Understanding the link between depression and pain

1.1 Aim

A point of contention in the relationship between depression and pain however, is the chicken-or-the-egg debate: is depression primary or secondary to pain. In this critical review we will examine three relevant questions in light of the latest methodological and theoretical developments reported in the literature:

1. First, does depression actually trigger pain? We examine whether the presence of depression makes the onset of a pain problem more likely.

2. Second, does treating one problem alone result in improvements in the other? For example, does treatment aimed decreasing depression result in less pain and improved function? And, if pain is treated does the depression automatically reside? Such questions have special significance for treating pain problems since depression is often not detected in patients with pain and may also go untreated in the belief that once the pain is reduced so will the depression [1].

3. Third, what is the likely mechanism or pathway by which depression and pain are linked? To this end, we present a new model as a heuristic that may help us to understand better this relationship and to stimulate questions for future research. This may pave the road for developing new treatment strategies.

3.1 Conclusion

Although there is a strong link between the presence of depression and the development of chronic pain, the evidence that depression actually triggers the onset of pain is relatively limited. First, some studies fail to show a relationship at all. Second, the size of the relationship predicting future pain or chronicity is small compared to the effect sizes noted between depression and pain in chronic sufferers. Third, even the best studies have had difficulties pinpointing the exact time point for the onset of the pain, making it impossible to know whether the depression actually preceded the onset. Therefore, while a relationship has been shown, it is not clear that the depression actual causes (e.g. preceded) the onset of the pain. Consequently, the temporal binding between depression and the onset of a pain problem was not clearly demonstrated so that it could be considered as a trigger.

Rather than a trigger, it might be more accurate to view depression as a force that catalyzes pain problems. When depression is present during the early stages of pain, it is clearly linked to a higher probability of the development of longstanding pain problems.

4.1 Pharmacological treatment

The most common treatment for depression in patients with a pain condition is antidepressant drugs. One classic review of this literature found 22 studies where most used tricyclic antidepressants, but the methodological quality of the studies was said to be low [1]. While most of the investigations reported improvements in both pain and depression, there was considerable variability in findings. A recent Cochrane review of the use of antidepressant medication for chronic low back pain found no evidence that these drugs were more effective than placebo [30]. However, the review did not focus on whether depression actually decreased, but instead examined the effects on pain intensity. Although it is widely believed that antidepressants result in pain relief, this statement appears in need of modification. In addition to the Urquhart review above, another recent review for acute and chronic pain concludes that the effects of antidepressants for chronic pain are small [31]. The data are more encouraging for some other types of pain, e.g. rheumatic problems where antidepressants are said to reduce pain symptoms [32]. However, this review also underscores that while antidepressants for pain problems, e.g. fibromyalgia are nearly equivalent to analgesics, both have quite limited effects. Therefore, management can seldom rely entirely on antidepressants. One study deserves particular mention since it examined the treatment of depression in primary care in patients as it demonstrates that those suffering co-morbid pain problems had the least benefit of pharmacological treatment [33]. The results then, of pharmacological treatments of depression in pain patients is quite mixed and there is an apparent lack of evidence demonstrating a consistent and significant effect when depression is treated in patients suffering both pain and depression. Even though pharmacological treatment is often recommended for clinical use, it has limited effects on pain and few documented effects on function.

Interestingly, few studies actually observed whether the depression was affected by the medication as the focus of most studies was on the effect of the pharmacological treatment on pain intensity. Nonetheless, one review of antidepressants for syndromes where both pain and depression were assessed found that pain symptom improvement did not correlate very well with depression improvement [34]. This underscores that we do not understand the relationship in terms of mechanisms. Thus, there is an obvious need to conduct research into how various treatments impact on both depression and pain. This would promote a better understanding of the relationship and provide insights into how the combination of depression and pain might best be tackled clinically.

4.2 Psychological and combined treatments

Cognitive-behavioral therapy (CBT) is an evidence-based treatment for depression [35,36,37] that might be utilized since it also is applied to pain. Psychological treatment is even effective for subclinical levels of depression and may prevent the development of a full blown depression [38]. Clearly, since many pain services report having a cognitive behavioral approach, then it might be possible to offer CBT therapy for depression. However, it is difficult to ascertain to what extent this is done in existing programs since most describe an array of treatment techniques most of which are not specific to depression. Moreover, in multimodal programs there are methodological problems in determining whether improvements might be due to the treatment for depression or instead due to the other treatments provided for pain. Although there may be difficulties in administration, the effects of a more concentrated psychological approach to depression for those suffering pain and depression would provide essential data shedding light on the role of depression in pain. Another approach would be to actively target both the depression and the pain.

One recent study did attempt to look at the effects of addressing both depression and pain problems in primary care patients suffering depression and pain [39]. Patients first received antidepressants (12 weeks) to deal with the depression and then behavioral self-management (6 weeks) to deal with other aspects of the pain. A continuation phase followed that aimed to consolidate the two steps. Results showed significantly fewer participants with depression in the group receiving both treatments in comparison to the Treatment As Usual control group. This also coincided with significantly larger pain improvements as 40% achieve a pain reduction of 30% or more (ES = .5). However, only 38% of the treated group had a decrease in depression of more than 50% and 41% still were deemed to have major depression despite the antidepressant treatment. This study illustrates that a combination of treatment approaches may be one way of achieving better results. At the same time it accentuates the difficulty in obtaining large decreases in depression as well as in pain. Indeed, only 26% of the patients receiving the combined treatment achieved combined benefits in both depression and pain. Further, it utilized medications for depression and it would be of value to know how psychological treatment might be utilized for both depression and pain problems.

A second study of relevance utilized a national survey to examine the effects prospectively of depression treatment for patient with both depression and chronic pain [40]. Receiving specific treatment for depression was associated with improved mental health and less interference of pain on work. These authors conclude that it is important for clinicians to understand that specific treatment for depression can enhance outcomes. Interestingly, a main effect was on function, i.e. being able to return to work. Again, the mechanism for the effects is not known, but future research might elucidate ways to treat depression and pain in an integrative approach.

4.3 Is there a need for early treatment of depression?

Some evidence suggests that it may be critical to specifically address depression in pain treatment programs, particularly early on in treatment. Consider two pertinent Canadian studies [14,16] where initial levels of depression severity were studied in relation to treatment response. Patients with musculoskeletal pain were also assessed on level of depression and then followed with one measurement point each before, during, and after a rehabilitation treatment. The results underscored two things. First, those high on depression severity at the start of treatment were still likely to be depressed at the end of treatment. Second, early interventions that reduced depression were associated with the best return to work rates during the follow up. To be sure, for those with normal scores on the depression scale, 91% returned to work as compared to only 26% for those still having high scores at the termination of the treatment [14]. There is no indication in these studies however, as to what is driving the depression, or how exactly it is influencing the pain and disability; catastrophizing levels are not even reported. Moreover, these studies are based on a limited sample and on only three measurement points; hence they do not fully isolate the process. However, a study from our clinic expands some of the findings as it shows that a failure to obtain improvements in key psychological factors, e.g. depression, catastrophizing, and worry is associated with poorer outcomes [47]. More research is therefore needed to clarify the relationship of depression and pain over the course of treatment, and to elucidate the actual mechanism by which depression may impact on pain.

In summary there is considerable evidence suggesting that there is a need to treat both depression and pain in order to maximize outcome results. Although some studies have shown limited effects using psychopharmacological approaches, cognitive-behavioral therapy might provide an important avenue for addressing the depression. Indeed, since cognitive-behavioral programs for pain already exist these might be united with specific psychological approaches to depression. However, research is direly needed to study whether this can be achieved. Further, it appears that addressing the depression as early as possible in treatment may be advantageous for results of the treatment as well as to prevent the further development of the depressive disorder. Although a good deal of work has been done that focuses on treating depression in patients suffering both depression and pain there is no clear model showing the mechanism by which the one impacts on the other. Hence, while it seems apparent that addressing depression is important, especially early on, our lack of understanding the mechanisms has prohibited the development of a potent treatment package. There is a consequent need to develop models that will stimulate research into the system of how depression and pain impact on each other.

5.1 The possible role of catastrophizing

Catastrophizing is a central concept in models of both depression and musculoskeletal pain. In the cognitive model of depression, for example, catastrophizing is seen as one important form of “distortion” or “cognitive error” [42,43]. In short, catastrophizing distorts reality and leads to negative affect which in turn may amplify the catastrophizing. It is one explanation for the development of negative affectivity that earmarks depression. Therefore, high levels of depression are, in general, associated with high levels of catastrophizing. By comparison, catastrophizing also takes an important spot in the fear and avoidance model of the development of chronic back pain [44,45]. In the fear and avoidance model, catastrophizing is related to enhanced pain, but also to fear, worry, attention and avoidance. Indeed, the research suggests that catastrophizing is related to the development of long-term pain problems [45]. Interestingly, the fear and avoidance model also includes depressed mood where catastrophizing is implicated as a driver of depression as it is placed earlier in the fear-avoidance chain.

The relationship of catastrophizing in depression and pain was underscored in a study of 812 chronic pain patients in a crosssectional survey [18]. The results, while controlling for the possible effects of age, gender and pain duration, showed that catastrophizing accounted for about 40% of the variance in depression severity. Interestingly, the same study found that pain severity and fear of movement were not significant predictors of depression severity. These findings were interpreted to mean that depression in people with chronic pain is related more to cognitions like catastrophizing than to pain severity.

In another study from our clinic, we analyzed data from two samples of pain patients where we could contrast and compare the effects of catastrophizing and depression on pain. We classified participants seeking care for pain as to whether they had high scores (as compared to a clinical norm) on catastrophizing and depression [46]. We found that those with a single elevated factor (only a high level of catastrophizing or alternatively only a high level of depression) had more intense pain and greater dysfunction. However, those with the combination of both high levels of depression and catastrophizing had the highest levels of pain and dysfunction. Indeed, those suffering the combination were as much as 14 times more likely to suffer poor function. Consequently, the combination of catastrophizing and depression was associated with worse outcomes. Another recent study from our group, analyzing data from patients participating in physical therapy treatment, supports these findings. This study indicates that elevated levels of either catastrophizing or depression is associated with a more unfavorable outcome after six months, and elevated levels of both these factors is in some cases associated with the worst outcome in terms of dysfunction [47].

In sum, we see that catastrophizing may play a distinct role in both depression and pain and thus may be one “transdiagnostic” mechanism by which pain and depression impact on each other.

5.2 The possible role of emotion regulation

We suggest that there is also accumulating evidence that points to emotion regulation as a “transdiagnostic” process that ties pain and depression into the same system. Emotion regulation is a theoretical concept that postulates a striving towards emotional balance [48]. Consequently, when situations create intense emotions, this system comes into play to regulate them. Unlike homeostasis which is automatic and controlled by neural systems, emotion regulation involves psychological processes such as covert behavior (e.g. cognitions) and overt behavior. Emotion regulation includes many “coping” strategies such as taking deep breaths, engaging in another activity, relaxation, or distraction techniques. The context is extremely important in determining whether the emotion regulation will be successful. For example, distraction may be a good way of regulating emotions if they are not too intense, but a poor method if the pain is extremely intense. Thus, emotion regulation involves a variety of behavioral “coping” strategies for up-regulation of positive emotions and down-regulation of negative emotions [49]. This system of balance helps us to respond appropriately to emotions, but yet prevent extreme responses such as an emotional outburst.

Pain and depression share a number of interesting similarities concerning the regulation of emotions including mood and negative affect. For example, considerable evidence shows that the experience of pain is influenced by a variety of emotions including anger, frustration, anxiety and mood [50,51]. Further, it is interesting to note that for both depression and chronic pain, problems with emotion dysregulation have been implicated such as difficulties with anxiety and distress [5,6,52]. Indeed, both depression and pain involve regulating recurrent negative affect. In fact, a central psychological risk factor is so-called “emotional distress” [53,54]. This signifies symptoms of being upset and includes depressive symptoms as well as strong physiological reactions associated with anxiety. In other words, negative affect. Emotional distress underscores that emotion regulation is likely to be an important process in the development of chronic pain problems. From a transdiagnostic view pain is also a significant stressor which taxes the emotion regulation system. The next section provides a new model that might help to explain how these mechanisms work.

5.3 The Örebro Behavioral Emotion Regulation Model

We have constructed a model that highlights how emotion regulation may be a central function negotiating the interrelationship between pain and depression. It also emphasizes the role of catastrophizing as a mediator. Fig. 1 illustrates the model for chronic pain.

Fig. 1

The Örebro Model of Behavioral Emotion Regulation for Pain which highlights the role of catastrophizing, negative affect and emotion regulation in relapses of pain and/or depression. Note that there are two vicious circles whereby catastrophizing increases negative emotion and more catastrophizing (pink arrows) increasing the likelihood of relapse and a second which underscores that a relapse is linked through learning to the trigger and in turn linked to emotion regulation making a relapse more probably in the future.

The Örebro Behavioral Emotion Regulation Model illustrates how pain and mood may trigger a flare-up that in the long run acts as a central factor maintaining chronic disability. To understand this, let us begin with the interesting fact that both chronic pain and depression are characterized by a recurrent, cyclic nature [55,56,57]. Hence, there are periods when the patient feels relatively well, but also new episodes or flare-ups of depressed mood and/or pain. To be sure, these are frustrating events that may trigger negative feelings since these tend to occur when the patient believes that (s)he has been cured. In our model, this cyclic pattern is viewed as a crucial part of the development and maintenance of long-term musculoskeletal pain. The flare-ups represent an episode of either negative mood (i.e. an increase of negative mood or a decrease of positive mood) or pain (i.e. increased pain intensity). As the model shows, the flare-up reactivates catastrophic worry from earlier experiences and this in turn enhances the negative affect and perception of pain quality and intensity. Unlike a jolt of pain or emotion, this flare-up has an undetermined duration that is not momentary but extended in its nature. It might go on for a day or a week. The flare-up strains the emotion regulation system. The success or failure of the emotion regulation strategies is defined by outcome, that is, whether the individual is able to proceed with goal-oriented behavior and maintain activity levels despite the flare-up. If, on the contrary, the individual becomes preoccupied with negative emotions and catastrophic worry and is unable to pursue goals and maintain activities then the emotion regulation strategies would be considered unsuccessful. When emotion regulation is successful, difficult emotions are thus dealt with by employing various coping methods and no relapse occurs. However, if successful emotion regulation is not achieved then negative affect develops that may in turn trigger significant depression and pain. The figure illustrates how a vicious circle of catastrophic worry and negative emotions along with unsuccessful attempts at regulating these emotions is initiated (shown by the grey circle in Fig. 1), which in turn might lead to a full relapse. This is related to the long learning history the patient has where these emotions are tied to catastrophizing on the one hand and more pain and negative affect on the other. A vicious circle of catastrophic worry, negative affect and emotion dysregulation results then in increasing levels of distress, pain, and ultimately a relapse. The figure underscores that relapse may be associated with numerous attempts at regulating emotions. It is not simply one single unsuccessful attempt at regulation emotions but rather the inflexible use of unsuccessful emotion regulation behaviors that results in unsuccessful regulation and ultimately a relapse. This describes a significant part of the process of chronification. Emotion regulation consequently plays a central role in negotiating the volatile affective situation created by the flare-up.

A singular feature of our model is that the learning that fuels this process coincides with the recurrent nature of pain and depression. The learning featured in the model creates a system that helps perpetuate the connections in the model. That is, when behavioral emotion regulation is successful, this is reinforced by being able to maintain activity levels. Therefore, mood and pain flare-ups do not become powerful stimuli that trigger catastrophic worry. This may be the typical case for most people suffering various pain problems. On the other hand, when behavioral emotion regulation is not successful, the association between the trigger stimulus (mood and pain) is strengthened and a “vicious circle” may develop with negative emotion regulation and relapses leading to more flareups. This means that in the future it is more likely that mood or pain flare-ups will trigger the catastrophic worry and tax behavioral emotion regulation. It also means that in the future, the emotion regulation strategies may be given up more quickly since the person has learned that they are nevertheless unsuccessful. Taken together these help perpetuate the vicious circle and catalyze the problem towards chronicity. Our model then, highlights the role of emotions and their regulation. Rather than viewing depressed mood or pain as an outcome our model features them as important stimuli that may trigger emotional responses like catastrophizing that in turn increase the likelihood of increased disability.

Our model suggests then, that emotion regulation is a mechanism by which depression and pain may impact on each other. Moreover, it provides mediators for this process, e.g. catastrophizing. Although the cyclic nature of pain and depression are well known by clinicians and epidemiologists, this has seldom been utilized in models. In our model these cycles are important since they stress the emotion regulation system. It also explains why other factors like stress at the workplace might make emotion regulation even more difficult. Finally, the model implies that assessment and treatment might target the emotion regulation system. For example, one important aspect of the system is how catastrophic thoughts are dealt with. Early assessment would seem essential for identifying the problem but also for isolating how emotion regulation might be enhances. This might provide new insights into how depression and pain might be addressed in the same treatment.