D B Allison

University of Alabama at Birmingham, Birmingham, Alabama, United States

Are you D B Allison?

Claim your profile

Publications (256)1408.46 Total impact

  • A E Ivanescu · P Li · B George · A W Brown · S W Keith · D Raju · D B Allison
    [Show abstract] [Hide abstract]
    ABSTRACT: Deriving statistical models to predict one variable from one or more other variables, or predictive modeling, is an important activity in obesity and nutrition research. To determine the quality of the model, it is necessary to quantify and report the predictive validity of the derived models. Conducting validation of the predictive measures provides essential information to the research community about the model. Unfortunately, many articles fail to account for the nearly inevitable reduction in predictive ability that occurs when a model derived on one dataset is applied to a new dataset. Under some circumstances, the predictive validity can be reduced to nearly zero. In this overview, we explain why reductions in predictive validity occur, define the metrics commonly used to estimate the predictive validity of a model (e.g., R(2), mean squared error, sensitivity, specificity, receiver operating characteristic, concordance index), and describe methods to estimate the predictive validity (e.g., cross-validation, bootstrap, adjusted and shrunken R(2)). We emphasize that methods for estimating the expected reduction in predictive ability of a model in new samples are available and this expected reduction should always be reported when new predictive models are introduced.International Journal of Obesity accepted article preview online, 09 October 2015. doi:10.1038/ijo.2015.214.
    International journal of obesity (2005) 10/2015; DOI:10.1038/ijo.2015.214 · 5.00 Impact Factor
  • J A Dawson · K A Kaiser · O Affuso · G R Cutter · D B Allison
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: It has not been established whether control conditions with large weight losses (WLs) diminish expected treatment effects in WL or prevention of weight gain (PWG) randomized controlled trials (RCTs). Subjects/methods: We performed a meta-analysis of 239 WL/PWG RCTs that include a control group and at least one treatment group. A maximum likelihood meta-analysis framework is used in order to model and understand the relationship between treatment effects and control group outcomes. Results: Under the informed model, an increase in control group WL of one kilogram corresponds with an expected shrinkage of the treatment effect by 0.309 kg [95% CI (-0.480, -0.138), P=0.00081]; this result is robust against violations of the model assumptions. Conclusions: We find that control conditions with large weight losses diminish expected treatment effects. Our investigation may be helpful to clinicians as they design future WL/PWG studies.International Journal of Obesity accepted article preview online, 09 October 2015. doi:10.1038/ijo.2015.212.
    International journal of obesity (2005) 10/2015; DOI:10.1038/ijo.2015.212 · 5.00 Impact Factor
  • P L Capers · A D Fobian · K A Kaiser · R Borah · D B Allison
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent epidemiological and ecological trends in humans indicate a possible causal relationship between sleep duration and energy balance. We aimed to find experimental evidence that has tested this relationship between sleep duration and measures of body composition, food intake or biomarkers related to food intake. We conducted a systematic literature review using six databases throughout 7 August 2014. We sought reports of randomized controlled trials where sleep duration was manipulated and measured outcomes were body weight or other body composition metrics, food intake, and/or biomarkers related to eating. We found 18 unique studies meeting all criteria: eight studies with an outcome of body weight (4 - increased sleep, 4 - reduced sleep); four studies on food intake; four studies of sleep restriction on total energy expenditure and three of respiratory quotient; and four studies on leptin and/or ghrelin. Few controlled experimental studies have addressed the question of the effect of sleep on body weight/composition and eating. The available experimental literature suggests that sleep restriction increases food intake and total energy expenditure with inconsistent effects on integrated energy balance as operationalized by weight change. Future controlled trials that examine the impact of increased sleep on body weight/energy balance factors are warranted. © 2015 World Obesity.
    Obesity Reviews 06/2015; 16(9). DOI:10.1111/obr.12296 · 8.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: International Journal of Obesity is a monthly, multi-disciplinary forum for papers describing basic, clinical and applied studies in biochemistry, genetics and nutrition, together with molecular, metabolic, psychological and epidemiological aspects of obesity and related disorders
    International journal of obesity (2005) 04/2015; 39(7). DOI:10.1038/ijo.2015.81 · 5.00 Impact Factor
  • Nutrition 03/2015; 31(7-8). DOI:10.1016/j.nut.2015.02.008 · 2.93 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: That one or multiple measures of metabolic rate may be robustly associated with, or possibly even causative of, the progression of aging-resultant phenotypes such as lifespan is a long-standing, well-known mechanistic hypothesis. To broach this hypothesis, we assessed metabolic function and spontaneous locomotion in two genetic and one dietary mouse models for retarded aging, and subjected the data to mediation analyses to determine whether any metabolic or locomotor trait could be identified as a mediator of the effect of any of the interventions on senescence. We do not test the hypothesis of causality (which would require some experiments), but instead test whether the correlation structure of certain variables is consistent with one possible pathway model in which a proposed mediating variable has a causal role. Results for metabolic measures, including oxygen consumption and respiratory quotient, failed to support this hypothesis; similar negative results were obtained for three behavioral motion metrics. Therefore, our mediation analyses did not find support that any of these correlates of decelerated senescence was a substantial mediator of the effect of either of these genetic alterations (with or without caloric restriction) on longevity. Further studies are needed to relate the examined phenotypic characteristics to mechanisms of aging and control of longevity. © 2015 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
    Aging cell 02/2015; 14(3). DOI:10.1111/acel.12318 · 6.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Researchers and participants' expectations can influence treatment response. Less is known about the effects of researchers' expectations on the accuracy of data collection in the context of a weight loss trial.Methods Student raters (N = 58; age = 20.1 ± 2.3 years) were recruited to weigh individuals who they thought were completing a 12-month weight loss trial, although these ‘participants’ were actually standardized patients (SPs) playing these roles. Prior to data collection, student raters were provided information suggesting that the tested treatment had been effective. Each student rater received a list of 9–10 ‘participants’ to weigh. While the list identified each person as ‘treatment’ or ‘control’, this assignment was at random, which allowed us to examine the effects of non-blinding and expectancy manipulation on weight measurement accuracy. We hypothesized that raters would record the weights of ‘treatment participants’ as lower than those of ‘control participants’.ResultsContrary to our hypothesis, raters recorded weights that were 0.293 kg heavier when weighing ‘treatment’ vs. ‘control’ SPs, although this difference was not significant (P = 0.175).Conclusions This pilot study found no evidence that manipulating expectancies about treatment efficacy or not blinding raters biased measurements. Future work should examine other biases which may be created by not blinding research staff who implement weight loss trials as well as the participants in those trials.
    02/2015; 5(1). DOI:10.1111/cob.12083
  • Childhood Obesity 12/2014; 10(6):542. DOI:10.1089/chi.2014.0081 · 1.87 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Energy intake (EI) and physical activity energy expenditure (PAEE) are key modifiable determinants of energy balance, traditionally assessed by self-report despite its repeated demonstration of considerable inaccuracies. We argue here that it is time to move from the common view that self-reports of EI and PAEE are imperfect, but nevertheless deserving of use, to a view commensurate with the evidence that self-reports of EI and PAEE are so poor that they are wholly unacceptable for scientific research on EI and PAEE. While new strategies for objectively determining energy balance are in their infancy, it is unacceptable to use decidedly inaccurate instruments, which may misguide health care policies, future research, and clinical judgment. The scientific and medical communities should discontinue reliance on self-reported EI and PAEE. Researchers and sponsors should develop objective measures of energy balance.International Journal of Obesity accepted article preview online, 13 November 2014. doi:10.1038/ijo.2014.199.
    International journal of obesity (2005) 11/2014; 39(7). DOI:10.1038/ijo.2014.199 · 5.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background:Public health and clinical interventions for obesity in free-living adults may be diminished by individual compensation for the intervention. Approaches to predict weight outcomes do not account for all mechanisms of compensation, so they are not well suited to predict outcomes in free-living adults. Our objective was to quantify the range of compensation in energy intake or expenditure observed in human randomized controlled trials (RCTs).Methods:We searched multiple databases (PubMed, CINAHL, SCOPUS, Cochrane, ProQuest, PsycInfo) up to 1 August 2012 for RCTs evaluating the effect of dietary and/or physical activity interventions on body weight/composition. Inclusion criteria: subjects per treatment arm ≥5; ≥1 week intervention; a reported outcome of body weight/body composition; the intervention was either a prescribed amount of over- or underfeeding and/or supervised or monitored physical activity was prescribed; ≥80% compliance; and an objective method was used to verify compliance with the intervention (for example, observation and electronic monitoring). Data were independently extracted and analyzed by multiple reviewers with consensus reached by discussion. We compared observed weight change with predicted weight change using two models that predict weight change accounting only for metabolic compensation.Findings:Twenty-eight studies met inclusion criteria. Overfeeding studies indicate 96% less weight gain than expected if no compensation occurred. Dietary restriction and exercise studies may result in up to 12-44% and 55-64% less weight loss than expected, respectively, under an assumption of no behavioral compensation.Interpretation:Compensation is substantial even in high-compliance conditions, resulting in far less weight change than would be expected. The simple algorithm we report allows for more realistic predictions of intervention effects in free-living populations by accounting for the significant compensation that occurs.
    International journal of obesity (2005) 10/2014; 39(8). DOI:10.1038/ijo.2014.184 · 5.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We evaluated whether the obesity-associated years of life lost (YLL) have decreased over calendar time. We implemented a meta-analysis including only studies with two or more serial body mass index (BMI) assessments at different calendar years. For each BMI category (normal weight: BMI 18.5 to <25 [reference]; overweight: BMI 25 to <30; grade 1 obesity: BMI 30 to <35; and grade 2-3 obesity: BMI ≥ 35), we estimated the YLL change between 1970 and 1990. Because of low sample sizes for African-American, results are reported on Caucasian. Among men aged ≤60 years YLL for grade 1 obesity increased by 0.72 years (P < 0.001) and by 1.02 years (P = 0.01) for grade 2-3 obesity. For men aged >60, YLL for grade 1 obesity decreased by 1.02 years (P < 0.001) and increased by 0.63 years for grade 2-3 obesity (P = 0.63). Among women aged ≤60, YLL for grade 1 obesity decreased by 4.21 years (P < 0.001) and by 4.97 years (P < 0.001) for grade 2-3 obesity. In women aged >60, YLL for grade 1 obesity decreased by 3.98 years (P < 0.001) and by 2.64 years (P = 0.001) for grade 2-3 obesity. Grade 1 obesity's association with decreased longevity has reduced for older Caucasian men. For Caucasian women, there is evidence of a decline in the obesity YLL association across all ages.
    Obesity Reviews 06/2014; 15(8). DOI:10.1111/obr.12191 · 8.00 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We sought to evaluate whether residence at high altitude is associated with the development of obesity among those at increased risk of becoming obese. Obesity, a leading global health priority, is often refractory to care. A potentially novel intervention is hypoxia, which has demonstrated positive long-term metabolic effects in rats. Whether or not high altitude residence confers benefit in humans, however, remains unknown. Using a quasi-experimental, retrospective study design, we observed all outpatient medical encounters for overweight active component enlisted service members in the U.S. Army or Air Force from January 2006 to December 2012 who were stationed in the United States. We compared high altitude (>1.96 kilometers above sea level) duty assignment with low altitude (<0.98 kilometers). The outcome of interest was obesity related ICD-9 codes (278.00-01, V85.3x-V85.54) by Cox regression. We found service members had a lower hazard ratio (HR) of incident obesity diagnosis if stationed at high altitude as compared to low altitude (HR 0.59, 95% confidence interval [CI] 0.54-0.65; p<0.001). Using geographic distribution of obesity prevalence among civilians throughout the U.S. as a covariate (as measured by the Centers for Disease Control and Prevention and the REGARDS study) also predicted obesity onset among service members. In conclusion, high altitude residence predicts lower rates of new obesity diagnoses among overweight service members in the U.S. Army and Air Force. Future studies should assign exposure using randomization, clarify the mechanism(s) of this relationship, and assess the net balance of harms and benefits of high altitude on obesity prevention.
    PLoS ONE 04/2014; 9(4):e93493. DOI:10.1371/journal.pone.0093493 · 3.23 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Aim: Body composition changes among elite athletes may influence competitive performance. This study aimed to characterize the body composition changes at the molecular, cellular, tissue, and whole-body level of analysis in elite junior basketball players during the course of a season. Methods: Twelve males and 11 females (16 to 17 years) were evaluated. Dual-energy X-ray absorptiometry (DXA) was used to assess bone mineral (Mo) and lean-soft tissue (LST). Total-body water (TBW) and extracellular water (ECW) were assessed using isotope dilution techniques, and extracellular (ECF) and intracellular fluids (ICF) were calculated. Fat mass (FM) and fat-free mass (FFM) were assessed with a four-compartment model. Body cell mass was calculated (LST - (ECF + ECS)). Skeletal muscle (SM) was estimated using appendicular LST (ALST) as: (1.19 x ALST) - 1.65. At the whole-body level, weight, sum of 7 skinfolds, and muscle circumferences (Mc) were measured. The handgrip and the countermovement jump tests were used for performance assessment. Results: Males increased FFM (4.4±2.3%), TBW (3.5±4.6%), SM (4.5±2.3%), and arm (3.4±2.7%) and thigh (3.8±3%) Mc. Females increased SM (5.9±4.6%) and arm (3.6±3.8%) and thigh (4±5.2%) Mc and decreased ICF (-9.7±13.6%). FFM components differed from the established values based on cadaver analysis. Both genders increased their performance and associations were found between changes in molecular and whole-body components with performance. Conclusion: In conclusion the season was associated with an improved body composition profile in males and few changes in females.
    The Journal of sports medicine and physical fitness 04/2014; 54(2):162-73. · 0.97 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective State-level estimates of obesity based on self-reported height and weight suggest a geographic pattern of greater obesity in the Southeastern US; however, the reliability of the ranking among these estimates assumes errors in self-reporting of height and weight are unrelated to geographic region. Design and Methods We estimated regional and state-level prevalence of obesity (body mass index ≥ 30 kg/m2) for non-Hispanic black and white participants aged 45 and over were made from multiple sources: 1) self-reported from the Behavioral Risk Factor Surveillance System (BRFSS 2003-2006) (n = 677,425), 2) self-reported and direct measures from the National Health and Nutrition Examination Study (NHANES 2003-2008) (n = 6,615 and 6,138 respectively), and 3) direct measures from the REasons for Geographic and Racial Differences in Stroke (REGARDS 2003-2007) study (n = 30,239). Results Data from BRFSS suggest that the highest prevalence of obesity is in the East South Central Census division; however, direct measures suggest higher prevalence in the West North Central and East North Central Census divisions. The regions relative ranking of obesity prevalence differs substantially between self-reported and directly measured height and weight. Conclusions Geographic patterns in the prevalence of obesity based on self-reported height and weight may be misleading, and have implications for current policy proposals.
    Obesity 01/2014; 22(1). DOI:10.1002/oby.20451 · 3.73 Impact Factor
  • Andrew W Brown · Michelle M Bohan Brown · David B Allison
    American Journal of Clinical Nutrition 01/2014; 99(1):213. DOI:10.3945/ajcn.113.077354 · 6.77 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Obesity is a global public health problem that is linked with morbidity, mortality, and functional limitations and has limited options for sustained interventions. Novel targets for prevention and intervention require further research into the pathogenesis of obesity. Consistently, elevated markers of inflammation have been reported in association with obesity, but their causes and consequences are not well understood. An emerging field of research has investigated the association of infections and environmental pathogens with obesity, potential causes of low grade inflammation that may mediate obesity risk. In this study, we estimate the possible association between Toxoplasma gondii (T. gondii) infection and obesity in a sample of 999 psychiatrically healthy adults. Individuals with psychiatric conditions, including personality disorders, were excluded because of the association between positive serology to T. gondii and various forms of serious mental illness that have a strong association with obesity. In our sample, individuals with positive T. gondii serology had twice the odds of being obese compared to seronegative individuals (p = 0.01). Further, individuals who were obese had significant higher T. gondii IgG titers compared to individuals who were non-obese. Latent T. gondii infection is very common worldwide, so potential public health interventions related to this parasite can have a high impact on associated health concerns.
    Frontiers in Public Health 12/2013; 1:73. DOI:10.3389/fpubh.2013.00073
  • Gordon Fisher · Gary R Hunter · David B Allison
    International Journal of Epidemiology 12/2013; 42(6):1845-8. DOI:10.1093/ije/dyt163 · 9.18 Impact Factor
  • David B Allison
    The British journal of nutrition 10/2013; 111(03):1-3. DOI:10.1017/S0007114513003309 · 3.45 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Four agents - acarbose (ACA), 17-α-estradiol (EST), nordihydroguaiaretic acid (NDGA), and methylene blue (MB) - were evaluated for lifespan effects in genetically heterogeneous mice tested at three sites. Acarbose increased male median lifespan by 22% (P < 0.0001), but increased female median lifespan by only 5% (P = 0.01). This sexual dimorphism in ACA lifespan effect could not be explained by differences in effects on weight. Maximum lifespan (90th percentile) increased 11% (P < 0.001) in males and 9% (P = 0.001) in females. EST increased male median lifespan by 12% (P = 0.002), but did not lead to a significant effect on maximum lifespan. The benefits of EST were much stronger at one test site than at the other two and were not explained by effects on body weight. EST did not alter female lifespan. NDGA increased male median lifespan by 8-10% at three different doses, with P-values ranging from 0.04 to 0.005. Females did not show a lifespan benefit from NDGA, even at a dose that produced blood levels similar to those in males, which did show a strong lifespan benefit. MB did not alter median lifespan of males or females, but did produce a small, statistically significant (6%, P = 0.004) increase in female maximum lifespan. These results provide new pharmacological models for exploring processes that regulate the timing of aging and late-life diseases, and in particular for testing hypotheses about sexual dimorphism in aging and health.
    Aging cell 10/2013; 13(2). DOI:10.1111/acel.12170 · 6.34 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Study-level design characteristics that inform the optimal design of obesity randomized controlled trials (RCTs) have been examined in few studies. A pre-randomization run-in period is one such design element that may influence weight loss. We examined 311 obesity RCTs published between 1 January 2007 and 1 July 2009 that examine d weight loss or weight gain prevention as a primary or secondary end-point. Variables included run-in period, pre-post intervention weight loss, study duration (time), intervention type, percent female and degree of obesity. Linear regression was used to estimate weight loss as a function of (i) run-in (yes/no) and (ii) run-in, time, percent female, body mass index and intervention type. Interaction terms were also examined. Approximately 19% (18.6%) of the studies included a run-in period, with pharmaceutical studies having the highest frequency. Although all intervention types were associated with weight loss (Mean = 2.80 kg, SD = 3.52), the inclusion of a pre-randomization run-in was associated with less weight loss (P = 0.0017) compared with studies that did not include a run-in period. However, this association was not consistent across intervention types. Our results imply that in trials primarily targeting weight loss in adults, run-in periods may not be beneficial for improving weight loss outcomes in interventions.
    Obesity Reviews 09/2013; 15(1). DOI:10.1111/obr.12111 · 8.00 Impact Factor

Publication Stats

17k Citations
1,408.46 Total Impact Points


  • 1995–2015
    • University of Alabama at Birmingham
      • • Department of Biostatistics
      • • Nutrition Obesity Research Center (NORC)
      • • Department of Psychology
      • • Department of Nutrition Sciences
      Birmingham, Alabama, United States
  • 2009
    • Duke University Medical Center
      • Department of Psychiatry and Behavioral Science
      Durham, North Carolina, United States
  • 1994–2005
    • Columbia University
      • College of Physicians and Surgeons
      New York, New York, United States
    • Virginia Commonwealth University
      Ричмонд, Virginia, United States
  • 1999–2003
    • Johns Hopkins University
      • Department of Medicine
      Baltimore, Maryland, United States
  • 2002
    • Laval University
      Quebec City, Quebec, Canada
  • 2001–2002
    • University of Alabama
      Tuscaloosa, Alabama, United States
    • University of Maryland, Baltimore
      Baltimore, Maryland, United States
  • 1993–2001
    • CUNY Graduate Center
      New York, New York, United States
    • St. Luke's Hospital
      CID, Iowa, United States
  • 1995–1999
    • Aurora St. Luke's Medical Center
      Milwaukee, Wisconsin, United States
  • 1993–1999
    • Saint Luke's Hospital (NY, USA)
      New York, New York, United States
  • 1991
    • Hofstra University
      • Department of Psychology
      New York City, NY, United States