How Do Risk Factors Work Together? Mediators, Moderators, and Independent, Overlapping, and Proxy Risk Factors

Kraemer, Helena C.; Stice, Eric; Kazdin, Alan E.; Offord, David R.; Kupfer, David J.

doi:10.1176/appi.ajp.158.6.848

Cited by 1,412 publications

(884 citation statements)

References 54 publications

Supporting

Mentioning

852

Contrasting

Unclassified

Order By: Relevance

“…For example, it is possible that these risk processes would operate differently among much older youth (Arria et al, 2008) or urban youth where access to cannabis may be easier. Next, we could not fully explore complex causal or temporal processes (Kraemer et al, 2001), nor could we fully clarify social and/or psychological mechanisms linking adolescent characteristics and cannabis offers: our results are consistent with several plausible explanations, but to which extent each of those possibilities played a role cannot be fully ascertained from these crosssectional data primarily because we could not rule out self-selection. Further, the social context of these cannabis offers was unknown (Moon et al, 1999) as information concerning the source and circumstances were not available.…”

Section: Strengths and Limitationsmentioning

confidence: 64%

See 1 more Smart Citation

Who receives cannabis use offers: A general population study of adolescents

Andreas¹,

Pape²

2015

Drug and Alcohol Dependence

View full text Add to dashboard Cite

Background: Drug use is predicated on a combination of "willingness" and "opportunity". That is, independent of any desire to use drugs, a drug use opportunity is required; be it indirect (i.e., being in a drug-use setting) or direct (i.e., receiving a direct drug offer). However, whether some youth are more likely to encounter such direct drug use opportunities is not fully known. Aims:We examined whether certain characteristics placed adolescents at greater risk for being offered cannabis, after accounting for a number of demographic-, contextual-, interpersonal-, and personal-level risk factors. Methods:We utilized data from a Norwegian school survey (N=19,309) where the likelihood of receiving cannabis offer in the past year was estimated using logistic regression models. Substantive focus was on the individual and combined effects of personal (i.e., delinquency) and interpersonal (i.e., cannabis-using close friend) risk factors. Separate models were fit for middle-and high-school students.Results: Delinquency was a significant risk factor for receiving cannabis offers, as was a cannabis-using best friend. In addition, peer cannabis use increased the risk of cannabis offers mostly for adolescents on the lower delinquency spectrum, but less so for highly-delinquent adolescents. These interaction effects were primarily driven by the middle-school cohort. Conclusions:Cannabis offers were more likely to be extended to youth of certain high-risk profiles.Targeted prevention strategies can therefore be extended to a general profile of younger adolescents with externalizing problems and cannabis-using peers.

show abstract

Section: Strengths and Limitationsmentioning

confidence: 64%

“…These age-related effects were also reiterated when we examined how the risk factors worked together (Kraemer et al, 2001). The results of the interaction models revealed that having cannabis-using best friends primarily placed lower-delinquent youth at risk for cannabis offers.…”

Section: Cannabis Offersmentioning

confidence: 77%

Who receives cannabis use offers: A general population study of adolescents

Andreas¹,

Pape²

2015

Drug and Alcohol Dependence

View full text Add to dashboard Cite

show abstract

“…We have assumed simple linear regression models in our discussion, but more complicated nonlinear alternatives should be considered for binary variables (Collins et al, 1998;Kraemer, Stice, Kazdin, Offord, & Kupfer, 2001). Bootstrap methods can be used with nonlinear models as well as the linear models we have considered in our illustrations.…”

Section: Concluding Commentsmentioning

confidence: 99%

Mediation in experimental and nonexperimental studies: New procedures and recommendations.

Shrout¹,

Bolger²

2002

Psychological Methods

9,276

127

7,247

View full text Add to dashboard Cite

Mediation is said to occur when a causal effect of some variable X on an outcome Y is explained by some intervening variable M. The authors recommend that with small to moderate samples, bootstrap methods (B. Efron & R. Tibshirani, 1993) be used to assess mediation. Bootstrap tests are powerful because they detect that the sampling distribution of the mediated effect is skewed away from 0. They argue that R. M. Baron and D. A. Kenny's (1986) recommendation of first testing the X → Y association for statistical significance should not be a requirement when there is a priori belief that the effect size is small or suppression is a possibility. Empirical examples and computer setups for bootstrap analyses are provided.

show abstract

“…Association of each potential moderator with the predictor obesity status was no more than trivial based on a rule using a phi statistic: 23 phi(age, obesity) ¼ 0.03; phi(sex, obesity) ¼ 0.17; and phi(race, obesity) ¼ 0.09. Thus, these variables qualify for a potential moderator in the sense of Kraemer et al 24 …”

Section: Predictormentioning

confidence: 99%

Depressive mood and obesity in US adults: comparison and moderation by sex, age, and race

et al. 2005

View full text Add to dashboard Cite

Objective: Sustained depressive mood is a gateway symptom for a major depressive disorder. This paper investigated whether the association between depressive mood and obesity differs as function of sex, age, and race in US adults after controlling for socio-economic variables of martial status, employment status, income level and education level. Methods: A total of 44 800 nationally representative respondents from the 2001 Behavioral Risk Factor Surveillance Survey were studied. Respondents were classified as having experienced a depressive mood if they felt sad, blue, or depressed at least for 1 week in the previous month. The depressive mood was operationalized in terms of duration and sustenance, both defined based on number of days with depressive mood: 7 þ and 14 þ days. Age groups were classified as young (18-64 years) and old (65 þ years). Obesity status was classified as: not overweight/obese (BMIo25); overweight (25pBMIo30); obese (BMIX30). Results: Prevalence of prior-month depressive mood was 14.3 and 7.8% for 7 þ and 14 þ days, respectively. Controlling for race and socio-economic variables, both young overweight and obese women were significantly more likely to have experienced depressive mood than nonoverweight/nonobese women. Young overweight, but not obese, men were significantly more likely to have experienced depressive mood than nonoverweight/nonobese men. Young obese women were also significantly more likely to have a sustained depressive mood than nonoverweight/nonobese women. For old respondents, depressive mood and its sustenance were not associated with obesity in either sex. Discussion: The relationship between the depressive mood and obesity is dependent upon gender, age, and race. Young obese women, Hispanics in particular, are much more prone to depressive mood than nonobese women. Future studies testing associations between depression and obesity should be sensitive to the influence of these demographic and socio-economic variables.

show abstract

How Do Risk Factors Work Together? Mediators, Moderators, and Independent, Overlapping, and Proxy Risk Factors

Cited by 1,412 publications

References 54 publications

Who receives cannabis use offers: A general population study of adolescents

Who receives cannabis use offers: A general population study of adolescents

Mediation in experimental and nonexperimental studies: New procedures and recommendations.

Depressive mood and obesity in US adults: comparison and moderation by sex, age, and race

Contact Info

Product

Resources

About