Domestic Violence During the Pandemic: Prevalence Surge, Neurobiological Impact, Risk Factors, and Intervention Barriers

The COVID-19 pandemic created an unprecedented natural experiment in the epidemiology of intimate partner violence (IPV) and domestic violence (DV). Within weeks of the first global lockdowns in March 2020, the United Nations Population Fund coined the term "shadow pandemic" to describe the explosive increase in reports of violence within households. The convergence of economic destabilization, enforced cohabitation, social isolation, disrupted access to services, and heightened psychological distress created what researchers have characterized as a perfect storm of risk factor amplification for domestic violence perpetration and victimization.

This article examines the pandemic DV surge through a clinical lens, integrating epidemiological data on prevalence changes, the neurobiological mechanisms through which trauma exposure produces psychiatric sequelae, evidence-based treatment modalities for survivors, and the systemic barriers that impeded intervention during the pandemic period. Understanding these dynamics is essential not only for historical accounting but for preparedness planning, as the pandemic revealed profound structural vulnerabilities in domestic violence response systems worldwide.

Domestic violence encompasses intimate partner violence (physical, sexual, psychological, and economic abuse by a current or former partner), child maltreatment, and elder abuse within household settings. The WHO defines IPV as behavior within an intimate relationship that causes physical, psychological, or sexual harm, including acts of physical aggression, sexual coercion, psychological abuse, and controlling behaviors. The DSM-5-TR does not classify DV exposure as a standalone diagnosis but recognizes its consequences under Posttraumatic Stress Disorder (PTSD; 309.81), Acute Stress Disorder, Adjustment Disorders, and the V/Z-code Partner or Spouse Violence, Confirmed (T74.11XA) or Suspected (T76.11XA), which serve as relational and contextual codes critical for clinical documentation.

Pre-pandemic global estimates from the WHO's 2018 multi-country study indicated that approximately 30% of women worldwide had experienced physical and/or sexual violence by an intimate partner in their lifetime, with 12-month prevalence rates of approximately 5–12% in high-income countries and up to 30% in some low- and middle-income countries. These figures provided the baseline against which pandemic changes were measured.

A landmark systematic review and meta-analysis by Piquero et al. (2021), published in the Journal of Criminal Justice, analyzed 18 studies across multiple countries and found a statistically significant increase in domestic violence during stay-at-home orders, with a pooled estimate suggesting an 8.1% increase in DV reports during the initial lockdown period. However, this figure likely represents an underestimate, as the same lockdowns that increased violence simultaneously reduced victims' ability to report it.

Country-specific data revealed dramatic patterns:

• United States: A study by Boserup, McKenney, and Elkbuli (2020) documented increases of 21–35% in DV-related calls to police departments in major metropolitan areas during the first months of lockdown. The National Domestic Violence Hotline reported a 9% increase in contacts between March and May 2020 compared to the same period in 2019, with many callers specifically identifying COVID-19 and quarantine as factors.
• United Kingdom: The charity Refuge reported a 25% increase in calls and contacts in the first week after lockdown was announced, rising to a 65% increase by late April 2020. Killings of women by current or former partners doubled in the first three weeks of lockdown compared to the average rate from the preceding decade.
• France: The Ministry of Interior reported a 30% increase in DV reports within the first 10 days of lockdown.
• China: Reports from Jianli County, Hubei Province, indicated DV reports to police tripled in February 2020 compared to February 2019.
• Australia: Google Trends data analyzed by Berniell and Facchini (2021) showed a significant spike in DV-related search queries, and frontline services reported increases of 40–75% in demand.

A critical epidemiological nuance is the paradox of reduced formal reporting alongside increased actual violence. Many jurisdictions reported decreases in formal DV reports to police and courts during lockdowns, even as helpline calls, emergency department presentations for assault-related injuries, and eventual post-lockdown disclosures surged. Peterman et al. (2020), in a widely cited UN Women working paper, estimated that for every 3 months of lockdown, an additional 15 million cases of IPV could be expected globally. This discrepancy between formal reporting and actual incidence reflects the profound reporting barriers created by continuous cohabitation with the perpetrator, court closures, and reduced access to mandated reporters (teachers, healthcare providers, social workers).

Importantly, the pandemic did not create domestic violence de novo — it amplified pre-existing patterns and pulled new populations into risk. Emerging data suggest that approximately 15–20% of pandemic DV cases involved first-time perpetration, while the majority represented escalation in frequency and severity among already-abusive relationships.

The pandemic amplified virtually every established risk factor for domestic violence perpetration and victimization. These factors operate across biological, psychological, social, and structural levels, consistent with the ecological model of violence endorsed by the WHO and originally proposed by Heise (1998).

Individual-Level Risk Factors

• Alcohol and substance use: Global alcohol sales increased by 20–55% in the early pandemic period depending on jurisdiction. Nielsen data for the US showed a 54% increase in national alcohol sales in the week ending March 21, 2020. Alcohol intoxication is implicated in approximately 40–60% of IPV incidents, functioning as both a proximal risk factor and a disinhibitor of aggressive impulses. The relationship is bidirectional — substance use increases perpetration risk, and victimization increases substance use as a coping mechanism.
• Pre-existing psychopathology: Perpetrators with Antisocial Personality Disorder (ASPD; prevalence 3–5% in men, highly overrepresented among DV perpetrators at 15–30%), Borderline Personality Disorder (BPD), and substance use disorders were at elevated risk during lockdown. Among victims, pre-existing depression, anxiety disorders, and PTSD were exacerbated by entrapment.
• Economic stress and unemployment: Job loss is one of the strongest situational predictors of IPV perpetration. US unemployment reached 14.7% in April 2020 — the highest since the Great Depression. Studies from prior recessions consistently show that male unemployment increases IPV perpetration by 30–50%, mediated through mechanisms including threatened masculine identity, increased time at home, and financial strain.

Relationship-Level Factors

• Enforced cohabitation and reduced exit options: Stay-at-home orders eliminated the temporal separation that many survivors relied upon (partner's work hours, children's school attendance) to access safety resources. Shelter capacity was simultaneously reduced by 30–50% due to social distancing requirements.
• Power and control intensification: Perpetrators leveraged pandemic conditions as new tools of control — restricting access to masks, sanitizer, and medical care; threatening to expose partners to the virus; weaponizing quarantine requirements to justify surveillance.

Community and Societal Factors

• Service disruption: Court system closures led to delays in protection orders. Many jurisdictions reported 50–70% reductions in restraining order filings during the initial lockdown period — not because fewer were needed, but because the mechanisms to obtain them were inaccessible.
• Reduced bystander intervention: Social isolation removed protective factors such as family visits, neighbor observation, and workplace contacts who might notice signs of abuse and intervene.

The psychiatric sequelae of domestic violence are mediated by well-characterized neurobiological mechanisms involving the hypothalamic-pituitary-adrenal (HPA) axis, limbic circuitry, prefrontal regulatory systems, and neuroinflammatory pathways. The pandemic context compounded these mechanisms through the additive stress of pandemic-related fear and uncertainty layered onto the chronic trauma of domestic violence.

HPA Axis Dysregulation

Chronic IPV exposure produces a characteristic pattern of HPA axis dysregulation. Initial trauma exposure activates the stress response via corticotropin-releasing hormone (CRH) release from the hypothalamus, stimulating adrenocorticotropic hormone (ACTH) from the anterior pituitary, and ultimately cortisol release from the adrenal cortex. In acute stress, this cascade is adaptive. However, chronic, unpredictable threat — the defining characteristic of DV — produces allostatic overload, a concept articulated by McEwen (2000). Over time, DV survivors develop a blunted cortisol response to acute stressors (hypocortisolism) alongside elevated basal CRH levels, a pattern also observed in chronic PTSD. This neuroendocrine signature has been specifically documented in IPV-exposed women by studies including those by Griffin et al. (2005), who found that IPV survivors with PTSD showed significantly lower cortisol reactivity compared to traumatized women without PTSD.

Amygdala-Prefrontal Circuit Disruption

Neuroimaging studies of DV-exposed populations reveal structural and functional changes in key neural circuits. The amygdala, the brain's primary threat detection hub, shows hyperactivation in response to threat-related stimuli in IPV survivors, particularly those meeting criteria for PTSD. Simultaneously, the medial prefrontal cortex (mPFC) and anterior cingulate cortex (ACC) — regions responsible for top-down regulation of emotional responses and extinction of conditioned fear — show reduced activation and, in some studies, reduced gray matter volume. Fonzo et al. (2010) demonstrated that women with IPV-related PTSD showed significantly greater amygdala reactivity and reduced ventromedial PFC activation to fearful faces compared to trauma-exposed controls without PTSD.

This circuit-level imbalance produces the clinical phenomenology of PTSD: hypervigilance (tonic amygdala activation), exaggerated startle (brainstem nuclei dysregulation), emotional numbing (prefrontal over-suppression as a compensatory mechanism), and impaired fear extinction (mPFC failure to inhibit conditioned fear responses). In the pandemic context, this neural architecture meant that survivors already primed for threat detection were living in an environment of continuous, inescapable threat — exacerbating neural sensitization.

Neurotransmitter Systems

Multiple neurotransmitter systems are disrupted in DV-related trauma:

• Noradrenergic system: Elevated norepinephrine (NE) from the locus coeruleus drives hyperarousal symptoms. Chronic DV exposure sensitizes the noradrenergic system, producing exaggerated NE release to mild stressors. This is the mechanistic basis for the efficacy of prazosin (an alpha-1 adrenergic antagonist) in treating trauma-related nightmares, with the Raskind et al. (2003) trial demonstrating significant efficacy.
• Serotonergic system: Reduced serotonergic tone, particularly at 5-HT1A receptors, is associated with both trauma vulnerability and depressive comorbidity in DV survivors. SSRIs target this system and remain first-line pharmacotherapy for PTSD.
• Glutamate and the NMDA system: Excessive glutamatergic signaling during trauma exposure facilitates fear memory consolidation in the basolateral amygdala. Emerging research on NMDA receptor modulators (e.g., D-cycloserine as an adjunct to exposure therapy) targets this mechanism.
• Endocannabinoid system: Reduced endocannabinoid tone (particularly anandamide) has been documented in PTSD populations and may contribute to impaired fear extinction and emotional regulation difficulties.
• Oxytocin system: Chronic DV exposure disrupts the oxytocin system, which normally facilitates social bonding, trust, and stress buffering. Reduced oxytocin levels have been found in IPV-exposed women, potentially contributing to difficulties in forming trusting therapeutic relationships — a critical barrier to treatment engagement.

Epigenetic Mechanisms and Intergenerational Transmission

A particularly concerning neurobiological dimension is the epigenetic modification induced by chronic DV exposure. Methylation changes in genes regulating the glucocorticoid receptor (NR3C1), FKBP5 (a co-chaperone of the glucocorticoid receptor complex), and brain-derived neurotrophic factor (BDNF) have been documented in trauma-exposed populations. The landmark work of Yehuda et al. (2015) on intergenerational trauma demonstrated epigenetic changes in offspring of trauma-exposed parents, and analogous mechanisms are implicated in the well-established intergenerational transmission of domestic violence. Children exposed to DV during the pandemic — estimated at 275 million globally even pre-pandemic according to UNICEF — are at risk for epigenetically mediated alterations in stress responsivity that increase their vulnerability to both future victimization and perpetration.

The psychiatric burden of DV exposure is extensive and multi-diagnostic. Clinicians assessing pandemic DV survivors must be prepared for complex comorbidity profiles rather than single-disorder presentations.

PTSD and Complex PTSD

PTSD is the most characteristic psychiatric consequence of DV, with prevalence rates of 31–84% among IPV survivors depending on the population studied and assessment methodology — substantially higher than the 6.8% lifetime prevalence in the general US population (National Comorbidity Survey Replication). The ICD-11 diagnosis of Complex PTSD (CPTSD) — which includes the core PTSD symptom clusters plus disturbances in self-organization (affect dysregulation, negative self-concept, and disturbances in relationships) — may be particularly appropriate for chronic DV survivors. Research by Karatzias et al. (2017) found that approximately 36% of trauma-exposed individuals meeting criteria for a PTSD-spectrum diagnosis were better characterized by CPTSD, with interpersonal trauma being the strongest predictor of CPTSD over standard PTSD.

Major Depressive Disorder

Depression co-occurs with DV exposure at rates of 37–54%, substantially exceeding the 7.8% 12-month prevalence in the general population. The depression associated with DV often has a particular phenomenological profile: pronounced hopelessness, guilt (often perpetrator-instilled), psychomotor retardation, and suicidal ideation. Pandemic isolation likely worsened depressive symptoms by removing the social contacts that serve as protective factors.

Anxiety Disorders

Generalized anxiety disorder (GAD), panic disorder, and social anxiety disorder are comorbid in an estimated 30–50% of DV survivors. During the pandemic, the additive anxiety of viral threat compounded DV-related anxiety, creating layered threat processing demands.

Substance Use Disorders

Approximately 18–25% of IPV survivors develop substance use disorders, often as self-medication for trauma symptoms. This bidirectional relationship complicates treatment, as active substance use impairs engagement with trauma-focused therapies while untreated trauma drives continued substance use. Pandemic-era increases in substance availability and isolation-related coping exacerbated this pattern.

Traumatic Brain Injury

An underrecognized comorbidity is traumatic brain injury (TBI). Estimates suggest that 60–92% of IPV survivors have experienced at least one head injury, with many experiencing repetitive subconcussive and concussive impacts. Valera and Berenbaum (2003) documented that IPV-related TBI severity predicted neuropsychological deficits independent of PTSD. The cognitive sequelae of TBI — including impaired executive function, memory deficits, and processing speed reduction — can be misattributed to depression, PTSD, or malingering if the TBI history is not assessed.

Differential Diagnosis Pitfalls

Several diagnostic errors are common in clinical work with DV survivors:

• Misdiagnosis as Borderline Personality Disorder (BPD): The affective instability, relational disturbance, identity confusion, and self-harm seen in chronic DV survivors overlap substantially with BPD criteria. Herman (1992) argued that many cases diagnosed as BPD are better understood as Complex PTSD secondary to interpersonal trauma. Clinicians should carefully assess trauma history before assigning personality disorder diagnoses.
• Failure to identify coercive control: Psychological and economic abuse without physical violence may not trigger DV screening protocols, yet produces equivalent or greater psychiatric morbidity. The pandemic amplified coercive control mechanisms.
• Somatic symptom presentations: DV survivors frequently present with chronic pain, gastrointestinal disturbance, headaches, and gynecological complaints. Without screening, the DV etiology may be missed entirely.
• Attribution of symptoms to pandemic stress alone: During the pandemic, clinicians may have normalized severe anxiety, depression, and hypervigilance as proportionate pandemic responses rather than recognizing them as indicators of DV-related trauma.

Treatment of DV-related psychological trauma follows the broader evidence base for PTSD and trauma-related disorders while incorporating DV-specific safety considerations. Safety planning must precede and accompany any trauma-focused treatment — a principle that became more complex during lockdowns.

Trauma-Focused Cognitive Behavioral Therapy (TF-CBT)

Cognitive Processing Therapy (CPT) and Prolonged Exposure (PE) are the two most extensively validated trauma-focused CBT modalities. In the IOM (Institute of Medicine, 2008) and APA (2017) Clinical Practice Guidelines for PTSD, both receive the strongest recommendation grade. CPT has been specifically evaluated in IPV populations: Resick et al. (2002) demonstrated that CPT produced PTSD remission rates of approximately 53% at post-treatment and improvements maintained at 5–10 year follow-up. Effect sizes (Cohen's d) for CPT and PE typically range from 1.0–1.5 relative to waitlist controls — large effects by any standard. However, dropout rates in DV populations range from 20–40%, which is higher than in other trauma populations and reflects the ongoing safety concerns and instability characteristic of DV contexts.

Eye Movement Desensitization and Reprocessing (EMDR)

EMDR has Level A evidence for PTSD treatment from multiple guidelines. A meta-analysis by Chen et al. (2015) found EMDR and TF-CBT to be roughly equivalent in PTSD symptom reduction, though head-to-head comparisons specifically in IPV populations are limited. EMDR's advantage may lie in its reduced requirement for homework and detailed trauma narration, which can be impractical or dangerous for clients still cohabitating with perpetrators. During the pandemic, EMDR was adapted for telehealth delivery with emerging evidence of maintained efficacy.

Pharmacotherapy

Pharmacotherapy for DV-related PTSD follows the broader PTSD evidence base:

• SSRIs (sertraline, paroxetine): FDA-approved for PTSD. The SSRI evidence base shows response rates of 50–60% versus 30–40% for placebo, yielding an NNT of approximately 4.5–5 based on meta-analytic data (Stein et al., 2006). Effect sizes are modest (Cohen's d ≈ 0.3–0.5) compared to trauma-focused psychotherapy.
• SNRIs (venlafaxine): Strong evidence for PTSD symptom reduction, with effect sizes comparable to SSRIs.
• Prazosin: For trauma-related nightmares, prazosin showed efficacy in the Raskind et al. (2003, 2013) trials, though the larger PACT trial (2018) produced negative results in a veteran sample, creating debate about its generalizability. Clinical experience suggests continued benefit in selected patients.
• Atypical antipsychotics: Low-dose quetiapine or risperidone are sometimes used adjunctively for severe hyperarousal or dissociative symptoms, but evidence is limited and side-effect burden is significant.

Integrated DV-Specific Interventions

Several interventions have been developed specifically for IPV survivors:

• Helping to Overcome PTSD through Empowerment (HOPE): This cognitive-behavioral intervention, designed specifically for IPV survivors in shelters, showed significant PTSD symptom reduction in a randomized trial by Johnson et al. (2011), with effect sizes of d = 0.67 for PTSD symptoms.
• Strength at Home: A preventive intervention targeting IPV perpetration risk, studied primarily in military populations, with evidence of reduced physical and psychological aggression perpetration at follow-up.

Comparative Effectiveness

Head-to-head comparisons in IPV populations specifically are limited. The broader PTSD literature suggests that trauma-focused psychotherapies outperform pharmacotherapy alone (effect sizes approximately 2–3 times larger), that CPT, PE, and EMDR are roughly equivalent in efficacy, and that combined psychotherapy and pharmacotherapy may offer marginal additional benefit over psychotherapy alone in some populations. The STAR*D-equivalent for PTSD has not been conducted; the closest approximation is the VA/DoD collaborative care research showing that measurement-based, stepped-care approaches improve outcomes.

The pandemic created a unique constellation of barriers to DV intervention that simultaneously increased need and reduced service capacity. Understanding these barriers is essential for preparedness planning.

Shelter and Housing Barriers

DV shelters — a critical first-response resource — faced devastating capacity reductions. Social distancing requirements reduced bed capacity by an estimated 30–50% in many facilities. Some shelters closed entirely during initial lockdown periods. The National Network to End Domestic Violence's annual census documented that on a single day in September 2020, 11,047 requests for DV services went unmet — the majority for housing. Simultaneously, the economic disruption eliminated independent housing options, as unemployment and eviction moratoriums created housing market bottlenecks.

Healthcare System Diversion

Emergency departments — often the primary site of DV identification — pivoted overwhelmingly to COVID-19 response. Routine screening protocols were disrupted. Telehealth visits, while expanding access in some domains, were counterproductive for DV screening because providers could not ensure the patient was in a private, safe location. Studies documented significant reductions in DV identification through healthcare systems during the first pandemic year.

Legal System Disruption

Court closures and delays had immediate safety implications. Many jurisdictions suspended in-person hearings for protection orders. While some adapted to virtual hearings, others experienced backlogs of weeks to months. In some US states, protection order filings dropped by 50–70% during initial lockdowns. For survivors depending on the legal system for safety, this represented a life-threatening gap.

Technology-Facilitated Abuse and the Telehealth Paradox

The rapid shift to telehealth created a paradox for DV survivors. While telehealth expanded access to mental health services in general, for DV survivors cohabitating with perpetrators, conducting therapy sessions from home was impossible or dangerous. Perpetrators also leveraged technology for increased surveillance — monitoring phone calls, tracking internet searches for DV resources, and using smart home devices for eavesdropping. Research by Woodlock et al. (2020) documented the escalation of technology-facilitated abuse during the pandemic.

Barriers for Marginalized Populations

The pandemic amplified pre-existing disparities in DV service access:

• Immigrant populations: Fear of deportation, language barriers, and the shutdown of immigration services created compounding barriers. Undocumented survivors faced the prospect of seeking help from systems they perceived as threatening.
• LGBTQ+ individuals: Forced return to unsupportive family environments (particularly for young adults) and the scarcity of LGBTQ+-affirming DV services were exacerbated.
• Rural populations: Already underserved by DV resources, rural survivors faced additional barriers from reduced transportation options and spotty internet connectivity for telehealth services.
• Disabled individuals: Those dependent on abusive caregivers faced near-total entrapment when home health aides and day programs were suspended.

Children's exposure to DV during the pandemic represents a major public health concern with long-term developmental implications. Pre-pandemic, UNICEF estimated that 275 million children globally were exposed to domestic violence annually. School closures, which affected over 1.5 billion children at peak, eliminated the primary institutional setting for identifying child abuse and DV exposure, as teachers and school counselors are mandated reporters and account for approximately 20% of child maltreatment reports in the US.

The neurobiological impact of DV exposure on children is mediated by developmental timing. Exposure during critical periods of brain development produces lasting alterations:

• Early childhood (0–5 years): DV exposure disrupts attachment formation and HPA axis calibration during the sensitive period when stress response systems are being programmed. Elevated cortisol levels in infancy are associated with hippocampal volume reductions and altered amygdala development, increasing risk for anxiety disorders and PTSD later in life.
• Middle childhood (6–12 years): Exposure affects development of the prefrontal cortex and executive function networks. Children in this age group show elevated rates of conduct problems, academic difficulties, and social withdrawal. Approximately 40–60% of children in DV shelters meet criteria for at least one psychiatric diagnosis.
• Adolescence: DV-exposed adolescents show elevated rates of depression (2–3x baseline), substance use initiation, self-harm, and dating violence perpetration or victimization. The pandemic context of social isolation from peers removed critical developmental supports.

Emerging research documents that pandemic school closures led to a 27–40% decrease in child maltreatment reports to child protective services during lockdown periods — universally interpreted not as a decrease in actual maltreatment but as a catastrophic failure of detection systems. When schools reopened, many jurisdictions saw a surge in reports, confirming this interpretation.

Clinical research has identified several factors that predict recovery trajectories in DV survivors, which is essential for treatment planning and resource allocation.

Factors Associated with Better Outcomes

• Cessation of violence exposure: The single strongest predictor of psychiatric symptom improvement is achieving safety from the perpetrator. Survivors who successfully separated from abusive partners showed 2–3 times greater PTSD symptom improvement over 12 months compared to those who remained in the relationship (Anderson et al., 2003).
• Social support: Strong social networks are consistently associated with better recovery. The pandemic severely undermined this protective factor through enforced isolation.
• Engagement with trauma-focused therapy: Completion of evidence-based treatment (CPT, PE, or EMDR) is associated with PTSD remission rates of 50–70%, substantially better than supportive counseling or pharmacotherapy alone.
• Economic independence: Financial self-sufficiency is associated with lower rates of returning to abusive relationships and better long-term mental health outcomes. The pandemic disproportionately affected women's employment, undermining this factor.
• Absence of TBI: Survivors without significant head injury histories show better treatment response and faster recovery trajectories.

Factors Associated with Poorer Outcomes

• Ongoing violence exposure: Continued contact with the perpetrator — whether voluntary or coerced — is the strongest predictor of chronic symptom course.
• Childhood trauma history: Survivors with adverse childhood experiences (ACEs) in addition to adult DV show more severe and treatment-resistant presentations. The dose-response relationship between ACE score and health outcomes documented in the Felitti et al. (1998) ACE Study applies directly: survivors with ACE scores ≥4 had dramatically worse psychiatric and medical outcomes.
• Complex comorbidity: Co-occurring substance use disorders, TBI, and multiple psychiatric diagnoses predict longer treatment courses and lower remission rates.
• Dissociative subtype PTSD: Survivors with prominent dissociative symptoms (depersonalization, derealization, emotional numbing) may require modified treatment approaches and show slower initial response to standard exposure-based interventions, though ultimate outcomes may be comparable with extended treatment (Wolf et al., 2012).
• Systemic barriers: Immigration status, poverty, housing instability, and racial/ethnic discrimination predict worse access to care and poorer outcomes even when treatment is received. These structural determinants were amplified by the pandemic.

A complete clinical picture requires attention to perpetrator intervention, though the evidence base is notably weaker than that for survivor treatment. The most common perpetrator intervention model is the Batterer Intervention Program (BIP), typically based on the Duluth Model (psychoeducational, feminist-informed) or cognitive-behavioral approaches. The evidence for BIPs is sobering: a landmark meta-analysis by Babcock, Green, and Robie (2004) found small effect sizes (d = 0.09 for the Duluth model; d = 0.12 for CBT-based programs), translating to approximately a 5% improvement over no treatment. Recidivism rates for DV perpetration remain high — approximately 30–60% over 2–5 years following BIP completion, depending on the study and definition of recidivism.

Emerging approaches show more promise:

• Motivational Interviewing (MI)-enhanced programs: Adding MI to BIPs has shown improved treatment engagement and modest reductions in dropout (from approximately 50% to 30–35%).
• Substance use-integrated treatment: Given the high co-occurrence of substance use and DV perpetration, integrated programs addressing both simultaneously show better outcomes than sequential treatment.
• Neurobiological interventions: Preliminary research on oxytocin administration, transcranial direct current stimulation (tDCS) targeting prefrontal cortex function, and pharmacological approaches to reducing reactive aggression (e.g., SSRIs, mood stabilizers) represents an early frontier, but no neurobiological intervention has sufficient evidence for clinical recommendation in DV perpetration specifically.

The pandemic essentially halted most perpetrator intervention programs, as in-person group sessions — the standard delivery format — were impossible during lockdowns. Virtual adaptation of BIPs raised significant safety and confidentiality concerns, as perpetrators might use program content to manipulate victims or might attend sessions from shared living spaces.

The pandemic DV surge has catalyzed several active research frontiers with implications for clinical practice and policy.

Digital and Technology-Based Interventions

The pandemic accelerated development of technology-based DV interventions, including safety planning apps (e.g., myPlan), AI-powered chatbots for crisis support, and text-based hotlines that can be used more covertly than phone calls. The myPlan app, developed by Glass and colleagues at Johns Hopkins, showed efficacy in a randomized trial in helping women make safer decisions about their relationships. However, the digital divide means these innovations disproportionately benefit survivors with smartphone access and digital literacy.

Screening in Healthcare Settings

The pandemic exposed critical gaps in DV screening protocols. The USPSTF recommends screening all women of reproductive age for IPV, but compliance was poor even pre-pandemic and deteriorated further during the pandemic. Research is examining how to embed effective, safe screening into telehealth encounters, including the use of coded language systems and online intake forms that can be completed privately.

Neurobiological Treatment Targets

Advances in understanding the neurobiology of both trauma response and aggression are generating novel treatment targets. MDMA-assisted psychotherapy for PTSD, which received Breakthrough Therapy designation from the FDA based on phase 2 trials by the Multidisciplinary Association for Psychedelic Studies (MAPS), showed PTSD remission rates of 67% at 2-month follow-up in phase 3 trials (Mitchell et al., 2021), though its application specifically to DV-related PTSD has not been studied and raises unique safety and ethical considerations. Stellate ganglion block (SGB), ketamine-assisted therapy, and neurofeedback are additional emerging modalities being investigated for treatment-resistant PTSD.

Limitations of Current Evidence

Several significant limitations constrain the evidence base on pandemic DV:

• Measurement heterogeneity: Studies used vastly different DV definitions, data sources (police reports, hotline calls, surveys, hospital records), and comparison periods, making cross-study comparison difficult.
• Ecological fallacy: Much pandemic DV data is ecological (area-level) rather than individual-level, limiting causal inference.
• Reporting bias: The paradox of reduced formal reporting alongside increased actual violence means that any dataset based on official records underestimates the true increase.
• Limited prospective data: Most pandemic DV research is cross-sectional or retrospective. Prospective longitudinal studies tracking survivors from the pandemic period are underway but results are still emerging.
• Generalizability: The vast majority of DV research focuses on cisgender women in heterosexual relationships. DV in LGBTQ+ relationships, against men, and in non-partner household contexts (elder abuse, child maltreatment) remains understudied.

The pandemic DV experience yields several clinical imperatives for mental health professionals:

• Universal screening: All patients presenting with anxiety, depression, PTSD, somatic complaints, or substance use should be screened for current and lifetime DV exposure, with particular attention to the pandemic period. Validated screening tools include the HITS (Hurt, Insult, Threaten, Scream) scale and the WAST (Woman Abuse Screening Tool).
• Safety-first treatment planning: Trauma-focused treatment should not be initiated until a safety assessment confirms that the patient has a viable safety plan and that treatment activities (homework, between-session emotional activation) will not increase danger. This principle required adaptation during the pandemic and must remain a priority.
• Trauma-informed care as an organizational standard: Healthcare systems should adopt trauma-informed frameworks that assume a high base rate of DV exposure in patient populations and structure all interactions accordingly — from intake procedures to physical examination practices.
• Assessment for TBI: Given the high prevalence of head injury in IPV survivors, neuropsychological screening should be integrated into DV-related clinical assessments. Tools such as the HELPS Brain Injury Screening Tool, adapted for DV populations by Valera, can identify survivors who may benefit from neurological referral.
• Preparedness planning: The pandemic revealed that DV response systems had no pandemic preparedness plans. Healthcare systems, DV services, and governmental agencies should develop protocols for maintaining DV services during future public health emergencies, including digital service delivery models, PPE stockpiles for shelters, and fast-tracked virtual protection order procedures.
• Addressing perpetration risk: Clinicians seeing patients with known DV perpetration histories should assess for pandemic-related escalation and provide referrals to evidence-based intervention where available. This includes assessing for acute risk factors such as recent job loss, increased substance use, access to firearms, and escalating controlling behaviors.

The COVID-19 pandemic laid bare both the pervasiveness of domestic violence and the fragility of the systems designed to address it. The neurobiological, psychological, and social consequences of the pandemic DV surge will reverberate for decades, particularly through the developmental impact on children exposed during critical periods. Clinical and systemic response must match the scale of this impact.