Introduction
Depression is common in adolescence. It is estimated that 11.7% of 13- to 18-year-olds will experience a major depressive disorder before they reach adulthood (Merikangas et al. Reference Merikangas, He, Burstein, Swanson, Avenevoli, Cui, Benjet, Georgiades and Swendsen2010). Individual CBT is recommended as a first-line treatment for moderate to severe depression in this age group (NICE, 2015). To help deliver CBT effectively, it is important to understand how young people actually experience key features of depression such as rumination. Each individual will have their own experience of rumination, but research can help elicit potential common themes to guide therapists’ questions. Previous research has examined adults’ experience of rumination, but the current study is the first to find out how young people with depression experience rumination and to consider how this might influence CBT in practice.
Rumination involves repetitively thinking about depressive symptoms, the cause of these symptoms and their consequences (Nolen-Hoeksema, Reference Nolen-Hoeksema, Papageorgiou and Wells2004). This abstract thinking style is characterized by global ‘why’ questions about self-worth (Nolen-Hoeksema et al. Reference Nolen-Hoeksema, Wisco and Lyubomirsky2008) and negative comparisons with others (Treynor et al. Reference Treynor, Gonzalez and Nolen-Hoeksema2003). It is thought that rumination is used to cope with negative mood (Lyubomirsky & Nolen-Hoeksema, Reference Lyubomirsky and Nolen-Hoeksema1993) or when unattainable goals cannot be relinquished (Carver & Scheier, Reference Carver and Scheier1981).
In adults, rumination has been associated with depression both cross-sectionally (e.g. Harrington & Blankenship, Reference Harrington and Blankenship2002; Kuyken et al. Reference Kuyken, Watkins, Holden and Cook2006) and over time (e.g. Spasojevic & Alloy, Reference Spasojevic and Alloy2001; Sarin et al. Reference Sarin, Abela and Auerbach2005). While questionnaire-based and experimental studies are key in determining the role of rumination in the onset and maintenance of depression, it is helpful to develop a more nuanced understanding of the process using qualitative methods (Silverstein et al. Reference Silverstein, Auerbach and Levant2006). Pearson et al. (Reference Pearson, Brewin, Rhodes and McCarron2008) interviewed chronically depressed adults about how they experienced rumination. Rumination was reported to be a complex process: usually focused on two topics or life events; often on the past, present and future simultaneously; associated with emotions ranging from hopelessness to feeling trapped; and typically ending through distraction. Drawing on this wealth of research into rumination in depressed adults, rumination-focused CBT has been developed (Watkins et al. Reference Watkins, Mullan, Wingrove, Rimes, Steiner, Bathurst, Eastman and Scott2011).
By contrast, research into rumination in young people with depression remains at an early stage. A direct association between rumination and adolescent depression has been demonstrated in cross-sectional studies, longitudinally (see Rood et al. Reference Rood, Roelofs, Bögels, Nolen-Hoeksema and Schouten2009 for a meta-analysis) and experimentally (e.g. Park et al. Reference Park, Goodyer and Teasdale2004; Hilt & Pollak, Reference Hilt and Pollak2013). However, there have been no qualitative studies to date investigating the phenomenology of rumination in young people with depression. Given the unique cognitive, social and developmental context of adolescence, some aspects of young people with depression's experience of rumination may differ from adults. For example, adolescents’ ongoing cognitive development of executive function may mean they find rumination harder to control than do adults (Blakemore & Choudhury, Reference Blakemore and Choudhury2006). As they prepare for adulthood by developing a social identity (Paus, Reference Paus2005), rumination topics may be more strongly focused on interpersonal relationships. Co-rumination may occur in a different context to adults, involving close friendships (Jose et al. Reference Jose, Wilkins and Spendelow2012) and families (Waller & Rose, Reference Waller and Rose2010) rather than romantic relationships (Calmes & Roberts, Reference Calmes and Roberts2008). The present study addressed these unknowns with the first exploratory investigation of the phenomenology of rumination in depressed youth.
Method
Participants
Seven young people (aged 16–18 years, mean = 16.85, s.d. = 0.90; 2 male, 5 female) with a primary diagnosis of depression took part in the study. They had been receiving CBT at a national and specialist mood disorder clinic in the UK for 1–13 months when interviewed (mean = 6.14 months, s.d. = 4.78). Mood and Feelings Questionnaire (MFQ – Long; Angold & Costello, Reference Angold and Costello1987) scores at initial clinical assessment ranged from 19 to 53 (mean = 33.17, s.d. = 13.75), indicating clinical levels of depression for the majority of participants (cut-off = 29; Daviss et al. Reference Daviss, Birmaher, Melhem, Axelson, Michaels and Brent2006). Primary diagnoses of depression were confirmed for all participants following multi-disciplinary clinical interviews with the young person and their family as part of the initial clinical assessment. No further assessments were carried out before the study interview, but all participants remained in active CBT treatment and their therapist indicated that they still met clinical criteria for depression. Co-morbid secondary diagnoses included obsessive compulsive disorder (n = 2), post-traumatic stress disorder (n = 1), general anxiety disorder (n = 1), social phobia (n = 1) and anorexia nervosa (n = 1). Young people with bipolar disorder, psychosis, a learning disability, or an autism spectrum disorder were excluded.
Ten young people attending the clinic met eligibility criteria and were invited by their therapist to meet with a research assistant to find out more about the study. The therapist and research assistant emphasized that participation or refusal would have no impact on treatment. It was also made clear that all participants would be interviewed by the same research assistant, who was not part of their treatment team and was not known to them. Seven young people (and parents of those aged <18 years) gave consent to participate.
Design
The study used a qualitative, semi-structured interview design. The research protocol was approved by an NHS research ethics committee (12/LO/0264) and local clinical ethics bodies.
Interview
In a semi-structured interview adapted from Pearson et al. (Reference Pearson, Brewin, Rhodes and McCarron2008), the young person was asked to talk about their experience of rumination. Rumination was defined as repetitive, negative thinking when in a low mood, and this definition was written on a visible whiteboard. Participants were asked to describe rumination topics and the associated emotions, sensory images and verbal thoughts. Other questions covered start and stop triggers, co-rumination experiences and the frequency and duration of rumination episodes. The interview ended with a discussion of benign topics. The interview script was amended after piloting with two young people with depression. They recommended that questions about frequency and duration of rumination episodes should be moved to the end of the interview.
Analysis
Interview transcripts were anonymized with pseudonyms and then analysed using interpretative phenomenological analysis (IPA; Smith et al. Reference Smith, Jarman, Osborn, Murray and Chamberlain1999). IPA allows examination of the richness and diversity of participants’ experience and is appropriate for a sample group of seven participants (Smith, Reference Smith2004). Transcripts were analysed individually, in sequence. Emergent themes were clustered into superordinate, master and subordinate levels. Themes were based predominantly on richness of experience and were considered fully supported if mentioned by over half of the sample (Smith, Reference Smith2011).
Reliability was checked with five participants, who gave unanimous approval that the draft list of themes matched their experience. Inter-rating on 50 sample quotes was carried out between the principal researcher and two members of the research team who had not been at the interviews or seen the transcripts. All three inter-raters identified the same subordinate theme on 43 out of 50 sample quotes and at least two inter-raters agreed on the remaining seven quotes.
Results and Discussion
Participants experienced rumination as a disorientating cognitive battle that was emotionally complex and had a significant interpersonal component. Figure 1 shows a visual summary of these themes, with example quotes. A more detailed list of themes, with numbered superordinate, master and subordinate themes is provided in Table 1.
Fig. 1. How young people with depression describe their experience of rumination – summary of themes.
Table 1. How young people with depression describe their experience of rumination – numbered themes
* Italicized themes were mentioned by less than half the participants, so are not considered fully supported (Smith, Reference Smith2011).
Cognitive experience of rumination: a disorientating battle
In line with findings in adult depression studies (Papageorgiou & Wells, Reference Papageorgiou and Wells2003), most young people in the current study expressed a lack of confidence in their ability to fight and control rumination (1.1)Footnote † . Rumination was characterized as an attacking force (1.1.1).
It can be more crushing than other things (Julia, line 610).
It's trying to get out of its little drawer in your head and it's really, like, disruptive (Bryony, line 1494).
The sense of uncontrollability was heightened, for some participants, by the intrusion of emotive memories (1.1.2).
And it's like, kind of, try and not push it out, but it's really hard to, like, especially with images from the past that have a kind of emotional link to you (Tim, line 157).
Positive beliefs about rumination were not endorsed, in contrast with findings for depressed adults (Watkins & Moulds, Reference Watkins and Moulds2005).
Participants described experiencing thoughts related to the past, present and future, as shown in Table 2. Consistent with rumination models (e.g. Treynor et al. Reference Treynor, Gonzalez and Nolen-Hoeksema2003), young people described these thought processes as cyclical (1.2.1).
Table 2. Overview of individual participants’ cognitive experience of rumination
You think ‘Well there's nothing good to look back on. Is there anything good to look forward to?’, and then it leads back into the other two [topics], sort of like, ‘Well, everyone else is so happy, why aren't you?’ (Bryony, line 223).
Two months later, I might still ruminate the same subject (Samantha, line 1428).
Participants also described linear progression (1.2.2) from specific event-related thoughts to abstract and global thoughts
I think, if I feel guilty for doing something, I mean, anything, it just makes you think about everything you kind of feel guilty for (Tim, line 1137).
Overgeneralized autobiographical memory was evident, as with adult studies (e.g. Watkins & Teasdale, Reference Watkins and Teasdale2001).
I will be upset at the [examination] result and therefore angry at myself . . . ‘If I can't do anything, why am I here? I'm worthless and useless and I have no future’ (Liz, line 1126).
Emotional experience of rumination: a complex association with mood
Participants’ emotional experience of rumination was predominantly sad, with elements of anger and anxiety. Specific emotions included despair, helplessness, guilt, fear, paranoia and hatred. No positive emotions were reported.
A complex relationship between rumination and mood emerged, with some participants describing a one-way relationship (2.1) and others perceiving a feedback loop (2.2). Most young people described low mood serving as a trigger for ruminative thoughts (2.1.1):
I'll know like if I'm having a bad day, ’cause I'll just, I'll just be down. Then I'd start thinking then (Melanie, line 633).
This is consistent with the description of rumination as a response to distress (Nolen-Hoeksema, Reference Nolen-Hoeksema, Papageorgiou and Wells2004). For two participants, however, rumination pre-empted a drop in mood (2.1.2).
I'm thinking about more and more things, so it's like piling on top, so, the mood just kind of keeps going down (Liz, line 1073).
Many participants perceived rumination and mood as having a reciprocal relationship (2.2.1).
If I dwell on, think about ‘What's the point?’, I know that I'll end up in a low mood. But then if I'm in a low mood, I can't help myself from what I'm doing, from like, my mind from wandering down that way and then, sort of, reinforcing the low mood (Bryony, line 677).
Behavioural experience of rumination: other people have significant impact
Interpersonal stress is known to influence both depression and rumination (Hammen, Reference Hammen2006; Hilt & Pollak, Reference Hilt and Pollak2013) and ruminators’ social interactions tend to be characterized by friction (Nolen-Hoeksema et al. Reference Nolen-Hoeksema, Wisco and Lyubomirsky2008). Unsurprisingly, then, all the young people in this study reported that unhelpful conversations (3.1) initiated periods of rumination (3.1.1).
While they're shouting at me, they may say something, even sometimes insult me . . . all these negative thoughts come flooding in (Liz, line 955).
Figure 2 summarizes the main start triggers for rumination identified by participants, of which the majority were interpersonal.
Fig. 2. Start triggers for rumination.
Most confirmed that co-rumination with others was a negative experience (3.1.2).
[My sister] was like ‘Oh, our family's awful’ . . . and then we both just kind of carried on talking about the past and feeling bad about it (Tim, line 919).
Most also described themselves as engaging in imagined future conversations or new versions of past conversations while ruminating (3.1.3).
You can see yourself in different scenarios . . . trying to go into the classroom and people's faces or the teacher's surprise (Julia, line 406).
Here, the experience of rumination was suffused with worry about future events.
Many participants described actively seeking help from other people to end rumination bouts (3.2.1), which generally occurred daily and could last from 5 minutes to 2 hours.
If I start thinking negatively, like I don't want to be around anymore, that scares me . . . I get up and go and tell my mum that I'm thinking like that and she'll calm me down (Melanie, line 718).
This could be a positive strategy, but may actually reflect the dependent interpersonal style of ruminators (Nolen-Hoeksema et al. Reference Nolen-Hoeksema, Wisco and Lyubomirsky2008), especially if help is not available (3.2.2).
I'll think, ‘What do I do if my parents aren't around?’, because they've always been there through all this, to get me through it, so I wonder ‘If they're not there, what do I do, what happens?’ (Melanie, line 530).
Only two participants experienced rumination as a process they could end without help from others (3.2.3).
I reach like a breaking point where I'm either going to go really, really, really dark or . . . I'll just sort of pick myself up again slowly (Bryony, line 931).
Stepping back when I'm ruminating or something, I think that I may forget what I'm ruminating about (Sean, line 1112).
Methods used by these participants to stop rumination episodes are shown in Figure 3, along with a summary of other stop triggers.
Fig. 3. Stop triggers for rumination.
Conclusion
Overall, the young people with depression in this study reported experiencing rumination in a similar way to adults with depression. As in Pearson and colleagues’ adult study (2008), a range of negative emotions were experienced, distraction was a stop trigger and rumination focused on distinct topics comprising worry about the future, thoughts about past events and current difficulties with depression. Participants also reported abstract, global thoughts, which have been found to affect emotional processing, problem solving and overgeneral autobiographical memory in experimental studies with adults (e.g. Watkins & Teasdale, Reference Watkins and Teasdale2001; Watkins & Moulds, Reference Watkins and Moulds2005; Watkins, Reference Watkins2008).
This first indication that young people with depression may experience rumination in a broadly similar way to adults has both clinical and research implications. Abstract thinking was evident in several young people in this study, so could be targeted in psychological treatment for young people with depression, consistent with existing third-wave CBT approaches. Several participants described ruminating about the future, so it would be valuable to explore experimentally how worry and rumination, two very similar repetitive thinking processes (Watkins et al. Reference Watkins, Moulds and Mackintosh2005), interact in young people with depression. Developing a better understanding of the interaction between worry and rumination is particularly important because, as with the participants in this study, co-morbid anxiety disorders are prevalent in young people with depression (Garber & Weersing, Reference Garber and Weersing2010).
As with the adult participants in Pearson et al.’s (Reference Pearson, Brewin, Rhodes and McCarron2008) study, most of the young people in the current study described moving and still images as a feature of their rumination episodes. Mental imagery is implicated as a maintaining factor in emotional disorders, including depression, in adults (Patel et al. Reference Patel, Brewin, Wheatley, Wells, Fisher and Myers2007; Holmes et al. Reference Holmes, Lang and Deeprose2009). In young people, Meiser-Stedman et al. (Reference Meiser-Stedman, Dalgleish, Yule and Smith2012) found that intrusive memories were common after negative events among secondary school students and they were associated with subsequent depressed mood. It would be valuable to further investigate the relationship between intrusive memories and rumination. New treatments for depression have been developed that target intrusive memories (Brewin et al. Reference Brewin, Wheatley, Patel, Fearon, Hackmann, Wells, Fisher and Myers2009), and it would be worthwhile exploring the extent to which these treatments reduce both rumination and intrusive memories in young people.
Certain elements of young people's experience of rumination emerged as distinct from adults. There was a particularly vivid sense of being attacked by rumination, with a strong negative emotional impact. Asking questions about experiences of being overwhelmed by rumination may be particularly important when working therapeutically with depressed young people. It may be, for example, that a sense of being overwhelmed by repetitive, negative thoughts makes it hard to react appropriately to other people, contributing to irritability, an attribute associated with depression in young people (Stringaris et al. Reference Stringaris, Maughan, Copeland, Costello and Angold2013). Further studies could explore whether an ability to control rumination, and so avoid being overwhelmed by it, is associated with lower levels of irritability.
The influence of other people on rumination behaviour seemed more prominent for young people in this study than for adults (Pearson et al. Reference Pearson, Brewin, Rhodes and McCarron2008). Questioning and discussion in treatment that pays particular attention to systemic influences may prove helpful when targeting rumination in treatment for young people with depression. In particular, therapists may wish to explore with the young person what types of interaction they have with parents and peers. Waller et al. (Reference Waller, Silk, Stone and Dahl2014) found that young people with depression were more likely than healthy controls to engage peers and parents in negative co-rumination, which can extend depressive episodes. In our study, Tim succinctly described this effect: ‘[My sister] was like “Oh, our family's awful” . . . and then we both just kind of carried on talking about the past and feeling bad about it’ (Tim, line 919). Where possible, therapists could encourage young people with depression to engage in active problem-solving with others rather than co-rumination.
The study had two main limitations. A small sample was used to allow detailed analysis of participants' experience (Smith, Reference Smith2004). However, the gender and age balance of the sample was weighted towards females and late adolescents. This reflects the sample population at the national and specialist CAMHS recruitment site, but the findings may not represent male and younger adolescents' experience of rumination. A further limitation was the length of treatment participants had received prior to interview. Some participants had received almost a full course of CBT, which includes discussions about the rumination process. For these individuals both the ruminative process and their views of it may have been altered. Future studies could recruit participants earlier in treatment and aim for a more balanced gender and age representation. Further improvements for future research might be to use diary methods to supplement retrospective interview responses and re-administer the MFQ or other formal depression measures at the start of the study to support clinical observations that the primary diagnosis on admission remained valid.
In conclusion, the present study provides rich detail about the experience of rumination amongst a small group of depressed young people. Their accounts confirm the repetitive, abstract nature of the process and point to the complex relationship rumination has with a range of emotions. The study provides new insights into the importance of the interpersonal context of rumination and the sometimes overwhelmingly negative emotions it elicits in young people with depression. These aspects now warrant further quantitative investigation with larger samples. CBT-based interventions specifically targeting the ruminative process may be important for young people with depression, particularly interventions which consider the impact of family members or other systemic factors on rumination behaviour.
Summary
To gain a rich understanding of the phenomenon of rumination among adolescents, the present study employed IPA to analyse the transcripts of interviews with seven depressed young people. As a first step in mapping the territory of an under-researched area such as rumination in youth, qualitative methods are useful. The main themes that emerged from the current analysis were that participants experienced rumination as a disorientating cognitive battle; it was perceived to be emotionally complex; and the process was influenced strongly by other people. Interpersonal interaction played a key role in starting and stopping rumination episodes. CBT-based interventions for young people with depression may benefit from a focus on rumination and the effect other people have on starting or stopping a rumination episode.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Acknowledgements
The authors acknowledge the kind assistance given by Victoria Cole and Professor Paul Chadwick during the study.
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Declaration of Interest
None.
Learning objectives
-
(1) To gain an awareness of how rumination is experienced by young people with depression.
-
(2) To increase understanding of the complex interaction between thoughts, feelings and behaviours during young people's rumination episodes.
-
(3) To highlight how important other people can be in prompting and ending young people's rumination episodes.
-
(4) To give practitioners ideas about what to ask young people about when discussing their experience of rumination.
Comments
No Comments have been published for this article.