Family Check-Up Online effects on parenting and parent wellbeing in families of toddler to preschool-age children

Abstract

Online parenting interventions hold promise for increasing access to behavioral support for families with low income and who reside in rural areas. The current study evaluates the efficacy of a mobile app-based parenting support program, the Family Check-Up Online (FCU-O) with telehealth coaching support, for parents of children 1.5–5 years old at risk of experiencing parenting challenges. We tested effects of the FCU-O on parent wellbeing (parenting self-efficacy and symptoms of depression and anxiety) and parent-reported parenting skills (support for positive behavior, limit-setting, and proactive parenting) at the 3-month follow-up in a randomized controlled trial (FCU-O versus control). We also tested relationships between parents’ baseline characteristics and their engagement with the app and telehealth coaching. Eligibility criteria included endorsing depressive symptoms or current or past substance misuse. Participants (N = 356) were predominantly White (72%) and low-income, with 43% of families residing in a rural area. Approximately one third of participants reported clinically significant symptoms of anxiety or depression and one third endorsed a lifetime history of opioid misuse. Intent to treat analyses tested effects of the FCU-O on parent wellbeing and parenting skills. The FCU-O was associated with improvements to parents’ support for positive behavior, limit-setting, and proactive parenting skills, as well as parenting self-efficacy and depressive symptoms. Parents’ initial symptoms of depression and anxiety significantly predicted greater telehealth coach engagement; low levels of initial self-reported skills positive behavior support and limit-setting significantly predicted greater app engagement. Results provide support for the FCU-O as a preventive parenting intervention for parents experiencing mental health challenges, with implications for reducing barriers to accessing parenting support for underserved families.

Responsive parenting and parent-child relationship quality in early childhood are consistently linked with children’s cognitive, social, and emotional growth and development (Knauer et al., 2019; Landry et al., 2003; Landry et al, 2001). Parents with mental health challenges, including depression, anxiety, and substance use, are at increased risk of experiencing parenting challenges (Kim et al., 2010; Lovejoy et al., 2000) and increased parenting stress (Aviles et al., 2024), which can impact parent-child relationship quality and subsequently child developmental outcomes (Goodman et al., 2011; Urizar & Muñoz, 2022). An estimated one in five parents of children under 18 in the U.S. experienced past-year mental illness (Stambaugh et al., 2017), and one in eight children in the U.S. lives with a parent with a substance use disorder (Lipari & Van Horn, 2017). Moreover, rates of home removals for children due to parental substance use has risen dramatically since 2012 (Meinhofer & Angleró-Díaz, 2019). Parental substance use impacts families across the spectrum of income and racial and ethnic identity, with research indicating that Native American and non-Hispanic White children experienced both the highest levels and fastest growth of home removals involving parental substance use (Meinhofer et al., 2020). Therefore, providing accessible and effective parenting and mental health resources for parents with young children at risk for substance use is crucial.

There is growing evidence that stressors associated with the COVID-19 pandemic (Brown et al., 2020) significantly affected parents of young children (e.g., toddler, pre-school age; Gadermann et al., 2021; Racine et al., 2022) with pre-existing vulnerabilities, consistent with Prime and colleagues’ (2020) theoretical framework on family risk and resilience. Increasingly, pandemic-related social disruptions are shown to have potentially long-lasting deleterious effects on parent wellbeing, parenting behaviors, and child development (Robertson et al., 2021; Thompson et al., 2022). As such, the pandemic has significantly exacerbated an existing mental health crisis caused in part by a severe shortage of mental health providers, particularly in rural areas of the U.S. (Health Resources and Services Administration, 2024; Kuehn, 2022). This rural-urban inequality in mental healthcare access has dire consequences for parents of infants and toddlers as the risk of perinatal depression is higher for women in rural compared with urban areas (Nidey et al., 2020). In addition, rural communities have been particularly adversely affected by the opioid epidemic (Palombi et al., 2018). For parents who are using opioids and other substances, stigma presents a major barrier to receiving behavioral health services for themselves and for their children, with White parents more likely to report stigma as a barrier to treatment than non-White parents (Stringer & Baker, 2018). In sum, there is a critically urgent need to address the limited availability of efficacious behavioral health treatment and parenting support for parents of toddler and pre-school aged children, particularly those who have historically had difficulty accessing treatment (e.g., who live in rural areas, have low income, who experience stigma due to substance use; Anderson et al., 2013; Castro-Ramirez et al., 2021; Saldana et al., 2021; Stringer & Baker, 2018).

Research suggests that early intervention and preventative resources for families can help to mitigate some of the negative child outcomes (e.g., being removed from the home, higher levels of challenging behaviors; Aviles et al., 2024; Roscoe et al., 2021) associated with parent mental health challenges, including substance misuse (Roscoe et al., 2021). Evidence-based parenting support programs can help to decrease parental mental health problems and substance misuse, and improve adaptive parenting skills, family functioning, and child wellbeing (Connell & Stormshak, 2023; Jones et al., 2014; Lindsay et al., 2011; Phelan et al., 2013; Saldana et al., 2021). Theoretically, parenting support programs work by improving parenting skills and increasing parenting self-efficacy, leading to improvements in child behavior over time (Sanders, 2023). However, there are multiple barriers to accessing these programs, including stigma, cost, and transportation, particularly for families in rural communities (Boydell et al., 2006; Weisenmuller & Hilton, 2021; Wright et al., 2023). For example, Saldana and colleagues found that the Families Actively Improving Relationships (FAIR) program, a comprehensive home visiting and community-based treatment model designed to support families involved in the child welfare system for parental substance abuse, was associated with significant reductions in parent substance use, mental health challenges, parenting stress, and risk for child neglect in a sample of parents who were predominantly (68%) non-Hispanic White (Saldana et al., 2021). However, authors noted that the cost of travel for home visits for families living more than 20 miles from providers would likely be an unsustainable model under a Medicaid structure.

Online parenting intervention and prevention programs hold great promise for addressing the growing need for parenting support to reduce parenting stress and improve parenting skills and parent-child relationship quality, as they address barriers related to provider shortages in rural and other underserved parts of the country (Andrilla et al., 2018) and reduce reliance on licensed behavioral health clinicians (Graham et al., 2021). Increasing access to parenting support during early childhood (i.e., the toddler to preschool years) is especially critical, as responsive caregiving during this time has been found to uniquely predict children’s long-term social and cognitive development (Landry et al., 2003).

Online parenting support programs

Online parenting support programs may include services delivered via telehealth, as well as parents’ self-directed or coach-supported engagement in parent training program content via computer or mobile device. Widespread implementation of online parenting interventions increases the reach of high-quality parenting support at a lower cost to both consumers and providers (Le et al., 2021). Existing online parenting programs vary across several domains, including how parents engage with content (e.g., whether a clinician is available to support parent’s engagement), how programs have been optimized for use (e.g., computer, mobile device, video conferencing), and the extent to which they are evidence-based.

Several existing evidence-based parenting interventions for caregivers of young children have been adapted to online formats (e.g., Breitenstein et al., 2021; Day & Sanders, 2018; McAloon & Armstrong, 2024; Spencer et al., 2020), with some evidence for associated improvements in parenting skills and child behavior. In a study testing an online version of the Triple P - Positive Parenting Program in predominantly White parents of 2–8-year-old children from Australia, Day and Sanders (2018) found promising effects on key intervention targets, particularly for parents who were randomly assigned to also receive telephone support sessions from a practitioner. Specifically, comparisons between the control group and practitioner-supported online Triple P yielded large effects (i.e., Cohen’s d ≥ .80) on child behavior problems and small to moderate effects (i.e., Cohen’s d = .20-.79) on negative parenting practices and parenting self-efficacy immediately post-intervention. Randomization to the practitioner-supported group was also found to increase parents’ engagement in online parenting content. In a randomized controlled trial of ezPARENT, a digital delivery adaptation of the Chicago Parent Program, parents of children ages 2–8 engaged in a fully self-directed version of the program (Breitenstein et al., 2021). The majority of parents were African American or Latine and recruited from urban primary care clinics. No main effects on parenting or child behavior were found; however, there was some evidence of positive outcomes for parents experiencing greater levels of initial stress and child behavior problems. In a recent meta-analysis of online behavioral parenting interventions that included 14 randomized controlled trials, McAloon and Armstrong (2024) found small to moderate effect sizes on child behavior, parenting self-efficacy, and parent stress. The inclusion of therapeutic support in online parenting programs was found to significantly increase intervention effects on parent stress. Taken together, these findings suggest that the effectiveness of online parenting interventions are bolstered when parents are also connected with a trained clinician, typically via telehealth, to motivate engagement and provide tailored support. Of note, no online parenting interventions were found that were developed for or tested on parents living in rural areas or with a history of substance use, groups that may be especially likely to benefit from the online format due to stigma of seeking traditional, in-person parenting support for parents using substances (Stringer & Baker, 2018) and limited access to such programming in rural areas (HRSA, 2024).

Overall, online parenting interventions have been found to be effective in reducing child behavior problems and improving parenting skills, with comparable effect sizes (i.e., in the small to moderate range) to traditional, in-person parenting interventions (Florean et al., 2020; Flujas-Contreras et al., 2019; Spencer et al., 2020). However, few if any evidence-based online interventions for parents of young children were designed specifically for delivery via mobile phone app (Florean et al., 2020). Mobile app-based parenting intervention delivery has the potential to further reduce barriers to parents’ accessing quality behavioral healthcare for their families, given that many families, particular those with lower income, do not have a computer or broadband internet access at home (e.g., 43% of families making less than $30,000 per year, 26% of families making $30,000 – 49,000; Pew Research Center, 2021). Although there are a number of commercially available mobile apps offering parenting skills support to parents of young children, few if any have been subjected to empirical testing (David et al., 2024; Virani et al., 2019). Moreover, a recent review of commercially available parenting skills apps targeted toward parents of school-age children found that few adhered to evidence-based behavioral parent training principles and strategies (Li et al., 2023). In another recent review of 53 commercially available parenting intervention apps for parents of children aged 2–12, the authors concluded that overall, apps lacked both personalization (e.g., use of self-report measures to assess symptoms and current skills) and crucial psychoeducational components (e.g., providing context and rationale for therapeutic techniques; David et al., 2024).

The Family Check-Up Online

The original Family Check-Up (FCU) is a brief, cost-effective, and strengths-based parenting intervention that is individually tailored to the needs of families (Dishion & Stormshak, 2007). Parents take part in an ecological assessment, which consists of standardized questions about parenting and family management skills, as well as important contextual factors such as stress, social support, and health behaviors, including substance use. Following the assessment, a family coach engages the parent in a feedback session in which they collaboratively review the results of the assessment, focusing on family strengths and the parent’s goals for change. The family coach uses motivational interviewing strategies to engage the parent in all sessions. Following the feedback, the parent can elect to engage in subsequent sessions with the family coach, which typically focus on parenting skills (e.g., limit-setting, positive reinforcement) and include a strengths-based approach to coaching the parent (i.e., focusing on the parent’s strengths to motivate and engage the parent in completing their goals). Findings from several randomized controlled trials indicate that the FCU is associated with a multitude of positive family outcomes, including gains in parenting skills (Sitnick et al., 2015), reduced child behavioral and emotional problems (Dishion et al., 2008; Fosco et al., 2016; Shelleby et al., 2018), and improvements to parents’ mental health and wellbeing (Resnik et al., 2023; Shaw et al., 2009). The original FCU model has been implemented with parents of children across the developmental spectrum, including infants (Shaw et al., 2021), toddlers (Dishion et al., 2008), early school-age (Garbacz et al., 2022), and adolescents (Galán et al., 2023; Stormshak et al., 2011). Prior randomized controlled trials have tested the FCU’s effectiveness among families experiencing poverty and other contextual stressors (e.g., Dishion et al., 2008; Galán et al., 2023; Stormshak, Caruthers et al., 2019), with a primary intervention target of decreasing the risk of adolescent substance use. The FCU was designed to be flexible and adaptive to diverse racial, ethnic, and cultural groups (Dishion & Stormshak, 2007), with research suggesting that it is equally effective across White, African American, and Hispanic/Latine families (Berkel, Fu, et al., 2021; Smith et al., 2014).

The FCU was later adapted as a telehealth model, the FCU Online (FCU-O). The FCU-O was created as part of a randomized controlled trial in which mostly White, suburban parents of middle-school children received an online version of the FCU with or without telehealth coaching (Stormshak, Seeley et al., 2019). Findings from this trial indicated that parents randomized to the FCU-O with coaching engaged more with the online program than parents without access to coaching, and that the FCU-O with coaching was associated with greater improvements on several key outcomes, including child emotional problems, child effortful control, and parenting self-efficacy. Next, a mobile app version of the FCU-O was created for parents of middle school students and tested in a randomized controlled trial. The FCU-O school-age version includes five “modules” or content areas (e.g., positive parenting, rules and consequences). Parents randomized to the FCU-O also received telehealth meetings with a parent coach. Assignment to the FCU-O was associated with improvements to parents’ anxiety, depression, and stress, as well as parenting skills and family functioning at 2-month follow-up as compared with a control group (Connell & Stormshak, 2023). In the initial study of the FCU-O school-age version, most parent participants (85%) identified as European American and generally endorsed low levels of socioeconomic risk (Connell & Stormshak, 2023). More work was therefore needed to understand intervention effects of the FCU-O with a higher-risk and lower-resourced sample.

The Current Study

This study is the first to test effects of the early childhood version of the FCU-O, with a specific focus on including parents experiencing high levels of stress, substance use problems, and/or difficulty accessing behavioral health supports. Our primary goal was to test initial intervention effects on parenting and parent mental health outcomes at the 3-month follow-up. To our knowledge, the FCU-O is the first online parenting program that is optimized for delivery via mobile device and incorporates telehealth coaching. The early childhood FCU-O was designed as a preventive intervention for parents with mental health and/or substance use challenges and difficulty accessing mental health and parenting supports due to rural residence. Notably, residents of the rural U.S. predominantly identify as non-Hispanic White (76%; Johnson & Lichter, 2022), which is consistent with the population of the current study. The FCU-O content was adapted based on feedback from family and community service groups in rural areas, as well as interviews with parents with a history of opioid use (Stormshak et al., 2021).

Given that both depression (Beeber et al., 2014) and substance use problems (Kim et al., 2010) are risk factors for parenting challenges, and that the stigmatized nature of substance use in parents may present a barrier to seeking treatment for both substance use and mental health or parenting concerns (Stringer & Baker, 2018), we included parents who were currently experiencing depressive symptoms and/or current or past substance misuse. We also made substantial efforts to recruit participants residing in rural parts of the state.

As the FCU-O’s content was designed to provide psychoeducation on parenting skills during early childhood in the specific areas of positive parenting, proactive parenting, and limit-setting, we hypothesized that intervention group assignment would be associated with improvements to parenting skills in these three areas. We also hypothesized that FCU-O group assignment would be associated with improvements to parenting self-efficacy at the 3-month follow-up, as research suggests that use of positive parenting skills is associated with increased parenting self-efficacy over time (Dumka et al., 2010; Glatz & Buchanan, 2015). Finally, we hypothesized that parents randomly assigned to the FCU-O would experience improvements in their symptoms of depression and anxiety from baseline to 3-month follow-up. The FCU-O offered explicit content on parent wellness and self-care, and prior research suggests intervention effects of the original FCU model on parental depressive symptoms (Shaw et al., 2009).

Since this online parenting intervention may be particularly beneficial for families with low income and residing in rural areas, an exploratory aim was to test whether participant demographic characteristics (including family income and rurality status) and family composition would predict engagement in both the app and telehealth coaching.

Material and Methods

Design

Parent participants were randomly assigned to the FCU-O – Parenting Young Children version or a control group. Data were collected from both groups at baseline and three months following the baseline assessment. This study was pre-registered at ClinicalTrials.gov (NCT05180487).

Participants

Participants were the primary caregivers (N = 356) of children aged 1.5 to 5 years. A variety of recruitment strategies were used and included partnerships with community agencies and in-person events, as well as advertising on social media. To meet eligibility criteria, parents had to endorse any one of the following to be eligible for the study: binge drinking or any recreational drug use (including cannabis) in the past year, lifetime history of opioid misuse, or endorsement of depressive symptoms on the Patient Health Questionnaire-2 (PHQ-2; Löwe et al., 2005). Parents also had to be living with their child at least 50% of the time, have a smart phone with text messaging capability and access to email, and reside in the state in which the research was being conducted. Participants were recruited in rural and metropolitan areas of the Pacific Northwest region of the U.S.

Most parent participants identified as female (93%) and White, Non-Hispanic (72%). The median annual family income bracket was $35,000-$49,999, substantially lower than the median household income in the U.S., which was approximately $75,000 in 2022, the year that recruitment began (U.S. Census Bureau, 2023). Over two thirds of the sample (70%) reported that their family received some government assistance in the past year, including Temporary Assistance for Needy Families (TANF), Supplemental Nutrition Assistance Program (SNAP) benefits, and/or Women, Infants, and Children (WIC) program benefits. The average age of the target children in the study at baseline was 43.40 months (SD = 14.2), or approximately 3.5 years old. Nearly half of participants (43%) resided in a zip code designated as rural according to the Oregon Office of Rural Health geographic definitions. Nearly ten percent (9.3%) of the sample elected to complete research procedures in Spanish. Additional demographic data are presented in Table 1, separated by group assignment. Of the 356 parents who enrolled in the study, 324 (91.0%) participated in the 3-month follow up assessment.

Table 1.

Participant demographic characteristics

Family Demographic Characteristics	Control Group (N = 176)	FCU-O Group (N = 180)	Whole Sample (N = 356)
	M (SD) or % endorsed	M (SD) or % endorsed	M (SD) or % endorsed
Child age (months)	42.80 (14.8)	43.79 (13.7)	43.30 (14.2)
Parent sex - female	92.5%	93.9%	92.7%
Parent race/ethnicity
White, Non-Hispanic/Latine	70.5%	72.8%	71.6%
Hispanic/Latine	21.6%	20.6%	21.1%
American Indian/Alaska Native	6.8%	4.4%	5.6%
Asian	2.8%	6.1%	4.5%
Black/African American	5.7%	2.8%	4.2%
Native Hawaiian or Other Pacific Islander	0.6%	3.3%	2.0%
Parent age (years)	33.59 (6.3)	33.77 (5.6)	33.70 (6.0)
Parent’s education level
Less than high school degree	6.8%	8.9%	7.8%
High school degree or GED	19.3%	18.9%	19.1%
Technical school, partial college, or associate’s degree	41.5%	38.3%	39.9%
Bachelor’s degree	14.8%	22.2%	18.5%
Graduate training or degree	17.6%	11.7%	14.6%
Annual family income
< $15,000	20.4%	21.1%	20.8%
$15,000 - $34,999	19.9%	22.8%	21.3%
$35,000 - $49,999	12.5%	12.2%	12.4%
$50,000 - $74,999	18.2%	16.7%	17.4%
$75,000 - $99,999	10.2%	7.8%	9.0%
>$100,000	13.0%	16.1%	14.6%
Parent employment status
Employed, full-time or part-time	58.3%	55.1%	56.6%
Unemployed, looking for job	14.9%	12.9%	13.6%
Unemployed, not looking for job	26.9%	32.0%	29.7%
Family resides in rural area	45.4%	40.6%	43%
Spanish language	9.1%	9.4%	9.3%

Procedures

Interested parents contacted the study team via phone call, email, or a form on the project’s website. A staff member subsequently contacted the parent to explain the study, assess eligibility, and review informed consent if eligibility criteria were met. Following informed consent, parents completed baseline assessment questionnaires via phone interview or online survey, depending on parent preference. Next, they were randomized to the intervention or control condition. Parents in both groups were able to continue accessing other substance use, mental health, and/or parenting support services during the assessment period. Parents in both groups completed a 3-month follow-up assessment. They were compensated $100 for each completed assessment. All study procedures were approved by the IRB.

FCU Online– Parenting Young Children version.

The FCU-O – Parenting Young Children version used in the current study was adapted from the initial online iteration (Stormshak, Seeley et al., 2019), which is grounded in the original FCU model (Dishion & Stormshak, 2007). Stormshak and colleagues (2021) provide a detailed summary of the original, in-person FCU model as compared with the FCU-O. In short, while the original FCU model focuses primarily on ecological assessment and feedback (see above), with follow-up sessions devoted to parenting skills, the FCU-O incorporates assessment, feedback, and parenting skills content into each module within the app, as well as each telehealth coaching session.

Like the original FCU and subsequent online version (Connell & Stormshak, 2023), the Parenting Young Children version of the FCU-O incorporates self-assessment and feedback on parenting skills and family management. The Parenting Young Children version was specifically developed for parents of toddler and preschool-age children who are at high risk of experiencing mental health and/or parenting challenges, including current or past substance use. Content was adapted from the Everyday Parenting curriculum (Dishion et al., 2012), with additional focus on parent wellbeing and substance use psychoeducation. The FCU-O was adapted for early childhood with feedback from focus groups of parents and community providers (Stormshak et al., 2021). Content is divided into five modules: wellness and self-care, parenting and substance use, positive parenting, proactive parenting, and rules and consequences. Consistent with the original FCU model’s emphasis on assessment and feedback, at the beginning of each module, parents complete a brief assessment (10–12 questions) followed by automated feedback within the app that identifies specific areas of strength and growth. Parents had access to the app for one year after their enrollment in the program.

Parents in the intervention group were assigned to work with a telehealth family coach after completing the baseline research assessment. The role of the family coach was to support parent’s practice of skills learned through the app and work with the parent on applying the skills to their personal parenting goals and family situation. The family coach contacted the parent for an initial meeting to enroll them in and orient them to the FCU-O. Family coaches used motivational interviewing strategies to engage families, set goals, reinforce parents’ existing strengths, and implement evidence-based parenting strategies. The family coach has access to “backend” data on parents’ progress within the online program, including modules completed, time spent on modules, in-app assessment and feedback data, and activities completed. They used this information to provide strengths-based feedback, as in the original FCU model (Gill & Shaw, 2020). Family coaches were master’s and doctoral level clinicians trained to deliver the original FCU model with fidelity using the COACH system (Smith et al., 2013).

Telehealth coaching sessions were intended to flexibly correspond with parent’s completion of each of the five app modules, with approximately five to six 20-minute coaching sessions over a period of approximately three months. However, the frequency and duration of coaching sessions were tailored to parents’ personal needs and goals. As such, parents may have continued to have follow-up sessions with their family coach at the time that they completed the 3-month follow-up assessment. Both the FCU-O and telehealth coaching were available in Spanish.

Control Condition.

Caregivers randomized to the control group were informed that they would receive the intervention at the conclusion of the assessment period and that they would be contacted by research staff in three months to complete the 3-month follow-up questionnaires. In other words, they completed the 3-month follow-up without having been exposed to any intervention components. Approximately one year after completing the baseline assessment, the study team contacted parents in the control condition to offer them access to the FCU-O. A staff member assisted the parent in enrolling in the FCU-O and oriented them to the app’s structure and content. Telehealth coaching sessions were not offered as part of the control condition.

Measures

Demographic Characteristics.

Information about parent demographic characteristics, including parent race, ethnicity, income, and family composition, was collected through a demographic questionnaire administered at the baseline assessment. Family rurality status was determined based on whether their reported zip code was designated as rural according to the Oregon Office of Rural Health geographic definitions (Oregon Office of Rural Health, 2024).

Intervention Engagement.

Information on app engagement, including total minutes engaging with app content and number of modules that parent engaged with, was extracted. We used number of telehealth coaching sessions and total minutes of phone contact to represent parents’ engagement with the telehealth coaching intervention component.

Parenting Skills.

Parents completed the Parenting Young Children scale (PARYC; McEachern et al., 2012), a 21-item validated self-report measure of parenting skills. The PARYC asks parents to rate the frequency of their use of parenting skills in the past month on a scale from 1 (“not at all”) to 7 (“most of the time”). It is comprised of three subscales, each containing seven items: Supporting Positive Behavior (e.g., “reward your child when they did something well or showed a new skill”), Setting Limits (e.g., “explain what you wanted your child to do in clear and simple ways”), and Proactive Parenting (e.g., “avoid struggles with your child by giving clear choices”). A mean score for each of the three subscales was calculated for the baseline and 3-month follow-up assessments. Internal consistencies in the current sample were α = 0.71 (Supporting Positive Behavior), α = 0.75 (Setting Limits), and α = 0.77 (Proactive Parenting).

Parenting Self-Efficacy.

Parents completed a version of the Parenting Tasks Checklist (Sanders & Wooley, 2005) that was adapted and abbreviated for the current study. Parents were asked to rate 9 parenting skills (e.g., “providing praise and encouragement for good behavior”; “applying consequences when rules are broken”) on how confident they felt practicing the skills on a scale from 1 (“not at all confident”) to 5 (“very confident”), and how important they perceived the skill to be on a scale from 1 (“not important at all”) to 5 (“very important”). In the current study, we used the parenting confidence subscale and calculated a mean score for both the baseline and 3-month follow-up assessments. Internal consistency in the current sample was α = 0.84.

Parent Depressive Symptoms.

Parents completed the Patient Health Questionnaire-9 (PHQ-9; Kroenke et al., 2001), a 9-item self-report, validated screening measure of depressive symptoms experienced during the past two weeks. Item responses range from 0 (“not at all”) to 3 (“nearly every day”). A mean score was calculated to represent total depressive symptoms at both baseline and 3-month follow-up assessments. Internal consistency in the current sample was α = 0.89.

Parent Anxiety Symptoms.

Parents completed the Generalized Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006), a 7-item self-report, validated screening measure of anxiety symptoms experienced during the past two weeks. Like the PHQ-9, item responses range from 0 (“not at all”) to 3 (“nearly every day”). A mean score was calculated to represent total anxiety symptoms at both baseline and 3-month follow-up assessments. Internal consistency in the current sample was α = 0.92.

Data Analytic Plan

All analyses were conducted using MPlus (Muthén & Muthén, 1998–2017). Intent-to-treat (ITT) analyses were used to test intervention effects at the 3-month follow up on parenting skills, confidence, and mental health (i.e., anxiety and depression symptoms) using multivariate multiple regression analysis. Using an intent-to treat approach for overall intervention effects, a total of 254 families (after accounting for 15% attrition) were needed to provide sufficient power (i.e., > .80) to detect a condition effect of Cohen’s d = 0.3 or larger (Faul et al., 2009). We calculated Cohen’s d effect sizes for each outcome variable as mean change score between groups divided by the pooled pre-treatment standard deviation (Morris, 2008). We used Cohen’s guidelines to assess the magnitude of group differences, with d = 0.2 indicative of a small effect, d = 0.5 a moderate effect, and d = 0.8 a large effect (Cohen, 1998). All analyses included covariates of parent ethnicity (parent identifies as Hispanic/Latine or not), child age, rural status, and family income. Bivariate correlations were used to assess associations between family demographic variables and intervention engagement.

Missing Data

We tested whether participants without 3-month follow-up assessment data differed from those with complete data on a range of demographic characteristics and parent outcome variables measured at the baseline assessment using χ2 and t-tests; none were statistically significant. Missing data were handled using full information maximum likelihood estimation.

Results

Descriptive data

No significant differences emerged between the intervention and control groups on demographic characteristics or outcome variables measured at baseline. At baseline, parents’ mean score on the PHQ-9 was 7.67 (SD = 5.9), with 31% of the sample scoring at or above the recommended clinical range of 10 on this screening measure (Kroenke et al., 2001). The mean score on the GAD-7 was 7.70 (SD = 5.5), with 31% of the sample scoring at or above the clinical cutoff of 10 (Spitzer et al., 2006). One quarter (25%) of the sample endorsed marijuana use in the past 30 days. Approximately one third (30%) of the sample endorsed a lifetime history of opioid misuse.

Across the whole sample, parents reported receiving other substance use, mental health, and/or parenting support services that may have taken place during the study assessment period. At the 3-month follow-up assessment, 12% of parents reported that their child had received case management services in the past year. Just under half (44%) of parents reported having received therapy or counseling not specific to substance use (e.g., couples counseling, family therapy) in the past 12 months, and 15% of parents shared that they received treatment for substance use in that same time period. There were no significant differences in service utilization between the intervention and control groups.

Parents in the FCU-O condition spent an average of 87.84 minutes engaging with app content (SD = 54.0, range = 0-320 minutes), with the majority (73%) of parents engaging with content from all five of the modules. Parents completed an average of 5.99 coaching sessions (SD = 4.7, range = 0–36), with an average of 178.43 minutes of time spent in phone contact with the family coach (SD = 195.8, range = 0–1725).

Parents’ rating of their skills in supporting their child’s positive behavior at baseline was significantly negatively correlated with both minutes of app engagement and number of app modules that the parent engaged with (r’s = −.17, p’s = .03 and .02, respectively). Similarly, parents’ baseline rating of their limit-setting skills was significantly negatively correlated with minutes of app engagement (r = −.15, p = .04). These significant correlations both indicate that that parents with initially low levels of parenting skills engaged more with app content. Parental depression and anxiety at baseline were both significantly correlated with total minutes of coaching support and number of coaching sessions (depression: r’s = .23 and .29, p = .002 and < .001, respectively; anxiety: r’s = .15 and .19, p = .05 and p = .01, respectively).

Family income was significantly negatively correlated with parent depressive and anxiety symptoms (r = −.18 and r = −.14, respectively, p < .001 and p = .01, respectively), indicating that lower income was associated with higher levels of anxiety and depression. Parents who identified as Hispanic/Latine reported significantly greater skills in the areas of limit-setting and proactive parenting at baseline (r = .13 and r = .14, p = .02 and p = .01, respectively). Having an older target child in the study was significantly correlated with less use of positive behavior support at baseline (r = −.12, p = .03). See Table 2 for a complete summary of correlations among covariates, baseline outcome variables, and intervention engagement.

Table 2.

Correlations among study variables

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Parent is Hispanic/Latine	--
2. Family income	.02	--
3. Child age	.03	−.07	--
4. Rural residence	.12*	−.07	−.14*	--
5. Positive parenting T1	.08	−.06	−.12*	.11	--
6. Limit-setting T1	.13*	.09	−.06	.00	.52*	--
7. Proactive parenting T1	.14*	−.01	.02	−.01	.50*	.62*	--
8. Parenting self-efficacy T1	−.03	.03	−.07	−.03	.57*	.55*	.54*	--
9. Parent depression T1	.10	−.18*	.05	.14*	−.16*	−.12*	−.15*	−.30*	--
10. Parent anxiety T1	.13*	−.14*	.02	.13*	−.11*	−.12*	−.09	−.24*	.76*	--
11. Minutes engaged in app†	−.15	.03	−.04	.05	−.17*	−.15*	−.07	−.13	.07	.00	--
12. Number of app modules parent engaged with†	−.06	.13	−.14	.08	−.17*	−.05	−.02	−.01	.05	−.03	.71*	--
13. Minutes of coaching support sessions†	−.01	−.09	−.12	.01	−.04	−.06	.00	−.05	.23*	.15*	.50*	.34*
14. Number of coaching support sessions completed†	−.07	−.06	−.14	.05	−.06	−.07	−.04	−.07	.29*	.19*	.49*	.39*	.88*

^*p < .05

^†Intervention group only

Intent-to-Treat Analyses

Results for all ITT analyses are presented in Table 4. The FCU-O was associated with statistically significant improvements to all parenting skills (supporting positive behavior, setting limits, and proactive parenting) and parenting self-efficacy. It was also associated with statistically significant reductions in symptoms of depression, but not anxiety. Associated effect sizes (Cohen’s d) were in the small to medium range, with the strongest effect on parenting self-efficacy (d = .38), followed by proactive parenting and limit-setting skills (d = .31 and .33 respectively).

Table 4.

Multivariate multiple regression results assessing intervention effects

Variable	Beta	SE	p	β	Cohen’s d effect size
Supporting positive behavior	1.38	.56	.013*	.12	.18
Setting limits	2.49	.55	< .001*	.23	.31
Proactive parenting	3.13	.61	< .001*	.26	.33
Parenting self-efficacy	.26	.06	< .001*	.21	.38
Parent depression	−0.94	.47	.045*	−.09	−.22
Parent anxiety	−.67	.47	.15	−.07	−.12

Note: Regression analysis included baseline levels of each outcome variable, as well as covariates of family income, rural status, child age, and parent ethnicity also included in this analysis.

^*p < .05

Discussion

Overall, our findings provide initial support for the efficacy of the FCU-O in improving parent-reported parenting skills, parenting self-efficacy, and depressive symptoms within a socioeconomically and geographically diverse sample. Most families in this study were White, consistent with rural populations in the broader U.S. (Johnson & Lichter, 2022), and reported an income below the U.S. median family income. Almost half of the parents resided in rural areas and many were experiencing mental health challenges. Approximately one third of parents reported clinically significant levels of anxiety, depression, and/or had a history of substance abuse. The current study’s findings support prior research on the FCU and FCU-O’s efficacy in improving proactive and positive parenting, as well as limit-setting strategies (Connell & Stormshak, 2023; Gardner et al., 2007). FCU-O modules and coaching intervention sessions were explicitly designed to promote these parenting behaviors, and, consistent with study hypotheses, our findings showed increases in positive behavior support and proactive parenting (Dishion et al. 2008; Stormshak, Seeley et al., 2019). Overall, the FCU-O demonstrated multiple benefits to families experiencing mental health and socioeconomic challenges.

The FCU-O’s effect on parenting self-efficacy is consistent with prior research demonstrating increased self-efficacy in parents of middle schoolers who received the FCU-O with coaching (Stormshak, Seeley et al., 2019). As in the current study, the sample was predominantly non-Hispanic White and the majority were from suburban or urban areas, so it is not clear whether the benefits of the FCU-O for improving parenting self-efficacy would extend to more ethnically and racially diverse parent populations. Parenting self-efficacy has previously been found to mediate the relationship between the FCU and proactive parenting skills. Specifically, assignment to the kindergarten version of the FCU was associated with higher levels of parenting self-efficacy in second grade, which subsequently led to greater use of proactive parenting strategies when children were in third grade (Resnik et al., 2023). Findings from the current study contribute to research demonstrating the benefits of the FCU for improving parenting self-efficacy, an important component of parenting and family context that has been found to longitudinally predict parenting behaviors (Rominov et al., 2016). Further, given the high proportion of low-resourced, rural families present in the sample, findings indicate that the FCU-O increased parent self-efficacy among the population the program was designed to support.

We found a decrease in parent depression at the 3-month follow-up for families in the FCU-O that is consistent with prior research on the original FCU (Connell & Stormshak, 2023; Shaw et al., 2009). Notably, almost three quarters of the parents in the intervention group viewed and interacted with content from all five parenting modules, and higher levels of parental depression and anxiety at baseline were associated with more total minutes of coaching support. It is possible that having the opportunity to interact with a parenting coach in the FCU-O may have contributed to decreased depression due to accessing this social support. This would be consistent with prior work finding that social support predicted lower parental depression among high-risk and low-resourced parents with young children (Herbers et al., 2023). The increased usage of the app among parents with more symptoms of depression and anxiety speaks to the high accessibility and relevance of the app content for parents with current mental health challenges. Further, emotional and informational support can be less available to families living in rural areas (Bardach et al., 2011), which may additionally explain high app usage in the current study. The magnitude of the effect of the FCU-O on improving parenting skills, parenting-self-efficacy, and parent depressive symptoms were in the “small” range, ranging from approximately .20 (supporting positive behavior; depressive symptoms) to .38 (parenting self-efficacy). These effect sizes are similar to the effects of other online behavioral parenting interventions on improving parenting behaviors (Hedge’s g = 0.34; Florean et al., 2020) and parent self-efficacy (Hedge’s g = 0.37; McAloon & Armstrong, 2024). Although our sample consisted of parents at risk for experiencing challenges with parenting and/or child behavior on the basis of reporting current or past substance use and/or depressive symptoms, the current study utilized a preventive design as parents were not actively seeking treatment. Effect sizes in the current study are therefore consistent with prior work on the original FCU model (e.g., Brennan et al., 2013) and other preventive parenting interventions (e.g., Gross et al., 2009).

Results of the current study also indicated that parents with low levels of positive parenting and limit-setting skills at baseline engaged more with app content. Few prior studies have tested whether parenting skills at baseline are associated with engagement in online parenting modules. One study did not find an association between baseline parenting skills and completing the minimum dose of the Triple P Online program (Baker & Sanders, 2017). Another study of a telehealth-based program for parents of children with autism spectrum disorder found that parents’ baseline depressive symptoms were associated with not completing the online program (Ingersoll & Berger, 2015). In the current study, correlations between parents’ baseline symptoms of depression and anxiety and online program engagement were not significant. This suggests that common parent mental health challenges, such as depression and anxiety, are not a significant barrier to engaging with FCU-O program content.

Our findings also provide support for the FCU-O’s effective engagement of families living in rural areas. Although increasing access to parenting support for families in rural areas has been identified as a goal for online parenting programs (e.g., Holtrop et al., 2023), to our knowledge, no study to date has focused on online interventions directly targeting families residing in rural areas. Notably, less than half of adults with mental health concerns, and less than 10% with substance use disorders, received treatment over the course of the year in 2020, indicating a critical need for affordable and accessible supports for families (Counts, 2022). In the current study, one third of parents resided in a rural area, and the correlations between rurality status and intervention engagement variables were not significant. However, overall intervention engagement was very high, indicating that the FCU-O effectively engaged parents in rural areas similarly to those living in urban and suburban areas. This is important given that families living in rural areas face several challenges to parenting and barriers to care that may make an online parenting program a particularly good fit. For example, transportation is reported to be a major barrier to accessing healthcare (Arcury et al., 2005), and children living in rural areas are more likely to have a parent with mental health challenges (Robinson et al., 2017). In addition, parents who use substances in rural areas have been found to experience significantly lower odds of receiving treatment than their urban counterparts (Ali et al., 2022). Future research on this intervention should include testing whether living in a rural area operates as a moderator of intervention effectiveness and/or engagement.

Surprisingly, no differences in parent anxiety were found among those who participated in the intervention. It is possible that due to the stressors experienced by parents in this sample (e.g. low income, less access to resources, mental health challenges), parent anxiety remained high despite having access to the intervention. Interestingly, we found evidence for differential effectiveness of the FCU-O on reducing parents’ anxiety symptoms, whereby the intervention was associated with reduced anxiety only for parents with a history of adolescent-onset cannabis use (Hails et al., 2024).

Findings of the current study report on outcomes from the first follow-up (3-month), which typically occurred immediately following the intervention. The current study’s findings require replication at the 6-month and 1-year follow-up assessments. A deeper investigation of app engagement and usage across time will help to better understand the mediators and pathways between changes in parent mental health and parenting skills. Additionally, because we offered the app to parents in the control condition at the conclusion of the assessment period, we can compare both app engagement and parenting outcomes for parents who received a self-directed version of the app versus those who had access to additional support from a parenting coach.

Limitations

This study includes a few notable limitations. First, results include data from the first follow-up assessment at three months. Further analyses need to be conducted once the sample has completed all time points to investigate intervention effects longitudinally. Second, the study only includes parent-report data. Future studies of the FCU-O could include additional information such as observational data or teacher report of child behaviors in the classroom. Third, the sample is predominantly White, largely reflecting the demographics of the area in which families were recruited. While past research has found that race, ethnicity, and language of intervention delivery (i.e., English or Spanish) does not moderate the FCU’s effects (Berkel, Fu, et al., 2021; Smith et al., 2014), there is some evidence of baseline and intervention engagement differences between families and youth of different racial and cultural groups (Berkel, Mauricio, et al., 2021; Smith et al., 2014), suggesting that parenting needs may vary by race, ethnicity, culture, and language. Furthermore, past exploratory research on another parenting intervention – the home-based Triple P – found that race and ethnicity may moderate intervention effects among families specifically in rural areas (Abate et al., 2019). Taken together, findings of past research suggest that engagement with and outcomes of the FCU-O could vary in a sample more ethnically diverse than the one in this study. Therefore, it is critical for future researchers to investigate the FCU-O with a more racially and ethnically diverse sample. Finally, family coaches in the current study were master’s and doctoral clinicians. Future studies should investigate whether the FCU-O’s effects are maintained when family coaches are not trained mental health providers, which could further reduce costs associated with program implementation and other barriers to families’ accessing parenting support. Prior research suggests that effect sizes are similar for clinicians and non-clinical technicians in supporting patients on an internet-based cognitive behavioral therapy program for depression (Titov et al., 2010), but this needs to be tested with the FCU-O.

Conclusions

The current study demonstrates the effectiveness of the FCU-O intervention for improving parent-reported parenting skills, parenting self-efficacy, and depressive symptoms among a group of socioeconomically and geographically diverse families who were predominantly non-Hispanic White. Future research will test longitudinal outcomes for families (6-month and 1-year follow up assessments). The FCU-O for young children has the potential to support parents and families as a cost-effective and accessible app to promote family wellbeing.

Table 3.

Sample descriptive statistics by intervention assignment on key outcome variables

	Control			Intervention
	Baseline (N = 176) M (SD)	3-Month Follow-Up (N = 167) M (SD)	Mean Change Score Baseline - 3-Month Follow-Up	Baseline (N = 180) M (SD)	3-Month Follow-Up (N = 170) M (SD)	Mean Change Score Baseline - 3-Month Follow-Up
Supporting positive behavior	37.43 (6.5)	37.49 (6.5)	.06	37.35 (5.6)	38.51 (5.3)	1.16
Setting limits	35.04 (6.7)	36.35 (5.8)	1.31	35.40 (6.1)	38.72 (5.6)	3.32
Proactive parenting	35.49 (7.6)	36.68 (6.8)	1.19	36.45 (7.0)	40.03 (5.5)	3.58
Parenting self-efficacy	3.73 (.7)	3.78 (.7)	.05	3.74 (.7)	4.06 (.6)	.32
Parent depression	7.20 (5.7)	6.67 (5.4)	−.53	8.13 (5.9)	6.34 (5.6)	−1.79
Parent anxiety	7.41 (5.3)	6.70 (5.5)	−.71	8.00 (5.7)	6.63 (5.3)	−1.37

Highlights.

• The Family Check-Up Online (FCU-O) is a parenting program with telehealth coaching.

• Parent participants faced challenges including mental illness and substance misuse.

• Participants were primarily low income, and about one third resided in rural areas.

• Findings offer preliminary support for the FCU-O among parents of young children.

• The FCU-O improved parenting skills, parenting self-efficacy, and depressive symptoms.