This study adopted a formative mixed methods design, integrating the development of a serious pregnancy education game with a postrelease evaluation of users that included quantitative survey data and qualitative user feedback. The development period spanned from April 2024 to March 2025, and the game was released in April 2025 simultaneously on the iOS App Store, Google Play Store (Android), and Steam (Windows) platforms. The game was freely available without registration or cost, and users were recruited via an in-game prompt presented upon completing the game. A link to the anonymous postgame survey (hosted on Google Forms) was embedded within the game and displayed at the end of play. No systematic exclusion criteria were applied; all users who completed the game and consented to participate were eligible. The postgame survey was conducted between April 2025 and January 2026.

The game was developed by a multidisciplinary team led by the first author, a pediatrician specializing in neonatology. Scenario development, including field interviews and contextual refinement, was supported by a pediatrician with experience in maternal health and clinical communication. Medical and administrative content were reviewed iteratively at multiple stages of development to ensure accuracy and consistency with current prenatal care practices.

The implementation team consisted of 6 medical students who were primarily responsible for programming, branching logic, and image processing, and 1 student illustrator who created all character illustrations, background art, and visual assets. No professional programmers were involved in the development process.

The game was created using TyranoBuilder, a visual novel engine that enables interactive storytelling, branching narratives, animation, and bilingual text support without requiring advanced programming expertise.

The game development workflow comprised seven sequential steps: (1) scenario design by the first author and a pediatrician with expertise in maternal health; (2) script implementation and branching logic by 6 medical students; (3) illustration production (all characters, backgrounds, and user interface elements) by the student team; (4) medical and administrative review by obstetricians, midwives, and public health nurses; (5) bilingual implementation (Japanese and English); (6) platform optimization for iOS, Android, and Steam; and (7) quality and usability testing by medical students and perinatal care professionals.

This workflow was designed to be reproducible in educational and academic settings with limited technical and financial resources. This resulted in a game that provides a fully illustrated, narrative-driven educational experience with an estimated playtime of approximately 1 hour and is freely accessible without advertisements or in-app purchases.

The game consists of a sequential 9-chapter storyline that simulates the course of pregnancy from early recognition to late-pregnancy checkups and preparation for childbirth. Each chapter includes illustrated scenes, interactive dialogue choices, and an optional TIPS that provides concise medical or administrative information (Figure 1).

The 9 chapters cover the following topics: pregnancy recognition and confirmation (chapter 1), the first obstetric and gynecological visit (chapter 2), informal peer consultation (chapter 3), partner communication and shared decision-making (chapter 4), public health center consultations (chapter 5), mid-pregnancy medical checkups (chapter 6), daily life and safety during pregnancy (chapter 7), preparation of the home environment (chapter 8), and late-pregnancy checkups and delivery planning (chapter 9).

After completing the game, participants were invited to respond to an anonymous online questionnaire via Google Forms. The survey collected demographic information and evaluated gameplay experience and perceived educational usefulness. Survey items are summarized in Table 1.

For gameplay evaluation items (game length, difficulty, and gameplay interactivity), a score of 3 indicated an appropriate or balanced level (eg, appropriate length, appropriate difficulty, or a sufficient level of interactivity). Scores below 3 indicated the game was perceived as too short, too easy, or insufficiently interactive; scores above 3 indicated it was perceived as too long, too difficult, or overly interactive.

For educational usefulness items, higher scores indicated more positive evaluations, with 5 representing the most favorable response.

The game was simultaneously released on the Apple App Store (iOS), Google Play Store (Android), and Steam (Windows).

Analytical data were extracted at a single predefined time point after release to ensure consistency across platforms. Platform-specific metrics were extracted from App Store Connect (Apple Inc), Google Play Console (Google LLC), and the Steamworks Developer Dashboard (Valve Corporation).

Survey data and platform analytics were summarized using descriptive statistics. For Likert-scale items, given the ordinal nature of the data, we report means, SDs, SEs, medians, and IQRs. Categorical variables are summarized as frequencies and percentages. Inferential statistics were not used. All analyses were performed using Microsoft Excel. Participants with missing responses on individual items were included in analyses for all items they completed and excluded only from item-specific calculations. No pre-post testing or between-group comparisons were performed, in accordance with the formative nature of the study.

Steam reviews were analyzed using inductive content analysis conducted by the first author (Y Miyosawa). All 21 reviews available on the Steam platform were included; reviews from iOS and Android platforms were not included, as these platforms do not provide a structured text-based review system equivalent to that of Steam. The analysis followed established procedures for qualitative content analysis [13,14], with initial open coding of all review texts, grouping of codes into subcategories, and abstraction into overarching thematic categories [13,14]. A second reviewer (AF) independently coded a subset of 5 reviews to verify the emerging categories, and any interpretive differences were resolved through consensus discussion to enhance trustworthiness. No formal coding software was used. In our sample, many reviews contained limited substantive content suitable for thematic analysis; only reviews offering substantive qualitative feedback were included in the thematic synthesis. The median length of Steam reviews has been reported as 19 words for positive reviews and 40 words for negative reviews [15].

A total of 65 users completed the postgame questionnaire; their characteristics are summarized in Table 2. The most common age group was 10 to 19 years (38/65, 58.5%), followed by 20 to 29 years (14/65, 21.5%), 40 to 49 years (7/65, 10.8%), 30 to 39 years (5/65, 7.7%), and ≥60 years (1/65, 1.5%). Most participants identified as female (55/65, 84.6%), with 4 (6.2%) identifying as male and 6 (9.2%) selecting a sex option not listed as female or male. The sex question offered 3 response options (female, male, and an open category); future surveys should include more inclusive response options in line with best practices for gender-diverse reporting. The largest occupational group was high school students (32/65, 49.2%), followed by non–health care professionals (14/65, 21.5%). Health care professionals and students collectively accounted for 11 (16.9%) participants, comprising physicians (3/65, 4.6%), nurses (3/65, 4.6%), medical students (2/65, 3.1%), nursing students (2/65, 3.1%), and other health care workers (1/65, 1.5%). Smartphones were the most commonly used device (42/65, 64.6%), followed by tablets (12/65, 18.5%) and PCs (7/65, 10.8%). Regarding playtime, 30 (46.2%) completed the game in less than 1 hour, 18 (27.7%) in 1 to 2 hours, 3 (4.6%) in 2 to 4 hours, and 4 (6.2%) in more than 4 hours. Ten participants (15.4%) did not report playtime; these participants were retained in all other analyses.

Figure 2 depicts gameplay evaluation scores from the postgame survey conducted among users (Japan and international users) of the pregnancy education serious game from April 2025 to January 2026.

Gameplay-related ratings were generally near the neutral midpoint of 3.0 on the 5-point scale. For game length, the mean score was 3.37 (SD 0.96) and the median score was 3 (IQR 3-4). For difficulty, the mean score was 2.85 (SD 0.85) and the median score was 3 (IQR 2-3). For gameplay interactivity, the mean score was 3.31 (SD 1.10) and the median score was 3 (IQR 3-4).

Figure 3 depicts perceived educational usefulness scores from the postgame survey conducted among users (Japan and international users) of the pregnancy education serious game from April 2025 to January 2026.

Perceived educational usefulness indicators received consistently favorable ratings. For reduced anxiety, the mean score was 3.84 (SD 0.96) and the median score was 4 (IQR 3-4; n=57). For perceived educational usefulness, the mean score was 3.98 (SD 1.02) and the median score was 4 (IQR 3-4; n=65). For perceived knowledge acquisition, the mean score was 4.06 (SD 1.06) and the median score was 4 (IQR 3-4; n=65). For story empathy, the mean score was 3.80 (SD 1.11) and the median score was 4 (IQR 3-4; n=65). For overall satisfaction, the mean score was 4.05 (SD 1.04) and the median score was 4 (IQR 3-4; n=59). Higher scores indicate more positive evaluations, with 5 representing the most favorable response. The reduced anxiety and overall satisfaction items had slightly lower response counts (n=57 and n=59, respectively) due to missing data.

An exploratory subgroup analysis by age (teenagers vs adults) did not reveal any substantial differences in gameplay evaluation or perceived educational usefulness between the 2 groups (Table S1 in Multimedia Appendix 1).

Across iOS, Android, and Steam, the game had achieved 925 cumulative downloads by February 5, 2026. Statistics regarding downloads are summarized in Table 3 below.

On the iOS and Android platforms, downloads were predominantly recorded in Japan, whereas Steam downloads showed a more geographically diverse distribution.

A total of 21 reviews were posted on Steam, of which 20 (95.2%) were positive. Content analysis of the 21 Steam reviews indicated that the majority were short and contained limited substantive content for thematic analysis, consistent with previously reported median review lengths on the Steam platform [15]. Among the subset of reviews containing substantive content (5/21, 23.8%), recurring positive themes included the game’s educational value and the accessibility of pregnancy-related information. Representative positive comments included: “Very educational and cute!”; “Learned a lot!!! its very good and provides thorough explanations on the experience of pregnancy”; and “Interesting and fun.” The single negative review cited technical usability concerns, including auto-advancing text messages, audio interruptions, and unexpected language switching, which were identified as priority targets for future improvement. iOS reviews comprised 2 five-star ratings and 1 one-star rating; Android received no reviews.

This formative study demonstrated that a serious pregnancy education game developed through a low-cost clinician-student collaborative model achieved broad user acceptance and high perceived educational usefulness. The game attracted a diverse user population, with particularly high engagement among teenagers, who constituted most survey respondents. Gameplay characteristics were rated near the neutral midpoint, reflecting balanced but modest interactivity, while indicators of perceived educational usefulness and overall satisfaction received consistently favorable ratings. The game achieved substantial international reach, with cumulative downloads from more than 40 countries across 3 platforms, and received predominantly positive ratings on Steam. Together, these findings suggest that narrative-based serious games may serve as accessible and scalable tools for maternal health education, even when developed with limited technical and financial resources.

A distinctive feature of our study was the high proportion of teenage users (58% aged 10‐19 years), most of whom were nonpregnant high school students. This demographic diverges from the game’s primary design intent of simulating the pregnancy experience for expectant individuals but underscores a valuable secondary use case: reproductive health literacy education among adolescents. Early exposure to accurate pregnancy-related information during adolescence may contribute to future health literacy and informed decision-making, consistent with growing evidence that digital game interventions can improve knowledge, attitudes, and self-efficacy among young people. It should be noted that the current findings primarily reflect acceptability and perceived usefulness among youth and nonpregnant users; generalizability to pregnant adults—the intended target population—remains to be established. Future deployment strategies should consider differentiated approaches for these 2 populations: school-based or community programs for adolescent reproductive health literacy, and prenatal clinic integration for pregnant women seeking to navigate the pregnancy journey.

Although overall evaluations were positive, feedback from some users indicated a desire for increased gameplay interactivity or additional minigame components, reflected in the near-neutral interactivity score (mean 3.31, SD 1.10 and median 3, IQR 3-4). Similar critiques have been noted in women’s health serious game literature, where insufficient interactivity may limit engagement and educational impact [11]. Consistent with recommendations from prenatal gamification trials, future iterations may consider incorporating immersive minigames, nonlinear or branching decision paths, audio narration or animation, and adaptive learning sequences tailored to user characteristics. Enhancing gameplay depth may further strengthen educational engagement and potential behavioral impact, as suggested by previous serious game interventions for pregnancy and adolescent sexual health [20-22].

This study has several limitations. Survey participation was voluntary, raising the possibility of self-selection bias. The sample size (n=65) was modest and skewed toward teenagers. No preintervention or postintervention comparisons or control group designs were used. Platform analytics, particularly for Android, contained inconsistencies requiring reliance on the most stable snapshot. Long-term retention and behavioral outcomes were not evaluated, echoing the gaps identified in prior game reviews, including limited long-term follow-up and heterogeneous evaluation designs. Additionally, playtime data were self-reported via the survey rather than platform tracked; participants who abandoned the survey midway would not have contributed playtime data, precluding a meaningful correlation analysis between gameplay scores and missing playtime values.

Future work may incorporate additional interactivity, minigames, and audiovisual enhancements to further enrich the learning experience. More rigorous evaluation designs, including pretesting and posttesting or randomized comparisons, are important to better quantify the educational impact. Localization and cultural adaptation may broaden international reach. Building on these findings, collaboration with schools and community health organizations could facilitate integration into adolescent reproductive health and perinatal education programs, in line with prior evidence supporting the feasibility and impact of serious games and digital interventions for youth health promotion.

This study demonstrates that a low-cost, clinician-student collaborative serious game can deliver meaningful pregnancy education and achieve broad user engagement, including at the international level. The positive reception across platforms suggests that narrative-based simulations represent a promising approach to maternal health education and public outreach. Future research should incorporate preintervention and postintervention assessments or randomized comparisons and explore opportunities for localization, cultural adaptation, and integration into school-based or community reproductive health programs to enhance educational impact and broaden international applicability.