Increasing Parenting Self-Efficacy Through a Community Partnership in Akron, Ohio

Background: The Nurturing Family Program (NFP) is a family-centered educational curriculum designed to equip parents with tools and techniques to support a nurturing relationship with their child. While there have been many evaluations of the NFP, no evaluations to date measure how the curriculum may influence parenting self-efficacy, how confident a caregiver feels about their ability to foster their child’s development and success, or parenting stress. Methods: This community-based interventional study used a convenience sample of families recruited from community agencies or within a pediatric medical center. Families with one or more children aged 0 to 5 years, premature infants or an infant discharged from the neonatal intensive care unit (NICU), and/or adolescent parents were invited to participate in NFP. Participants were recruited from community agencies that were most likely to work with populations meeting program eligibility criteria. The NFP was offered as in-home and community-based sessions. The primary outcome of interest was perceived parental self-efficacy, measured using the Tool to Measure Parenting Efficacy (TOPSE) at weeks 1, 5, and 10. The secondary outcome of interest was perceived level of stress, as measured by a visual analog scale each week. Results: Participants of community-based sessions (n = 45) experienced a significant increase in parenting self-efficacy at week 10, compared to week 1 and week 5 (p < 0.01). Overall, participants of the community-based sessions experienced an average increase of 17.7 points on the TOPSE scale (p = 0.014). Across all participants (n = 79), each session attended resulted in a modest, but significant, 3% reduction in stress (p = 0.021). Conclusion: The NFP improved parental self-efficacy among participants of the community-based sessions and reduced perceived stress for all participants.


INTRODUCTION
Early life experiences matter not only in terms of early childhood development but also because of the impact on educational, health, Parenting is influenced by multiple factors across the individual, interpersonal, community, and societal levels. 3 Parenting selfefficacy can be defined as how confident a caregiver or parent feels about their ability to successfully foster their child's development and success. 4,5 Parental self-efficacy is a core life skill that has been found to be associated with promoting positive child behavioral, emotional, physical health, and social outcomes. 6,7 Parenting difficulties are a major source of stress for parents, and parenting self-efficacy has been shown to be an important buffer against parenting stress. Individuals with strong parental self-efficacy are able to guide their children through the developmental stages they face without serious problems or undue strain on their relationship with their spouse or partner. Individuals low in parental selfefficacy may struggle to meet familial demands and are at risk of stress and depression. 8 Programs that seek to support and build parental self-efficacy are critical opportunities for promoting relational health characterized by safe, stable, and nurturing relationships (SSNR). The American Academy of Pediatrics (AAP) suggests that such programs are an important component of adopting a comprehensive public health approach, and require effective collaboration between the pediatric health system, families, and community partners. 2 This article describes outcomes of the Nurturing Family Program (NFP) offered through a partnership between a pediatric medical center, families, and key community partners. The NFP is a parenting program that provides tools and techniques to parents to support nurturing relationships. 9 While the NFP has been found to reduce the likelihood of being reported for maltreatment among program participants, specifically those who attend a high number of sessions, 10 no NFP evaluations to date have examined the potential effects of the curriculum on parenting self-efficacy, [11][12][13][14] and almost no studies have looked at stress reduction as an NFP outcome.
The primary objective of the study was to explore whether participation in the NFP changed parent perceptions of self-efficacy. Our secondary objectives focused on understanding whether participation in the NFP changed parents' perceived level of stress.

METHODS
All study methods were reviewed and approved by the Akron Children's Hospital institutional review board (IRB # 17-2160).

Procedures/Program Description
The NFP, also referred to as the Nurturing Parenting Program, is a family-centered educational curriculum designed to equip parents with tools and techniques to support a nurturing relationship with their child. 9 The NFP can be customized to needs of specific populations, such as children with special needs or parents who are incarcerated. 15,16 Parents and their children participated in weekly NFP sessions for 10 weeks, with each session lasting approximately 2.5 hours. A description of topics discussed during each session is outlined in the Appendix. Both the in-home and community-based groups received the same curriculum. Families in both the in-home and community-based groups were also provided a kit, which included developmentally appropriate toys and books, educational materials, and a community resources guide. This program was offered 7 times over 2 years (n= 10 sessions x 7 programs = 70 total sessions) and all participants were invited to engage in the research study component. The program facilitators were employees of the pediatric medical center, and the community partners provided the space for the group sessions. Importantly, all parents who participated in NFP received the same intervention. Only parents who elected to participate in the research study component were asked to complete assessments to measure parenting self-efficacy and stress.
This was the first time NFP was administered in Akron. The NFP was chosen to respond to increasing rates of child abuse and neglect, identified as a community health need through recent assessments. 17,18 Setting Akron is an urban city with a strong, cross-sector collaborative approach to improving community health. As the fifth largest city in Ohio, slightly more than 190 000 people live in Akron. 19 According to the 2020 Census, 6% of residents are under 5 years of age, with 21% of residents under the age of 18 years. 19 Nearly 60% of people living in Akron self-identify as White, 30% as Black or African American, and 5% as Asian. 19 About half of the homes in Akron are owner-occupied and median monthly rent is $759. 19 Neighborhoods throughout Akron experience multiple poor outcomes that endanger relational health and support the need for parenting programs, including preterm birth, low birthweight, lead exposure, social isolation, changing family structure, chronic poverty, parenting stress, family disorganization, violence, parent history of neglect and abuse in family of origin, and parental beliefs and knowledge about the role that they play in the development of their children. 20

Community, Participant Characteristics, Recruitment
Members of the collaborative clinical, community, and local governmental partnership played an active role in recruitment and implementation. Families were referred to the NFP through community agencies, as well as through Akron Children's Hospital's Maternal and Fetal Medicine (MFM) and Neonatal Intensive Care Unit (NICU). Families (also referred to as participants) were invited to participate in the program if they met one or more inclusion criteria: (1) were parents of one or more children aged 0 to 5 years, (2) were parents of a premature infant or an infant discharged from the NICU, and/or (3) were adolescent parents. Families involved with the court system, or child protective services, were excluded due to challenges in obtaining consent to partici-pate in either the NFP or the research study component. Additionally, families for whom English was not their primary language were also excluded due to lack of accurately translated and validated teaching materials.
Families recruited through community partners selected their preferred location for NFP, either individual sessions in their home or group sessions at a community partner site, depending on which was most convenient for their family. Parents who were recruited through hospital providers and staff were offered individual sessions in their home, also referred to as the in-home NFP program, but could also choose group sessions at a community partner site, referred to as the community-based NFP program. Parents recruited through both strategies were invited to participate in the research study component of the program. Participation in NFP was voluntary and engagement in the research study component was optional. At the beginning of the first session, informed consent was obtained by research staff from parents who elected to participate in the study. Staff reviewed the entire consent form with each parent and answered any questions before obtaining their signature. The research study component began October 1, 2017, and concluded September 30, 2019.

Measures/Outcomes
Demographic data were collected at week 1, and attendance was tracked weekly. Outcome data focused on changes in perceived parenting self-efficacy and changes in reported stress. Parenting self-efficacy was measured at weeks 1, 5, and 10 by the Tool to Measure Parenting Efficacy (TOPSE). 21,22 Based on self-efficacy theory, the TOPSE instrument consists of 48 items divided into 8 subscales/parenting dimensions: emotion and affection (I am able to show affection towards my child); play and enjoyment (Playing with my child comes easily to me); empathy and understanding (I am able to comfort my child); control (I can remain calm when facing difficulties); discipline and setting boundaries (I am able to reason with my child); pressures (It is difficult to cope with other people's expectations of me as a parent); self-acceptance (I can manage the pressures of parenting as well as other parents do); and learning and knowledge (I am able to learn and use new ways of dealing with my child). 21,22 Internal reliability for the subscales ranges from 0.80 to 0.89 with the overall reliability at 0.94. 22 Stress was measured weekly through the use of the visual analog scale (VAS), an efficient, empirically based method widely used to measure perceived stress. 17 As participating families filled out their attendance form at each session, they were asked to "Indicate on this scale how stressed you feel at this time" by marking on a 10 cm line. The left endpoint was labeled "no stress" and the right endpoint was labeled "high stress." Trained research staff measured the number of centimeters starting from the left endpoint to the nearest quarter of a centimeter. All data were collected through paper and pencil, and acquired prior to the start of the NFP session. Data were then entered by trained research staff into an Excel database for analysis. A table outlining the data collection timeline is available in the Appendix.

Statistical Analysis
Data were imported into SPSS statistical software 23 and analyzed via 2-sided statistical testing with p < 0.05 considered statistically significant. Participant characteristics were summarized by program location using frequencies and percentages for categorical data and means and standard deviations for numeric data. Comparisons of the distribution of categorical data between program locations were performed via Pearson chi-square or Fisher exact test depending on cell sample size. For analysis of a possible effect of the program location (in-home versus community-based), TOPSE scores were compared via repeated measures ANOVA with a between group effect for program location. The repeated measures analysis of variance (ANOVA) model had an unfulfilled assumption of sphericity verified via significant Mauchly test of sphericity. A Greenhouse-Geisser correction for non-sphericity was then employed in determining the effect of study time point on the TOPSE score outcome.
Since the effect of time was significant in the repeated measures ANOVA model, data were compared across locations at each of the study time points. Data were compared for rank equivalence between location at each of the week 1, 5, and 10 study time points via Mann-Whitney U tests. These tests were employed due to failed normality assumptions required for independent samples t tests. Data were subsequently aggregated across the 2 program locations due to insignificant location main effects. Pairwise leastsignificant difference tests determined which study time points were significantly different in mean TOPSE scores. The primary analysis of the mean change in TOPSE scores from week 1 to week 10 study time point was performed on the aggregated cohort across locations. A paired samples t test compared the mean change for equality to zero since the aggregated sample size allowed for invocation of the central limit theorem. A sensitivity analysis was also performed to determine the relative influence of outliers and those who dropped out prior to week 10. Missing data were imputed using the week 5 TOPSE value or week 1 TOPSE value if unavailable carried forward to week 10 for those subjects with missing week 10 TOPSE values. Changes in specific domain scores of the TOPSE from week 1 to week 10 were also analyzed similarly to the overall change score.
The other measures involving numeric metrics (demographics, attendance, VAS stress) were correlated with the change in TOPSE score to week 10 to determine factors that might have significantly influenced gains in self-efficacy. Changes in VAS stress from week 1 to week 10 were assessed using a paired samples t test and the correlation with class attendance determined. This significant correlation prompted the use of a linear regression model to determine the influence of class attendance on reductions in parental stress.

ojph.org
Ohio Public Health Association

Participant Characteristics
Over two-thirds of families participating in the NFP also elected to participate in the research study component (79/117 = 68%). Characteristics of families who participated in the research study component are presented in Table 1. Across both groups, the majority of participants were single (in-home: 84%, communitybased: 80%), had a high school diploma (in-home: 50%, community-based: 55%), and an annual household income less than $15 000 (in-home: 83%, community-based: 67%). The inhome and community-based groups did not differ significantly on marital status (p = 1.000), education level (p = 0.335), or annual household income (p = 0.269). However, the percentage of participants currently pregnant was significantly different between the in-home and community-based groups (25% vs 50%, p = 0.031).

Attendance
The average (SD) class attendance out of 10 classes was 6.8 (3.29) classes across the 2 groups. There was a significant difference in the mean (SD) attendance between the in-home and communitybased groups (8.1 (2.23) vs 5.8 (3.63), p = 0.001).

Parental Self-Efficacy
Changes in self-efficacy were measured by the changes in TOPSE overall scores from week 1 to week 10. In-home participants were significantly less likely to drop out early (p < 0.001, Table 2). The effect of program location on the total TOPSE score across the three study time points was insignificant (p = 0.799, partial eta squared effect size = 0.02) so analysis was conducted on the aggregated program data across study arms. However, the repeated effect for time was highly significant, indicating that TOPSE scores differed across the 3 measurement time points (p = 0.003 via Greenhouse-Geisser adjustment for sphericity violation).
A post hoc least-significant differences (LSD) procedure was performed to determine pairwise distinction across the 3 time points. Weeks 1 and 5 were comparable on average (p = 0.642); however, each differed significantly from week 10 (p < 0.01 for each). Participants exhibited a significant increase in parenting self-efficacy during the second half of the program. The significant mean (SD) increase of 17.7 (46.55) (p = 0.014, Table 2) reaffirms the repeated measures results to indicate significant improvement in TOPSE for those participants completing the week 10 assessment. A sensitivity analysis was also performed imputing a value of zero improvement for those not completing the week 10 assessment to reveal similarly significant results (p = 0.015). The sensitivity analysis indicates that the conclusions of significant improvement in TOPSE are consistent even after adjustment for the influence of attrition. The significant gains in TOPSE at week 10 relative to week 1 were further probed by looking at the specific domains that comprise the validated metric. Each of the 8 domains exhibited mean improvement; significant gains were determined specifi- ojph.org Ohio Public Health Association cally for discipline and boundaries, control, and empathy and understanding (p < 0.001, p = 0.011, p = 0.023 respectively).

Parenting Stress
Parenting stress significantly decreased from week 1 (mean=3.74) to week 10 (mean=2.48; p=0.041, Table 2). This reduction represents a 34% decrease in stress over the 10-week study period for those providing week 10 data. The correlation between number of classes and VAS scale was calculated to explore a per class trend in stress reduction to include all participants and not just those who provided week 10 data. A significant negative correlation (r = -0.100, p = 0.021) indicates that class attendance was significantly associated with reduced stress, regardless of whether the participant completed the program. Each class attended contributed a modest but significant reduction in stress of 0.109 (3%) as measured by the VAS.

DISCUSSION
The purpose of this study was to evaluate the effects of a parenting program offered through a collaboration between a pediatric hospital and community agencies that included local housing authorities and a mental health agency. The study looked at changes in 2 key parenting variables shown to be critical determinants of relational health: self-efficacy and stress. Parenting self-efficacy in both program locations improved significantly over time. Changes were especially strong in 3 domains: empathy and understanding; control; and discipline and boundaries. Parenting stress in both locations also significantly decreased from week 1 to week 10. The increases in self-efficacy and the reduction in stress over time reflect the interconnectedness among the 3 principles to improve child and family outcomes: supporting responsive relationships, strengthening core life skills, and reducing sources of stress for children and adults. 1 Program attrition rates have consistently been a challenge in evaluating parenting programs serving vulnerable populations. Multiple studies have indicated attrition rates ranging from 30% to 70%. 24 In this study, the use of sensitivity analysis indicated that the changes in TOPSE scores continued to be significant even after adjusting for the influence of attrition. However, attrition was indeed higher in the community-based group, suggesting possible barriers such as transportation, child care, or feeling uncomfortable in a group setting.
The findings did show that the more sessions parents attended (or the greater the "dosage" of intervention), the greater was the increase in self-efficacy scores and the greater the decrease in parenting stress levels. The significant changes appeared to have occurred between week 5 and week 10. Perhaps the changes were due not only to simply "showing up" week after week, but also due to increased engagement over time with the program. Engagement is affected by a variety of provider, parent, program and neighborhood characteristics, and is more multidimensional than "showing up;" it includes aspects such as listening, asking questions, asking for advice, and applying the information between ojph.org Ohio Public Health Association 99 sessions. 25 Any future parenting program evaluation should consider a closer look at potential engagement factors and measure more than attendance.
While this study has many strengths, results should be considered in light of its limitations. First, as described in the Methods section, the study used a convenience sample to evaluate the program. Next, collection of additional demographic data from participants might have identified additional mediating engagement variables that could have helped explain, for example, why the inhome group and community-based group differed in terms of length of participation. Long-term follow-up after session 10 was not possible, so it is not known if participants were able to maintain the short-term significant changes in self-efficacy and stress in the months after program completion. Additionally, the study was completed in a single geographic and urban area and may not be generalizable to communities across the state. Lastly, analyses were limited by consistency in program participation. Nevertheless, the finding that increased participation in a parenting program led to stronger outcomes such as self-efficacy and decreased stress is consistent with other recent research. 26 As a result, the collaborative partners continued the program throughout the pandemic and continue to offer the program today. Importantly, the NFP was administered in a virtual environment at the onset of the pandemic and continues to be administered in virtual individual and group sessions today. Future evaluations should explore whether the same positive outcomes will be experienced by families participating in a virtual setting.

PUBLIC HEALTH IMPLICATIONS
The 2021 AAP Policy Statement (Preventing Childhood Toxic Stress: Partnering with Families and Communities to Promote Relational Health) describes a vision for a comprehensive public health approach to building "relational health," a key predictor of many aspects of wellness later in life. 2 The AAP vision emphasizes a public health approach that not only addresses toxic stress but also addresses, promotes, and repairs relational health, and is collaborative in nature so that a variety of interventions are integrated both vertically (within the health care system) and horizontally (with community agencies). A public health approach is crucial since no one system alone can successfully reduce the burdens of toxic stress and promote or repair relational health in vulnerable communities. 27 Although small in size, the study can be seen as a promising step toward building a successful and comprehensive community public health approach to improve outcomes for children and families. The development and implementation of the program was in response to data from a community-wide public health assessment that identified neighborhoods experiencing poor outcomes for children and families. The effort was led by the pediatric hospital in collaboration with community partners serving those neighborhoods. The parenting program was chosen based on its focus on building relational health and SSNRs. Although outside funding supported training for the session leaders and the initial implementation and evaluation of the program, the program is now financially supported by the hospital. The collaboration with community partners continues to grow.
This study describes an approach and model for building "relational health" not only between parents and children, but also between a medical center and surrounding community agencies.
Communities considering replicating such an approach will have many decisions to make: What does our community needs assessment data tell us? What population shall we serve? What program or intervention shall we choose? Who will take the lead? Who will work in partnership with the lead? How can we recruit families and keep them engaged? How will we know if our approach is successful? Such an effort is not without its challenges and bumps in the road; however, the hard work involved in designing a response to community needs assessment data can have long-lasting positive effects for children, families, and communities.
ojph.org Ohio Public Health Association

Detailed Curriculum
Session 1: Welcome Session 1 introduces the program themes, discusses nurturing parenting principles and how parents impact a child 's emotional development, describes personality traits and how they relate to parenting, defines ways parents can become nurturing parents, and encourages parents to discuss their hopes and fears.

Session 2: Building Joy and Praise
Session 2 is designed to improve parents' and children's self-worth and self-concept. Participants discuss the definitions of self-concept, self-esteem, and self-worth; realize their ability to impact a child's overall feeling of worth; learn about labels and how they hurt children; and explore ways parents can improve their children's self-concept, self-esteem, and self-worth.

Session 3: Discipline
Session 3 is designed to increase parents' awareness of the other ways to discipline besides spanking. Participants discuss why parents spank their children, learn how spanking is detrimental to children, and identify alternatives to spanking.

Session 4: Warmth, Needs and Nurturing
Session 4 is designed to increase parents' skills in developing family morals, values, and rules. Participants discuss family morals, give examples of family morals, relate family morals to family discipline, define family values, define the purpose of family rules, and practice making family rules.

Session 5: Problem Solving
Session 5 focuses on brain development and the functions of the brain. Parents learn methods on how to problem solve using a series of techniques and how to handle conflict.

Session 6: Ages and Stages
Session 6 is designed to increase parent's awareness of appropriate expectations of children. Participants increase their knowledge of age-appropriate expectations of children and stages of child development.

Session 7: Handling Stress and Anger
Session 7 is designed to increase parents' ability to recognize and handle stress. Participants explore the meaning of stress, identify ways adults create stress, learn about stressors children face and how to help them cope, and learn ways to help parents and children reduce stress.

Session 8: Feelings
Session 8 is designed to help parents recognize and understand their feelings and the feelings of their children. Participants discuss feelings of comfort and feelings of discomfort, learn how to help their children understand feelings, learn to recognize feelings in children, and learn how to teach children how to manage their feelings.

Session 9: Communication
Session 9 is designed to help parents recognize and communicate their feelings and the feelings of their children. Participants practice how to communicate their feelings in a healthy way, define I-Statements and You Messages, learn negotiation skills and ways to confront without criticizing, and learn to model appropriate communication styles.