Feasibility, Acceptability, and Short-term Behavioral Impact of the MySmileBuddy Intervention for Early Childhood Caries
Objectives. To evaluate acceptability, feasibility, and short-term behavioral impact of an early childhood caries (ECC) intervention. Methods. Predominantly low-income Hispanic parent/child (2–6 years) dyads attending a busy pediatric dental clinic in New York City completed a single administration of the iPad-based technology-assisted education, goal-setting, and behavior change MySmileBuddy program. Self-reported behavior change was assessed via telephone survey one month post-intervention. Results. Of 113 parent/child dyads approached, 108 (95.6%) participated and all completed MySmileBuddy in its entirety. Over 96% (n = 76) of 79 parents reached for follow-up recalled MySmileBuddy; 63.3% (n = 50) recalled their diet-and/or oral hygiene-related behavioral goal; and 79.7% (n = 79) reported taking action to initiate behavior change. Conclusions. Findings suggest that MySmileBuddy was feasibly implemented in a busy clinic, acceptable to this high-risk population, and effectively promoted preliminary ECC-related behavior changes. Larger, long-term studies are warranted to further investigate the impact of the MySmileBuddy program.
Early childhood caries, oral health, diet, feasibility, acceptability, pediatric dentistry
Early childhood caries (ECC) is a highly prevalent disease with significant health consequences for children and their families. As a chronic infectious disease of the oral cavity, ECC causes tooth decay in children under six years old and can lead to pain, serious infections, difficulty chewing and speaking, and diminished quality of life.1–5 Low-income and minority children are disproportionately affected, experiencing both higher rates and greater severity of this largely preventable disease.6–14 Despite national public health initiatives to combat caries,15 ECC remains the most common chronic [End Page 59] disease of young U.S. children,16 affecting nearly one-quarter (23%) of 2–5 year olds.17 Due to its significant impact and high prevalence, ECC is one of the most serious and costly health conditions affecting children today.18
Traditional treatment via surgical dental repair may be effective in restoring physical function of the tooth. However, it fails to address the underlying etiology and natural history of the disease, leading to high recurrence rates post-treatment. It is increasingly recognized that disease management strategies addressing the primary mediators of ECC etiology (i.e., diet-and fluoride-related behaviors) hold promise to both improve health and reduce treatment costs.19–21 Despite research that suggests parents prefer preventive behavioral strategies (e.g., tooth brushing) that are child-focused, promote healthy habits, and employ home care techniques,22,23 implementation strategies for behavioral management remain largely unstudied.8,24,25 There has been scant research on interventions that promote suppression of cariogenic behavior, particularly among high-risk Hispanic immigrant populations. Even fewer studies have evaluated the feasibility and acceptability of such interventions.
To address this void, a multidisciplinary team of researchers conducted an ancillary analysis of data from the Diet and Early Childhood Caries (DECC) study. The DECC study sought to validate the technology-assisted ECC risk reduction program, MySmileBuddy. The MySmileBuddy program, described by C. L. Custodio-Lumsden, et al. (2016), capitalizes on the recognized benefits of health information technology (Health IT) and is designed to circumvent literacy and language barriers by using a highly visual, bilingual (English and Spanish) iPad-based software suite.26–28 The My SmileBuddy technology facilitates ECC risk assessment and guides users to provide targeted counseling and behavior change recommendations based on individual markers of risk.26,29,30 By presenting science-based, culturally-specific education and videos, MySmileBuddy teaches parents about how tooth decay develops and how they might decrease their child's individual risks for caries.26,29,30
Sitting side-by-side with a parent, the individual administering the MySmileBuddy technology (a health professional and/or lay health worker) navigates five assessment modules in the iPad app—diet, feeding practices, thoughts and feelings, fluoride, and family history—to evaluate ECC risk.26,29,30 Each module delivers a sequence of targeted questions in a highly visual, interactive, and structured manner. MySmileBuddy then calculates a risk score based on prior literature on cariogenicity and clinical expertise.26,29,30 The risk score is revealed upon completion of the electronic assessment modules and serves as a cue to action to reduce risk through salutary behavior change. MySmileBuddy then guides parents to formulate an individualized action plan in partnership with the health worker with whom they are working.
Feasibility and acceptability studies, common in the fields of nutrition and health promotion, are largely lacking in the oral health literature. However, such studies are essential to the successful dissemination of effective health interventions. Thus, the purpose of this ancillary analysis of DECC study data is to evaluate the acceptability, feasibility, and short-term impact of a single session of the MySmileBuddy program in a busy dental clinic. Findings will be used to inform MySmileBuddy implementation in an upcoming clinical trial. [End Page 60]
The DECC study was conducted in a predominantly Spanish-speaking, low-income, immigrant population in New York City.26,31 Demographic data describing the sample, previously published, revealed that children were 4.12 years of age on average, with a slight majority of females (56.5%).26 Most parents were mothers (88.9%), Spanish-speaking (72.2%), and foreign-born (75.9%). The majority reported receiving Supplemental Nutrition Assistance Program (SNAP) benefits and all but one of the children were Medicaid beneficiaries (98.2%).
The DECC study protocol was approved by the Institutional Review Boards at Teachers College Columbia University and Columbia University Medical Center and written consent was obtained from all participants. This ancillary study utilized data collected during implementation of the DECC study between August 2012 and February 2013.
The study sample was drawn from individuals presenting to the Columbia University Pediatric Dental Clinic for routine dental examinations, comprising both initial and recall examinations. Recruitment was limited to non-special needs children 2–6 years of age (the lower-bound age [=2 years] at which full dentition is typically achieved and the upper limit [= 6 years], by definition, for early childhood caries1). Recruitment of parents was limited to Spanish-and/or English-speaking primary caregivers over the age of 18 years, competent to consent. When parents presented with more than one eligible child, the youngest was selected for participation. Parents were recruited in the clinic waiting and examination rooms by study investigators, who subsequently administered the intervention in examination rooms and an adjacent conference room. A transit card (valued at $10) was offered in appreciation of study participation.
The intervention in the DECC study consisted of a single administration of the MySmileBuddy technology (approximately 15–20 minutes) by one of four bilingual investigators. Participants completed all study activities (including MySmileBuddy videos, app modules, and surveys) in their preferred language (English or Spanish). Investigators were trained to follow a specific protocol outlined in a data collection guide created for the study. Investigators additionally received training on the navigation and proper administration of the MySmileBuddy technology, as well as a basic introduction to the etiology of ECC and its relationship to diet-and fluoride-related behaviors.
Approximately one month post-intervention, parents were telephoned to complete a 5–10 minute follow-up survey. Multiple attempts were made, as needed, and voice messages left when possible. Investigators followed a structured telephone script to ensure consistency in survey delivery. Parents were asked (yes/no) if they recalled setting a behavioral goal via MySmileBuddy to reduce ECC risk. If "yes," they were asked to state their chosen goal and to describe any steps they had taken to achieve their goal. If "no," parents were reminded of their goal and asked if they had made any changes to their (or their child's) behavior to achieve the stated goal. In addition, all parents were asked if they had made any other changes to their child's diet and/or oral hygiene behaviors as a result of the information they received during participation in the study.
Descriptive statistics were used to detail characteristics of the study sample and to evaluate behavioral outcomes one month post-intervention. All categorical variables [End Page 61] were assigned numeric codes to aid in statistical analysis using SPSS software (Version 21.0, 2012, IBM Corp. Armonk, NY).
A total of 113 child/parent dyads were approached for study participation, of which 108 (95.6%) agreed and were subsequently enrolled (Figure 1). The five parents who declined participation stated time constraints as the primary reason for declination. All participating parents completed administration of the MySmileBuddy technology in its entirety during, or immediately following, their child's dental examination.
Attempts were made to contact all participating parents for the one-month follow-up survey; in the end, 79 (73.1%) were reached and completed the telephone survey an average of 41 (SD = 7.66) days post-intervention (Figure 1). Of those lost to follow-up, seven (6.5%) had incorrect or non-working telephone numbers, despite collection of
[End Page 62]
both primary and secondary (when available) contact numbers at time of enrollment. Additionally, five parents (4.6%) declined participation in the follow-up survey, and another 17 (15.7%) were unable to be reached (no response after multiple call attempts). Overall attrition was 26.9% for the follow-up survey. Comparison of demographic data (i.e., race/ethnicity, immigration status, language preference, mother's educational achievement, and WIC/SNAP participation) via chi-square analysis did not reveal any statistically significant differences between those lost to follow-up and those completing the one-month follow-up survey (data not shown).
Of the 79 parents who completed the follow-up survey, 96.2% (n = 76), recalled completing MySmileBuddy (Table 1). Most (63.3%) recalled setting a behavioral goal through the MySmileBuddy software and were able to recall their goals (60.8%). Over two-thirds (68.4%) reported taking action to achieve their MySmileBuddy goals. Examples of reported actions included, "… buy Cheetos® [only] once per month and start buying yogurt," "Water is added to juice … avoid buying sweets," "Replace juice for water." (Table 2) Approximately half (53.2%) reported having initiated additional behavioral changes in their child's diet or oral health behaviors, including nine who reported not taking steps to achieve their original chosen MySmileBuddy behavioral goals. In total, 63 parents (79.7%) reported engaging in some kind of action to change behavior post-intervention.
Evincing both acceptability and feasibility, nearly all parents (96%) presented with the opportunity to join the DECC study agreed to participate and also completed the intervention session in its entirety. The high participation rate likely speaks to the widespread interest of parents to learn more about optimizing the oral health of their young children. Other health-related studies conducted within the same geographic area and target population also had high participation rates, likely related to parental interest.32,33 In addition to the data collected via the MySmileBuddy technology and the one-month follow-up survey, the DECC study investigators recorded noteworthy [End Page 63]
comments from parents. Comments recorded included those regarding parental interest in the study topic, reported history of oral health problems, and general comments that may have affected caries risk or study outcomes. Over three quarters of the relevant comments recorded (those regarding study procedures were excluded) expressed high parental interest in oral health or diet education, and/or a desire to prevent oral health problems in their children due to significant personal (or child) history of poor oral health. Consistent with previous literature suggesting that parents prefer preventive behavioral strategies (e.g., tooth brushing) that promote healthy habits and home care techniques over traditional dental treatment for caries,22,23 parents in this descriptive study appeared eager to learn about salutary diet-and oral health-related behaviors. Additionally, administration of the MySmileBuddy technology was found to be feasible within the setting of a busy urban dental clinic. All parents completed one full session of the MySmileBuddy program either within the time allotted for their child's dental appointment or immediately following completion of the appointment. Thus, the 15–20 minute intervention required little, if any, additional time for administration beyond the dental examination. This suggests MySmileBuddy may be a feasible addition to initial or recall dental appointments in the clinic setting.
Data from the one-month follow-up survey are promising, revealing that nearly all parents reached for follow-up recalled administration of the MySmileBuddy technology. The majority of parents also remembered, and could accurately state, the behavioral [End Page 64] goal they set. This may be in part due to the fact that MySmileBuddy was very well received by parents, likely a result of its visual appeal and interactive design. The Institute of Medicine recognized the potential of Health IT—technologies to store, share, and analyze health information—to address health literacy barriers, improve quality of care, and reduce health disparities.27,28 By utilizing mobile Health IT, MySmileBuddy is able to communicate ECC counseling messages in a highly visual, interactive way with minimal reliance on literacy skills. The MySmileBuddy technology was also designed to be culturally, linguistically, and literacy appropriate for the DECC study target population. It is colorful, interactive, and incorporates images of children, caregivers, and food/dental products familiar to this low-income, Hispanic, urban population. By utilizing a mobile Health IT software platform, these features could be easily modified for future application in a variety of other settings and populations.
Additionally, MySmileBuddy provides recommendations for behavior change goals based on individual risk and self-reported behaviors. The provision of such targeted messages and features likely further enhanced parental recollection of MySmileBuddy goals. The large majority of parents reached for follow-up in this study reported that they were actively working towards achieving their MySmileBuddy goal or an additional ECC-related behavior goal. These encouraging findings may also be reflective of My SmileBuddy's grounding in theoretical models of behavior change and its inclusion of various determinants of behavior that inform the individualized guidance it presents.21 However, a future study examining the ability of MySmileBuddy to exert influence on determinants of behavior change is needed.
The sizeable attrition rate at one-month follow-up (26.9%) speaks to the challenge of working with a poor and low-income, immigrant population. Gaps in patient/clinic communication are frequent. Current telephone contact information is difficult to maintain. Many of the families in the New York City communities from which this study sample was drawn (Washington Heights and Inwood) return to their native countries for extended periods of time during summer months and holidays. The study sample reflected the population of the geographic area from which it was drawn, where nearly half (48%) of the residents are immigrants, two thirds of whom are from the Dominican Republic, and the median household income is significantly less than New York City as a whole.31 Families temporarily relocate, move, change cell phone carriers, or use intermittently activated phones. Other research with the same target population as the DECC study have had success with contacting participants using text messages, which may have been a more successful approach to maintain contact and minimize attrition than using telephone calls alone.32,33 Though analysis of demographic characteristics revealed that those parents who completed the one-month follow-up survey did not significantly differ from those who were lost-to-follow-up, it is recognized that those parents who were not reached for follow-up may be precisely the ones that need the intervention reinforcement the most.
The limitations of this descriptive analysis of DECC study data include those related to the parent project. Namely, the study recruited a small (n = 108) demographically homogenous sample, so generalizability of findings may be limited. The DECC study also recruited from within a dental clinic, so parents may have already placed a high value on oral health and may have received behavior change guidance from dental providers [End Page 65] at previous visits. It is not possible in the present analysis to distinguish between the influence of MySmileBuddy versus that of the dental providers' routine counseling. Additionally, the MySmileBuddy technology was only administered to parents once and no other follow-up beyond the telephone survey was conducted. Thus, the impact of the intervention must be viewed within this context. More intense follow-up and reinforcement of MySmileBuddy goals would likely have resulted in more robust findings related to behavior change. Lastly, because clinical outcomes were not assessed post-intervention, it is not possible to determine whether MySmileBuddy exerted an impact on clinical markers of disease progression in this study.
While these findings are encouraging, larger and longer-term studies are necessary to determine the efficacy and effectiveness of MySmileBuddy to induce positive behavior change and ultimately reduce ECC experience. Repeated MySmileBuddy contacts would permit establishment of rapport with parents and allow continued and updated tailored behavior change support. Further investigation into the application of the MySmileBuddy technology as a useful platform for education, goal setting, and positive behavior change is warranted. The follow-up survey data support the potential utility of MySmileBuddy as a valuable behavior change tool.
Disclosure of Funding
There is no funding or support for the research described in the manuscript entitled, "The Diet and Early Childhood Caries (DECC) Study: Feasibility, Acceptability, and Short-term Impact of a Novel ECC Intervention."
Drs. LUMSDEN and EDELSTEIN are affiliated with Columbia University College of Dental Medicine, Section of Population Oral Health. Drs. WOLF, CONTENTO, BASCH, ZYBERT, and KOCH are affiliated with Teachers College Columbia University, Department of Health and Behavior Studies.