Subthreshold symptomatology and adjustment disorder

In addition to those who develop diagnosable mental health conditions, a considerable portion of individuals experience subthreshold symptomatology. Subthreshold symptomatology refers to those whose symptoms do not meet criteria for disorder but who do report high levels of symptoms or distress. Self-reported rates of psychological distress can be as high as 42% around 12 months following bushfire exposure.^{Reference McFarlane, Clayer and Bookless9} Rates of subthreshold PTSD have been found to be double those of full PTSD in the early aftermath of earthquake (28.0%, 13.8% respectively^{Reference Çorapçioğlu, Tural, Yargiç and Kocabaşoğlu10} and 19.0%, 10.3%^{Reference Lai, Chang, Connor, Lee and Davidson11}) and terror attack exposure (15.7%, 4.8%^{Reference Motreff, Baubet, Pirard, Rabet, Petitclerc and Stene12}) and in the 12 months following terror attack exposure (17.5%, 9.3%^{Reference Chen, Salim, Rodriguez, Singh, Schechter and Dasaro13}). Although PTSD and MDD are the most frequently studied clinical outcomes in disaster literature,^{Reference Messiah, Vaiva, Gokalsing, Tempier, Jean and Acuna14} research suggests that a broad range of subthreshold symptoms may be the most prevalent mental health outcome following disaster.

Examining the occurrence and course of subthreshold symptomology in the aftermath of disaster is important, as such symptoms are associated with many negative psychological outcomes in both trauma-exposed and non-trauma-exposed populations, including reduced quality of life,^{Reference Rodríguez, Nuevo, Chatterji and Ayuso-Mateos15} increased suicide risk^{Reference Marshall, Olfson, Hellman, Blanco, Guardino and Struening16} and social and occupational impairment.^{Reference Zlotnick, Franklin and Zimmerman17} Of critical concern is research that demonstrates subthreshold symptomology to be associated with an equal degree of dysfunction and functional impairment when compared with both MDD^{Reference Kessler, Zhao, Blazer and Swartz18,Reference Howland, Schettler, Rapaport, Mischoulon, Schneider and Fasiczka19} and PTSD.^{Reference Zlotnick, Franklin and Zimmerman17} This again speaks to the importance of directing research and clinical attention towards understanding subthreshold symptomology following disaster exposure.

In recognition of the importance of subthreshold psychopathology, DSM-5 has reclassified adjustment disorder to sit alongside PTSD in the ‘Trauma and stressor-related disorders’ chapter. In response to a psychosocial stressor, symptoms of PTSD or major depressive episode (MDE) that do not meet their respective DSM-5 diagnostic thresholds are now directed to adjustment disorder.²⁰ Emerging evidence supports this approach, indicating that adjustment disorder may sit on a continuum of symptom severity between no disorder and other psychiatric disorders.^{Reference O'Donnell, Alkemade, Creamer, McFarlane, Silove and Bryant21}

Post-trauma adjustment disorder research

Despite recognition that adjustment difficulties and subclinical symptoms are an important outcome following trauma, most research exploring adjustment difficulties has not been conducted after trauma. As a result, there is little understanding pertaining to the prevalence, trajectories and risk factors of adjustment difficulties in the aftermath of exposure to potentially traumatic events such as disasters, nor evidence pertaining to effective interventions for those with these difficulties.^{Reference Bachem and Casey22–Reference O'Donnell, Agathos, Metcalf, Gibson and Lau24} There is little understanding of what potential risk factors implicate an individual's trajectory towards adjustment difficulties versus a more severe disorder following trauma, although one study found physical proximity to the traumatic event to differentiate a diagnosis of PTSD from adjustment disorder.^{Reference Mahat-Shamir, Ring, Hamama-Raz, Ben-Ezra, Pitcho-Prelorentzos and David25}

In one of the few longitudinal studies to date, O'Donnell and colleagues (2016) examined adjustment disorder in a sample of traumatic injury survivors (i.e. motor vehicle accident, fall, assault), with prevalence rates estimated at 19% 3 months post-injury and 16% 12 months post-injury.^{Reference O'Donnell, Alkemade, Creamer, McFarlane, Silove and Bryant21} Those with adjustment disorder (relative to no disorder) at 3 months were found to be 2.7 times more likely to meet criteria for a more severe psychiatric disorder (PTSD, MDD or anxiety) at 12 months, evidencing the importance of adjustment disorder as a potential gateway to more severe disorder over time. Further, the majority of trauma survivors who developed adjustment disorder did so after the 3-month post-injury period, contravening the DSM-5 diagnostic criterion, which requires the onset of adjustment disorder to commence within 3 months of the stressor. This finding presents the possibility that individuals may develop adjustment difficulties well after the occurrence of a traumatic event, when the personal, social or financial consequences associated with the event are realised. This is a particular issue in post-disaster contexts, where acute post-disaster stressors (e.g. displacement, property damage) that may have immediate impacts on mental health^{Reference Neria, Nandi and Galea26} are often compounded by longer-term post-disaster stressors (e.g. ongoing financial concerns, relationship conflicts, insurance claim difficulties) that play additional roles in the course of psychological symptoms.^{Reference Messiah, Vaiva, Gokalsing, Tempier, Jean and Acuna14} Although exposure to a disaster itself may act as a trigger stressor for the development of adjustment disorder, the ongoing financial, health or interpersonal stressors^{Reference Cerdá, Bordelois, Galea, Norris, Tracy and Koenen27,Reference Lowe, Tracy, Cerdá, Norris and Galea28} that ensue in the months following the disaster may play a key role in increasing risk for the persistent course of adjustment disorder.

The Beyond Bushfires study

In January and February 2009, Australia experienced wide-ranging bushfires, with the most severe occurring on 7 February, subsequently referred to as the Black Saturday bushfires. The bushfires caused widespread damage across the state of Victoria, resulting in 173 fatalities, 3500 buildings damaged, including over 2000 homes, and severe impact on community infrastructure.^{Reference Teague, McLeod and Pascoe29} The Beyond Bushfires study is a programme of research undertaken by the University of Melbourne in partnership with community, government and emergency agencies to evaluate the longitudinal mental health and community outcomes in regions of Victoria affected by the fires.^{Reference Gibbs, Waters, Bryant, Pattison, Lusher and Harms30} The study conducted three waves of data collection, across 10 years following exposure to the fires. Mental health outcomes for Wave 1, Wave 2 and Wave 3 have been reported elsewhere.^{Reference Bryant, Waters, Gibbs, Gallagher, Pattison and Lusher31–Reference Bryant, Gibbs, Colin Gallagher, Pattison, Lusher and MacDougall33}

The aim of the present study was to explore the longitudinal trajectory and risk factors of probable adjustment disorder across the 10 years following exposure to the 2009 Black Saturday bushfires. The term ‘probable’ disorder was used because we used simple, self-reported screening instruments (not clinical diagnostic measures) to measure mental health symptoms. This was achieved by examining (a) the prevalence of probable adjustment disorder at each wave; (b) trajectories of probable adjustment disorder over time; and (c) factors associated with the development of probable adjustment disorder 3–4 years after the fires, relative to no psychiatric disorder and probable other psychiatric disorder including PTSD and MDE.

Adjustment disorder as a risk for more serious disorder

Consistent with previous research that found subthreshold symptoms to be more prevalent than psychiatric disorder in the immediate^{Reference Lai, Chang, Connor, Lee and Davidson11,Reference Motreff, Baubet, Pirard, Rabet, Petitclerc and Stene12} and long-term^{Reference Chen, Salim, Rodriguez, Singh, Schechter and Dasaro13} aftermath of disaster, the prevalence rate of probable adjustment disorder at each wave in the current study exceeded rates of probable fire-related PTSD and MDE. Differences in prevalence rates became most prominent 10 years after the fires, when the rate of probable adjustment disorder (18.6%) were approximately three times that of fire-related PTSD (6.4%) and double that of MDE (8.9%). Consideration of probable adjustment disorder in the present study points to the substantially increased mental health burden of disasters, relative to estimates based on prevalence of psychiatric disorders alone.

The Sankey diagram revealed that approximately one-third of those with a probable diagnosis of adjustment disorder at each wave continued to have a probable diagnosis at the subsequent wave. This rate is comparable to that in prior adjustment disorder research in a sample of traumatic injury survivors^{Reference O'Donnell, Alkemade, Creamer, McFarlane, Silove and Bryant21} and emphasises that some people with adjustment disorder may experience persistent distress and impairment, which raises important questions regarding the diagnosis of adjustment disorder as a largely transient condition. Further, this finding underscores the importance of long-term assessment and monitoring of adjustment difficulties in populations exposed to community trauma and disasters.

Building on emerging evidence implicating adjustment disorder as a potential gateway to development of more severe disorders,^{Reference O'Donnell, Alkemade, Creamer, McFarlane, Silove and Bryant21} results showed that probable diagnosis of adjustment disorder predicted a five-fold increase in risk for escalating to a diagnosis of more severe psychiatric disorder (i.e. incident cases of fire-related PTSD and/or MDE). In other words, for a subset of those with probable adjustment disorder, their symptoms accumulated and escalated over time, indicating that adjustment disorder may be an early marker for severe disorder. This finding accords with previous reports that subthreshold PTSD often escalates to PTSD over time^{Reference Bryant, O'Donnell, Creamer, McFarlane and Silove41} and highlights that presence of subthreshold distress may be a useful marker to identify disaster survivors who could benefit from early intervention in order to disrupt this trajectory.

In delineating how adjustment disorder differs from more severe disorder, fearing for one's life during the Black Saturday bushfires emerged as a differential risk factor for probable (PTSD or MDE) psychiatric disorder. Consistent with research that has found fearing for one's safety to be a key predictor of PTSD,^{Reference Bryant and Harvey38} this finding highlights that fear may be a key factor distinguishing those who develop more severe disorder as opposed to adjustment difficulties following trauma.

Risk factors for adjustment disorder

The current study found that stressors related to income and relationship changes were associated with an increased risk of having a probable diagnosis of adjustment disorder. This is largely consistent with post-disaster research linking the experience of ongoing financial, health and interpersonal stressors to negative long-term post-traumatic mental health outcomes.^{Reference Cerdá, Bordelois, Galea, Norris, Tracy and Koenen27,Reference Lowe, Tracy, Cerdá, Norris and Galea28} The conservation of resources (COR) theory^{Reference Hobfoll42} provides a useful framework for understanding how these longer-term stressors may be contributing to post-disaster mental health. The COR theory conceptualises stressors as losses of personal, social and financial resources, and inherent to the theory is the tenet that persistent mental health symptoms are perpetuated by the process of ‘loss spirals’, whereby initial stressors and losses can lead to subsequent stressors in the same or different life domains.^{Reference Hobfoll42} Moreover, the ‘loss spirals’ principle inherent to the COR theory suggests potential reciprocal relationships between stressors and post-disaster mental health, whereby either direction of causal effects between two variables of interest is plausible. When interpreted in light of current findings, this also points to the possibility that negative changes in income and relationships may operate as both a precipitant and a consequence of post-disaster mental health outcomes. This finding points to the important role of service providers in the provision of long-term support when responding to ongoing stressors in the disaster aftermath. Service responses that provide ongoing, long-term practical support to directly target post-disaster stressors, such as resources to assist with income difficulties and/or assisting with securing employment or permanent housing, have the potential to protect against mental health difficulties arising in the immediate and long-term aftermath of a disaster.

Potential interventions for adjustment disorder

The identification of factors associated with the adjustment difficulties may facilitate the targeting of interventions to subclinical presentations. Given emerging evidence suggests that adjustment disorder sits on continuum of symptom severity between no disorder and severe psychiatric disorders,^{Reference O'Donnell, Alkemade, Creamer, McFarlane, Silove and Bryant21} low-intensity interventions may be particularly useful for targeting adjustment difficulties following adversity. A number of low-intensity interventions are gaining an evidence base, particularly Problem Management Plus (PM+)^{Reference Bryant, Schafer, Dawson, Anjuri, Mulili and Ndogoni43} and the Skills for Life Adjustment and Resilience program (SOLAR).^{Reference O'Donnell, Lau, Fredrickson, Gibson, Bryant and Bisson44–Reference Lotzin, Hinrichsen, Kenntemich, Freyberg, Lau and O'Donnell46} Both of these interventions are brief transdiagnostic interventions that can be delivered by non-specialist laypeople in a task-shifting model, which is relevant to post-disaster contexts. SOLAR may be particularly useful because it is trauma-informed and has a module for addressing trauma memories. These low-intensity, brief transdiagnostic interventions may play a key role in mitigating the long-term course of adjustment difficulties in the wake of disaster.

Limitations

Several limitations of the current study need to be acknowledged. The sample was restricted to participants from communities characterised as having suffered high and medium levels of bushfire impact (as defined by evidence of multiple fatalities and significant burning), and thus excluded participants classified as having experienced low impact (no evidence of burning). This is consistent with prior studies of the same data-set, which have used this area-based indicator to operationalise bushfire exposure,^{Reference Bryant, Waters, Gibbs, Gallagher, Pattison and Lusher31,Reference Cowlishaw, Metcalf, Varker, Stone, Molyneaux and Gibbs47} but does not account for individual variability in disaster experiences (including subjective experiences of threat). When this study was initiated, there was lack of brief instruments designed to assess DSM-5 criteria for adjustment disorder,^{Reference Bryant, O'Donnell, Creamer, McFarlane and Silove41} and for consistency over time, the same measures were repeated. Consequently, a concise index of probable adjustment disorder was created using the K-6, and thus the probable diagnosis lacks reliability or validity information. The probable adjustment disorder index was centred on the clinical significance symptoms central to the diagnosis; however, there were limitations in the degree to which the probable diagnosis mapped onto remaining DSM-5 adjustment disorder diagnostic criteria. Specifically, we only excluded probable PTSD and MDE but not other psychiatric disorders (such as generalised anxiety disorder), which may overstate or complicate our estimates of probable adjustment disorder. Probable adjustment disorder and reports of post-fire stressors were both measured at Wave 1 (3–4 years after the fires), and there was no measure of pre-disaster mental health status. We were therefore unable to assess the temporal association of these variables and thus cannot determine at what point the onset of probable adjustment disorder occurred. Moreover, the causal associations between probable adjustment disorder and income and relationship stressors is unclear. This suggests that either direction of effects is plausible, highlighting the likely reciprocal relationships indicative of the described process of ‘loss spirals’ inherent to the COR theory. Last, the recruitment strategy resulted in only 14% of residents in the targeted regions participating in the study, potentially limiting the generalisability of the findings. These participants were more likely to be older, women and better educated (when compared with census data), with such characteristics also associated with varying rates of mental health problems over time.^{Reference Bryant, Waters, Gibbs, Gallagher, Pattison and Lusher31}

Implications

This research speaks to the importance of symptoms that fall below diagnostic thresholds, into the subclinical space. Given our finding that those with probable adjustment disorder are at risk for later disorder, and are particularly vulnerable to financial and relationship stressors, it is essential that disaster responses include a focus on services for those with adjustment difficulties in addition to those with other psychiatric disorders.