Article Text
PDF
Abstract
Objectives This study aimed to examine the preliminary efficacy of the FirstStep2Health versus usual care control on improving parents’ lifestyle behaviours (moderate to vigorous physical activity, screen time, fruit/vegetable and fibre intake, skin carotenoids), nutrition and physical activity knowledge, self-efficacy, support, parenting style, feeding practices, home environment, anthropometric outcomes (body mass index, % body fat) and blood pressure from baseline to postintervention after adjusting for random cluster effects.
Design A cluster randomised controlled trial with 10 Head Start daycare centres (five intervention, five control) was conducted using computer-generated randomisation after baseline data collection.
Setting US Head Start daycare centres.
Participants 95 parent-child dyads (53 intervention, 42 control).
Interventions The 16-week, dyadic, FirstStep2Health intervention included: (1) a daycare-based child programme on healthy mindful eating and physical activity, (2) child letters to parents to connect school learning with home practice, (3) social media-based parent programme to assist parents to promote healthy eating and physical activity at home, (4) virtual group parent meetings via Zoom on topics related to healthy eating and physical activity and (5) weekly motivational messages to increase parental motivation to build a healthy home environment.
Results Mixed-effect models were used to examine intervention effects, adjusting for baseline outcome and cluster effects at the daycare and classroom levels. Intervention parents engaged in more moderate to vigorous physical activity (B=0.49, p=0.874) postintervention than controls, although not significantly. However, intervention parents showed significantly higher nutrition knowledge (B=0.87, p=0.009), physical activity knowledge (B=0.95, p=0.049), nutrition self-efficacy (B=0.74, p=0.025) and physical activity self-efficacy (B=0.86, p=0.013) compared with controls at postintervention. Fibre intake was also significantly higher (B=2.99, p=0.049), and intervention parents had lower % body fat (B=−2.56, p=0.005) and systolic blood pressure (B=−10.98, p=0.005) postintervention. No significant effects were found for fruits/vegetables intake, parental support for healthy behaviours, home physical activity environment or authoritative parenting style.
Conclusions Future endeavours to proactively engage parents in a dyadic childhood obesity prevention approach such as the FirstStep2Health intervention are warranted to improve outcomes among both children and parents.
Trial registration number NCT04164277.
- obesity
- behavior
- primary prevention
- blood pressure
- family
- parents
Data availability statement
Data are available on reasonable request. Study data can be requested from the corresponding author.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The study is a cluster randomised controlled trial to account for cluster effects of daycare centres and classrooms.
The study applied objective measurements including height, weight, % body fat, blood pressure and skin carotenoids.
The study sample is ethnically and racially diverse with low-socioeconomic status.
The study occurred under a global pandemic context, limiting the generalisability.
The study had a relatively small sample size (n=95 Head Start parents) with 91.6% being female.
Introduction
Parents significantly influence children’s health behaviours.1 2 Childhood obesity experts emphasise implementing obesity prevention strategies with parents during preschool years rather than later in childhood, because parental attitudes, beliefs and behaviours play a critical role in shaping young children’s lifestyles and weight status.3 Children with parents who are overweight or obese have a higher risk of developing similar conditions both during childhood and into adulthood.4 Especially before age 12 years, children’s trajectories for developing overweight or obesity are largely influenced by their parents’ weight status.5 For example, a recent meta-analysis with 23 studies4 found a pooled OR of 1.97 (95% CI 1.85, 2.10) for overweight or obesity between parents and children.
With a high prevalence of overweight (30.7%) or obesity (42.4%) among US adults,5 children nowadays are at an elevated risk of developing these conditions unless timely and effective parent-involved interventions are implemented. Dyadic interventions targeting both parents and children, rather than focusing on one group alone,6 are shown to have greater sustaining effects on reducing children’ body mass index (BMI).7 However, effective obesity interventions targeting young children and parents living in poverty remain limited.8 Furthermore, active parent engagement in childhood obesity prevention interventions has been linked to improvements in children’s physical activity and diet quality.9 Yet, the specific benefits parents experience in dyadic interventions remain relatively obscure.10
Since the family, particularly parents, heavily influences children’s diet and physical activity choices,1 obesity is seen as a family issue. Familial factors—such as genetics and shared environments—contribute to childhood obesity, but focusing on the modifiable shared obesogenic environment can better address intergenerational obesity transmission.11 12 This environment is usually highly conducive to unhealthy lifestyles, such as high screen time, low physical activity and inadequate intake of fruits/vegetables (F/V) and fibre, among both parents and children.13 Thus, to eliminate childhood obesity within a family context, effective dyadic obesity prevention interventions should aim to cultivate a healthy home environment that fosters nutritious eating and regular physical activity.
Most dyadic interventions focus on improving parents’ knowledge of physical activity, nutrition and feeding practices, but rarely address parents’ lifestyle behaviours or anthropometric outcomes.3 As strong correlations exist between parents’ and children’s lifestyles and anthropometrics,4 14 thoroughly understanding the impact of dyadic interventions on parents’ lifestyle behaviours and anthropometrics (eg, BMI, % body fat) is crucial in mitigating obesity-related comorbidities from a family perspective. Recognising how these interventions benefit both parents and children is valuable, as it highlights the reciprocal influences within the family system.2 Moreover, this understanding can lead to the development of more effective interventions for improving family health, fostering supportive home environments and promoting positive behaviour changes within the entire family system.
Given that % body fat is a stronger predictor of cardiovascular risk compared with BMI,15 and that systolic and diastolic blood pressure (BP) are key indicators of cardiovascular health,16 evaluating these metrics in parents is fundamental for generating evidence to improve family cardiovascular health. Therefore, the purpose of this study was to evaluate the preliminary efficacy of FirstStep2Health, a school-based and home-based healthy lifestyle intervention, on improving parents’ moderate to vigorous physical activity (MVPA), diet quality (ie, F/V and fibre intake, skin carotenoids), screen time, proportion of overweight/obesity, BMI, % body fat, BP, nutrition and physical activity knowledge, self-efficacy, parental support, parenting style, feeding practices and home environment after adjusting for random cluster effects.
Methods
Conceptual framework
The intervention development was grounded in both the Actor-Partner Interdependence Model and the Social Cognitive Theory.17 18 The Actor-Partner Interdependence Model endorses the reciprocal influences between a parent and a child within the family context.19 20 Based on the Social Cognitive Theory, individuals’ behaviours can be modified by their knowledge, self-efficacy, social support and environment. Parents require sufficient knowledge and self-efficacy to effectively promote and support their children in adopting healthy lifestyles.21 Additionally, they need to employ effective parenting practices to integrate healthy foods and regular physical activity into their daily family routines.22 Building on these theories and evidence, our FirstStep2Health dyadic intervention targeted parents to enhance their knowledge, self-efficacy and support to improve their parenting style and feeding practices, fostering a nurturing and healthy home environment. To enhance the influences from children to parents, we implemented an innovative approach that connected school learning with home practices to further promote healthy eating and physical activity behaviours at home.
Overview of the FirstStep2Health study
This cluster randomised controlled trial was conducted with 10 Head Start daycare centres in the Midwestern USA from fall 2021 to spring 2023. A cluster randomised controlled trial was used to minimise contamination because children are clustered within daycare classrooms and centres. The intervention period ran from November to March each year. The intervention effects on improving children’s outcomes (primary aim of the trial) were published in a prior manuscript23 and this current manuscript focused on examining the effects on parents’ outcomes (secondary aim). The study statistician randomly assigned the 10 centres into intervention (five centres) or control (five centres) groups using computer-generated random numbers after participant enrolment and baseline data collection. The study was registered in ClinicalTrials.gov (NCT04164277).
We mainly used recruitment flyers to recruit participants. Classroom teachers helped to distribute the study recruitment flyers to families. Interested families were instructed to use the survey link or QR code on the flyer to complete the screening and enrolment survey and contact the study team with any questions. A face-to-face recruitment method was also used. The trained recruitment team went to each daycare centre to recruit families during daycare drop-off and pick-up times. A US$5 e-gift card was provided to each family who completed the screening and enrolment survey regardless of their participation status.
Head Start daycare centres serving children aged 3–5 years were eligible to participate in the study. Inclusion criteria for parents included: (1) being the primary adult caregiver (parent or legal guardian aged ≥18 years) of a Head Start child, (2) having at least weekly internet access to use Facebook or the study’s private website to participate in the intervention activities and (3) being able to understand and speak English. All eligible and consented children and parents from participating daycare centres were enrolled in the study.
Assuming an overall cluster effect intraclass correlation coefficient (ICC)=0.01 based on prior literature,24 no variations between clusters (ie, differences among clusters are minimal, as all clusters are urban Head Start daycare centres operated by the same organisation), and a significance level of 0.05, a sample size of six daycare centres, 24 classrooms and 130 participants would provide a power of 0.80 to detect an effect size of 0.50 in the trial’s primary outcome of children’s MVPA.23 The selected effect size of 0.50 was based on the results observed in our previous quasi-experimental feasibility study.25
Patient and public involvement
Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of our research.
Intervention
The 16-week, dyadic, FirstStep2Health intervention was tailored to economically marginalised families with young children.23 We actively engaged Head Start teachers and parents to develop culturally appropriate content for the intervention through focus groups and interviews.25 26 The intervention included: (1) a 16-week (total of 32 sessions) daycare-based child programme on healthy mindful eating and physical activity, tailored to young children’s developmental stage; (2) 32 child letters to parents to connect school learning with home practice; (3) a 16-week social media-based parent programme to assist parents in promoting healthy eating and physical activity at home; (4) three virtual group parent meetings at weeks 1, 8 and 16 via Zoom on topics related to healthy eating and physical activity. The meeting delivery format was changed from in-person to virtual Zoon sessions due to the COVID-19 pandemic and (5) weekly motivational messages (three times/week for 16 weeks) to increase parental motivation to build and sustain a healthy home environment. The daycare-based child programme was delivered in-person by trained interventionists to all children enrolled in the intervention daycare centres instead of pulling only the study children out. This approach was highly recommended by Head Start administrators and can help reduce potential feelings of insecurity and anxiety among children. Make-up sessions were offered to ensure the full dose of 32 sessions was achieved when intervention classes were closed due to the spread of COVID-19 cases. To promote mindful eating and regular physical activity at home, each intervention family received a healthy cookbook containing budget-friendly family recipes, adult and child MyPlate plates to guide appropriate portion size and a community resource booklet to help sustain healthy behavioural changes. Table 1 provides details of the intervention.
The dyadic FirstStep2Health intervention for parent-child dyads
The usual care control group did not receive any intervention during the study period. After postintervention data collection, a 2-week mini programme was offered. The 2-week mini programme included: (1) children participating in a 2-week daycare-based child programme in the daycare centers, (2) children completing two letters to parents to share what they had learnt from the 2-week child programme and (3) parents receiving a programme handbook containing all intervention instructions and flyers, as well as supplies including a healthy cookbook, two MyPlate plates and a community resource booklet.
Intervention implementation fidelity monitoring
To monitor the implementation fidelity of the child programme, trained independent process evaluators observed six randomly selected sessions (two sessions at weeks 1–4, two sessions at weeks 7–10 and two sessions at weeks 13–16) at each intervention classroom. Feedback and suggestions were provided to the interventionists for further improvement. A remediation plan with additional training would be implemented if the total score was <4 out of a 6-point range. The project manager conducted three parent meetings following developed scripts to ensure consistent delivery. The study process evaluators observed each parent meeting to evaluate the delivery fidelity in comparison with the developed scripts.
Data collection
We collected individual outcome data at baseline (before intervention began) and immediately postintervention. At each time point, parents received a survey link via text message or email to complete an online survey via Qualtrics. A face-to-face appointment was then scheduled to complete measurements of skin carotenoids, height, weight, % body fat and BP, as well as to distribute ActiGraph GT3X-plus accelerometers to assess physical activity. The face-to-face data collection appointment mainly occurred at a private daycare classroom within participating Head Start centres. However, we also offered home visiting data collection options for participants who were uncomfortable leaving home due to pandemic concerns. Some in-person data collection events were held at a local university conference room when a participating Head Start centre was closed due to the spread of COVID-19 cases. Trained and blinded from randomisation status, data collectors collected data from participants.
Measures
Parents completed reliable and valid instruments in an online survey measuring their demographics, F/V and fibre intake, screen time in hours per day, nutrition and physical activity knowledge, self-efficacy, support, parenting style, feeding practices and home eating and physical activity environment (table 2).
Study outcomes and measures
MVPA was measured using the ActiGraph GT3X-plus accelerometer with the wrist-worn approach.27 ActiLife software was used to initialise each ActiGraph and set it to begin data collection at 05:00 hours on the first day. We instructed parents to wear the ActiGraph on their dominant wrist (attached to an adjustable woven nylon wristband) from the time they got out of bed in the morning until going to sleep at night for seven consecutive days (excluding when bathing or swimming). An auto text message reminder was sent to parents every morning at 07:00 hours via our developed Health Insurance Portability and Accountability Act (HIPAA) compliant platform Twilio (www.twilio.com). Parents’ physical activity was reported as: sedentary 0–99, light 100–2019, moderate 2020–5998 and vigorous ≥5999 counts per minutes.28 Data were determined valid when wear time by the parents was at least 8 hours per day for at least 2 days.29 30
Skin carotenoids were measured by the pressure-mediated reflection spectroscopy with the Veggie Meter (Longevity Link Corporation). Skin carotenoids are being used as a non-invasive biological marker for F/V intake due to their consistent positive correlation with plasma/serum carotenoids in the adult population.31 Evidence has supported the sensitivity of using Veggie Meter to detect changes in skin carotenoids in experimental studies among ethnically/racially diverse populations.32
Height and weight were measured by the Shortboards Stadiometer to the nearest 0.1 cm and Seka model 874 scale to the nearest 0.01 kg, respectively. BMI was calculated using measured height and weight: weight (kg)/[height (m)]2. We used the BC-533 InnerScan Body Composition Monitor to measure parents’ % body fat. During the measurements, parents were asked to take off shoes, socks or bulky clothes. For measuring % body fat, each parent stepped onto the scale surface and aligned feet with the four electrodes.
Parents’ BP was measured by the Omron HEM-705-CP digital BP monitor following the protocol established by the American Heart Association.33 The day before data collection, parents were informed to not smoke, exercise or injest caffeine during the 30 min before their BP was taken. During BP measurement, parents were instructed to keep their left arm on the measurement table (right arm was used if the left arm had a surgery or injury), sit upright, back straight, with feet flat on the floor.
To increase measurement reliability and validity, two measurements were taken for height, weight, % body fat and BP. A third measurement would be taken, when the two measurements differed by ≥0.5 cm for height, ≥0.5 kg for weight, ≥1% for % body fat and ≥5 mm Hg for BP. To calculate the final measurement value, the two closest measurements were averaged.
Data analysis
IBM SPSS Statistics V.27 was used for all data analyses following the intention-to-treat principle to preserve the original randomisation. That is, data were analysed based on participants’ originally assigned group regardless of whether they received the intervention or not. Statistically significant level was set at 0.05. Means, SD, ranges, frequencies and percentages were calculated to describe study variables. To compare group differences in individual demographics while adjusting for cluster effects of daycare centres and classrooms, mixed-effect models were applied and identified significant demographics were adjusted in the further analyses. Mixed-effect models with restricted maximum likelihood estimation method and Satterthwaite approximations for df34 were applied to examine intervention effects. In the mixed-effect models, postintervention outcome was the dependent variable, fixed effect independent variable was group (intervention vs control) and covariate was baseline outcome and any identified significant demographics. Random cluster effects included both daycare centres and classrooms, as parents’ children were nested in daycare centres and classrooms. The regression coefficient B was estimated to indicate the differences in dependent variables between the intervention and control groups, while accounting for covariates and random cluster effects. Furthermore, when the dependent variable was the proportion of overweight/obesity, generalised linear mixed-effect model with logic link was performed to examine the group differences at postintervention after controlling for baseline weight status and cluster random effects. The advantage of using mixed-effect models is that correct inference can be estimated even when missing data exist.35 Effect sizes of Cohen’s d and OR were calculated to describe the intervention effects: small (d=0.2, OR=1.68), medium (d=0.5, OR=3.47) and large (d=0.8, OR=6.71).36 37
Results
Participants
The study’s flow diagram was published in a prior manuscript reporting the intervention effects on improving children’s outcomes.23 No daycare centre or classroom was lost to follow-up. The loss to follow-up rate among children was 10.5%, primarily due to children leaving Head Start daycare programme. A total of 95 parents (53 intervention and 42 control) participated in the study. Parents’ mean age was 30.42 years, and majority (91.6%) were female. About 8.4% were Hispanic and 36.8% were Black. Over half of the parents were single, and 66.3% of the families had an annual family income <US$20 000. Nearly half of the parents were unemployed, and 57.9% had an education level of high school or less. On average, each family had three children living in the household. As shown in table 3, parents’ demographics are very comparable between the intervention and control groups. No adverse events were observed during the study.
Demographic characteristics of parents
Intervention attendance
On average, children participated in 21 sessions of the daycare-based programme and completed 21 letters to parents, with a range from 0 to 31. About 77.4% of the children (n=41) attended 16 or more sessions and completed 16 or more letters. Parents, on average, attended one parent meeting and participated in 6 weeks of the social media-based programme. Seventeen (32.1%) parents attended all three parent meetings, and 34.0% participated in more than >8 weeks of the social media-based programme. The average daycare-based child programme attendance rate was 66.1%, parent meeting participation rate was 43.5% and social media-based programme participation rate was 38.6%.
Intervention effects
As presented in table 4, the intervention did not result in significant effects on improving parents’ MVPA (d=−0.36; 95% CI=−0.77, 0.05). The effects on improving their light physical activity were small and non-significant (d=0.23; 95% CI=−0.18, 0.63).
Description of parents’ outcomes over time
However, the intervention had significant medium to large effects on increasing parents’ nutrition knowledge (d=0.86; 95% CI=0.44, 1.29) and physical activity knowledge (d=0.51; 95% CI=0.10, 0.92), and decreasing their systolic BP (d=−1.11; 95% CI=−1.54, −0.67) and diastolic BP (d=−0.58; 95% CI=−0.99, −0.17). The significant effects on improving parents’ physical activity self-efficacy (d=0.47; 95% CI=0.06, 0.88), and decreasing their perceived parental weight (d=−0.46; 95% CI=−0.87, −0.05) and permissive parenting style (d=−0.49; 95% CI=−0.90, −0.08) were small to medium.
The non-significant effects on increasing nutrition self-efficacy (d=0.37; 95% CI=−0.04, 0.78), parental support on healthy eating (d=0.21; 95% CI=−0.19, 0.62) and physical activity (d=0.25; 95% CI=−0.16, 0.66) were small to medium. Additionally, small to medium non-significant effects were also observed on decreasing parental monitoring on child’s eating (d=−0.39; 95% CI=−0.80, 0.02) and authoritarian parenting style (d=−0.30; 95% CI=−0.70, 0.11). Although participants in both groups experienced increases in both BMI and % body fat, the increases in the intervention group were slightly smaller than those in the control group (BMI: d=−0.34; 95% CI=−0.74, 0.07; % body fat: d=−0.32; 95% CI=−0.73, 0.09).
Intervention effects after adjusting for cluster random effects
Table 5 demonstrates the results from mixed-effect models after adjusting for the cluster random effects of daycare centres and classrooms. In comparison with the control group, intervention parents engaged in more MVPA (B=0.49, p=0.874) and light physical activity (B=4.06, p=0.308) postintervention; however, these results were not statistically significant. The ICC for the random effects of daycare centres and classrooms was 0.06.
Intervention effects on parents’ outcomes after adjusting for random cluster effects of daycare centres and classrooms
Compared with the control group, parents in the intervention group had significantly higher nutrition knowledge (B=0.87, p=0.009), physical activity knowledge (B=0.95, p=0.049), nutrition self-efficacy (B=0.74, p=0.025) and physical activity self-efficacy (B=0.86, p=0.013) postintervention. In addition, fibre intake among intervention parents was significantly higher than control parents (B=2.99, p=0.049). At postintervention, parents in the intervention group had significantly lower % body fat (B=−2.56, p=0.005) and systolic BP (B=−10.98, p=0.005) than those in the control group. Although not statistically significant, the intervention showed positive effects on improving parents’ F/V intake, parental support on healthy eating and physical activity, home physical activity environment and authoritative parenting style. The ICCs ranged from 0 to 0.43 for these outcomes.
Discussion
The goal of this study was to examine the preliminary efficacy of the FirstStep2Health versus usual care control among parents in mitigating familial obesogenic risk factors (MVPA, dietary quality, parental knowledge, self-efficacy, support, parenting style, feeding practice and environment) and their anthropometric changes (BMI, % body fat and BP). Overall speaking, although the FirstStep2Health intervention failed to result in significant improvements in parents’ MVPA in this study, it significantly mitigated many familial obesogenic risk factors particularly with regard to parental knowledge and self-efficacy in promoting physical activity and healthy eating behaviours. These significant improvements in parental knowledge (nutrition and physical activity) and self-efficacy (nutrition) are congruent with previous studies.3 Although not statistically significant, the positive increases in parental support and home environment (d=0.19–0.25) among intervention parents are meaningful because building a supportive home environment that is more conducive for healthy eating and physical activity is an essential first step to prevent childhood obesity.38
For the accelerometer-measured MVPA and light physical activity, both intervention and control parents experienced deceases in the study. The most plausible explanation for the unsuccessful results may be that the global pandemic has caused low attendance of the intervention and significant decreases in physical activity and increases in sedentary activity.39 Although many strategies (eg, virtual intervention delivery format, make-up sessions) were applied to handle the challenges due to the COVID-19 pandemic, the overall intervention participation rates were still lower than what we had achieved in the feasibility study (39%–66% vs 77%–87%) conducted before the pandemic.25 However, in the previous feasibility study, we also provided a small incentive (US$5 per week) to compensate parents’ time and efforts in participating in the weekly social media-based programme. Given the hardships of poverty and the lack of spouse or partner support (62% were not married or partnered) among these vulnerable families, financial compensations may be necessary to overcome their unique challenges (eg, being a single parent, a tight family budget) for intervention participation,26 especially when facing worsening financial challenges within the context of a global pandemic.
Although parents in both groups experienced some increases in their BMI and % body fat during the study, intervention parents had a smaller increase than control parents (d=−0.34, −0.32). In addition, intervention parents’ % body fat was significantly lower than control parents at postintervention after controlling for the baseline values and random cluster effects. This result is noteworthy because % body fat is a stronger indicator of cardiovascular risks than BMI.40 More importantly, intervention parents had decreases in both systolic and diastolic BP, indicating a reduced risk for cardiovascular disease.41 These results are somewhat surprising given that the intervention was intended to actively engage parents to reduce obesogenic risks among children rather than parents. One potential explanation is that children play a significant role in influencing their parents’ food purchasing and consumption behaviours at home,42 and our innovative approach of enhancing the influences from children to parents via child letters was effective. However, due to the large proportion (39%) of parents not providing these outcome data, the results may be biased as parents who did not benefit from the intervention might have been more likely to skip data collection. Although further investigations with a larger sample size are needed, the study’s results show some favourable effects of a childhood healthy lifestyle intervention on parents’ cardiovascular risk factors including % body fat and BP.
Parental beliefs, attitudes and practices regarding child feeding within the context of their children’s obesity proneness improved among the intervention group. Most importantly, intervention parents’ feeding practices were less likely to be influenced by their own perceived weight status (d=−0.46), as compared with control parents. Although not statistically significant, the intervention group’s parental control practices and attitudes regarding child feeding (restriction, pressure to eat or monitoring) decreased (d=−0.05 to −0.39) over time. These positive results are encouraging as evidence indicates that parents’ use of controlling restrictive feeding practices could result in negative effects on children’s diet quality, and the detrimental effects were even stronger when fathers implemented the restrictive feeding practices than mothers.43 Thus, the father-mother differences in feeding practices should be carefully considered when engaging parents into improving diet quality among children.
The positive changes in intervention parents’ feeding practices are also reflected in their improvements in parenting style including increases in authoritative parenting and decreases in both authoritarian and permissive parenting. The significant decreases in using permissive parenting style (d=−0.49) among intervention parents are critical because permissive parenting was reported to contribute to children’s unhealthy lifestyles.44 Given that authoritative parenting is associated with healthy lifestyle behaviours and lower BMI in children,45 46 promoting authoritative parenting can be an effective approach to prevent childhood obesity, particularly for achieving long-term sustaining outcomes. Unfortunately, very few obesity prevention interventions have assessed the changes in parenting styles,3 9 limiting the comparison between this study’s results on parenting style and findings from previous literature. Regardless, the FirstStep2Health intervention’s effects on enhancing positive parenting can help children to have a healthier lifestyle and better mental health, which can subsequently lead to fewer behavioural problems.47
In terms of parents’ diet quality changes, intervention parents demonstrated positive but not significant changes in increasing F/V and fibre intakes (d=0.25 and 0.26, respectively). The positive effects on increasing parents’ F/V intake were smaller than the effects (d=0.40) observed in the previous feasibility study,25 which may be due to the adverse effects of pandemic and low intervention engagement. The decreases in skin carotenoids (d=−0.25) may have occurred because the control parents had a much lower level of skin carotenoids at baseline (236.83 vs 256.33), as compared with intervention parents. Another reason is the potential existing of unmeasured confounders, as skin carotenoids were reported to be associated with smoking and sun exposure.48
Limitations
This study has some limitations. First, this study had a relatively small sample size (n=95 Head Start parents) with 91.6% being female. Thus, generalising to the general public or male caregiving adults would be inadvisable. Moreover, families were recruited based on their acceptance of the study recruitment flyer and voluntary agreement, which may limit the generalisability of results to non-participating parents. Second, this study was conducted under a global COVID-19 pandemic context, the quarantine mandate and fear of virus transmission may contribute to the relatively low enrolment and participation rates in intervention sessions and in-person data collection for MVPA, height, weight and % body fat among parents. Moreover, the potential additional stress that parents had encountered during the pandemic may have lowered the intervention effects. Thus, caution is needed when interpreting the results, especially their generalisability to a different social context. Third, we could not possibly control for every potential sociodemographic and lifestyle factor and the self-reported nature and short form of some measurements leave the possibility of residual confounding. In addition, the study sample size was small due to the challenging recruitment under a global pandemic context, although we managed to oversample 10 daycare centres instead of the originally planned six, a small ICC of 0.01 was used in power analysis, the significant level was not adjusted for multiple comparisons when examining all outcomes and potential confounders were not controlled for during the analyses, which may threaten the statistical conclusion validity and increase the risk of false negatives. Finally, the study did not include a long-term follow-up evaluation, so future research should focus on assessing the long-term sustaining effects of the intervention.
Conclusions
The positive impacts of the dyadic FirstStep2Health intervention on parents are promising. In addition to the improvements in parental knowledge and self-efficacy, the intervention’s contributions to improving parenting style and feeding practices as well as their anthropometric and cardiovascular outcomes (slowed the decreases in BMI and % body fat, decreased systolic and diastolic BP) are vital given the detrimental impacts of obesity-related comorbidities (eg, hypertension, type 2 diabetes, cardiovascular diseases and cancers) in the USA. The dyadic intervention’s positive effects on improving both children and parents’ outcomes23 further imply the bidirectional and intergenerational relationships within a family system. Therefore, dyadic interventions targeting both children at school and parents at home may have created a virtuous cycle, enhancing parents’ supportive influences on children’s behaviours and children’s positive impact on parenting practices.20 Future endeavours should proactively engage parents in childhood obesity prevention efforts and evaluate the effects among both children and parents.
Data availability statement
Data are available on reasonable request. Study data can be requested from the corresponding author.
Ethics statements
Patient consent for publication
Ethics approval
The study was approved by the Michigan State University Biomedical and Health Institutional Review Board (approval number: STUDY00001629). Parents provided written consent for themselves and their children to participate in the study. In addition, children aged 5 years or older were required to provide verbal assent in order to be enrolled with their adult caregivers.
Acknowledgments
We would like to express our sincere gratitude to our community partner, Capital Area Community Services, for their invaluable support and assistance in completing this study. A special thanks to the graduate and undergraduate students from Michigan State University who contributed to the study. We also extend our heartfelt appreciation to the parents and children who participated in the study, for their willingness, interest and engagement
References
Footnotes
Contributors JL contributed to the study design and implementation, drafted the paper and served as the Principal Investigator. T-SAK contributed to data interpretation, drafted the paper and critically edited drafts. LBR and JMK contributed to the conceptualisation, edited drafts and served as co-investigators. NZ contributed to data analysis, edited drafts and was a co-investigator. YS developed the programme's private website and text messaging system, critically reviewed the paper and served as a consultant. All authors approved the final version of the manuscript. JL is responsible for the overall content as the guarantor.
Funding Research reported in this publication was supported by the National Institute of Nursing Research of the National Institutes of Health under Award Number R21NR017958.
Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.