The purpose of this study was to examine whether chronic diseases and pain are associated with recurrence of depressive and anxiety disorders and if so, whether these associations are partly mediated by subthreshold depressive and anxiety symptoms. In this large study with a mean follow up of 2,5 years, 37.8% of people experienced a recurrence, which corresponds reasonably well to recurrence rates found in previous research [1–11]. During follow-up a quarter of all participants had a depression recurrence, while a third of participants with severe or disabling pain had a recurrence. We did not detect an association between chronic disease and recurrence of depressive and anxiety disorders. Severe neck, chest and abdominal pain, a higher number of pain locations and higher pain severity of pain were significantly associated with recurrence of depressive disorder, but not anxiety. Subthreshold depressive symptoms mediated the associations between pain and depression recurrence.
The finding that chronic diseases are not associated with the recurrence of depression or anxiety in this study is in line with previous studies [3, 4, 8, 33]. Iosifescu et al. found that increasing cumulative chronic disease ratings did predict relapse in a treatment trial . Contrary to the measurements used by Iosifescu et al., we had no information on the severity of the chronic diseases and our study was based on a relatively healthy population with on average 0.6 chronic diseases. Also, the present study and the above mentioned studies investigated rather young populations with mean age around forty years [3, 4, 8, 32]. Chronic diseases often progress over time so associations with depression and anxiety recurrence might be different in studies examining older multi-morbid populations.
This is the first observational study to systematically examine the associations between pain symptoms and depression and anxiety recurrence. We found that several pain characteristics are associated with a higher risk of depression recurrence, which concurs with results of Fava et al. who found that higher levels of pain severity predicted relapse of depression in trial subjects . Since all pain locations were found to be borderline significant or significant, the particular location of the pain seems less relevant in predicting depressive disorder recurrence. Our results point to a dose–response relationship, where increasing number of pain locations and increasing pain severity seem to heighten the risk of depression recurrence. Our findings, with subthreshold depression levels largely mediating the associations, indicate that a mutually reinforcing mechanism between pain and depression might exist . In support of such relationship, another study found that change in pain was a strong predictor of subsequent depression symptom severity and vice versa . A vicious circle could arise with negative coping strategies in response to impaired physical and social role functioning caused by both the pain and affective symptoms [47, 48]. Furthermore, evidence for various shared pathophysiological pathways has been found. Neuroimaging studies have shown overlap of the neuronal networks of emotion and pain, particularly in the (pre)frontal cortical regions . Pain may cause changes such as dysregulation of the HPA axis (increasing cortisol levels), and of the autonomic nervous system (increased sympathetic or decreased parasympathetic function) leading to a new depressive disorder episode. It might also be possible that through these same pathways, previous depressive episode(s) have made patients more vulnerable to pain [22, 27, 47, 50–52].
Contrary to what we expected, we did not find a significant relationship between pain and anxiety recurrence. Although the same pathophysiological pathways for depression and pain are mentioned in research on anxiety and pain, associations might be different or they might be linked through depressive disorders. For instance, HPA-axis abnormalities were found to be strongly associated with depressive disorder and co-morbid depression and anxiety, but were not present for most types of anxiety disorders [52, 53]. A cross-sectional study found that depression partly mediated the relationship between pain and anxiety . A previous NESDA study found that anxiety predicted the recurrence of anxiety only, whereas depression predicted the recurrence of both depression and anxiety . Since research on anxiety and co-morbid depressive and anxiety disorders is still limited compared with depressive disorder alone, more research, particularly on anxiety, is needed to further explore such links.
This study shows that pain makes a patient vulnerable to recurrence of depressive disorders. Pain is among the most common reasons to consult a physician, so patients with pain who also have a depressive history are very likely to contact their physician. The physician then has the opportunity to enquire after depressive symptoms and start treatment, if necessary. Previous studies have shown that pain negatively impacts depression treatment and vice versa [27, 55]. The current study suggests that it may be of benefit to prolong depression treatment until full remission, especially for those patients who also experience severe pain symptoms, as they are a risk group to remit into full-blown disorders. Also, specific relapse prevention trials could focus on distinctions between patients with and without pain in order to find new strategies to prevent depression relapse in patients with pain. Some evidence has been found for alleviating chronic pain and depressive symptoms with collaborative care initiatives for chronic pain patients [56, 57]. Collaborative care trials have proven their efficacy on sustained recovery of depressive disorder and continuation and maintenance therapies can reduce recurrence [58, 59]. Tailored collaborative care interventions with maintenance therapies for patients who recovered from a depressive disorder but who are experiencing pain might reduce recurrence risk in this particular subpopulation of vulnerable patients. Based on ever growing evidence on shared neurobiological pathways new effective interventions need to be explored.
Strengths of our study are the large sample size, the prospective design and the systematic investigation of the role of various chronic diseases and pain on the recurrence of both depression and anxiety. Also, we used full diagnostic interviews to assess the presence of depressive and anxiety disorders with questions on the recency of episodes which enabled us to assess time to recurrence more accurately. This study also had some limitations. The prevalences of specific chronic diseases (myocardial infarction, epilepsy etc.) were quite low, so our capacity to study the role of all diseases individually was limited. This study was conducted with patients of 18 to 65 years old; findings could be different in populations with younger individuals or elderly people. We analysed depressive and anxiety disorders separately, but did not analyse differences between specific depressive or anxiety disorders. The mediating variable, depressive symptoms, may partly measure the same construct as the outcome variable. The final limitation is that the statistical mediation model assumes temporal direction of the independent variable preceding the mediator, which precedes the outcome variable, but both pain and subthreshold symptoms were measured at the same assessment.