br Table shows the relationships
Table 4 shows the relationships between attachment anxiety, at-tachment avoidance, demoralization, and depression. Both subscales were positively associated with demoralization (models 2a and 3a), with the anxiety subscale showing a large eﬀect (β = 0.52, 95% CI: 0.44 to 0.60) and the avoidance subscale showing a moderate eﬀect (β = 0.36, 95% CI: 0.27 to 0.45). Testing the interaction with symptom burden revealed a significant moderator eﬀect for attachment anxiety, but no evidence that attachment avoidance acted as a moderator be-tween symptom burden and demoralization. Fig. 2 shows the associa-tion of physical symptoms and demoralization for low and high levels of attachment anxiety (A) and attachment avoidance (B).
Repeating all analyses with depression as the dependent variable revealed a small significant eﬀect of attachment security on depression while controlling for age, gender, and symptom count (β = −0.10, 95% CI: −0.19 to −0.02). There was no evidence that attachment security moderated the association of physical symptoms and depres-sion (β = 0.01, 95% CI: −0.07 to 0.09). Interaction plots for depression are presented in the supplement (Fig. 3).
The existential threats of progressive disease may activate attach-ment needs for security and intimacy and the seeking of support to minimize or modulate distress. Those with less attachment security may have less capacity to seek or make flexible use of attachment relation-ships and therefore may be more likely to experience demoralization, a state of existential distress involving perceived pointlessness and meaninglessness. In a study of 382 patients with advanced cancer, we found that attachment security was highly associated with demor-alization and that it 848035-21-2 further moderated the association of physical symptoms with demoralization. Those with less attachment security were more likely to report demoralization in response to greater phy-sical symptom burden. When depression was the dependent variable, the eﬀect of attachment security was small, and no evidence of a moderation eﬀect was found.
Correlations of predictors and dependent variables (N = 382).
Gendera Physical symptom count Attachment security Attachment anxiety Attachment avoidance Demoralization Depression
Results from the multiple linear regression models of the relationship between attachment security, demoralization, and depression (N = 382).
Predictors βa 95% CI p
a Standardized regression coeﬃcients.
Physical Symptom Burden
Fig. 1. Association between physical symptom burden and demoralization for high and low levels of attachment security. Attachment security significantly moderated the association of symptom burden and demoralization.
The present study showed significant associations of attachment anxiety and attachment avoidance with demoralization. This associa-tion was large for attachment anxiety and moderate for attachment avoidance. Previous studies found somewhat lower associations of
attachment security with hopelessness , death anxiety [23,24] and spiritual well-being . We found higher associations between at-tachment anxiety, avoidance and demoralization compared to the pooled associations of attachment anxiety and attachment avoidance with depression and anxiety reported by Nissen et al. . This may be explained by the particular importance of attachment security to sus-taining hope and the sense of purpose in the context of advanced dis-ease. For many patients with severe illness, the meaning of personal relationships is substantially heightened  and spending time with significant others becomes a much more valued goal . Those with less attachment security may fear that their relational needs will not be met and may be less able to obtain felt support, and, for both reasons, be substantially more vulnerable to experience a loss of morale and inability to pursue what is important to them in life. This potential pathway to a loss of meaning and purpose is coherent with the view that feeling alone and isolated, without help available from others can be a significant source of demoralization .
The association between attachment avoidance and demoralization was relatively large in our study population, compared to its association with other psychological outcomes reported in previous studies. To the extent that avoidantly attached individuals find hypertension diﬃcult to tolerate dependence on others, the reality of progressing illness may leave them feeling demoralized and unable to manage . This is consistent with earlier studies in both cancer and non-cancer populations reporting an association of avoidant attachment and suicidal ideation [29,30]. These findings suggest that demoralization may be a link between avoidant attachment and suicidal ideation in the context of perceived threat and need to rely on others. Moreover, the presumed link of attachment avoidance to fears of dependency and to the need for control  is