Skip to main content
. Author manuscript; available in PMC: 2024 Feb 1.
Published in final edited form as: J Psychiatr Res. 2022 Dec 24;158:231–244. doi: 10.1016/j.jpsychires.2022.12.025

Table 5.

Risk/Protective Factors Examined in Relation to Eating Disorder Psychopathology, Ranked in Order of Strongest Risk Factor to Strongest Protective Factor

Reference Risk/Protective Factor k N Pooled Effect Size I2 AMSTAR 2 Rating GRADE Rating
Metric Estimate 95% CI p < .05 eOR
Potential risk factors (eOR > 1.00)
Kehayes et al., 2019 Evaluative concerns perfectionism 4 1,586 r 0.16RE (0.10, 0.22) Yes 1.80 32% ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Family/peer modeling of body image and eating disturbances 2 449 r 0.16FE NR Yes 1.80 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Dieting 9 10,190 r 0.15FE NR Yes 1.73 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Kehayes et al., 2019 Concern over mistakes 3 571 r 0.14RE (0.06, 0.22) Yes 1.67 0% ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Body dissatisfaction 12 18,366 r 0.13FE NR Yes 1.61 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Perceived pressure to be thin 5 7,895 r 0.12FE NR Yes 1.55 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Negative affect 12 18,445 r 0.09FE NR Yes 1.39 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Thin-ideal internalization 4 15,128 r 0.08FE NR Yes 1.34 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Impulsivity 4 1,669 r 0.07FE NR Yes 1.29 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Substance use 5 2,972 r 0.07FE NR Yes 1.29 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Kehayes et al., 2019 Socially prescribed perfectionism 2 409 r 0.06RE (−0.04, 0.16) No 1.24 0% ★☆☆☆ ⊕ ◯ ◯ ◯
Kehayes et al., 2019 Personal/parental standards 5 1,418 r 0.06RE (0.01, 0.11) Yes 1.24 0% ★☆☆☆ ⊕ ◯ ◯ ◯
Kehayes et al., 2019 Personal standards 4 1,394 r 0.04RE (−0.06, 0.13) No 1.16 66% ★☆☆☆ ⊕ ◯ ◯ ◯
Stice, 2002 Body mass 11 11,063 r 0.04FE NR Yes 1.16 NR ★☆☆☆ ⊕ ◯ ◯ ◯
Potential protective factors (eOR < 1.00)
Colmsee et al., 2021 Self-esteem 9 3,711 r −0.09RE (−0.14, −0.03) Yes 0.73 61% ★☆☆☆ ⊕ ◯ ◯ ◯

Note. CI = confidence interval; eOR = equivalent odds ratio; OR = odds ratio; RR = risk ratio; RE = random effects model; FE = fixed effects model; NR = not reported.

k represents the number of samples; N represents the number of participants; I2 represents heterogeneity. AMSTAR 2 rating indicates overall methodological quality of the review, where ★☆☆☆ = critically low, ★★☆☆ = low, ★★★☆ = moderate, and ★★★★ = high. GRADE rating indicates quality of evidence for the specific risk/protective factor, where ⊕ ◯ ◯ ◯ = very low, ⊕ ⊕ ◯ ◯ = low, ⊕ ⊕ ⊕ ◯ = moderate, and ⊕ ⊕ ⊕ ⊕ = high.