This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.
Neonatal hyperbilirubinemia is one of the leading causes of neonatal readmission—especially severe hyperbilirubinemia and its complications—and it influences disease burden as well as neonatal and maternal health. Smartphones have been shown to have satisfactory accuracy in screening neonatal bilirubin levels, but the impact of this technology on neonatal health care service and maternal health outcomes is still unknown.
The aim of this study was to evaluate the impact of a smartphone-based out-of-hospital neonatal jaundice screening program on neonatal readmission rates for jaundice and related maternal anxiety.
This was a 2-arm, unblinded, randomized controlled trial with 30 days of intervention and follow-up periods. From August 2019 to August 2020, healthy mother-infant dyads were recruited on-site from 3 public hospitals in Hainan, China. Intervention group mothers used the smartphone app to routinely monitor neonatal jaundice at home under the web-based guidance of pediatricians. Control group participants received routine care. The primary study outcome was the neonatal readmission rate due to jaundice within 30 days of the first hospital discharge. The secondary outcome was the maternal anxiety score associated with neonatal jaundice. The data were collected through a self-assessed questionnaire. All participants were included in the analysis (intention-to-treat).
In this study, 1424 mother-infant dyads were recruited, comprising 1424 mothers and 1424 newborns. The median age
of the mothers was 29 (IQR 26-32) years, and there were 714 (50.1%) male neonates. These mother-infant dyads were randomly
assigned to the intervention group and the control group, with 712 dyads in each group; only 1187 of these dyads completed the
follow-up. We found that the adjusted 30-day neonatal readmission rate due to jaundice reduced by 10.5% (71/605, 11.7% vs
141/582, 24.2%; 95% CI 5%-15.9%; odds ratio 0.4, 95% CI 0.3-0.5;
Our study shows that the smartphone-based out-of-hospital screening method for neonatal hyperbilirubinemia decreased the neonatal readmission rate within 30 days from the first discharge and improved maternal mental health to some degree, thus demonstrating the usefulness of this screening app for follow-up in pediatric care.
China Clinical Trial Registration Center, ChiCTR2100049567; http://www.chictr.org.cn/showproj.aspx?proj=64245
Neonatal hyperbilirubinemia (or neonatal jaundice) is a clinical condition in which the skin, sclera, and mucous membranes become yellowish as a result of elevated serum bilirubin levels caused by abnormalities in bilirubin metabolism [
Early identification of high-risk neonates, appropriate follow-ups, and timely intervention contribute to preventing the deterioration of neonatal hyperbilirubinemia [
Mobile communication devices have been widely used in the health care sector [
This study was an unblinded, parallel-assignment RCT conducted at 3 public hospitals in Hainan, China from August 2019 to September 2020. Participants were randomly assigned to either the smartphone-based intervention group or the routine care control group and were followed up for 30 days (
The ethics committee of the Hainan Women and Children’s Medical Center in Hainan approved the trial protocol (institutional review board approval HNWCMC201605), and all participants provided written informed consent. The trial registration number is ChiCTR2100049567 in the China Clinical Trial Registration Center database.
Neonates born in Hainan Women and Children’s Medical Center, Wanning People’s Hospital, and Chengmai People’s Hospital and their mothers were recruited as participants. All mother-infant dyads met the hospital discharge criteria as judged by the bedside doctor, and they were about to be discharged home. The eligible mother was at least 16 years old with clear understanding and communication skills, owned a smartphone, and was able to use it. Exclusion criteria included multiple births (eg, twins); neonates with congenital disease, perinatal asphyxia, neonatal infection, hemolytic disease (positive Coombs test), or other serious organ damage; or family not living in Hainan within the study period. Qualified mother-infant dyads were identified and screened face-to-face by trained obstetric nurses prior to discharge from the hospital. All enrolled persons were aware of the trial objective and process, and written informed consent was obtained from mothers at study entry.
After baseline information was collected, the trained obstetric nurses carried out simple randomization (draw of lots) and allocated each mother-infant dyad to the intervention and control groups. Because of the nature of the intervention, it was not possible to blind participants or the pediatric nurses who were responsible for participant recruitment and follow-up.
Participants randomized to the control group received routine care. Under the National Basic Public Health Service Program in China [
Participants randomized to the intervention group received routine care and the smartphone-based, family-physician collaborative neonatal jaundice screening program, that is, under the remote guidance of a pediatrician, mothers routinely monitored neonatal jaundice at home by using a smartphone app. The jaundice mobile monitoring app used in this study is a publicly available, free downloadable software in China that allows remote, noninvasive, and self-service jaundice monitoring and early warnings. It is used in combination with a jaundice colorimetric card, which is required to be placed on the chest of the child. After judging the image quality, light environment, and skin area, the app will automatically scan and take photos to obtain a clear image of the newborn skin. Once the cloud server receives the image data, the jaundice value will be automatically calculated and displayed (automated image-based bilirubin test), along with an indication of the risk level of neonatal jaundice (low risk, 2.6-10.1 mg/L; medium risk, 10.2-17.1 mg/L; high risk: ≥17.2 mg/L). A clinical trial conducted to evaluate the screening accuracy of this app found strong concordance between automated image-based bilirubin test and TSB levels (
Intervention group mothers received free appropriative jaundice colorimetric cards and installed the jaundice mobile monitoring app on their smartphones at hospital discharge. They were instructed by trained obstetric nurses about the data uploading method. They were required to detect neonatal bilirubin values by using the app, when needed, after returning home. We assigned each mother a pediatrician from the hospital where the delivery took place and bound the mother’s and the doctor’s versions of the jaundice mobile monitoring app together; both parties could use the app platform for communication and consultation. Pediatricians would be able to access the dynamic changes and risk prompts of neonatal bilirubin values synchronously, so as to judge the infants’ situation and guide the mother to take the correct preventive measures. If neonatal bilirubin level was considered to be at a high-risk level or the condition was quite severe, the pediatrician would alert the mother to bring the child to the hospital for further examination and specialist treatment. In addition, pediatricians would supervise and remind mothers to measure newborn infants’ jaundice level every day through the app.
On the 31st day after the mother-infant dyad was discharged from the hospital, the maternal and neonatal outcomes were collected by trained obstetric nurses through telephone follow-up, and the last questionnaire survey was completed in September 2020. The primary outcome was the neonatal readmission rate due to jaundice, which was defined as the ratio of the number of neonates readmitted to hospitals for jaundice within 30 days of the first discharge to the total number of neonates in each group. Mothers were asked to answer, “Has your baby gone to the hospital again because of jaundice in the past 30 days?” and the corresponding options were yes or no. The secondary outcome was the maternal anxiety score associated with neonatal jaundice, which was measured by a self-designed scale (
A sample size calculation was performed based on the primary outcome. Based on previous studies [
All analyses followed the intention-to-treat principle, including all the randomly assigned mother-infant dyads. We examined the distributions of the baseline characteristics by using descriptive statistics and Pearson chi-square test; 2-sided Student
From August 2019 to August 2020, we screened 5635 mother-infant dyads in 3 public hospitals and 1424 dyads were eligible (
Consolidated Standards of Reporting Trials (CONSORT) diagram for this study.
The median age of the mothers was 29 (IQR 26-32) years; 32.3% (460/1424) of the mothers lived in the rural areas, and the mean maternal anxiety score associated with neonatal jaundice was 19.6 (SD 6.3) (
Baseline characteristics of the mothers.a
Variables | Overall (N=1424) | Intervention group (n=712) | Control group (n=712) | ||||
Age (years), median (IQR) | 29 (26-32) | 29 (26-32) | 29 (26-32) | ||||
|
|||||||
|
Han nationality | 1338 (94.1) | 670 (94.4) | 668 (93.8) | |||
|
Minority | 84 (5.9) | 40 (5.6) | 44 (6.2) | |||
|
|||||||
|
Junior high school or below | 376 (26.4) | 190 (26.7) | 186 (26.1) | |||
|
High school or technical secondary school | 332 (23.3) | 167 (23.5) | 165 (23.2) | |||
|
Junior college | 338 (23.7) | 170 (23.9) | 168 (23.6) | |||
|
Undergraduate or above | 378 (26.5) | 185 (26) | 193 (27.1) | |||
|
|||||||
|
Formally employed | 634 (44.6) | 315 (44.3) | 319 (44.8) | |||
|
Temporary employed | 77 (5.4) | 33 (4.6) | 44 (6.2) | |||
|
Unemployed | 712 (50) | 363 (51.1) | 349 (49.1) | |||
|
|||||||
|
Yes | 1254 (88.2) | 621 (87.5) | 633 (88.9) | |||
|
No | 168 (11.8) | 89 (12.5) | 79 (11.1) | |||
|
|||||||
|
Urban | 964 (67.7) | 479 (67.3) | 485 (68.1) | |||
|
Rural | 460 (32.3) | 233 (32.7) | 227 (31.9) | |||
|
|||||||
|
Yes | 1246 (87.6) | 616 (86.8) | 630 (88.5) | |||
|
No | 176 (12.4) | 94 (13.2) | 82 (11.5) | |||
|
|||||||
|
High income | 51 (3.6) | 25 (3.5) | 68 (9.6) | |||
|
Middle income | 1230 (86.5) | 612 (86.1) | 618 (86.8) | |||
|
Low income | 142 (10) | 74 (10.4) | 26 (3.7) | |||
Maternal anxiety scoreg, mean (SD) | 19.61 (6.3) | 19.79 (6.3) | 19.42 (6.3) |
a
bNationality was unknown for 2 intervention group mothers.
cEmployment status was unknown for 1 intervention group mother.
dMedical insurance was unknown for 2 intervention group mothers.
eConvenience of access to health care represents self-assessed convenience of access to health care facilities from residence by mothers; this was unknown for 2 intervention group mothers.
fEconomic status was unknown for 1 intervention group mother.
gIncluding mothers whose children developed jaundice at baseline (604 in the intervention group, 608 in the control group, as shown in
Baseline characteristics of the neonates.a
Variables | Overall (N=1424) | Intervention group (n=712) | Control group (n=712) | ||||
|
|||||||
|
Male | 714 (50.1) | 362 (50.8) | 352 (49.4) | |||
|
Female | 710 (49.9) | 350 (49.2) | 360 (50.6) | |||
Gestational ageb (weeks), median (IQR) | 39.1 (38.57-39.9) | 39.1 (38.43-39.86) | 39.1 (38.6-40.0) | ||||
Birth weightc (grams), median (IQR) | 3200 (2900-3450) | 3150 (2950-3450) | 3200 (2900-3450) | ||||
|
|||||||
|
Breastfeeding | 1062 (74.6) | 523 (73.5) | 539 (75.8) | |||
|
Nonbreastfeeding | 28 (2) | 13 (1.8) | 15 (2.1) | |||
|
Mixed feeding | 333 (23.4) | 176 (24.7) | 157 (22.1) | |||
|
|||||||
|
First child | 630 (44.2) | 317 (44.5) | 313 (44) | |||
|
Second child | 677 (47.5) | 338 (47.5) | 339 (47.6) | |||
|
Third child or later | 117 (8.2) | 57 (8) | 60 (8.4) | |||
|
|||||||
|
Not present | 211 (14.8) | 108 (15.2) | 103 (14.5) | |||
|
Present | 1212 (85.2) | 604 (84.8) | 608 (85.5) | |||
|
|||||||
|
Natural delivery | 899 (63.1) | 476 (66.9) | 482 (67.7) | |||
|
Cesarean section | 466 (32.7) | 236 (33.2) | 230 (32.3) |
a
bGestational age was unknown for 1 intervention group neonate and 1 control group neonate.
cBirth weight was unknown for 2 intervention group neonates and 2 control group neonates.
dFeeding pattern was unknown for 1 control group neonate.
eStatus of jaundice before discharge represents if the neonate developed jaundice at baseline; this was unknown for 1 control group neonate.
Within 30 days of the first hospital discharge, 68.1% (412/605) and 70.1% (408/582) of the neonates in the intervention and control groups, respectively, who were successfully followed up showed jaundice symptoms, with no statistical difference between the 2 groups. In comparison with the control group, the smartphone-based intervention group was significantly associated with a decrease (141/582, 24.2% vs 71/605, 11.7%, respectively; risk difference=12.5%, 95% CI 8.2%-16.8%) in the neonatal readmission rate due to jaundice (OR 0.4, 95% CI 0.3-0.6). After adjusting the model, our intervention remained observably effective in reducing the risk of readmission (OR 0.4, 95% CI 0.3-0.5), although the degree of reduction was smaller than before (risk difference=10.5%, 95% CI 5%-15.9%) (
Differences in the primary and secondary outcomes between the 2 groups.
Outcomes | Intervention group (n=605) | Control group (n=582) | Unadjusted | Adjusted | ||
|
|
|
Odds ratio (95% CI) or β (95% CI) | Differencea (95% CI) | Odds ratio (95% CI) or β (95% CI) | Differencea (95% CI) |
Primary outcome, n (%) | 71 (11.7) | 141 (24.2) | 0.4 (0.3 to 0.6)b | 12.5% (8.2% to 16.8%) | 0.4 (0.3 to 0.5)b | 10.5% (5% to 15.9%) |
Secondary outcomec, mean (SD) | 16.8 (4.2) | 20.5 (7.4) | –3.7 (–4.6 to –2.9)b | –3.7 (–4.6 to –2.9) | –3.6 (–4.5 to –2.8)b | –3.6 (–4.4 to –2.8) |
aDifference represents risk difference for primary outcome and mean difference for secondary outcome. Risk difference represents the absolute value of the difference in the neonatal readmission rates between the 2 groups. Mean difference represents the difference in the mean maternal anxiety scores between the 2 groups.
b
cIncluding mothers whose children developed jaundice symptoms after hospital discharge (intervention group, n=412; control group, n=408).
Maternal anxiety scores associated with neonatal jaundice in the smartphone-based intervention group were apparently lower than those in the control group (mean difference=–3.7, 95% CI –4.6 to –2.9). The adjusted model showed a similar intervention effect (mean difference=–3.6, 95% CI –4.5 to –2.8) (
Changes in the adjusted mean maternal anxiety score from baseline to the 31st day of follow-up (n=695). Points and error bars show the least squares mean and 95% CIs, respectively, that were derived from the adjusted linear mixed-effects repeated measures model.
More than half of intervention group mothers rated the jaundice mobile monitoring app as relatively or very convenient (331/580, 57.1%) and relatively or very credible (295/597, 51%) in measuring their child’s jaundice. Approximately 87.1% (504/579) of the mothers were willing to recommend this app to people in need, and 87.7% (508/579) held a positive attitude toward the necessity of promoting this app on a wide scale. In terms of satisfaction, only 11.1% (64/579) of the mothers showed dissatisfaction with the app (
In this smartphone-based RCT, we found that the intervention had a significant impact on reducing the neonatal readmission for jaundice within 30 days of the initial discharge, wherein the readmission rate in the intervention group was 10.5% (95% CI 5%-15.9%) lower than that in the control group. In contrast with those in the control group, maternal anxiety symptoms induced by neonatal jaundice were also less severe in mothers who had installed and used the jaundice mobile monitoring app. Following-up monitoring of neonatal bilirubin levels is conducive to improving the outcome of neonatal hyperbilirubinemia, including reducing readmission rates [
According to some surveys, the shorter the average length of postpartum hospital stay, the higher was the risk of 30-day neonatal readmission [
Our intervention has proved to be effective in maintaining maternal mental health. Other studies have shown that family-centered parent-involved daily neonatal health care is positively associated with clinical benefits for infants (decreased readmission rate) and improved parental outcomes (decreased anxiety, depression, and stress) [
The application of the smartphone app in the monitoring of neonatal jaundice fits in with the widespread use of mobile technology currently; the global internet usage rate has reached 65.6% as of March 31, 2021 [
Although this jaundice mobile monitoring app can be regarded as a good auxiliary screening tool for neonatal jaundice in some cases, there is greater scope for improving its acceptability, because the proportion of mothers who explicitly expressed satisfaction with the app was less than 50%. The potential reason for this low satisfaction rate may be that the app was developed by a non–health care provider, leading to some concerns about the accuracy and credibility of its information. In this regard, we suggest that on the premise of ensuring the quality of jaundice screening technology, telemedicine guidance from health care professionals should be facilitated and related units should optimize their publicity and promotion strategies.
This study, to our knowledge, is the first RCT to analyze the effects of a smartphone app on neonatal health resource utilization and maternal mental health. Another strength of this study is the inclusion of large samples and enough statistical power to test the effects of the intervention.
Although our research results have important clinical implications, several limitations should be recognized. First, as we did not restrict the hospitals to which participants were readmitted, it was not possible to obtain and check all neonates’ visit records in the 3 study hospitals; therefore, self-reported data were uniformly used in all outcomes. Although some recall bias was inevitable, the short follow-up period (only 30 days) and the fact that parents usually attach importance to neonatal health and remember relatively well whether their child visited a doctor again minimized the recall bias. Second, due to the unavailability of the return visit records, we could not assess whether our intervention increased the readmission rates related to worsening neonatal hyperbilirubinemia, despite a significant drop in the overall rate. However, our intervention approaches are unlikely to delay admission to hospital for serious jaundice under the guidance of pediatricians. Finally, the trial was limited by the study site in comparison to that in other reports [
Our study showed that a smartphone-based jaundice screening app was effective for decreasing the risk of neonatal readmission due to jaundice within 30 days of discharge from hospital and the related maternal anxiety. Smartphones are becoming increasingly used around the world widely, as they are highly accessible and can act as low-cost, self-service testing tools for measuring neonatal bilirubin levels. More studies are needed to analyze the effectiveness of this intervention and to ensure large-scale implementation in other settings, subject to adequate resources.
CONSORT-eHEALTH checklist (V 1.6.1).
Items of the self-designed maternal anxiety due to neonatal jaundice scale.
Differences in the demographic characteristics between participants who completed and did not complete the trial.
Sensitivity analyses.
Evaluation of the jaundice mobile monitoring app.
odds ratio
randomized controlled trial
transcutaneous bilirubinometry
total serum bilirubin
This work was supported by the National Natural Science Foundation of China (grant 71663016), Hainan Province Clinical Medical Center, and the Excellent Talent Team of Hainan Province (QRCBT202121). The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication. We appreciate all the medical staff in the Hainan Women and Children’s Medical Center, Wanning People’s Hospital, and Chengmai People’s Hospital for their active cooperation in this study.
QY, YG, and QL contributed equally. LF and YG are joint corresponding authors. LF, XY, YG, QL, and QY conceptualized and designed the study. QL, HW, JF, KX, YH, CH, and LF collected the data, which were supervised by LF and YG. QY, QL, and YG analyzed and interpreted the data. QY, QL drafted the paper, QY, YG mainly completed the paper revision. All authors involved in revising the manuscript for important intellectual content and approved the final paper.
None declared.