powder — sleep; replenish magnesium depleted by Metformin; higher-normal serum.

Recently removed

  • Metformin — 500 mg · Rx (AM & PM) — removed by doctor due to increasing ALT. obesity; hyperinsulinemia. Had increased 500→1,000 mg Dec 2025. Removed Jan 2026.
  • NAC — 500 mg · Thorne (AM & PM) — removed (not tolerated). glutathione/ALT/oxidative-stress support; coincided with Metformin increase. Removed Jan 2026.

Glossary

PCOS
polycystic ovary syndrome
insulin resistance
cells respond poorly to insulin so the body makes more — the suspected root driver of this cluster
ALT
alanine aminotransferase — a liver enzyme; a rising level signals liver irritation
NAFLD
non-alcoholic fatty liver disease (fat in the liver)
OGTT
oral glucose-tolerance test — the most accurate way to measure insulin resistance in PCOS
atomoxetine
a non-stimulant ADHD medication (brand: Strattera)
What's missing from what you shared — a few details would sharpen this

We deliberately didn't guess on these — and several flags below shift depending on them:

  • Your age (some hormonal shifts aren't fully addressable by diet/supplements)
  • Your actual ALT value and trend (mild vs significant changes the urgency)
  • Recent A1C, potassium (you're on spironolactone), and vitamin D level now vs before

Reply to our email with any of these and we'll sharpen this audit for you. Otherwise, raise them with your doctor — or weigh them in your own analysis.

Output · the evidence audit

The four flags below are interrelated — see the diagram.

Flag 1 Relevance · upstream

Measure the root cause — insulin resistance

Ask your doctor

"Can we measure my insulin resistance directly — an OGTT (most accurate in PCOS), or at least an A1C to screen for prediabetes?"

"And if it is the driver, should the plan focus more on treating it directly — rather than managing each downstream symptom on its own?"

The clinical evidence

Insulin resistance is the upstream driver of this whole cluster — genetic (Mendelian-randomization) evidence shows higher BMI and fasting insulin cause PCOS, not the reverse (Brower 2018), and lowering it improves real PCOS outcomes while treating the symptoms alone doesn't (Luque-Ramírez 2018). Yet it's often not measured directly.

Brower 2018 · Hum Reprod (Mendelian randomization). "Increasing BMI appears to be causal for PCOS but having PCOS does not appear to affect BMI" — 1-SD higher BMI → PCOS OR 4.89; reverse direction null. PubMed ↗
Luque-Ramírez 2018 · Hum Reprod Update (33 RCTs). Head-to-head, HOMA-IR MD −0.46 favoring metformin over anti-androgens — treating the androgens does not fix insulin resistance. PubMed ↗
Lowering insulin resistance moves real PCOS outcomes — four independent levers, same direction:
Metformin — ovulation OR 2.55, clinical pregnancy OR 1.93, live birth OR 1.59 (Cochrane, 48 RCTs) Morley 2017 ↗
Low-GI diet — regular cycles in 95% vs 63% at equal weight loss Marsh 2010 RCT ↗
Weight loss — free-androgen index −1.11 (Cochrane) Lim 2019 ↗; GLP-1 → natural pregnancy RR 1.72 Ge 2023 ↗
Inositol — menstrual normalization RR 1.79, lowers testosterone, non-inferior to metformin Greff 2023 ↗
(Hard fertility / live-birth endpoints are weaker and less certain for inositol and diet alone.)
Flag 2 Precision

The metformin removal over a rising ALT

Ask your doctor

"Was my rising ALT worked up for fatty liver — and is metformin really the cause, given it's usually liver-safe?"

The clinical evidence

Metformin is an uncommon cause of liver-enzyme rises and may even lower ALT in fatty liver (LiverTox). In PCOS + insulin resistance the likely cause is fatty liver (NAFLD), which PCOS roughly doubles-to-triples (Rocha 2017; Yao 2023), and metformin is guideline-recommended for your metabolic profile (2023 PCOS guideline). The drug that was kept, atomoxetine, carries its own rare liver-injury signal (LiverTox).

If metformin stays off, it's worth asking about inositol (myo- + D-chiro-inositol) — a studied insulin-sensitizing option in PCOS — since otherwise nothing in the plan is acting on the root driver.

LiverTox · Metformin (NBK548726). "Metformin has not been linked to serum enzyme elevations during therapy… may actually lower elevated aminotransferase levels in patients with fatty liver disease… need not be avoided in patients with mild, preexisting serum enzyme abnormalities." NIH ↗
Rocha 2017 · J Endocrinol Invest (meta-analysis). PCOS↔NAFLD pooled OR 2.54 (95% CI 2.19–2.95), 17 studies. PubMed ↗
Yao 2023 · Endokrynol Pol. PCOS↔NAFLD OR 2.93 (2.38–3.62); "All patients with PCOS should undergo appropriate diagnostics for early detection of fatty liver and fibrosis." PubMed ↗
2023 International PCOS Guideline (Teede) · JCEM. Metformin recommended for PCOS + BMI ≥25 (insulin resistance, glucose/lipids, weight). PubMed ↗
LiverTox · Atomoxetine (NBK548671, likelihood C) + FDA label. "Atomoxetine has been linked to serum aminotransferase elevations in a small proportion (~0.5%)… several reports of clinically apparent acute liver injury." FDA: "discontinue in patients with jaundice or laboratory evidence of liver injury." NIH ↗
Flag 3 Recall

Screen for obstructive sleep apnea

Ask your doctor

"Given my PCOS, insulin resistance, and fatigue, should I be screened for sleep apnea?"

The clinical evidence

The 2023 PCOS guideline recommends OSA screening for symptomatic women (Teede 2023); OSA is ~3.8× more common in PCOS (Kahal 2020), insulin resistance is a stronger risk factor than weight (Vgontzas 2001), and untreated OSA impairs attention and focus — overlapping the symptoms treated as ADHD (Olaithe & Bucks 2013).

2023 International PCOS Guideline (Teede) · JCEM. "Women with PCOS have significantly higher prevalence of obstructive sleep apnoea compared with women without PCOS, independent of BMI" — screen symptomatic women. PubMed ↗
Kahal 2020 · Sleep & Breathing (meta-analysis). OSA prevalence 35.0% (CI 22.2–48.9%); OR 3.83 (1.43–10.24) vs controls. PubMed ↗
Vgontzas 2001 · JCEM. "Insulin resistance is a stronger risk factor than is body mass index or testosterone for [sleep-disordered breathing] in PCOS women." PubMed ↗
Olaithe & Bucks 2013 · Sleep (meta-analysis). Executive-function subdomains impaired in adult OSA, improving with CPAP. PubMed ↗
Flag 4 · more speculative Relevance · downstream

Atomoxetine may be feeding the loop

Ask your doctor

"Could atomoxetine be contributing to my insomnia — and might it belong on the differential for my ALT?"

The clinical evidence

Per the FDA label, atomoxetine causes insomnia in 15% vs 8% and raises heart rate (FDA label), and it has a rare liver-injury signal (LiverTox). So a symptom (focus) treated with atomoxetine may, in turn, worsen sleep and — rarely — the ALT: a loop worth untangling. A model, not a verdict — don't change any medication on your own.

FDA STRATTERA label · adult pooled RCT data. Insomnia 15% vs 8%; heart-rate rise ≥20 bpm 22.4% vs 8.3%; decreased appetite 16% vs 3%. DailyMed ↗
LiverTox · Atomoxetine (NBK548671). Rare clinically-apparent liver injury, typically 3–12 weeks after starting; FDA bolded severe-liver-injury warning. NIH ↗

The big picture — two models for your second opinion

① Before — what your plan assumes

Status-quo model: the plan treats each PCOS symptom on its own (spironolactone, minoxidil, supplements) so the cluster is considered managed, and assumes the rising ALT was caused by metformin — so metformin is stopped. There is no upstream driver in this picture.⤢ Expand

Each treatment (blue) points to the one symptom it's meant to manage. The targets sit side by side — nothing connects them, and no shared cause is named. And the one drug that acted on the metabolic root, metformin, was stopped (blamed for a rising ALT), leaving insulin resistance (the dark box) untreated. The "After" diagram shows what all of these actually have in common.

② After — the audited causal model

Audited model: insulin resistance at the hub; upstream exposures (diet, weight); the symptom cluster; the self-reinforcing loop (sleep apnea → ADHD/atomoxetine → ALT → metformin stopped → insulin resistance); and where each treatment acts. Flags 1–4 are labeled on the diagram.⤢ Expand

The same picture, audited. Dark = the root (insulin resistance, Flag 1); amber = upstream exposures you can influence; red = the self-reinforcing loop (the focus problem reads as ADHD → atomoxetine → which can worsen sleep and, rarely, the ALT that led to stopping metformin → worsening insulin resistance); blue = treatments and where each acts. The four flags are labeled on it. Show both to your doctor and ask whether they agree or work from a different model — comparing the two pictures is exactly what a second opinion is for. (click a diagram to expand it full-screen)

Appendix — every claim, classified (how the audit reasons)

For the curious: every claim in your plan, classified — this is the audit's underlying logic, summarised. The simple flags above are what matters; this is the receipts.

Each claim gets a plain-English verdict: Supported · Needs a look (in your plan, but depends on facts we don't have) · Missing (not in your plan, but worth raising).

ClaimTypeIn plan?Verdict — whyFlag
Explicit care-plan claims (what you wrote)
"Metformin had to be stopped because it raised my ALT"💊 RxexplicitNeeds a look — metformin is an uncommon ALT culprit; fatty liver & atomoxetine belong on the differential2
"My supplement stack covers my PCOS / insulin resistance"💊 RxexplicitNeeds a look — most of it treats symptoms; nothing is acting on the root driver since metformin came off1
"Atomoxetine for ADHD / focus"💊 RxexplicitNeeds a look — may worsen sleep and (rarely) the ALT; the focus problem may overlap untreated OSA3, 4
"Spironolactone for hirsutism"💊 RxexplicitSupported — standard for PCOS androgen symptoms (keep monitoring potassium)
"Omega-3 / vitamin D / B12 / K2 repletion"💊 RxexplicitSupported — reasonable for the stated deficiencies / IR-inflammation adjunct
Implicit care-plan claims (assumed, unstated)
"The rising ALT was caused by the metformin"🔬 DximplicitNeeds a look — PCOS roughly triples NAFLD; metformin can even lower ALT in fatty liver2
"Treating each symptom on its own is enough"🔬 DximplicitNeeds a look — insulin resistance is the upstream driver of the whole cluster1
"My focus problem is ADHD"🔬 DximplicitNeeds a look — untreated sleep apnea impairs attention and overlaps these symptoms3, 4
Missing — not in the plan, but needed (the audit adds these)
Measure insulin resistance directly — OGTT (best in PCOS) or A1C🔬 DxmissingMissing — the root driver is never measured1
Work up the ALT for fatty liver (NAFLD)🔬 DxmissingMissing — the likelier cause in PCOS + IR2
Screen for obstructive sleep apnea🔬 DxmissingMissing — guideline-recommended; OSA ~3.8× more common in PCOS3
Consider inositol (myo- + D-chiro) if metformin stays off💊 RxmissingMissing — an insulin-sensitizing lever when nothing else is working the root2
Lower glycemic load / weight — upstream IR levers💊 RxmissingMissing — moves the whole cluster, not one symptom1

This is a real, de-identified audit shown as an example — research and education, not medical advice. Want your own care plan audited? Email ops@nobsmed.com.