Evaluation of eating habits and practice of physical activity in colorectal cancer patients in the active phase of treatment

Colorectal cancer is one of the most prevalent in men and women, and its development is associated with several risk factors, such as a sedentary lifestyle, obesity, alcohol consumption, and eating habits. This study aimed to present the results of applying the questionnaire on eating habits and physical activity from the Food Guide: how to have a healthy diet in the first chemotherapy session, after signing the consent form, and eight weeks after the first application. All patients with colorectal cancer under active treatment from July to December 2022 were invited to participate in the study. The Ethics Committee approved this project. During the research, patients underwent a new monitoring model that included the use of a chatbot and wearable device for eight weeks. The model aimed to ensure greater patient engagement with their treatment through interactions about symptoms and adverse effects, eating habits, and physical activity practice. As a result, there was a statistical reduction in the consumption of alcoholic beverages and fast foods and a statistical increase in fruit consumption. In addition, the findings indicate that the model contributed to a greater practice of physical activity by patients. For future work, we suggest applying this model to patients with other types of cancer to assess and extend the benefits of the model in different contexts of cancer treatment.


Introduction
Colorectal cancer (CRC) is the third most common cancer and the fourth most common cause of cancer-related death.
The risk of developing CRC is associated with personal characteristics such as age, gender, history of chronic diseases, and lifestyle factors such as consumption of carbohydrate-rich and lipid-rich foods and low intake of fruits, vegetables, and fiberrich foods (Mármol et al., 2017).This disease is influenced by several factors, including age (being ≥50 years one of the main risk factors), overweight, obesity, and unhealthy dietary habits such as low intake of fruits, vegetables, and fiber-rich foods, as high consumption of processed and red meats (Queiroz, Carneiro, et al., 2022), (Wilde et al., 2019).Furthermore, genetic factors, exposure to ionizing radiation, family history, previous presence of intestinal, ovarian, uterine, or breast cancer, smoking, and excessive alcohol consumption also play a significant role in the growth and development of this tumor (Calle & Kaaks, 2004).
There are protective factors that are associated with a lower incidence of CRC, such as regular physical activity and a diet rich in fruits, vegetables, fiber, folate, calcium, dairy products, vitamin D, and vitamin B6, as well as intake of magnesium, fish, and garlic (Thanikachalam & Khan, 2019).In addition, physical activity plays an essential role as a protective factor, as it reduces body fat and therefore has a beneficial effect on the risk of CRC, possibly by decreasing insulin resistance and inflammation -both associated with the development of CRC.Other mechanisms by which physical activity may reduce the risk of CRC include stimulating digestion and reducing transit time through the intestine, although robust data to support this mechanism in humans are limited (AICR & WCRF, 2018;WCRF, 2023).
Given current technological advances, chatbots have the potential to play an increasingly important role in health and medical care, assisting physicians during a consultation, supporting patients in changing behavior, or helping in their own homes.Patients can, for example, receive advice and information provided by chatbots and describe their current symptoms, even if they are away from the medical care center (Wolters et al., 2016).Similarly, patients can use chatbots to check symptoms and monitor their mental health (Hoermann et al., 2017;Miner et al., 2016).Ly et al. (Ly et al., 2017) assessed the efficacy and compliance of a smartphone application designed to teach strategies from optimistic psychology aimed at enhancing happiness and decreasing negative symptoms.
IoT has emerged as a promising tool in healthcare and has the potential to reduce costs, improve user experience and patient monitoring, and increase the quality of life (Costa et al., 2018), (Gralha et al., 2022).Personalized interventions combined with IoT have contributed to improving the quality of life and mitigating symptoms and adverse effects in cancer patients undergoing treatment (Queiroz et al., 2021).The use of wearable devices demonstrated that the better clinical condition and quality of sleep of the patient were associated with improvement in the practice of physical activity (Chung et al., 2019), (Nyrop et al., 2018), (Queiroz et al., 2021).A higher number of steps was associated with reduced chemotherapy-related symptoms such as fatigue, anxiety, and distress data (Chung et al., 2019), (Nyrop et al., 2018).
Thus, this study aimed to develop and evaluate a tool for monitoring patients with CRC during the active phase of treatment using the Internet of Things and Artificial Intelligence techniques.The model aimed to ensure greater patient engagement with their treatment through interactions about symptoms and adverse effects, eating habits, and physical activity practice.This study presents the results of applying a questionnaire on eating habits and physical activity from the Food Guide: how to have a healthy diet -Ministry of Health (Brazil, 2013) before and after the intervention and is part of an ongoing broader study.

Study type
A prospective clinical study aims to evaluate the benefits of the SMT (Smart Monitoring Tool) model to patients regarding eating habits and physical activity practice.The study population was CRC patients undergoing active treatment at the Sinop Cancer Center (Cecans) in Sinop, Mato Grosso, Brazil.Data were collected from July 2022 to December 2022 at Cecans.This study is part of a broader study that evaluated the benefits of the model by comparing patients who used the SMT model with patients who followed the traditional model adopted by the clinic.The questionnaire was applied in person in the field during chemotherapy sessions and data collection was based on a quantitative approach (Pereira et al., 2018).Patients were instructed to choose one of the available options in each question, when available, inform the amount of consumption or answer yes/no.

Inclusion and exclusion criteria
Inclusion criteria were CRC patients; aged over 18 years; under active cancer treatment; staging from I to IV; Patients must have smartphone skills.Exclusion criteria were patients with restrictions to exercise due to severe cardiovascular, pulmonary, or renal diseases, have cognitive impairment that prevents using smartphones or are unable to give verbal consent.

Intervention
Patients who met the inclusion criteria were invited to participate in the research during the 1 st chemotherapy session.
Patients who agreed to participate signed the Informed Consent Form.Patients were asked to answer a questionnaire containing 18 questions about eating habits and physical activity from the Food Guide: how to have a healthy diet -Ministry of Health (Brazil, 2013).The questionnaire was applied in the 1st chemotherapy session and eight weeks after the first application.In the first application, patients were instructed to self-report their eating habits and physical activity before the cancer diagnosis.And in the second application, patients were asked to self-report their conduct after participating in the research.

Data analysis
The influence of the SMT model on the patient's eating habits and physical activity was evaluated.Statistical analysis: data were presented as mean ± standard deviation (SD) or in percentage (%).The results were statistically evaluated by the Student's  test, the Wilcoxon test, or the chi-square test ( 2 ) using the GraphPad Prism 7 Program.The minimum acceptable significance level was  < 0.05.

Ethical aspects
Ethical aspects are considered following Resolution 466/12 of the Ministry of Health, which establishes ethical standards governing research involving human beings.The execution of the project was authorized by Cecans and by the physician responsible for the Oncology Wing and approved by the Ethics Committee in Research with Human Beings (CAAE 48258421.7.0000.5344).

Results and Discussion
In total, 19 patients were invited to the intervention group.Unfortunately, four patients did not accept to participate because they did not have technological skills or were not interested in participating, and two patients withdrew during the research due to complications resulting from the treatment.In addition, the mean age of these patients was 58.2 years old, and most patients (83.3%) were over 50 years old.Finally, 13 patients completed the study in the intervention group.All patients had private health insurance or underwent treatment privately.

Clinical and epidemiological profile of intervention group of CRC patients
According to the epidemiological data, body weight mean, and height median were 73.8 kg and 1.68m, respectively.
The body mass index mean was 26.1 kg/m², with 5 (38.5%) patients with a BMI between 20kg/m² and 25kg/m², considered regular/eutrophic, and 8 (61.5%) patients over 25kg/m², classified as overweight/obese.The data corroborate the literature, showing that overweight/obese patients are at greater risk of developing CRC (Loosen et al., 2022), (INCA, 2022).The oximetry means, and temperature median were 97.0 % and 36.4 º C, respectively.Regarding blood pressure, the mean systolic blood pressure was 123.8 (SD 16.1) mmHg, and the median diastolic was 80 (80 -80) mmHg.All these data were within the reference parameters considered normal.
According to the clinical data, most patients were diagnosed with stage 3 (69.2%),had a family history (69.2%), and had surgery and chemotherapy as indicated type of treatment (76.9%), with adjuvant chemotherapy (69.2%).The mFolfox6 protocol (69.2%) was the most recommended.All patients presented adenocarcinoma-type CRC (100.0%).Most had affected lymph nodes (76.9%), were diagnosed with primary cancer (92.3%), and did not have metastasis (92.3%).Studies have shown that, due to the characteristics of CRC, patients typically discover the disease at a more advanced stage, consequently, the staging will be greater, and the clinical prognosis will be more critical (Queiroz, Alegranci, et al., 2022;Schult et al., 2021).Screening through colonoscopy is one of the methods that can contribute to discovering this type of cancer at an early stage (Bai et al., 2022).Family history is also one of the risk factors.Individuals with first-degree relatives diagnosed with CRC have a two to four times higher risk than the general population (Bai et al., 2022).

Evaluation of eating habits and practice of physical activity
The meals most consumed by patients before the intervention were breakfast (84.6%), lunch (92.3%), and dinner (100.0%), as seen in Table 1.Moreover, a minority of patients self-reported consuming morning snacks (38.5%) and afternoon snacks (46.2%), and supper (7.7%).After the intervention, it was observed, based on the patient's self-report, higher consumption of breakfast (92.3%), morning snack (46.2%), afternoon snack (69.2%), and lunch (100%).Studies report that one of the symptoms of cancer and the adverse effects of treatment with antineoplastic drugs is decreased food intake and lack of appetite, which can lead to anorexia (Lazarus & Bays, 2022), (Ferioli et al., 2018).However, our results indicate that patients improve their eating habits by eating more frequently during the day.The amount of water consumed before or after the intervention was very similar.No statistical differences were identified.However, a statistical decrease in alcohol consumption (p=0.0472) was observed.Before the intervention, most patients self-reported alcohol consumption (76.9%), and after the intervention, only 30.8% reported alcohol consumption.The diagnosis, medical advice, and participation in the study may have contributed to this relevant reduction in consumption.Several studies have shown an association between alcohol consumption and CRC incidence (Thanikachalam & Khan, 2019), (Ghazaleh Dashti et al., 2017).
Table 1 -Analysis of the regular consumption of each type of meal, the daily number of glasses of water, and the frequency of alcoholic beverages.Data were extracted before and after the intervention.Table 3 shows that, before and after the intervention, most patients self-reported consuming pasta, carbohydrates, cereals (before and after: 100%), and bread (before and after: 76.9%).Furthermore, before the intervention, 38.5% reported rarely consuming sweets and sugary drinks, and 23.1% reported rarely consuming fast food, sausages, and fried food.However, after the intervention, most patients reported rarely consuming fast food, sausages, and fried food (61.5%).This is essential data since consuming processed meats, such as hamburgers, sausages, and bacon, contributes to increased cancer risk (Wilde et al., 2019).
Table 3 -Analysis of pasta, carbohydrates, cereals, industrialized beverages, fast food, and sweets consumption by patients before and after the intervention.Results are expressed as median (interquartile of 25% and 75%) or in number of individuals and percentage (n (%)).Statistical analysis: chisquare test (X 2 ), # Fisher test, and & Wilcoxon test.Source: elaborated by the authors.

Variable
Most patients self-reported consuming meat or eggs (before: 100.0%; after: 92.3%), the most common daily consumption being two pieces or eggs (before: 61.5%; after: 53.8%, p=0.5866), as seen in Table 4. Before the intervention, 5 (38.5%) patients self-reported that they usually removed visible fat from the meat.After the intervention, there were 10 (76.9%) patients (p=0.1107).The frequency of fish consumption remained similar before and after the intervention, but there was an increase in the consumption of milk and dairy products (before: 61.5%; after: 84.6%; p=0.3783).Whole milk was the most common type of milk consumed by patients (before: 53.8%; after: 69.2%; p=0.3928).Consumption of meat, fish, dairy products, and milk remained similar before and after the intervention.The habit of removing visible fat from meat was the only change in animal protein consumption, most patients self-reported that they had this behavior after the intervention.
Red meat consumption is one of the main risk factors for CRC (Wilde et al., 2019), and the consumption profile of patients before diagnosis may have contributed to the development of cancer.Participation in the intervention did not change the amount of meat consumed.Results are expressed in number of individuals and percentage (n (%)).Statistical analysis: chi-square test (X 2 ), # Fisher test.Source: Elaborated by the authors.
Table 5 shows the types of activity and the frequency performed by the patients before and after the intervention.Before the intervention, 5 (38.5%) patients self-reported practicing physical activity, with two reporting practicing walking and five reporting practicing other physical activities, such as running, bodybuilding, beach tennis, and pilates.After the intervention, 7 (53.8%) patients reported practicing physical activity, with walking being the most common activity (6/13; 46.1%).As expected, walking was the most practiced physical activity after the intervention since the physicians at the clinic advised all patients to prioritize walking, light, and low-impact physical exercise.As a positive result, we had an increase in the practice of physical activity by patients, even during the active phase of treatment, although no statistical difference was observed.This result suggests that the model stimulated the practice of physical activity by the patients.However, during treatment, patients usually experience various adverse effects due to the antineoplastic drugs administered, which typically contribute to decreased physical exercise.Table 6 presents the general score of the questionnaire self-reported by the patients before and after the intervention.
From 0 to 28 points, the standard feedback in the questionnaire is "You need to make your diet and life habits healthier!".From 29 to 42, the feedback is "Be careful with your diet and other habits such as physical activity and fluid consumption.".Finally, 43 points or more, "Congratulations!You're on the way to healthy living.".Patients' scores indicate an exciting improvement in eating habits before and after the intervention, corroborating that the use of the new monitoring model may have contributed to a better quality of life for the patient.

Conclusion
In conclusion, it was observed that the model may have contributed to a change in the patient's behavior.Most patients consumed carbohydrates, fast food, red meat, and alcohol, and few practiced physical activities.The results suggest that the intervention through the application of the model contributed to an increase in the consumption of fruits and the practice of physical activity, in addition to helping to reduce the intake of alcoholic beverages and the consumption of fast food.For future work, we suggest applying this model to patients with other types of cancer to assess and extend the benefits of the model in different contexts of cancer treatment.

Table 2 -
Analysis of patients' behavior before and after the intervention regarding consuming fruits, vegetables, and grains.
(Hurtado- Barroso et al., 2020)application of the questionnaire during the intervention may have contributed to the improvement in fruit consumption, as patients were encouraged to reflect on their current behavior.Fruit consumption is a crucial factor in primary cancer prevention.Consumption has been associated with a lower mortality rate for cancer survivors(Hurtado- Barroso et al., 2020).Most reported eating two or more fruits a day (61.6%),eatingvegetables (84.6%), and grains (92.3%).Results are expressed in number of individuals and percentage (n (%)).Statistical analysis: chi-square test (X 2 ) and # Fisher test.*p<0.05.Source: Elaborated by the authors.

Table 4 -
Analysis of animal protein (meat, fish, milk, and dairy products) consumption by patients before and after the intervention.

Table 5 -
Distribution of the type and frequency of physical activity the patient performed before and after diagnosis.Results are expressed in number of individuals and percentage (n (%)).Statistical analysis: chi-square test and # Fisher test.Source: Elaborated by the authors.

Table 6 -
Overall score of the Food Guide: how to have a healthy diet questionnaire.