Introduction

Rectal cancer is among the most common type of cancer globally, the trend for age of onset becoming younger [1,2,3]. With the promotion of early screening and advances in medical technology, the mortality resulting from rectal cancer decreases year-on-year [3]. The 5-year survival rate has reached 65% [4]. Therefore, the quality of life and long-term outcomes have received increasing attention as essential aspects of care.

Notable advances in the techniques used in colorectal surgery and rectal anastomosis have safely expanded the frontiers of sphincter preservation, preventing increased numbers of patients suffering from permanent abdominal colostomy. The anterior resection has been widely used in clinical practice and has become the standard treatment for mid and low rectal cancers. However, the benefits of such surgery for preservation of anal sphincter function in recent years have been controversial. A review and meta-analysis from Cochrane found that patients undergoing anterior resection had no higher quality of life indicators than patients who underwent abdominal resection/Hartmann surgery [5]. One larger reason is that 50 to 90% of patients undergoing sphincter-preservation surgery experience varying degrees of postoperative bowel dysfunction, including altered stool consistency, urgency, fecal incontinence, and emptying difficulties, collectively known as low anterior resection syndrome (LARS) [6]. A meta-analysis found that the estimated prevalence of major LARS was 41% (95% CI: 34–48) [7]. Repeated episodes of these symptoms caused the patients physical discomfort, emotional imbalance, and social isolation; some patients complained that they experience “toilet-dependency” [8, 9]. Although patients with LARS showed signs of improvement over a 2-year period, nearly half of the patients reported severe impairment in daily activities at long-term follow-up examination [10].

With the new surgical techniques of total mesorectal excision (TME) and transanal-total mesorectal excision (TaTME) gradually replacing traditional abdominoperineal resection in the treatment of low and ultra-low rectal cancer, LARS management becomes a new concern of colorectal cancer surgery team. Optimal management of LARS is the cornerstone of achieving maximum quality of life of these rectal cancer patients. However, for potential LARS patients, the recovery process principally occurs at home, which lacks guidance or assistance from healthcare professionals [8]. Although the Internet represents a powerful tool for the provision of health information, a recent study that evaluated current LARS-related websites found that the online information related to LARS was suboptimal due to great heterogeneity between websites, the lack of important content, and the required reading levels of patients [11]. Thus, patients have to rely on information gathered during hospitalization and follow-up to cope with potential physical and psychological problems. The awareness and practice of LARS management by healthcare professionals have a direct impact on the rehabilitation of such patients.

Few studies have so far investigated LARS management among healthcare professionals, and the results of such studies have not been ideal [12, 13]. An international study has shown that colorectal cancer experts lack a thorough understanding of patients’ bowel dysfunction following sphincter-sparing treatment and its impact on patients’ quality of life [12]. Another study came to a similar conclusion and found that physicians severely underestimated the prevalence of LARS [13]. As important members of the colorectal surgical team and a bridge between patients and the healthcare system, colorectal surgery nurses (CSNs) play an important role in preoperative education, postoperative monitoring, observation, education, and follow-up for these patients, placing high demands on their comprehension and skill to manage LARS. Therefore, the present study adopted a knowledge, attitudes, and practices model to investigate the status of CSNs in LARS management, with a view to providing the basis for subsequent training and education. Furthermore, the study was undertaken to inspire more nurses to focus their attention on the new clinical problem of LARS so as to improve health outcomes in these patients.

Methods

Design and setting

A multicenter cross-sectional study was conducted using a paper-based questionnaire with CSNs enrolled in six tertiary referral hospitals across four cities in Jilin Province, China (Changchun, Jilin, Siping, and Yanbian), between June and November 2019.

Participants

The study cohort comprised certified registered nurses who had worked in the colorectal surgery department for more than 12 months. Training nurses from other hospitals and student nurses were excluded.

Sample size analysis and sampling technique

As a previous study that was insufficiently considered was available for reference, sample size estimation for the present study was based on available resources and the results of a pilot study. It was estimated that the number of colorectal surgical nurses working in tertiary hospitals in Jilin Province was approximately 1500, and the standard deviations of their knowledge, attitude, and practice scores were 2.28, 3.05, and 5.38, according to the results of a pilot study. Given α = 0.05, confidence level 1-α = 0.95, and distance from mean to limit(s) = 0.5, the required sample sizes were calculated to be 79, 133, and 345, using PASS 11 software (NCSS, LLC. Kaysville, UT, USA) [14]. The largest sample size was adopted, plus an additional 10% for incomplete or invalid responses; a target sample of 380 was selected.

Random cluster sampling was used in the study. Six hospitals were randomly selected from 25 tertiary hospitals in Jilin Province. All CSNs working in the selected hospitals during the study period were cluster-sampled, forming a cohort of 392 participants.

Data collection instruments

The questionnaire consisted of five sections incorporating sociodemographic information; three sections that posed questions about their knowledge, attitude, and practice of LARS management; and finally, a training needs survey. These are detailed below:

  • Section 1: Sociodemographic survey questions captured age, gender, educational level, professional title, etc.

  • Section 2: The knowledge questionnaire consisted of ten items concerning the duration (Q1), etiology (Q3), incidence (Q6), risk factors (Q2, Q3, Q7, Q8), and treatment of LARS (Q4, Q5, Q9, Q10). Options for response were “true,” “false,” or “do not know.” The knowledge score was defined as the total number of correct answers to the ten items yielding a possible score of 0–10 (correct answer = 1, incorrect or unknown answer = 0).

  • Section 3: The attitude questionnaire consisted of four items concerning the attitude of individuals towards LARS, their responsibility for management of LARS, confidence in their ability to manage LARS, and the impact of LARS. The total score of these items was calculated using the 5-point Likert scale (“strongly agree = 5,” “agree = 4,” “no opinion = 3,” “disagree = 2,” and “strongly disagree = 1”); the total score ranges from 5 to 20.

  • Section 4: The practice questionnaire consisted of six items concerning the practice of LARS management in three stages: preoperative, postoperative, and follow-up. The total score of six items was calculated using the 5-point Likert scale (“always = 5,” “frequently = 4,” “sometimes = 3,” “seldom = 2,” and “never = 1”); the total score therefore rangres between 6 and 30. In addition, two further multiple-choice questions not counting towards the final score are posed at the end of the questionnaire to determine which assessment tools and treatment methods were used or recommended by participants during practice.

  • Section 5: The training needs survey comprised 2 questions, namely: “Do you want to receive LARS training in the future?” (single-choice question, options for response included “Yes,” “No,” or “Unsure”) and “Which of the following training methods would you like to receive?” (multiple-choice question, options for response included “eLearning,” “instructor-led training,” “brochure,” “lectures,” or “other methods”).

The study tool was developed by the study team, following an extensive literature review. Content validity was evaluated using a double Delphi procedure by an expert panel of six senior nurses or nursing professors. Items that did not meet the requirements were modified or deleted. The item-level content validity index of each item in the three questionnaires was greater than 0.78 [15], suggesting that the questionnaires contained valid qualified content. To evaluate the feasibility and construct validity of the questionnaires, a pilot study was performed on 50 CSNs working at a tertiary referral hospital using a convenient sampling method in March 2019. For items that were answered dichotomously, the internal consistency was studied by calculating KR-20 [16]. The KR-20 measure of the knowledge questionnaire was 0.706. Where more than two alternative responses were possible for each item, internal consistency was studied by calculating Cronbach’s α [16]. Cronbach’s α of the attitude and practice questionnaires were 0.709 and 0.816, respectively. The results above were all greater than 0.7 and less than 0.9 [16], suggesting that the questionnaires possessed qualified internal consistency. After 2 weeks, participants in the pilot study were asked to participate in a test-retest procedure. The reliability coefficients of the three questionnaires were 0.857, 0.860, and 0.882, and as all were greater than 0.7 [17], the questionnaires were considered to have qualified stability.

Data collection

A researcher personally distributed the questionnaire to all CSNs in all selected tertiary referral hospitals on the scheduled date. Uniform instructions were provided to explain the aims of the study. All participants were required to not use any resources or ask others when completing the questionnaire. Each nurse was informed that the study was anonymous and voluntary. After a week, the head nurse of the colorectal surgery department in each hospital collected the complete questionnaires with consent forms and returned them to the study researcher by mail.

Data analysis

All data analyses were conducted using SPSS v23.0 software (SPSS Inc., Chicago, IL, USA). Descriptive statistics, an independent t test, one-way analysis of variance (ANOVA), Scheffe’s test, and multiple linear regression analysis were performed. Descriptive analysis included frequency and percentage distribution for categorical variables and means and standard deviations for numerical variables. To determine the association between participants’ sociodemographic characteristics and scores of knowledge, attitudes, and practices regarding LARS, and independent t test and ANOVA were used. Post hoc testing for pairwise comparisons was conducted using Scheffe’s procedure. Independent variables were also introduced into the multiple linear regression model that was significant in the independent t test and ANOVA. The collinearity of independent variables was investigated using the variance expansion factor (VIF) index. If the VIF was greater than 10, multicollinearity was detected. Finally, variables in multiple linear regression with a P value < 0.05 were considered significantly associated with the outcome variable.

Results

Descriptive statistics

A total of 387 qualified nurses agreed to participate in the survey. Of these, 382 completed the questionnaire. After excluding invalid questionnaires, 361 eligible questionnaires were included in the analysis. Table 1 summarizes the sociodemographic characteristics of the participants. The vast majority of respondents (95.3%) were women. Except for a few older nurses (1.1%) with a technical secondary level of education only, the majority of participants (71.2%) had a Bachelor’s degree. A clear majority of participants (71.5%) had never received training on LARS.

Table 1 Sociodemographic characteristics of eligible respondents (N = 361)
Full size table

Knowledge

The responses of CSNs to knowledge-related statements is displayed in Fig. 1. The mean (SD) score obtained by the participants was 4.72 (1.91). The highest mean scores were achieved on the themes of etiology (0.83 ± 0.38), which contained only one item (i.e., “Internal anal sphincter dysfunction is an important pathophysiological factor of LARS”). The lowest mean score was attained on the theme of duration (0.17 ± 0.38), which also included only one item (i.e., “All symptoms of LARS will be completely resolved within 6 months of surgery”). Of the four items in the theme of treatment, fewer than 50% of participants selected the correct answer for half of the questions, and only 16.1% of participants knew that transanal irrigation (TAI) had a significant clinical effect on LARS.

Fig. 1
figure 1

Colorectal surgery nurses’ knowledge of low anterior resection syndrome

Full size image

Attitudes

The response of participants to attitude-related statements is presented in Fig. 2. The mean (SD) score obtained by the nurses was 12.77 (2.61). Of the four questions, the lowest score was for question 3 which was regarding confidence in the ability to manage LARS, with a mean score of 2.57 (SD = 1.00). The majority of participants (61.22%) disagreed that they felt confident managing LARS. Most nurses (61.49%) believed that LARS had a strong negative impact on the quality of life of patients, but fewer than half (48.75%) considered it to be an important clinical disorder.

Fig. 2
figure 2

Colorectal surgery nurses’ attitude to low anterior resection syndrome

Full size image

Practice

The response of participants to practice-related statements is shown in Fig. 3. The mean (SD) score obtained by the nurses was 18.05 (5.54). Of the six questions, the lowest score related to questions 1 (mean = 2.81, SD = 1.18) and 2 (mean = 2.71, SD = 1.24), with almost half of the participants having rarely or never assessed the risk factors of patients for LARS, or had provided LARS-related education prior to surgery. However, a significant number of participants stated that they often educated patients before discharge (41.55%) and assessed change in bowel function during the follow-up period (48.76%). These two items scored the highest. A considerable number of participants stated that they rarely or never assessed patient bowel function (35.18%) or discussed treatment options for LARS during follow-up (35.46%). The largest number of nurses used the “Wexner Incontinence Score” to assess patients (46.2%), followed by “The Fecal Incontinence Severity Index (FISI)” (43.9%). The number of participants adopting the Low Anterior Resection Syndrome Score (LARS score), a professional assessment tool, was small, accounting for only 26.8% of nurses. The majority of participants recommended biofeedback therapy (83.7%) and pelvic floor muscle training (78%) to patients, while TAI (14.7%) was recommended less often.

Fig. 3
figure 3

Colorectal surgery nurses’ practice on low anterior resection syndrome

Full size image

Factors associated with knowledge

The relationship between sociodemographic characteristics and knowledge score for LARS is presented in Table 1. Significant differences were observed in knowledge score depending on gender, educational level, form of employment, and whether training on LARS had been received. Further analysis with multiple linear regression revealed that there was a significant linear positive relationship between educational level and knowledge score, indicating that participants with a higher educational level had a higher knowledge score (β = 0.67, p = 0.000). Knowledge scores for staff nurses were higher than those of contract nurses (β = 0.61, p = 0.012), and nurses who had received training scored higher than those who had not (β = 0.95, p = 0.000) (Table S1).

Factors associated with attitude

The relationship between sociodemographic characteristics and attitude score for LARS is shown in Table 1. Significant differences were found for age, work experience in colorectal surgery, professional title, form of employment, and whether training had been received with attitude score for LARS. Further analysis using multiple linear regression revealed that years of practice in colorectal surgery significantly predicted high attitude scores, suggesting that the longer participants worked in colorectal surgery, the higher their attitude scores (β = 0.43, p = 0.036). The attitude scores of staff nurses were higher than those of contract nurses (β = 1.19, p = 0.001), and nurses who had received training higher than those who had not (β = 1.04, p = 0.001) (Table S2).

Factors associated with practice

The relationship between sociodemographic characteristics and practice scores for LARS is shown in Table 1. Significant differences were found between educational level, whether training had been received, and practice score for LARS. Further analysis with multiple linear regression revealed that educational level was positively associated with practice scores, indicating that nurses with higher educational levels obtained higher practice scores (β = 3.13, p = 0.000). Nurses that were trained scored higher than those who were not (β = 2.95, p = 0.000) (Table S3).

Training needs

The vast majority of participants stated that they wanted to receive training for LARS in the future (78.1%). Most participants were willing to receive “instructor-led training” (70.9%), followed by “brochure” (64.3%) and “eLearning” (58.7%).

Discussion

As the principal practitioner of patient education and communication, the awareness of CSNs and their practice of LARS management are the foundation of provision of optimal management of LARS for patients. The purpose of this study was to survey and assess the knowledge, attitude, and practice of LARS management by CSNs. The overarching study outcomes indicated that the majority of nurses were ill-informed about LARS, and the results of attitude and practice were also discouraging.

Unlike the underestimation of surgeons regarding the incidence of LARS, the principal problem with the knowledge that Chinese CSNs had about LARS was that they underestimated its duration [13, 18]. The vast majority of nurses believed that all symptoms of LARS would be completely resolved within 6 months of surgery. However, such intestinal sequelae last for more than 2 years in approximately half of patients, although all symptoms improve over time in the first 2 years [19, 20]. Follow-up actions following discharge are mainly conducted by nurses, so they directly grasp the recovery status and problems of patients, having greater understanding of the incidence of LARS and negative impact based on experience. However, their awareness is also limited by the duration of follow-up; thus, they can easily underestimate the duration of LARS. This may also explain the ambivalent attitude of a number of participants, that is, they agreed that LARS seriously damages the quality of life of patients, but disagreed LARS was an important clinical disorder.

With regard to risk factors and treatment, the results also appeared negative with a considerable number of participants having insufficient knowledge of risk factors and treatment methods. Lack of awareness of risk factors would inevitably impair the quality of preoperative evaluation and education, having an adverse impact on patient’s treatment choices and expectations. The unfamiliarity with treatment is reflected by more than half of nurses lacking confidence in the management of LARS, and a considerable number never or rarely discussed treatment options with patients during follow-up. Biofeedback therapy and pelvic floor muscle training were most often recommended, while TAI, as a proven method with excellent clinical outcomes, technically easy to learn, and of low cost, was poorly understood and seldom recommended [21,22,23].

Preoperative assessment and education are crucial for patient education to be effective, but was most often overlooked by participants. Adequate preoperative assessment would assist nurses with the early identification of potential symptoms and problems that may damage the quality of life of patients and even influence their treatment decisions, providing forward-looking guidance and interventions. Through preoperative education, patients could fully anticipate and prepare for possible poor prognosis and problems. The lack of relevant knowledge undoubtedly restricts good practice. Assessment following surgery and during follow-up was performed better than preoperative evaluation. Similar to doctors’ concerns [12], incontinence has also been the main focus of postoperative bowel dysfunction among CSNs, reflected in the two tools used for fecal incontinence being used most frequently. A study comparing specialists’ versus patients’ perspectives towards bowel dysfunction after rectal cancer treatment found that experts vastly overestimated the negative impact of stool incontinence and frequent bowel movements on the quality of life, while they definitely underestimated the impact of clustering and urgency [12]. LARS is a multifactorial and multifaceted syndrome [7, 22, 24], and consideration of a single symptom for assessing this complex syndrome is markedly one-sided and inappropriate. LARS score, having been developed and validated for 8 years, is recognized as the most appropriate questionnaire for capturing postoperative bowel function and has been translated into several languages, conducive to international communication and cooperation [25, 26]. The Chinese version was validated in 2015 [27], but is still not widely used.

Whether training on LARS had been received was an important factor affecting the knowledge of nurses, their attitude and practice, and those who had received training had more comprehensive knowledge, a more positive attitude, and higher levels of best practice. However, most participants stated they had not participated in training (71.5%), and more participants stated that they would like to receive training on LARS in the future (78.1%). These highlight the necessity of LARS training. CSNs in the survey identified a preference for training in the form of “instructor-led training.” In any case, it is concerning that research has shown that those with low levels of education had poor knowledge and failed to undertake best practice, with contract nurses having lower levels of knowledge and attitudes than staff nurses, demonstrating the need for reinforcement of training for these people. Finally, the number of studies related to LARS is growing, with new methods being discovered and previous conclusions being overturned. This requires guidelines to be developed to regulate nursing practice and meet their educational needs, with training that needs to be updated on an ongoing basis.

Limitations

The selection of participants is the main limiting factor for this study. The survey was limited to CSNs in Jilin Province, China, and one should not draw conclusions for other countries or regions based on these results. In addition, the setting selected in our study was tertiary referral hospitals where a large number of high-level medical professionals are based and high-quality medical services provided, which may overestimate the current level of knowledge, attitude, and practice for LARS among CSNs in general.

Conclusion

As frontline providers, CSNs must have sufficient up-to-date clinical knowledge and skills to ensure they provide the patients at risk of LARS with optimal management. This study indicates that the knowledge, attitude, and practice of Chinese CSNs regarding LARS were discouraging, with gaps between the requirements and reality being evident. The number of those receiving LARS training was small but they had better knowledge, attitude, and behavior. Training is undoubtedly a key strategy to address this gap.