Development and Psychometric Evaluation of a Knowledge, Attitude, and Practice Questionnaire Regarding Sexual Health Care Among Gynecologic Oncology Nurses

1. Introduction

The global burden of gynecologic malignancies has been rising steadily over the past several decades. In 2022, 1.47 million women worldwide were newly diagnosed with gynecologic cancers. With continuous advances in early cancer screening, precision diagnosis and treatment, and rehabilitation support systems, the survival of patients with gynecologic malignancies has improved markedly, with the 5-year survival rates for cervical cancer, endometrial cancer, and ovarian cancer reaching 62%, 81%, and 40%, respectively [1]. Owing to cancer location, the extent of surgery, damage caused by radiotherapy and chemotherapy, and changes in hormone levels, the rate of sexual dysfunction among patients diagnosed with gynecologic malignancies is as high as 93.7%, which is substantially higher than that among patients with other types of cancer [2]. Increasing attention has been paid by the medical community and society to patients’ quality of life, mental health, and sexual health. As a key dimension of overall health and rehabilitation, sexual health directly affects family functioning, social adaptation, and long-term prognosis [3,4]. The vast majority of patients with gynecologic cancers have sexual health care needs, and healthcare professionals should proactively assess these needs, provide scientific guidance, and deliver standardised sexual health support [4,5].

As integral participants of the oncology care team, nurses serve as frontline providers in sexual health assessment, patient-provider communication and health education, rehabilitation interventions, and interdisciplinary care linkage [6,7,8]. The Knowledge, Attitude, and Practice (KAP) theoretical framework holds that knowledge forms the foundation for behavioural change, attitude functions as the intrinsic impetus behind behavioural transformation, and practice is the ultimate goal of translating knowledge and beliefs into action [9]. Oncology nurses’ knowledge base, attitudes, and practical competence in sexual health nursing directly determine the quality and accessibility of sexual health care in clinical practice [10]. However, problems such as insufficient knowledge, conservative attitudes, and inadequate practice in sexual health nursing are widespread among gynaecologic oncology nurses, and there remains a lack of unified, scientific, and highly targeted evaluation instruments, making it difficult to comprehensively evaluate their knowledge, attitudes, and practice regarding sexual health care [11]. Therefore, guided by the KAP theoretical framework and using a combination of a systematic literature review, semi-structured interviews, and the Delphi expert consultation method, the present study developed a Knowledge, Attitude, and Practice (KAP) Questionnaire on Sexual Health Care for Gynaecologic Oncology Nurses and evaluated its reliability and validity. This study aimed to establish a rigorous and dependable instrument for the scientific evaluation of gynaecologic oncology nurses’ sexual health care competence, the targeted implementation of specialist training, and the promotion of standardized clinical sexual health care.

2. Methods

2.1. Establishment of the Study Group

The study group comprised two master’s supervisors in nursing, one chief nurse, one associate chief nurse, two postgraduate nursing students, and two clinical nurses. The team’s core tasks involved carrying out a literature review, conducting semi-structured interviews, developing, distributing, and retrieving expert consultation questionnaires, integrating expert feedback, and performing data analysis and statistical processing.

2.2. Construction of the Questionnaire Item Pool

2.2.1. Theoretical Basis

The KAP (Knowledge, Attitude, and Practice) framework maintains that knowledge provides the necessary foundation, constructive attitudes and beliefs act as internal drivers, and the adoption and evolution of practice behaviors mark the final objective [10]. Based on this theoretical framework, the research team organized the questionnaire into three dimensions: knowledge, attitude, and practice. (1) Knowledge: this dimension assessed gynecologic oncology nurses’ understanding of basic theoretical knowledge related to sexual health, manifestations of sexual dysfunction, and factors influencing sexual health in cancer patients. (2) Attitude: this dimension assessed attitudes toward sexual health issues among cancer patients, attitudes toward the provision of sexual health care, and self-perceived role awareness in sexual health care practice. (3) Practice: this dimension assessed gynecologic oncology nurses’ competence in the assessment, management, and nursing practice of sexual health issues in cancer patients.

2.2.2. Literature Review

The search terms included “nurse/health caregiver”, “cancer*/tumor*/neoplasms”, and “sexuality/sexual dysfunction/sexual adjustment/sexual life/sexual behavior/sexual problem”. A systematic search was conducted in the following databases: PubMed, Embase, The Cochrane Library, Web of Science, China National Knowledge Infrastructure, Wanfang Data, VIP Database, and SinoMed. In addition, the reference lists of included studies were manually screened for supplementary retrieval. The search period covered database inception to August 2024. Based on the KAP theoretical model, the included literature was reviewed and analyzed to construct the initial structural framework and item pool of the questionnaire. The preliminary questionnaire comprised three dimensions—knowledge (12 items), attitude (13 items), and practice (17 items)—with a total of 42 items.

2.2.3. Semi-Structured Interviews

We adopted purposive sampling to recruit 12 nursing professionals from a tertiary A-level oncology-specialized hospital in Tianjin, and conducted semi-structured interviews with them in September 2024. Participants satisfying the following criteria were included: (1) possess a current registered nurse license; (2) being a gynecologic oncology nurse or a nurse with experience in caring for patients with gynecologic malignancies; (3) have accumulated at least 1 year of work experience in the field; and (4) agreeing to participate in this research. Individuals were excluded if they met the following criteria: (1) nurses currently in training programs or internships; or (2) staff who were absent from work due to sick leave.

The interview guide included the following questions: (1) When providing sexual health-related care services for patients diagnosed with gynecologic cancers, what knowledge and theories do you consider essential to understand and master? (2) What are your attitudes, views, feelings, and experiences regarding sexual health issues and sexual health care provided to patients with gynecologic malignancies? (3) What do you consider to be the basic content and core competencies required for sexual health care provided to patients with gynecologic malignancies? (4) What suggestions do you have regarding sexual health care practice?

Drawing on the literature review and qualitative interview results, the research group iteratively refined and revised the measurement tool. The initial version of the scale included three core dimensions: knowledge (12 items), attitude (12 items), and practice (20 items), with a total of 44 items.

2.3. Delphi Expert Consultation

Purposive sampling was used to conduct Delphi expert consultation from September to October 2024. Experts satisfying the following criteria were included: (1) primary clinical or research experience in gynecologic oncology medicine, nursing, or sexual health medicine of no less than 10 years; (2) having obtained a bachelor’s degree or above; (3) holding an intermediate-level professional title or higher; (4) familiarity with the Delphi consultation process and questionnaire development procedures; and (5) agreeing to take part in this research.

The Delphi expert consultation questionnaire was structured into three sections: an instruction sheet, a form collecting experts’ basic information, and the consultation rating form. The instruction sheet included the research background, objectives, significance, and guidance on how to complete the questionnaire. A basic information form gathered details on demographic and professional characteristics, including age, educational level, and years of work experience, as well as experts’ self-rated familiarity (Cs) and judgment basis (Ca). The consultation form employed a 5-point Likert scale for assessing the significance of all dimensions and items within the questionnaire.

Questionnaires were sent out and gathered through both email and on-site administration. The consultation process was terminated when expert opinions tended to converge, and the initial version of the Knowledge, Attitude, and Practice Questionnaire on Sexual Health Care for Gynecologic Oncology Nurses was then finalized. The selection criteria for items included: a mean importance score > 3.5, a coefficient of variation < 0.35, and consensus criteria including ≥ 70% agreement [12].

2.4. Pilot Survey

Using convenience sampling, gynecologic oncology nursing staff from a tertiary A-level oncology-specialized hospital in Tianjin were recruited in October 2024 for a pilot survey. The criteria for participant inclusion and exclusion remained identical to those applied in the semi-structured interviews. By asking participants about their understanding of each questionnaire item and their response selection process, the wording of the questionnaire was further refined to improve clarity, readability, and accuracy.

2.5. Formal Survey

Using convenience sampling, gynecologic oncology nursing staff from nine hospitals across six provincial-level regions in China, including Tianjin, Shanghai, Hebei, Xinjiang, Hunan, and Hainan, were recruited between November and December 2024. Participants were divided into two independent groups, with one subset designated for exploratory factor analysis and the other for confirmatory factor analysis. Participant inclusion and exclusion criteria remained identical to those applied in the semi-structured interviews.

According to the principles of scale development, the required sample size for exploratory factor analysis is 5 to 10 times the number of items, whereas for confirmatory factor analysis, the sample size should be larger than that used in exploratory factor analysis and also exceed 200 [13]. The trial version of the questionnaire contained 42 items. Considering a projected 10% proportion of unusable responses, the sample size needed for this investigation was determined to be at least 462 individuals. Ethical clearance was obtained from the Institutional Review Board of Tianjin Medical University Cancer Institute and Hospital (bc20250246) on 15 January 2025.

The survey instruments comprised a general information collection form and the trial version of the Knowledge, Attitude, and Practice Questionnaire on Sexual Health Care for Gynecologic Oncology Nurses. The general information questionnaire collected data on sex, age, length of work experience, educational attainment, professional rank, position, whether the participant was a specialist nurse, and department. The pilot questionnaire employed a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) for all items. Higher summed scores reflected better knowledge and more positive attitudes and practices regarding sexual health care.

With permission secured from the nursing managers of each hospital, trained members of the research team distributed the questionnaire link to participants via an online survey platform. The opening section detailed the research goals and offered guidance on how to fill out the survey. Submission of the questionnaire was permitted only after every item had been answered. After the submission deadline, the data were exported in a timely manner and cross-checked independently by two researchers. Questionnaires completed in less than 1 minute or showing obvious response patterns or logical inconsistencies were excluded.

Reliability and validity testing of the scale

Item analysis (1) Critical ratio method: Questionnaire responses were sorted in descending order based on their total scores. The highest-scoring 27% were designated as the high-score group, and the lowest-scoring 27% as the low-score group. Differences between the two groups were evaluated via an independent-samples t test. Only items that met the criteria of a critical ratio ≥ 3 or a significance level of P ≤ 0.05 were retained in the questionnaire. (2) Correlation coefficient method: Pearson correlation coefficients were computed between each item and the overall questionnaire score. Items that met the criteria of a correlation coefficient > 0.4 or a p-value ≤ 0.05 were kept for inclusion in the final scale. (3) Cronbach’s α coefficient method: Items were assessed based on their impact on the scale’s reliability. Any item that, when deleted, caused the overall Cronbach’s α value to rise was removed from the questionnaire [14].

Validity testing (1) Content validity: Content validity was evaluated by calculating the item-level content validity index (I-CVI) and the scale-level content validity index (S-CVI). An I-CVI ≥ 0.78 and an S-CVI ≥ 0.80 were deemed to demonstrate adequate content validity.(2) Construct validity: Exploratory factor analysis was conducted using principal component analysis with varimax rotation. The criteria were as follows: eigenvalue > 1, cumulative variance explained > 50%, and a factor loading > 0.40 for each item on a single factor. Confirmatory factor analysis was performed using the maximum likelihood method. The evaluation criteria were standardized factor loadings > 0.45, chi-square/degrees of freedom (χ²/df) < 3, root mean square error of approximation (RMSEA) < 0.08, and goodness-of-fit index (GFI), comparative fit index (CFI), incremental fit index (IFI), and normed fit index (NFI) all > 0.90, indicating an acceptable model fit.(3) Convergent validity: Convergent validity was considered acceptable when the average variance extracted (AVE) was > 0.50 and the composite reliability (CR) was > 0.70.(4) Discriminant validity: Discriminant validity was considered acceptable when the square root of the AVE for each factor exceeded the correlation coefficient between that factor and any other factor [14].

Reliability testing (1) Cronbach’s α coefficient: A Cronbach’s α coefficient ≥ 0.70 for each dimension and for the overall questionnaire, together with no substantial increase in Cronbach’s α after deletion of any individual item, was considered indicative of good internal consistency. (2) Split-half reliability: Split-half reliability for each dimension and for the overall questionnaire was calculated using the odd-even split method. (3) Test–retest reliability: A sample of 50 nurses working in gynecologic oncology units was randomly selected to retake the questionnaire after a 2-week period. The test-retest reliability was deemed satisfactory if the correlation coefficient > 0.70 [14].

2.6. Statistical Analysis

All data were double-entered and analyzed by SPSS version 27.0 and AMOS version 26.0. Categorical data were presented as frequencies and percentages. Continuous variables following a normal distribution were reported as the mean ± standard deviation, whereas those with a non-normal distribution were expressed using the median and interquartile range. The degree of expert participation was calculated from the effective questionnaire response rate. The authority coefficient (Cr) served as the measure for evaluating expert authority, variability was described by the coefficient of variation, and the degree of coordination was assessed using Kendall’s coefficient of concordance. A two-sided P < 0.05 was considered statistically significant.

3. Results

3.1. Results of the Delphi Expert Consultation

The Delphi consultation was conducted in two rounds. For the first round, questionnaires were sent to 16 experts, with 15 valid responses returned (93.75%). The second round of the survey achieved full participation, with a valid response rate of 100%.

The 15 experts were enrolled from tertiary grade A general hospitals, specialized hospitals, and medical universities in multiple provinces and municipalities, including Beijing, Tianjin, Hebei, and Hainan. Their average age was 42.50 ± 2.00 years, with a mean of 17.50 ± 1.80 years of professional experience. Among them, 5 were engaged in gynecologic oncology medicine, 9 in gynecologic oncology nursing, and 1 was a specialist in sexual medicine. In terms of educational level, 6 possessed doctoral degrees, 6 held master's degrees, and 3 had bachelor's degrees. Regarding professional title, 5 held senior professional titles and 10 held associate senior-level titles. In the two consultation rounds, the expert authority coefficients were 0.862 and 0.897, respectively. The coefficients of variation for the items ranged from 0 to 0.211 in the first round and from 0 to 0.196 in the second round. The Kendall’s W values for the two rounds were 0.261 and 0.314, respectively, both reaching statistical significance (P < 0.001).

Using the item selection criteria and input from experts, the questionnaire items were revised. Specifically, after the first round of expert consultation, two items were removed, three new items were introduced, six items were combined, and the phrasing of eleven items was adjusted. Following the second round, no items were removed or added, while four items underwent wording refinements. After these two rounds of expert consultation, the initial questionnaire comprised three dimensions—knowledge (12 items), attitude (10 items), and behavior (20 items)—with a total of 42 items.

3.2. Results of the Pilot Survey

In the pilot survey, 20 questionnaires were sent out. Each participant completed the questionnaire in 5–10 minutes and indicated no difficulty in reading or understanding the items. Consequently, no items were removed, added, or revised.

3.3. General Characteristics of the Participants

530 questionnaires were sent out, and 512 valid ones were retrieved, corresponding to a valid return rate of 96.60%. The valid questionnaires were then randomly split into two subsamples, 245 for item analysis and exploratory factor analysis, and 272 for confirmatory factor analysis and criterion-related validity analysis.

Among the 512 participants, their ages spanned from 24 to 54 years, with an average of 35.35 ± 7.10 years. Regarding educational background, 67 participants (13.09%) held a junior college diploma, 442 (86.33%) possessed a bachelor’s degree, and 3 (0.59%) had a master’s degree or higher. As for professional title,75 (14.65%) were nurses, 256 (50.00%) were senior nurses, 169 (33.01%) were nurse-in-charge, 7 (1.37%) were associate chief nurses, and 5 (0.98%) were chief nurses. A total of 103 participants (20.12%) were oncology specialist nurses. Their work experience spanned 2 to 34 years, averaging 12.09 ± 7.75 years.

3.4. Item Analysis

The results of item analysis were as follows. (1) Critical ratio method: For all items in the 245 questionnaires, the differences between the high-group and low-scoring group were statistically significant. The t values for all items were > 3, and all P values were < 0.01. Consequently, none of the items were deleted. (2) Correlation coefficient method: The correlation coefficients for all items were > 0.4, and all P values were < 0.01. (3) Internal consistency coefficient method: After deleting each item one by one, the overall Cronbach’s α coefficient did not increase substantially. Taken together, all 42 items were retained.

3.5. Validity Testing

3.5.1. Content Validity

The item-level content validity index (I-CVI) fell between 0.889 and 1.00, while the scale-level content validity index (S-CVI) was 0.956, reflecting satisfactory content validity for the questionnaire.

3.5.2. Construct Validity

Exploratory factor analysis: The results of the first exploratory factor analysis showed that the Kaiser-Meyer-Olkin (KMO) value was 0.867, and Bartlett’s test of sphericity produced a χ² value of 5112.80 (P < 0.001), suggesting that the data were suitable for factor analysis. By employing principal component analysis with orthogonal rotation, four common factors with eigenvalues exceeding 1 emerged, cumulatively accounting for 71.437% of the variance. Two items showed cross-loadings and were therefore deleted.

In the second exploratory factor analysis, the KMO value was 0.926, and Bartlett’s test of sphericity produced a χ² value of 5888.751 (P < 0.001). Three common factors with eigenvalues > 1 were extracted, with a cumulative variance explained of 73.288%. Factor loadings for the retained items ranged from 0.578 to 0.928, and no cross-loading was observed; therefore, no further items were deleted. The extracted common factors were consistent with the predefined dimensions. The three factors were named knowledge (Items 1-10), attitude (Items 11-20), and practice (Items 21-40), comprising a total of 40 items (Table 1).

Confirmatory factor analysis: The results showed that the model had a χ²/df of 2.538, a root mean square error of approximation (RMSEA) of 0.067, a goodness-of-fit index (GFI) of 0.917, a comparative fit index (CFI) of 0.956, an incremental fit index (IFI) of 0.926, and a non-normed fit index (NNFI) of 0.934. These findings suggested a satisfactory model fit and a robust factor structure for the questionnaire.

3.6. Convergent Validity and Discriminant Validity

The average variance extracted (AVE) values for the three dimensions were 0.832, 0.573, and 0.653, respectively, and the composite reliability (CR) values were 0.972, 0.930, and 0.974, respectively, indicating satisfactory convergent validity of the questionnaire. In addition, the square root of the AVE for each dimension was greater than the correlations between factors, suggesting that the items in each dimension accurately reflected the construct intended to be measured. These results demonstrated good discriminant validity of the questionnaire (Table 2).

3.7. Reliability Testing

The overall Cronbach’s α coefficient of the questionnaire was 0.980, and the Cronbach’s α coefficients for the three dimensions were 0.970, 0.927, and 0.973, respectively. The overall split-half reliability coefficient was 0.919, and the split-half reliability coefficients for the three dimensions were 0.981, 0.935, and 0.955, respectively. The overall test–retest reliability coefficient was 0.982, and the test–retest reliability coefficients for the three dimensions were 0.976, 0.923, and 0.975, respectively.

4. Discussion

In this study, the questionnaire was constructed using the Knowledge, Attitude, and Practice (KAP) theory as the guiding conceptual. As an established theory in the field of behavior change, the KAP model clarifies the progressive relationship among knowledge, attitude, and practice, thereby providing a solid theoretical basis for the dimensional structure of the questionnaire [9]. In the process of developing the questionnaire, the research team performed a thorough literature review to systematically identify relevant domestic and international evidence on sexual health care provided to patients with gynecologic malignancies, thereby ensuring that the questionnaire items covered core domains such as basic sexual health knowledge, attitudes toward care, and practical skills. In addition, semi-structured interviews with frontline nursing staff were conducted to thoroughly explore the needs and challenges encountered in clinical practice, allowing the questionnaire items to better reflect real-world nursing contexts. Furthermore, two consecutive Delphi rounds involving senior experts in gynecologic oncology medicine, nursing, and sexual medicine were conducted. The expert authority coefficients were both greater than 0.85, and Kendall’s coefficients of concordance were statistically significant, indicating good agreement among experts and further supporting the scientific rigor and professional relevance of the questionnaire content.

The reliability analysis revealed that the overall Cronbach’s α coefficient for the questionnaire was 0.980, and the Cronbach’s α values for all dimensions exceeded 0.92, reflecting excellent internal consistency. The overall split-half reliability coefficient was 0.919, with dimension-level coefficients all above 0.93, suggesting good stability of the questionnaire items. The overall test–retest reliability coefficient was 0.982, and the test–retest reliability coefficients for all dimensions exceeded 0.92, indicating strong reproducibility of the measurement results and minimal influence of time-related factors. With respect to validity, the scale-level content validity index (S-CVI) was 0.956, and the item-level content validity indices (I-CVIs) ranged from 0.889 to 1.00, indicating that the questionnaire items were highly relevant to the study objectives and adequately reflected the core constructs of gynecologic oncology nurses’ knowledge, attitudes, and practices regarding sexual health care [15]. Exploratory factor analysis successfully extracted three common factors, which were fully consistent with the predefined dimensions of knowledge, attitude, and practice. The cumulative variance explained reached 73.288%, and item factor loadings ranged from 0.578 to 0.928, with no cross-loadings observed. Confirmatory factor analysis further showed that the model fit indices were acceptable, with χ²/df = 2.538 and RMSEA = 0.067, while GFI, CFI, and other fit indices were all above 0.91, indicating that the questionnaire had a stable structure and good fit with the theoretical framework [16]. In addition, the average variance extracted (AVE) values for all dimensions were greater than 0.50, the composite reliability (CR) values were all above 0.93, and the square root of the AVE for each dimension exceeded the inter-factor correlations. These findings indicate that the questionnaire had satisfactory convergent validity and discriminant validity and was able to effectively distinguish among different measurement domains.

The questionnaire was structured around three dimensions—knowledge, attitude, and practice—and consisted of 40 items in total, featuring comprehensive content and a clear focus. The knowledge dimension covered core topics such as basic sexual health theory, the influence of cancer treatments on sexual function, and the assessment of sexual dysfunction, enabling a comprehensive evaluation of nurses’ professional knowledge reserves [17]. The attitude dimension focused on nurses’ perceptions of sexual health care, their acceptance of patients’ sexual health needs, and their self-perceived professional role, reflecting the intrinsic motivation to provide such care [18,19]. The practice dimension encompassed practical components such as sexual health assessment, communication skills, implementation of interventions, and continuity of care, thereby comprehensively evaluating nurses’ clinical practice [20]. This survey instrument was tailored to the unique working characteristics of nurses in gynecologic oncology settings. The item phrasing was straightforward and comprehensible, the 5-point Likert scale was easy to administer, and the questionnaire took only a short time to complete, making it appropriate for large-scale survey studies [21,22,23]. In addition, the questionnaire has broad applicability and may be used to evaluate the knowledge, attitudes, and practices concerning sexual health care of gynecologic oncology nurses with different professional ranks and years of work experience. It may also provide a scientific basis for healthcare institutions to develop targeted training programs and serve as a standardized measurement tool for related research, highlighting its important practical value.

In contrast to existing general oncology nursing tools, this questionnaire is uniquely tailored to the specific clinical context of gynecologic malignancies, covering sexual dysfunction mechanisms, treatment-related impacts, communication skills, rehabilitation guidance, and continuity of care—domains underrepresented in previous instruments [24,25,26]. Differences from international scales may reflect cultural variations in attitudes toward sexual health, institutional care protocols, and specialty-specific nursing roles; the present tool was developed using domestic clinical evidence, frontline interviews, and multidisciplinary expert consensus to enhance contextual and cultural relevance. In clinical practice, the questionnaire supports systematic identification of knowledge deficits, attitude barriers, and practice gaps, enabling nursing managers and educators to design targeted, evidence-based training programs; it also facilitates quality improvement, standardization of sexual health care pathways, and enhanced survivorship care for women with gynecologic malignancies.

Despite these strengths, the study has identifiable limitations, including the use of convenience sampling from nine hospitals across six provincial-level regions in China, which may introduce selection bias and restrict generalizability to nurses from lower-level hospitals, smaller departments, or other geographic areas; the questionnaire has not been tested for predictive validity or longitudinal sensitivity to training interventions; and cross-cultural or cross-linguistic validation has not yet been performed. For future research, broader sampling across more regions, hospital levels, and department types is recommended to reduce selection bias and improve representativeness; longitudinal studies are needed to examine the questionnaire’s responsiveness to training and its predictive value for clinical performance; further investigation into factors influencing KAP levels will support precision intervention strategies; cross-cultural adaptation and validation can extend the tool’s international applicability; and development of a short-form or digital version may improve usability in busy clinical settings.

5. Conclusions

The Knowledge, Attitude, and Practice Questionnaire on Sexual Health Care for Gynecologic Oncology Nurses demonstrates excellent reliability, high content validity, stable construct validity, and satisfactory convergent and discriminant validity. This instrument provides a rigorous, standardized tool for assessing nurses’ knowledge, attitudes, and practices related to sexual health care in gynecologic oncology settings.