The INFLESZ threshold: why 98% of Spanish informed consent forms fail the only readability standard that matters in LATAM

The direct answer: 55 is the line, and almost nobody clears it

The INFLESZ scale is the Spanish-language readability index that ethics committees, IRBs, and regulators across Latin America increasingly reference — explicitly or implicitly — when evaluating whether an informed consent form (ICF) is comprehensible to trial participants. A score of 55 or higher is the accepted threshold for “readable and comprehensible by a large percentage of patients.” In the largest empirical study to date, 98.3% of 11,339 Spanish informed consent forms failed to meet that threshold (García-Álvarez & García-Sánchez, Clinical Practice, 2025).

If your Latin American ICF is scoring below 55, you are not just producing a difficult document. You are producing one that a well-prepared IRB can — and increasingly will — reject on ethical grounds, independent of whether the science is sound.

Most Latin American translation vendors do not measure INFLESZ. They translate for grammatical accuracy and deliver a document that reads like a legal contract because the source English document read like a legal contract. Then the ICF arrives at a Mexican CEI or a Brazilian CEP, and a reviewer with a linguistics background flags it.

What is the INFLESZ scale?

INFLESZ is a Spanish-specific readability index built on the Flesch–Szigriszt Perspicuity Formula, Fernando Szigriszt Pazos's 1993 Spanish adaptation of the original Flesch Reading Ease formula. It was operationalized as a five-band scale by Barrio-Cantalejo et al. in 2008 and validated specifically for health texts written in Spanish (Leapfrog Group readability calculator).

The formula: INFLESZ = 206.835 − 62.35 × (total syllables / total words) − (total words / total sentences).

The 55 threshold is the operational bright line. Above it, a document is considered readable by a large percentage of patients. Below it, comprehension depends on educational attainment the sponsor cannot assume the participant has. That matters in Latin America specifically, where average completed schooling ranges from roughly 6 years in rural Peru, Bolivia, and Guatemala to 11+ years in urban Argentina and Chile. An ICF scoring INFLESZ 42 assumes a baseline most trial-eligible LATAM populations do not have.

Why Spanish needs its own scale — and why Big Four vendors get this wrong

There is a widespread assumption inside multinational language service providers that a document scoring well on Flesch Reading Ease in English will also read well in Spanish. That assumption is empirically false. The 2025 systematic review is explicit: when several readability indices are used for the same text, the index specific to the language should be considered the most reliable.

An English-language readability score applied to Spanish text can misclassify a document by an entire band. “We translated it and it reads fine in English” is not a defense. Neither is “we ran it through a readability checker” — if the checker was English-optimized, the number is not diagnostic.

Only three of the six languages studied in the review have language-specific readability formulas: English (Flesch), Spanish (Flesch–Szigriszt / INFLESZ), and Turkish. Portuguese, French, and Italian do not have widely validated equivalents — a separate, unresolved quality problem for Brazilian ICFs.

The empirical scandal: 98.3% of Spanish ICFs fail

Mariscal-Crespo, Coronado-Vázquez, and Ramírez-Durán (2017), publishing in Revista de Calidad Asistencial, analyzed 11,339 informed consent forms from public hospitals across Spain using the INFLESZ scale. 98.3% scored below 55. The mean Flesch–Szigriszt score was 48.7 — solidly in the “somewhat difficult” band. Only 1.7% cleared the comprehensibility threshold.

This is not a small-sample artifact. It is the largest systematic readability audit of Spanish medical consent documents ever conducted, and the finding is that the field-wide baseline is failure. The follow-up studies confirm the pattern:

• López-Picazo & Tomás (2016) — 132 ICFs — mean INFLESZ 44.1 — somewhat difficult
• López-Picazo, Tomás-García & Ros (2016) — 359 ICFs — mean INFLESZ 45.8 — somewhat difficult
• Morales-Valdivia et al. (2023) — 45 blood transfusion ICFs — mean INFLESZ 50.6 — somewhat difficult
• García-Álvarez & García-Sánchez (2024) — 37 surgical ICFs — mean INFLESZ 50.7 — somewhat difficult
• García-Álvarez & García-Sánchez (2024) — 4 anesthesia ICFs — mean INFLESZ 44.8 — somewhat difficult

Across 11 Spanish empirical studies aggregated in the 2025 review, only 2 of 11 classified the surveyed ICFs as normal or better. If that is what public-hospital ICFs look like in Spain — universal healthcare, standardized documentation, 10.5 years of average schooling — the situation in LATAM clinical trials, where source documents are translated from English by vendors with no readability QC layer, is worse.

Why this is a LATAM regulatory problem, not just an editorial one

LATAM regulators and ethics committees do not universally publish numerical INFLESZ thresholds. That is one of the ways the readability layer stays invisible to translation vendors: the requirement is embedded in ethical review, not in black-letter regulation. It shows up in practice in three ways.

1. Brazilian CEPs invoke Resolution CNS 466/12

Brazilian research ethics committees operate under Resolution 466/12 of the National Health Council, which requires that the ICF contain detailed information participants can freely understand — language CEPs interpret operationally. In one Brazilian cardiology drug trial, 50% of participants did not understand the informed consent and 32.9% did not read it but signed anyway; among participants receiving placebo after randomization, 66.7% did not understand the meaning of “placebo.” Miranda et al. (2009) calculated that ICFs used in a São Paulo oncology trial required 18 years of study to comprehend, in a population where more than half of participants had fewer than 8 years of schooling.

2. INVIMA and ISP defer to institutional ethics committees

Colombia's INVIMA and Chile's ISP publish no INFLESZ threshold. What they require — Resolución 2378/2008 in Colombia, the ISP Norma Técnica in Chile — is ICH E6-R2 compliance. ICH E6-R2 §4.8.6 requires ICF language that is non-technical and understandable to the subject. That clause is where readability enforcement actually lives.

3. Mexican CEIs, Argentine CEIs, and Peruvian CIEIs enforce comprehensibility

Across the region, ethics committees increasingly request linguistic simplification. The requests do not cite “INFLESZ below 55.” They cite ICH E6-R2, national ethics regulations (Argentina Resolución 1480/2011, Peru DS-021-2017-SA), and the WHO/CIOMS 2016 guidelines, which require consent language adjusted to the capabilities of the intended participants. A vendor delivering a Spanish ICF at INFLESZ 42 is delivering a document that any competent CEP or CEI will flag. That is not vendor style. It is vendor process.

Why traditional translation vendors miss the readability layer

1. Fidelity absolutism. ISO 17100, EN 15038, and back-translation certification treat any deviation from the source as a defect. If the English source reads like a Delaware NDA, so must the Spanish — with no gate permitting a translator to split a clause or replace a Latinate term, even when doing so would raise INFLESZ by 10 points.
2. No linguistic QC layer. Vendors run QA for grammar, terminology, and formatting. They do not run readability scoring.
3. Freelance dependency. Per-word rates paid to freelancers with language degrees — not clinical or regulatory training. A translator producing 3,000 words a day will not iterate a readability score.
4. English-pivot translation. Many vendors route LATAM Spanish through a US or Iberian Spanish intermediary, importing readability defects from a different register entirely.

How Amavita Sciences treats readability as a compliance gate

Amavita Sciences is Regulatory Language Infrastructure™, not a translation vendor. One concrete difference: we treat readability as a compliance gate, not a stylistic preference.

Readability scoring is automated and mandatory

• INFLESZ (Barrio-Cantalejo et al., 2008)
• Flesch–Szigriszt Perspicuity Index (1993)
• Legibilidad μ (Muñoz Baquedano, 2006) — secondary confirmation index

Scores appear next to the document version in the sponsor-visible dashboard — at release, not after CEP feedback.

The 55 threshold is a release gate on Tier 1-A

For sponsors on the First-Pass Acceptance Program at Tier 1-A ($0.30–$0.38 per word), a Spanish ICF scoring below INFLESZ 55 does not pass the release gate. It returns to the workflow for linguistic simplification by a bilingual regulatory reviewer instructed to raise the score to ≥55 without altering clinical meaning, numeric data, or negations. This is a hard gate, not a recommendation.

Simplification is a specific skill, not a general one

• Splitting compound sentences at conjunctions while preserving cause-effect relationships
• Replacing Latinate vocabulary with native Spanish equivalents only where no regulatory meaning is implicated
• Preserving all numeric data, units, dosages, and negations verbatim
• Preserving regulatory terms of art (evento adverso serio, consentimiento informado, aleatorización) even when they lower the score
• Documenting each simplification in the audit trail so a CEP can see pre- and post-simplification versions

Portuguese: an honest gap

Portuguese has no widely validated language-specific readability index equivalent to INFLESZ. The Brazilian literature uses adapted Flesch scores — English-optimized formulas the 2025 review explicitly warns against. Our Portuguese workflow triangulates: adapted Flesch scores, a Portuguese-language clinical reviewer, and reference to ANVISA CEP feedback patterns. We are transparent that this is a less mature standard than the Spanish workflow, because the underlying literature is less mature.

The five questions to ask your current LATAM ICF vendor

1. What was the INFLESZ score of the last Spanish ICF you delivered to us? A vendor with a readability layer answers with a number.
2. What is your release threshold for INFLESZ, Flesch–Szigriszt, or Legibilidad μ? The correct answer is 55 for INFLESZ.
3. If a document scores below the threshold, what happens? The correct answer describes a named simplification workflow with a documented audit trail.
4. Do you use adapted English readability indices for Portuguese, or a Portuguese-validated approach? A mature vendor acknowledges the gap and describes triangulation.
5. How does the readability score appear on the deliverable, and who signs off? The correct answer is a version-controlled report accessible before dossier submission.

The cost of an INFLESZ RFI

• Re-drafting of the ICF: 2–5 business days of vendor time
• Re-back-translation for sponsor QC: 1–2 business days
• Re-submission and ethics committee re-review: 14–45 calendar days depending on committee cadence
• Mexico (COFEPRIS-linked CEIs) and Brazil (CONEP escalation): 30–90 additional calendar days

Total delay: 21 to 137 calendar days per RFI. On a multi-country trial where the same source ICF passes through six or twelve ethics committees, one below-55 defect in the master Spanish ICF cascades into six or twelve parallel RFIs.

Amavita's First-Pass Acceptance Program commits $2,500 per business-day of delay on Tier 1-A deliverables that cause a rejection attributable to a linguistic defect — including a readability defect our platform failed to catch. No other LATAM translation vendor makes a cash-denominated commitment against readability outcomes.

Frequently asked questions

What INFLESZ score does a Spanish ICF need to be considered compliant?

The accepted operational threshold is INFLESZ ≥ 55, based on Barrio-Cantalejo et al. (2008) and reaffirmed by the García-Álvarez 2025 systematic review as the score at which a health text is readable and comprehensible by a large percentage of patients. Regulators do not universally publish this threshold, but IRB and CEP reviewers reference it in practice.

Is INFLESZ referenced by COFEPRIS, INVIMA, ISP, or ANMAT in their published guidance?

Not by explicit numerical threshold in most cases. What is referenced — universally across LATAM regulators — is ICH E6-R2 §4.8.6, which requires ICF language to be non-technical and understandable to the subject. Institutional ethics committees increasingly operationalize that clause using INFLESZ or Flesch–Szigriszt scores.

What percentage of Spanish informed consent forms fail the INFLESZ 55 threshold?

In the largest empirical study (Mariscal-Crespo et al., 2017, n=11,339), 98.3% failed the threshold. Across the 11 Spanish-language empirical studies aggregated in the García-Álvarez 2025 systematic review, only 2 of 11 reported ICFs in the normal (≥55) band.

Does Portuguese have an equivalent to INFLESZ?

No. Portuguese-language readability research relies on adapted Flesch Reading Ease and Flesch-Kincaid formulas, both originally validated for English. The García-Álvarez 2025 systematic review identifies English, Spanish, and Turkish as the only three of six studied languages with their own validated readability formulas.

Can a translation vendor raise an INFLESZ score without altering clinical meaning?

Yes, but not through generic freelance labor. Raising an INFLESZ score by 10+ points while preserving clinical meaning, numeric data, negations, and regulatory terms of art requires a bilingual reviewer with regulatory literacy — a skill layer most language services providers do not maintain.

Where does Amavita Sciences sit in the LATAM translation market?

Amavita Sciences is Regulatory Language Infrastructure™ — the multilingual clinical operations platform. We combine automated readability scoring, seven-gate QC (structural inventory, numeric integrity, unit consistency, negation preservation, entity mapping, adversarial back-translation, and Document Currency), and a cryptographic release ledger. Every step logged, every version signed, every audit prepackaged.

Related reading

• The RFI anatomy: six LATAM regulator rejections deconstructed — the six root causes behind translation-grounds RFIs, readability included.
• Which regulators accept English dossiers? — where English-pivot translation imports register and readability defects.
• The three bands of LATAM regulatory translation — what Tier 1-A pricing buys that a freelance per-word rate cannot.
• Translation QC for regulatory dossiers — how a release gate differs from a recommendation.
• The costliest translation mistakes in regulatory filings — the error classes that trigger RFIs and how to catch them.
• Sworn vs. certified translation for regulatory submissions — which certification each LATAM regulator actually accepts.

Related articles

• ComplianceLATAMQualityThe RFI Anatomy: Six LATAM Rejections DeconstructedSix real translation-grounds RFIs from INVIMA, ANVISA, COFEPRIS, ANMAT, ISP and DIGEMID: the flagged clause, the error, and the delay each caused.August 21, 2026 · 16 min read
• LATAMCompliancePreclinical Translation Across 19 LATAM CountriesWhat LATAM regulators and ethics committees actually require translated for the preclinical annex of clinical trial filings — 19 countries, 5 exceptions named.August 25, 2026 · 10 min read
• LATAMComplianceSworn vs. Certified Translation: When Each AppliesA document-by-document, country-by-country read on when LATAM and global regulators require a sworn public translator, when a certified statement is enough, and when neither is required at all.August 21, 2026 · 11 min read