Energy expenditure

Recommendations for Determining the Validity of Consumer Wearables and Smartphones for the Estimation of Energy Expenditure: Expert Statement and Checklist of the INTERLIVE Network

BACKGROUND

Consumer wearables and smartphone devices commonly offer an estimate of energy expenditure (EE) to assist in the objective monitoring of physical activity to the general population. Alongside consumers, healthcare professionals and researchers are seeking to utilise these devices for the monitoring of training and improving human health. However, the methods of validation and reporting of EE estimation in these devices lacks rigour, negatively impacting on the ability to make comparisons between devices and provide transparent accuracy.

OBJECTIVES

The Towards Intelligent Health and Well-Being Network of Physical Activity Assessment (INTERLIVE) is a joint European initiative of six universities and one industrial partner. The network was founded in 2019 and strives towards developing best-practice recommendations for evaluating the validity of consumer wearables and smartphones. This expert statement presents a best-practice validation protocol for consumer wearables and smartphones in the estimation of EE.

METHODS

The recommendations were developed through (1) a systematic literature review; (2) an unstructured review of the wider literature discussing the potential factors that may introduce bias during validation studies; and (3) evidence-informed expert opinions from members of the INTERLIVE network.

RESULTS

The systematic literature review process identified 1645 potential articles, of which 62 were deemed eligible for the final dataset. Based on these studies and the wider literature search, a validation framework is proposed encompassing six key domains for validation: the target population, criterion measure, index measure, testing conditions, data processing and the statistical analysis.

CONCLUSIONS

The INTERLIVE network recommends that the proposed protocol, and checklists provided, are used to standardise the testing and reporting of the validation of any consumer wearable or smartphone device to estimate EE. This in turn will maximise the potential utility of these technologies for clinicians, researchers, consumers, and manufacturers/developers, while ensuring transparency, comparability, and replicability in validation.

TRIAL REGISTRATION

PROSPERO ID: CRD42021223508.

New equations to estimate resting energy expenditure in obese adults from body composition

AIMS

The aims of this study were: to develop new equations for predicting resting energy expenditure (REE) in obese Italian subjects according to body composition parameters; to compare them with predicted values estimated by other REE prediction equations; and to cross-validate our equations using a validation set cohort.

METHODS

Four hundred patients were enrolled and divided into three groups. Besides anthropometry and REE (indirect calorimetry), total body fat and lean were evaluated by dual X-ray absorptiometry, and fat mass and fat-free mass by bioelectrical impedance analysis.

RESULTS

The subjects eligible to participate were 330. Group 1 (n = 174) was used to develop (R ² = 0.79) and (R ² = 0.77). Group 2 (n = 115) was used to generate (R ² = 0.85) and (R ² = 0.81). Group 3 (n = 41) was used to cross-validate the equations.

CONCLUSION

Equations 1 and 3 are reliable to measure REE from calorimetry and better than other equations that use anthropometric variables as predictors of REE. Further analysis in different populations is required before it can be applied in clinical practice.

Indirect Calorimetry: From Bench to Bedside

Accurate determination of energy expenditure (EE) is vitally important yet often neglected in clinical practice. Indirect calorimetry (IC) provides one of the most sensitive, accurate, and noninvasive measurements of EE in an individual. Over the last couple of decades, this technique has been applied to clinical circumstances such as acute illness and parenteral nutrition. Beyond assessing the nutritional needs, it has also shed light on various aspects of nutrient assimilation, thermogenesis, the energetics of physical exercise, and the pathogenesis of obesity and diabetes. However, because of little or no experience with IC provided during medical education, the benefits of IC are poorly appreciated. Newer technology, cost-effectiveness, and a better understanding of how to interpret measurements should lead to more frequent use of IC. This review focuses on the physicochemical background of IC, the various indications for use, techniques and instruments, potential pitfalls in measurement, and the recent advances in technology that has adapted the technique to long-term studies in humans.

Assessment of physical activity and energy expenditure: an overview of objective measures

The ability to assess energy expenditure (EE) and estimate physical activity (PA) in free-living individuals is extremely important in the global context of non-communicable diseases including malnutrition, overnutrition (obesity), and diabetes. It is also important to appreciate that PA and EE are different constructs with PA defined as any bodily movement that results in EE and accordingly, energy is expended as a result of PA. However, total energy expenditure, best assessed using the criterion doubly labeled water (DLW) technique, includes components in addition to physical activity energy expenditure, namely resting energy expenditure and the thermic effect of food. Given the large number of assessment techniques currently used to estimate PA in humans, it is imperative to understand the relative merits of each. The goal of this review is to provide information on the utility and limitations of a range of objective measures of PA and their relationship with EE. The measures discussed include those based on EE or oxygen uptake including DLW, activity energy expenditure, physical activity level, and metabolic equivalent; those based on heart rate monitoring and motion sensors; and because of their widespread use, selected subjective measures.

Invited review: Aging and energy balance

Humans over 70 yr of age often lose weight. This appears to be due to a physiological anorexia of aging as well as a loss of lean mass (sarcopenia) and, to a lesser extent, fat mass. The causes of the physiological anorexia of aging include changes in taste and smell and a decrease in adaptive relaxation of the fundus of the stomach, which leads to more rapid antral filling and early satiation. In addition, basal and stimulated levels of the satiating hormone, cholecystokinin, are increased. In men, the decline in testosterone leads to an increase in leptin and a loss of lean mass. Although resting metabolic rate declines with aging, this is mainly due to the decline in lean body mass. Energy metabolism is also decreased due to a decline in Na+-K+-ATPase activity, decreased muscle protein turnover, and possibly changes in mitochondrial membrane protein permeability. Physical energy expenditure declines with aging. Meal-induced thermogenesis shows a delay to peak, possibly due to a delay in gastric emptying. Inadequate data are available on the effect of aging in humans on other energy-producing mechanisms such as adaptive thermogenesis. These physiological changes place older men and women at major risk of developing pathological weight loss when they develop disease states, especially those associated with cytokine elaboration.

Energy metabolism of young rats after early postnatal overnutrition

Early postnatal overnutrition (PNO) induced by restricting litter size in rats leads to increased body-weight (BW) and body-fat gain in later life. PNO rats are used as an animal model of moderate obesity and early hyperinsulinism. We investigated whether the increased adiposity could be due to a decreased energy expenditure. Male newborn Wistar rats were raised in litters of either two (SL) or twelve pups (NL), weaned at 4 weeks of age and subsequently fed ad libitum. BW was recorded continuously until 12 weeks of age. Daily energy intake, total daily energy expenditure (EE, measured by indirect calorimetry) and body composition were measured in weaned pups at 5, 8 and 12 weeks of age. SL rats displayed increased BW compared with NL rats from week 2 to 5 and again from week 10 to 12. Lean body mass, body fat and protein content and total EE were increased in SL rats at week 5. The same linear correlation described the relationship between BW and total EE in NL and SL rats. At week 8 to 12 no differences in energy metabolism could be found, but the total fat content was increased in SL rats at week 12. Energy balance, i.e. assimilated energy minus EE, was no different between SL and NL at any time that it was measured. We conclude that although PNO rats display increased adiposity in early life, there seem to be no long-lasting effects on energy metabolism in later life, even if a tendency to increased adiposity can still be detected.

Habitual fat intake and basal fat oxidation in obese and non-obese Caucasians

OBJECTIVE

To examine the relationship between habitual fat intake and basal fat oxidation in obese and non-obese Caucasian men and women.

METHODS

Habitual fat intake was assessed by 7-day weighed dietary records and resting fat oxidation was determined after an overnight fast in 132 weight stable non-diabetic subjects (38 males, 94 females). All subjects were characterized for weight, height, waist-to-hip ratio, physical activity, plasma glucose and insulin response to an oral glucose load, plasma catecholamine and leptin levels. Under-reporters, defined according to plausibility of the relationship between energy expenditure and energy intake, were excluded from the analyses.

RESULTS

The mean age was 53.1+/-10.6 y (19-72 y) and mean body mass index (BMI) was 30.7+/-5.8 kg/m(2) (19.4-45.8 kg/m(2)). Sixty-eight subjects were obese (BMI>30 kg/m(2)). Univariate regression analysis revealed a significant, albeit modest, relationship between absolute fat intake and BMI (r(2)=0.06; P<or=0.05) but not between fat intake and fat mass (r(2)=0.026; P=0.08). However, multiple regression analysis revealed significant effects of body fat mass (FM) and sex on basal fat oxidation (bFO) explaining 33% of the variation of bFO (P<or=0.0001; radical s.e.=18.0 g/24 h). In univariate regression analysis, habitual fat intake was significantly related to adjusted fat oxidation, explaining 12% of the variation (P<or=0.0001; radical s.e.=11.7 g/24 h).

CONCLUSION

Habitual fat intake has a significant, albeit modest, effect on basal fat oxidation, even when adjusted for sex and body FM. The rather modest effect of habitual fat intake on fat oxidation may in part explain the increased propensity to gain FM on a high-fat diet.

The neuroendocrinology of obesity

The regulation of energy balance is enormously complex, with numerous genetic, hormonal, neural/behavioral, and societal influences. Although the current epidemic of obesity has its underpinnings in the changes in culture during the last half century, the role of the neuroendocrine system in the genesis of obesity is physiologically and therapeutically unavoidable. Increased understanding of this system has suggested organic etiologies (and therapies) for some rare and not-so-rare forms of obesity. With so many inputs, it is not implausible that dysfunction of other parts of this feedback system will be found to explain other forms of obesity in the future. Fortunately or unfortunately, diet and exercise remain the mainstays of obesity therapy. Most diet-exercise programs result in an acute 11-kg weight loss in adults; the question is whether it can be sustained without significant long-term behavior modification. In the European Sibutramine Trial of Obesity Reduction and Maintenance (STORM), 42% of treated patients dropped out; of those remaining, 77% of subjects lost more than 5% of initial body weight, but only 43% of these individuals maintained greater than 80% of this loss over 2 years. Could there be an organic component in persons who do not respond? Obesity pharmacotherapies sometimes have beneficial acute effects, but these effects are impermanent; discontinuation tends to result in a rebound weight gain, suggesting that the etiology of the obesity is still present. A useful guiding principle is that patients who do not respond to diet and exercise should undergo an initial medical evaluation, including assessments of birth weight, past medical history, weight history, family history, diet, exercise, and fasting insulin and thyroid levels. As the nosology of obesity improves, diagnostic efficiency and therapeutic success should increase, leading to a decrease in associated morbidity, mortality, and socioeconomic ramifications.

The neuroendocrinology of childhood obesity

The regulation of energy balance is enormously complex, with numerous genetic, hormonal, neural and behavioral, and societal influences. Although the current epidemic of obesity clearly has its underpinnings in the changes in culture during the past half-century (see other articles in this issue), the role of the neuroendocrine system in the genesis of obesity, as described in this article, is physiologically and therapeutically unavoidable. An understanding of this system has suggested organic causes (and therapies) for some rare and not-so-rare forms of obesity. With so many inputs, it is not far-fetched to assume that dysfunction of other parts of this feedback system will be found to explain other forms of obesity in the future. What does this mean for obese children entering the pediatrician's office? Fortunately or unfortunately, diet and exercise are the mainstays of obesity therapy for children and adults. Most diet-exercise programs result in an acute 11-kg weight loss in adults; the question is whether it can be sustained without significant long-term behavioral modification. For instance, the European Sibutramine Trial of Obesity Reduction and Maintenance trial showed that 42% of treated subjects drop out; of those remaining, 77% of subjects lost more than 5% of initial body weight, but only 43% of those maintained more than 80% of this over 2 years. Could there be an organic component in those who do not respond? Of course, obesity pharmacotherapies sometimes have beneficial acute effects, but these drugs work for only as long as they are consumed; discontinuation tends to result in a "rebound" weight gain, suggesting that the cause of the obesity is still present. Furthermore, in 2001, there are no obesity drugs approved for children. A useful guiding principle is that children deserve at the minimum an initial medical evaluation, including birth weight, medical history, family history, dietary evaluation, and exercise assessment. Perhaps the most important feature that can distinguish "organic" from "behavioral" weight gain in childhood is the age of the "adiposity rebound." The Centers for Disease Control and Prevention now supplies BMI charts for boys and girls at www.cdc.gov/growthcharts. Plotting of the BMI versus age allows pediatricians to determine the age at which the BMI starts to increase (mean, 5.5 years). The earlier the adiposity rebound, the more likely the child will be obese as an adult, and the more likely that an organic cause can be determined. In such patients, thyroid levels and fasting insulin and leptin levels should be measured. An initial attempt at diet and exercise is essential; patients who do not respond with BMI stabilization should be investigated for a more ominous cause of their obesity. As the nosology of obesity improves, pediatricians will be able to increase the diagnostic efficiency and therapeutic success of this unfortunate, debilitating, and expensive epidemic.