Predictive Margins with Survey Data

Graubard, Barry I.; Korn, Edward L.

doi:10.1111/j.0006-341x.1999.00652.x

Cited by 782 publications

(564 citation statements)

References 13 publications

Supporting

Mentioning

556

Contrasting

Unclassified

Order By: Relevance

“…We used logistic regression to estimate relative risks from abuse for BMI 30 and 40. Because odds ratios from logistic regression may overestimate relative risks, we used predictive margins 16 to estimate relative risks from predicted values produced by logistic regression. Standard normal confidence intervals for logrelative risks were calculated using bootstrap methods with 200 replications, 17 then transformed back to the original relative risk scale.…”

Section: Discussionmentioning

confidence: 99%

Body weight and obesity in adults and self-reported abuse in childhood

et al. 2002

View full text Add to dashboard Cite

BACKGROUND:Little is known about childhood factors and adult obesity. A previous study found a strong association between childhood neglect and obesity in young adults. OBJECTIVE: To estimate associations between self-reported abuse in childhood (sexual, verbal, fear of physical abuse and physical) adult body weight, and risk of obesity. DESIGN: Retrospective cohort study with surveys during 1995 -1997. PATIENTS: A total of 13 177 members of California health maintenance organization aged 19 -92 y. MEASUREMENTS: Body weight measured during clinical examination, followed by mailed survey to recall experiences during first 18 y of life. Estimates adjusted for adult demographic factors and health practices, and characteristics of the childhood household. RESULTS: Some 66% of participants reported one or more type of abuse. Physical abuse and verbal abuse were most strongly associated with body weight and obesity. Compared with no physical abuse (55%), being 'often hit and injured' (2.5%) had a 4.0 kg (95% confidence interval: 2.4 -5.6 kg) higher weight and a 1.4 (1.2 -1.6) relative risk (RR) of body mass index (BMI) 30. Compared with no verbal abuse (53%), being 'often verbally abused' (9.5%) had an RR of 1.9 (1.3 -2.7) for BMI 40. The abuse associations were not mutually independent, however, because the abuse types strongly co-occurred. Obesity risk increased with number and severity of each type of abuse. The population attributable fraction for 'any mention' of abuse (67%) was 8% (3.4 -12.3%) for BMI 30 and 17.3% ( 7 1.0 -32.4%) for BMI 40. CONCLUSIONS: Abuse in childhood is associated with adult obesity. If causal, preventing child abuse may modestly decrease adult obesity. Treatment of obese adults abused as children may benefit from identification of mechanisms that lead to maintenance of adult obesity.

show abstract

Section: Discussionmentioning

confidence: 99%

Body weight and obesity in adults and self-reported abuse in childhood

et al. 2002

View full text Add to dashboard Cite

show abstract

“…Slightly more than half of television beer ads appear on sports programming37 , 38 . Of these, a substantial portion appears on ESPN (Nielsen Media Research), a cable network.…”

Section: Methodsmentioning

confidence: 99%

Early Adolescent Exposure to Alcohol Advertising and Its Relationship to Underage Drinking

Collins¹,

Ellickson²,

McCaffrey³

et al. 2007

Journal of Adolescent Health

176

140

View full text Add to dashboard Cite

Purpose-To determine whether early adolescents who are exposed to alcohol marketing are subsequently more likely to drink. Recent studies suggest that exposure to alcohol ads has a limited influence on drinking in mid-adolescence. Early adolescents may be more vulnerable to alcohol advertising effects.Methods-Two in-school surveys of 1,786 South Dakota youth measured exposure to television beer advertisements, alcohol ads in magazines, in-store beer displays and beer concessions, radiolistening time, and ownership of beer promotional items during sixth grade, and drinking intentions and behavior at seventh grade. Multivariate regression equations predicted the two drinking outcomes using the advertising exposure variables and controlling for psychosocial factors and prior drinking.Results-After adjusting for covariates, the joint effect of exposure to advertising from all six sources at Grade 6 was strongly predictive of Grade 7 drinking and Grade 7 intentions to drink. Youth in the 75 th percentile of alcohol marketing exposure had a predicted probability of drinking that was 50% greater than that of youth in the 25 th percentile.Conclusions-Although causal effects are uncertain, policy makers should consider limiting a variety of marketing practices that could contribute to drinking in early adolescence.

show abstract

“…The regressions followed the generalized linear model with log link and gamma distribution, and adjusted for patient characteristics, including demographics (age, sex, and region of residence), CCI, and subtypes of SLE (discoid/ cutaneous lupus and lupus nephritis), during the study period. Based on regression results, the incremental total costs were estimated for each of the GC cohorts, relative to no GC use, such that the expected instantaneous change in costs was a function of a change in GC variables while keeping all other covariates constant (21). Analyses were performed using SAS, version 9.2.…”

Section: Significance and Innovationsmentioning

confidence: 99%

Glucocorticoid Use in Patients With Systemic Lupus Erythematosus: Association Between Dose and Health Care Utilization and Costs

Chen

Choi

et al. 2015

Arthritis Care & Research

View full text Add to dashboard Cite

Objective. To investigate the determinants of health care utilization and costs with use of glucocorticoid (GC) drugs among adult systemic lupus erythematosus (SLE) patients. Methods. This cross-sectional study analyzed established SLE patients identified by International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes from a large US insurance claims database in 2007-2011. Five patient groups were defined by their oral GC use during a 1-year period: non-GC users, <60 days of GC use, and ‡60 days of GC use in low dosage (£7.5 mg/day), medium dosage (>7.5 to £15 mg/day), or higher dosage (>15 mg/day). Annual health care utilization and costs were compared across these groups. Incremental costs of GC groups, calculated as the difference in total health care costs compared with those of the non-GC group, were estimated from multivariable regressions adjusting for demographic/clinical characteristics and stratified by concomitant immunosuppressant use. Results. A total of 50,230 SLE patients were identified (52% non-GC users, 20% <60 days of GC use, and 10% low dose, 10% medium dose, and 8% higher dose of ‡60 days of GC use). GC users had higher health care utilization and costs. Incremental costs were significant (all P < 0.01) for medium-dose ($5,319 and $6,913) and higher-dose ($12,517 and $15,019) GC groups, regardless of concomitant immunosuppressant use. The incremental costs for the low-dose GC group with concomitant immunosuppressants ($1,285; P 5 0.04) were smaller than the incremental costs for the low-dose GC group without concomitant immunosuppressants ($2,514; P < 0.01). Conclusion. GC use, especially at higher doses, was associated with higher health care utilization and costs. Findings in users with concomitant immunosuppressants suggest that therapies with a GC-sparing effect may be associated with lower economic burden in SLE treatment.

show abstract

Predictive Margins with Survey Data

Cited by 782 publications

References 13 publications

Body weight and obesity in adults and self-reported abuse in childhood

Body weight and obesity in adults and self-reported abuse in childhood

Early Adolescent Exposure to Alcohol Advertising and Its Relationship to Underage Drinking

Glucocorticoid Use in Patients With Systemic Lupus Erythematosus: Association Between Dose and Health Care Utilization and Costs

Contact Info

Product

Resources

About