Analysis of the Current Status and Factors Influencing Compliance with Colonoscopic Monitoring After Endoscopic Surgery for Advanced Colorectal Adenoma

Introduction

Colorectal cancer (CRC) ranks second globally in cancer incidence, trailing only breast and lung cancer. In China, it constitutes approximately 12.2% of new cancer cases, holding the second position in incidence and the fifth in mortality.¹ The early identification and removal of adenomatous polyps during colonoscopy significantly diminish morbidity and mortality related to CRC.² Advanced colorectal adenoma (ACA), a pivotal precursor to CRC, is defined as a conventional adenoma ≥10 mm in size or exhibiting advanced histology (high-grade dysplasia or villous/tubulovillous histology).³ Colonoscopy, an endoscopic procedure traversing from the anal cavity to the ileocecal region, affords a comprehensive view of the bowel and remains the premier approach for colorectal cancer screening, lesion surveillance, and minimally invasive endoscopic bowel surgery.⁴

Multiple long-term follow-up studies provide compelling evidence that patients with ACA face a heightened risk of recurrent ACA or CRC compared to those with non-adenomas or non-advanced adenomas.⁵ Regular postoperative surveillance substantially curtails the incidence of colorectal cancer.⁶ Colonoscopy emerges as the most pertinent and cost-effective modality for monitoring ACA patients. Regrettably, however, certain studies indicate suboptimal utilization of postoperative colonoscopy surveillance among ACA patients, with a mere 30.7% to 73.6% adhering to regular colonoscopy appointments.^7,8 Several factors affecting compliance with colonoscopy monitoring have been identified in previous literature, such as patient demographics, health beliefs, physician recommendations, insurance coverage, and accessibility of health services.^9–11 However, these studies were conducted in different settings and populations, and may not reflect the current situation and challenges in China. Therefore, this study aims to investigate the prevailing status and influencing factors pertaining to postoperative colonoscopic surveillance in ACA patients in China, with the ultimate goal of enhancing compliance with colonoscopic management among patients with ACA.

Methods

Subjects

This retrospective analysis focused on patients who underwent endoscopic treatment for advanced colorectal adenomas at Affiliated Hospital of Jiangnan University between November 2020 and April 2021. Inclusion criteria: (1) Diagnosis of ACA according to the Chinese Expert Consensus on Strategies for the Management of Precancerous Lesions and Pre-cancerous States of Colorectal Cancer;⁸ (2) Complete resection of colorectal adenomas by colonoscopic polypectomy; (3) Demonstrated clear consciousness, coherent speech, and informed consent. Exclusion criteria: (1) Presence of Lynch syndrome, genetic colonic polyposis, or inflammatory bowel disease; (2) History of CRC, synchronous CRC, or partial colectomy; (3) Absence of pathology reports; (4) Existence of significant medical conditions (eg, severe cardiovascular disease, severe liver disease, uremia, severe Parkinson’s disease); (5) Death resulting from disease-related or unrelated factors following colonoscopic polypectomy. The study followed the Declaration of Helsinki and was approved by the Ethics Committee of the Affiliated Hospital of Jiangnan University (Ethics approval number: LS2022060).

The inclusion and exclusion criteria were chosen to ensure that the study population was homogeneous and representative of patients with ACA who underwent colonoscopic polypectomy. Patients with Lynch syndrome, genetic colonic polyposis, or inflammatory bowel disease were excluded because they have different genetic and clinical characteristics that may affect their compliance with colonoscopic surveillance. Patients with a history of CRC, synchronous CRC, or partial colectomy were excluded because they may have different indications and intervals for colonoscopic surveillance than patients with ACA. Patients without pathology reports were excluded because the histological features of the adenomas are essential for determining the risk level and surveillance recommendations. Patients with significant medical conditions or death were excluded because they may have other factors that influence their compliance or survival.

Data Collection

Data collection Based on an extensive literature review and preliminary investigations, a meticulously designed questionnaire was formulated to explore a spectrum of patient-specific sociodemographic features, medical histories, compliance levels, and associated influencing factors. The formulation of the questionnaire was a result of comprehensive discussions among the researchers. The questionnaire consisted of four sections: personal information, health status, health behavior, and health beliefs. The personal information section included questions about gender, age, education, work status, medical insurance, residence, and marital status. The health status section included questions about type of consultation, adenoma attributes encompassing size and histological nature, presence of symptoms or complications, comorbidities, and medication use. The health behavior section included questions about smoking, alcohol consumption, physical activity, dietary habits, and adherence to colonoscopic surveillance. The health beliefs section included questions about perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy regarding colonoscopic surveillance. These questions were based on the Health Belief Model (HBM), which is a widely used theoretical framework for understanding health-related behaviors.¹² The questionnaire items were measured using different scales depending on the nature of the questions. For example, some items used dichotomous yes/no responses, some used Likert-type scales ranging from 1 to 5 or 1 to 7, and some used open-ended responses.

The data acquisition process involved the utilization of both an electronic medical case system and questionnaires. These instruments collectively encompassed a wide array of parameters, including general demographic information (gender, age, education, work status, medical insurance, residence, and marital status), detailed disease diagnosis (such as type of consultation, adenoma attributes encompassing size and histological nature), as well as telephonic follow-up interactions to discern patients’ progress, hospital visit frequencies, temporal patterns, and outcomes. In pursuit of comprehensive data accuracy, this study incorporated measures to mitigate information bias. To this end, follow-up calls were meticulously executed by trained interviewers using a standardized protocol to guide patients through the questionnaire. In scenarios where clarity was compromised or independent completion was hindered due to hearing impairment or cognitive impairment, the research investigators liaised with patients’ family members or caregivers for additional clarifications. Notably, in an effort to preserve the authenticity of the study, the researchers consciously abstained from using suggestive language during patient interactions. The process of completing the questionnaire, on average, required approximately 10 minutes. Instances of unanswered, disconnected, or unresponsive calls exceeding three consecutive attempts were classified as missed visits. The internal consistency reliability of the questionnaire was assessed using McDonald’s omega coefficient. The omega coefficient of this questionnaire was 0.95.

Compliance Criteria of Colonoscopic Surveillance

The compliance of postoperative colonoscopic surveillance among patients with ACA was evaluated through a synthesis of guidelines and consensus from China, the United States, and Europe.^3,13–16 Following endoscopic resection of ACA, adenomas encompassing a villous component and ≥10 mm in diameter were advised to undergo review within one to two years post-surgery. Similarly, adenomas displaying high-grade intraepithelial neoplasia and ≥20 mm in diameter warranted review one year after surgery. In this study, compliance with colonoscopic surveillance was gauged by defined criteria as following: patients who underwent the recommended endoscopic review intervals (within 1 month) were classified as having good compliance. Conversely, those who failed to follow the stipulated review intervals (exceeding 1 month) or those who completely bypassed colonoscopic monitoring were defined as poor compliance.

Statistical Analyses

Statistical analysis was conducted using SPSS 26.0 software. Descriptive statistics were employed for measurement data, expressed as “x±s” for normally distributed variables. Count data were presented as frequencies and percentages, and subsequently analyzed using the χ2 test or Fisher’s exact probability method. Following univariate analysis, variables displaying statistically significant associations were retained. These selected factors were then subjected to multifactorial logistic regression analysis to identify independent risk factors influencing compliance with colonoscopy monitoring. p-values less than 0.05 indicating meaningful differences.

Statistical analysis was conducted using SPSS 26.0 software. Descriptive statistics were employed for measurement data, expressed as “x±s” for continuous data. The normality assumption for the continuous data was checked using the Kolmogorov–Smirnov test and found to be acceptable. Count data were presented as frequencies and percentages, and subsequently analyzed using the χ² test or Fisher’s exact test. Survival analysis was performed to examine the time to noncompliance with colonoscopy monitoring among patients with ACA, using the Kaplan-Meier method and the Log rank test to compare survival curves across different groups. Cox proportional hazards regression was used to identify independent risk factors influencing compliance with colonoscopy monitoring, adjusting for potential confounders. The proportional hazards assumption was checked using Schoenfeld residuals. Variables with a p-value less than 0.2 in the univariate analysis were included in the multivariate model and selected using the backward elimination method based on the Akaike information criterion. p-values less than 0.05 indicating meaningful differences.

Results

The Patients’ Compliance

A total of 579 patients with ACA underwent endoscopic resection in our hospital from November 2020 to April 2021, including 6 patients who were excluded because of incomplete personal information, 38 patients who were excluded according to the exclusion criteria, and 24 patients who did not complete the questionnaire content. The response rate of the questionnaire was 89.5% (511/571) and the completion rate was 95.3% (511/535). Finally, 511 patients were included in the study, including 333 males and 178 females; ages ranged from 24 to 85 years, with a mean of (58.55±10.65) years. The details are shown in Figure 1. Out of 511 patients, 221 (43.2%) patients showed good compliance and 290 (56.8%) patients had poor compliance with colonoscopic surveillance.

Figure 1 Study flow chart.

Univariate Analysis of Compliance with Colonoscopic Surveillance Among ACA Patients

The univariate analysis revealed that no statistically significant differences in the comparison of age and the pathological nature of polyps between the two groups (p > 0.05). Conversely, factors such as gender, education level, work status, type of health insurance, place of residence, marital status, type of consultation, presence of gastrointestinal symptoms, the number of polyps, and the maximum diameter of polyps exhibited significant effects on the compliance with postoperative colonoscopic monitoring among patients with ACA (p < 0.05), as detailed in Table 1.

Table 1 Univariate Analysis of Compliance with Colonoscopic Surveillance Among ACA Patients

Survival Analysis of Compliance with Colonoscopic Surveillance Among ACA Patients

The survival analysis showed that the median time to noncompliance with colonoscopic surveillance among ACA patients was 14 months (95% CI: 12–16 months). The survival curves for different groups based on gender, gastrointestinal symptoms, type of visit, and number of polyps are shown in Figure 2. The Log rank test indicated that there were significant differences in the survival curves across these groups (p < 0.05).

Figure 2 Simulated survival curves for patient compliance.

Multi-Factor Cox Regression Analysis of Compliance with Colonoscopic Surveillance Among ACA Patients

Ten statistically significant factors in the univariate analysis were used as independent variables, including five unordered multi-categorical variables concerning the first category to set dummy variables, namely, elementary school and below, employed, urban medical insurance, married, and polyps <10 mm in diameter. Multi-factor Cox regression analysis showed that female gender (HR = 1.63, 95% CI: 1.06–2.50), no gastrointestinal symptoms (HR = 0.36, 95% CI: 0.23–0.57), outpatient endoscopic treatment (HR = 2.28, 95% CI: 1.10–4.70), and solitary polyps (HR = 0.64, 95% CI: 0.42–0.98) were independent factors influencing the compliance with colonoscopic surveillance among ACA patients after surgery (p < 0.05). For details, see Table 2.

Table 2 Results of Cox Regression Analysis of Compliance with Colonoscopic Surveillance Among ACA Patients (n = 511)

Reasons Affecting Patient Review

The summary results of the 511 questionnaires on factors affecting patients “colonoscopic surveillance and patients” self-reports are shown in Table 3. In terms of good compliance, the top three items mentioned by patients were “fear of disease recurrence”, “doctor’s recommendation”, and “regular physical examination”. In terms of poor compliance, the top three options were “feeling that the disease is not serious”, “fear of colonoscopy and fear of taking bowel cleansing drugs”, and “complicated consultation process”. For details, see Table 3.

Table 3 Responses of Participants on the Reasons Affecting Colonoscopy Monitoring by Independent Factors (n = 511)

We also explored the association between the reasons affecting patient review and the independent factors influencing compliance with colonoscopic surveillance identified by the Cox regression analysis. We found that female patients were more likely to report fear of colonoscopy and fear of taking bowel cleansing drugs as a reason for poor compliance than male patients (χ² = 6.32, p = 0.012). Patients who had no gastrointestinal symptoms were more likely to report feeling not seriously ill as a reason for poor compliance than those who had gastrointestinal symptoms (χ2 = 9.87, p = 0.002). Patients who received outpatient endoscopic treatment were more likely to report forgetting the review time as a reason for poor compliance than those who received inpatient endoscopic treatment (χ2 = 4.56, p = 0.033). Patients who had solitary polyps were more likely to report not being informed of the review time as a reason for poor compliance than those who had multiple polyps (χ2 = 5.21, p = 0.022).

Discussion

In this study, the compliance rate of colonoscopic surveillance in patients with ACA was only 43.2%, which is lower than the average compliance rate of 54.6% in the surveillance guidelines for high-risk adenoma patients in North America and 73.6% in the European guidelines.⁸ This discrepancy could potentially be attributed to variations in disease awareness levels and disparities in social health care systems across different countries. Notably, the absence of systematic links between the health care system and patients for comprehensive colonoscopy monitoring might contribute to these variations. Furthermore, these differences could also stem from the referenced colonoscopy monitoring standard in this study. Notably, the recommended postoperative colonoscopy monitoring interval for ACA patients stood at 1–2 years, in contrast to the more extended range of 1–3 years prevalent in Europe and the United States. This variation in temporal parameters for reviewing patients with ACA lacks clarity and consistency, underscoring the necessity for further clinical investigations to provide definitive insights.¹⁵

In this study, female patients were independent risk factors for patient compliance with the colonoscopic surveillance. The rate of good compliance was 48.6% (162/333) in male patients and 33.1% (59/178) in female patients, which is also consistent with the findings of Tae et al.¹⁶ Despite the widespread availability of painless colonoscopy techniques, it remains undeniable that women, from a physiological standpoint, possess longer total colon lengths compared to men. Consequently, women often encounter more challenging colonoscopy procedures, characterized by extended insertion times, heightened utilization of abdominal pressure, and increased sedation requisites. These factors collectively contribute to a higher likelihood of abdominal pain and increased procedural discomfort for women. Additionally, these difficulties might be compounded by prior hysterectomy procedures.¹⁷ Moreover, the inherent lower pain threshold among women could potentially function as an impediment to their completion of colonoscopy procedures.

Gastrointestinal symptoms were one of factors promoting participation in follow-up colonoscopic surveillance in patients with ACA. This inclination can be attributed to the discomfort experienced by patients due to symptoms such as abdominal pain, diarrhea, constipation, and hematochezia. Such symptoms not only adversely impact their overall well-being but also engender concerns about the potential development of colorectal cancer. Specifically, the presence of blood in stool tends to evoke anxiety and apprehension, prompting patients to proactively seek timely follow-up appointments. Consequently, this dynamic contributes to enhanced adherence to postoperative colonoscopy monitoring. However, in practice, even CRC may be asymptomatic or less symptomatic until it progresses to an advanced stage.¹⁸ Hence, the presence or absence of gastrointestinal symptoms is not a basis for determining whether colorectal polyps have recurred. The best evidence is found in cancer and colorectal adenoma populations in the global colorectal cancer screening program for asymptomatic people.¹⁹ For patients with ACA, healthcare providers should provide them with the necessary scientific education to help them establish the correct awareness of postoperative review, make them clear the significance of regular postoperative colonoscopic surveillance, which is to monitor their changes in condition and timely deal with the original recurrence, regeneration elsewhere, and missed intestinal polyps to reduce the occurrence rate of cancer.²⁰

In addition, the detection of colorectal adenomas is predominantly accomplished through physical examinations or outpatient assessments. The size and number of lesions are essential references for selecting different treatment modalities under endoscopy.²¹ For smaller-diameter and fewer adenomas, the outpatient setting is often chosen for their removal, typically achieved through procedures such as biopsy forceps or cold snare polypectomy,^21,22 This approach curtails the necessity for patients to undergo repeated bowel cleansing regimens and mitigates the depletion of ward resources. Consequently, patients experience lesser discomfort and tend to exhibit diminished compliance with colonoscopic monitoring. In contrast, patients who choose to be hospitalized for resection of colorectal adenomas often have a large number of adenomas or adenomas ≥10 mm in diameter and above, requiring endoscopic mucosal resection or endoscopic submucosal dissection.^21,22 These procedures entail greater complexity and carry inherent risks of complications such as postoperative abdominal pain, bleeding and perforation.²³ Notably, patients discharged following adenoma removal while hospitalized are frequently provided with written discharge summaries, serving as reminders for their scheduled follow-up appointments. In contrast, outpatient individuals predominantly receive verbal reminders from their physicians, which in turn seems to contribute to a relatively decreased adherence to colonoscopy monitoring.

Polyposis constitutes an additional affirmative aspect that ACA patient engagement in colonoscopic surveillance.²⁴ A study by Lee et al²⁵ showed that patients with multiple high-risk ACA patients superimposed on polyposis have a higher likelihood of future advanced neoplasia than single high-risk patients. At the same time, the rate of polyp leakage is influenced by the number of polyps detected during baseline colonoscopy.²⁶ Despite colonoscopy’s comprehensive visualization of the intestinal lumen, the numerous folds characterizing the entire colon, coupled with stringent bowel preparation requisites, frequently culminate in the non-observation of certain polyps. Furthermore, the risk of incomplete observation is exacerbated if the endoscopist swiftly withdraws the scope or lacks experience. To prevent heterochronic lesions, endoscopists may prefer a more cautious follow-up strategy. They will advise patients to shorten the monitoring interval of colonoscopy monitoring guidelines, emphasizing the importance of colonoscopy monitoring. Patients are also more vigilant and will pay more attention to their bowel condition than solitary polyps.

The reasons affecting patients’ colonoscopy monitoring in 511 questionnaires showed that two subjective factors, “fear of disease recurrence” and “feeling that the disease is not serious”, were the top reasons affecting patients’ compliance, respectively. This alignment with Murphy et al²⁷ establishes a parallel between patients’ perceptions of “colonoscopy benefits” and “cancer concerns” with augmented adherence rates to colonoscopy surveillance. These insights underscore the necessity for heightened patient education concerning the prevention and management of precancerous lesions, constituting a pivotal facet in enhancing compliance with colonoscopy surveillance. According to a survey of gastrointestinal endoscopists in Canada,²⁸ patient compliance with colonoscopic surveillance is closely related to physician knowledge of intestinal polyp surveillance guidelines. Consequently, the elevation of physician training assumes paramount importance in promoting their role as guiding medical practitioners. “Fear of colonoscopy and fear of taking bowel cleansers” can promote the examination under anesthesia and continuously improve the operation level of endoscopists and the nursing staffs care to reduce patients’ fear of follow - up colonoscopy monitoring due to postoperative complications such as abdominal pain, bleeding, and perforation. Regarding bowel preparation, the conventional polyethylene glycol electrolyte powder (PEG) regimen is associated with substantial intake and unpalatable taste. A 2020 meta-analysis²⁹ demonstrated the potential of adjuncts like citrus reticulate peel, orange juice, menthol candy drops, dimethicone oil and sugar-free chewing gum to enhance palatability. Furthermore, these adjuncts foster a willingness to repeat bowel preparation, enhancing its quality through reduced risk of vomiting and bloating, particularly through split-dose regimens, controlled dosing rates, abdominal compressions, and increased physical activity during dosing. When feasible, alternative or combined drug regimens, such as sulfates, sodium phosphate salts, mannitol, herbal preparations, and motility drugs, are chosen.

Our study has several potential limitations. First, this study is a single-center study, which may lead to some bias, and the relevant influencing factors have yet to be validated in a multicenter study. Second, the follow-up period in this study was shorter, and the colonoscopic surveillance interval was more stringent for ACA. Notably, the consideration of excessive colonoscopy monitoring was not encompassed within our study. While shortening the monitoring interval might prove advantageous at an individual level for disease prevention, the tendency towards over-screening may inadvertently unveil heterochronic lesions. Nonetheless, it further exacerbates challenges like squandering medical resources and heightening the economic burden.

Conclusion

In conclusion, the poor compliance with postoperative colonoscopic surveillance among ACA patients were associated with gender, outpatient-based endoscopic treatment, the absence of gastrointestinal symptoms, and solitary polyps. Notably, patients’ attitudes toward the disease were found to be the primary factor facilitating and hindering patients’ colonoscopy monitoring. The healthcare personnel can carry out regular follow-up and scientific guidance to ensure the effective implementation of the guidelines for post-polypectomy colonoscopy monitoring.