The interaction between coping styles with relationship breakdowns, loss, and conflict and the frequency of self-harm thoughts and behaviours: a longitudinal analysis of 21,581 UK adults

Background

Self-harm thoughts and behaviours are significant public health problems and significant risk factors for future death by suicide,^1–3 particularly when repeated.^{4, 5} Relationship difficulties with romantic partners and family members are the most frequent stressors precipitating self-harm,^{6, 7} and diathesis-stress models of self-harm thoughts and behaviours emphasise the importance of these life events as proximally occurring risk factors for self-harm.^{8, 9} However, the vast majority of people who experience a relationship breakdown do not think of death or harming themselves.

According to diathesis-stress models of suicide and self-harm risk, stable and enduring factors such as maladaptive ways of coping with stress form a long-term vulnerability to thinking about and engaging in self-harm (the diathesis).^{8, 10} Acute risk for self-harm is then greatest when stressful life events such as relationship breakdowns and conflict act on the underlying diathesis to exacerbate their impact.^{8, 9} Different coping styles, or the cognitive and behavioural strategies individuals use to deal with stress and adversities,¹¹ have shown differential associations with self-harm risk. Coping strategies which are characterised by avoiding a stressful situation or difficult experience (e.g., using denial, distraction or using drugs or alcohol), so-called ‘avoidance strategies’, have been shown to increase the likelihood of self-harm thoughts and behaviours and other mental health problems,^{12, 13} whilst problem oriented coping (e.g., taking steps to actively make the situation better), emotion-focused coping (e.g., positive reframing), and emotional support from others (socially supported coping) decrease their likelihood.^{14, 15}

However, despite evidence for these direct associations with self-harm thoughts and behaviours, hardly any research has examined interactions between these long-term factors and stressful life events such as relationship dissolution and conflict occurring in closer proximity to self-harm thoughts and behaviours. The COVID-19 pandemic provides a natural experiment through which to explore this topic. It placed additional strains on interpersonal relationships that may have resulted in more frequent relationship breakdowns and conflicts compared to before the pandemic. For example, there were reported increases across the pandemic in relationship dissolution and conflict.^{16, 17} But research testing whether these stressors interact with longer-standing influences to increase or decrease risk for self-harm thoughts and behaviours is lacking. Further, evidence for how these risk factors influence the frequency of self-harm thoughts and behaviours over time is needed, as most studies of self-harm thoughts and behaviours have been insufficiently powered. This is especially important because risk for future death by suicide is especially strong when self-harm thoughts and behaviours are more frequent.⁵

This study directly addresses these needs by examining whether relationship breakdowns, loss and conflict were associated with increasing frequency of self-harm thoughts and behaviours over the course of the second year of the pandemic in the UK. We chose the second year of the pandemic as it presented a period after the initial psychological shock had passed and during which time there were fewer social restrictions than in Year 1, thereby presenting more ordinary patterns of social behaviours. We then explored whether these associations between relationship factors and self-harm thoughts and behaviours were attenuated or exacerbated by four different coping styles: emotion focused, problem focused, socially supported, and avoidant coping. The results have the potential to deepen our understanding of the circumstances in which relationship breakdowns, loss, and conflict lead to self-harm thoughts and behaviours and to make clinical assessments more meaningful.

Methods

Results

Discussion

Using longitudinal data from a large sample of UK adults, we examined whether relationship breakdowns, loss, and conflict increased the frequency of self-harm thoughts and behaviours over the course of the second year of the COVID-19 pandemic. Then, we investigated whether different coping styles, suggested by theoretical models of self-harm risk, interacted with these stressors in relation to these outcomes. All four relationship variables generally increased the risk for more frequent self-harm thoughts and behaviours, as did the use of more avoidant coping strategies (e.g., substance use and denial). Emotion-focused coping (e.g., positive reframing, acceptance, humour) reduced the risk for more frequent self-harm thoughts and behaviours. However, results from interactions between relationship breakdowns, loss, and conflict with these factors provided mixed support for theoretical models; avoidant coping attenuated the impact of predictor variables on self-harm thoughts (but not behaviours).

Amongst the four variables representing relationship dissolution, loss, and conflict increased the frequency of outcomes, associations were stronger for family problems, separation or divorce, and the breakdown of an ‘other’ relationship compared to those for death of a close relative or friend. This is consistent with research showing romantic relationship difficulties and family problems as the most frequent stressors precipitating self-harm.^{6, 7} Stress as a result of the COVID-19 pandemic may have had a ‘spillover effect’ into interpersonal relationships, placing pressure on dyadic and individual coping resources.^{16, 17} The pandemic has also likely placed additional strains on family members and romantic partners to meet social needs that were previously met elsewhere (e.g., in the workplace) when restrictions on face-to-face contact were not in place. Although the use of more avoidant coping strategies increased the risk for the frequency of self-harm thoughts and behaviours, they attenuated risk for self-harm thoughts in interaction with many of the exposure variables, but not for self-harm behaviours. It could be that the use of these coping strategies captured in the scale used, which were relinquishing efforts to deal with the situation, engaging in denial, or using substances to cope, were helpful in the short term for reducing the psychological stress associated with these events and redirecting thoughts away from those stressors. But it is notable that there was no protective association with self-harm behaviours, reinforcing previous research suggesting that such coping strategies are not adaptive in the long-term.

Another key finding was that emotion-focused coping exacerbated the impact of having had a breakdown of an ‘other’ relationship on self-harm thoughts. This was not entirely unexpected, as emotion-oriented coping has been found to be associated with self-harm thoughts and behaviours in adolescents.²⁶ The emotion-focused coping scale used in the current study involved reframing the event more positively, using humour to deal with the situation, and using religious or spiritual beliefs or practices to cope. However, in the context of the COVID-19 pandemic, such strategies may have been insufficient for coping with a stressor as significant as a relationship breakdown, which could have involved having to make plans to relocate or change one’s living circumstances, for which opportunities were limited due to restrictions on household mixing and moving house.

Additionally, we found that when the socially supported coping scale included venting (e.g., saying things to let one’s unpleasant feelings escape), this way of coping increased risk for both outcomes. This is not entirely surprising, given prior work showing that venting negatively impacts mental health.^{12, 22} However, when the two venting items were removed, this way of coping reduced the frequency of both outcomes, and attenuated the impact of several predictors (e.g., divorce or separation, family problem) on outcomes, as predicted by previous literature. ^{12, 22} The remaining four items comprising the socially supported coping scale reflected emotional (e.g., getting comfort and understanding from someone) and instrumental support (e.g., getting help and advice from other people). Inadequate social support in the context of psychological distress is a key aspect of theoretical models of self-harm risk.^{8, 9} In the UK Biobank study, a lack of emotional support explained longitudinal associations between living alone and later death by suicide in men (but not women).²⁷ Conversely, higher emotional support is associated with reduced likelihood of self-harm thoughts.²⁸

This study has several strengths, including the use of a large well-stratified sample and repeated measures of the outcome variables. However, there are also several limitations. First, due to data limitations, we did not include measures of other factors thought to form the diathesis for suicide and self-harm risk, such as impulsivity and proneness to aggression.^{10, 29} Second, although the wording of our self-harm behaviours question was developed to be similar to our self-harm thoughts question, the way the question was phrased (“How often have you been bothered by…”) may have led to underreporting of self-harm behaviours. This is because first-hand accounts of reasons for self-harm suggest the behaviour is often perceived positively and used as a coping mechanism by individuals who self-harm.³⁰ The same can be said of thoughts of self-harm or suicide, which can, like self-harm behaviours, provide an individual experiencing distress with a sense of soothing and self-regulation.³¹ Although well-stratified across major demographic groups and heterogenous, the sample was not representative of the UK general population. We may have therefore failed to capture individuals who were experiencing the most extreme distress. Further, although this study included regular monthly follow-ups, there is still a possibility that some who were excluded were engaging in more severe self-harm thoughts and behaviours and were therefore unable to participate. However, the comparison of excluded vs included participants did not indicate that the former group had experienced more relationship breakdowns or self-harm thoughts and behaviours. Further research is also needed to unpack the heterogeneity within the ‘other’ relationship breakdown category, as question wording prevented us from considering this. Finally, whilst this study gives particular insight into circumstances surrounding relationship breakdowns and self-harm during the COVID-19 pandemic, replication in non-COVID times is needed.

Conclusions

Our findings demonstrate the importance of interpersonal loss and conflict for more frequent self-harm thoughts and behaviours during the second year of the COVID-19 pandemic. Avoidant coping strategies increased the risk for self-harm thoughts and behaviours. However, we found little evidence from statistical interactions to support predictions from theoretical models^{8, 9} that these long-standing factors increase the likelihood of thinking about and engaging in self-harm when these stressful life events occur. Most people who experience breakdowns or conflict in relationships do not think about or engage in self-harm, so ongoing research to understand the factors which attenuate or exacerbate this risk is a research priority.

Declarations

Supplemental Materials