Interventions Targeting Bottle and Formula Feeding in the Prevention and Treatment of Early Childhood Caries, Overweight and Obesity: An Integrative Review

Overweight, obesity and early childhood caries (ECC) are preventable conditions affecting infants and young children, with increased prevalence in those formula-fed. Previous research has focused on distinct outcomes for oral health and healthy weight gain. However, the aetiology may be linked through overlapping obesogenic and cariogenic feeding behaviours, such as increased sugar exposure through bottle propping and overfeeding. Best-practice bottle feeding and transition to cup use may concurrently reduce overweight, obesity and ECC. This integrative review aimed to identify interventions supporting best-practice formula feeding or bottle cessation and examine the intervention effects on feeding, oral health and weight outcomes. The reviewers searched nine databases and found 27 studies that met the predetermined inclusion criteria. Eighteen studies focused on populations vulnerable to ECC or unhealthy weight gain. All studies focused on carer education; however, only 10 studies utilised behaviour change techniques or theories addressing antecedents to obesogenic or cariogenic behaviours. The outcomes varied: 16 studies reported mixed outcomes, and eight reported worsened post-intervention outcomes. While some studies reported improvements, these were not maintained long-term. Many study designs were at risk of bias. Effective intervention strategies for preventing ECC and child obesity require the holistic use of interdisciplinary approaches, consumer co-design and the use of behavioural change theory.


Introduction
Overweight, obesity and early childhood caries (ECC) are preventable conditions affecting infants and young children. ECC are dental caries occurring in children aged under 6 years, defined as the presence of one or more primary teeth affected by decay, tooth loss or tooth fillings [1]. Globally, the age-standardised prevalence of untreated caries in primary (baby) teeth of children up to 14 years of age ranged from 4.9% (Australia) to 10.8% (the Philippines) in 2010 [2], with a worldwide prevalence of 7.8% and 126 million age-standardised cases in 2015 [3]. The health consequences of ECC include poor child growth from eating problems and poor nutrition, impaired speech development, and impaired sleep, play, learning, concentration, school performance and attendance due to caries-related pain [4][5][6]. Children who experience ECC are likely to be at increased risk of later dental problems [7].
Overweight and obesity in children under 5 years of age, defined respectively as two and three standard deviations above the World Health Organization weight-for-height growth standard median [8], affected an estimated 38.2 million children in 2019 [8]. Longitudinal data show that the trajectory of infant weight gain increases the risk of obesity in childhood and adulthood [9,10]. Obesity during childhood increases the risk of chronic disease, such as type 2 diabetes, cardiovascular disease and non-alcoholic fatty liver disease [11]; children with obesity are over five times more likely to have obesity into adulthood [12].
These health conditions may be linked due to overlapping obesogenic and cariogenic feeding behaviours [13]. Increased exposure to sugar increases the risk for both dental caries and excessive caloric intake through various practices: propped bottle feeding in bed for younger infants or bedtime bottle use for older children who can hold bottles [6,[14][15][16]; the use of sugar-sweetened beverages in bottles [14]; the addition of fermentable carbohydrates, such as sugar, syrup, honey or cereal, in bottles [16,17]; and frequent exposure to sugar, such as through snacking or drink sipping throughout the day [14,15]. Bottle use past the age of 12 months can entrench constant drink sipping throughout the day, particularly of sugar-sweetened beverages, which can also contribute to dental caries and obesity [15]. Research has found prolonged bottle use at 24 months of age associated with obesity at 5.5 years of age [18]; late bottle cessation after 18.8 months of age is associated with an increased risk of overweight and obesity at 3-5 years old [19]. Two meta-analyses of children aged up to 6 years found an increased risk of ECC for children who were above a healthy weight, although the results were inconsistent across the weight categories of overweightness, obesity and combined overweightness and obesity [13,20].
Preventative strategies for both ECC and excessive weight gain include breastfeeding until 6 months of age, avoiding added and free sugar, using responsive bottle feeding, avoiding infant overfeeding, using cups from 6 months of age and eliminating infant bottle use at one year of age [15,21]. A systematic review by Appleton and colleagues identified additional formula feeding practices to reduce the risk of infant overweight and obesity, including choosing infant formula with lower protein content, avoiding 'follow on' formulas marketed at infants aged 6 months and above, avoiding the addition of fermentable carbohydrates in bottles and using smaller infant bottles to avoid overfeeding [16].
Healthcare professionals working with infants and their families are well-placed to discuss infant feeding that promotes healthy practices, reduces the risk of overweight and obesity, and reduces cariogenic behaviours [1]. A review on primary preventative oral healthcare for young children or childbearing women and delivered by nurses or midwives found that 14 out of 21 trials reported improved the outcomes of dental caries prevalence, oral health and dietary behaviours and dental service use [22]. Similarly, a review of nondental health professionals providing preventative dental care found that more effective caries prevention and improved health and dietary behaviours were associated with longer intervention periods with education reinforcement, multiple avenues of verbal and written education and counselling, and comprehensive interventions with education, oral health toolkits and counselling [5]. An oral health model implemented in two Women, Infants and Children (WIC) program centres targeting low-income mothers with children aged under five in the USA demonstrated that allied health clinicians can expand their practices to oral health risk screening, assessment, and fluoride varnish application [23].
Interdisciplinary approaches by healthcare professionals in dental, medical, nursing and allied health settings that address formula feeding and infant bottle use may help to prevent both ECC and the risk of overweight and obesity. Previous research on this topic has existed in two distinct academic silos: that is, dental interventions report outcomes, such as bottle use, as an oral health behaviour but not an obesity risk behaviour and obesity interventions report outcomes, such as the type of fluid consumed, as a dietary but not a oral health risk behaviour. To our knowledge, this is the first integrative review conducted by an interdisciplinary research team to address the dental and nutritional approaches to infant bottle and formula feeding on dental and obesity outcomes. This integrative review aimed to identify interventions, trials and programs undertaken to support best-practice formula feeding or bottle cessation in infants and children and to examine their effectiveness in formula feeding practice, bottle cessation, oral health and/or child weight outcomes.

Research Questions
To capture the breadth of the research across disciplines, this review addressed the questions: 1.
What interventions, trials or programs have been undertaken to support best-practice formula feeding or bottle cessation in infants and children, focusing on either oral health or weight-related outcomes? 2.
What are the impacts of these interventions on formula feeding practice, bottle cessation, oral health and/or child weight outcomes?
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist was used to structure the presentation of this manuscript (Supplementary File S1).

Eligibility Criteria
The Population/Problem, Interest and Context framework used to structure the inclusion and exclusion criteria of the search strategy [24,25] is attached as Supplementary File S2. This review included studies focused on interventions aimed at parents or carers of infants and young children to improve formula, bottle or caries-preventing practices that encourage bottle cessation and cup transition and that measured health or behaviour outcomes relating to feeding practices, parent or carer knowledge, and infant anthropometry.

Information Sources
The database searches were conducted in CINAHL (via EBSCO), MEDLINE (via Ovid), EMBASE (via Ovid), Global Health (via Ovid), Maternity and Infant Care Database, Scopus, ProQuest, PubMed and Web of Science. The searches were conducted in July to August 2020, with weekly checking for relevant updates until June 2021. Hand searching the reference lists and forward citations of the included studies was undertaken to further identify relevant studies [26].

Search Strategy
The Population/Problem, Interest and Context framework was used to devise the search strategy. Eligible studies addressed: (1) interventions on education about infant formula or bottle use, (2) infant formula or bottle use, including bottle or formula cessation, and (3) interventions about infant formula or bottle use and dental caries. An example of the search strategy structured for the CINAHL database, using the Population/Problem, Interest and Context framework, is provided in Supplementary File S3. The search strategies for all the databases are available in Supplementary File S4.

Study Selection
All references were downloaded to Endnote X9. The references were screened by abstract and title for relevance, then assessed for full-text eligibility by author HC. For studies where eligibility was unclear-such as intervention content-the authors discussed their eligibility by using their clinical knowledge and expertise on infant feeding and oral health.
The study selection process is illustrated in a PRISMA diagram ( Figure 1).

Data Extraction and Synthesis
Data on the first author; publication year; country; study aim, design and period or duration; study setting; participants; intervention and comparator group conditions; and study findings were extracted and tabulated. The data was extracted by author HC. The statistical data, where available, were presented as the mean ± standard deviation or median (interquartile range).

Quality Assessment
A quality appraisal of the included studies was undertaken by authors HC and CR using the Mixed Methods Appraisal Tool (MMAT) [27]. The MMAT is an appropriate evaluation tool that accounts for the diverse study types included and rates each article on the specific criteria relevant for its study type. Figure 1 shows the PRISMA study flow diagram. A total of 12,147 references were identified through database searching, and 53 references were identified through hand searching. After removal of the duplicates, 8377 references were screened by the abstract and title for relevance, with 8137 references excluded. Two hundred and forty references were assessed for full-text eligibility. A total of 209 references did not meet the inclusion criteria; of these, 32 references were interventions that were terminated or had minimal formula feeding or bottle cessation support in the wider context of the study intervention. Finally, 31 references were included for analysis, reporting 27 studies, programs, trials or interventions. The study findings with outcomes on bottle and cup use, caries prevalence and caries-related dietary behaviours are presented in Table 1.

Results
The summary details of the 32 references that were excluded and which addressed formula feeding, bottle cessation or cup use briefly in the wider context of the study intervention or experienced early study termination are available in Supplementary File S5.

Included Studies
Twelve studies were undertaken in the USA, 4 studies in the UK and 4 studies in Canada. The remaining 11 studies were undertaken in Australia, Germany, Israel, the Netherlands, Syria and Thailand. Thirteen studies utilised a randomised controlled trial (RCT) or cluster-RCT design, five were community-wide programs with quasi-experimental designs and the remaining nine studies were quasi-experimental pre-post-trials or nonrandomised controlled trials. Four studies were pilot studies that eventuated into two clinical trials or established evidence for a larger-scale government-run program.
To determine program effectiveness of parents' oral health knowledge and behaviours for their young children. Syria.

Quasi
To investigate the impact of an integrated oral health promotion intervention, within a national immunisation programme, on tooth-brushing and bottle-feeding termination practices.   UK.
To determine the effect of dental health education on caries incidence in infants, through regular home visits by trained dental health educators over a period of 3 years.
Quasi-experimental trial with cohort design of interventions of varying intensity; dates NR.
Mothers with children born between Jan and September 1995, residing in a deprived area with high caries prevalence.   Canada.
To evaluate the effectiveness of the dental hygiene-coordinated prenatal nutrition program, delivered by community-based nutrition educators, on dental health and child feeding habits; child oral health status; and early childhood obesity.
Cross-sectional longitudinal evaluation of community program; launched mid−1996.    Israel.

Carers of infants born in
To measure the effect of a community health education program on reported infants' bottle-feeding practices and toothbrushing behaviour, with or without distribution of toothpaste and toothbrushes.
Mother and child health centres, providing services to 6-12 months infants.
Stratified by religion profile: secular-moderately religious mixed or predominantly Orthodox. USA.
To assess effect of policy, systems and environmental change strategies to promote responsive bottle feeding on RWG risk.
Mothers with new-born infants in USA WIC program.
IG: Policy, systems and environmental change of WIC program. Retooling of infant feeding assessment to be inclusive of responsive bottle feeding; development of assessment tools, counselling probes resources; responsive bottle-feeding online education and text message support; rebranding of infant feeding helpline.
CG: Usual care, including timely and tailored breastfeeding support; breastfeeding support resources (helpline, online education, responsive text message support).
The duration of interventions in the primary healthcare setting ranged from oneoff education sessions [28,[30][31][32]45] or a resource handout [44,45,51] to long-term care embedded in routine well-child visits, health visits or vaccination clinics, with the longest lasting until children were aged 3 years in two studies [41,42]. The community program intervention durations ranged from one year [49] to ongoing, with the Healthy Smile, Happy Child project initiated in 1999 and continuing at time of writing [35,60].
The behaviour change counselling described in six studies was motivational interviewing, goal-setting or action planning. In three studies, motivational interviewing comprised two telephone calls in addition to outpatient counselling [52], embedding motivational interviewing into the usual health visit processes [48], and four sessions of counselling from pregnancy to 18 months of age [38][39][40]. Goal-setting occurred as part of motivational interviewing [48] or educational classes [47] or through the use of self-directed worksheets [42]. Three studies used action planning to initiate behaviour change goals or to identify and resolve barriers to change [38][39][40]48,50].

Intervention Design and Stakeholder Engagement
Five studies described the design of their interventions [28,29,47,50,56], involving stakeholder engagement for program development. Four studies used focus groups and interviews with interventionists and carers in resource development [61,62], the planning and acceptability of intervention messages [63,64] and determining the acceptability of intervention delivery [64].
There were varying levels of community engagement across the studies, such as communitybased interventions or community health workers as interventionists. This included community championships to enable education in early childhood caries prevention [35][36][37]49], media, advertising and campaign promotion in the community [34,36,37,46,56], and building capacity in the existing childhood and family community programs and services to deliver intervention activities [34,35,49]. The Contra Caries program utilised community-based education classes with community health educators [47].

Weight or Anthropometry
Of the six studies that reported on weight or anthropometry outcomes, two studies reported an improvement, with significantly fewer children with a BMI above the 85th percentile in one intervention cohort [34] and a decreased risk of rapid weight gain for infants in an intervention trial arm [33]. The remaining studies reported no significant effect on their anthropometric parameters [28,40,50] or worsened effects, with intervention children in two studies reporting significantly greater weight gains than children in the control groups [32,40].

Dental Caries Prevalence
The intervention impact on dental caries tended towards improvements or no significant difference across five of the 11 studies. Six studies reported a decreased prevalence of ECC [34,36,37,39,[41][42][43]. However, limitations existed in two studies, where a decreased ECC prevalence was found post-intervention at 2 years of age but not at follow-up at 5 years of age [38], and the discontinuation of program counselling and community-based education activities after funding cessation resulted in nonsignificant changes in ECC prevalence in seven American First Nations communities after four years [36]. The evaluation of one community program that only had one cohort for long-term examination found significant differences in ECC prevalence at 4 years of age but not at 2 and 3 years of age [34], which is reflective of the developmental stages presence of tooth eruption and the natural slow progressing nature of dental caries.
Five studies found no significant differences in dental caries prevalence [31,35,49,53,55]. Notably, one program evaluation across four Canadian First Nation communities found a significantly increased prevalence of ECC and severe ECC-where caries patterns were atypical, progressive, acute or rampant [6]-in rural and remote communities compared to an urban community [35].

Carer Knowledge and Awareness
In nine studies that reported on carer knowledge and awareness outcomes generally reported improved understanding of bottle feeding [51], ECC [56], bottle cessation [53], oral health knowledge (including topics on bottle use or cariogenic drinks) [30,31,34,35,47], effect of cariogenic drinks on dental health [47,49], and recall of interventionist education [44]; however, it was not significantly improved in one study [58].

Bottle and Cup Use
The outcomes on bottle and cup use were mixed. In ten studies reporting on bottle cessation, seven studies reported an improvement with increased bottle cessation [42,43,45,46,52,56] or earlier cessation [54], while three studies reported no differences in prevalence of bottle cessation or age of cessation [29,34,53]. Two studies reported an earlier start to cup use by 2 to 3 months compared to their comparative groups [34,54]. In nine studies that reported on bottle use or bottle feeding, five studies found a decreased use of bottles in prevalence and frequency of use [29,46,48,54,56], while four studies reported no differences [28,33,35,58]. In three studies that reported on the frequency or prevalence of cup use, two studies found no differences [28,34], and one study found increased cup use [35]. Lawrence and colleagues found lower rates of carer ever using bottles for feeding but no difference in the rates of breastfeeding and no difference in the prevalence of continued bottle use at 2-5 years of age [34]. Ventura and colleagues found that, although exclusive breastfeeding decreased, mixed formula feeding and breastfeeding decreased, and exclusive formula feeding increased from birth to six months; there was no interaction between responsive bottle-feeding intervention strategies and time, indicating that the intervention strategies did not inadvertently promote greater levels of bottle feeding [33]. Bottle use during sleep is a risk factor for dental caries. In four studies with outcomes on bedtime or sleep time use of bottles and cups, one study found an improvement, with a decreased use of non-water cups at bedtime but no difference in non-water bottles at bedtime or bottle cessation at bedtime [42], while three studies found no difference in bottle use during sleep [49,55,57].
Increased exposure to potentially cariogenic drinks through increased bottle or cup uses outside of meal sessions is a risk factor for dental caries. Of two studies, which did not report on which types of drinks were contained in bottles, one study found no difference in bottle use during meals and between meals [57], and the other study found increased bottle use by intervention children outside of meals but decreased bottle use to soothe crying [34].

Cariogenic Dietary Behaviours
The increased frequency of exposure to cariogenic foods and beverages increases dental caries risk. Cariogenic dietary behaviours, external to bottle or cup use, include the frequency of snacking, types of beverages consumed, and the addition of cariogenic foods into infant milk bottles. Across 16 studies, the outcomes on cariogenic behaviours tended towards no significant effects. Twelve studies found no significant differences in the amount or frequency of intake of infant formula [32], milk [29,54], juice [29,31,49,54] and cariogenic or discretionary foods or beverages [29,31,34,40,[47][48][49]55,58]; the total fluid intake from bottles and cups [28]; the addition of cariogenic foods to bottles [48]; and the use of sugar-or honey-dipped baby pacifiers [34]. Five studies reported positive outcomes with decreased cariogenic beverage intakes in cups or at daytime [43,58], more children limiting their intake of sweet beverages [47] or foods [41], a decreased addition of sugar or sweeteners to infant foods or bottles [34,58], and an increased use of nonsweetened beverages (cow's milk and formula) with a decreased use of sugar-sweetened beverages (condensed milk) in one cohort [34]. Two studies found worsened outcomes, with increased snacking between meals [42] or the increased use of bottles with added sugar for feeding [57].

Obesogenic Dietary Behaviours
Bottle feeding can be associated with pressuring feeding behaviours, such as encouraging infants to feed until the bottle is empty, which increases the risk of overfeeding, instead of feeding in response to infant satiety cues. Ventura and colleagues found that responsive feeding styles, pressuring feeding styles or the encouragement of bottle emptying were not impacted by strategic changes to promote responsive bottle feeding in six WIC clinics [33].

Healthcare Professional Practice
Three studies reported changes in healthcare professional practices. Hamilton and colleagues found that more mothers recalled health visitor talking about bottle transition to cup use and limiting sugary food and beverages at the 8-month well-child visit by 39% and 29%, respectively [44]. Strippel and colleagues' structured oral health education intervention significantly increased the discussion of oral health prevention topics addressed by clinicians during routine paediatric examinations; however, this constituted half of the 15 topics outlined in the intervention protocols being addressed [58]. The evaluation of the Bottle it up!-take a cup campaign found that more child health staff discussed transitioning from bottle to cup use (from 15% to 27%), but fewer child health staff warned against incorrect bottle use (75% to 24%) [56].

Critical Appraisal
A critical appraisal with the MMAT is summarised in Supplementary File S6. Of 13 studies (16 references) with an RCT or cluster-RCT design, almost all (11 out of 13) had unclear risks of bias on adherence to the assigned intervention, with adherence not reported. Six studies were at a high risk of bias from the inadequate description of participant randomisation. Five studies (six references) were at a high risk of bias with the groups not being comparable at the baseline: differences in the risk of overweight or obesity [28], cultural background [43], socioeconomic status [45], maternal age [45], infant feeding intentions [54] and registration for private health insurance [30,31]. Only six studies retained ≥80% of participants at the intervention end or follow-up and were at a low risk of bias for incomplete outcome data. Most studies had outcome assessors blinded to the participant conditions, with only four studies at a high risk of bias from unblinded assessors.
Of 14 studies (15 references) with a quasi-experimental non-randomised study design, almost all (13 out of 14) used appropriate measurements for measuring intervention exposure and participant outcomes. Eleven studies (12 references) involved participants who were representative of the target population, while the remaining three studies provided minimal information about parent demographics. Six studies were at a high risk of bias for incomplete outcome data, with ≤80% retention of participants at the intervention end or follow-up. Nine studies (10 references) were at a high risk of bias, as confounders were not accounted for in the designs and statistical analyses. There was a mixed risk of bias of adherence to the intervention administered.

Discussion
This integrative review synthesised 27 studies that investigated formula feeding or bottle cessation in infants and young children, and their effects on anthropometry, caries prevalence and dietary behaviours. This review assessed the effectiveness to date of interventions with a dual focus on oral health and child weight outcomes. A range of intervention strategies were used, primarily focused on education and resource distribution. Education for carers addressed various topics on oral health, feeding and dietary behaviours, dental care attendance and tooth brushing. The resources used to facilitate health behaviours included information handouts for carers and interventionists, children's drinking cups and oral health kits with toothbrushes and toothpaste. Intervention effectiveness was mixed: most studies reported mixed or non-statistically significant outcomes, and eight studies demonstrated worsened post-intervention outcomes. Notably, over half of all studies were targeted at infants and young children with risk factors predisposing them to ECC and excessive weight, including financial or social disadvantages, cultural factors and inequity related to First Nations backgrounds.
This literature review has several strengths. First, it integrates research from multidisciplinary research streams that are usually separate but that have parallel goals. The findings and recommendations are relevant to strengthen practices and develop effective interventions across disciplines. Clinical practices used frequently in certain disciplines-for example, the distribution of child drinking cups to support bottle cessation-can improve intervention designs. Second, it demonstrates the importance of interdisciplinary practice, particularly where interventions are focused on vulnerable populations at increased risk of dental caries, overweight and obesity in early childhood. Third, this review adopted a comprehensive search strategy identifying over 12,000 references, although it is possible that some may have been missed, especially older publications and smaller-scale interventions in local areas not included in peer-reviewed publications.
The limitations of this review include the inability to draw definitive conclusions or conduct meta-analyses due to the breadth or the diversity of study designs and reported outcomes. The study quality varied, with 21 studies displaying a high risk of bias in at least one study dimension, most frequently due to incomplete outcome data (seven RCT/cluster-RCT studies and six non-randomised experimental trials) and in confounders not being accounted for in the design and analysis (nine experimental trials). Furthermore, the impact of these interventions was mixed. Moreover, evaluation with the MMAT indicates limitations in study designs, which may contribute to the misestimation of differences between the intervention and comparator outcomes. Only two studies reported improvements in anthropometry, with fewer children who were overweight or at risk of being overweight in one of two participant cohorts [34] and a decreased risk of rapid weight gain for infants in an intervention arm [33]. Of 11 studies with caries outcomes, six reported improvements; however, of these studies, two experienced no long-term effects after funding cessation or trial completion [36,38], and four reported a high risk of bias for incomplete outcome data [34,39,41,43]. Carer knowledge and awareness of dental caries and feeding behaviours were reported as the most consistent improvements in nine studies. Across 16 studies on cup and bottle use, 11 studies found increased cup use, reduced bottle use and earlier bottle cessation or start of cup use; however, nine of these studies reported a high risk of bias, most commonly in comparable baseline data, complete outcome data and confounders in their designs [29,33,34,43,45,46,52,54,56]. Few clinically significant changes were found for bottle use during sleep or bottle use outside of meal sessions as a potential contributor for cariogenic exposures. Across 16 studies with outcomes on cariogenic dietary behaviours, only five studies had positive outcomes with a decreased exposure to cariogenic foods and beverages but also reported a high risk of bias across at least one MMAT domain [34,41,43,47,58]. Whilst the breadth of the outcome measures did not enable a meta-analysis, the range of the study findings indicates strengths present in both disciplines that should be used to inform future research and intervention designs. Utilising or establishing core outcomes in infant feeding, dietary intake or oral health interventions [65,66] can support the standardised reporting and comparison of meaningful effects.
Health-related behaviour changes are difficult. Behaviour changes should not be dependent on individual-level actions-such as expecting information and knowledge to change established behaviours-without also understanding the underlying factors [67]. Interventions informed by behaviour change theory and that address multifaceted factors underpinning behaviours may support long-lasting changes [68]. Although many studies focused on parent or carer education and/or resource distribution [28][29][30][31][32][33][34]41,[43][44][45][46]51,[53][54][55]57,58], only six studies utilised motivational interviewing, action planning or goal setting in addition to education [38][39][40]42,47,48,50,52]. Similarly, only six studies included behaviour changes or educational theory [30][31][32]47,49,50,61]. Antecedents of obesogenic and cariogenic bottle, beverage and formula-feeding behaviours relate to knowledge gaps and cultural preferences, including child soothing or settling [69,70], increasing weight gain from perceived poor appetite [71,72], the preference for children with larger body sizes [73,74], and misconceptions on the cariogenicity of drinks [71,75,76]. Barriers to non-cariogenic drink consumption, such as water, by infants and young children may include a child's dislike of water, a child's preference for sweet cariogenic drinks, a carer's concerns about the safety of tap water, a carer's belief that drinking water shows poverty and the inability to purchase drinks, a carer's belief that milk is a meal instead of a drink, and norms that do not support drinking tap water [75][76][77]. Without addressing these factors, education focusing on how to undertake best-practice behaviours may be insufficient to promote improved oral health and feeding behaviours.
To support behaviour changes, motivational interviewing as a client-centred approach can address underlying behaviours and develop parent-directed goal setting. Three studies in this review utilised motivational interviewing as part of carer education [38][39][40]48,52] and found some significant improvements in bottle use and ECC prevalence. A metaanalysis by Borrelli and colleagues found that motivational interviewing targeting parents or parent-child dyads in health interventions improved children's oral health hygiene, physical activity, screen time use and diet [78]. In an oral health context, motivational interviewing is typically delivered in clinical settings; however, this is often too late-as children present for clinical care after experiencing caries-associated pain-and is not feasible as a population-wide strategy, as it is time-and labour-intensive. Further, although prevention behaviour change interventions have included patient-focused dietary and oral hygiene counselling delivered alongside operative clinical interventions in clinical settings, these interventions are time-and workforce-intensive, expensive to deliver and, without regular and repeated exposure, have shown inconsistent results on sustainably improved dental caries outcomes [79,80]. This provides the impetus for collaborative and interdisciplinary approaches for disease prevention prior to children presenting for clinical treatment.
The maintenance of these interventions and long-term follow-up are essential. A longitudinal study design in future interventions should be considered, as dental caries is a progressive disease. This was exemplified in two studies: one cohort of an First Nations Canadian community program that found significant difference in ECC prevalence at 4 years of age but not at 2 and 3 years of age [34] and a RCT with First Nations Australian children that found a significant decrease in ECC prevalence at 2 years but not at 5 years of age [38][39][40]. Furthermore, this integrative review highlights the importance of behaviour change theory underpinning long-term intervention designs, as well as the importance of interdisciplinary approaches alongside consumer involvement when developing holistic long-term health education interventions.
Consumer involvement is essential in developing appropriate messages and strategies acceptable to target communities. Five studies [28,29,47,50,56] involved user engagement as part of the intervention design of resources [61,62], intervention messages [63,64] and intervention delivery [64]. Equally essential is community ownership and participation. First Nations community members and health workers in Australia, Canada and the USA were engaged in building the organisational capacity, delivering interventions, acting as community champions, and implementing local solutions in four oral health promotion programs [34][35][36][37][38][39][40]. Likewise, in-group community health workers across three studies supported program delivery to culturally and linguistically diverse communities in Canada and the USA [46][47][48]. Community engagement does not guarantee successful outcomes, with one study concluding that a culturally sensitive health promotion intervention did not warrant service-based implementation [40]; however, this remains important to the study design and may contribute to how Healthy Smile, Happy Child remains an ongoing community-led program in Manitoba, Canada [81]. Future interventions undertaken in populations vulnerable to ECC and child overweight and obesity should involve user codesign of interventions, particularly where cultural appropriateness is the key in designing intervention messages and supporting behaviour changes [82].
Interdisciplinary approaches to the prevention of ECC and child overweight and obesity by addressing best-practice formula feeding and bottle use can strengthen and focus preventative care. In the authors' local health area, an Early Childhood Oral Health Program integrates oral healthcare into general health interventions by child health professionals [83,84], and the surgical treatment of ECC in public hospitals requires attendance with an oral health therapist and dietitian in ECC prevention education. A midwiferyinitiated oral health service with antenatal dental treatment in the Greater Western Sydney region of Australia improved maternal oral health knowledge, oral hygiene and health and the uptake of dental services, where the process evaluation reported positive experiences by midwives, dental health professionals and mothers [85][86][87]; further, it has since developed state government prenatal oral health resources, been adopted into a policy in the state of Victoria, and been integrated into the national body of midwifery's continuing education program [88,89]. Similarly, the Healthy Tums Healthy Gums program, delivered by social workers, oral health staff and dietitians to vulnerable families, improved oral health and childhood nutrition knowledge, including cup use from 6 months onwards, the cessation of bottle use by 12 months of age, and identifying cariogenic and non-cariogenic foods and drinks [90]. An integrated obesity and ECC prevention approach to promoting best-practice formula feeding and bottle use, as exhibited in emerging research [48,91,92], is a promising novel approach that addresses the risk factors identified by child health professionals as contributing to obesogenic formula-feeding behaviours [93].

Conclusions
This integrative review of 27 studies combined research from disciplines that share similar goals regarding infant formula and bottle use, with implications for long-term metabolic and oral health outcomes. The intervention strategies in primary healthcare, community settings and a combination of both ranged from one-off education sessions or resource handouts to long-term care embedded into the usual care practices. While intervention effectiveness was mixed and most studies reported mixed or non-statistically significant outcomes, a range of intervention strategies was demonstrated. This included education, behaviour change counselling, resource distribution and stakeholder engagement.
The findings and recommendations of this integrative review are relevant to strengthen practices by emphasising the need for collaborative interdisciplinary approaches that incorporate dental and nutrition messages to prevent ECC and child overweight and obesity. Specific disciplinary strategies, such as targeted resource use for supporting behaviour changes, should be used to develop effective interventions across disciplines. This review emphasised the need to use behavioural change theory, stakeholder involvement and co-design in intervention development in order to support vulnerable populations at combined increased risks of dental caries, overweight and obesity in early childhood.  Data Availability Statement: Data sharing is not applicable to this article, as no new data were created or analysed in this study.