Colon cancer is the third most common cancer worldwide [2]. Diagnosis and subsequent treatment decisions are usually dependent on “hard” clinical factors such as tumor stage or comorbidities of the patient [3]. There is an increasing demand to include self-reported symptoms and functional limitations not only as an outcome after treatment, but also before and during therapy [10]. The tools of choice to measure such symptoms and functional limitations are patient-reported outcomes (PRO). These are standardized questionnaires that patients answer themselves and thus provide information about patients’ own assessment of their health status and the symptoms and functional limitations they encounter. In oncology, the European Organisation for Research and Treatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30) is widely used [1], which can be combined with a disease-specific module; for colorectal cancer, this is EORTC QLQ-CR29 [12]. Pioneering work by Bash et al. was able to show that early inclusion of PROs into treatment planning and monitoring leads to a statistically significant increase in life expectancy [4]. Nevertheless, patient-reported outcomes are still very rarely used in colorectal oncologic care [9]. This seems to partly be due to the fact that the questionnaires are very long and thus perceived as impractical for clinical routine (taking the generic module together with the colorectal cancer-specific form, a total of 35 different symptoms and functional scores are captured by the EORTC QLQ questionnaires) [5]. Therefore, it seems to be of high clinical relevance to select those scores that are particularly important from the point of view of the treating clinicians in the pretherapeutic assessment of colorectal cancer patients.

Materials and methods

The CHERRIES framework for reporting online surveys was used for this study [7].

Design, participants, and recruitment process

Coordinators and directors of certified colorectal cancer centers which participated in a PRO study using the EOTRC QLQ-C30 and CR29 questionnaires (EDIUM centersFootnote 1) were asked to participate in a closed online survey. The survey was hosted on the platform sosci-survey.de between February and March 2021.

All participants gave their informed consent based on information regarding the time required for the survey (approximately 5 min), contact information to study group, and the pseudonymization process before accessing the online survey.

Directors and coordinators of all EDIUM centers were contacted via mail with information about the survey and study purposes. If they were willing to participate, they could access a personalized link. All eligible directors and coordinators were reminded biweekly three times. There were no additional incentives. The study group set up an internal trust office led by SS for contacting and reminding the eligible directors and coordinators. SS did not take part in any data analysis for pseudonymization purposes.

The survey was part of the evaluation process of the EDIUM study and as such has an ethical approval by the Ethics Committees of the Berlin Chamber of Physicians (Eth-19/18).

Questionnaire

The questionnaire was developed by NTS. A first pre-test was performed by CB and CK, both investigators of the EDIUM study group. The questionnaire was then pre-tested by two clinicians of the independent scientific board of the EDIUM study. Any ambiguities relating to the survey questions were revised afterwards and reworded by NTS, if necessary. The final questionnaire consisted of three main questions followed by two optional fields for any additional comments on the survey or the EDIUM study in general.

The first two questions were (translated) “Which of the following dimensions (disease-specific symptoms and function) of the EORTC questionnaires QLQ-C30 and CR29 are most relevant regarding clinical assessment, treatment preparation, and discussion before a definitive treatment (e.g., surgery) of COLON carcinoma?”Footnote 2 and “Which of the following dimensions (disease-specific symptoms and function) of the EORTC questionnaires QLQ-C30 and CR29 are most relevant regarding clinical assessment, treatment preparation, and discussion before a definitive treatment (e.g., surgery) of RECTAL carcinoma?”Footnote 3. Both main questions had 35 response options—the number of EORTC QLQ-C30 and CR29 dimensions—and the participants were asked to choose the five most relevant dimensions for them. Afterwards, the participants were to indicate which specialty (internal medicine, surgery, other) they work in.

The survey offered completeness checks and reminded the participants if any question was not yet answered. However, completeness was not compulsory for submitting the survey. After submitting, the respondents were not able to re-submit the questionnaire or change their responses.

Data analysis

Data were descriptively analyzed using R version 4.0.2 (2020.06.22; R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/). A summarizing, qualitative analysis of the two “further comments” sections (text fields) was also performed.

Results

Participants and response rates

A total of 203 directors and coordinators of 103 EDIUM centers were asked to participate in the survey. After three biweekly reminders, 96 of the contacted persons submitted the questionnaire, resulting in a response rate of 45.8%. Most of the respondents were surgeons (83 in total, 86%). The frequencies of the different specialities can be found in Table 1.

Table 1 Frequencies of specialities
Full size table

On average, respondents needed 3.6 min (standard deviation 1.9 min) to answer the survey.

Relevant EORTC QLQ-C30 and CR29 items

Colon cancer

For colon carcinoma, the “quality of life” score was most often chosen to be one of the five most relevant scores (70 of 83 surgeons, 8 of 9 specialists for internal medicine). The other four most frequently chosen scores were: “pain” (36 surgeons, 2 specialists for internal medicine), “physical function” (30 surgeons, 5 specialists for internal medicine), “constipation” (30 surgeons, 1 specialist for internal medicine), and “abdominal pain” (27 surgeons, 4 specialists for internal medicine). The scores “dry mouth,” “dysuria,” “hair loss,” “trouble with taste,” and “urinary frequency” were not chosen by any participants. For the exact frequencies, compare Fig. 1 and the supplementary Appendix.

Fig. 1
figure 1

Most relevant EORTC QLQ-C30 and CR29 scores for colon cancer (arranged by surgeons’ responses)

Full size image

Rectal cancer

For rectal cancer, the “quality of life” score was most often chosen to be one of the five most relevant scores (65 of 83 surgeons, 7 of 9 specialists for internal medicine). The other four most frequently chosen scores were: “fecal incontinence” (54 surgeons, 6 specialists for internal medicine), “pain” (23 surgeons, 3 specialists for internal medicine), “constipation” (22 surgeons, 1 specialist for internal medicine), and “physical function” (20 surgeons, 4 specialists for internal medicine). The scores “dry mouth,” “hair loss,” “insomnia,” and “trouble with taste” were not chosen by any of the respondents. For the exact frequencies, compare Fig. 2 or the Appendix.

Fig. 2
figure 2

Most relevant EORTC QLQ-C30 and CR29 scores for rectal cancer (arranged by surgeons’ responses)

Full size image

Additional remarks

The participants had the possibility to fill out a text field if they had any additional remarks. Two participants highlighted the importance of the patients’ perception of stomal therapy:

“How stressful do patients find a stoma depending on their age and life situation?” (Specialist for internal medicine)

Another participant doubted that any dimensions of EORTC QLQ-C30 and -CR29 should be regarded on their own, as the overall picture of the patient is most relevant:

“I find the selection problematic. We actually want to have an overall picture. Many of the individual factors listed must be evaluated by the patient and then discussed, whereas others require direct action.” (Surgeon)

Moreover, one participant named “sexual life” as an important additional piece of information about the patients, another “psychiatric comorbidities.”

Discussion

To the authors’ knowledge, this is the first summary and presentation of clinicians’ assessment of the relevance of the EORTC QLQ-C30 and -CR29 scores for treatment planning in colorectal cancer patients. First, it should be emphasized that a general summary score such as the quality of life score is most relevant for both rectal and colon cancer from the clinicians’ perspective. Differences in the localization of these tumors explain why fecal incontinence is mentioned in second place for rectal cancer (seventh place for colon cancer). Dyspareunia and impotence, which are frequently mentioned in rectal cancer, can also be explained by the location of the tumor. Symptoms occurring after cancer treatment (e.g., chemotherapy) were not reported at all or rarely (e.g., “trouble with taste” or “hair loss” in both entities).

Our results underline the importance of standardized recording of symptoms and functions in patients with colorectal cancer. However, based on our survey and other findings, it is recommended to make an appropriate selection of EORTC-QLQ scores in order to not overburden patients or clinicians with unnecessary information [8, 11]. The clinicians’ particularly strong focus on quality of life also underscores this information’s relevance for treatment planning.

Often, however, quality of life is not recorded in a standardized way, but may only be taken into account during the medical history taking. In the interests of equal treatment of all patients and quality assurance in oncologic care, it is advisable to query patients in a structured and thus binding manner before the start of a therapy rather than to rely on an unstructured anamnesis. If collected in a standardized way, information can only not be lost less easily, but it can also be used by the entire treatment team instead of by a single physician who talked to the patient. Here, too, it seems particularly relevant to restrict oneself to the clinically important scores, in order to promote acceptance among clinicians beforehand—even though one participant strengthened the importance of non-restriction as an additional remark.

Moreover, if patients are actively involved during treatment planning and treatment decision-making by contributing information about their disease burden themselves, and thus by being able to control what is particularly important to them for treatment, the communication between clinicians and patients is strengthened. This contributes to an improvement of colorectal cancer treatment [6].

These results may help to select relevant scores to be queried in one’s own care unit. Due to the low level of participating specialists in internal medicine, conclusions can be drawn from the current survey mainly for surgical departments. However, since colorectal cancer patients often first encounter surgeons, it seems reasonable to set a main focus on surgically important PROs. Since the EORTC-QLQ instruments were only introduced in many participating centers during the EDIUM study, it is possible that responders are comparably inexperienced in the use of PROs. Further research may investigate whether responses differ in samples with more PRO experience. Moreover, the authors strongly encourage further research in patient-relevant symptoms and functions, which might be others than those important to clinicians. Nevertheless, this is the first listing and survey of the most relevant symptoms and functions—which can be self-reported by the patient—for the treatment decision in colorectal cancer and, as such, should also be included in the clinical oncology daily routine.