Abstract
-
Purpose
This study compared healthy lifestyle (HLS) practices and awareness regarding the use of chatbots (A-uC) for health management between childhood and adolescent cancer survivors (CACSs) and their parents, with the aim of assessing the feasibility of tailored artificial intelligence (AI) chatbot-based interventions for holistic survivorship care.
-
Methods
A descriptive comparative design was employed involving 80 CACSs and 80 parents (N=160) recruited through the Korean Pediatric Cancer Foundation. HLS practices were assessed using a validated seven-domain instrument encompassing physical activity, nutrition, interpersonal relations, stress management, positive life perspective, health responsibility, and spiritual health. A-uC was evaluated using an extended technology acceptance model-based tool. Responses to the open-ended question addressing unmet HLS practices were analyzed using latent Dirichlet allocation topic modeling.
-
Results
No significant differences were observed between CACSs and parents in overall HLS (CACSs: 3.16±0.80; parents: 3.18±0.36, p=.74). While perceptions across most A-uC domains did not differ significantly, parents demonstrated a significantly higher “intention to use” chatbots for health management than CACSs (p=.03). The mean A-uC scores exceeded 4 (out of 5) in both groups, reflecting positive perceptions of chatbot-based HLS support. Topic modeling identified “exercise,” “healthy diet,” and “regular lifestyle” as common unmet areas.
-
Conclusion
CACSs and their parents share largely concordant views on HLS and A-uC, with a strong interest in chatbot interventions. These findings underscore the potential of tailored AI chatbot programs to address unmet lifestyle needs and promote holistic survivorship care.
-
Key words: Adolescent; Artificial intelligence; Cancer survivors; Childhood; Health
INTRODUCTION
Childhood and adolescent cancer survival rates in Korea have steadily improved, increasing from 77.3% in 2006 to 85.6% in 2016 and 86.3% in 2021 [
1]. However, despite the improved survival rates, childhood and adolescent cancer survivors (CACSs) experience long-term side effects, referred to as “late effects,” after the completion of treatment [
2,
3]. A systematic review of health problems experienced by CACSs found that they experience physical changes, such as fatigue, infections, and hair loss, resulting from surgery, radiotherapy, and chemotherapy [
4], as well as unmet physical, psychosocial, and developmental needs, such as depression, anxiety, low self-esteem, confusion about identity, school adjustment, future careers, and problems forming intimacy with the opposite sex [
5,
6].
Childhood and adolescence represent a critical developmental transition between early childhood and adulthood, during which the foundations of healthy adulthood are formed, emphasizing the importance of balanced, holistic development. Humans are motivated by physical, psychosocial, and spiritual needs related to the promotion of well-being and consequently engage in health-promoting behaviors through the realization of their individual potential and modification of their environment [
4]. Grounded in the concept of human wholeness, the goal of healthcare for CACSs is the development of healthy adults and requires positive changes in their lives through holistic health promotion strategies that address the physical, psychological, social, and spiritual dimensions of health as well as discovering the power of positivity and hope [
7].
A text network analysis, integrative review, and meta-analysis of research trends in the health needs and management of CACSs [
2,
8,
9] revealed that most research falls into two main areas: First, studies related to risk management with cancer treatment side effects. Research topics include studies that examined physical issues such as physical activity, fatigue, symptom control, and sleep disorders, as well as studies related to psychological and social health, such as low self-esteem, anxiety, school adjustment, and resilience. Second, studies on health-related quality of life with supportive care that addressed self-care issues. The meta-analysis also focused on the physical outcome variables of supportive interventions. However, few studies have been conducted on healthy lifestyle (HLS) based on the health needs of CACSs, and only a few systematic reviews and meta-analyses [
2,
8,
9], holistic health needs analyses [
10], and intervention studies [
11] have attempted to address holistic healthcare access for CACSs since 2022.
In the United States, European countries, and Korea, self-management of HLS aimed at returning to daily life, including complex unmet needs following the completion of treatment for CACSs, is included in the standard of care [
12]. Kang et al. [
10] suggested that the health management of CACSs requires a holistic approach, including health responsibility, nutrition, physical activity, and spiritual health, along with the establishment of a professional information and communication system to motivate CACSs to self-direct HLS management. Another study reported the significant effects of a mobile functional game-based HLS program on the overall health of adolescent cancer survivors. It found that mobile functional games were highly learner-driven and engaging, reflecting the characteristics of the “game generation” of adolescent learners, but were limited in terms of interactivity [
11]. These findings suggest a need for integrated interactive interventions to advance holistic HLS.
On the other hand, systematic reviews have examined the adoption of generative artificial intelligence (AI) chatbots in healthcare [
13-
15]. AI-based support systems can be introduced for self-directed HLS management and interactive supportive care. The 2020s are expected to represent an era of AI chatbots that are capable of understanding and learning from data through deep learning AI technology to generate new results based on user requests, with applications expanding across all sectors, including education, healthcare, industry, finance, art, and the military. Currently, the most active applications are in education, while research in the healthcare sector is beginning to focus on areas that require ongoing healthcare or support roles. AI chatbots have the advantage of enabling real-time self-directed management and interaction for those who require continuous healthcare at home, such as CACSs.
This study investigates the scientific feasibility of applying AI chatbots to the HLS management of CACSs by comparing the degree of HLS management and awareness of the use of chatbots (A-uC) among CACSs and parents. It sought to (1) compare the degree of CACSs’ HLS as self-reported and as perceived by their parents, (2) compare children’s and parents’ reports regarding the A-uC, and (3) identify vulnerable aspects of HLS reported by CACSs and their parents.
1. Theoretical Framework
To identify the HLS of CACSs, we adopted the health promotion model [
16], which posits that humans engage in health promotion behaviors by realizing their individual potential and modifying their environment, synchronized with physical, psychosocial, and spiritual needs to promote well-being, and emphasizes the active role of patients in health promotion behaviors. The HLS measure applied in this study was based on seven domains, namely, physical activity, nutrition, interpersonal relationships, stress management, positive thinking, health responsibility, and spiritual growth, of the Child Healthy Lifestyle Profile [
17], which was developed based on the health promotion model.
The A-uC assessment was grounded in the technology acceptance model (TAM), an information systems theory that explains and predicts users’ acceptance and use new technologies [
18]. The model delineates key determinants of technology adoption, with particular emphasis on three key factors: “perceived usefulness,” “perceived ease of use,” and “intention to use” [
19]. In this study, we applied the A-uC measurement tool developed by Kang et al. [
20] by adding the three TAM factors and the concept of “perceived value” based on previous studies (
Figure 1).
METHODS
Ethical statements: This study was approved by the Institutional Review Board of Sahmyook University (IRB No. SYU 2025-05-039-002). Informed consent was obtained from all participants.
1. Study Design
This descriptive comparative study aimed to compare the perceptions of CACSs and their parents regarding HLS and generative A-uC as a health management support tool. A cross-sectional survey method was employed, and the reporting of this study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies [
21].
The study participants consisted of CACSs and their parents who were recruited online after obtaining consent from the Korean Pediatric Cancer Foundation. The participants were registered with the foundation and included CACSs and their parents. The reason for including not only CACSs but also their parents as the target population was that, due to their developmental characteristics, parents are consistently and directly involved in care and actively participate in their care processes.
The eligibility criteria for CACSs were as follows: (1) being of Korean nationality and between the ages of 10 and 16 years, (2) having completed all acute-phase anticancer treatments following a cancer diagnosis more than six months ago, (3) being capable of reading and understanding the survey questions, (4) having voluntarily agreed to participate in the study, and (5) having received written parental consent for their participation.
The parent group was selected from among the parents of the recruited CACS who understood the purpose and procedures of the study and agreed to participate. Those with severe mental or cognitive impairments that would hinder their participation in the survey were excluded.
The appropriate sample size was calculated based on a previous study that investigated the HLS among childhood cancer survivors and their parents [
10]. The sample size was calculated using G*Power ver. 3.1.9 (Heinrich-Heine-Universität Düsseldorf) [
22], with the calculation based on an independent-samples t-test to test for mean differences between the groups. Although the participants comprised CACSs and their parents, this analysis focused on comparing the overall perceptions and HLS characteristics of the two groups as independent cohorts rather than as matched pairs. This approach was intentionally selected to identify the distinct tendencies and needs of each group. The parameters entered into the calculation had a significance level of α=.05, power 1–β=.80, and an intermediate effect size d=0.5. The effect size was set at d=0.5, following Cohen’s medium effect criteria [
23] and the methodological design of a previous study [
10], because this value represents a clinically meaningful threshold for identifying intervention needs in this population. The total sample size was 128 participants (64 per group). Considering a dropout rate of 10%–20%, we aimed to recruit 75 participants per group. The final data analysis included 160 participants, comprising 80 participants in the CACS group and 80 participants in the parent group.
In this study, a structured self-report questionnaire was used to assess HLS behaviors and A-uC.
HLS behaviors were assessed using an HLS measurement tool developed and validated for childhood cancer survivors [
17]. The tool comprises seven subdomains: health responsibility (four items), physical activity (four items), nutrition (four items), positive life perspective (five items), stress management (five items), interpersonal relations (six items), and spiritual health (three items). Thus, the total number of items was 31. Each item was scored on a 4-point Likert scale (1=not at all, 4=very much), with higher scores denoting a higher level of practice of HLS behaviors in each domain. At the time of development, Cronbach’s α was .87, and the Cronbach’s α in this study was also .87.
A-uC was measured using a tool that was modified and supplemented to assess A-uC for violence prevention education among elementary school students [
20]. This tool was developed based on the extended TAM [
18]. The tool under consideration consists of 22 items divided into the following subcategories: perceived value (11 items: professionalism, three items; reliability, four items; empathy, four items), perceived usefulness and ease of use (seven items), and intention to use (four items). “Professionalism” refers to the extent to which a chatbot is perceived as providing accurate, knowledgeable, and expert-level information in a credible and appropriate manner. “Reliability” reflects the extent to which the chatbot is perceived as providing consistent, dependable, and trustworthy information that users can rely on. “Empathy” refers to the extent to which the chatbot is perceived as understanding users’ emotions and responding in a caring, supportive, and emotionally appropriate manner [
18,
20]. Each item is measured using a 5-point Likert scale (1=not at all, 5=very much), with higher scores denoting a more positive perception of AI-based chatbots. In a previous study, Cronbach’s α was .96 [
20], and in this study, Cronbach’s α was .95.
Furthermore, to identify areas of HLS that require improvement, an open-ended question was added: “What is not being done for HLS?” The responses to this question were analyzed via topic modeling to derive common themes.
4. Data Collection
Data were collected from June 18 to July 1, 2025, following approval from the IRB of Sahmyook University. The study was conducted exclusively with participants who voluntarily agreed to participate after receiving a detailed explanation of the research objectives and methods through an online recruitment notice targeting CACSs registered with the Korean Pediatric Cancer Foundation and their parents. For minor CACSs, additional written consent (assent) was obtained from both the legal guardians and participants. Parental consent was obtained through online signatures from parents who agreed to participate in the study after receiving an explanation via telephone.
All participants were informed that their involvement was entirely voluntary and that they could withdraw from the study at any time. The survey was administered online using a web-based platform (Google Forms; Google LLC) without face-to-face contact, with an average response time of approximately 15–20 minutes. Responses were collected anonymously and were not used for any purpose other than this study. To compensate the participants for their involvement in the study, an online gift card with a value of 10,000 Korean won was provided.
5. Data Analysis
The collected data were analyzed using IBM SPSS Statistics ver. 29.0 (IBM Corp.) [
24] and NetMiner ver. 4.5.0 (Cyram Inc.) [
25]. The demographic characteristics of the participants were calculated using frequencies, percentages, means, and standard deviations. Differences in HLS and A-uC scores between the groups were compared using an independent-samples t-test. Prior to conducting the test, the assumptions of normality and homogeneity of variance were verified using the Shapiro-Wilk test and Levene’s test, respectively. The open-ended responses to the question “What is not being done for HLS?” were analyzed using latent Dirichlet allocation (LDA)-based topic modeling with NetMiner ver. 4.5.0 to extract core keywords and name them by topic [
26]. To produce distinct and sparse topic distributions, LDA analysis was conducted with parameters ⍺=.01 and β=.01, and 1,000 iterations were performed to ensure model convergence. The optimal number of topics was determined by evaluating topic coherence and exclusivity indicators across multiple simulations [
25,
27]. In light of these findings, two researchers with expertise in pediatric nursing independently conducted a dual review of the degree of topic differentiation and interpretability. Any discrepancies between the researchers regarding topic categorization or naming were resolved through in-depth discussions until a 100% consensus was reached. The final number of topics was arrived at through this iterative consensus process. Each topic was named by identifying the first and second core keywords, as well as keywords appearing at least twice, followed by a thorough review of the original responses with high topic probabilities. To visually present the relative frequency and importance of the analyzed keywords, a word cloud was generated to enable the intuitive identification of keywords by topic.
RESULTS
1. General Characteristics of the Study Participants
The participants consisted of 80 pairs of CACSs and their parents (a total of 160 individuals) (
Table 1).
The average age of the CACSs was 12.1±2.0 years, with a majority of females: 41 (51.2%). Thirty participants (37.5%) were second-born children, and the average age at diagnosis was 7.48±3.6 years. The most common type of cancer diagnosed was leukemia, accounting for 47 cases (58.8%), and the most common (29 participants, 36.3%) time period since the last treatment was 1 to 5 years. The most common type of treatment was chemotherapy, with 67 participants (45.9%) receiving this treatment and 66 participants (82.5%) having no relapse. The health status perceived by CACS patients was “healthy” (28 participants, 35.0%). The primary source of HLS knowledge was parents (64 participants, 38.8%), and the most common source of HLS education was school (69 participants, 40.1%). Among the CACS patients, 44 (55.0%) had no experience with AI chatbots, but 49 (61.3%) responded that AI chatbots were “necessary” for HLS education.
The average age of parents was 44.7±4.2 years, with mothers accounting for 76 (95.0%) of the total. The most common place of residence was Gyeonggi-do (20 participants, 25.0%), and 63 parents (78.7%) were college graduates or higher. Seventy-five participants (93.8%) were married, and 63 (78.8%) rated their socioeconomic status as “middle.” In terms of religion, 35 participants (43.8%) answered “none.” The subjective health status of CACS perceived by parents was “moderate” for 41 participants (51.2%). The primary source of HLS knowledge was “mass media” for 77 participants (44.8%), and the primary source of HLS education was “mass media” for 44 participants (32.6%). While 48 participants (60.0%) had no experience with AI chatbots, 59 (73.8%) responded that AI chatbots were “necessary” in HLS education.
2. Degree of CACSs’ HLS and A-uC Reported by Themselves and Their Parents
No statistically significant differences were found in the total HLS or scores of the seven subdomains between the CACSs and their parents (
Table 2). Both groups scored an average of 3 or above out of 4 points for spiritual health, stress management, interpersonal relations, and positive life perspective, whereas they averaged less than 3 points for nutrition, health responsibility, and physical activity. The overall HLS score was similar between CACSs (3.16±0.80) and parents (3.18±0.36) (t=0.33,
p=.74).
The scores for A-uC were high, averaging 4 out of 5 or above for both groups, and no significant differences were observed in most subdomains (perceived usefulness and ease of use; perceived value: professionalism, reliability, and empathy). However, in the “intention to use” domain, parents scored significantly higher at 4.21±0.68 points compared with CACSs (3.94±0.90 points) (t=2.19, p=.03). Nevertheless, the CACSs scores exceeded 3.9 out of 5 points, indicating an overall positive response.
3. Vulnerable Areas in the Practice of HLS
To identify areas where HLS practice is vulnerable, topic modeling analysis was conducted on responses to the open-ended question “What’s not being done for HLS?” (
Table 3,
Figure 2). The topics named by the authors were determined based on core keywords and their frequencies identified through the topic modeling results.
When integrating the responses from CACSs and parents, the most frequently mentioned theme was “exercise” (57 documents), which included keywords such as “physical,” “activity,” and “strength.” Next was “healthy diet” (52 documents), with keywords such as “overeating,” “healthy food,” and “fast food,” followed by “a regular lifestyle” (51 documents), which appeared alongside keywords such as “change habit,” “smartphone,” and “sleep well.”
By subgroup, parents most frequently mentioned “a regular lifestyle” (30 documents), followed by “exercise” (27 documents) and “healthy diet” (13 documents). Conversely, CACSs most frequently mentioned “exercise” (34 documents), followed by “a regular lifestyle” (25 documents) and “healthy diet” (21 documents).
The three themes (exercise, healthy diet, and a regular lifestyle) were commonly derived from both CACSs and parents, and researchers interpreted and named them based on core keywords extracted through topic modeling (first and second keywords and keywords appearing with a frequency ≥2 times). In the word cloud (
Figure 2), these keywords were visualized in a larger font size than the other words. This finding suggests that these were not areas where HLS management maintained good practices.
DISCUSSION
The following discussion is based on the findings of this study.
For CACSs, schools were the primary source of information and education about HLS, followed by the mass media, whereas for parents, the mass media was the primary source of information. These results suggest that CACSs need an HLS-centered life for their ongoing health management and school adjustment and that mass media plays a key role in this management system. It also underscores the need to provide reliable and scientifically based data for school education, as well as for education delivered through mass media. One study reporting the effectiveness of nurse-led video-coaching interventions also showed medium-to-large improvements among CACSs in quality of life, lifestyle (lifestyle intervention), self-efficacy, and self-management [
28]. The active intervention role of nurses as educators in health education for CACSs is emphasized. They are encouraged to actively intervene in education through the mass media.
CACSs and parents did not differ significantly in their responses to the HLS total score or any of the seven subscales. These findings are consistent with those of Kang et al. [
10], who conducted an HLS survey with a different sample of CACSs and their parents. Given the developmental aspects of children’s dependence on their parents, a close organic relationship between parents and children is observed in the management of HLS in CACSs, suggesting that HLS interventions should adopt a holistic approach that considers both parents and children.
When the seven HLS subscales are analyzed, both groups ranked the perception of HLS management the same, with “spiritual health,” “stress management,” “interpersonal relations,” and “positive life perspective” ranking above 3 (4-point Likert scale), but “nutrition,” “health responsibility,” and “physical activity” ranking below 3. The open-ended question (“What’s not being done for HLS?”) also revealed that neither group managed “exercise,” “healthy diet,” and “a regular lifestyle” well. “Nutrition,” “health responsibility,” and “physical activity” also scored below 3 in the HLS survey of CACSs, as reported by Kang et al. [
10], meaning that these areas are very important to CACSs but are weak in terms of actual health management. For the HLS of CACSs, it is thought that “exercise,” “healthy diet,” and “a regular lifestyle” can be promoted by strengthening health responsibility. In contrast, both CACSs and parents scored lowest on “spiritual health” [
10], unlike the present study. In the area of “spiritual health,” repeated studies are needed to conduct an accurate unmet needs survey. In a systematic review that identified physical activity and diet interventions for adolescent and young adult cancer survivors, most interventions were related to physical activity, with only four studies including nutrition or diet interventions, and digitally based physical activity interventions were commonly used [
29]. In this respect, the ideal model of transitional survivorship care for CACSs should include a patient navigator who promotes provider flexibility, consistent communication, and proactive comprehensive care that encompasses both medical and psycho-social well-being [
30]. The adoption of AI-powered chatbots, in addition to traditional training modalities, is an active issue.
Close to half of the participating CACSs and parents had experience using chatbots and recognized the need for chatbots in HLS education. When analyzed specifically for chatbot usage perceptions, the A-uC composite score (based on a 5-point scale) was above 4 and was not significantly different between groups. However, the only sub-item that showed a significant difference between the two groups was “intention to use,” for which parents had a higher level of intention than CACSs, but the CACSs scores were also close to 4. In this study, a high demand for education through chatbots was noted for both CACSs and parents, suggesting a growing need for chatbots as an educational medium for CACSs to improve HLS maintenance. Sezgin et al. [
31] found that caregivers of young childhood cancer survivors preferred multimodal interactions (voice and text), particularly those valuing flexibility based on context. Both systematic review [
32] and scoping review [
33] studies indicated that, as a useful and cost-effective intervention tool, chatbots under the supervision of medical professionals have the potential to improve patient outcomes, including in the field of oncology. In contrast, it was suggested that more scientific evidence is needed in the field of education. The results of this study can help identify the needs and specific areas of educational intervention for chatbots in the healthcare field.
This study discusses the management of HLS in CACS based on holism according to eight subdomains. When CACSs healthcare is integrated with physical, socio-psychological, routine self-care, and preventive healthcare, true health recovery and promotion of CACSs will be achieved [
2]. This study’s identification of CACSs’ holistic healthcare status provides important scientific evidence to guide the specific management of CACSs’ HLS and to support their development into adults with a balanced lifestyle and healthy minds and bodies. It can also serve as a useful resource for parents to raise their awareness of holistic healthcare. This is the first study in Korea to investigate A-uC among CACSs and their parents based on the TAM model [
18]. The study revealed little substantive evidence or understanding of human–chatbot interactions, such as from participant observations or in-depth interviews. These results indicate that technology design activities (platform design, communication modality, and content) are needed to adopt chatbots in education [
33]. In this study, A-uC was assessed in a general sense rather than being limited to the use of a chatbot designed for a specific topic or content area. Although further evaluation is necessary after the development of a chatbot specifically tailored for CACSs, the purpose of the present assessment was to examine participants’ overall perceptions of chatbot use. Future studies should reevaluate perceptions after the development and implementation of a chatbot designed specifically for CACSs.
We summarize the limitations of this study and offer suggestions for future research. First, it is challenging to conduct a large-scale survey of CACSs and their parents. Children and their parents who were willing to participate after being informed through the Korean Childhood Cancer Foundation were selected. To compensate for the limited generalizability of the results owing to the small sample size, we investigated the awareness of HLS management and A-uC by including parents and CACSs. In addition, incorporating parents’ perspectives alongside those of CACSs helped address the limitations of presenting CACSs’ awareness alone. Second, our sample had a limited capacity to identify differences in HLS management based on diagnosed pediatric malignancies and different pediatric malignancies, including differences according to prior experience with chatbot use. We suggest that future studies conduct a replication study with a larger sample size of CACSs.
CONCLUSION
Various healthcare interventions are expected to be introduced and applied based on chatbots. Considering the developmental characteristics of CACSs who need continuous holistic healthcare, the results of this study can serve as a useful resource for developing customized chatbot interventions based on the HLS and A-uC of CACSs and their parents.
ARTICLE INFORMATION
Figure 1.Theoretical framework. (A) From Kang KA. Child Health Nurs Res. 2023;29(1):37-50 [
17] and (B) from Kang KA. Child Health Nurs Res. 2022;28(4):291-298 [
20]. CACSs, childhood and adolescent cancer survivors.
Figure 2.Word cloud of unmet heath lifestyle (HLS) practices from topic modeling. (A) Childhood and adolescent cancer survivors (CACSs) and parents (total). (B) Parents. (C) CACSs.
Table 1.General characteristics of participants (N=160)
|
Characteristic |
Category |
N (%) |
|
Child-related characteristics (n=80) |
|
|
|
Sex |
Male |
39 (48.8) |
|
Female |
41 (51.2) |
|
Child age (yr) |
|
12.1±2.0 |
|
Birth order |
First child |
28 (35.0) |
|
Second child |
30 (37.5) |
|
Third or more |
10 (12.5) |
|
Only child |
12 (15.0) |
|
Age at the time of cancer diagnosis |
|
7.48±3.6 |
|
Type of cancer diagnosed |
Leukemia |
47 (58.8) |
|
Lymphoma |
7 (8.6) |
|
Bone tumor |
5 (6.3) |
|
Brain tumor |
4 (5.0) |
|
Nephroblastoma |
3 (3.8) |
|
Others |
14 (17.5) |
|
Last treatment |
Less than 12 months |
16 (20.0) |
|
1 year to 5 years |
29 (36.3) |
|
More than 5 years have passed |
11 (13.8) |
|
Treatment in progress |
24 (30.0) |
|
Treatment receiveda)
|
Chemotherapy |
67 (45.9) |
|
Hematopoietic stem cell transplantation |
28 (19.2) |
|
Surgery |
26 (17.8) |
|
Radiotherapy |
21 (14.4) |
|
Other |
4 (2.7) |
|
Relapse experience |
Yes |
14 (17.5) |
|
No |
66 (82.5) |
|
Health status perceived by CACS |
Very healthy |
10 (12.5) |
|
Healthy |
28 (35.0) |
|
Moderate |
27 (33.8) |
|
Not healthy |
14 (17.5) |
|
Very unhealthy |
1 (1.3) |
|
Source of HLS knowledgea)
|
Parents |
64 (38.8) |
|
Mass media |
49 (29.7) |
|
School |
43 (26.1) |
|
Siblings & friends |
5 (3.0) |
|
Others |
4 (2.4) |
|
Source of HLS educationa)
|
School |
69 (40.1) |
|
Mass media |
55 (31.9) |
|
Parents |
28 (16.3) |
|
Others |
20 (11.6) |
|
AI chatbot experienced |
Yes |
36 (45.0) |
|
No |
44 (55.0) |
|
AI chatbot needed for HLS education |
Necessary |
49 (61.3) |
|
Neither necessary nor unnecessary |
29 (36.3) |
|
Unnecessary |
2 (2.5) |
|
Parental characteristics (n=80) |
|
|
|
Relationship with children |
Father |
4 (5.0) |
|
Mother |
76 (95.0) |
|
Parents’ age (yr) |
|
44.7±4.2 |
|
Place of residence |
Seoul |
17 (21.3) |
|
Gyeonggi-do |
20 (25.0) |
|
Gangwon-do |
3 (3.8) |
|
Chungcheong-do |
6 (7.5) |
|
Jeolla-do |
3 (3.8) |
|
Gyeongsang-do |
10 (12.5) |
|
Jeju-do |
3 (3.8) |
|
Other metropolitan areas |
18 (22.5) |
|
Parents’ education |
Less than high school |
17 (21.3) |
|
College graduation & graduated school |
63 (78.7) |
|
Marital status |
Married |
75 (93.8) |
|
Separation (divorced, bereaved) |
5 (6.2) |
|
Socioeconomic status |
High |
3 (3.8) |
|
Middle |
63 (78.8) |
|
Low |
14 (17.5) |
|
Religion |
Protestant |
21 (26.3) |
|
Buddhist |
13 (16.3) |
|
Catholic |
11 (13.8) |
|
None |
35 (43.8) |
|
Health status perceived by parents |
Very healthy |
6 (7.5) |
|
Healthy |
17 (21.3) |
|
Moderate |
41 (51.2) |
|
Not healthy |
15 (18.8) |
|
Very not healthy |
1 (1.3) |
|
Source of HLS knowledgea)
|
Mass media |
77 (44.8) |
|
School |
32 (18.6) |
|
Friends |
10 (5.9) |
|
Others |
8 (4.7) |
|
Source of HLS educationa)
|
Mass media |
44 (32.6) |
|
School |
39 (28.9) |
|
Others |
15 (11.1) |
|
AI chatbot experienced |
Yes |
32 (40.0) |
|
No |
48 (60.0) |
|
AI chatbot needed for HLS education |
Necessary |
59 (73.8) |
|
Neither necessary nor unnecessary |
19 (23.8) |
|
Unnecessary |
2 (2.5) |
Table 2.Comparative analysis of HLS and A-uC between groups (N=160)
|
Subdomain |
CACSs (n=80) |
Parents (n=80) |
t
|
p
|
|
M±SD |
|
HLS (4-point scale) |
|
|
|
|
|
Spiritual health (3 items) |
3.47±0.61 |
3.41±0.58 |
−0.577 |
.565 |
|
Stress management (5 items) |
3.36±0.51 |
3.44±0.44 |
–0.730 |
.465 |
|
Interpersonal relations (6 items) |
3.36±0.48 |
3.30±0.50 |
1.029 |
.305 |
|
Positive life perspective (5 items) |
3.20±0.58 |
3.18±0.52 |
–0.258 |
.797 |
|
Nutrition (4 items) |
2.91±0.57 |
2.93±0.63 |
0.230 |
.818 |
|
Health responsibility (4 items) |
2.64±0.70 |
2.79±0.47 |
1.601 |
.112 |
|
Physical activity (4 items) |
2.55±0.89 |
2.53±0.81 |
–0.116 |
.908 |
|
Total (31 items) |
3.16±0.80 |
3.18±0.36 |
0.328 |
.743 |
|
A-uC (5-point scale) |
|
|
|
|
|
Perceived ease of use & usefulness (7 items) |
4.19±0.65 |
4.26±0.55 |
0.696 |
.488 |
|
Perceived value (11 items) |
|
|
|
|
|
Professionalism (4 items) |
4.12±0.75 |
4.05±0.67 |
−0.582 |
.562 |
|
Reliability (3 items) |
4.08±0.83 |
4.09±0.68 |
0.104 |
.917 |
|
Empathy (4 items) |
4.00±0.77 |
4.00±0.69 |
0.027 |
.978 |
|
Intention to use (4 items) |
3.94±0.90 |
4.21±0.68 |
2.187 |
.030 |
|
Total (22 items) |
4.06±0.68 |
4.12±0.55 |
0.595 |
.533 |
Table 3.Topic modeling analysis of unmet HLS practices
|
Topic |
No. of document |
Keyword |
Frequent keywords (≥2) |
Named topic by authors |
|
1st keyword |
2nd keyword |
|
CACSs+parents |
1 |
57 |
Exercise |
Activity |
Body, playing, fatigue, posture, walking, activity, exercise, physical strength |
Exercise |
|
3 |
52 |
Eat well |
Fast food |
Limit, soda, overeating, picky eating, water, diet, healthy food, snacks, fast food, eat well, Malatang |
Healthy diet |
|
2 |
51 |
Sleep well |
Early bed time |
Change habit, smartphone, stress, early bed time, sleep well, life, management, regular life |
A regular lifestyle |
|
Parents |
3 |
30 |
Exercise |
Diet |
Smartphone, stress, early bedtime, sleep well, life, management |
A regular lifestyle |
|
2 |
27 |
Exercise |
Sleep well |
Rest, activity, exercise, physical strength, walking |
Exercise |
|
1 |
13 |
Fast food |
Eat well |
Limit, soda, water, picky eating, healthy food, diet, snacks, eat well, fast food, overeating |
Healthy diet |
|
CACSs |
2 |
34 |
Exercise |
Sleep well |
Body, fatigue, playing, walking, posture, exercise, activity |
Exercise |
|
3 |
25 |
Exercise |
Fast food |
Change habit, sleep well, early bedtime, stress |
A regular lifestyle |
|
1 |
21 |
Eat well |
Exercise |
Healthy food, snacks, fast food, eat well, overeating, water, diet |
Healthy diet |
REFERENCES
- 1. Korean Central Cancer Registry. Annual report of cancer statistics in Korea in 2021. National Cancer Center; 2023. Report No.: 11-1352000-000145-10.
- 2. Kim HJ, Lee E. An integrated review of health care in child and adolescent cancer survivors based on Roy’s adaptation model. Asian Oncol Nurs. 2024;24(2):82-93. https://doi.org/10.5388/aon.2024.24.2.82
- 3. Bhakta N, Liu Q, Ness KK, Baassiri M, Eissa H, Yeo F, et al. The cumulative burden of surviving childhood cancer: an initial report from the St Jude Lifetime Cohort Study (SJLIFE). Lancet. 2017;390(10112):2569-2582. https://doi.org/10.1016/S0140-6736(17)31610-0
- 4. Choi Y, Rhee H, Flannery M. Health behaviors in adolescent survivors of cancer: an integrative review. J Pediatr Nurs. 2022;66:e100-e115. https://doi.org/10.1016/j.pedn.2022.05.002
- 5. Oh SM, Park SY, Lee HJ, Lee JH. Measurement properties of self-report questionnaires measuring the social adjustment for youth after treatment of childhood cancer: systematic review. Child Health Nurs Res. 2018;24(1):78-90. http://doi.org/10.4094/chnr.2018.24.1.78
- 6. Ju EY, Min HY. An integrative review of the return to school of childhood and adolescent cancer survivors in Korea. J Korea Acad Ind Coop Soc. 2025;26(4):127-137. https://doi.org/10.5762/KAIS.2025.26.4.127
- 7. Richter F, Kronziel LL, König I, Langer T, Gebauer J. Implementation of regular lifestyle counseling during long-term follow-up care of childhood cancer survivors: monocentric prospective study. JMIR Cancer. 2024;10:e59614. https://doi.org/10.2196/59614
- 8. Kim HY, Kang KA, Han SJ, Chun J. Web-based research trends on child and adolescent cancer survivors over the last 5 years: text network analysis and topic modeling study. J Med Internet Res. 2022;24(2):e32309. https://doi.org/10.2196/32309
- 9. Kang KA, Han SJ, Chun J, Kim HY, Oh Y, Yoon H. Healthy lifestyle interventions for childhood and adolescent cancer survivors: a systematic review and meta-analysis. Child Health Nurs Res. 2023;29(2):111-127. https://doi.org/10.4094/chnr.2023.29.2.111
- 10. Kang KA, Kim SJ, Song I. Healthy lifestyles in childhood cancer survivors in South Korea: a comparison between reports from children and their parents. Child Health Nurs Res. 2022;28(3):208-217. https://doi.org/10.4094/chnr.2022.28.3.208
- 11. Kang KA, Kim HH, Kim SJ, Song IH, Lee MJ, Lee SY, et al. Effectiveness of a healthy lifestyle program based on a mobile serious game for childhood cancer survivors: a quasi-randomized trial. J Pediatr Nurs. 2024;77:35-44. https://doi.org/10.1016/j.pedn.2024.03.006
- 12. An H, Lee S. Returning to social life: development of social identity for adolescent and young adult survivors of leukemia in Korea. J Pediatr Oncol Nurs. 2019;36(1):35-43. https://doi.org/10.1177/1043454218810145
- 13. Joe Y, Kim M. Analysis of trends in domestic and international research on educational use of chatbots: systematic literature review. J Educ Res [Internet]. 2023 [cited 2025 Jan 5];45(3):1-32. Available from: https://scholar.kyobobook.co.kr/article/detail/4040068798284
- 14. Ahn J, Park HO. Development of a case-based nursing education program using generative artificial intelligence. J Korean Acad Soc Nurs Educ. 2023;29(3):234-246. https://doi.org/10.5977/jkasne.2023.29.3.234
- 15. Li J, Dada A, Puladi B, Kleesiek J, Egger J. ChatGPT in healthcare: a taxonomy and systematic review. Comput Methods Programs Biomed. 2024;245:108013. https://doi.org/10.1016/j.cmpb.2024.108013
- 16. Pender NJ, Murdaugh C, Parsons MA. Health promotion in nursing practice. 4th ed. Prentice-Hall; 2002.
- 17. Kang KA, Kim SJ. Psychometric validation of the Child Healthy Lifestyle Profile in South Korea: a cross-sectional study. Child Health Nurs Res. 2023;29(1):37-50. https://doi.org/10.4094/chnr.2023.29.1.37
- 18. Venkatesh V, Davis FD. A theoretical extension of the technology acceptance model: Four longitudinal field studies. Manag Sci. 2000;46(2):186-204. https://doi.org/10.1287/mnsc.46.2.186.11926
- 19. Hussain A, Zhiqiang M, Li M, Jameel A, Kanwel S, Ahmad S, et al. The mediating effects of perceived usefulness and perceived ease of use on nurses’ intentions to adopt advanced technology. BMC Nurs. 2025;24(1):33. https://doi.org/10.1186/s12912-024-02648-8
- 20. Kang KA, Kim SJ, Kang SR. Elementary school students’ awareness of the use of artificial intelligence chatbots in violence prevention education in South Korea: a descriptive study. Child Health Nurs Res. 2022;28(4):291-298. https://doi.org/10.4094/chnr.2022.28.4.291
- 21. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10):e296. https://doi.org/10.1371/journal.pmed.0040296
- 22. Faul F, Erdfelder E, Lang AG, Buchner A. G* Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175-191. https://doi.org/10.3758/BF03193146
- 23. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Lawrence Erlbaum Associates; 1988.
- 24. IBM Corp. IBM SPSS Statistics for Windows. Version 29.0 [software]. IBM Corp.; 2022.
- 25. Cyram Inc. NetMiner. Version 4.5 [software]. Cyram Inc.; 2022 [cited 2025 Jan 5]. Available from: https://www.netminer.com
- 26. Blei DM, Ng AY, Jordan MI. Latent Dirichlet allocation. J Mach Learn Res. 2003;3:993-1022.
- 27. Song IH, Kang KA. Research trends over 10 years (2010-2021) in infant and toddler rearing behavior by family caregivers in South Korea: text network and topic modeling. Child Health Nurs Res. 2023;29(3):182-194. https://doi.org/10.4094/chnr.2023.29.3.182
- 28. Bouwman E, Stollman I, Wilbers J, Claessens JJ, van Spronsen DJ, Bongaerts A, et al. Feasibility and potential effectiveness of nurse-led video-coaching interventions for childhood, adolescent, and young adult cancer survivors: the REVIVER study. BMC Cancer. 2024;24(1):722. https://doi.org/10.1186/s12885-024-12430-3
- 29. Vasilopoulou M, Asimakopoulou Z, Velissari J, Vicha A, Rizogianni M, Pusa S, et al. Interventions about physical activity and diet and their impact on adolescent and young adult cancer survivors: a Prisma systematic review. Support Care Cancer. 2024;32(6):342. https://doi.org/10.1007/s00520-024-08516-0
- 30. Sadak KT, Gemeda MT, Grafelman M, Neglia JP, Freyer DR, Harwood E, et al. Identifying metrics of success for transitional care practices in childhood cancer survivorship: a qualitative interview study of survivors. BMC Cancer. 2020;20(1):898. https://doi.org/10.1186/s12885-020-07360-9
- 31. Sezgin E, Jackson DI, Kaufman K, Skeens MA, Gerhardt CA, Moscato E. Perceptions about the use of virtual assistants for seeking health information among caregivers of young childhood cancer survivors. Digit Health. 2025;11:20552076251326160. https://doi.org/10.1177/20552076251326160
- 32. Xu L, Sanders L, Li K, Chow JC. Chatbot for health care and oncology applications using artificial intelligence and machine learning: systematic review. JMIR Cancer. 2021;7(4):e27850. https://doi.org/10.2196/27850
- 33. Wilson L, Marasoiu M. The development and use of chatbots in public health: scoping review. JMIR Hum Factors. 2022;9(4):e35882. https://doi.org/10.2196/35882
