Child Abuse, Child Development, Child Neglect, Parenting
Conditions
Brief summary
An intervention study to provide in-the-moment parenting tips with the goal of increasing healthy parent-child interactions leading to resiliency in high-risk children. Specifically, in a sample of parents participating in HV programs, the investigators will use a smartphone app (mHealth app) to deliver daily tailored messages with tips on monitoring and promoting child development. Daily assessments of parents' emotions, parenting behaviors, and interactions with their children will also be collected via the app. During the 4-week EMA study, parents will receive either 2 weeks of the mHealth intervention + EMA data collection followed by 2 weeks of only EMA data collection or to receive 2 weeks of only EMA data collection followed by 2 weeks of mHealth intervention + EMA data collection.
Detailed description
This is an intervention study to provide in-the-moment parenting tips with the goal of educating parents about monitoring and promoting child development, thereby increasing healthy parent-child interactions leading to resiliency in high-risk children. Specifically, in a sample of parents participating in a HV program, the investigators will use a smartphone app (mHealth app) to deliver twice-daily tailored messages with tips on monitoring and promoting child development. Ecological momentary assessments (EMA) of parents' emotions, parenting behaviors, and interactions with their children will also be collected via the mHealth app. Building off previous work the investigators expect that parents will evidence greater engagement in positive parenting practices on days when they receive the mobile-based parenting tips and strategies relative to on days when they do not receive this content. The investigators hypothesize that in-the-moment parenting tips delivered around peak times of parent-child interactions (i.e., before work/school, mealtimes, and bedtime) will promote positive parenting practices in a sample of parents at increased risk of adversity exposure. The investigators aim to use EMA methodology to measure both group level differences (i.e., intervention vs. services as usual) and within-person fluctuations in harsh parenting, children's emotional and behavioral functioning, and positive parenting practices. The study will evaluate the following aims and hypotheses: Aim 1. Conduct a 2-arm, randomized, controlled crossover trial to examine the feasibility and effectiveness of a mobile-based parenting app with parents participating in HV services. Hypothesis 1a. Parents will report high rates of satisfaction and usability. Hypothesis 1b. Parents will evidence greater engagement in positive parenting practices on days when they receive the mobile-based parenting tips and strategies relative to days when they do not receive this content. Hypothesis 1c. Groups will evidence increases in positive parenting subscale scores after a month of app engagement. Aim 2. Utilize EMA data and follow-up assessments to examine the impact of positive parenting practices on the promotion of child development. Hypothesis 2a. Children will demonstrate less challenging and more positive behavioral functioning on days when their parents receive positive parenting tips. Hypothesis 2b. Groups will evidence increases in developmental scores after a month of app engagement.
Interventions
BEHAVIORALmHealth Intervention
A smartphone app (mHealth app) will deliver daily tailored messages with tips on monitoring and promoting child development and positive parenting tips.
Sponsors
University of Oklahoma
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
OTHER
Masking
NONE
Intervention model description
The statistical model is outcome dependent. Some outcomes were measured at pre-intervention, mid-intervention, and post-intervention assessment points allowing for classic crossover study design analyses of intervention arm differences. Other outcomes were only measured at pre-intervention and post-intervention assessment points and allow for classic within-group pre-post change analyses. Feasibility and usability outcomes were effectively measured at post-intervention only and were analyzed with one-sample Wilcoxon signed rank tests to detect moderate or greater levels of feasibility/utility.
Eligibility
Sex/Gender
ALL
Age
0 Years to 99 Years
Healthy volunteers
No
Inclusion criteria
* Any new or currently enrolled parent in Oklahoma or Tulsa County in MIECHV-funded home visiting programs.
Exclusion criteria
* Parents residing outside of Oklahoma or Tulsa Counties will not be recruited (although if they move to another county between baseline and follow-up, they may be eligible to complete the follow-up survey at the discretion of the Principal Investigator). * Parents who cannot read/speak English or Spanish at an 8th-grade level or higher will not be recruited. * Parents under age 16 will not be recruited. Parents ages 16-17 meeting other eligibility requirements may be recruited using the assent form with signatures from both of his/her parents.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Recommend the App to a Friend | Post-Intervention Assessment | Participants responded on a scale from Extremely unlikely (scored -3) to Extremely likely (scored 3). |
| Positive Parenting Practices: EMA Data Time Spent With Child Daily (in Hours) | Daily EMA Data (28 days) | Mean number of hours spent with the child daily during the periods with and without app-delivered parenting tips. |
| Helped me Interact Positively With Child | Post-intervention Assessment | Participants responded on a scale from Not at all helpful (scored -1) to Extremely helpful (scored 3). |
| Interest in Using the App in the Future | Post-Intervention Assessment | Participants responded on a scale from Not at all interested (scored -1) to Extremely interested (scored 3). |
| Positive Parenting Practices: EMA Data Positive Parenting Activity Engagement | Daily EMA Data (28 days) | Participants select the positive parenting activities they engaged in with their child each day during the EMA study (0 is the lowest and represents no engagement in positive parenting practices and a higher number reflects more activities reported). Analyses compared average activity counts during periods with and without app-delivered parenting tips. |
| Parenting Behavior: Alabama Parenting Questionnaire- Positive Parenting Subscale | Change from baseline to post-intervention | The Alabama Parenting Questionnaire contains 3 subscales previously used with parents of young children that represent different dimensions of parenting. Each item is rated on a 1 (never) to 5 (always) Likert scale. A higher score on each subscale represents more engagement in that parenting behavior. This analysis focused on the Positive Parenting Subscale (12 items; score range: 12-60). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Child Behavior: EMA Challenging Child Behaviors | Daily EMA Data (28 days) | Parents report the number of challenging (difficult to manage) child behaviors daily. Analyses compared average behavior counts during periods with and without app-delivered parenting tips. |
| Child Behavior: EMA Positive Child Behaviors | Daily EMA Data (28 days) | Parents report the number of positive (good) child behaviors daily. Analyses compared average behavior counts during periods with and without app-delivered parenting tips. |
| General Child Development Score | Change from baseline to post-intervention | Developmental Profiles-4 parent/caregiver interview contains five subscales: social-emotional, physical, communication, cognitive, and adaptive behavior. Each subscale score is age standardized with an average of 100 and a standard deviation of 15 (minimum score of 40, maximum score of 160). The General Development Score used in this analysis represents a standardized composite of the 5 subscale scores (minimum score of 46, maximum score of 150). A higher composite score indicates more advanced child development. |
Countries
United States
Participant flow
Recruitment details
Recruitment of clients involved the assistance of up to 3 home visiting agencies in two urban counties in the SW U.S. All agencies were participating in a quality improvement network focused on developmental monitoring and promotion. Agencies that elected to design QI projects that promoted the study parenting app were asked to deliver an initial study pitch to willing and eligible clients. Clients that opted-in were later contacted by the study team for more comprehensive study recruitment.
Participants by arm
| Arm | Count |
|---|---|
| No Parenting Tip, Then Parenting Tips In this crossover arm, participants are assigned to an EMA Only condition for the first two weeks (14 days) of study involvement. During this time, parents receiving daily alerts to complete an ecological momentary assessment (EMA) that asks participants to provide information on recent parent-child interactions, current stressors or challenges, and their child's emotional and behavioral functioning. For the last two weeks of study involvement (days 15 through 28), this group continues to complete the once-daily EMA survey, but they also receive app-delivered parenting tips twice daily. While in this condition phase, a tip notification is delivered to the participant's mobile phone once in the morning and once in the evening. Participants must click on the notification to reveal the parenting tip within the app interface.
mHealth Intervention: A smartphone app (mHealth app) will deliver daily tailored messages with tips on monitoring and promoting child development and positive parenting tips. | 11 |
| Parenting Tips, Then no Parenting Tips In this crossover arm, participants are assigned to the EMA + mHealth intervention (i.e., app-delivered parenting tips twice daily) condition for the first two weeks (14 days) of study involvement. During this time, parents receiving daily alerts to complete an ecological momentary assessment (EMA) that asks participants to provide information on recent parent-child interactions, current stressors or challenges, and their child's emotional and behavioral functioning. During this time, a tip notification is also delivered to the participant's mobile phone once in the morning and once in the evening. Participants must click on the notification to reveal the parenting tip within the app interface. For the last two weeks of study involvement (days 15 through 28), this group continues to complete the once-daily EMA survey, but the parenting tips are no longer delivered.
mHealth Intervention: A smartphone app (mHealth app) will deliver daily tailored messages with tips on monitoring and promoting child development and positive parenting tips. | 13 |
| Total | 24 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Post-Intervention Assessment | Lost to Follow-up | 0 | 1 |
| Week 1 App Use (1+ Survey and/or Tips) | No app interactions logged | 1 | 0 |
| Week 3 App Use (1+ Survey and/or Tips) | No app interactions logged | 0 | 3 |
| Week 4 App Use (1+ Survey and/or Tips) | No app interactions logged | 0 | 5 |
Baseline characteristics
| Characteristic | No Parenting Tip, Then Parenting Tips | Parenting Tips, Then no Parenting Tips | Total |
|---|---|---|---|
| Age, Continuous | 32.5 years STANDARD_DEVIATION 6.9 | 31.0 years STANDARD_DEVIATION 8.2 | 31.7 years STANDARD_DEVIATION 7.5 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 9 Participants | 8 Participants | 17 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 2 Participants | 5 Participants | 7 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Black or African American | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) More than one race | 1 Participants | 1 Participants | 2 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 10 Participants | 11 Participants | 21 Participants |
| Region of Enrollment United States | 11 participants | 13 participants | 24 participants |
| Sex/Gender, Customized Female | 11 Participants | 13 Participants | 24 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 11 | 0 / 13 |
| other Total, other adverse events | 0 / 11 | 0 / 13 |
| serious Total, serious adverse events | 0 / 11 | 0 / 13 |
Outcome results
Primary
Helped me Interact Positively With Child
Participants responded on a scale from Not at all helpful (scored -1) to Extremely helpful (scored 3).
Time frame: Post-intervention Assessment
Population: As explained under Study Design, feasibility and usability outcomes were only measured at post-intervention and were analyzed with one-sample Wilcoxon signed rank tests to detect moderate or greater levels of feasibility/utility. This restriction applies to this outcome; thus, no test for differences between the experimental arms was conducted.~Also, 1 participant skipped this item at the post assessment. 1 participant did not complete the post-assessment (lost to follow-up).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Helped me Interact Positively With Child | 1.64 Units on a scale | Standard Deviation 1.18 |
Comparison: Null hypothesis was that the app was only slightly helpful (scored 0). A Wilcoxon signed rank test was first performed to evaluate potential differences across randomization groups (early vs late app-delivered parenting tips). When the test of a randomization group difference was not significant, the analysis continued with a one-sample Wilcoxon signed rank test.p-value: <0.01Wilcoxon (Mann-Whitney)
Primary
Interest in Using the App in the Future
Participants responded on a scale from Not at all interested (scored -1) to Extremely interested (scored 3).
Time frame: Post-Intervention Assessment
Population: As explained under Study Design, feasibility and usability outcomes were only measured at post-intervention and were analyzed with one-sample Wilcoxon signed rank tests to detect moderate or greater levels of feasibility/utility. This restriction applies to this outcome; thus, no test for differences between the experimental arms was conducted.~Also, 1 participant skipped this item. 1 participant did not complete the post-assessment (lost to follow-up).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Interest in Using the App in the Future | 1.41 Units on a scale | Standard Deviation 1.26 |
Comparison: Null hypothesis was that there would only be slight interest (scored 0) in using the app in the future. A Wilcoxon signed rank test was, first, performed to evaluate potential differences across randomization groups (early vs late app-delivered parenting tips). When the test of a randomization group difference was not significant, the analysis continued with a one-sample Wilcoxon signed rank test.p-value: <0.01Wilcoxon (Mann-Whitney)
Primary
Parenting Behavior: Alabama Parenting Questionnaire- Positive Parenting Subscale
The Alabama Parenting Questionnaire contains 3 subscales previously used with parents of young children that represent different dimensions of parenting. Each item is rated on a 1 (never) to 5 (always) Likert scale. A higher score on each subscale represents more engagement in that parenting behavior. This analysis focused on the Positive Parenting Subscale (12 items; score range: 12-60).
Time frame: Change from baseline to post-intervention
Population: As explained under Study Design, some outcomes were only measured at pre- and post-intervention assessment points and, therefore, only allow for classic within-group pre-post change analyses. This restriction applies to this outcome; thus, no test for differences between the experimental arms was conducted.~Also, 1 participant skipped too many items at baseline, and a subscale calculation was not warranted. 1 participant did not complete the post-intervention assessment (lost to follow-up).
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Parenting Behavior: Alabama Parenting Questionnaire- Positive Parenting Subscale | 47.6 Units on a scale | Standard Error 1.28 |
| Receiving Parenting Tips | Parenting Behavior: Alabama Parenting Questionnaire- Positive Parenting Subscale | 47.6 Units on a scale | Standard Error 1.28 |
Comparison: Null hypothesis is that the mean positive parenting subscale score would not change from the baseline to the post-intervention assessment. Mixed effects models were performed regressing mean subscale scores on a timeframe (baseline vs post) indicator and a randomization group (early vs late app-delivered parenting tips) indicator. A randomization group difference over time was evaluated. This interaction effect was not included in the final statistical model unless statistically significant.p-value: <1Mixed Models Analysis
Primary
Positive Parenting Practices: EMA Data Positive Parenting Activity Engagement
Participants select the positive parenting activities they engaged in with their child each day during the EMA study (0 is the lowest and represents no engagement in positive parenting practices and a higher number reflects more activities reported). Analyses compared average activity counts during periods with and without app-delivered parenting tips.
Time frame: Daily EMA Data (28 days)
Population: Three participants did not provide data on this measure during their No Parenting Tips period.
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Positive Parenting Practices: EMA Data Positive Parenting Activity Engagement | 3.75 Daily number of activities | Standard Error 0.25 |
| Receiving Parenting Tips | Positive Parenting Practices: EMA Data Positive Parenting Activity Engagement | 2.38 Daily number of activities | Standard Error 0.24 |
Comparison: Null hypothesis is the number of positive activities engaged in with the child will not differ during the periods with and without app-delivered parenting tips. Mixed effects models were performed regressing mean activity counts on a 'Tips Active' indicator and a measurement-period (1st two weeks vs last two weeks) indicator.p-value: <0.01Mixed Models Analysis
Primary
Positive Parenting Practices: EMA Data Time Spent With Child Daily (in Hours)
Mean number of hours spent with the child daily during the periods with and without app-delivered parenting tips.
Time frame: Daily EMA Data (28 days)
Population: Three participants did not complete EMA assessments during their No Parenting Tips period.
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Positive Parenting Practices: EMA Data Time Spent With Child Daily (in Hours) | 2.82 Daily number of hours | Standard Error 0.07 |
| Receiving Parenting Tips | Positive Parenting Practices: EMA Data Time Spent With Child Daily (in Hours) | 2.72 Daily number of hours | Standard Error 0.07 |
Comparison: Null hypothesis is that the number of hours spent with the child will not differ during the periods with and without app-delivered parenting tips. Mixed effects models were performed regressing mean number of hours on a 'Tips Active' indicator and a measurement-period (1st two weeks vs last two weeks) indicator.p-value: <0.14Mixed Models Analysis
Primary
Recommend the App to a Friend
Participants responded on a scale from Extremely unlikely (scored -3) to Extremely likely (scored 3).
Time frame: Post-Intervention Assessment
Population: As explained under Study Design, feasibility and usability outcomes were only measured at post-intervention and were analyzed with one-sample Wilcoxon signed rank tests to detect moderate or greater levels of feasibility/utility. This restriction applies to this outcome; thus, no test for differences between the experimental arms was conducted.~1 participant did not complete the post-assessment (lost to follow-up).
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Recommend the App to a Friend | 1.48 Units on a scale | Standard Deviation 1.81 |
Comparison: Null hypothesis was that the participant was neither likely nor unlikely (scored 0) to recommend the app to friends. A Wilcoxon signed rank test was, first, performed to evaluate potential differences across randomization groups (early vs late app-delivered parenting tips). When the test of a randomization group difference was not significant, the analysis continued with a one-sample Wilcoxon signed rank test.p-value: <0.01Wilcoxon (Mann-Whitney)
Secondary
Child Behavior: EMA Challenging Child Behaviors
Parents report the number of challenging (difficult to manage) child behaviors daily. Analyses compared average behavior counts during periods with and without app-delivered parenting tips.
Time frame: Daily EMA Data (28 days)
Population: Three participants did not provide EMA data during the No Parenting Tips period.
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Child Behavior: EMA Challenging Child Behaviors | 0.50 Daily number of challenging behaviors | Standard Error 0.06 |
| Receiving Parenting Tips | Child Behavior: EMA Challenging Child Behaviors | 0.33 Daily number of challenging behaviors | Standard Error 0.06 |
Comparison: Null hypothesis is that the number of challenging child behaviors will not differ during the periods with and without app-delivered parenting tips. Mixed effects models were performed regressing mean behavior counts on a 'Tips Active' indicator and a measurement-period (1st two weeks vs last two weeks) indicator.p-value: <0.01Mixed Models Analysis
Secondary
Child Behavior: EMA Positive Child Behaviors
Parents report the number of positive (good) child behaviors daily. Analyses compared average behavior counts during periods with and without app-delivered parenting tips.
Time frame: Daily EMA Data (28 days)
Population: Three participants did not provide EMA data during the No Parenting Tips period.
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | Child Behavior: EMA Positive Child Behaviors | 2.54 Daily number of positive behaviors | Standard Error 0.13 |
| Receiving Parenting Tips | Child Behavior: EMA Positive Child Behaviors | 2.21 Daily number of positive behaviors | Standard Error 0.13 |
Comparison: Null hypothesis is that the daily number of positive child behaviors will not differ during the periods with and without app-delivered parenting tips. Mixed effects models were performed regressing mean behavior counts on a 'Tips Active' indicator and a measurement-period (1st two weeks vs last two weeks) indicator.p-value: <0.01Mixed Models Analysis
Secondary
General Child Development Score
Developmental Profiles-4 parent/caregiver interview contains five subscales: social-emotional, physical, communication, cognitive, and adaptive behavior. Each subscale score is age standardized with an average of 100 and a standard deviation of 15 (minimum score of 40, maximum score of 160). The General Development Score used in this analysis represents a standardized composite of the 5 subscale scores (minimum score of 46, maximum score of 150). A higher composite score indicates more advanced child development.
Time frame: Change from baseline to post-intervention
Population: As explained under Study Design, some outcomes were only measured at pre- and post-intervention assessment points and, therefore, only allow for classic within-group pre-post change analyses. This restriction applies to this outcome; thus, no test for differences between the experimental arms was conducted. 1 participant did not complete the post-intervention assessment (lost to follow-up).
| Arm | Measure | Value (LEAST_SQUARES_MEAN) | Dispersion |
|---|---|---|---|
| No Parenting Tips | General Child Development Score | 99.6 Units on a scale | Standard Error 3.92 |
| Receiving Parenting Tips | General Child Development Score | 101.0 Units on a scale | Standard Error 3.94 |
Comparison: Null hypothesis is that the mean general development scale score would not change from the baseline to the post-intervention assessment. Mixed effects models were performed regressing mean scale scores on a timeframe (baseline vs post) indicator and a randomization group (early vs late app-delivered parenting tips) indicator.p-value: <0.54Mixed Models Analysis