Defining and distinguishing promotive and protective effects for childhood externalizing psychopathology: a systematic review.

PURPOSE: We examine evidence for whether decreases in externalizing behaviors are driven by the absence of risk (e.g., lack of poor housing quality) or the presence of something positive (e.g., good housing quality). We also review evidence for whether variables have promotive (main) effects or protective (buffering) effects within contexts of risks.
METHODS: We conducted a systematic review of longitudinal studies. First, we review studies (n = 7) that trichotomized continuous predictor variables. Trichotomization tests whether the positive end of a variable (e.g., good housing quality) is associated with lower delinquency compared with the mid-range, and whether mid-range scores are associated with fewer problems than the "risky" end (e.g., poor housing quality). We do not review dichotomous variables, because the interpretation of results is the same regardless of which value is the reference group. To address our second aim, we review studies (n = 53) that tested an interaction between a risk and positive factor.
RESULTS: Both the absence of risk and the presence of positive characteristics were associated with low externalizing problems for IQ, temperament, and some family variables. For other variables, associations with low delinquency involved only the presence of something positive (e.g., good housing quality), or the absence of a risk factor (e.g., community crime). The majority of studies that tested interactions among individual and family characteristics supported protective, rather than promotive, effects. Few studies tested interactions among peer, school, and neighborhood characteristics.
CONCLUSIONS: We discuss implications for conceptual understanding of promotive and protective factors and for intervention and prevention strategies.