WO2024084459A1 - A system and method for predicting resting metabolic rate - Google Patents

Abstract

A system, method, and non-transitory computer-readable storage medium designed to predict an individual's Resting Metabolic Rate (RMR) and make wellness recommendations based thereon. The system comprises a memory and at least one processor, receiving patient information, including age, sex, Fat-Free Mass (FFM), and Fat Mass (FM). The predictive algorithm calculates the patient's predicted RMR. In Model 1, RMR calculation depends on age, FFM, and FM. Model 2 includes FFM, FM, and FFM*FM. The system can also generate wellness recommendations to adjust caloric intake, exercise plans, or drug prescriptions based on the patient's predicted RMR or through comparison of the patient's predicted RMR with the patient's actual RMR.

Description

A SYSTEM AND METHOD FOR PREDICTING RESTING METABOLIC RATE

STATEMENT OF PRIORITY

[0001] This application claims priority to U.S. Provisional App. No. 63/418,304 for a method for predicting resting metabolic rate based on body composition.

BACKGROUND

[0002] According to the World Health Organization (WHO), the global prevalence of overweight and obesity has nearly tripled over the last five decades, which is a significant public health concern. Body weight reflects the balance between energy intake and energy expenditure. Assessment of resting metabolic rate (RMR), the mam component (50-70%) of total daily energy expenditure (TDEE), is used to determine individual energy requirements.

[0003] Assessing an individual’s RMR has many practical applications. For example, it can be used for calculating a person’s caloric needs, designing a personalized weight management, nutrition, exercise or weight loss plan, and making clinical assessments of a patient to provide medical recommendations.

[0004] Indirect calorimetry is the gold standard to measure RMR but is rarely used in clinical settings due to its high cost Several prediction methods have been developed to predict RMR: Harris-Benedict, Food and Agricultural Organization/WHO/United National University, Mifflin-St. Jeor, and Owen et al., have been developed as methods to assess RMR in clinical practice. Most of these methods estimate RMR based on sex and body weight, with the Harris- Benedict and Mifflin-St. Jeor methods also including height and age.

[0005] Notably, body composition has a strong influence on energy expenditure, with a 4.5 kcal/day contribution per kg of fat, compared to a 13 kcal/day contribution (~x4) per kg of muscle. Thus, using weight as a predictor, rather than body composition components, leads to a greater range of variance when estimating RMR and may provide a reasonable fit at the group level, but with a relatively large variance among individuals.

[0006] A significant issue is that while some studies report a trend to underestimation of RMR, there may also be problems with overestimation, depending on the characteristics of the study population. For instance, in one study conducted among 125 women, the methods resulted in an underestimate, although they yielded overestimates in another three studies with an African American population or Hispanic Women.

[0007] The Cunningham RMR prediction method relies on fat- free mass (FFM) because of the strong correlation (r = 0.7) between FFM and RMR. The method was developed in 1980 and is based on the data of 223 subjects from various studies published by Harris and Benedict in 1919. The prediction ability of this method has been found to be better for athletes than for the general population specifically among male as compared to female athletes. To make the system more generally applicable, some studies have proposed models that integrate body composition measurements, but again these have primarily been based on small numbers of participants from specific populations.

[0008] Therefore, despite the wide use of the common methods, there remains a need to establish a novel prediction method for RMR with a better fit at the individual level. The current study provides an evaluation of the common prediction methods on a large and diverse dataset and presents a novel prediction model that incorporates body composition parameters.

SUMMARY

[0009] Current methods for predicting resting metabolic rate (RMR) were validated in a relatively small sample with high individual variance. The present disclosure determined the accuracy of five common RMR methods and proposed a novel prediction method, including body composition. A total of 3001 participants (41 ± 13 years; BMI 28.5 ± 5.5 kg/m2; 48% males) from nutrition clinics in Israel were evaluated by indirect calorimetry to assess RMR. Dual-energy X-ray absorptiometry were used to evaluate fat mass (FM) and free-fat mass (FFM). Accuracy and mean bias were compared between the actual or measured RMR and the prediction methods. A random training set (75%, n = 2251) and a validation set (25%, n = 750) were used to develop a new prediction model. All the prediction methods underestimated RMR. The Cunningham method obtained the largest mean deviation (-16.6%; 95% level of agreement (LOA) 1.9, -35.1), followed by the Owen (-15.4%; 95% LOA 4.2, -22.6), Mifflin-St. Jeor (-12.6; 95% LOA 5.8, -26.5), Harris-Benedict (-8.2; 95% LOA 11.1, -27.7), and the WHO/FAO/UAU (-2.1; 95% LOA 22.3, -26.5) methods. [0010] The presently disclosed model includes sex, age, FM, and FFM and successfully predicted 73.5% of the explained variation, with a bias of 0.7% (95% LOA -18.6, 19.7). The present disclosure demonstrates a large discrepancy between the common prediction methods and measured RMR and presents a novel method for predicting RMR which includes both FM and FFM and presents relevant applications of said method.

[0011] The presently disclosed method can be used for predicting a person’s RMR and their caloric needs, designing a personalized weight management, nutrition, exercise or weight loss plan, and making clinical assessments of a patient for providing medical recommendations.

[0012] Based on the foregoing and continuing description, the subject invention in its various embodiments may comprise one or more of the following features in any non-mutually- exclusive combination:

[0013] A system comprises a memory and at least one processor to execute computerexecutable instructions to receive patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM);

[0014] A system comprises a memory and at least one processor to execute computerexecutable instructions to execute a predictive algorithm to generate the patient’s predicted Resting Metabolic Rate (RMR);

[0015] A system comprises a memory and at least one processor to execute computerexecutable instructions to generate at least one wellness recommendation based on the patient’s predicted RMR;

[0016] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein, the predictive algorithm is selected from Model 1 or Model 2;

[0017] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein, the Model 1 predictive algorithm for males is calculated as RMR (kcal/24 h) = 775.8 - (age 5) + (FFM 20.5) + (FM* 7.7);

[0018] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the Model 1 predictive algorithm for females is calculated as RMR (kcal/24 h) = 709 - (age* 5) + (FFM 20.5) + (FM 7.7); [0019] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the Model 2 predictive algorithm for males is calculated as RMR (kcal/24 h) = 891.7 - (age 5) + (FFM 18.5) + (FM 3.5) + (FFM* FM* 0.07);

[0020] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the Model 2 predictive algorithm for females is calculated as RMR (kcal/24 h) = 824 - (age* 5) + (FFM 20.5) + (FM 7.7) + (FFM* FM* 0.07);

[0021] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the predictive algorithm is based on Model 1 ;

[0022] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the predictive algorithm is based on Model 2;

[0023] A system comprises a memory and at least one processor to execute computerexecutable instructions comprising instructions to receive the patient’s actual RMR;

[0024] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein when the patient’s actual RMR is greater than the patient’s predicted RMR, generating at least one wellness recommendation to decrease the patient’s actual RMR;

[0025] A system comprises a memory and at least one processor to execute computerexecutable instructions comprising receiving the patient’s actual RMR;

[0026] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein when the patient’s actual RMR is less than the patient’s predicted RMR, generating at least one wellness recommendation to increase the patient’s actual RMR;

[0027] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein the at least one wellness recommendation is selected from the group consisting of modifying the patient’s calorie intake, modifying the patient’s exercise plan, and modifying the patient’s drug prescription;

[0028] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein modifying the patient’s calorie intake recommends the patient increase daily calorie intake when actual RMR is higher than predicted RMR; [0029] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein modifying the patient’s calorie intake recommends the patient decrease daily calorie intake when actual RMR is lower than predicted RMR;

[0030] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein modifying the patient’s exercise plan recommends the patient increase the number of calories burned per day through exercise when the patient’s actual RMR is lower than the patient’s predicted RMR;

[0031 ] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein modifying the patient’s exercise plan recommends the patient decrease the number of calories burned per day through exercise when the patient’s actual RMR is higher than the patient’ s predicted RMR;

[0032] A system comprises a memory and at least one processor to execute computerexecutable instructions wherein modifying the patient’s drug prescription recommends the patient increase the number of calories burned per day through exercise when the patient’s actual RMR is lower than the patient’s predicted RMR;

[0033] A method, comprising receiving, by at least one processor, patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM);

[0034] A method, comprising executing, by the at least one processor, a predictive algorithm based on Model 1 or Model 2 to generate the patient’s predicted Resting Metabolic Rate (RMR);

[0035] A method wherein the predictive algorithm is based on Model 1;

[0036] A method wherein the predictive algorithm is based on Model 2;

[0037] A method comprising receiving, by the at least one processor, the patient’s actual RMR;

[0038] A method wherein, when the patient’s actual RMR is greater than the patient’s predicted RMR, generating, by the at least one processor, at least one wellness recommendation to decrease the patient’s actual RMR;

[0039] A method comprising receiving, by the at least one processor, the patient’s actual RMR; [0040] A method wherein when the patient’s actual RMR is less than the patient’s predicted RMR, generating, by the at least one processor, at least one wellness recommendation to increase the patient’s actual RMR;

[0041] A method wherein the at least one wellness recommendation is selected from the group consisting of modifying the patient’s calorie intake, modifying the patient’s exercise plan, and modifying the patient’s drug prescription;

[0042] A non-transitory computer-readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, the operations comprising receiving patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM);

[0043] A non-transitory computer- readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, the operations comprising executing a predictive algorithm based on Model 1 or Model 2 to generate the patient’s predicted Resting Metabolic Rate (RMR);

[0044] A non-transitory computer-readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, wherein the predictive algorithm is based on Model 1 ;

[0045] A non-transitory computer- readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, wherein the predictive algorithm is based on Model 2;

[0046] A non-transitory computer-readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, the operations comprising generating at least one wellness recommendation based on the patient’s predicted RMR.

DESCRIPTION OF THE FIGURES

[0001] Figure 1. Study flow chart. Participants were excluded from the study if they were younger than 20 years old (n = 21), lactating (n = 7), maternal women (n = 3), users of steroidal drugs (n = 18), or had undergone a surgical procedure that could affect body composition [e.g., amputated organs (n = 4), breast augmentation or reduction (n = 9), and liposuction surgeries (n = 3)].

[0047] Figures 2(a) - 2(j). Bland-Altman analysis presenting the bias and the 95% level of agreement between each prediction method and the measured RMR. The x-axis represents the mean of the measured and estimated RMR, and the y-axis expresses the difference in percentage between the RMR predicted using each of the methods and the RMR measured as follows: (pRMR-mRMR/mRMR)/100). Lin’s concordance correlation was used to determine the R² and the linear method between each common method and the measured RMR. Figs. 2(a), 2(b) Cunningham [1, 2]; Figs. 2(c), 2(d) Harris-Benedict [3]; Figs. 2(e), 2(f) Food and Agricultural Organization/WHO/United National University [4]; Figs. 2(g), 2(h) Mifflin-St. Jeor [5]; Figs 2(i), 2(j) Owen et al. methods [6,7],

[0048] Figures 3(a) - 3(d). Bland-Altman analysis of the proposed new models applied to the validation set (n = 750). The x-axis represents the mean of the measured and estimated RMR, and the y-axis expresses the difference in percentage between the RMR predicted using each of the methods and the RMR measured as follows: (pRMR - mRMR/mRMR)/100). Lin’s concordance correlation was used to determine the R² and the linear method between each estimated method and the measured RMR. (a,b) model 1 , includes sex, age, FM, and FFM. (c,d) model 2, includes sex, age, FM, FFM, and the interaction between FM and FFM (FM*FFM).

[0049] Figure 4. A block diagram of a system for predicting a patient’s RMR, the system comprising a processor and a memory on which the predictive algorithm and other decision making logic, user interface or application is stored.

[0050] Figure 5. An illustrative example method for generating a predictive RMR according to an example of the instant disclosure.

[0051] Table 1. Continuous variables are presented as mean ± SD, while categorical and dichotomous variables are expressed as prevalence. Independent samples, Student’s t-test, and Pearson’s Chi-squared test were employed to compare the results between females and males. Categories for BMI classification include underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5-24.9 kg/m2), overweight (BMI 25-29.9 kg/m2), obesity class 1 (BMI 30-34.9 kg/m2), obesity class 2 (BMI 35-39.9 kg/m2), and obesity class 3 (BMI > 40 kg/m2). The key variables in the analysis are FFM (fat- free mass), FM (fat mass), and RMR (resting metabolic rate).

[0052] Table 2. This table presents two new models developed to predict RMR, based on the training set (n = 2251, 75%). Adjusted R² was 74.5 % for model 1, and 74.6% for model 2 calculated by linear regression based on stepwise elimination. FFM, fat-free mass; FM, fat mass; RMR, resting metabolic rate.

[0053] Table 3. Cross-validation of a new general population resting metabolic rate prediction method based on body composition study characteristics. This table presents study characteristics between the training set and validation set. Continuous variables percentage as mean ± SD, and prevalence for categorical and dichotomic variables. *For Training set vs. validation set comparison independent samples, Student's t-test or Person’s Chi-squared test were used, respectively. BMI, body mass index; FFM, fat-free mass; FM, fat mass; RMR, resting metabolic rate.

DETAILED DESCRIPTION

[0002] In this single-center cross-sectional observational study, 3001 participants attending a private nutrition clinic in the center of Israel were enrolled between October 2015 and October 2020. Study participants visited the clinic for nutrition consulting to improve their well-being, change lifestyle habits, or lose weight. Among the entire study population, 51.9% of the participants were male with a large range of BMI (14.7-59 kg/m2), and from different races, which represent well the general population in Israel. Participants were excluded from the study if they were younger than 20 years old (n = 21), lactating (n = 7), maternal women (n = 3), users of steroidal drugs (n = 18), or had undergone a surgical procedure that could affect body composition [e.g., amputated organs (n = 4), breast augmentation or reduction (n = 9), and liposuction surgeries (n = 3)].

[0003] Demographic parameters, including age and sex, were collected, as well as the medical history of bariatric surgery (n = 136), thyroid disorders (n = 46), or diabetes (n = 115) based on the clinical records. Furthermore, the participants (n = 2937, 97.9% of the study sample) were classified based on self-reported previous weight loss attempts: never (n = 2109), one or two attempts (n = 625), or three or more attempts (n = 203). The study flow chart is shown in Figure 1. Data collection and analysis were conducted in accordance with the relevant ethical codes and were approved by the Ethics Committee of Tel Aviv University.

[0004] The same clinician conducted all the anthropometric measurements. While most participants (>95%) were tested on the same day, some conducted the metabolic and body composition assessments one week apart. Weight (±0.1 kg) was recorded on a digital scale while subjects were dressed in shorts and a T-shirt. A standard, wall-mounted stadiometer was used to measure the height (±0. 1 cm) without shoes, and the BMI (kg/m2) was calculated accordingly. Neck and abdomen circumferences were measured (cm) with a flexible tape at the levels of laryngeal prominence and umbilicus, respectively.

[0005] Body composition measurements, FFM (kg), FM (kg), and FM (%), were measured using dual-energy X-ray absorptiometry and analyzed using the integrated software (enCORE 2011, v. 13.60.; GE™, Madison, WI, USA). Participants were instructed to arrive at the clinic after at least a 4 h fast. Before the scan, participants were asked to remove all metal items. Each whole-body scan took ~7 min. Quality control calibration procedures were conducted on a spine phantom each morning.

[0006] RMR was measured in a metabolic cart using an indirect-calorimeter device, Quark™ RMR (Cosmed™, Rome, Italy). Participants were instructed to arrive at the clinic by the morning after an overnight fast (12 h) and avoid any exercise training 24 h before the measurement. In addition, participants were restricted from consuming nicotine products for at least 2 h before the measurement. To ensure rest state when measuring RMR, according to the guidelines [8], participants were at rest 20 min before the measurement. Turbine calibration and gas calibration were performed before each test, according to the manufacturer's instructions.

[0007] During the measurement, the subjects lay awake in a supine position, in a quiet room with stable temperature (22-24 °C). All measurements were conducted by an adjusted size face mask. Measurements were 21 min long, with a 5 min adaptation phase excluded from the analysis, and the mean of the final 16 mm calculated as the mean RMR. In case of major movement or falling asleep during the measurement, this period was excluded, and the measurement continued to achieve the 16 min stability. Nearly all measurements (>90%) were at least 5 min steady state (10% or less) coefficient of variation in VO2 and VCO2) [8], VO2 and VCO2 were recorded every five seconds. The Weir method was used to convert respiratory gas measures to energy expenditure, with acceptable respiratory exchange ratio ranged from 0.68 to 0.90 [8],

[0008] Statistical Analysis

[0009] The normality of the distribution of each continuous variable was assessed using histograms and QQ plots and by the Kolmogorov-Smirnov Test. Variables found to have nonnormal distributions were subjected to traditional transformations: square root for left-tail distributions and log-normal transformation for right-tail distributions. Participant characteristics were presented as the mean ± SD for continuous variables and by prevalence for categorical and dichotomic variables. Student’s t-test or Pearson’s Chi-squared test were used to compare the sex differences.

[0010] Bland-Altman analysis was used to determine the accuracy and the level of agreement of five common RMR methods (Cunningham [1, 2], Harris-Benedict [3], Food and Agricultural Organization, WHO, United National University [4], Mifflin-St. Jeor [5], and Owen [6]) with 95% level of agreement and mean bias [(RMR measured - RMR estimated / RMR measured) x 100] between the estimated RMR of each method and the experimentally measured RMR from our data. Lin’s concordance correlation was used to examine the concordance between the r² and the linear method of each estimated method and the experimentally measured RMR values. The level of inter-method agreement was compared using Bland-Altman plots, with a 95% level of agreement for mean bias.

[0011] To develop a multivariable prediction model for RMR, the dataset was randomly split using the train_test_split function of the scikit-learn Python package. A random subset of 75% of the participants (n = 2251) was assigned to a training set and 25% (n = 750) to a validation set. The two sets were matched in age, sex, BMI, and body composition (Table 3). The inclusion variable for RMR predictors was set according to the Pearson’s coefficient between RMR and each variable and based on the least squares method to maximize r². Parameters were included in the model according to a stepwise method, based on Pearson’s coefficient between RMR and each variable, and by the least squares method to maximize r². Pearson’s correlation between potential variables was examined to avoid the exclusion of a multicollinearity variable with a high correlation (>0.7) in the model, and according to the variance inflation factor (VIF) [0012] The new model was evaluated on the validation set (n = 750), using Bland- Altman plots (with 95% level of agreement for mean bias), and using Lin’s concordance correlation. A good model fit was defined as having maximum mean absolute error ±200 kcal/24 h (~10%). Data were collected using Microsoft® Excel v. 16. 16.27 and analyzed using IBM® SPSS Statistics v.27.

[0013] Results

[0014] Characteristics of the 3001 participants across sex are presented in Table 1. The mean age was 41 ± 13 years, between 20-95 years of age, with the following age distribution: 742 participants were between 20-30 years old, 712 were 30-40 years old, 806 participants were 40-49 years old, 478 participants were 50-59 years old, 200 participants were 60-69 years old, and 63 participants were 70 years old or older. The mean was BMI 28 ± 5.5 kg/m2 (range: 14.7- 59 kg/m2), mean measured RMR was 1841 ± 365 kcal day-1, and 52% were females. As expected, compared to females, males had significantly higher RMR (2075 ± 325 kcal day-1 vs. 1615 ± 236 kcal day-1, p < 0.001) and FFM (64.1 ± 9.1 vs. 43.4 ± 6.4, p < 0.001), and lower FM (29.2 ± 9.3 vs. 39.5 ± 9.1, p < 0.001). The largest category of subjects was overweight (35.1%), followed by obesity (32.8%), normal weight (32.3%), and underweight (0.9%). The measured RMR, FFM, and FM increased significantly with increases in BMI (p < 0.001 for all).

[0015] Table 1 as shown in the figures, provides the characteristics of the study population according to sex. Continuous variables are presented as mean ± SD, while categorical and dichotomous variables are expressed as prevalence. To compare the results between females and males, independent samples, Student’s t-test, and Pearson’s Chi-squared test were employed. Categories for BMI classification include underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5-24.9 kg/m2), overweight (BMI 25-29 9 kg/m2), obesity class 1 (BMI 30-34.9 kg/m2), obesity class 2 (BMI 35-39.9 kg/m2), and obesity class 3 (BMI > 40 kg/m2). The key variables in the analysis are FFM (fat- free mass), FM (fat mass), and RMR (resting metabolic rate).

[0016] Bland-Altman analysis presenting the bias and the 95% LOA of each of the common prediction method is presented in Figure 2. The largest difference was obtained for the Cunningham method, which is based on FFM as the single predicting factor (1521 ± 280 kcal day-1, -16.6%; 95% LOA 1.9, -35.1). This was followed by the Owen method (1542 ± 281 kcal day-1, -15.4%; 95% LOA 4.2, -22.6), the Mifflin-St. Jeor method (1593 ± 284 kcal day-1, -12.6; 95% LOA 5.8, -26.5), and the Harris-Benedict method (1676 ± 313 kcal kg-1, -8.2%; 95% LOA 11.1, -27.7). The lowest mean deviation was obtained by the WHO/FAO/UAU method (1792 ± 701 kcal kg-1, -2.1%; 95% LOA 22.3, -26.5).

[0017] The explained variation (r²) ranges from 0.63 to 0.70, in the following order: Mifflin-St. Jeor > Harris-Benedict > WHO/FAO/UAU > Owen > Cunningham (Figure 2a, c, e, g). A high Pearson coefficient was found between RMR and weight (p = 0.74, p < 0.001), height (p = 0.65, p < 0.001), and neck circumference (p = 0.74, p < 0.001), with the highest coefficient obtained with FFM (p = 0 824, p < 0.001). In addition, a high Pearson coefficient was obtained between FFM and height (p = 0.8, p < 0.001), neck circumference (p = 0.86, p < 0.001), and weight (p = 0.65, p < 0.75). These were not combined as RMR predictors in the model in order to avoid multicollinearity.

[0018] Figure 2 shows a Bland- Altman analysis presenting the bias and the 95% level of agreement between each prediction method and the measured RMR. The x-axis represents the mean of the measured and estimated RMR, and the y-axis expresses the difference in percentage between the RMR predicted using each of the methods and the RMR measured as follows: (pRMR-mRMR/mRMR)/100). Lin’s concordance correlation was used to determine the R² and the linear method between each common method and the measured RMR. (a,b) Cunningham [1, 2]; (c,d) Harris-Benedict [3]; (e,f) Food and Agricultural Organization/WHO/United National University [4]; (g,h) Mifflin-St. Jeor [5]; (i,j) Owen et al. methods [6,7],

[0019] The two newly proposed models developed to predict RMR based on the training set are presented in Table 2 (n = 2251, 75%). Model 1 includes age (years), sex, FFM (kg), and FM (kg) as predictors with an R² value of 0.745. For males, RMR (kcal/24 h) = 775.8 - (age 5) + (FFM 20.5) + (FM* 7.7). For females, RMR (kcal/24 h) = 709 - (age* 5) + (FFM 20.5) + (FM 7.7)

[0020] Model 2 includes an interaction variable of FFM*FM, which increases the explained variation by 0.001. For males, RMR (kcal/24 h) = 891.7 - (age 5) + (FFM 18.5) + (FM 3.5) + (FFM* FM* 0.07). For females, RMR (kcal/24 h) = 824 - (age* 5) + (FFM 20.5) + (FM 7.7) + (FFM* FM* 0.07). [0021] In both models, the FFM predictor has the most significant contribution to RMR prediction (standardized P coefficients of 0.73 and 0.65, respectively; p < 0.001). Male sex was related to an increase of 9% in predicted RMR (p < 0.001); each year of age was related to a decrease of 5 kcal day-1, and 1 kg of FFM contributed 20.5 kcal day-1 to the RMR, which is a three-fold increase over the value for FM kg (7.7 kcal day-1, p < 0.001 for both). A sensitivity analysis for participants with thyroid disorders or following bariatric surgery revealed similar results in the accuracy of the model. Moreover, no significant change in the prediction accuracy or mean bias was found when the self-reported previous weight loss attempts were added to the model.

[0022] Next, Bland-Altman analysis was applied to the validation set to determine the mean bias and level of agreement of both models (Figure 3). The new model (model 2) successfully predicted 73% of the explained variation of the measured RMR (p < 0.001). The mean deviation percentage of model 1 is -0.7% (p = 0.049), and no significant difference was found for the average deviation of model 2 and the measured RMR (average deviation -0.6%: 0.55, p = 0. 123). The 95% level of agreement ranges from -18.6 to 19.7 for both proposed models.

[0023] Figure 3 shows a Bland- Altman analysis of the proposed new models applied to the validation set (n = 750). The x-axis represents the mean of the measured and estimated RMR, and the y-axis expresses the difference in percentage between the RMR predicted using each of the methods and the RMR measured as follows: (pRMR - mRMR/mRMR)/100). Lin’s concordance correlation was used to determine the R² and the linear method between each estimated method and the measured RMR. (a,b) model 1 , includes sex, age, FM, and FFM. (c,d) model 2, includes sex, age, FM, FFM, and the interaction between FM and FFM (FM*FFM).

[0024] Discussion

[0025] The present disclosure is the largest cross-sectional study to examine the accuracy of several common RMR prediction methods using body composition parameters, in addition to height, weight, sex, and age. As a result of the investigation, a novel prediction model for RMR is proposed. The known RMR prediction methods produced a large variation (-0.7% to -16.6%) in the mean bias with an explained variation of 0.63 to 0.70. In contrast, novel method of predicting RMR, which includes both fat mass and fat-free mass, can successfully predict 73% of the explained variation of the measured RMR, with a mean bias of -0.7%.

[0026] These findings demonstrate that the common prediction models underestimate experimentally measured RMR by -2.1 ± 12.4% to -16.1 ± 9.4%. While population characteristics may influence the accuracy of a model, the characteristics of the 3001 Israeli participants comprising the study population employed are very similar to those of the general population. For example, the average BMI of the study participants was 28.5 kg/m², similar to the mean BMI among the USA (28.5 kg/m²) and slightly higher than the average BMI in Israel of (26.3 kg/m²). In addition, the sample size we used to determine the accuracy of the common methods and to develop the new prediction model is larger than that used in any previous study. The Harris-Benedict method was developed based on 239 Caucasian participants with normal body weight [3], Furthermore, these previous measurements were conducted under resting and not basal conditions, with no representation of elderly participants. Similarly, although the WHO/FAU/UNU methods were based on a large number of participants (2526), this population differed from the general population in that 90% of the participants were men and mostly young members of the military or police forces [4], The Owen method was based on 60 men and 44 women, with an age range from 18 to 82 years and 18 to 65 years, respectively, excluding more elderly women [6,7], Similarly, the population used to develop the Mifflin-St. Jeor method comprised 498 participants, with members of all the BMI categories, and ages 19-78 but did not include the oldest old group (>80). In addition, the development was based on data from the RENO Diet-Heart study, which involved a five-year follow-up. It might have introduced biases related to the induction process [5], Accordingly, the presently proposed method is the most appropriate for the general population with a reasonable external validity.

[0027] The presently disclosed novel method exhibits higher accuracy with values of - 0.6%; 0.55, and r² of 0.73 compared to the 0.63-0.71 range of explained variation obtained from the commonly used methods. Notably, an extensive systematic review of validated common methods revealed large deviations between the predicted and measured values of RMR, with both under- and overestimations, depending on the method and the study population. The deviation generated by the Mifflin-St. Jeor ranged from an underestimate of 18% to an overestimate of 15%; the deviation generated by the Harris-Benedict method ranged from an underestimate of 65% to an overestimate of 43% among obese individuals [32] ; and the Owen method results ranged from 24% underestimation to 28% overestimation of the measured RMR [31], Furthermore, the overestimation in predicting RMR tended to be particularly in people with obesity due to the higher fat mass, which has a lower metabolic rate. The presently disclosed results did not find such an association. One possibility could be the large age distribution in the present study population, with the likelihood that some obese people were younger and had a high metabolic rate compared to older adults without obesity.

[0028] The present disclosure found that combining demographic indices (age and sex) together with body composition indices increases the prediction accuracy and reduces the range of deviation. RMR prediction models based on body composition parameters have been previously proposed in the existing literature, but most are based on specific populations, e.g., women or athletes only. The most common formula for use among the general population is the Cunningham method, whose development was based on nine databases from nine different studies (a total of 1483 observations), where the data concerning body composition and energetic expense were collected using different methods in each study. When the accuracy level and LOA were examined, it was found that of all the common methods examined, the Cunningham formula generated the most significant underestimate, 16.6% (95% LOA -35.1, -1.9). This finding emphasizes that, despite the high correlation (r = 0.7) between FFM and RMR [20], combining the demographic indices, age, and sex with body composition parameters contributes to the accuracy of the individual RMR assessment and reduces the average deviation. Furthermore, it was found that using body composition parameters (both FM and FFM) allows the method to be applied to a wide range of populations, including individuals with obesity.

[0029] The model proposed in the present disclosure predicts 73% of the variation in RMR. Several factors influence RMR variation and may contribute to an increase in the explained variation. Several studies have shown that past calorie restriction attempts may lead to less-than-predicted RMR by losing muscle and respiring mass. The metabolic effect of 5% weight loss remained even up to 6 years after the initial weight loss. On the other hand, other studies did not find a long-term metabolic effect on weight reduction.

[0030] Although it was hypothesized that past calorie restriction attempts would increase the explained variance, the presently disclosed models did not find an association between weight loss attempts and predicted RMR. Another variance in RMR may relate to the lack of ability to measure organ metabolic rate and size. The brain, liver, and kidneys have a relatively high mass-specific metabolic rate of ~240, 200, and 400 kcal/kg/day and account for about 22%, 21%, and 8% of RMR, respectively, in an average adult Moreover, studies in humans and mice found that weight loss or calorie restriction is associated with reduced internal organ size (except brain mass which increases in response to weight loss), which leads to metabolic adaptation. Therefore, the high variability and residuals in RMR may be explained, at least partially, by differences in the size and the metabolic rate of internal highly metabolic organs.

[0031] The present disclosure has several features. The large sample size, with the representation of a wide age range, minimizes the potential selection bias, as the study of the Israeli population has similar characteristics to the general population (same BMI and equal sex distribution).

[0032] The strengths of the present disclosure include DXA measurements of body composition and the use of a pre-defined protocol with the same metabolic cart for the RMR measurements for all participants. The disclosed new prediction method relies on DXA measurement to assess body composition, a gold standard but expensive assessment rarely used in a clinical setting. Because BIA is correlated with DXA measurements of body composition, it is considered that it can be used in the RMR method of the present disclosure. Accordingly, as a feature of the disclosure the proposed model can be used based on measurements obtained from BIA devices.

[0033] The presently disclosed method of predicting a person’s RMR can be used for predicting a person’s caloric needs, designing a personalized weight management, nutrition, exercise or weight loss plan, and making clinical assessments of a patient for providing medical recommendations. The average RMR has a wide range

[0034] Resting Metabolic Rate (RMR) serves a multifaceted role in various aspects of health and wellness. At its core, RMR provides a baseline measurement of the number of calories an individual's body requires while at rest. This fundamental information is instrumental in a range of applications, from calculating daily caloric needs to designing personalized dietary plans for weight management, weight loss, or weight gain. Nutritionists and healthcare professionals use RMR to fine-tune calorie recommendations for their clients, ensuring a balance between energy intake and expenditure. Additionally, RMR plays a pivotal role in weight management, helping individuals understand the relationship between their caloric consumption and weight changes. Fitness enthusiasts leverage RMR data to set appropriate exercise goals and tailor workouts to their specific objectives. Clinical settings use RMR measurements to assess patients' nutritional status and guide treatment for conditions like obesity, eating disorders, and malnutrition. Furthermore, RMR supports research efforts to investigate the impact of various factors on metabolism, while its data can be crucial for medical interventions in cases of extreme weight fluctuations. In summary, RMR provides a quantitative foundation for informed decisions about nutrition, exercise, and overall well-being, making it an indispensable tool in the fields of nutrition, fitness, and healthcare.

[0035] The presently disclosed method can be employed to predict a person’s RMR, allowing practitioners to make informed recommendations regarding a person’s health. Accordingly, the present disclosure relates to a method for predicting a person’s RMR and applying said RMR to recommend calculating a person’s caloric needs, designing a personalized weight management, nutrition, exercise or weight loss plan, and making clinical assessments of a patient to provide wellness recommendations. Wellness recommendations can be based solely on the predicted RMR of a patient or by comparing a patient’s actual RMR to their predicted RMR. Wellness recommendations based solely on a predicted RMR allow practitioners to compare a patients current calorie intake, exercise plan, and dietary choices, with what they should be for someone with the patient’ s predicted RMR.

[0036] Said wellness recommendations can include recommending that a person increase or decrease their caloric intake, increase or decrease the amount of calories burned per day by exercise, increase or decrease consumption of calorie dense foods per day. For example, meat is more calorie dense than vegetables which are less calorie dense. Therefore, the lower a patient’s predicted RMR, a practitioner might prescribe more vegetables and less meat to prevent weight gam, obesity and associated comorbidities, and other conditions.

[0037] Wellness recommendations can also be made by comparing a patient’s actual RMR with their predicted RMR. For example, said wellness recommendations can include modifying caloric intake by increasing caloric intake when a person’s actual RMR is higher than the patient’s predicted calorie intake. Said wellness recommendations can also include modifying caloric intake by decreasing a person’s calorie intake when the person’s actual RMR is lower than its predicted RMR. Achieving a lower daily calorie intake can be achieved by decreasing the relative daily consumption of calorie dense foods or increasing the relative consumption of less calorie dense foods. Achieving a higher daily calorie intake can be achieved by increasing the relative daily consumption of calorie dense foods or decreasing the relative consumption of less calorie dense foods.

[0038] Said wellness recommendations can also include modifying a patient’s exercise by increasing the number of calories burned per day through exercise when a person’s actual RMR is lower than the person’s predicted RMR. Said wellness recommendations can also include modifying a patient’s exercise plan by decreasing the number of calories burned per day through exercise when a person’s actual RMR is higher than the person’s predicted RMR.

[0039] Certain medications can have diverse effects on RMR. Thyroid medications such as Levothyroxine, Liothyronine, Armour Thyroid, Euthyrox are prescribed to manage hypothyroidism and can elevate RMR, potentially leading to weight loss. Conversely, corticosteroids such as Prednisone, Methylprednisolone, Dexamethasone, Prednisolone, used for inflammatory conditions, can decrease RMR causing weight gain and alterations in metabolism, often resulting in increased fat deposition. Some antidepressants, particularly selective serotonin reuptake inhibitors (SSRIs) like fluoxetine, sertraline citalopram, escital opram, paroxetine, and duloxetine, can either increase or decrease RMR, affecting weight changes, appetite, or fat storage patterns. Beta-blockers such as Metoprolol, Propranolol, Atenolol, Carvedilol, common for heart conditions and hypertension, tend to lower RMR and could lead to weight gain.

[0040] Antipsychotic medications, especially atypical antipsychotic drugs such as Olanzapine, Risperidone, Quetiapine, Aripiprazole, Clozapine, can decrease RMR and are associated with significant weight gain and can impact insulin sensitivity. Oral contraceptives such as those containing estrogen and progestin, Progestin, Ethinyl estradiol/levonorgestrel, Norgestrel/ethinyl estradiol, while typically causing modest effects, may lead to changes in RMR or weight in some women, often due to hormonal influences on water retention and fat storage. Anti-diabetic medications, including insulin, metformin, and sulfonylureas, glibenclamide, glimepiride, and sitagliptin, can have varying effects on RMR and affect glucose metabolism and contribute to varying weight changes. Medications designed for weight loss, such as Orlistat, Phentermine-topiramate, Buproprion-naltrexone, Liraglutide, Phentermine, increase RMR and energy expenditure and suppress appetite. In addition, certain antihistamines such as Diphenhydramine, Loratadine, Cetirizine, Fexofenadine, Desloratadine can induce drowsiness and temporarily reduce RMR.

[0041] Therefore, said wellness recommendations can also include recommendations regarding the prescription of such drugs affecting RMR when a patients actual RMR is outside of the predicted RMR provided by the models disclosed herein. Said wellness recommendations can include modifying a patient’s drug prescription by increasing or decreasing administration frequency or dosage of the aforementioned types of medications or recommending that a patient start or stop taking said medications. For example, if a patient’s actual RMR is lower than its predicted RMR based on the disclosed models, a practitioner can recommend modifying a patient’s drug prescription by starting or stopping medication or increasing or decreasing dosage or frequency of a medication to raise the patients actual RMR to better fit the predicted RMR. Conversely, if a patients actual RMR is higher than its predicted RMR based on the disclosed models, a practitioner can recommend modifying a patient’s drug prescription by starting or stopping medication or increasing or decreasing dosage or frequency of a medication to lower the patients actual RMR to better fit the predicted RMR.

[0001] According to some embodiments, the disclosure includes a system 100, such as a modern computing system, having a memory 104 and at least one processor 102 designed to execute computer-executable instructions. The system 100 can be a computing system the components of which, the processor 102, and the computer readable media 104, are in operational communication. Processor 102 can include any general purpose processor as well as a specialpurpose processor where software instructions are incorporated into the actual processor design. Processor 102 may essentially be a completely self-contained computing system, containing multiple cores or processors, a bus, memory controller, cache, etc. A multi-core processor may be symmetric or asymmetric.

[0042] These instructions enable the system to receive patient information comprising a set of variables including a patient's age, sex, Fat-Free Mass (FFM), and Fat Mass (FM), as used in model 1, above. In other embodiments, the set of variables can also include FFM*FM, as used in model 2. Using a predictive algorithm based on model 1 or model 2, above, the system 100 generates the patient's predicted Resting Metabolic Rate (RMR). Based on the predicted RMR, the system can generate at least one wellness recommendation. The predictive algorithm operates differently for male and female patients. According to some embodiments, for males, the RMR calculation is as follows: RMR (kCAL/24h) = 775.8 - (age5) + (FFM20.5) + (FM7.7).

According to some embodiments, for females, the RMR calculation is: RMR (kCAL/24h) = 709 - (age5) + (FFM20.5) + (FM*7.7).

[0043] The memory 104, for example, computer storage media, includes non-transitory storage memory, volatile media, nonvolatile media, removable media, and/or non-removable media implemented in a method or technology for storage of information, such as computer/machine-readable/executable instructions, data structures, program modules, or other data. Communication media may embody computer/machine-readable/executable instructions, data structures, program modules, or other data and include an information delivery media or system, both of which are hardware.

[0044] According to some embodiments, the system 100 can further receive the patient's actual RMR and provide wellness recommendations based on a comparison of the patient’s actual RMR and the patient’s predicted RMR. By way of example, when the patient's actual RMR is higher than the predicted RMR, the system 100 can generate wellness recommendations aiming to reduce the patient’ s actual RMR. Conversely, if the patient's actual RMR is less than the predicted RMR, the system 100 can generate wellness recommendations to increase the actual RMR. According to certain non-limiting embodiments, wellness recommendations can include modifying the patient's calorie intake, adjusting their exercise plan, or altering their drug prescription.

[0045] Any of the steps, operations, functions, or processes described herein may be performed or implemented by a combination of hardware and software services or services, alone or in combination with other devices. In some embodiments, the method can be performed via a service that can be software that resides in memory of a device and/or one or more servers and perform one or more functions when a processor executes the software associated with the service. In some embodiments, a service is a program or a collection of programs that carry out a specific function. In some embodiments, a service can be considered a server. The memory can be a non-transitory computer-readable medium. [0046] In some embodiments, the computer-readable storage devices, mediums, and memories can include a cable or wireless signal containing a bit stream and the like. However, when mentioned, non-transitory computer-readable storage media expressly exclude media such as energy, carrier signals, electromagnetic waves, and signals per se.

[0047] Methods according to the above-described examples can be implemented using computer-executable instructions that are stored or otherwise available from computer-readable media. Such instructions can comprise, for example, instructions and data which cause or otherwise configure a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Portions of computer resources used can be accessible over a network. The executable computer instructions may be, for example, binaries, intermediate format instructions such as assembly language, firmware, or source code. Examples of computer-readable media that may be used to store instructions, information used, and/or information created during methods according to described examples include magnetic or optical disks, solid-state memory devices, flash memory, USB devices provided with non-volatile memory, networked storage devices, and so on.

[0048] Devices implementing methods according to these disclosures can comprise hardware, firmware and/or software, and can take any of a variety of form factors. Typical examples of such form factors include servers, laptops, smartphones, small form factor personal computers, personal digital assistants, and so on. The functionality described herein also can be embodied in peripherals or add-in cards. Such functionality can also be implemented on a circuit board among different chips or different processes executing in a single device, by way of further example.

[0049] The instructions, steps, media for conveying such instructions, data input and output devices, computing resources for executing them, and other structures for supporting such computing resources are means for providing the functions described in these disclosures.

[0050] The disclosure also includes a method 200 including step (a) 202 for receiving, by at least one processor, patient information comprising a set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM), and step (b) 204 executing, by the at least one processor, a predictive algorithm to generate the patient’s predicted Resting Metabolic Rate (RMR). The method further includes step (c) 206 generating, by the at least one processor, at least one wellness recommendation based on the patient’s predicted RMR. Based on the predicted RMR, the method can generate at least one wellness recommendation. The set of variables may also encompass FFM*FM.

[0051 ] According to some embodiments, the method includes step (d) 208 receiving the patient's actual RMR. The method can further include step (e) 210, comparing the actual RMR with the predicted RMR and make a wellness recommendation based on whether the patient’s actual RMR should be raised or lowered. When the actual RMR is greater than the predicted RMR, the method can provide wellness recommendations to decrease the actual RMR. Conversely, when the actual RMR is lower than the predicted RMR, the method can offer wellness recommendations to increase the actual RMR.

[0052] According to other embodiments, the disclosure includes a non-transitory computer-readable storage medium containing instructions that, when executed by a computing device, enable the device to receive patient information, calculate the patient's predicted RMR using a predictive algorithm. The system can generate wellness recommendations based on the patient’s predicted RMR or based on a comparison of the patient’s predicted RMR with the patient’s actual RMR. In some embodiments, the set of variables can encompass FFM*FM.

[0053] It is believed that the present disclosure and many of its attendant advantages will be understood by the foregoing description, and it will be apparent that various changes may be made in the form, construction and arrangement of the components without departing from the disclosed subject matter or without sacrificing all of its material advantages. The form described is merely explanatory, and it is the intention of the following claims to encompass and include such changes.

[0054] While the present disclosure has been described with reference to various embodiments, it will be understood that these embodiments are illustrative and that the scope of the disclosure is not limited to them. Many variations, modifications, additions, and improvements are possible. More generally, embodiments in accordance with the present disclosure have been described in the context of particular implementations. Functionality may be separated or combined in blocks differently in various embodiments of the disclosure or described with different terminology. These and other variations, modifications, additions, and improvements may fall within the scope of the disclosure as defined in the claims that follow. [0055] References

[0056] All publications, patents, patent applications and other references cited in this application are incorporated herein by reference in their entirety for all purposes to the same extent as if each individual publication, patent, patent application or other reference was specifically and individually indicated to be incorporated by reference in its entirety for all purposes. Citation of a reference herein shall not be construed as an admission that such is prior art to the present invention. References below are cited throughout the disclosure by their number as [#].

[0057] [1] Cunningham, J. J. A Reanalysis of the Factors Influencing Basal Metabolic Rate in Normal Adults. Am. J. Clin. Nutr. 1980, 33, 2372-2374. https ://doi. org/10.1093/aj cn/33. 11.2372.

[0058] [2] Broad, E.M.; Newsome, L.J.; Dew, D.A.; Barfield, J.P. Measured and Predicted Resting Energy Expenditure in Wheelchair Rugby Athletes. J. Spinal Cord Med. 2020, 43, 388-397. https://doi.org/10.1080/10790268.2019.1608062.

[0059] [3] Harris, J. A.; Benedict, F.G. A Biometric Study of Human Basal Metabolism. Proc. Natl. Acad. Sci. USA 1918, 4, 370-373. https://doi.Org/10.1073/pnas.4.12.370.

[0060] [4] LIVESEY, G. Energy and Protein Requirements the 1985 Report of the 1981 Joint FAO/WHO/UNU Expert Consultation. Nutr. Bull. 1987, 12, 138-149. https://doi.Org/10.l l l l/j.1467-3010.1987.tb00040.x.

[0061] [5] Mifflin, M.D.; St Jeor, S.T.; Hill, L.A.; Scott, B.J.; Daugherty, S.A.; Koh, Y.O. A New Predictive Method for Resting Energy Expenditure in Healthy Individuals. Am. J. Clin. Nutr. 1990, 51, 241-247. https://doi.org/10.1093/ajcn/5E2.241.

[0062] [6] Owen, O.E.; Kavle, E.; Owen, R.S.; Polansky, M.; Caprio, S.; Mozzoli, M.A.; Kendrick, Z.V Bushman, M.C.; Boden, G. A Reappraisal of Caloric Requirements in Healthy Women. Am. J. Clin. Nutr. 1986, 44, 1-19. https://doi.Org/10.1093/ajcn/44.l.l.

[0063] [7] Owen, O.E.; Holup, J.L.; D’Alessio, D.A.; Craig, E.S.; Polansky, M.; Smalley, K.J.; Kavle, E C.; Bushman, M.C.; Owen, L.R ; Mozzoli, M.A.; et al. A Reappraisal of the Caloric Requirements of Men. Am. J. Clin. Nutr. 1987, 46, 875-885. https://doi.Org/10.1093/ajcn/46.6.875. [0064] [8] Fullmer, S.; Benson-Davies, S.; Earthman, C.P.; Frankenfield, D C ; Gradwell, E.; Lee, P.S.P.; Piemonte, T.; Trabulsi, J. Evidence Analysis Library Review of Best Practices for Performing Indirect Calorimetry in Healthy and Non-Critically Ill Individuals. J. Acad. Nutr. Diet 2015, 115, 1417-1446.e2. https://doi.Org/10.1016/j.jand.2015.04.003.

[0065] Abbreviations

[0066] BEE - basal energy expenditure

[0067] BIA - bioelectrical impedance analysis

[0068] BMI - body mass index

[0069] DXA - dual-energy X-ray absorptiometry

[0070] FAO - Food and Agricultural Organization

[0071] RMR - resting metabolic rate

[0072] T2DM - type 2 diabetes

[0073] TDEE - total daily energy expenditure

[0074] TEE - total energy expenditure

[0075] UNU - United National University

[0076] VIF - variance inflation factor

[0077] WHO - World Health Organization

Claims

1. A system comprises a memory and at least one processor to execute computer-executable instructions to: receive patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM); execute a predictive algorithm to generate the patient’s predicted Resting Metabolic Rate (RMR), wherein, the predictive algorithm is selected from Model 1 or Model 2; wherein, the Model 1 predictive algorithm for males is calculated as RMR (kcal/24 h) = 775.8 - (age 5) + (FFM 20.5) + (FM* 7.7); wherein, the Model 1 predictive algorithm for females is calculated as RMR (kcal/24 h) = 709 - (age* 5) + (FFM 20.5) + (FM 7.7); wherein, the Model 2 predictive algorithm for males is calculated as RMR (kcal/24 h) = 891 .7 - (age 5) + (FFM 18.5) + (FM 3.5) + (FFM* FM* 0.07); and wherein, the Model 2 predictive algorithm for females is calculated as RMR (kcal/24 h) = 824 - (age* 5) + (FFM 20.5) + (FM 7.7) + (FFM* FM* 0.07).

2. The system of any of claim 1, wherein the predictive algorithm is based on Model 1.

3. The system of any of claims 1, wherein the predictive algorithm is based on Model 2

4. The system of any of claims 1 - 3, further comprising instructions to: receive the patient’s actual RMR; wherein, when the patient’s actual RMR is greater than the patient’s predicted RMR, generating at least one wellness recommendation to decrease the patient’s actual RMR.

5. The system of any of claims 1 - 3, further comprising: receiving the patient’ s actual RMR; wherein, when the patient’s actual RMR is less than the patient’s predicted RMR, generating at least one wellness recommendation to increase the patient’s actual RMR.

6. The system of any of claims 4 - 5, wherein: the at least one wellness recommendation is selected from the group consisting of modifying the patient’s calorie intake, modifying the patient’s exercise plan, and modifying the patient’s drug prescription.

7. The system of claim 6, wherein: modifying the patient’s calorie intake recommends the patient increase daily calorie intake when actual RMR is higher than predicted RMR; or wherein, modifying the patient’s calorie intake recommends the patient decrease daily calorie intake when actual RMR is lower than predicted RMR.

8. The system of claim 6, wherein: modifying the patient’s exercise plan recommends the patient increase the number of calories burned per day through exercise when the patient’s actual RMR is lower than the patient’s predicted RMR.

9. The system of claim 6, wherein: modifying the patient’s exercise plan recommends the patient decrease the number of calories burned per day through exercise when the patient’s actual RMR is higher than the patient’s predicted RMR.

10. The system of claim 6, wherein: modifying the patient’s drug prescription recommends the patient increase the number of calories burned per day through exercise when the patient’s actual RMR is lower than the patient’s predicted RMR.

11. A method, comprising: receiving, by at least one processor, patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM); executing, by the at least one processor, a predictive algorithm based on Model 1 or Model 2 to generate the patient’s predicted Resting Metabolic Rate (RMR).

12. The method of any of claims 11, wherein: the predictive algorithm is based on Model 1

13. The method of any of claims 11, wherein: the predictive algorithm is based on Model 2

14. The method of any of claims 11 - 13, further comprising: receiving, by the at least one processor, the patient’s actual RMR; wherein, when the patient’s actual RMR is greater than the patient’s predicted RMR, generating, by the at least one processor, at least one wellness recommendation to decrease the patient’s actual RMR.

15. The method of any of claims 11 — 13, further comprising: receiving, by the at least one processor, the patient’s actual RMR; wherein, when the patient’s actual RMR is less than the patient’s predicted RMR, generating, by the at least one processor, at least one wellness recommendation to increase the patient’s actual RMR.

16. The method of any of claims 14 — 15, wherein: the at least one wellness recommendation is selected from the group consisting of modifying the patient’s calorie intake, modifying the patient’s exercise plan, and modifying the patient’s drug prescription.

17. A non- transitory computer-readable storage medium comprising instructions stored thereon that, when executed by a computing device to perform operations, the operations comprising: receiving patient information comprising a set of variables, said set of variables comprising a patient’s age, sex, Fat-Free Mass (FFM), and Fat Mass (FM); executing a predictive algorithm based on Model 1 or Model 2 to generate the patient’s predicted Resting Metabolic Rate (RMR).

18. The non-transitory computer-readable storage medium of claim 17, wherein: the predictive algorithm is based on Model 1 .

19. The non-transitory computer-readable storage medium of claim 17, wherein: the predictive algorithm is based on Model 2.

20. The non-transitory computer-readable storage medium of claim 17, further comprising: generating at least one wellness recommendation based on the patient’s predicted RMR.