Clinical Outcome Assessments (COAs) and Patient-Reported Outcome (PRO) measures are essential for evaluating treatment effectiveness and patient well-being. However, when assessments are poorly designed, patients experience cognitive overload, leading to fatigue, response bias and inaccurate data.

In this article, we apply concepts from cognitive psychology, such as Miller’s Magic Number and digital amnesia in the form of the Google Effect and the still little understood LLM effect. We also follow regulatory guidance from the FDA and EMA, to demonstrate how assessments can be optimised for clarity, accuracy and usability.

We examine how cognitive overload manifests in different clinical areas and provide specialised recommendations for mental health, pain management and oncology.

Cognitive Overload in Mental Health Outcome Measures

Case study 1: Long recall periods in depression and anxiety assessments.

Many mental health PROs ask patients to recall symptoms over extended periods, which can be difficult due to fluctuating moods and memory biases.

Example:

“Over the past month, how often have you felt down, depressed or hopeless?”

Cognitive Load Issue:

·       The recall period is too long. Patients may struggle to report accurately mood fluctuations.

·       The Google Effect may lead participants to search online for symptoms before answering, affecting self-perception.

Better Question: shorter recall periods, FDA-aligned:

“In the past week, how often have you felt down, depressed or hopeless?”

Why this works:

·       Shorter recall periods reduce bias and improve response accuracy.

·       More frequent assessments provide better symptom tracking over time.

Case Study 2: Double-Barrelled Questions in PTSD Screening

Many PTSD assessments combine multiple concepts into a single question, increasing cognitive load.

Example:

“In the past month, how often have you had nightmares or flashbacks related to a traumatic event?”

Cognitive Load Issue:

·       Nightmares and flashbacks are separate symptoms. Patients may experience one but not the other.

·       Combining two symptoms forces respondents to separate mentally experiences, increasing cognitive strain.

A better worded question would have single-concept items, EMA-aligned.:

“In the past month, how often have you had nightmares related to a traumatic event?”

Why this works:

·       Breaking the question into two separate items improves accuracy and response reliability.

·       Patients can focus on one symptom at a time, reducing cognitive strain.

Cognitive Overload in Pain Management Outcome Measures

Case Study 3: Complex pain descriptors in chronic pain assessments.

Chronic pain patients often struggle with assessments that use medical jargon or too many response options, leading to decision fatigue.

Example question:

“Describe your pain using the following terms: aching, stabbing, burning, cramping, crushing, shooting, tingling, radiating, numb, dull, sharp, gnawing, heavy, throbbing, cutting, pressing, exhausting, sickening, punishing.”

Cognitive load issue:

·       Patients have to process and differentiate between many similar terms, increasing mental effort.

·       Too many options violate Miller’s 7±2 rule, leading to decision fatigue.

Better worded question, by reducing response options and aligning to FDA guidance:

·       “Which of the following best describes your pain? (Select up to 3)

o   Aching

o   Stabbing

o   Burning

o   Cramping

o   Tingling.”

Why this works:

·       Limiting choices to 5-7 prevents decision overload.

·       Allowing multiple selection ensures patients don’t feel forced into a single category.

Case Study 4: Long, fatiguing pain scales.

Pain scales often ask too many questions in a row, leading to mental exhaustion and rushed responses.

Example: “Rate your pain on a scale from 0-10 in the morning, afternoon, evening and nighttime.”

Cognitive load issue:

·       Patients must recall multiple time periods, increasing memory strain.

·       Repetitive questions extend survey length, increasing fatigue and dropout rates.

Better worded question could have progressive pain tracking, EMA-aligned:

“In the past 24 hours, when was your pain the worst?”

“What was your worst pain level on a scale of 0-10?”

Why this works:

·       It reduces the number of questions while still capturing meaningful data.

·       It helps patients focus on the most significant pain episode, reducing mental effort.

Cognitive Overload in Oncology Outcome Measures

Case Study 5: Multi-symptom fatigue questions in cancer PROs

Cancer-related fatigue assessments often combine multiple aspects of fatigue (physical, mental and emotional), making it difficult for patients to respond accurately.

Example: “In the past week, how often have you felt physically and mentally exhausted due to your cancer treatment?”

Cognitive load issue:

·       Physical and mental fatigue are distinct experience.

·       Forcing patients to mentally separate these categories increases effort.

A better worded question would single-symptom focus, FDA-aligned:

·       “In the past week, how often have you felt physically exhausted?”

·       “In the past week, how often have you felt mentally exhausted?”

Why this works:

·       It breaks down complex fatigue into separate, easier-to-answer questions.

·       Improves data reliability for oncology symptom tracking.

Case study 6: Emotional burden cancer quality of life measures.

Cancer patients already experience high emotional stress; surveys should be designed to minimise additional cognitive burden.

Example:

“How much has your cancer diagnosis affected your ability to work, socialise and enjoy hobbies?”

Cognitive load issue:

·       The question asks about three separate life domains (work, socialising and hobbies).

·       This is an emotionally taxing question that may lead to avoidance or rushed responses.

A better question would be to use gradual, separate items that are EMA-aligned:

·       “How much has your cancer affected your ability to work?”

·       “How much has your cancer affected your social life?”

·       “How much has your cancer affected your hobbies?”

Why this works?

·       It simplifies responses by focusing on one area at a time.

·       It reduces emotional strain, thus improving response quality.

Conclusion: Best Practices for Reducing Cognitive Load in COAs

·       Use short recall periods: FDA and EMA recommend 1-2 weeks max.

·       Do not use multi-concept questions, involving separate physical, mental and emotional symptoms.

·       Limit response options to 5-7 (in accordance with Miller’s Magic Number; 4-5 in accordance with Cowan’s Number)

·       Ensure clear, concise language, i.e. do not use medical or study-specific jargon.

·       Design digital assessments with progress indicators to minimise fatigue.

By implementing these principles, clinical researchers and healthcare providers can improve patient experience, increase data accuracy and enhance the effectiveness of COAs across different medical conditions.

Thank you for reading,

Mark Gibson, Lancaster, United Kingdom

Nur Ferrante Morales, Ávila, Spain

August 2025