Background

Renal replacement therapy (RRT) is fluid removal and replacement for maintaining solute, acid–base, and electrolyte balance using dialysis and/or hemofiltration [1,2,3]. The types of venous RRT include continuous renal replacement therapy (CRRT), intermittent renal replacement therapy (IRRT), sustained low-efficiency dialysis (SLED), and peritoneal dialysis, which is more common in areas with limited resources and is easier than venous RRT to administer, but it does not allow for control of the fluid removal rate, and there are risks of protein loss, peritonitis, hyperglycemia, and potential respiratory impairment [1,2,3]. Some patients will eventually require kidney transplantation. The selection of RRT should be made by evaluating the entire clinical scenario, the presence of factors that can be modified with RRT, and trends of laboratory tests when deciding to start RRT instead of relying on blood urea nitrogen (BUN) and creatinine thresholds alone [1,2,3]. In addition to guidelines, individualized factors, including the patient’s financial situation, compliance, caregivers, and residual urine output, should also be considered.

Still, whether nephrologists are well acquainted with the principles of RRT and apply them in practice is poorly understood. Emotional burden is often encountered during and influences RRT decision-making [4, 5]. It has been reported that many nephrologists feel uncomfortable not offering dialysis for reasons they poorly understand, but it might include prognostic uncertainty and discomfort with a possible death [5, 6]. The Knowledge, Attitude, and Practice (KAP) framework is a quantitative method based on standardized questionnaires that provide quantitative and qualitative data that can help unravel the misconceptions and misunderstandings posing obstacles in a specific clinical activity and help identify specific points for behavior changes [7, 8].

Therefore, this study aimed to examine the KAP of nephrologists on the decision of RRT. The results could help pinpoints areas that would benefit from additional training.

Methods

Study design and participants

This multicenter cross-sectional study was conducted between July 1, 2022, and August 20, 2022, and enrolled all qualified nephrologists who volunteered to participate. Nephrologists on maternity or sick leave or temporary workers were excluded. This study was approved by the Ethics Committee of the Hunan Provincial People’s Hospital (as the lead center). All participants signed the informed consent.

Procedures

According to the 17th Acute Disease Quality Initiative International Consensus Conference: introducing precision renal replacement therapy and KDIGO clinical practice guideline for acute kidney injury [9]. This questionnaire was modified following the comments of two senior nephrologists. Initially, the knowledge dimension consisted of 12 questions, while the attitudes and practices dimensions had 12 and 10 questions, respectively. After consultation with experts, the knowledge dimension was reviewed, and four questions were added, resulting in 16 questions. The attitude dimension was expanded with seven more questions, resulting in 19. The practice dimension was modified to assess all three decision-making settings, and some of the questions were deemed inappropriate or inaccurate and were replaced. For instance, the original questionnaire referred to “palliative care specialists,” which may not be available in some areas. Hence, it was modified to “nutrition specialists,” resulting in 11 questions.

The questionnaire was pre-tested and had a Cronbach’s α of 0.938 and a KMO of 0.885, suggesting high internal consistency. The final questionnaire was the Chinese version and included four dimensions: 1) the demographic data of the participants, including gender, age, marital status, level of education, grade of hospitals, professional title, working experience, methods of RRT available in hospitals, population with dialysis (including peritoneal dialysis and hemodialysis) in hospitals, and region of hospitals; 2) 16 questions about the knowledge for the decision of RRT (scored 1 point for correct answers and 0 points for incorrect or unclear answers, ranging from 0 to 16 points); 3) 19 items (5-point Likert scale) about the attitude of nephrologist on the decision of RRT, with 15 questions scored from very positive (5 points) to very negative (1 point) (total score ranging from 15 to 75 points) and four open questions; 4) 11 items on the practice of nephrologist on decision of RRT, including six questions scored on a 5-point Likert scale scored from always (5 points) to never (1 point) (total score ranging from 6 to 30 points) and five open questions.

An online questionnaire with a QR code was established using the WeChat-based Questionnaire Star applet to collect data through WeChat. The participants logged in via WeChat by scanning the QR code and completing the questionnaire. In order to ensure quality and completeness, a given IP address could only submit the questionnaire once, and all questions had to be answered. The research team checked all questionnaires for completeness, internal coherence, and rationality.

Statistical analysis

SPSS 26.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Continuous data were expressed as means ± standard deviation (SD) and compared by t-test or ANOVA, while the least-significant difference (LSD) was used as a post hoc test. Categorical data were expressed as n (%). Pearson correlation was used to analyze the correlations between knowledge, attitude, and practice scores. Logistic regression was used to analyze influencing factors of practice. The median of the practice score (= 20) was used as the cut-off value. Factors with P < 0.001 in the univariable analyses were included in the multivariable logistic regression analysis. Two-sided P-values < 0.05 were considered statistically significant.

Results

Ultimately, 327 nephrologists participated in this study, including 104 (31.80%) males and 223 (68.20%) females. Most were 31–40 years of age (61.16%), unmarried (86.54%), bachelor’s degree education (57.49%), public tertiary hospital (65.44%), with intermediate professional titles (51.68%), and with 5–10 years of working experience (35.17%). Peritoneal dialysis was available for 78.59% of the nephrologists, hemodialysis for 98.47%, kidney transplantation for 23.55%, and none for 1.53%. Most nephrologists had a patient volume of 100–300 per year (37.31%) (Table 1).

Table 1 Baseline characteristics and KAP scores
Full size table

The total knowledge, attitude, and practice scores were 12.03 ± 2.11/16, 58.39 ± 6.62/75, and 27.15 ± 2.74/30, respectively. The knowledge scores were influenced by the level of education (P < 0.001), grade of hospitals (P < 0.001), professional title (P < 0.001), the available RRT methods (all P < 0.001), and patient volume (P < 0.001). No factor was significantly associated with the attitude scores (all P > 0.05). The available RRT methods were associated with the practice scores (all P < 0.05) (Table 1). Table 2 presents the responses to the knowledge dimension items. Table 3 presents the answers to the items of the attitude dimension, while Table 4 presents the answers to the practice dimension items. The Pearson correlation analyses showed that the knowledge scores correlated with the practice scores (r = 0.239, P < 0.001) and the consideration score of peritoneal dialysis (r = 0.171, P = 0.002). The attitude scores correlated with the practice scores (r = 0.251, P < 0.001), the consideration score of peritoneal dialysis (r = 0.453, P < 0.001), the consideration score of hemodialysis (r = 0.425, P < 0.001), and the consideration score of kidney transplantation (r = 0.407, P < 0.001). The practice scores correlated with the consideration score of peritoneal dialysis (r = 0.145, P = 0.009) and the consideration score of kidney transplantation (r = 0.119, P = 0.031). The consideration score of peritoneal dialysis correlated with the consideration score of hemodialysis (r = 0.597, P < 0.001) and the consideration score of kidney transplantation (r = 0.489, P < 0.001). The consideration score of hemodialysis correlated with the consideration score of kidney transplantation (r = 0.673, P < 0.001) (Supplementary Table 1).

Table 2 “Knowledge” dimension
Full size table
Table 3 “Attitude” dimension
Full size table
Table 4 “Practice” dimension
Full size table

The multivariable logistic regression analysis showed that the attitude score (OR = 1.19, 95%CI: 1.13–1.25, P < 0.001), 41–50 years of age (vs. < 30, OR = 0.45, 95%CI: 0.21–0.98, P = 0.045), and > 50 years of age (vs. < 30, OR = 0.27, 95%CI: 0.08–0.84, P = 0.024) were independently associated with the consideration score of peritoneal dialysis (Table 5).

Table 5 Logistic regression (consideration scores of peritoneal dialysis, hemodialysis, and kidney transplantation)
Full size table

The attitude score (OR = 1.14, 95%CI: 1.09–1.19, P < 0.001), female gender (OR = 1.66, 95%CI: 1.04–2.66, P = 0.034), 41–50 years of age (vs. < 30, OR = 0.27, 95%CI: 0.12–0.60, P = 0.001), > 50 years of age (vs. < 30, OR = 0.31, 95%CI: 0.10–0.97, P = 0.043), senior professional title (vs. junior, OR = 0.34, 95%CI: 0.12–0.96, P = 0.042), and ≥ 16 years of working experience (vs. ≤ 5 years, OR = 0.45, 95%CI: 0.23–0.88, P = 0.020) were independently associated with the consideration score of hemodialysis (Table 5).

The knowledge score (OR = 1.89, 95%CI: 0.79–0.995, P = 0.041), attitude score (OR = 1.12, 95%CI: 1.07–1.16, P < 0.001), 41–50 years of age (vs. < 30, OR = 0.45, 95%CI: 0.20–0.97, P = 0.042), > 50 years of age (vs. < 30, OR = 0.24, 95%CI: 0.08–0.77, P = 0.016), and below public tertiary hospital (vs. public tertiary hospital, OR = 2.22, 95%CI: 1.30–3.79, P = 0.004) were independently associated with the consideration score of kidney transplantation (Table 5).

The multivariable logistic regression analysis showed that the knowledge scores (OR = 1.08, 95%CI: 1.03–1.13, P = 0.001) were independently associated with the practice scores (Table 6).

Table 6 Logistic regression (practice)
Full size table

Discussion

This study suggests that good attitudes may lead nephrologists to make more considerations when choosing between peritoneal dialysis, hemodialysis, and kidney transplantation, while senior physicians may consider relatively less when making decisions, and in addition, both good knowledge and good attitudes may lead to good practice.

In the present study, the knowledge, attitude, and practice scores of Chinese nephrologists regarding the decision for RRT were 12.0 ± 2.1/16, 58.4 ± 6.6/75, and 27.2 ± 2.7/30, respectively, which could be considered moderate, low, and high. Ockhuis & Kyriacos [10] (South Africa) reported knowledge, attitude, and practice scores of 10.8 ± 3.1/18, 25.9 ± 6.0/41, and 35.8 ± 6.0/50 for the safe use of unfractionated heparin during RRT. A study in Nepal showed relatively poor to moderate knowledge regarding kidney diseases but relatively good attitude and practice [11]. A study in Pakistan revealed that only 18% of physicians had good knowledge about kidney diseases [12], while good knowledge was seen in 24% of physicians in Sudan [13]. The need for training of Brazilian palliative care physicians regarding RRT has been emphasized by a study [14]. The relatively higher KAP scores observed in the presented study could be because only nephrologists were enrolled.

In this study, only the knowledge scores independently influenced the practice scores. On the other hand, since peritoneal dialysis, hemodialysis, and kidney transplantation have different indications and target patient populations, more factors were associated with those three scores: the attitude score and age were independently associated with the consideration score of peritoneal dialysis; the attitude score, female gender, age, senior professional title, and ≥ 16 years of working experience were independently associated with the consideration score of hemodialysis; the knowledge score, attitude score, age, and below public tertiary hospital were independently associated with the consideration score of kidney transplantation. Ochkuis & Kyriacos [10] reported that the category of professionals, knowledge, and years of experience influenced the quality of dialysis practice, but their study focused on using unfractionated heparin. The middle-adulthood age range (40–65 years) is considered stage 7 of psychosocial development, characterized by an experience of stagnation and a feeling of unproductiveness [15, 16], which could explain why older age is consistently associated with lower KAP scores. A qualitative study by Greer et al. identified hospital resources, provider skills, and patient attitudes as the main barriers to RRT [17]. Wu et al. [18] reported that the main barriers to implementing AKI management in China were inadequate knowledge, inadequate training, absence of clinical protocols, and insufficient multidisciplinary cooperation. It is also supported by studies from developing countries [19,20,21]. A worldwide survey identified the patients, the nephrologists, geography, and the healthcare systems as barriers to RRT in 78%, 71%, 72%, and 73% of the countries, respectively [22]. Work experience was not associated with the KAP dimensions, which was a little surprising. Indeed, it can be expected that one’s knowledge will increase with experience, but work experience is only a metric of how long an individual has been involved in a particular work. It does not evaluate his knowledge, motivation, enthusiasm, and attitude. A young physician can be very enthusiastic at the idea of helping patients, while another might be only attracted by the social status of being a physician. Some older physicians can still be very enthusiastic in their job, while others might be tired and waiting for retirement. Another factor could be that continuous education, favorable attitudes, and practice according to the guidelines are similar across work experience. Unfortunately, the available data do not allow delving deeper into why work experience is not associated with any KAP dimensions.

Hence, the present study identified categories of nephrologists who might benefit from training on RRT to improve their KAP. Considering that the current clinical trend is that the selection of RRT should be made by evaluating the entire clinical scenario, modifiable factors, and laboratory tests when deciding to start RRT instead of relying on BUN and creatinine alone [1,2,3], training should be implemented to improve the nephrologists’ awareness of the factors to be considered for starting RRT. Group discussions to share past experiences could be of use. Postgraduate training and continuous education should be emphasized. Multidisciplinary management should also be explored.

This study has some limitations. Considering the number of nephrologists in China, the sample size was relatively small, and the nephrologists were mainly from Eastern China. Future studies should enroll a sample size more representative of the entire country. The most effective treatment for CKD remains kidney transplantation, but only a few questions were about transplantation. Of course, a KAP survey has limitations. It can only identify deficits related to the asked questions, and the points not covered by the questionnaire items will remain unknown. In addition, a KAP questionnaire is usually specifically designed for a given hospital, province, or country, and the results are difficult to generalize. Nevertheless, they can give ideas to researchers from around the globe for performing KAP surveys and implementing improvement training.

In conclusion, the knowledge, attitude, and practice scores of nephrologists regarding RRT were moderate, low, and high, respectively. This study also revealed factors associated with the KAP of nephrologists regarding RRT. It identified areas that could be targeted by additional training. Future studies should examine the implementation of different training methods to improve the KAP of nephrologists toward RRT.