Background: Diet and lifestyle-related illnesses like obesity and functional gastrointestinal disorders (FGIDs) are rapidly emerging health issues worldwide. Research has focused on addressing FGIDs via in-person cognitive-behavioral therapies and lifestyle modifications focusing on diet modulation and pharmaceutical intervention. However, there is a paucity of research reporting on the effectiveness of digital care based on genome SNP and gut microbiome markers to guide lifestyle and dietary modulations on FGID associated symptoms and on modeling diseased groups or outcomes based on a combination of these markers. Objective: This study sought to model subjects with FGID symptoms vs. those that do not present them, using demographic, genetic, and baseline microbiome data. Additionally, we aimed at modeling changes in FGID symptom severity of subjects at the time of achieving 5% or more of body weight loss in a digital therapeutics care program compared to baseline symptom severity. Methods: A group of 177 adults with 5% or more weight loss on the Digbi Health personalized digital care program was retrospectively surveyed about changes in the symptomatology of their FGIDs and other comorbidities. The FGID subgroup rated their symptom severity on a scale of 1 to 5 at the beginning of the program and after successfully achieving >5% body weight decrease. During the intervention, personalized coaching for lifestyle changes, including diet and exercise, was delivered by both human and digital coaching. The demographic, genomic, and baseline microbiome data of the subgroup of participants (n=104) who self-reported any of six FGIDs (IBS, diarrhea, constipation, bloating, gassiness, and cramping) were compared with those who did not report FGIDs (n=73) and used as variables for a logistic model. The sum of reductions in symptom severity and IBS, diarrhea, and constipation symptom severity reduction were analyzed using the same variables in linear regression models. Results: Gut microbiome taxa and demographics were the strongest predictors of FGID status. The digital therapeutics program implemented effectively reduced the summative severity of symptoms for 89.92% of users who reported FGIDs, with a highly significant reduction in severity (Wilcoxon signed-rank test, p=4.89e-17*). A mixture of genomic and microbiome predictors modeled the best reduction in summative FGID symptom severity and IBS symptom severity, whereas reduction in diarrhea symptom severity and constipation symptom severity were best modeled by microbiome predictors only. Conclusion: A digital therapeutics program, informed by genomic SNPs and baseline gut microbiome and their interaction with participant diet and lifestyle, can effectively reduce functional bowel disorder symptomatology. While further research is needed for validation, demographics, microbiome taxa, and genetic markers can effectively inform models aiming at classifying subjects with FGIDs vs. those that do not have FGIDs and models assessing the reduction in symptom severity experienced by FGID sufferers. The methods and models presented here can readily be implemented to study other comorbidities where genetics and gut microbiome play a central role in disease etiology.

Background The COVID-19 pandemic has highlighted the urgency of addressing an epidemic of obesity and associated inflammatory illnesses. Previous studies have demonstrated that interactions between single-nucleotide polymorphisms (SNPs) and lifestyle interventions such as food and exercise may vary metabolic outcomes, contributing to obesity. However, there is a paucity of research relating outcomes from digital therapeutics to the inclusion of genetic data in care interventions. Objective This study aims to describe and model the weight loss of participants enrolled in a precision digital weight loss program informed by the machine learning analysis of their data, including genomic data. It was hypothesized that weight loss models would exhibit a better fit when incorporating genomic data versus demographic and engagement variables alone. Methods A cohort of 393 participants enrolled in Digbi Health’s personalized digital care program for 120 days was analyzed retrospectively. The care protocol used participant data to inform precision coaching by mobile app and personal coach. Linear regression models were fit of weight loss (pounds lost and percentage lost) as a function of demographic and behavioral engagement variables. Genomic-enhanced models were built by adding 197 SNPs from participant genomic data as predictors and refitted using Lasso regression on SNPs for variable selection. Success or failure logistic regression models were also fit with and without genomic data. Results Overall, 72.0% (n=283) of the 393 participants in this cohort lost weight, whereas 17.3% (n=68) maintained stable weight. A total of 142 participants lost 5% bodyweight within 120 days. Models described the impact of demographic and clinical factors, behavioral engagement, and genomic risk on weight loss. Incorporating genomic predictors improved the mean squared error of weight loss models (pounds lost and percent) from 70 to 60 and 16 to 13, respectively. The logistic model improved the pseudo R2 value from 0.193 to 0.285. Gender, engagement, and specific SNPs were significantly associated with weight loss. SNPs within genes involved in metabolic pathways processing food and regulating fat storage were associated with weight loss in this cohort: rs17300539_G (insulin resistance and monounsaturated fat metabolism), rs2016520_C (BMI, waist circumference, and cholesterol metabolism), and rs4074995_A (calcium-potassium transport and serum calcium levels). The models described greater average weight loss for participants with more risk alleles. Notably, coaching for dietary modification was personalized to these genetic risks. Conclusions Including genomic information when modeling outcomes of a digital precision weight loss program greatly enhanced the model accuracy. Interpretable weight loss models indicated the efficacy of coaching informed by participants’ genomic risk, accompanied by active engagement of participants in their own success. Although large-scale validation is needed, our study preliminarily supports precision dietary interventions for weight loss using genetic risk, with digitally delivered recommendations alongside health coaching to improve intervention efficacy.

Diet and lifestyle-related illnesses including functional gastrointestinal disorders (FGIDs) and obesity are rapidly emerging health issues worldwide. Research has focused on addressing FGIDs via in-person cognitive-behavioral therapies, diet modulation and pharmaceutical intervention. Yet, there is paucity of research reporting on digital therapeutics care delivering weight loss and reduction of FGID symptom severity, and on modeling FGID status and symptom severity reduction including personalized genomic SNPs and gut microbiome signals. Our aim for this study was to assess how effective a digital therapeutics intervention personalized on genomic SNPs and gut microbiome signals was at reducing symptomatology of FGIDs on individuals that successfully lost body weight. We also aimed at modeling FGID status and FGID symptom severity reduction using demographics, genomic SNPs, and gut microbiome variables. This study sought to train a logistic regression model to differentiate the FGID status of subjects enrolled in a digital therapeutics care program using demographic, genetic, and baseline microbiome data. We also trained linear regression models to ascertain changes in FGID symptom severity of subjects at the time of achieving 5% or more of body weight loss compared to baseline. For this we utilized a cohort of 177 adults who reached 5% or more weight loss on the Digbi Health personalized digital care program, who were retrospectively surveyed about changes in symptom severity of their FGIDs and other comorbidities before and after the program. Gut microbiome taxa and demographics were the strongest predictors of FGID status. The digital therapeutics program implemented, reduced the summative severity of symptoms for 89.42% (93/104) of users who reported FGIDs. Reduction in summative FGID symptom severity and IBS symptom severity were best modeled by a mixture of genomic and microbiome predictors, whereas reduction in diarrhea and constipation symptom severity were best modeled by microbiome predictors only. This preliminary retrospective study generated diagnostic models for FGID status as well as therapeutic models for reduction of FGID symptom severity. Moreover, these therapeutic models generate testable hypotheses for associations of a number of biomarkers in the prognosis of FGIDs symptomatology.

BACKGROUND In this age of global COVID-19 pandemic, the urgency of addressing an epidemic of obesity and associated inflammatory illnesses has come to the fore. Studies have demonstrated that interactions between single nucleotide polymorphisms (SNPs) and lifestyle interventions like food and exercise may vary metabolic outcomes, contributing to obesity and therapeutic response. However, there is a paucity of research relating outcomes from digital therapeutics to inclusion of genetic data in care interventions. OBJECTIVE This study aims to describe and model weight loss of subjects enrolled in a precision digital weight loss program informed by machine learning analysis of subject data, including genomic. It was hypothesized that weight loss models would exhibit better fit when incorporating genomic data than utilizing demographic and engagement variables alone. METHODS A cohort of 393 participants enrolled in Digbi Health’s personalized digital care program for 120 days was analyzed retrospectively. Care protocol included the use of subject genomic and gut microbiome data informing precision coaching by mobile app and personal coach. Two linear regression models of weight loss in this cohort (pounds lost, percentage lost) as a function of demographic and behavioral engagement variables were fit. Genomic-enhanced models were built by adding 197 SNPs from subject genomic data as predictors, then refitting, employing Lasso regression on SNPs for variable selection. Success/failure logistic regression models were also fit, with and without genomic data. RESULTS 72% of subjects in this cohort lost weight, while 17% maintained stable weight. 142 subjects lost 5% within 120 days. Models describe the impact of demographic and clinical factors, behavioral engagement, and genomic risk on weight loss. The addition of genomic predictors improved the mean squared error of weight loss models (pounds lost and percent) from 70 to 60 and 16 to 13 respectively. The logistic model improved pseudo R2 from 0.193 to 0.285. Gender, engagement and specific SNPs were significantly associated with weight loss. SNPs within genes involved in metabolic pathways that process food and regulate storage of fat were associated with weight loss in this cohort. This included rs17300539_G (insulin resistance, monounsaturated fat metabolism), rs2016520_C (BMI, waist circumference, cholesterol metabolism), and rs4074995_A (calcium-potassium transport, serum calcium levels). Models described greater average weight loss for subjects having more of these risk alleles. Notably, coaching for dietary modification was personalized to these genetic risks. CONCLUSIONS Adding genomic information in modeling outcomes of a digital precision weight loss program greatly enhanced model accuracy. Interpretable weight loss models pointed to efficacy of coaching informed by subjects’ genomic risk, accompanied by active engagement of subjects in their own success. While large-scale validation is needed, our study preliminarily supports precision dietary interventions for weight loss utilizing genetic risk, with digitally delivered recommendations alongside health-coaching to improve intervention efficacy.