Using simulation to enhance primary care sexual health services for breast cancer survivors: a feasibility study

To evaluate the impact of a virtual simulation game (VSG) to improve primary care sexual health services for breast cancer survivors. We developed a VSG to help primary care providers (PCPs) address sexual health disturbances among breast cancer survivors. We used a pretest–posttest design with a series of validated tools to assess the feasibility and perceived impact of the VGS, including an open-ended question about participants’ perceptions. Quantitative data was analyzed using descriptive and inferential statistics and qualitative data through an inductive content analysis approach. Of the 60 participants, the majority were nurse practitioner students (n = 26; 43.3%), female (n = 48; 80%), and worked full-time (n = 35; 58.3%). Participants perceived the VSG as feasible and potentially effective. The intervention elicited an improvement in PCPs’ perception of knowledge between pretest and posttest surveys (z =  − 1.998, p = 0.046). Professional background and previous exposure to sexual health training were predictors of knowledge perception. Participants described the intervention as an engaging educational strategy where they felt safe to make mistakes and learn from that. VSGs can be a potentially effective educational approach for PCPs. Our findings indicate that despite being an engaging interactive strategy, VSG interventions should be tailored for each professional group. This intervention has potential to improve the knowledge and practice of PCPs related to breast cancer follow-up care to support comprehensive care for survivors, resulting in a better quality of life and patient outcomes.


Introduction
One in eight females will be diagnosed with breast cancer during their lifetime [1].Despite this high incidence, breast cancer has relatively low mortality and the five-year net survival can reach nearly 90% [1].Even though many breast cancer survivors may be free of disease, they face several challenges while trying to readjust to life after cancer [2].These include physical, social, emotional, financial, and spiritual effects of their cancer experience, which can persist for years and negatively affect their quality of life and health outcomes [2].The prevalence and intensity of these challenges can vary widely by individual, their diagnosis, treatment, and support received [3].The most common psychosocial issue that affects female cancer survivors is sexual problems, but they are often poorly addressed by healthcare providers [4][5][6][7][8].
Many cancer centers are progressing with the transition of patients affected by breast cancer from oncologists to primary care providers (PCPs) for follow-up care [9][10][11].In Canada, PCPs are usually nurse practitioners (NPs) or medical doctors (MDs) [12], and the survivorship care provided by those practitioners plays a significant role in delivering quality care and improving access to primary health [13][14][15][16].However, PCPs often report knowledge and practice gaps related to the management of the long-term effects of breast cancer and its treatment [4].These deficits can negatively affect the quality of life and health outcomes of survivors [17].This is in addition to evidence that healthcare 576 Page 2 of 9 professionals lack the knowledge and skills to address breast cancer survivors' sexual health concerns [4].It is recommended that these gaps be addressed through educational interventions [4,18].Still, different healthcare professional groups may have diverse educational needs and learning styles, and therefore separated investigation of each professional group can be useful to develop tailored and effective strategies.
Simulation-based education can improve healthcare professionals' knowledge and skills in a variety of areas [19].Additionally, virtual simulation and virtual simulation games (VSGs) are innovative knowledge translation strategies that can be more cost-efficient and as effective as inperson high-fidelity simulation and can positively impact knowledge and satisfaction with learning [20][21][22][23].Although it is important to understand the role of simulation-based education in relation to PCPs' knowledge and skills regarding cancer survivorship care, there is a gap in the literature related to this topic.Recent literature reviews [24,25] reveal there is no study exploring simulation use related to cancer survivorship care; this suggests that more research should be conducted to further understand the impact of simulation on cancer survivorship care.
To help address this gap, we developed a VSG focused on supporting PCPs to start sexual health discussions and perform screening/assessment and essential interventions.This VSG was developed using the Canadian Alliance of Nurse Educators using Simulation (CAN-Sim) design process [26], based on best practice guidelines [3,[27][28][29], and the content was reviewed by a panel of experts.The learning outcomes (Supplemental Material S1) guided the entire development of the intervention; specific details of the VSG creation process are published elsewhere [30].Therefore, following the development of that intervention (https:// can-sim.ca/ acces sjama/ sexual-health-cance r/#/), this study was conducted to (1) evaluate the feasibility, acceptability, appropriateness, and perceived effectiveness of a VSG to educate PCPs providing care to breast cancer survivors and (2) investigate the potential impact of a VSG on PCPs' perception of knowledge regarding sexual health disturbance in breast cancer survivors.

Method
Due to the lack of evidence on the use of simulation to educate professionals in cancer survivorship care [31], we used a feasibility approach to assess the implementation and potential impact of implementing a VSG with PCPs.We used a single-group pretest-posttest design to evaluate the influence of our VSG as this type of investigation allows assessments to predict directions and trends on whether the intervention can be feasible to yield better outcomes in clinical practice [32].This study is part of a multi-methods research design guided by the Knowledge to Action Framework [33] and Cancer Survivorship Care Quality Framework [3] to test a series of hypotheses (Supplemental Material S2).To guide the reporting of methods, the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [34] for cohort studies guidelines was used.Ethics clearance was provided by Queen's University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board (file number 6035252).

Setting and sampling
Participants were recruited using a combination of purposive and snowball sampling through recruitment emails, social media, and existing networks within Southeastern Ontario, Canada.The inclusion criteria were to (1) be an MD or NP (NP students holding RN diplomas or MD graduate students holding MD diplicas were also eligible), (2) be working or studying in Southeastern Ontario, (3) be fluent in English, and (4) be willing to participate in the research.

Procedures
The survey with the measures (described in the following sections) was hosted online on the Qualtrics Survey Platform.After recruitment, participants signed the informed consent and completed the pretest survey, which included questions related to demographic data (e.g., profession, age, and years of experience and previous training) and the Sexual Health Knowledge Scale (SHKS).Then, participants completed the VSG and were redirected to respond to the posttest survey, which included the SHKS, the Triple P tool, and the Simulation Effectiveness Tool-Modified (SET-M).
The survey was open for a period of 153 days (September 1, 2022, to February 1, 2023).Participants who completed all procedures received a gift card valued at CAD$25.To support the validity and reliability of our findings and avoid potential sources of bias, we used a standardized protocol for all procedures (e.g., the same assessment tools and procedures were used consistently among all participants) and regular quality checks between the first author and co-investigators.

Triple P tool
This tool is composed of three subscales (Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM)) intended to measure outcomes indicating implementation success [35].Each subscale has four items scored on a 5-point Likert scale (ranging from completely disagree-1 to completely agree-5) [35].The subscales are scored separately, where higher scores indicate greater levels of acceptability, appropriateness, and feasibility.We considered item scores greater than or equal to 4 as our cut-off value to indicate a positive response to the intervention [35].

SET-M
The SET-M is a modified version of the Simulation Effectiveness Tool that was adapted to be used for virtual simulation [36].This tool consists of 19 items, each scored on a 3-point Likert scale (ranging from strongly agree-3 to do not agree-1), and higher scores mean a higher level of agreement.This tool has no cut-off values; instead, the developers suggest that the investigator should evaluate each item individually [37].Additionally, the tool also has an open-ended item where participants can share qualitative feedback about their perceptions of the intervention.

SHKS
This scale was developed to assess healthcare providers' perception of their knowledge related to the assessment, evaluation, and management of common sexual side effects of cancer treatment to help survivors achieve the best possible sexual function.The SHKS is a five-item scale that is scored using a 3-point Likert scale (ranging from not so much knowledge-0 to a lot of knowledge-2).All items are summed to reach a total score ranging from 0 to 10, where higher scores indicate greater knowledge about sexual health.We used a cut-off value of 8 to indicate an elevated level of self-rated knowledge [38].

Data analysis
To test the hypothesis (Supplemental Material S2) of this study, quantitative data were entered into SPSS Software Version 25.0 for statistical analysis.Only participants who provided complete responses for both the pretest and posttest surveys were included in the analysis, and all procedures were peer-reviewed by a professional statistician.We used a priori power analysis to determine our sample size considering an effect size of 0.8, alpha level of 0.05, equal group sizes, and 80% statistical power, which gave a requirement of 54 participants, still, to account for potential attrition and incomplete data, we recruited 60 participants (achieved power for an effect size of 0.805 was approximately 99%).
To address aim 1, descriptive statistics (e.g., frequency and mean) were used to evaluate the scores of the Triple P tool and the SET-M to evaluate the potential effectiveness and feasibility of the VSG (research question 1a).To test if there was a statistically significant difference among NPs, MDs, and NP students (for both the Triple P tool and the SET-M), we used the Kruskal-Wallis test and to identify the group that presented a different result from the others, and Mann-Whitney U tests were performed with two groups in each test (research question 1b).The qualitative data from the open-ended item of the SET-M were imported into NVivo® for a qualitative descriptive analysis approach where we could categorize our findings for reporting (research question 1a).
To address aim 2, we used repeated measures of the Wilcoxon signed-rank test to compare SHKS scores between pretest and posttest (research question 2a).Additionally, to model the relationship between the mean of the outcome variable (SHKS score) and the covariates which were defined as professional group (NPs, MDs, and NP students), gender (male, female, and non-binary/third gender), age (≤ 25 years, 26-35 years, and ≥ 36 years), and previous training (presence or absence) between observations (pretest and posttest), we used the generalized estimating equations (GEE) with Poisson distribution with a log link function (research question 2b).If any effect was found to be significant, the Bonferroni test for post hoc was chosen to reduce the probability of type I error as we were testing multiple hypotheses.

Results
Most of the participants were NP students (n = 26; 43.3%), female (n = 48; 80%), worked full-time (n = 35; 58.3%), and had attended specific training related to sexual healthcare for patients with cancer (n = 36; 60%).Participants' timing between completion of the pretest and posttest surveys was on average two hours.Detailed characteristics of the participants can be found in Table 1.

Aim 1 (research questions 1a and 1b): feasibility, acceptability, appropriateness, and perceived effectiveness
Participants perceived the VSG as appropriate, acceptable, and feasible to use, as demonstrated by their mean scores on the Triple P tool: AIM 4.61 (SD = 0.471), IAM 4.47 (SD = 0.561), and FIM 4.60 (SD = 0.460), all out of 5 (Supplemental Material S3).Additionally, our results indicated the VSG was a potentially effective educational strategy as the SET-M score (Supplemental Materials S4, S5) was relatively high and close to reports from normative samples [39] with an overall mean of 2.62 (SD = 0.441) out of 3.For the items scored, "understanding of medications" had the lowest mean of 2.46 (SD = 0.750), while "ability to teach patients about illness and interventions" had the highest mean of 2.77 (SD = 0.500).
For the following variables (Table 2), there was at least one group of professionals with a significant difference in scores (p < 0.005): AIM; IAM; SET-M 1, 4, 6, 7, 9, 11, 12, 13, 16, 17, and 19.After further analysis with two groups in each test (Supplemental Material S6, S7, S8), there was no significant difference between MDs and NPs; however, NP students presented significantly lower scores when compared to the other groups.
Nineteen participants provided qualitative feedback on their perceptions of the effectiveness of the VSG.Overall, participants had a positive view of the game as a "great learning," "engaging," and "choose your own adventure" experience.The aspects highlighted as being most interesting for participants were having a male healthcare provider, the possibility of trying different answers in the game while feeling comfortable in making mistakes, and the importance of debriefing for the learning process.However, participants also expressed concerns and provided suggestions for future interventions, including concerns about how the game would be updated when new guidelines are published, the need for more attention to body image disturbance, technical issues (e.g., program freezing), interest in more simulations approaching male patients, and adding a more in-depth discussion of other interventions to manage sexual health disturbance.

Aim 2 (research questions 2a and 2b): perception of knowledge related to sexual health disturbance
The difference in perception of knowledge was statistically significant (z = − 1.998, p = 0.046), indicating that participants had an improvement from the pretest to posttest assessment.Providers had a mean SHKS score of 6.34 (SD = 4.419; median = 9.00) out of 10 in the pretest and a mean of 7.18 (SD = 3.332; median = 9.00) out of 10 in the posttest, showing that the distribution of the SHKS scores was not symmetrical and that the tail of the distribution was skewed to the left or toward lower scores.Despite the improvement in providers' perceptions of knowledge, the posttest score was still lower than the cut-off value of 8 indicating a perception of knowledge below the ideal.We analyzed the data to evaluate the relationship between the mean of SHKS score and the covariates defined as professional group, gender, age, and previous training (Table 3, and Supplemental Material S9-S10) with participants with complete data (n = 56).When assessing for professional background, there was a significant association (p value < 0.001) between the SHKS score and professional groups at different points of data collection (pretest and posttest).Specifically, NP students' background was associated with lower scores in both the pretest and posttest when compared to MDs and NPs.Still, NP students were the only group that presented a statistically significant difference between scores (p value = 0.008), indicating that the VSG improved their knowledge between measures.Additionally, age (p value = 0.994) was not statistically significant with SHKS scores regardless of the point of data collection (both pretest and posttest).
When controlling for gender in the models, despite a significant difference (p value < 0.001), there was no association between the SHKS scores across gender in pretest and posttest measures (p value = 0.129).Specifically, while males were more likely to have a higher pretest score, their score decreased in posttest making the difference in the scores of the groups of female and non-binary individuals closer to that of the male group.Lastly, when controlling for previous sexual health training, we identified a statistically significant difference in the pretest and posttest scores of those who had previous training (p value = 0.028) and participants who attended sexual health education were more likely to have a higher perception of knowledge.

Discussion
This feasibility study implemented and tested the potential impact of a VSG to improve PCPs' knowledge perception about sexual health among breast cancer survivors.Participants perceived the VSG as feasible, acceptable, appropriate, and potentially effective.The VSG improved the perception of knowledge among participants, and the covariates that predicted a higher participants' perception of knowledge were professional background and previous sexual health training.
The scores for implementation success (Triple P tool) were high among participants, suggesting the VSG was a feasible intervention.Despite the lack of previous interventions in cancer survivorship care [24], our findings suggest that VSGs can meet PCPs' reported desire for short and interactive interventions to learn more about sexual health among breast cancer survivors [40].Similarly, participants' perception of effectiveness (SET-M) was also high, a finding that has been previously reported in the literature since VSGs are considered a potentially effective knowledge translation strategy [20][21][22][23].
The differences observed in our sample related to professional background and perception of knowledge suggest that the effectiveness of our VSG may vary according to the type of professional group.NP students' scores varied across the findings when compared to MDs and NPs.This finding is understandable given that the educational needs of healthcare students differ from healthcare professionals and so is important to use tailored interventions for each group of learners to increase the chances of effectiveness.Still, when analyzing the scores for the perception of knowledge, it was evident that NP students had lower baseline scores in comparison to MDs and NPs.This difference may be due to the lack of exposure to previous sexual health training, a covariable that was also a predictor for the perception of knowledge.While 100% (n = 12) of MDs and 80.95% (n = 17) of NPs previously attended sexual health training, only 19.23% (n = 5) of NP students had attended such training.Despite having lower scores in both the pretest and posttest, NP students presented the highest variation in both measures and were the only group whose scores had a significant improvement from pretest to posttest, indicating that VSGs can be particularly relevant for NP educational programs.Finding which goes toward previous studies where virtual simulation has been shown to support and advance graduate clinical education [41].
The small variation in MDs' and NPs' scores from pretest to test may be attributed to the fact that these groups already had a high score in the baseline measure for the perception of knowledge.It is not uncommon in the existing literature [42,43] to find cases where educational interventions do not yield improvements in the outcome measures, particularly when measuring for the perception of knowledge rather than actual knowledge, since the participants may rate their levels higher at baseline, while after exposure to the intervention they use the actual rates [42,44].Overall, these differences across professional groups highlight the complexity of the relationship between the intervention and the outcome and emphasize the importance of tailored interventions.
Despite meta-analytic findings substantiating the influence of gender and age [45,46] on training effectiveness, our results did not show any significant relationship in those covariates which may be due to the heterogeneous characteristics of our sample.Lastly, participants further expressed their views of the VSG's effectiveness by providing qualitative feedback on the open-ended item of the SET-M.Corroborating with the existing literature, our participants saw the intervention as an engaging educational strategy [20][21][22][23]; still, technological issues and the need for more interventions were identified as barriers to the use of VSGs [47].Collectively, our findings illuminate that VSGs can be a potentially effective educational strategy to meet the educational needs of PCPs; however, further studies are needed to confirm and extend our results.

Limitations
Despite the potential contribution of this study, there are also some limitations that are worth noting.We did not evaluate variables such as comfort, skills, performance, competency, and attitudes, and those could potentially be relevant to further explore the topic.This was a feasibility study testing a novel intervention in cancer survivorship care, and although our results showed that participants perceived the intervention as feasible and potentially effective, the levels Page 7 of 9 576 of perception of knowledge were still under the cut-off value and further investigation of actual knowledge levels would be beneficial.Also, we used a single posttest measure to evaluate the short-term effects of the intervention, and future studies should focus on the use of long-term measures as that could potentially yield further findings on the efficiency of VSG.Our study's findings may not be transferred to the broader population due to the overrepresentation of primary healthcare providers in southeastern Ontario, as the sample's regional exclusivity may limit broader generalizability.Lastly, technological limitations may have been a barrier for potential participants to enroll in this study as some participants may not be comfortable using technology or have access to that.

Implications to practice
This intervention has the potential to improve the knowledge and practice of PCPs related to breast cancer follow-up care to support comprehensive care for survivors, resulting in better quality of life and patient outcomes and reduced costs for the healthcare system.Additionally, this study was, to the best of our knowledge, the first to investigate a single simulation intervention used to improve the provision of healthcare to cancer survivors.As a result, it can be used to create a solid underpinning for the nursing and healthcare community to more interventions using a similar approach.Still, future studies should focus on evaluating the long-term effects of VSG, include more comprehensive variables (e.g., knowledge level and attitudes), and more focus on pathophysiology and interventions, including medication use, to help manage sexual health concerns among breast cancer survivors.Although this was a feasibility study, we achieved a high-power effect size of 0.805 (approximately 99%), which suggests that our sample size was adequate to detect a significant effect, and our results are likely to be dependable and capable of being reproduced.Lastly, various knowledge translation activities are being carried out to support the dissemination of this work.The VSG that was created for this research is part of the Open Access Inventory of CAN-Sim and is publicly available for the community and promptly accessible for implementation.The link to access the game can also be shared across multiple resources and supporters of this work (e.g., Canadian Association of Oncology Nurses).The authors from this work are working in further extend this investigation and create more VSG to improve the psychosocial domain of breast cancer survivorship care (e.g., body image disturbance).

Conclusion
This feasibility study provided preliminary evidence that the VSG can be feasible, acceptable, appropriate, and potentially effective to improve PCPs' perception of knowledge.Our results highlight the need for tailored interventions across different professional groups to increase the chances of effective implementation.This study, therefore, provides implications for healthcare practices, as the intervention can help improve breast cancer survivorship care provided by PCPs.Finally, the results of this study will be used to guide other researchers (including the lead authors) to develop more interventions in this area of investigation.

Table 1
Participants' demographic characteristics*A total of 71 participants responded to the pretest, and 60 responded to the posttest survey.For the data analysis, we only considered the 60 participants that responded to both measurements

Table 2
Kruskal-Wallis test for outcome measures (AIM, IAM, FIM, and SET-M individual scores) Note: The SET-M score cannot be combined and therefore individual scores were used; also, values in bold indicate a statistically significant test result (P ≤ 0.05)