The Development of Mood Repair Response Repertories: I. Age-Related Changes Among 7- to 14-Year-Old Depressed and Control Children and Adolescents

Abstract

The purpose of this study was to test developmentally informed hypotheses about regulatory responses to sadness that attenuate versus exacerbate it (adaptive versus maladaptive mood repair responses, respectively) across late childhood, early adolescence, and mid-adolescence. In a multi-site study in Hungary, clinic-based, 7- to 14-year-olds with Diagnostic and Statistical Manual of Mental Disorders’ (4th ed., text rev.) depressive disorders (N = 697; 55% male) and age/sex matched (at 1:2) nondepressed, school-based controls (N = 1,394) reported on their usual responses to sadness/dysphoria; parental reports were obtained separately. Adaptive and maladaptive response repertoire scores were compared across ages within and across subject groups, and by informant, controlling for confounds. Contrary to Hypothesis 1, older (vs. younger) youths in both groups reported fewer adaptive regulatory responses. Maladaptive response repertoires were unrelated to age among controls but significantly increased with age among depressed youths, particularly the girls. Partially supporting Hypothesis 2, subject groups differed in age-related trajectories of mood repair repertories, but not as expected (e.g., younger depressed children reported larger adaptive response repertoires than did controls). Parental reports revealed no developmental changes in offspring’s mood repair repertories. Parent-offspring reports were most discordant for younger (vs. older) offspring, tended to converge around age 11, and were consistently and significantly larger in the depressed sample. Self-reported adaptive mood repair repertories appear to have been laid down by late childhood and then undergo “trimming” across ages 7–14 years. The extensive maladaptive mood repair response repertoires of depressed youths, which increased with age, distinguish them primarily from controls. Therefore, reducing maladaptive regulatory responses to sadness should be a priority when treating depressed youths.

Funding

This study was supported by National Institute of Mental Health (NIMH) Program Project grant PO 1 MH056193; preparation of this article was supported by NIMH grants MH085722 and MH084938 to Dr. Kovacs. The CDI is published by MHS Inc., from which Dr. Kovacs receives royalties. The other authors report no conflict of interest.

Notes

¹ As articulated by Kraemer et al. (2000), cross-sectional comparisons of age groups will lead to erroneous developmental conclusions if the age groups differ in attributes that are known (or can be expected) to affect the outcome variable. Based on prior work, sex, depression severity, and the presence of comorbid anxiety disorder can influence FAM responses. Therefore, we examined the distribution of these variables across the various ages and their effects on the results. We confirmed that, overall, age is a significant predictor of FAM scores. Although the dramatic increase with age in MALADAPT scores in the DEP group appears to reflect primarily the impact of older female individuals, depression symptom severity and anxiety comorbidity make little or no contribution to this posited developmental trend.

Namely, the sex ratio (in both groups) shifted from about 60% male in the 7- to 12-year-old subset to about 60% female in the 13- to 14-year-old subset. Adding Sex × Group × Age interaction terms when modeling MALADAPT scores, we found a significant main effect of sex (F = 56.60, p < .001) and a Group × Sex interaction (F = 17.76, p < .001): Girls scored higher than boys, and the difference is greater in the DEP sample. Further significant interactions include Sex × Age (F = 12.66, p < .001) and Group × Age (F = 17.98, p < .001): According to estimated slopes, MALADAPT scores significantly increased with age in DEP girls (b = 1.14, SE = 0.19, p < .001), not significantly in DEP boys (b = 0.25, SE = 0.18, p = .2), but did not change in control girls (b = 0.14, SE = 0.14, p = .3) and boys (b = –0.12, SE = 0.13, p = .3).

Focusing on DEP subjects, depression severity (quantified via the CDI) significantly correlated with age and FAM scores (r_CDI-Age = .13, r_CDI-MALADAPT = .63, r_CDI-ADAPT = –.13, all ps ≤ .001). Depression severity was significant when added to the model of MALADAPT that included age, sex, and Age × Sex (F = 359.9, p < .001), but age continued to remain significant (F = 10.47, p = .001). In the entire DEP sample, 36% had a history of anxiety disorder. However, the rate of anxiety disorder did not vary significantly across the ages (χ²_MH = 2.6, p = .1) and was not analyzed further.

Modeling FAM ADAPT scores, we found a significant main effect of sex (F = 4.34, p = .037) and significant Group × Age (F = 6.88, p = .009) and Sex × Group × Age (F = 7.15, p = .008) interactions. Estimated slopes showed that ADAPT scores decreased significantly with age in DEP girls (b = –1.53, SE = 0.27, p < .001) and boys (b = –0.77, SE = 0.26, p = .003); in CONTs, the decrease was significant in boys (b = –0.78, SE = 0.18, p < .001), but not in girls (b = –0.31, SE = 0.20, p = .1). Although depressive symptoms (CDI scores) also affected ADAPT scores (F = 6.91, p = .009), age continued to remain significant (F = 26.31, p < .001).

² Although the initial overall “scope” of mood repair responses appears to have been set, there are age-related changes in the content of mood repair repertoires. For example, examining the group of FAM items that reflect the use of interpersonal processes for mood repair (“I talk to my parents,” “I call my friends”), we find that these scores in CONTs manifest significant linear, F(1, 1375) = 14.10, p < .001, ω² = .01) and quadratic, F(1, 1375) = 10.33, p = .001, ω² = .01, functions. Specifically, interpersonal strategy use declines around age 11 and then picks up again around age 14. Zimmermann and Iwanski (2014) reported similar findings. These trends are likely to mirror a developmental transition from relying on parents for interpersonal support to increasing use of peers to aid in combating sad affect.

Additional information

Funding

This study was supported by National Institute of Mental Health (NIMH) Program Project grant PO 1 MH056193; preparation of this article was supported by NIMH grants MH085722 and MH084938 to Dr. Kovacs. The CDI is published by MHS Inc., from which Dr. Kovacs receives royalties. The other authors report no conflict of interest.