We evaluated the usability and user-friendliness of Picterus JP through a prospective mixed methods study using the validated mHealth App Usability Questionnaire (MAUQ), open-ended questions, interviews, and focus groups.

This study was part of the Jaundice@Home project conducted under the Transforming Health and Care Systems program (project number: 1477) [24]. Participants were recruited from 3 distinct real-world newborn care settings: the maternity wards of 2 hospitals in the Netherlands (Frisius Medical Center, Leeuwarden, and University Medical Center Groningen [UMCG]) and 1 in Israel (Shamir Medical Center). This study was initially conducted in the Netherlands from August 2023 to December 2023 (period 1) and extended to Israel from October 2024 to January 2025 (period 2). To ensure comparability, this study was also repeated at UMCG in the Netherlands in the second period.

Parents and/or caregivers were invited to participate if their neonates were healthy, with a gestational age of 35 or more weeks and between the age of 12 hours and 7 days. At least one parent was required to have sufficient proficiency in Dutch (in the Netherlands) or Arabic or Hebrew (in Israel).

Parents were asked to use Picterus JP at least once during their hospital stay. Immediately afterward, parents completed a paper questionnaire about their experiences, which was collected within 24 hours. The app could only be used by parents during their hospital stay for practical and safety considerations. All participating neonates received standard care to ensure safe practice. This included visual inspection for signs of jaundice and TcB and/or TSB measurements as deemed necessary by the attending physicians or midwives. The use of the Picterus JP app did not affect standard care.

All parents received a Picterus calibration card, a download link, and the same instructions for using Picterus JP at their convenience. The instructions consisted of a brief verbal explanation; textual instructions (via the patient information letter or app); and pictograms, which were included in the app and printed out [25]. The study team did not provide explicit demonstrations or guidance before and during app use. Nurses received a demonstration of the app at the beginning of the study period. The study team and nurses were encouraged to leave the room while parents used the app but would sometimes remain as silent bystanders. Parents either installed the app on their own smartphones (the Netherlands) or used a Samsung Galaxy A50 smartphone provided by the researchers (Israel).

When starting Picterus JP, users were instructed to place the square, colorful Picterus calibration card on the infants’ bare chest (Figure 1). Picterus JP then guided users on-screen on how to correctly position their smartphone (ie, angle and distance). Once the app detected that the conditions were appropriate (ie, correct placement of the calibration card and positioning of the smartphone and sufficient ambient light), it would then automatically proceed to take 6 photos: 3 with flash and 3 without.

After capturing the images, the app either displayed a bilirubin value (in μmol/L or mg/dL depending on the setting) or issued an error message. If an error occurred, the app either stated that the measurement was unsuccessful or, if possible, provided a reason (eg, insufficient ambient light or shadows rendering the photo unsuitable for analysis). During the photo-taking phase, the app attempted to correct suboptimal conditions by offering simple on-screen instructions to the user. If image quality remained insufficient, Picterus JP stopped the standard 6-image count and either continued attempting automatic image capture or temporarily paused image capture. Image counting resumed only once image quality improved following user adjustment. If adequate image quality could not be achieved, the measurement attempt was eventually aborted and an error message was displayed, when available.

This study was conducted in accordance with the Declaration of Helsinki. In the Netherlands, the study was approved by the Central Ethical Review Committee of UMCG (research register number 10763). This study was classified as falling outside the scope of the Medical Research Involving Human Subjects Act. Local feasibility was confirmed by the Regional Ethics Review Committee at Frisius Medical Center (reference: RTPO 1160). In Israel, the study was approved by Shamir Medical Center Helsinki Committee (approval number 264-23-ASF). Informed consent was obtained from the parents of all participating neonates.

In total, 123 parents completed the MAUQ and open-ended questions. Table 1 shows participant characteristics. The sample included both first-time and experienced parents, with a relatively high proportion (78/123, 63.4%) of participating parents having completed higher education. In total, 16 parents were approached for in-depth interviews: 12 (75%) participated after 2 reminders across the 3 centers. Data saturation was reached after approximately 5 interviews.

The 2 focus groups included 12 nurses, all Dutch and female. Nurses had sometimes observed when parents used the app as they had been caring for mother and infant or had heard parents discuss app use afterward. Table 2 shows nurse characteristics. Overall, the nurses were relatively young and had few years of work experience.

Figure 2 shows the median MAUQ usability score of 6 (IQR 5-7) on a scale from 1 to 7 on the MAUQ. Similarly, the median scores for the subscales of ease of use, interface and satisfaction, and usefulness were 6 (IQR 5-7). There were no differences in overall median MAUQ scores or MAUQ scores per subscale between the 2 countries, among the 3 hospitals, or between both study periods.

Parents expressed the most positive attitudes toward the statements that Picterus JP was easy to learn and improved their access to health care, with a median MAUQ score of 7 (IQR 6-7). All other statements had a median MAUQ score of 6. Table S1 in Multimedia Appendix 1 shows all 17 MAUQ statements and their corresponding median scores.

We determined how parents perceived the usability and user-friendliness of Picterus JP, a novel smartphone app to allow for early detection of imminent SNH. Our results show that parents rated Picterus JP highly on usability, supported by a median MAUQ score of 6 (IQR 5-7) out of 7. Parents described Picterus JP as easy, simple, and quick to use.

A considerable number of parents reported either technical difficulties or movement of their baby as reasons for failing to obtain a result on the first use of Picterus JP.

Technical difficulties reported by users may be partly explained by the variation in smartphone devices used across study sites despite the fact that these devices passed Picterus JP’s compatibility check. However, some phones (eg, budget models) may still be more sensitive to ambient lighting conditions. In Israel, all parents used the same midrange phone provided by the study team, whereas in the Netherlands, varied personal devices were used. This difference may explain the lower incidence of technical difficulties in the Netherlands than in Israel. Furthermore, within the Dutch sites, technical difficulties decreased in the second study period, possibly due to ongoing software improvements of Picterus JP.

In addition to technical difficulties, some parents (22/120, 18.3%) expressed concerns regarding neonatal well-being during app use. Frequent topics were concerns about the infant becoming cold or upset. Most of these parents (20/23, 87%) reported having encountered difficulty using Picterus JP on their first attempt. This may suggest a possible connection between prolonged attempts and parental concerns, but this was not formally studied.

Although parents and nurses found the Picterus JP instructions clear, they reported that instructions could be overlooked or misunderstood. It is possible that, in some cases, unsuccessful attempts were related to skipped or misunderstood instructions rather than technical failure alone. Picterus JP offers multiple instructions: symbols on calibration cards and the app launch screen, online text, and in-app pop-ups. It is tempting to speculate that passive instructional methods may not fully support parents in busy postpartum settings.

Nonetheless, most parents (100/114, 87.7%), even among those who encountered technical difficulties or had concerns, stated that they would use the app at home. This implies that, despite initial barriers, overall acceptability and perceived usefulness of Picterus JP in a home setting are high, which are both essential prerequisites for its implementation. Given the near-universal smartphone use in our study population, eventual barriers to implementation are unlikely to stem from limited access to or lack of experience with smartphones [28].

Notably, the nurses participating in focus groups were young and had limited work experience, reflecting the composition of junior and senior staff typically present during shifts. Demographic characteristics, with younger nurses potentially being more technologically savvy, may influence generalizability but also improve comparability with similarly aged parents.

Nevertheless, all nurses had experience with SNH screening and regarded Picterus JP as relevant and useful. They emphasized that parents engaged more with Picterus JP when its relevance was clear to them too. Moreover, insights from nurses helped contextualize parental responses as they frequently observed parents during the use of Picterus JP. This underlines the importance of incorporating multiple perspectives in usability studies.

While many SNH screening applications have been developed over the years, our findings align with broader digital health literature showing that the main challenges of home-based screening tools lie in usability [29]. Despite the widely recognized need for accessible screening tools to detect SNH, still few clinically validated alternatives currently exist. Although noninvasive screening can be performed using TcB devices, these devices are rather costly and currently not available in home settings [7]. To save costs, 2-color and multicolor icterometers (ie, the BiliStrip and Bili-Ruler, respectively) have been developed with rather acceptable screening performances and interrater reliability [14,17]. In parallel with these icterometers, other smartphone apps to optically estimate hyperbilirubinemia have been developed and studied over the past years [7,15,16]. Except for Picterus JP, to our knowledge, none have currently published usability results or reached a clinical implementation phase and obtained a CE mark or Food and Drug Administration certification.

Our findings on the usability of Picterus JP are consistent with those of the 2 prior studies to date in which the use of Picterus JP was assessed by HCPs. HCPs perceived Picterus JP as an easy-to-use, useful, and time-saving tool in the management of neonatal jaundice [30,31]. Our study builds on previous work by including the perspectives of both HCPs and parents in high-income countries.

Usability studies are a relevant step for implementation given that apps that are considered easy to use have yielded higher acceptability and satisfaction ratings [32]. However, studies linking app usability to clinical outcomes in pediatrics remain scarce and are mainly focused on chronic disease management [33,34]. The MAUQ is a validated and reliable instrument specifically developed to evaluate mHealth apps used by patients and HCPs, unlike more general instruments such as the System Usability Scale and the Post-Study System Usability Questionnaire [26].

The MAUQ has increasingly been adopted in recent years and used for studies across diverse health care settings worldwide [35]. High MAUQ scores have been associated with proper test performance in another mHealth screening app [36] This suggests that high usability scores as measured using the MAUQ could be a relevant indicator for evaluating the feasibility of successful implementation of a screening mHealth tool such as Picterus JP.

This study has several limitations. First, we used convenience sampling, which may limit representativeness, although saturation was observed in the open-ended responses, interviews, and focus groups. The design may have led to selection bias: participants were motivated to take part, comfortable with smartphone use, less frequently first-time parents, and noticeably highly educated on average. We did not observe a clear association between educational level and MAUQ score or between educational level and parental experience (ie, failure to obtain a measurement). This does not eliminate the concern that less technologically savvy or lower-literacy parents may have been systematically underrepresented. Future studies should, therefore, specifically evaluate usability and feasibility in more diverse populations and care settings, including lower-income countries.

Second, the observational nature of our study design limited us to reporting descriptive findings rather than statistical cross-site comparisons or adjusting for potential confounders.

Third, MAUQ scores were generally high (median 6, IQR 5-7 on a 7-point scale) but with no complete concentration at the maximum score, suggesting no clear evidence of a ceiling effect.

Finally, the app was typically used for only 1 measurement in a hospital setting, with limited instruction with the purpose of generalizing the outcomes. One in-hospital measurement with either the study team or an HCP nearby may introduce performance bias and limit ecological validity. Although the study team was not always directly present during measurements, the potential influence of supervision was not systematically assessed. It is important to note that, in real-world settings, parents are also likely to use the app in interaction with HCPs such as maternity care assistants, who provide daily postpartum care in the Netherlands, or during telehealth consultations. The intended repeated daily use of Picterus JP in variable settings might influence user experience differently. Although repeated use might expose the user to other user- and device-related difficulties, familiarity through guidance and practice may also enhance perceived parental usability of the app.

To the best of our knowledge, this is the first study to assess the usability of a smartphone-based bilirubin screening tool from the perspective of both parents and HCPs. Furthermore, the multicenter and multicultural setting spanning 2 countries and 3 hospitals adds to the potential generalizability of the results to different health and care systems as similar usability patterns emerged across diverse user groups. In addition, we used a validated instrument (MAUQ) to evaluate usability, which enhances the reliability and comparability of our findings.

Our findings show that parents and nurses highly appreciated the Picterus JP app despite some device- and user-related difficulties. Our findings highlight the importance of ongoing technological improvements, clear user guidance, and proactive support to improve usability and user experience to successfully facilitate the implementation of parental use of Picterus JP in neonatal jaundice management.

If future studies confirm the clinical accuracy and reliability of parental use of Picterus JP in different populations and health and care systems, daily noninvasive home screening could become feasible. As such, the use of Picterus JP may enable earlier detection of imminent SNH, allowing for timely intervention for at-risk neonates, reducing the emotional and logistical burden on families, and supporting clinical decision-making by HCPs.