Feasibility and responsiveness to an electronic system to collect long-term patient-reported outcome measures after rectal cancer resection

Annalisa Maroli, Filippo Maria Bianchi, Caterina Foppa, Stefano De Zanet, Fabio Zangrandi, Michele Carvello, Carlotta La Raja, Antonino SpinelliBJS 2025 · DOI

TL;DR

This study investigated the feasibility of using electronic patient-reported outcome measures (ePROMs) to collect long-term data from rectal cancer surgery patients. They found ePROMs are feasible and can achieve good response rates, especially when patients are pre-alerted by research staff and have more severe symptoms, suggesting a potential for improved patient engagement and data collection in post-surgical care.

What Problem Does This Paper Address?

Collecting long-term patient-reported outcome measures (PROMs) is crucial for evaluating surgical quality and understanding patient recovery after complex procedures like rectal cancer resection. While electronic PROMs (ePROMs) offer theoretical advantages like efficiency and accessibility, their practical implementation can be hindered by hospital infrastructure and patient-specific factors. This paper addresses the challenge of understanding how feasible and effective an ePROM system is for collecting data on functional outcomes (specifically, the Low Anterior Resection Syndrome score) from patients long after their surgery, and what factors influence their willingness to participate.

Methods

The researchers conducted a retrospective cohort study involving 245 patients who underwent total mesorectal excision. They implemented an ePROM system that sent patients a survey via SMS and email, focusing on the Low Anterior Resection Syndrome (LARS) score. To assess the impact of proactive engagement, a subset of patients (34) received a telephone call from the research team before the survey was sent. For patients who initially didn't respond, the team attempted follow-up contact, including during outpatient appointments, to understand reasons for non-response and, if possible, obtain the LARS score. Statistical analysis, including multivariable logistic regression, was used to identify factors associated with survey responsiveness.

Datasets

The study utilized data from a cohort of 245 patients who underwent total mesorectal excision at a tertiary referral center between January 2015 and January 2022. No new datasets were introduced; the study evaluated the responsiveness of patients to an ePROM system applied to this existing patient population.

Key Insights and Ideas

  • Proactive Engagement is Key: Simply sending an electronic survey isn't enough. A simple telephone pre-alert significantly increased the likelihood of patients responding, highlighting the importance of human interaction in patient engagement.
  • Symptom Severity Matters: Patients experiencing more severe symptoms (indicated by a higher LARS score) were more likely to complete the survey. This suggests that patients who feel they have more to gain from reporting their outcomes are more motivated to participate.
  • Time is a Major Barrier: The most frequently cited reason for non-response was a lack of time, underscoring the need for efficient and user-friendly survey designs.
  • Potential for Selection Bias: While ePROMs are good for identifying patients with severe symptoms, the study cautions that they might introduce selection bias. The patients who respond may not be representative of the entire patient population, particularly those with milder symptoms or less engagement.

Key Results

  • A total of 125 patients (51.0%) completed the ePROM survey, with a median completion time of 4.00 minutes.
  • Three patients (1.2%) encountered technical issues.
  • The most common reason for non-response (83 patients, 33.8%) was a lack of time.
  • Patients who had been involved in previous research, received a survey pre-alert via telephone, and had major LARS symptoms were significantly more likely to respond.
  • Multivariable analysis confirmed that having major LARS symptoms was an independent factor associated with ePROM response (OR 5.51, P=0.001).
  • The multivariable model explained 16% of the variance in survey responsiveness.

Limitations and Future Work

The authors acknowledge that the multivariable model explained only 16% of the variance in survey responsiveness, suggesting that other factors not captured in this study likely contribute. They also note that the LARS score was retrieved from only 27 non-responders, limiting the depth of analysis for this group. The study's retrospective nature and the potential for selection bias are also limitations. Future work should include larger, prospective studies to confirm these findings and explore additional factors influencing ePROM responsiveness. The authors suggest that while ePROMs are valuable for promptly identifying patients with severe symptoms, they should be interpreted with caution when used for collecting prevalence data.

Summary updated

BibTeX
@inproceedings{maroli2025feasibility,
  author = {Annalisa Maroli and Filippo Maria Bianchi and Caterina Foppa and Stefano De Zanet and Fabio Zangrandi and Michele Carvello and Carlotta La Raja and Antonino Spinelli},
  title = {{Feasibility and responsiveness to an electronic system to collect long-term patient-reported outcome measures after rectal cancer resection}},
  booktitle = {BJS},
  year = {2025},
  doi = {10.1093/bjsopen/zraf053},
}
Figure