Narrative: Laboratorios

Lab Results – The Most Anxiety-Inducing Screen in Any Health App

The Story

María uploads a photo of her lab results from her phone’s camera. The preview looks good – clear text, no shadows. She taps “Subir.” A progress bar fills to 100%. The screen transitions to a soft pulse animation with a timeline: “Recibido” (check), “Procesando” (active), “Listo” (pending). Copy reads: “Estamos analizando tus exámenes. Tus resultados estaran listos en 30 min a 2 horas. Te avisamos por push.”

María goes about her day.

47 minutes later, her phone buzzes. “Tus exámenes están listos. Veamos como te fue.” She taps the notification. What she sees is not a table of numbers with reference ranges she does not understand. She sees a visual story of her health. Cards grouped by category: Metabólico, Lipidos, Hepatico. Each card shows a name, a value, and a colored indicator. Green for “óptimo.” Blue for “revisar.” She taps on Glucosa. A large number: 98 mg/dL. Below it, a horizontal bar with three zones. Her value sits comfortably in the green zone. A small marker labeled “Los médicos se preocupan aquí” sits far to the right, at 126. She is not near it. She exhales.

Below the bar, a trend chart shows her last four measurements: 126, 115, 105, 98. A downward line. A badge reads: “Cambio real detectado (-22%).” Below that, one sentence: “Tu glucosa ha mejorado de forma significativa. Mantener actividad física y dieta baja en azucar.”

María understands her health. She did not need a medical degree. She did not call her doctor in a panic. She did not google “glucosa 98 es malo” and spiral through search results.

Principle: Reduce Anxiety, Increase Understanding

Lab results are where health anxiety lives. The moment between “your results are in” and “here is what they mean” is one of the most emotionally loaded moments in any patient’s journey. A bad presentation does not just confuse – it causes harm.

A PMC systematic review (2024) proved that horizontal bars with colored zones significantly outperform numbers with reference ranges for patient comprehension. The finding is not marginal. Patients viewing near-normal results presented as raw numbers were significantly more likely to contact their doctor unnecessarily. The number “98” next to the range “70-100” does not tell you whether you should worry. A green bar with your value clearly inside the safe zone does.

Our 3-zone visualization:

NUMBERS ALONE (anti-pattern):
  Glucosa: 98 mg/dL (Rango: 70-100)
  --> Patient does not know if 98 is "almost bad" or "perfectly fine"
  --> Unnecessary doctor contacts increase

3-ZONE BAR (validated pattern):
  Glucosa: 98 mg/dL
  []
  |bajo    |óptimo        |alto|
                    ^tu valor    ↑ harm anchor (126)
  --> Immediate comprehension
  --> Unnecessary doctor contacts decrease significantly

The third element – the harm anchor – is what separates our implementation from basic colored bars. The marker that says “Los médicos se preocupan aquí” at 126 mg/dL gives the patient a frame of reference that no number-plus-range display can provide. Eye-tracking research found that patients who correctly interpreted their results used 11 fixations (median) compared to 32 fixations for those who did not understand. The bar supports rapid pattern recognition. The harm anchor provides clinical context. Together, they let the patient answer the only question that matters: “Should I worry?”

The answer, for a value of 98, is no. And the interface says so without the patient needing to figure it out.

The Anti-Nocebo Protocol

The word “no óptimo” causes panic. This is not speculation. It is a direct patient quote from a user-centered design study published in PMC. Patients described the term as capable of “causar pánico en cualquiera.”

The nocebo effect is real: negative framing of health information produces measurably worse patient outcomes, independent of the actual clinical findings. A patient told their result is “not optimal” experiences stress, cortisol elevation, and behavioral paralysis – even when the result is clinically unremarkable.

ADEN never uses the following terms:

This is not cosmetic. It was validated across 297 interpretations in the engine. Every piece of copy that touches a lab result runs through the anti-nocebo filter before reaching the patient. The filter is rule-based (Layer 1), not AI-generated, which means it is deterministic, consistent, and cannot hallucinate a scary word into existence.

The design principle: every screen that shows a health number must answer the question “should I worry?” before the patient has time to ask it. Context first. Number second. Action third.

RCV: Explaining Change Without Alarm

When María’s colesterol drops from 245 to 238, is that real improvement or random noise? Without Reference Change Value, neither she nor her doctor knows. The 3% decrease falls within the biological variability of cholesterol measurement. It might mean nothing.

But when her glucosa drops from 126 to 98, the 22% decrease exceeds the RCV threshold of 15%. This is a real, statistically significant change. Her body actually improved.

Most health apps show both changes the same way: a green arrow pointing down. This is misleading. The cholesterol “improvement” might reverse on the next test. The glucosa improvement almost certainly will not.

ADEN’s comparison screen and detail screen make this distinction explicit:

Glucosa: 126 --> 98 (-22%)
[Cambio real detectado]
"Tu glucosa bajo 22%. Este cambio es REAL y refleja una mejora significativa."

Colesterol: 245 --> 238 (-3%)
[Variabilidad normal]
"Tu colesterol bajó 3%. Esto está dentro de la variación normal de tu cuerpo."

This transparency builds trust that no other consumer health app offers. When ADEN says a change is real, the patient can trust it because they have seen ADEN honestly tell them when a change was not real. The system earns credibility by acknowledging uncertainty.

The patient does not need to understand the statistics behind RCV. They need to understand one sentence: “This change is real” or “This is normal variation.” The math is invisible. The clarity is not.

OCR Wait: Making Processing Feel Premium

The upload-to-results pipeline takes 30 minutes to 2 hours. This is an eternity in mobile UX. A spinner would be unacceptable. A blank screen with “processing” would generate support tickets.

Instead, we treat the processing time as three discrete stages, each with its own visual state:

Step 1: Recibido      [checkmark]   "Archivo recibido"
Step 2: Procesando    [pulse anim]  "Estamos analizando tus exámenes"
Step 3: Listo         [pending]     "Te avisamos por push"

The 3-step timeline borrows from e-commerce order tracking (a mental model 87% of smartphone users already understand). “Recibido” confirms the upload worked. “Procesando” shows activity is happening. “Listo” sets the expectation for completion.

The push notification copy is deliberate: “Tus exámenes están listos. Veamos como te fue.” The first sentence is informational. The second sentence is invitational – “veamos” (let us see) creates a sense of companionship and curiosity. Not “your results are ready” (clinical, cold) or “view your results now” (transactional, urgent). “Let us see how you did” frames the results as a shared discovery, not a verdict.

The timeout state (4+ hours) is equally considered. Instead of a generic error, the copy reads: “Tus resultados están tardando más de lo esperado. Nuestro equipo ya está revisando.” Active voice. Someone is working on it. The CTA offers two paths: contact support, or enter results manually. The patient is never stranded.

Decisions

Engine Connection

This is where the engine’s Layer 1 – 76 clinical rules, pure deterministic logic – directly touches the user interface.

Every “fuera de rango” flag that appears on María’s screen comes from a rule that considers more than just the number. The engine evaluates:

• Age and sex. A creatinine of 1.1 mg/dL is normal for a 35-year-old male but warrants attention for a 28-year-old female.
• Ethnicity and population calibration. María’s HOMA-IR threshold is 3.8 (calibrated for Latin American populations), not 2.5 (derived from European cohorts). A value of 3.2 that would flag as insulin resistant in a European system is correctly classified as normal in hers.
• Genetic variants. If María carries the ABCA1 R230C variant (exclusive to Americas populations, present in ~10% of Colombians), her HDL cholesterol is expected to be lower without carrying the proportional cardiovascular risk. A low HDL that would trigger concern in another patient is contextualized in hers.
• SLC16A11 status. If María carries this variant (30-50% prevalence in Amerindian populations), her diabetes risk profile is different from what standard models predict. A glucosa of 105 might warrant closer monitoring for her than for a patient without the variant.

The user never sees this complexity. They see a colored bar, a harm anchor, and a one-sentence recommendation. But behind that simplicity is the most personalized lab interpretation available in consumer health. The bar is not just “green because 98 is between 70 and 100.” It is green because 98, for María’s age, sex, genetic profile, and population calibration, is genuinely optimal.

No other consumer health platform can make this claim. Plataformas consolidadas personalize by age and sex, or group by system. None integrate pharmacogenomics, polygenic risk scores, and population-specific calibration into a single result interpretation. ADEN does all three, and the patient experiences it as a simple, calming screen that answers: “How am I doing?”

Design Philosophy: Interpretation Over Raw Data

Most consumer health platforms present data as charts without interpretation. A blood pressure reading appears as two numbers on a timeline. The user can see the trend, but the platform never says “this is good” or “this needs attention.” The position is philosophically consistent: the device collects data; the clinician interprets it.

ADEN takes the next step. Every number has context, explanation, and action.

The industry says: “Here is your data.” ADEN says: “Here is what it means for you.”

Generic consumer platforms operate under regulatory constraints that make interpretation risky at scale. But ADEN’s engine – with its deterministic rules, LATAM calibration, and pharmacogenomic integration – can offer personalized interpretation that no generic platform can.

The design ethos is restraint. The detail screen shows one number, one bar, one trend, and one recommendation. Not twelve widgets competing for attention. Not a dashboard that requires a medical degree to parse. The information hierarchy is absolute: value first (hero), context second (bar), history third (trend), action fourth (recommendation). Each layer answers a more specific question:

1. Value: What is the number?
2. Bar: Is it good or bad?
3. Trend: Is it getting better or worse?
4. Action: What should I do?

If the patient stops after the bar, they have enough. If they want more, the trend and recommendation are there. The interface respects the patient’s cognitive bandwidth. It never forces them to process more than they are ready for.

This hierarchy is the standard ADEN holds itself to.