Negative mood regulation (NMR) expectancies: a test of incremental validity

Abstract

Negative mood regulation (NMR) expectancies are beliefs in one’s ability to initiate behaviors to alleviate a negative mood or induce a positive mood. They are one of many cognitive factors that appear to contribute to individual differences in depressed mood. However, it remains unclear whether NMR expectancies predict depressed mood above and beyond related constructs, and whether they predict depressed mood over a time interval. We investigated the incremental validity of NMR expectancies in the prediction of depressed mood in a sample of 122 adults, 73 of whom were reassessed approximately one year later. At baseline, both NMR expectancies and negative affect (NA) accounted for non-overlapping variance in depressed mood. However, NMR expectancies were the only significant predictor of depressed mood at follow-up, even after controlling for NA and baseline depressed mood. These findings support the incremental validity of NMR expectancies as specific mechanisms involved in maintenance of depressed mood.

Introduction

Depressive symptoms affect a significant number of individuals at some time in their lives (Olfson et al., 2002). Individuals vary in the number, severity, and duration of depressive symptoms (Flett, Vredenburg, & Krames, 1997) and recent evidence indicates that subsyndromal symptoms (i.e., below the diagnostic threshold for Major Depressive Disorder [MDD; American Psychiatric Association, 1994]) are about three times more common than MDD level symptoms (Judd et al., 1998). Moreover, increasing levels of depressive symptoms have been associated with greater psychosocial dysfunction, even when these symptoms fall short of a diagnosis (Lewinsohn, Solomon, Seeley, & Zeiss, 2000). It is therefore of interest to understand variables that may exacerbate and/or maintain subsyndromal depressive symptoms. The purpose of the present study is to assess the extent to which negative mood regulation (NMR) expectancies account for individual differences in depressed mood measured both concurrently and over a one-year interval.

People differ not only in their levels of depressive symptoms, but in how they interpret and regulate these symptoms (Franko et al., 1985, Geisser et al., 2000). These individual differences in cognitive appraisals of depressed mood predict the longevity and severity of the depressive symptoms (e.g., Garnefski, Kraaij, & Spinhoven, 2001). One construct that may partly account for individual differences in depressed mood is the expectancy that some behavior or cognitive process will alleviate a negative mood state, or induce a positive one; these are referred to as NMR expectancies (Catanzaro & Mearns, 1990). People who have higher NMR expectancies believe they can initiate behaviors that will reduce their negative mood and/or induce a positive mood, in contrast to people with lower NMR expectancies. Catanzaro and Mearns (1990) suggested that attempts at negative mood regulation might have long-term consequences for adjustment. For a person with low NMR expectancies, attempts to alter negative moods will be feeble at best, and the negative mood state will continue or become worse. Moreover, failing to regulate these negative moods may serve to confirm expectancies about the inability to do so (Catanzaro & Mearns, 1990).

Since their initial conceptualization, NMR expectancies have been studied in relation to coping, depression, anxiety, and illness (e.g., Catanzaro, 1993, Catanzaro and Greenwood, 1994). NMR expectancies have been shown to predict the presence and severity of depression (Catanzaro et al., 2000, Kirsch et al., 1990). Moreover, higher NMR expectancies have been linked to improved mood, and positive mood influences physical health and emotional health (Cohen, 1996). NMR expectancies are predictors of avoidant coping, dysphoria, and somatic symptoms (Catanzaro and Greenwood, 1994, Flett et al., 1996). People with higher NMR expectancies handle stress better and avoid stressful situations less, perhaps because they feel emotionally equipped to successfully handle them (Catanzaro et al., 2000).

Controversy has arisen over whether factors such as NMR expectancies represent trait-like vulnerability factors that predispose individuals to depressive symptoms and/or episodes, or whether they are state-like phenomena that are simply artifacts of depressed mood and disappear with successful treatment (Ingram, Scott, & Siegle, 1999). To address this issue, the present study employed a longitudinal design, in which NMR expectancies were measured at baseline, along with concurrent depressive symptoms, and over a one-year interval after controlling for baseline depressed mood. If NMR expectancies are simply state-like artifacts of concurrent depressed mood, then they would not be expected to predict depressed mood one year later, particularly after controlling for depressed mood at baseline.

Another issue of concern with constructs such as NMR expectancies concerns the degree to which they evidence specific predictive validity when considered simultaneously with other related constructs. It is often the case that a construct will evidence a relationship to depressed mood, but this relationship will disappear or be considerably attenuated once other related constructs are considered simultaneously. For example, McWilliams, Cox, and Enns (2003) showed that a measure of coping styles predicted significant variance in psychological distress beyond that contributed by demographic and personality variables. Such studies are important to show the incremental validity of constructs hypothesized to predict depressed mood (see Sechrest, 1963 for a discussion of incremental validity).

To address the incremental validity of NMR expectancies for the prediction of both concurrent and future depressed mood, we included a measure of negative affect (NA) in the present study. Watson and Clark (1984) conceptualized NA as a general disposition to experience negative emotional states. Individuals high in NA are more likely to report distress, discomfort, and dissatisfaction over time and regardless of the situation, even in the absence of any overt or objective source of stress. High NA individuals appear to be more introspective and honest with themselves, and to dwell heavily on their failures and shortcomings (Mor & Winquist, 2002). High NA individuals also tend to focus on negative aspects of others and the world in general. Consequently, they have a less favorable view of themselves and other people, and are less satisfied with themselves and life in general (Watson and Clark, 1984, Watson et al., 1988a).

We investigated the incremental validity of NMR expectancies for the prediction of depressed mood. Participants underwent a baseline assessment of NMR expectancies, NA, and naturally occurring levels of depressed mood. Approximately 12 months after the baseline assessment session, participants’ levels of depressed mood were reassessed. We hypothesized that participants with higher levels of both NA and NMR at baseline would exhibit increased depressed mood at baseline and at follow-up than participants lower on these variables. Moreover, we predicted that NMR expectancies would explain specific (non-overlapping) variance in depressed mood, both at baseline and follow-up, even with NA considered simultaneously.