Knowledge, attitude and practices regarding postsurgical gastroparesis syndrome among at-risk patients following abdominal surgery in a tertiary hospital in China: a cross-sectional study

Abstract

Objectives

To assess the knowledge, attitude and practices (KAP) regarding postsurgical gastroparesis syndrome (PGS) among an at-risk population of patients who have undergone abdominal surgery and to identify factors associated with their KAP levels.

Design

A cross-sectional study.

Setting

Patients after abdominal surgery by administration of questionnaires.

Participants

The study was conducted at a tertiary hospital in the western region of Liaoning Province, China from July 2023 to December 2023.

Primary and secondary outcome measures

KAP scores regarding postoperative gastrointestinal function recovery (PGS) in at-risk patients; associations between demographic factors and KAP dimensions, including subgroup differences across age, education, residence and other clinical variables.

Results

555 valid questionnaires were analysed. The KAP scores were 2 (0, 5), 26 (22, 31) and 16 (13, 19), respectively. Knowledge positively correlated with attitude (r=0.710, p<0.001) and practice (r=0.731, p<0.001). Attitude correlated with practice (r=0.968, p<0.001). Multivariable logistic regression showed that older age (46–60 years: OR=0.351, p=0.017; >60 years: OR=0.160, p<0.001), county township residency (OR=0.516, p=0.024), rural residency (OR=0.357, p=0.005), higher education (OR=8.351, p<0.001), ‘slight’ postoperative nausea and vomiting (OR=1.975, p=0.033) and ‘very severe’ postoperative nausea and vomiting (OR=8.070, p=0.016) were independently associated with adequate practice.

Conclusions

Patients at risk for PGS following abdominal surgery exhibit critical deficiencies in knowledge, hold passive attitudes and engage in inadequate preventive practices. Healthcare providers must urgently develop and implement precision health education strategies tailored to these high-risk groups to improve awareness, foster positive attitudes and promote effective self-management for PGS prevention.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The validated questionnaire demonstrated high internal consistency (Cronbach’s α=0.930) and acceptable construct validity through confirmatory factor analysis.

• A robust sample size (n=555) exceeding the minimum required calculation (430) enhanced statistical reliability.

• The single-centre design and convenience sampling method may limit generalisability to other populations or settings.

• Reliance on self-reported data introduces potential recall and social desirability biases despite on-site completeness checks.

Background

Gastroparesis is a syndrome characterised by delayed gastric emptying and associated symptoms.^{1 2} Particularly, postsurgical gastroparesis syndrome (PGS) is recognised as an early common complication following upper abdominal surgery.^{3 4} The incidence of PGS has increased in recent years among abdominal surgery patients. The reported incidence in cases from China is as high as 2% to 10%, while international rates are slightly lower, ranging from 0.6% to 7.0%.⁵ Common surgeries associated with PGS include pancreaticoduodenectomy,⁶ gastrectomy^{7 8} and pancreatic surgery.⁹

Patients with gastroparesis manifest various symptoms such as early satiety, nausea, vomiting, epigastric abdominal pain and bloating.^{10 11} Studies have further indicated that gastroparesis significantly impacts nutritional status, leading to malnutrition and weight loss, contributes to psychological disorders such as anxiety and depression and considerably diminishes patients’ quality of life.^{12 13} Drug side effects often limit long-term treatment tolerance, leading to drug-refractory symptoms. This severity significantly interferes with daily life, restricts labour ability and aggravates the family’s financial burden.^{14 15} Current treatment strategies primarily include dietary modification, prokinetic agents, antiemetics, gastric electrical stimulation and surgical interventions, yet these approaches often have limited efficacy due to variable patient responses and adverse effects.^{16 17} Optimising patients’ nutritional status and psychological state preoperatively, and initiating prokinetic therapy early after surgery, are effective patient-centred measures to reduce PGS incidence.¹⁸

In the realm of knowledge, attitude and practice (KAP) theory, which delves into the dynamic interplay between an individual’s awareness, beliefs and actions, a conceptual framework has been established to grasp how patients respond and engage.^{19 20} By delving into the intricacies of patients’ KAP, a deeper understanding of their needs is facilitated, aiding the healthcare team in comprehending patient requirements and expectations. This, in turn, contributes to refining doctor–patient communication, fortifying the trust relationship between the medical team and patients, elevating the quality of healthcare services and fostering a more proactive approach to treatment. Previous studies applying KAP theory in gastroenterology, particularly from the patient perspective, have demonstrated its effectiveness in enhancing patient adherence to treatment regimens, optimising disease management strategies and improving patient satisfaction with medical services.^{21 22}

Current research related to PGS, particularly within the Chinese literature, is mostly centred on analysing clinical risk factors, treatment efficacies or routine care countermeasures. While the KAP framework has been used in managing other gastrointestinal diseases, there remains a critical research gap regarding its application to postoperative complications. Specifically, there is a lack of in-depth and systematic research exploring knowledge dissemination, attitudes towards treatment and preventive behavioural practices for PGS among patients who have undergone abdominal surgery in China. This study aims to evaluate the KAP regarding PGS among an at-risk population of patients who have undergone abdominal surgery.

Methods

Patient and public involvement

This study did not include patient or public involvement in its design, conduct or reporting.

Study design and subjects

This cross-sectional study investigated the KAP regarding PGS among at-risk patients who have undergone abdominal surgery. The study was conducted at the First Affiliated Hospital of Jinzhou Medical University in western Liaoning, China, from July to December 2023. To clearly define the characteristics of the surveyed population, the inclusion criteria were: (1) inpatients in the postoperative recovery period (specifically defined as 4 to 25 days postsurgery, representing the typical screening window from PGS onset to symptom alleviation)²³ who had undergone specific abdominal surgeries, including duodenectomy, pancreatic resection, radical or subtotal gastrectomy for gastric cancer and other tumour resections (radical gastrectomy here refers to oncological resections preserving a portion of the stomach); (2) patients aged >18 years; (3) individuals with normal verbal communication and literacy skills and (4) those who provided signed informed consent. The exclusion criteria were: (1) individuals with severe organic diseases or complications of malignant tumours; (2) those with mental disorders or unconsciousness that might affect the conduct of the study and (3) patients who refused to participate or withdrew. Exclusion data was verified through patient medical records and direct communication with the patients and their family members.

Procedures

The questionnaire was originally developed in Chinese and translated into English using a two-person two-way translation method to ensure semantic equivalence. The initial draft was developed by referencing relevant guidelines and literature, with subsequent modifications based on opinions from a panel of 15 experts. These experts, affiliated with various hospitals and higher medical institutions, specialised in gastrointestinal surgery clinical medicine, perioperative nursing management, nursing education and scale or assessment tool development. The panel comprised 10 nurses, 2 doctors and 3 teachers (9 with intermediate professional titles and 6 with senior titles; 13 holding master’s degrees and 2 holding doctoral degrees). Their ages ranged from 36 to 49 years (mean 42.47±3.96), and their work experience ranged from 10 to over 20 years (10 experts with 10–20 years and 5 with more than 20 years). After the initial draft, the questionnaire underwent two small-scale pilot tests. In the first pilot test, 51 questionnaires were distributed, yielding a Cronbach’s alpha of 0.662. Following this, the experts adjusted the questionnaire based on the Cronbach’s alpha coefficient results to ensure internal consistency and content validity. The adjusted version was then applied in the second pilot test, during which 45 questionnaires were distributed, yielding a Cronbach’s alpha of 0.767. The Cronbach’s α coefficient of the formal experiment feedback was 0.930 (knowledge dimension Cronbach’s α=0.838, attitude dimension Cronbach’s α=0.840, practice dimension Cronbach’s α=0.864). To further assess construct validity, confirmatory factor analysis was conducted. The model-fitting indicators are acceptable. Specifically, Chi-square/degrees of freedom (CMIN/DF) = 4.470, Root Mean Square Error of Approximation (RMSEA) = 0.079, Increasing Fitting Index (IFI) = 0.917, Tucker-Lewis Index (TLI) = 0.905, and Comparative Fit Index (CFI) = 0.917 (online supplemental table S1). Standardised factor loadings ranged from 0.184 to 0.867. Additionally, the Kaiser-Meyer-Olkin value was 0.928 (p<0.001), indicating an acceptable model fit (online supplemental table S2 and figure 1).

Figure 1. Confirmatory factor analysis path diagram for the KAP questionnaire. The path diagram illustrates the relationships between the latent dimensions (KAP) and their corresponding observed indicators. The model demonstrated an acceptable fit, CMIN/DF=4.470 (1-3: Excellent, 3-5: Good), RMSEA=0.079 (<0.08: Good), IFI=0.917 (>0.8: Good), and TLI=0.905 (>0.8: Good), CFI=0.917 (>0.8: Good), with standardised factor loadings ranging from 0.184 to 0.867. CFI, Comparative Fit Index, IFI, Increasing Fitting Index, KAP, knowledge, attitude and practice, RMSEA, Root Mean Square Error of Approximation, TLI, Tucker-Lewis Index.

The questionnaire comprised four dimensions: demographic information (age, gender, body mass index (BMI), marital status, occupational status, place of residence, education level, etc), knowledge dimension, attitude dimension and practice dimension (see online supplemental material: Parts 1–4). The knowledge dimension encompassed two aspects with nine questions, where correct answers scored 1 point and incorrect or uncertain responses scored 0 points, resulting in a score range of 0 to 9. The attitude dimension comprised eight questions rated on a 5-point Likert scale from very positive (5) to very negative (1), yielding scores between 8 and 40. The practice dimension included five questions, also using a 5-point Likert scale ranging from always (5) to never (1), resulting in scores between 5 and 25.

Participants have the option to complete the questionnaire either digitally through a QR code or on paper. Those who choose the paper version will be allocated a private and quiet location for completion. Electronic questionnaires, facilitated via WeChat and using the Questionnaire Star online platform (https://www.wjx.cn/vm/rxuPe7u.aspx), will be generated by the researchers and distributed through provided links. It is crucial for participants to complete the questionnaire within a specified time frame, and collected data should be organised on the same day whenever possible. Participants were selected using a convenience sampling method based on their availability and willingness to participate. During the data sorting and analysis phase, each retrieved dataset will be assigned a unique number to prevent duplicate entries. On-site checks will be conducted to ensure the completeness of respondent questionnaires, with immediate rectification of any identified issues. Incorrectly or incompletely filled questionnaires will be screened and eliminated. The collected data will undergo a two-person verification process before being input into an Excel table. Subsequently, the data will be imported into SPSS V.26.0 for analysis.

Sample size calculation

We determined the minimum required sample size (n) using the standard formula for cross-sectional studies, assuming the maximum sample size where the proportion (p) is 0.5:

$${\displaystyle {\displaystyle n=\frac{x_{(1-\alpha )/2}^{2}p(1-p)}{{\delta }^2}}}$$

The type I error (α) was set to 0.05, corresponding to a Z-score of 1.96 (for a 95% confidence level). The standard error (δ) was set at 0.05. Substituting these values yields a required sample size of 384. Assuming an effective questionnaire recovery rate of 90%, the final target is to collect at least 430 completed questionnaires.

Statistical analysis

Demographic data and question responses were presented as counts and percentages (N (%)). Given the non-normal distribution of data, dimension scores were described using the median, 25th percentile and 75th percentile. The Wilcoxon-Mann-Whitney test and Kruskal-Wallis test were applied to compare dimension scores across demographic groups. Correlation analyses used Spearman coefficients. Univariate and multivariate logistic regression analyses examined associations between demographic data and KAP dimensions; for this, scores were dichotomised at the median. Forward stepwise logistic regression was used in the multivariate analysis to address multicollinearity, with p<0.05 considered significant.

Results

Sociodemographic profile of participants

In this study, a total of 555 valid questionnaires were collected for this study. Of the patients who participated in the study, 289 (52.1%) were male, and 329 (59.3%) were above 60 years of age. In terms of education level, 220 (39.6%) had primary school education and below, 200 (36.0%) had a junior high school education and 135 (24.4%) had high school/secondary school education and above. Additionally, 224 (40.4%) underwent partial gastrectomy and 342 (61.6%) had a hospitalisation payment of medical insurance and minor out-of-pocket expenses. 174 (31.4%) had diabetes mellitus, 295 (53.2%) had a BMI in the range of 18.5–25 kg/m², 148 (26.7%) had slight postoperative nausea and vomiting, 126 (22.7%) had very mild postoperative epigastric fullness/bloating and abdominal bloating/distension. The median (25th percentile, 75th percentile) score of KAP was 2 (0, 5), 26 (22, 31) and 16 (13, 19), separately (table 1).

Table 1. Basic information of participants and KAP score.

	N (%)	Knowledge (K)		Attitude (A)		Practice (P)
		Median (25% quartile, 75% quartile)	P value	Median (25% quartile, 75% quartile)	P value	Median (25% quartile, 75% quartile)	P value
Total	555	2 (0, 5)		26 (22, 31)		16 (13, 19)
Gender			0.921		0.914		0.883
Male	289 (52.1)	2 (0, 4)		26 (22, 31)		16 (13, 18)
Female	266 (47.9)	2 (0, 5)		26 (22, 31)		16 (12, 19)
Age			<0.001		<0.001		<0.001
45 years old and below	62 (11.2)	6 (4, 7) a		34 (31, 37) a		21 (18, 22) a
46–60 years old	164 (29.5)	3 (0.5, 6) b		29 (25, 33) b		18 (15, 20) b
Above 60 years old	329 (59.3)	1 (0, 2) c		24 (20, 27) c		14 (12, 17) c
Education			<0.001		<0.001		<0.001
Primary school and below	220 (39.6)	0 (0, 2) c		21.5 (18, 25) c		12 (11, 15) c
Junior high school	200 (36.0)	1 (0, 3) b		26 (24.5, 29.5) b		16 (14, 18) b
High school/secondary school and above	135 (24.4)	6 (5, 7) a		34 (32, 37) a		21 (19, 22) a
Marital status			0.398		0.147		0.175
Married	497 (89.5)	2 (0, 4)		26 (22, 31)		16 (12, 18)
Other	58 (10.5)	2 (0, 6)		27.5 (22, 34)		17 (13, 20)
Type of surgery			0.972		0.642		0.634
Partial gastrectomy	224 (40.4)	2 (0, 5)		26 (22, 31)		16 (13, 19)
Pancreaticoduodenectomy (Whipple procedure)	107 (19.3)	2 (0, 5)		26 (24, 32)		16 (13, 19)
Gastroduodenostomy	83 (15.0)	2 (0, 3)		26 (21, 30)		15 (12, 18)
Laparoscopic radical gastrectomy for gastric cancer	67 (12.1)	2 (0, 4)		26 (21, 29)		15 (12, 19)
Colon resection	39 (7.0)	2 (0, 5)		27 (21, 31)		17 (13, 19)
Rectal resection	35 (6.3)	2 (0, 5)		27 (24, 31)		17 (15, 18)
Work status			<0.001		<0.001		<0.001
Employed	156 (28.1)	5 (2, 6)		32 (27, 35)		19 (17, 21)
Other	399 (71.9)	1 (0, 3)		25 (21, 28)		15 (12, 17)
Payment for hospitalisation			0.409		0.069		0.198
Medical insurance and major out-of-pocket expenses	186 (33.5)	2 (0, 4)		26 (21, 30)		15 (12, 18)
Medical insurance and minor out-of-pocket expenses	342 (61.6)	2 (0, 5)		27 (23, 31)		16 (13, 19)
Out-of-pocket expenses only	27 (4.9)	2 (0, 5)		26 (20, 32)		16 (12, 19)
Permanent residence			<0.001		<0.001		<0.001
City centre	135 (24.3)	4 (1, 6) a		31 (26, 35) a		18 (16, 21) a
Suburban area	104 (18.7)	2 (0, 3) b		26 (21.5, 30) b		16 (12, 18) b
County townships	210 (37.8)	2 (0, 3) b		26 (22, 30) bc		15 (12, 18) b
Rural	106 (19.1)	1 (0, 2) cb		24.5 (20, 27) c		14 (12, 17) cb
Diabetes			0.359		0.560		0.417
Yes	174 (31.4)	2 (0, 3)		26 (22, 30)		16 (13, 18)
No	381 (68.6)	2 (0, 5)		26 (22, 31)		16 (13, 19)
BMI			0.233		0.210		0.209
Below 18.5	148 (26.7)	1 (0, 3)		26 (21, 30)		16 (12, 18)
18.5–25	295 (53.2)	2 (0, 5)		26 (22, 31)		16 (13, 19)
Above 25	112 (20.2)	2 (0, 5.5)		26 (22.5, 32)		16 (13, 19)
Postoperative nausea and vomiting			0.647		0.099		0.213
None	175 (31.5)	2 (0, 4)		25 (21, 30)		15 (12, 19)
Very mild	143 (25.8)	2 (0, 4)		27 (24, 31)		16 (14, 19)
Slight	148 (26.7)	2 (0, 5)		27 (23, 31)		17 (13, 18)
Mild	62 (11.2)	2 (0, 5)		27.5 (23, 32)		16 (13, 19)
Severe	14 (2.5)	2 (0, 3)		23.5 (20, 27)		14.5 (12, 17)
Very severe	13 (2.3)	2 (1, 3)		27 (26, 29)		18 (15, 18)
Postoperative epigastric fullness/bloating and abdominal bloating/distension			0.137		0.017		0.011
None	182 (32.8)	1 (0, 3)		25 (21, 30)		15 (12, 18)
Very mild	126 (22.7)	2 (0, 4)		27 (22, 31)		16 (13, 19)
Slight	113 (20.4)	2 (0, 6)		27 (22, 33)		17 (13, 19)
Mild	87 (15.7)	2 (0, 5)		27 (24, 31)		16 (14, 19)
Severe	30 (5.4)	2 (0, 3)		24 (22, 27)		14 (12, 17)
Very severe	17 (3.1)	2 (1, 4)		27 (25, 28)		17 (15, 18)

Two-by-two comparisons: the same letter on the same group indicates a non-significant difference (p>0.05), while different letters indicate a significant difference (p<0.05).

BMI, body mass index.

Deficiencies in patient KAP of postoperative care

Substantial knowledge gaps were identified in postoperative dietary management and disease recognition. Substantial knowledge gaps were identified in postoperative dietary management and disease recognition. For instance, regarding the resumption of a normal diet after surgery (K8), only 34.2% of patients correctly disagreed with the statement that a normal diet could be resumed immediately. Conversely, 13.2% held this misconception, and a majority of patients (52.6%) expressed uncertainty, highlighting a critical area for patient education. Disease awareness remained deficient: merely 23.2% recognised PGS as a common complication (K1), and 24.1% could identify its cardinal symptoms (K2). Diagnostic knowledge was particularly inadequate, with ≤22.3% awareness of gastric emptying scintigraphy (K3) and endoscopy (K4) as confirmatory methods, over 50% responding ‘uncertain’. Therapeutic misconceptions prevailed: although 22.7% acknowledged standard pharmacotherapy (K6), 38.2% erroneously believed recurrence prevention solely required gastric medications (Item 9). Notably, alcohol-related knowledge showed polarised responses (K7: 30.5% Yes vs 38.0% No), whereas ≥50% uncertainty dominated other items, reflecting critical educational needs in perioperative care (online supplemental table S3).

Regarding attitudes towards PGS, 24.1% strongly agreed that a positive and optimistic mindset reduces the duration of morbidity (A1), and 25.6% strongly agreed that weight control is important to reduce the burden on the gastric system (A7). At the same time, 36.4% agreed that the disease was less scary after understanding its characteristics (A3), and 31.4% agreed that taking medication on time was important to control the disease (A6). In addition, 40.9% and 38.6% were neutral on the point that stress and worry can increase the duration of morbidity (A2), and that eating disorders caused by gastroparesis can affect the normal postoperative recovery (A5), respectively (online supplemental table S4).

When it comes to practices related to postoperative rehabilitation, only 16.9% always actively participate in the doctor’s or therapist’s recommendations and receive treatment on time, however, 39.6% do so only sometimes (P1). More notably, only 30.6% of respondents (21.6% often, 9.0% always) ensure eating smaller, more frequent meals (P5). Similarly, 36.7% (27.9% often, 8.8% always) ensure sufficient rest (P2), and 37.7% (28.8% often, 9.9% always) avoid eating spicy and indigestible foods (P3) (online supplemental table S5).

Correlations among KAP dimensions

In the correlation analyses, significant positive correlations were found between knowledge and attitude (r=0.710, p<0.001), knowledge and practice (r=0.731, p<0.001), as well as attitude and practice (r=0.968, p<0.001), respectively (table 2).

Table 2. Correlation analysis of KAP scores.

	Knowledge	Attitude	Practice
Knowledge	1.000	0.710 (p<0.001)	0.731 (p<0.001)
Attitude	0.710 (p<0.001)	1.000	0.968 (p<0.001)
Practice	0.731 (p<0.001)	0.968 (p<0.001)	1.000

KAP, knowledge, attitude and practices.

Multivariate logistic regression of factors influencing KAP

Multivariable logistic regression showed that older age (46–60 years: OR=0.348, 95% CI 0.162 to 0.747, p=0.007; >60 years: OR=0.157, 95% CI 0.073 to 0.339, p<0.001) and rural residency (OR=0.367, 95% CI 0.181 to 0.745, p=0.005) were independently and negatively associated with high knowledge scores. Higher education (junior high school and above) was independently and positively associated with high knowledge (OR=7.105, 95% CI 4.165 to 12.120, p<0.001) (table 3).

Table 3. Multivariate logistic regression analysis of factors independently associated with high knowledge scores regarding postsurgical gastroparesis syndrome (N=555).

Knowledge	Univariate		Multivariate (p<0.1)
	OR (95% CI)	P value	OR (95% CI)	P value
Gender
Male	ref.
Female	0.940 (0.667 to 1.323)	0.721
Age
45 years old and below	ref.		ref.
46–60 years old	0.240 (0.114 to 0.504)	<0.001	0.348 (0.162 to 0.747)	0.007
Above 60 years old	0.054 (0.026 to 0.111)	<0.001	0.157 (0.073 to 0.339)	<0.001
Education
Primary school and below	ref.		ref.
Junior high school/high school/secondary school and above	12.232 (7.633 to 20.453)	<0.001	7.105 (4.165 to 12.120)	<0.001
Marital status
Married	ref.
Other	1.224 (0.706 to 2.122)	0.471
Type of surgery
Partial gastrectomy	ref.
Pancreaticoduodenectomy (Whipple procedure)	0.889 (0.553 to 1.428)	0.626
Gastroduodenostomy	0.758 (0.447 to 1.285)	0.303
Laparoscopic radical gastrectomy for gastric cancer	0.944 (0.540 1.652)	0.840
Colon resection	1.036 (0.519 to 2.069)	0.921
Rectal resection	1.117 (0.543 to 2.296)	0.764
Work status
Employed	ref.		ref.
Other	0.177 (0.119 to 0.265)	<0.001	0.816 (0.499 to 1.335)	0.419
Payment for hospitalisation
Medical insurance and major out-of-pocket expenses	ref.
Medical insurance and minor out-of-pocket expenses	1.230 (0.850 to 1.781)	0.273
Out-of-pocket expenses only	1.250 (0.548 to 2.850)	0.596
Permanent residence
City centre	ref.		ref.
Suburban area	0.430 (0.255 to 0.725)	0.002	0.736 (0.393 to 1.380)	0.340
County townships	0.382 (0.245 to 0.596)	<0.001	0.730 (0.428 to 1.246)	0.249
Rural	0.160 (0.088 to 0.290)	<0.001	0.367 (0.181 to 0.745)	0.005
Diabetes
Yes	ref.
No	1.257 (0.866 to 1.824)	0.229
BMI
Below 18.5	ref.
18.5–25	1.366 (0.903 to 2.066)	0.140
Above 25	1.461 (0.879 to 2.428)	0.144
Postoperative nausea and vomiting
None	ref.
Very mild	0.961 (0.607 to 1.523)	0.867
Slight	1.216 (0.776 to 1.906)	0.393
Mild	1.524 (0.848 to 2.739)	0.159
Severe	0.964 (0.309 to 3.000)	0.949
Very severe	1.084 (0.340 to 3.454)	0.892
Postoperative epigastric fullness/bloating and abdominal bloating/distension
None	ref.
Very mild	1.082 (0.676 to 1.730)	0.743
Slight	1.446 (0.896 to 2.332)	0.131
Mild	1.074 (0.633 to 1.821)	0.791
Severe	0.879 (0.388 to 1.989)	0.757
Very severe	1.230 (0.447 to 3.385)	0.688

BMI, body mass index; ref, reference.

Older age (46–60 years: OR=0.295, p=0.007; >60 years: OR=0.191, p<0.001) and rural residency (OR=0.404, p=0.009) were independently and negatively associated with a positive attitude. Higher education was positively associated with a positive attitude (OR=6.034, p<0.001). Additionally, ‘very mild’ postoperative stomach bloating (OR=1.830, 95% CI 1.029 to 3.255, p=0.040) was also positively associated with a positive attitude (table 4). Older age (46–60 years: OR=0.351, 95% CI 0.149 to 0.826, p=0.017; >60 years: OR=0.160, 95% CI 0.068 to 0.374, p<0.001), county township residency (OR=0.516, 95% CI 0.290 to 0.916, p=0.024), and rural residency (OR=0.357, 95% CI 0.175 to 0.729, p=0.005) were independently and negatively associated with adequate practice scores. Higher education (OR=8.351, 95% CI 4.973 to 14.024, p<0.001), ‘slight’ postoperative nausea and vomiting (OR=1.975, 95% CI 1.058 to 3.687, p=0.033) and ‘very severe’ postoperative nausea and vomiting (OR=8.070, 95% CI 1.476 to 44.107, p=0.016) were independently and positively associated with adequate practice (table 5).

Table 4. Multivariate logistic regression analysis of factors independently associated with high attitude scores regarding postsurgical gastroparesis syndrome (N=555).

Attitude	Univariate		Multivariate (p<0.1)
	OR (95% CI)	P value	OR (95% CI)	P value
Gender
Male	ref.
Female	0.918 (0.658 to 1.281)	0.614
Age
45 years old and below	ref.
46–60 years old	0.209 (0.090 to 0.489)	<0.001	0.295 (0.121 to 0.716)	0.007
Above 60 years old	0.064 (0.028 to 0.145)	<0.001	0.191 (0.079 to 0.464)	<0.001
Education
Primary school and below	ref.
Junior high school	9.458 (6.320 to 14.431)	<0.001	6.034 (3.766 to 9.667)	<0.001
Marital status
Married	ref.
Other	1.484 (0.857 to 2.568)	0.159
Type of surgery
Partial gastrectomy	ref.
Pancreaticoduodenectomy (Whipple procedure)	1.014 (0.639 to 1.608)	0.954
Gastroduodenostomy	0.772 (0.464 to 1.286)	0.321
Laparoscopic radical gastrectomy for gastric cancer	0.959 (0.555 to 1.657)	0.880
Colon resection	1.441 (0.726 to 2.858)	0.296
Rectal resection	2.134 (1.012 to 4.497)	0.046
Work status
Employed	ref.
Other	0.183 (0.120 to 0.279)	<0.001	1.097 (0.683 to 1.762)	0.702
Payment for hospitalisation
Medical insurance and major out-of-pocket expenses	ref.
Medical insurance and minor out-of-pocket expenses	1.372 (0.958 to 1.966)	0.084	1.142 (0.735 to 1.773)	0.556
Out-of-pocket expenses only	1.230 (0.548 to 2.762)	0.615	1.153 (0.434 to 3.059)	0.776
Permanent residence
City centre	ref.
Suburban area	0.404 (0.237 to 0.687)	0.001	0.647 (0.339 to 1.237)	0.188
County townships	0.340 (0.215 to 0.537)	<0.001	0.600 (0.343 to 1.048)	0.073
Rural	0.180 (0.103 to 0.315)	<0.001	0.404 (0.205 to 0.796)	0.009
Diabetes
Yes	ref.
No	1.059 (0.739 to 1.517)	0.754
BMI
Below 18.5	ref.
18.5–25	1.122 (0.755 to 1.667)	0.569
Above 25	1.135 (0.694 to 1.856)	0.613
Postoperative nausea and vomiting
None	ref.
Very mild	1.688 (1.079 to 2.639)	0.022	1.622 (0.912 to 2.885)	0.100
Slight	1.753 (1.126 to 2.731)	0.013	1.708 (0.941 to 3.102)	0.079
Mild	1.911 (1.064 to 3.431)	0.030	1.939 (0.854 to 4.403)	0.113
Severe	1.180 (0.392 to 3.550)	0.768	2.133 (0.433 to 10.515)	0.352
Very severe	1.836 (0.592 to 5.694)	0.293	3.606 (0.765 to 17.002)	0.105
Postoperative epigastric fullness/bloating and abdominal bloating/distension
None	ref.		ref.
Very mild	1.910 (1.205 to 3.026)	0.006	1.830 (1.029 to 3.255)	0.040
Slight	1.611 (1.003 to 2.587)	0.048	0.979 (0.532 to 1.802)	0.945
Mild	1.880 (1.122 to 3.152)	0.017	1.381 (0.667 to 2.859)	0.385
Severe	0.764 (0.338 to 1.726)	0.517	0.437 (0.145 to 1.311)	0.140
Very severe	1.719 (0.634 to 4.661)	0.287	0.973 (0.230 to 4.109)	0.970

BMI, body mass index; ref, reference.

Table 5. Multivariate logistic regression analysis of factors independently associated with high practice scores regarding postsurgical gastroparesis syndrome (N=555).

Practice	Univariate		Multivariate (p<0.1)
	OR (95% CI)	P value	OR (95% CI)	P value
Gender
Male	ref.
Female	0.989 (0.707 to 1.383)	0.948
Age
45 years old and below	ref.
46–60 years old	0.257 (0.115 to 0.576)	0.001	0.351 (0.149 to 0.826)	0.017
Above 60 years old	0.055 (0.025 to 0.120)	<0.001	0.160 (0.068 to 0.374)	<0.001
Education
Primary school and below	ref.		ref.
Junior high school	14.465 (9.205 to 23.520)	<0.001	8.351 (4.973 to 14.024)	<0.001
Marital status
Married	ref.
Other	1.494 (0.866 to 2.577)	0.149
Type of surgery
Partial gastrectomy	ref.
Pancreaticoduodenectomy (Whipple procedure)	0.837 (0.526 to 1.333)	0.455
Gastroduodenostomy	0.653 (0.389 to 1.098)	0.108
Laparoscopic radical gastrectomy for gastric cancer	0.936 (0.541 to 1.619)	0.812
Colon resection	1.215 (0.615 to 2.399)	0.576
Rectal resection	1.222 (0.599 to 2.492)	0.582
Work status
Employed	ref.
Retirement	0.154 (0.101 to 0.235)	<0.001	0.836 (0.504 to 1.385)	0.486
Payment for hospitalisation
Medical insurance and major out-of-pocket expenses	ref.
Medical insurance and minor out-of-pocket expenses	1.149 (0.802 to 1.648)	0.449
Out-of-pocket expenses only	0.931 (0.410 to 2.116)	0.865
Permanent residence
City centre	ref.
Suburban area	0.369 (0.217 to 0.626)	<0.001	0.553 (0.283 to 1.081)	0.083
County townships	0.316 (0.201 to 0.498)	<0.001	0.516 (0.290 to 0.916)	0.024
Rural	0.174 (0.099 to 0.305)	<0.001	0.357 (0.175 to 0.729)	0.005
Diabetes
Yes	ref.
No	1.203 (0.837 to 1.730)	0.317
BMI
Below 18.5	ref.
18.5–25	1.171 (0.785 to 1.746)	0.439
Above 25	1.281 (0.781 to 2.101)	0.326
Postoperative nausea and vomiting
None	ref.		ref.
Very mild	1.567 (0.999 to 2.460)	0.051	1.522 (0.831 to 2.789)	0.174
Slight	1.826 (1.169 to 2.854)	0.008	1.975 (1.058 to 3.687)	0.033
Mild	1.667 (0.928 to 2.995)	0.088	1.964 (0.836 to 4.615)	0.122
Severe	0.711 (0.214 to 2.361)	0.578	1.129 (0.189 to 6.729)	0.894
Very severe	2.844 (0.892 to 9.066)	0.077	8.070 (1.476 to 44.107)	0.016
Postoperative epigastric fullness/bloating and abdominal bloating/distension
None	ref.
Very mild	1.624 (1.025 to 2.574)	0.039	1.487 (0.810 to 2.728)	0.200
Slight	1.832 (1.138 to 2.948)	0.013	1.109 (0.585 to 2.103)	0.752
Mild	1.494 (0.891 to 2.506)	0.128	0.942 (0.441 to 2.010)	0.877
Severe	0.610 (0.257 to 1.445)	0.261	0.307 (0.093 to 1.014)	0.053
Very severe	1.886 (0.695 to 5.120)	0.213	0.983 (0.210 to 4.600)	0.983

BMI, body mass index; ref, reference.

Discussion

Our study highlights that patients who have undergone abdominal surgery exhibit inadequate knowledge, negative attitudes and passive practices regarding key aspects of postoperative care for PGS, particularly dietary management, symptom recognition and adherence to recommended behaviours.

The present study offers valuable insights into the KAP of patients’ post-abdominal surgery concerning the improvement of PGS. These outcomes align with previous research highlighting the challenges patients face in comprehending and adhering to postoperative care recommendations.²⁴ Interestingly, the absence of significant gender-based differences contrasts with some studies suggesting gender-specific variations in health-related behaviours.²⁵ Consistent with literature linking higher socioeconomic status to better health resource access,²⁶ the observed advantages among employed and urban-dwelling patients emphasise the necessity of targeted interventions for rural and unemployed populations. Educational campaigns tailored to these groups, potentially delivered through accessible channels, could bridge the knowledge gap and improve attitudes and practices. Moreover, the disparities in age groups highlight the necessity of age-specific interventions.^{27 28} Younger individuals may benefit from more interactive and technology-driven educational approaches, while older individuals might prefer traditional formats.^{29 30} Collaborative efforts with healthcare providers and community organisations can enhance the effectiveness of these interventions.^{31 32}

The strong positive correlations across all KAP dimensions highlight their interconnectedness, reinforcing the necessity for holistic interventions. Furthermore, identifying independent predictors of poor KAP scores provides actionable insights for designing these tailored strategies. Addressing the educational needs of individuals with lower education levels is crucial. Implementing targeted educational programmes, perhaps incorporating multimedia elements to enhance comprehension, could effectively bridge this gap.^{33 34} Furthermore, the association between postoperative symptoms and attitudes and practices suggests the importance of symptom management in improving patient engagement. Interventions focusing on symptom alleviation, supported by healthcare professionals, may positively impact attitudes and practices.^{35 36}

While studying patients with a confirmed diagnosis of PGS would offer more direct clinical insights, the practical challenge of recruiting a sufficient number of such patients limited the feasibility of that approach. Therefore, this study focused on patients who had undergone abdominal surgery—an at-risk population that may experience symptoms resembling gastroparesis, such as delayed gastric emptying, nausea or bloating, even without a formal diagnosis. The primary objective was not to assess disease experience per se, but to evaluate risk awareness, preventive behaviours, and informational gaps related to this common postoperative complication. Understanding how this broader surgical population perceives and responds to PGS-related risks is critical for informing early-stage education, improving symptom management, and supporting the development of proactive perioperative care strategies.

Participants exhibited notable knowledge gaps regarding postoperative dietary management, symptom recognition, and alcohol consumption risks. Addressing these specific misconceptions through targeted educational interventions is crucial for improving postoperative care and patient outcomes. Considering the effectiveness of targeted education, implementing informative sessions and distributing educational materials can enhance patients’ understanding of postoperative care, promoting better outcomes.^{33 37} Indeed, the high proportion of ‘uncertain’ responses to questions involving diagnostic tools and terminology such as ‘gastroscopy’ or ‘gastric emptying scintigraphy’ highlights a significant knowledge gap and communication barrier between healthcare professionals and patients. While this may reflect the complexity of certain questionnaire items, it more importantly underscores the critical need for clear, accessible health education using patient-friendly language and visual aids. This finding reinforces the study’s conclusion that enhanced education is urgently required to support patients’ understanding and engagement in their postoperative care. Moreover, although the professional terms in the questionnaire were simplified during pretesting, their inclusion still allowed the study to capture the extent of patients’ unfamiliarity with essential medical concepts. This, in turn, highlighted specific areas where targeted health education is most urgently needed.

Participants’ attitudes towards postoperative recovery presented a nuanced perspective. Of significance were concerns related to psychological aspects, including anxiety about disease characteristics and apprehensions about the prolonged recovery time. This resonates with findings by Bian et al, who identified psychological distress as a common concern among postoperative patients.³⁸ However, participants also exhibited positive attitudes, recognising the importance of medication adherence and dietary modifications for effective recovery. Tailoring support programmes to address anxiety-related concerns can contribute to improved psychological well-being during the recovery process.^{39 40} Collaborative efforts involving healthcare providers, mental health professionals, dietetics, and patient support groups can play a pivotal role in fostering a positive attitude and enhancing overall postoperative experiences.^{41 42}

The examination of participants’ postoperative practices uncovered variations in adherence to recommended behaviours. Noteworthy was the diversity in responses, ranging from consistent adherence to dietary guidelines to sporadic avoidance of specific foods. Participants reported challenges in actively participating in prescribed treatments and ensuring adequate rest for postoperative recovery. This highlights a critical shortfall in implementing essential recovery practices, such as sufficient rest and appropriate dietary management before and after surgery.⁴³ Addressing these challenges necessitates improved nursing care, close supervision and enhanced interactions between nurses and patients preprocedure and postprocedure. Specifically, proactive preoperative interactions are essential to prepare patients psychologically and behaviourally, thereby significantly facilitating postoperative recovery. To bridge this gap, practical and context-specific recommendations are essential. Implementing personalised care plans, involving patients in goal-setting, and providing continuous support through telehealth initiatives are potential strategies. Such interventions, designed to improve both dietary management and rest periods, can enhance patient engagement and contribute to the effective implementation of postoperative practices, ultimately improving the overall quality of care.⁴⁴

Despite the valuable insights gained from our study, certain limitations must be acknowledged. First, this study used a cross-sectional design, which limits the ability to assess changes in KAP over time or establish causal relationships. Second, data were collected from a single tertiary care centre in China, which may limit the generalisability of the findings to broader populations. Third, the survey was conducted at a single postoperative time point, and may not reflect long-term practices or changes in perception. Additionally, the reliance on self-reported data through questionnaires may introduce response bias, as participants might provide socially desirable answers rather than reflecting their true experiences. Finally, there are inherent translation limitations regarding the nuances of clinical terminology. For instance, the terms slight and mild denote distinct clinical perceptions in the original Chinese context: slight refers to symptoms that are barely noticeable and do not affect activities, whereas mild indicates symptoms that are perceptible but do not completely interfere with daily life or postoperative recovery. Furthermore, to better align with clinical descriptions for English readers, the original Chinese terms for stomach and abdominal discomfort were optimised in translation to epigastric fullness/bloating and abdominal bloating/distension. Finally, there are inherent translation limitations regarding the nuances of clinical symptom severity. For example, the terms ‘slight’ and ‘mild’ used in the questionnaire denote distinct clinical perceptions in the original Chinese context: ‘slight’ refers to symptoms that are barely noticeable and do not affect activities, whereas ‘mild’ indicates symptoms that are perceptible but do not completely interfere with daily life or postoperative recovery. Despite these limitations, this paper’s strength lies in its comprehensive exploration of demographic factors associated with KAP levels and the identification of independent variables influencing patient perspectives. The large sample size, rigorous statistical analyses, and the inclusion of multivariate analyses enhance the robustness of our findings and contribute valuable information for targeted interventions to improve patient care in the context of PGS post-abdominal surgery. Although the survey was conducted only postoperatively, this was consistent with the study’s objective to understand patients’ perspectives during the recovery period following abdominal surgery. Moreover, by collecting data on postoperative symptoms such as nausea, vomiting, and abdominal distension, the study enabled comparative analyses across different symptom severity levels, partially addressing the absence of a preoperative control group. It is also important to note that patients who had undergone total gastrectomy were excluded from the study. Therefore, the findings are specifically applicable to individuals who underwent partial gastric resections, in whom the risk of delayed gastric emptying remains clinically relevant.

To translate these findings into clinical practice, several approaches are recommended. First, integrating KAP assessment into routine perioperative care could help identify high-risk patient groups (eg, rural residents, those with low education levels) who may benefit from targeted education. Second, educational interventions should be adapted to the literacy level and cognitive needs of different populations, using plain language, visual tools and culturally appropriate examples. Third, postoperative nursing pathways can be revised to include KAP-informed communication protocols that proactively address patients’ misconceptions and encourage positive behavioural change. Lastly, establishing feedback loops through follow-up visits or digital tools could reinforce key messages and improve long-term adherence.

In conclusion, our study highlights significant gaps in KAP regarding PGS among patients post-abdominal surgery. These individuals frequently exhibit insufficient understanding of dietary management and symptom recognition, hold attitudes negatively impacted by postoperative discomfort, and struggle with adhering to recovery guidelines. Notably, vulnerable groups—specifically the elderly, rural residents, the unemployed and those with lower educational levels—consistently exhibited significantly lower KAP scores. These findings provide concrete evidence highlighting the urgent need for precision health education and patient-centred care strategies tailored to these high-risk populations.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.