Our sources
Every source SafeMomma cites was checked for who runs it, who pays for it, and political or money pressure. Here is what we found and how we use each one.
Last audited 2026-10-14
How we decide
- Two-incentive rule: Every claim needs two sources with different owners and different money. Two US agencies count as one. Two summaries of the same study count as one.
- High-stakes or contested claims also need an Independent source or the original study, read in full.
- "Incentive exposure" describes pressure from money or politics. It is not a score for accuracy.
- Labels appear next to every citation on item pages.
Status legend
- IndependentPublic funding, no industry money, used as the independent check
- SupportingCited, but a second source with different owners and money has to agree
- LabeledCited only with its label showing, never on its own
- Search toolUsed to find studies; we cite the study itself
Independent
Being added. We have asked European public services for permission to list them.
Supporting
Who pays for it: ACOG's 2024 tax return reports $63,505,977 in revenue, including $13,263,216 in contributions, or 20.9%; donor sectors are not identified, and income from its advertised promotions was not verified.
Why this status: ACOG is clinician-led and publishes evidence-graded guidance, but it also sells advertising and the full methods for some guidance were not available for review. SafeMomma treats it as supporting evidence, not an independent check.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Cochrane's 2024 audited accounts report £10.52 million total income, including £7.045 million in Wiley Library royalties, about 70% of annual turnover; Groups estimate a further £10–15 million yearly from host institutions, governments and charities outside central accounts.
Why this status: Cochrane publishes transparent evidence reviews under restrictive conflict rules, but review age, author interests and publishing-income dependence still need checks. A relevant review is used as evidence synthesis, not as a stand-alone clinical directive.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: For the fiscal year ended May 31, 2025, Consumer Reports reported $244.422 million in operating revenue and support, including $208.333 million from subscriptions, newsstand and other sales and $32.531 million in contributions.
Why this status: Supporting because member-funded product testing has strong independence controls, but it is not a clinical authority. Product tests and chosen thresholds do not replace public-health guidance.
Incentive exposure: Low · Audited 2026-10-14
Who pays for it: EFSA's 2025 budget was €163.1 million in commitment appropriations and €150.9 million in payment appropriations and EU contribution; the reviewed budget materials identified no industry fees, advertising, or product-sales funding, while a separate private-contribution amount was unverified.
Why this status: EFSA receives public EU funding and has conflict rules, but it also evaluates company-submitted evidence and has faced contested study appraisals. It is cited with a second source that has a different owner and funding profile.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: IARC's audited 2025 statements list €24.867 million in assessed contributions and €25.087 million in voluntary contributions, including research grants, within €55.046 million total revenue.
Why this status: IARC publishes its funding and conflict process, but research grants and state contributions create financial and political exposure. Its Monographs classify cancer hazards, not risk at a particular exposure, so the findings are paired with a second source and exposure-specific evidence.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Region Stockholm provides and operates Janusmed, but the reviewed pages give no service budget, funding shares, or specific industry contribution; the amount is unverified.
Why this status: Janusmed describes itself as manufacturer-independent, but its full budget and author disclosures were not verified and some register data are not adjusted for background differences. SafeMomma uses it as an additional check, not as a stand-alone basis for important claims.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: FAO and WHO fund scientific advice through regular budgets and extra-budgetary resources, usually government contributions linked to specific activities; a current JECFA-specific budget or donor amount was not verified.
Why this status: JECFA has formal expert conflict procedures and primarily public or institutional funding, but it can assess data supplied by manufacturers, including proprietary data. It is used for food-chemical questions with its assessment date and data limits visible.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: LactMed is free to access, and its process page identifies NICHD development and NLM support; a LactMed-specific budget or industry funding was not identified, so amounts are unverified.
Why this status: LactMed has a named literature compiler and a peer-review panel, but its source-specific budget and author or record-level financial disclosures were not found. Important claims are paired with evidence from a source with a different owner and funding.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Lareb's 2027 budget proposal lists an institutional subsidy of €2.097 million for 2025 and €2.783 million for 2026 and each year 2027–2031; this is not TIS-only funding or total revenue, and the 2027 amount is proposed.
Why this status: Lareb's TIS is a public service, but its funding spans several tasks and its public overview gives limited detail on review methods and author disclosures. SafeMomma uses it with a separately owned source, and keeps TIS advice distinct from adverse-event reports.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: OTIS reports 16 sponsored pregnancy registries and, in FY2024, $2,346,081 in revenue, including $2,128,619 in contributions whose sources are not broken out.
Why this status: OTIS has specialist writers and publishes pregnancy research, but it also receives government and private funding and runs drug-company-sponsored registries. SafeMomma lists it as supporting, so claims need agreement from a source with different owners and funding.
Incentive exposure: Medium · Audited 2026-10-14
- NHS Specialist Pharmacy Service (SPS) / UK Drugs in Lactation Advisory Service (UKDILAS)UK governmentUnited Kingdom
Who pays for it: NHS England commissions SPS; the only amount found was a 2014 estimate of £7.1 million for all specialist pharmacy services in England, not SPS/UKDILAS alone, and funding after 31 March 2026 is unverified.
Why this status: UKDILAS provides specialist breastfeeding guidance based on several sources, medicine handling in the body, and expert judgment. Its post-March 2026 contract status is unverified, so SafeMomma uses it with another source for important claims.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: NTP/OHAT-specific total funding was unverified in the reviewed public materials, and no program-specific industry fees, sponsors, or donor revenue were identified.
Why this status: NTP/OHAT uses published systematic-review methods and external review, but it operates within the federal government and its program-specific funding remains unverified. It is citeds assessments with a second source and distinguishes hazard findings from exposure-level conclusions.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: REPROTOX earns subscription income; its fiscal year ending June 2026 reported $776,465 in revenue and $207,714 in contributions, whose sources are not named in the public summary.
Why this status: Its expert-written summaries cite studies and show update dates, but public pages do not name all contributors or disclose entry-level conflicts. SafeMomma lists REPROTOX as supporting if access is licensed, so key claims need an independent source and the cited research.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: No SCCS-only annual budget was identified; under 2024 Commission rules, members receive up to €450 per full working day from Commission funds, plus travel and subsistence.
Why this status: SCCS experts are Commission-funded and publish conflict declarations, with no product-company funding identified. Opinions may rely on applicant dossiers, so they are cited for the specific assessed use and checked against a source with a different owner.
Incentive exposure: Low · Audited 2026-10-14
Who pays for it: For July 2024 through June 2025, Silent Spring reported $5,758,023 in revenue, including $5,672,563 in contributions and grants (98.5%); the filing does not show donor sectors.
Why this status: Supporting because its peer-reviewed studies can inform research, while its advocacy mission and unknown donor-sector shares call for a separate source with different incentives. Hazard screening does not establish an individual's exposure or risk.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: PEAR-Net says TERIS subscriptions help sustain TERIS and Shepard's Catalog; current prices, financial statements, funders, and TERIS-specific amounts were not public in the reviewed pages.
Why this status: TERIS publishes expert consensus ratings and explains its risk method, but current finances and entry revision dates are not public. SafeMomma lists TERIS as supporting if access is licensed, so important claims need current research and a separate source.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: For the year ended March 31, 2025, BfN reported £2,801,689 in income; its accounts say local-authority, Department of Health and Scottish Government commissions are the majority of funding.
Why this status: Supporting because its pharmacist-maintained factsheets use research and national guidance, with a clear policy rejecting commercial infant-feeding support. Public commissioning is a major funding source, and medication pages need a separate current check.
Incentive exposure: Medium · Audited 2026-10-14
Labeled
Who pays for it: ABM's 2024 tax return reports $654,022 in revenue, including $305,303 in contributions, or 46.7%, with donor sectors unlisted; it permits reviewed, WHO Code-compliant meeting sponsors, but the amount was not verified.
Why this status: ABM publishes protocol-level methods and disclosures, but permits commercial meeting support and its current ethics-code file was not available for review. SafeMomma therefore cites ABM only with a visible label.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: AAP's tax return for the year ending June 2025 reports $138,498,070 in revenue; AAP told Undark corporate support was about 4% of revenue for July 2024 through June 2025, and its October 2025 page listed named supporters in giving bands.
Why this status: AAP lists corporate supporters, including companies in medicine and infant nutrition, and recent reporting raised questions about the visibility of some guideline-author disclosures. The audit found no evidence that sponsors dictated recommendations, but SafeMomma uses AAP only with a visible label.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: CDC's FY2025 core public-health program level was $9.248 billion; separately, the CDC Foundation reported $176.8 million in FY2025 grants and contributions, including $5.1 million from corporations and $98.8 million from government grants and contracts, without showing that donors funded the reviewed pages.
Why this status: CDC is a federal source, but the audit records high incentive exposure, policy changes, court action, and page-date differences. SafeMomma shows the government label and specific page date, and checks contested pregnancy or vaccine claims against another clinical source.
Incentive exposure: High · Audited 2026-10-14
Who pays for it: For the fiscal year ending December 2023, CLP reported $720,193 in revenue, including $668,335 in program-service revenue and $42,664 in contributions; CLP says donations, grants and brand certification fees support it.
Why this status: Labeled because brand certification fees are part of its funding and public reports do not release product-level results. Its screening reports are not individual exposure or clinical risk assessments.
Incentive exposure: High · Audited 2026-10-14
Who pays for it: CIR says PCPC funds the program, but the amount and share of PCPC's budget attributable to CIR were not identified in the reviewed materials.
Why this status: Labeled because PCPC funds CIR and chairs its policy committee, while a distinct expert panel makes the scientific conclusions. CIR reviews are not FDA approvals or product-level risk assessments.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: DailyMed is an NLM government service, but its specific budget was not identified; companies submit the labels, and FDA's FY2025 Prescription Drug User Fee Act (PDUFA) human-drug-review program had $1.759 billion in obligations, 77% ($1.360 billion) from user fees and 23% ($399.0 million) from appropriations, not funding for DailyMed or FDA as a whole.
Why this status: Drug companies submit the label text, and NLM says it does not review it before publication or guarantee it matches the latest FDA-approved label. SafeMomma cites it only for exact product-label wording, with product type and date visible.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: For the fiscal year ending December 2025, EWG reported $16,366,034 in revenue, including $13,191,054 in contributions and grants; it also earns licensing, certification and affiliate income.
Why this status: Labeled because EWG is an advocacy group with certification and affiliate revenue. Its rankings are relative screening tools, not personal exposure assessments.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: EMA's 2025 final budget was €600.23 million, with 91.4% of income from the broad category of medicine evaluation and related activities, 8.3% from the EU budget, and 0.3% from other sources; its 2026 page estimates 91.5%, about €563.2 million, from fees and charges.
Why this status: EMA is labeled as an EU regulator because it relies heavily on regulatory fees. Its public assessments and expert declarations are useful, but it is cited with an independent check, not alone.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: FDA's FY2025 PDUFA human-drug-review program had $1.759 billion in obligations, including $1.360 billion (77%) from pooled user fees and $399.0 million (23%) from appropriations; this is not FDA's total budget, and no separate budget for the FDA/EPA fish advice was identified.
Why this status: FDA is the primary source for its own notices and fish advice, but its audit rates incentive exposure as medium and identifies leadership exposure and pooled review fees. SafeMomma shows its government label and checks contested clinical interpretations against another source.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Springer sells the reference, and its catalog listed the 2027–2028 22nd edition from US$90 at review; sales revenue, author royalties and pharmaceutical underwriting were not disclosed.
Why this status: This is a paid reference with authorship overlapping InfantRisk, so the two are not independent checks. It is cited with a paid-reference label, edition and monograph date checks, and an outside cross-check.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Health Canada says most funding comes from Parliamentary appropriations; it reported CAD$317.4 million in fee revenue in FY2024–25, including CAD$98.3 million in human-drug submission examination fees.
Why this status: Health Canada has a published scientific-integrity policy, but it is a federal regulator and receives cost-recovery fees from regulated companies. Its notices are labeled as regulatory monitoring and checked against an independent lactation evidence source.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: HBBF has no separate public Form 990 or verified donor-sector breakdown; its May 2025 rice report says a Freedom Together Foundation grant supported the study, but does not state the amount.
Why this status: Labeled because HBBF combines commissioned testing with advocacy and its standalone finances and donor sectors are not verified. Its reports describe selected samples, not personal exposure or risk.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: InfantRisk advertises pharmaceutical and research partnerships, sponsored studies and consulting, while the Payne Foundation supports its free call center; sponsor-specific payment totals were not found.
Why this status: The Center markets research and consulting to pharmaceutical companies whose products may be assessed, but no sponsor-to-entry link was verified. It is cited with a visible commercial-exposure label and an independent check.
Incentive exposure: High · Audited 2026-10-14
Who pays for it: For the fiscal year ending March 2025, LLLI reported $1,502,702 in revenue, including $1,350,193 in contributions (about 89.9%); the filing does not classify donor sectors.
Why this status: Labeled because LLLI is a peer-support organization, not a clinical guideline developer, and its pages do not follow a uniform evidence-grading method. Its donor-sector split and detailed standalone conflict-policy terms could not be verified.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: Congress provided $170.4 million for FY2026; NCCIH also accepts unconditional and conditional research gifts, but its donation page did not list donor identities or totals.
Why this status: NCCIH is a federally funded research center with expert review and examples of reporting mixed results. It accepts conditional research gifts, has variable page-update timing, and faces direct budget and policy exposure, so SafeMomma labels it and corroborates important claims.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: The NHS says DHSC funds the website, which carries no advertising or corporate sponsorship; the sampled page's specific cost is not stated.
Why this status: NHS public pages have government funding and policy sign-off, and the sampled pregnancy page is a broad signpost rather than a medicine-by-medicine review. SafeMomma shows a UK-government label and does not use the page alone for clinical evidence.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: ODS reports a FY2024 budget of $28.5 million, including $20.4 million for research projects, with grant co-funding its largest allocation; direct industry contributions were not quantified.
Why this status: ODS is a federal office, and industry representatives took part in planning. The audit found no evidence that industry funds or controls ODS content, so SafeMomma labels the source and checks claims against independent evidence.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: SMFM's 2025 tax return reports $10,346,279 in revenue, including $2,126,055 in contributions with donor sectors unlisted; it also markets sponsorships and exhibits, but their revenue was not verified.
Why this status: SMFM uses GRADE, a system for rating evidence and recommendations, and publishes a detailed conflict-disclosure policy. It also accepts event sponsorships and has industry members, while the sponsor-revenue share is unknown, so SafeMomma uses it with a visible label.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: The FSA's 2024–25 accounts report £48.830 million operating income, including £48.293 million contract revenue and £0.537 million other income, against £193.049 million operating expenditure; no statement-specific funding was identified.
Why this status: The FSA has public scientific-adviser conflict registers and a risk-assessment remit, while it remains subject to government funding and appointments. Its COT material is labeled as UK food and chemical-risk guidance, not as a medication-in-milk database.
Incentive exposure: Medium · Audited 2026-10-14
Who pays for it: WHO's 2025 audited accounts report US$3.138 billion revenue, 83% from voluntary contributions and 13% assessed; 92% of voluntary contributions used for Programme Budget activities were earmarked, and the top 10 contributors supplied 57% of Programme Budget contributions.
Why this status: WHO uses formal guideline and conflict-management processes, while its funding is concentrated and its governance is political. It is cited with an international-agency label and the date and evidence basis of the specific guideline.
Incentive exposure: Medium · Audited 2026-10-14
Search tool
Who pays for it: PubMed is free and federally operated by NLM and NIH; a PubMed-specific budget or industry revenue was not identified, so amounts are unverified.
Why this status: PubMed is a search and indexing platform, not the publisher or author of most indexed papers. Inclusion does not rate article quality or mean NLM endorses a paper, so SafeMomma cites the study itself rather than PubMed as evidence.
Incentive exposure: Medium · Audited 2026-10-14
How we test sources over time
Valproate (seizure medicine) in pregnancy
Large studies from 2009 to 2013 linked it to birth defects and lower IQ scores. Regulators added stronger warnings in 2013 to 2018.
Some sources updated years after the regulators. Risk numbers differ between sources because they describe different outcomes.
Topiramate (seizure and migraine medicine) in pregnancy
Studies reported a higher rate of cleft lip in 2011 and neurodevelopment signals in 2022 to 2023. A 2024 study found no link to autism.
Two of the Nordic studies use overlapping data, so they count as one line of evidence. One source changed its wording without showing the disagreement.
Codeine while breastfeeding
FDA said in 2017 that breastfeeding is not recommended with codeine. Two journal columns behind earlier concern were retracted in 2020.
Sources still split: regulators advise against it, while several lactation references describe short, careful use. Even top sources sometimes overstate a study.
Ondansetron (nausea medicine) early in pregnancy
A 2018 study reported a small rise in oral clefts. EMA updated its advice in 2019.
Some sources mix up cleft types or leave out the size of the risk. Some updates came about 3 years later.
Acetaminophen (Tylenol) in pregnancy
In September 2025 US officials linked it to autism. FDA started a label process and said a cause was not established.
Public bodies split. CDC added avoidance wording. ACOG, SMFM, EMA, MHRA and WHO kept it as a first choice. Non-US public services help balance politically charged topics.
Fish and mercury
US advice moved from a limit only (2004) to a minimum and a limit (2014 draft) to a chart by fish type (2017).
Countries set different limits. One advocacy guide still used 2014 draft wording years later.
Caffeine
The UK lowered its limit to 200 mg a day in 2008 after one study. The study was corrected in 2010.
Limits differ: ACOG, EFSA and NHS say 200 mg, Health Canada says 300 mg. The numbers are cautious limits, not a sharp line.
Soft cheese and listeria
Outbreaks linked to pasteurized Hispanic-style soft cheese. CDC and FDA changed queso fresco wording in 2025.
Pasteurized does not mean listeria-free. Some US pages still carry older wording. Canada and Australia are stricter.
Retinoids in skincare
Studies of skin creams did not show more birth defects but were small. The EU limited retinol in cosmetics in 2024.
Retinoid pills and creams are different. One medical society page called a prescription retinoid an over-the-counter retinol.
Hair dye
IARC rates workplace exposure as probably cancer-causing, and personal use as not classifiable. A September 2026 review rated pregnancy evidence low to very low.
Workplace studies and personal use are different exposures. Reassurance about breastfeeding mostly comes from reasoning about low absorption, not direct milk studies.
Not used as evidence
- Health blogs and SEO content sites
- Forums, Reddit, or anecdotal posts
- Social media or influencer content
- AI chat answers without a primary source link
- Drug manufacturer marketing pages (official labeling via DailyMed is different)
- Supplement brand or seller sites
- Herbalist blogs
- Paywalled databases we cannot link to
Found an error?
Email support@safemomma.org. Corrections are dated on the source page.