📄 health systems and quality improvement

Balancing data quality and participant burden: A comparative analysis of abbreviated vs extended symptom diaries in the CanTreatCOVID trial

This secondary analysis of the CanTreatCOVID trial demonstrates that abbreviated and extended symptom diaries yield comparable adherence, symptom trajectories, and predictive performance for recovery outcomes in outpatient SARS-CoV-2 patients, suggesting that shorter formats may effectively reduce participant burden without compromising data quality.

Original authors: Hosseini, B., Mohammadrezaei, D., Balalaie, P., Sivayoganathan, K., Condon, A., da Costa, B. R., Daley, P., Greiver, M., Jüni, P., Lee, T. C., McBrien, K., McDonald, E. G., Murthy, S. C., Selby, P.

Published 2026-01-28

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Hosseini, B., Mohammadrezaei, D., Balalaie, P., Sivayoganathan, K., Condon, A., da Costa, B. R., Daley, P., Greiver, M., Jüni, P., Lee, T. C., McBrien, K., McDonald, E. G., Murthy, S. C., Selby, P., Pinto, A. D.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine you are running a massive experiment to see how people recover from a cold or the flu. To track their progress, you ask them to fill out a daily "symptom diary" for two weeks. But here's the catch: some people get a tiny, 9-question checklist, while others get a massive, 34-question encyclopedia of symptoms.

The big question was: Does the giant diary give us better data, or does it just make people quit filling it out because it's too much work?

This paper, based on a real study called the "CanTreatCOVID trial," compared these two approaches. Here is the story of what they found, explained simply.

The Setup: The Short List vs. The Long List

Think of the 9-item diary as a quick "pulse check." It asks about the most common symptoms (like fever, cough, or fatigue) and takes just a minute to fill out.

The 34-item diary (called FLU-PRO Plus) is like a deep-dive investigation. It asks about everything from nose and throat issues to stomach problems and mood. It's much more detailed, but it takes longer to complete.

The researchers wanted to know: If we give people the long list, will they get tired and stop answering? And if they do answer, does that extra detail actually help us understand their recovery better?

The Results: The "Burden" Myth

You might expect that the people with the long, 34-question diary would get annoyed and quit early, while the people with the short list would stick with it.

Surprisingly, that wasn't really the case.

Starting the Diary: Both groups started filling out the diaries at almost the exact same rate.
Finishing the Diary: Both groups finished the full two weeks at very similar rates. The short-list group did finish slightly more often, but the difference was so small it could have been just luck.
The Real Driver: The thing that actually made people stick with the diary wasn't the length of the questions. It was whether they got a specific medication (Paxlovid). People taking the medication were much more motivated to keep filling out their diaries than those who didn't. It's as if the treatment gave them a boost of energy to keep going, regardless of how long the form was.

The Data Quality: Is "More" Better?

The researchers then looked at the information gathered. They asked: "Does the long diary tell us a more complete story?"

Symptom Patterns: Both diaries showed the same story: symptoms got better over time. The long diary didn't reveal any "secret" symptoms that the short one missed in a way that changed the overall picture.
Predicting Recovery: The researchers built computer models to guess when a person would feel "back to normal." They used the short diary data and the long diary data.
- The Short List: Worked great.
- The Long List: Worked great, too.
- The Difference: Adding the extra 25 questions from the long list only gave a tiny, almost invisible boost to the accuracy of the predictions. It was like adding a pinch of extra spice to a soup that was already perfectly seasoned; you couldn't really taste the difference.

The "Feeling Recovered" Check

The study also looked at how people defined "recovery."

When people said, "I feel recovered," they almost always agreed with the statement, "I am back to my usual health." (This was a very strong match).
However, saying "I feel recovered" didn't always match perfectly with "I am back to my usual activities" (like going back to work or the gym). Sometimes people felt better but weren't ready to run a marathon yet. This was true for both the short and long diary groups.

The Bottom Line

The study concludes that you don't need the giant 34-question diary to get good results.

The short, 9-question diary was just as good at tracking recovery, predicting when people would get better, and keeping people interested in the study. The extra questions in the long diary didn't add enough value to justify the extra time and effort required to fill them out.

In simple terms: If you are designing a study or a health app, a short, sweet checklist is often just as powerful as a long, complicated one. It saves time for the participant and saves money for the researchers, without losing the important information.

Technical Summary: Balancing Data Quality and Participant Burden in Symptom Diaries

Problem Statement
Symptom diaries are essential tools in clinical trials for acute respiratory infections, providing longitudinal patient-reported data on severity and recovery. However, a fundamental tension exists in trial design: longer questionnaires offer a more comprehensive clinical picture but increase participant burden, potentially reducing adherence and data completeness. Conversely, shorter formats improve retention but risk omitting relevant symptoms. While meta-analyses suggest shorter questionnaires generally yield higher completion rates, there is a lack of direct evidence comparing long and short diary formats within the same trial setting regarding their impact on data quality, recovery estimates, and predictive modeling. This study addresses this gap by evaluating whether an abbreviated symptom diary can perform comparably to an extended format in an outpatient SARS-CoV-2 trial.

Methodology
This study is a secondary analysis of the CanTreatCOVID trial, an open-label, individually randomized, multi-center adaptive platform trial involving non-hospitalized individuals with confirmed SARS-CoV-2 infection in Canada. The analysis compared two diary formats over a 14-day follow-up period:

Abbreviated Diary: A 9-item tool capturing core symptom severity on a 4-point Likert scale.
FLU-PRO Plus Diary: A 34-item validated tool capturing the same 9 core symptoms plus 25 additional items across systemic, nasal, throat, gastrointestinal, and chest domains.

The study population included 712 participants (638 in the Abbreviated group and 74 in the FLU-PRO Plus group). Baseline characteristics were well-balanced between groups.

Analytical Approach
The authors employed a multi-faceted statistical approach to evaluate adherence, symptom trajectories, and predictive performance:

Adherence: Logistic regression and linear regression assessed diary initiation, full 14-day compliance, and completion rates. Survival analysis (Kaplan-Meier and Cox proportional hazards models) evaluated time to first missed entry (dropout). Weighted Generalized Estimating Equations (GEE) modeled daily response behavior.
Symptom Trajectories: Linear mixed-effects models analyzed changes in daily symptom severity scores, incorporating fixed effects for day, treatment, diary type, demographics, and interaction terms.
Recovery Agreement: Cohen's Kappa coefficient ( $\kappa$ ) measured agreement between self-reported "feeling recovered" and two other metrics: "return to usual health" and "return to usual activities."
Predictive Modeling: GEE models and Cox proportional hazards models were used to predict recovery outcomes and time to recovery, comparing models using only the 9 core symptoms against models utilizing the full 34-item set.

Key Results

Adherence: Diary type was not significantly associated with the likelihood of initiating a diary or completing all 14 days. While the Abbreviated group showed slightly higher completion rates (68.1% vs. 64.4%) and compliance rates, the difference was not statistically significant ( $p = 0.239$ for completion rate; $p = 0.202$ for compliance). In contrast, treatment assignment (Paxlovid vs. Usual Care) was a highly significant predictor of adherence ( $p < 0.0001$ ).
Symptom Trajectories: Symptom severity declined significantly over the 14-day period across all groups. Diary type alone did not significantly affect average symptom severity. However, significant three-way interactions (day × diary × treatment) were observed for shortness of breath, muscle ache, and mood disturbance.
Recovery Agreement: Agreement between "feeling recovered" and "return to usual health" was strong to excellent across both diary types ( $\kappa = 0.8371–0.8859$ ). Agreement with "return to usual activities" was moderate ( $\kappa = 0.5273–0.6583$ ).
Predictive Performance: Predictive models for recovery outcomes performed well for both diary formats, with Area Under the Curve (AUC) values ranging from 0.87 to 0.94. Including the extended FLU-PRO Plus items (34 symptoms) resulted in only marginal gains in predictive performance compared to models using the 9 core symptoms.
Predictors of Recovery: In time-to-event models, baseline cough was the only symptom consistently associated with delayed recovery.

Significance and Claims
The paper claims that in the context of an outpatient SARS-CoV-2 trial, abbreviated and extended symptom diaries produced comparable adherence, symptom trajectories, and predictive performance for recovery-related outcomes. The authors assert that their findings support the feasibility of using streamlined, targeted symptom monitoring tools (such as the 9-item Abbreviated Diary) in acute respiratory infection research. This approach allows for the maintenance of participant engagement without sacrificing the ability to track recovery trajectories or predict outcomes.

The study concludes that while longer diaries provide more granular data, the marginal gains in predictive performance do not necessarily justify the increased burden, suggesting that concise formats are sufficient for many trial objectives. The authors note limitations, including the relatively small sample size of the extended diary group and the 14-day follow-up period, recommending future research to extend follow-up durations and test these formats in more diverse and digitally underserved populations.

The Setup: The Short List vs. The Long List

The Results: The "Burden" Myth

The Data Quality: Is "More" Better?

The "Feeling Recovered" Check

The Bottom Line

More like this