The Families Improving Health Together (FIT) Program: Initial evaluation of retention and research in a multispecialty clinic for children with obesity

Abstract Background Obesity affects ∼17% of US children, with parallel increases in multiple comorbidities, especially among African‐, Asian‐, Hispanic‐, and Native‐Americans. Barriers to patient retention in pediatric obesity programs include lack of centralized care, and frequent subspecialty MD visits which conflict with patient school attendance and parental work attendance as well as with support service utilization. Lack of integration of multispecialty clinical care with interdisciplinary research is a major barrier to fuller exploration of the treatment, prevention, and understanding of obesity in childhood. Objective To test the hypothesis, a novel multispecialty/interdisciplinary clinical and research infrastructure with strong emphasis on a primary obesity care physician for children with early‐onset (<9 years) obesity (Families Improving health Together [FIT]) could promote lower patient attrition (primary goal) and foster productive research in pediatric obesity (secondary goal). Results Data support the hypotheses. Over 15 months, FIT reported a >90% participant retention (p < 0.001 vs. expected rate based on other studies of similar programs). Though 90% of children had at least one adiposity‐related comorbidity and 70% had at least two, there was no need for additional subspecialist visits with cardiologists, endocrinologists, gastroenterologists, or molecular geneticists. Three abstracts were presented at national meetings, and two manuscripts were published all with junior faculty as primary authors. Conclusion This pilot study suggests that an integrated multispecialty/interdisciplinary approach to children with obesity improves patient retention and can be integrated successfully with research.


| INTRODUCTION
Care for obesity and its comorbidities currently accounts for over $200 billion per year in healthcare costs (21% of total US healthcare budget) in the United States. 1 In New York State (NYS), 16%-17% of toddlers and adolescents have overweight and 11% of adolescents and 15% of toddlers have obesity. 2 The likelihood of sustained medically significant nonsurgical weight loss in adults has remained about 12%-20%, [3][4][5] despite multiple new interventions. The duration of reduced 10% or greater body fatness (by body mass index [BMI] z-score) is significantly improved in prepubertal children (success rates up to ∼50%) compared to adults enrolled in structured weight loss programs. [6][7][8][9][10] Obesity represents a complex interaction of biological, social, and environmental problems that require integrated care and research beyond individual discipline boundaries or "silos." 11,12 These silos may limit the coordination of clinical care and the collaboration of researchers to evaluate obesity as a multisystem disease. 12,13 The National Institutes of Health (NIH) has recognized the need for combined clinical care and interdisciplinary translational research ranging from individual microsystems, including genetics, to environmental macrosystems. [14][15][16][17][18] There are a number of suggested approaches to better integrate multispecialty clinical care and interdisciplinary research for children with obesity by diminishing barriers to participant retention. From a patient standpoint, compartmentalized referrals for each comorbidity (frequent MD visits), even within the same program, have been associated with poor child and family retention in treatment clinics. [19][20][21][22][23] Subspecialty referrals increase time demands and costs (missed school days and work days) for affected children and their families and diminish time available for participation in support services (e.g., dietary and exercise programming). In addition, noncentralized care often results in poor communication among physicians managing obesity, specialists managing comorbidities, dietitians, exercise programs, and social workers. From a provider standpoint, the most frequently reported barriers have included lack of time (75%), lack of awareness of referral options (70%), lack of coordinated care (58%), and costs (71%-85%). 24 These data suggest rates of attrition might be reduced in a program based on a single primary care physician who coordinates all care to minimize referrals and maximize communication regarding the child with obesity. [25][26][27][28] Families Improving Health Together (FIT) was a new program that was operational at the Children's Hospital of New York at Columbia University Irving Medical Center from March 2017 to July 2018. FIT focused on centralizing care of patients with obesity to a single physician supported by an on-site clinical multispecialty and academically interdisciplinary team. The interdisciplinary approach was distinct from other institutional subspecialty-based programs in which comorbidities frequently required referral to other subspecialists. This is a pilot study of a novel means to increase retention and encourage multispecialty/interdisciplinary research in a program for children with obesity. The primary hypothesis was that centralization of care for pediatric obesity would reduce the need for referrals, diminish attrition, and increase interactions with the primary care obesity physician, dietitian, and exercise programming.
A secondary hypothesis was that an integrated research and clinical environment with junior and senior faculties as well as standardized enrollment forms, a mineable database, and presentation of research goals would breakdown academic silos and stimulate interdisciplinary research.

| Funding
The FIT program was implemented at the Children's Hospital of New York at Columbia University Irving Medical Center with funding from the NY State Empire Clinical Research Investigator Program. FIT was funded as a pilot study to determine whether a multispecialty, interdisciplinary approach would improve patient retention and research. These results would then be incorporated into the infrastructure of a larger, more sustained, pediatric obesity initiative.

| Initial FIT design
Prior to opening the FIT clinic, the following resources were developed: 1. An overview curriculum discussing clinical research design, ethics, and statistical analyses for young investigators.
2. An extensive screening questionnaire (see Supplement) for each age group (ages 2-4, 5-7, and 8-10 years). 3. A database that interacted with the electronic medical records to allow direct downloading of vital signs, laboratory work, and ongoing treatment recommendations, as well as the data from the screening questionnaire. 5. Seminar schedule to familiarize all healthcare providers with issues related to obesity within their specialties and progress reports regarding their research.

| FIT infrastructure
The FIT infrastructure consisted of a coordinator, physicians (senior and junior faculties and fellows), a registered dietitian, and an exercise physiologist. The coordinator was responsible for scheduling 358of patients, ascertaining that the prescreening questionnaire was completed before patients were seen (see below), and for arranging regular seminars for FIT physicians. The clinical team consisted of senior and junior faculty from the Divisions of Pediatric "Cardiology," "Endocrinology," "Gastroenterology, Hepatology, and Nutrition," and "Molecular Genetics," as well as a dietitian and an exercise physiologist. Primary FIT physicians were assistant professors, associate professors, or fellows, and were responsible for organizing patient care and presenting patients to the entire team as new data became available. This organizational format allowed the more senior advisory group to directly supervise fellows and, in general, to provide early career introductions to obesity care and research. The

| FIT screening
Children with early-onset obesity (BMI > 95%ile by age 9; ages 2-10 at enrollment) were referred by their primary care physicians.
Primary care physicians were informed about FIT by advertisements that publicized the FIT website. Prior to arrival in the FIT clinic, parents were contacted by the FIT coordinator (K.L.) and asked to complete an extensive questionnaire (available in English and Spanish) to assess medical and family history, dietary habits, exercise habits, sleep habits, and the environment in which the children live (see Supplement). Growth charts, laboratory records, and any previous consultation records were obtained from the referring pediatrician before any appointment was made for a new patient in the FIT program. Written informed consent and assent forms were offered in English or Spanish to participate in FIT with options for biobanking of blood and DNA samples from participants and to be contacted regarding other research opportunities. All protocols, consents, and assents were approved by the Human Research Protection Office/ Institutional Review Board.

| FIT initiation
Each patient was assigned a primary FIT pediatrician from the (total time about 90 min). Vital signs including waist circumference, body fat percentage by Bio-electrical Impedance Analysis (BIA), and blood pressure (repeated three times) were measured for every child.
Within 2 weeks of the initial visit, children were scheduled to return for fasting laboratory testing 29 (see Figure 1).

| FIT follow-up
Parents returned to discuss the results of the screening laboratory tests with their FIT physician 2 weeks after labs were drawn. Prior to this, the entire FIT team, including the cardiology, endocrinology, gastroenterology, molecular genetics, and nutrition subspecialists, met to discuss the patient and their lab results. Team recommendations were communicated to parents by the FIT physician. Follow-up appointments were then made to meet with the dietitian every 1-2 months and their FIT physician at 3-4 month intervals. Routine fasting laboratory testing was done every 6-12 months unless the severity of a detected abnormal result (e.g., elevated glycosylated hemoglobin or thyroid disease) indicated the need for more frequent sampling. This routine laboratory testing functioned as a measure of co-morbidity presence/risk and as biomarkers that could be incorporated into discussions with families regarding the patient's health status. Letters were sent to referring primary care physicians after all MD visits.

| FIT exercise (HOP-UP)
The FIT exercise program was named "FIT HOP-UP" by Dr. Garofano who directed it. The goal of the exercise program was to monitor and facilitate participants' exercise during the sessions and in the home and to optimize available exercise opportunities in their communities.
Exercise during sessions (total session length, including 30-45 min of exercise depending upon age, was ∼60 min) consisted of ageappropriate moderate to vigorous physical activity (MVPA) at ROSENBAUM ET AL.

| FIT database
The FIT program data system was created by the data management

| Outcome variables and statistical analyses
The primary outcome variable was patient retention rates over 6 months (defined as at least one follow-up visit with the dietitian and/or physician after the initial visit with both the physician and dietitian and return for fasting blood testing with discussion of results). Because of the limited time that the program was operative (total of 12 months), longer follow-up data were not available for most participants. Secondary outcome variables were the number of successful research collaborations involving multiple investigators and multiple disciplines (defined as publications, grants, or abstract presentations at national or international meetings) and the number of patients who required referrals to subspecialty clinics to see a clinician other than their primary care obesity physician. An exploratory research project was to establish a means to share relevant information with the Washington Heights/Inwood community which was by the number of presentations to the surrounding health care community regarding the management of children with obesity in a primary care setting.

| Statistics and calculations
Rates of retention, defined as at least one visit by the child to the FIT physician and/or dietitian after the initial assessment and laboratory testing visits, and referral from FIT were compared with data obtained from results of other pediatric obesity programs 19,[34][35][36][37] using binomial tests of proportions. 38 There are, to our knowledge, no data available regarding frequency of subspecialty referral within a pediatric weight management program. Studies of primary care physicians report that 7%-25% of children with obesity are referred to subspecialists (excluding dietitians and pediatric obesity programs) with most referrals to pediatric endocrinologists. 21 Participant characteristics are presented in Table 1. Socioeconomic data (annual household income) data were also requested and are schematized in Figure 3. Because of the low number of participants who self-identified as non-Hispanic and the high number who did not provide reliable data regarding family history of obesity and socioeconomic status (see Table 1 and Figure 3), these data were not analyzed as correlates of attrition or comorbidities.

| Comorbidities
As shown in Table 2, over 90% of participants had at least one adiposity-related co-morbidity and over 70% had at least two comorbidities consisting of systolic and/or diastolic hypertension, dyslipidemia, evidence of insulin resistance, vitamin D insufficiency or deficiency, or laboratory evidence of nonalcoholic fatty liver disease (NAFLD). Even within this population of children with obesity, adiposity-related co-morbidity markers were often significantly correlated with adiposity (BMI z-score, see Table 3).

| Participant retention and referral
As shown in Figure 4, over 90% of participants returned for at least two visits (defined as at least one visit with the primary FIT physician or registered dietitian after laboratory testing had been performed and discussed with the primary FIT physician) to the FIT program.
The group discussed all patients. Other than three patients referred to ENT for evaluation of possible sleep apnea, no patient was referred to any subspecialty clinic outside of FIT for adiposity-related comorbidities. Rates of retention for two visits or more (as opposed to at least four visits in FIT-one for the initial screen, one for fasting laboratory testing, and one to discuss laboratory results, and a subsequent visit) reported in other pediatric obesity programs 39 ranged from approximately 50%-60% (p < 0.001 vs. FIT) for usual care 40,41 to 74% (p = 0.0016 vs. FIT) for enhanced care consisting of 2 h of inprogram lifestyle intervention per week for 12 weeks. 34 As noted, there are, to our knowledge, no available data on referral rates of children with obesity to subspecialists within obesity programs and reports from primary care programs for referrals, besides to dietitians or pediatric obesity programs ranges from 7% (p < 0.05 vs. FIT) to 25% (p < 0.001 vs. FIT). 21,36 Outline of Exercise Program (HOP-UP) Overview: The goal of the exercise program was to monitor and faciliate volunteer exercise during the sessions and in the home and to optimize available exercise opportunities outside in their communities. Exercise during sessions (total session length including 30-45 minutes of exercise depending upon age was ~60 minutes) consisted predominantly of aerobic exercise (HipHop dancing, running, jumping, walking) at gradually increasing levels with support to engage in aeroic exercise within the community based on the HOP-UP questionnaire. Sessions were divided by age and moderate exercise tailored to individual age and capabilities also derived from the questionnaire. Initial Meeting: Meet and greet, provide program overview, distribute HOP-UP questionnaire to identify existing knowledge of physical activity and opportunities to be phyiscally active at home and in the community, and determine family availability for subsequent sessions. Second Meeting: Equipment review. Each child is given a locker, instructed in what clothing to wear, and given an activity monitor (step counting accelerometer) to wear (sneakers etc.,.). Each child in the older group is assigned an iPad (provided by the program) which remains locked up when the children are not there and contains numerous simple questionnaires to document weekly physical activity and suggest new types of exercise. Months 1-2: Two sessions and one phone call per week. Separate sessions for age 2-7 and 7 and older. Exercise instruction is conducted at each session along with some review of previous exercises and discussion of how they have been used. Phone calls are to provide positive reinforcement for progress in engaging in physical activity.

Months 3-4:
One session per week and one phone call every other week.

Months 5-6:
One session every other week with one phone call in the intervening week.

Follow up:
One phone call quarterly.

F I G U R E 2
Outline of exercise intervention. The goal of this intervention was to give participants the knowledge to engage regularly in physical activity and acquaint them with the opportunities to do so in their communities ROSENBAUM ET AL. -361

| Research
Research efforts were productive with successful establishment of the database integrated with the electronic medical record, three nationally presented abstracts [42][43][44] with young investigators as primary authors, two manuscripts published 45,46 with FIT investigators as senior authors, and one manuscript in preparation. 47  Age (years) -Mean (SD) 6.6 (2.4) 6.6 (2.0) 6 Data available on only one parent in 35% of participants.

ROSENBAUM ET AL.
T A B L E 2 Mean (SD) values related to co-morbidities and percentage classified as having adiposity-related co-morbidities Note: Statistically significant differences from males are highlighted in bold.

T A B L E 3
Correlations of adiposity-related co-morbidity assessments and adiposity measured by BMI % above the 95%ile Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase.

F I G U R E 4
Number of visits completed by subjects. The initial visits with the physician and dietitian were classified as visit 1. Subsequent visits were defined as meeting with the primary care obesity MD and/or with the dietitian after the initial meeting. A return appointment for fasting blood tests obtained by a phlebotomist was not counted as a clinical visit and subjects who did not return for fasting blood work were considered as having had only one visit. All subjects had the opportunity to complete at least two visits and only 9% of subjects (orange segment) failed to do so. Those enrolled earlier had the opportunity to complete more visits. All subjects were enrolled for a sufficient period of time to generate at least two visits ROSENBAUM ET AL. -363 Other investigators have also noted higher short-and long-term retention rates and/or other outcomes in interdisciplinary programs 56,57 supporting the idea the centralization of care to a single program or even single physician is beneficial. Burton et al. 55  Finally, it should also be emphasized that this was not a clinical trial and there was no assessment of intervention efficacy or program financial sustainability.
These issues should be addressable in longer-term studies of this model hopefully utilizing innovations in telemedicine and other community outreach programming to meet the American Academy of Pediatrics guidelines for management of obesity in childhood. 64 Spence et al. 65