Female homicides in the state of São Paulo, Brazil: a time series analysis from 1980 to 2022.

Pain represents a complex sensory and emotional phenomenon related to actual or potential tissue injury (Raja et al., ). It is one of the most common neurological conditions and a significant source of disability and suffering globally, imposing substantial personal and social burdens (Disease, Injury, & Prevalence, ). Pain patterns and severity can vary widely among individuals. Despite similar levels of tissue damage or neuropathy, different individuals may perceive pain differently, complicating pain management and treatment (Meng et al., ; Ploner, Bingel, & Wiech, ). Therefore, universally accepted objective indicators and further advancements in its fundamental mechanisms are critically needed. The rapid development of neuroimaging techniques, exemplified by magnetic resonance imaging (MRI), has demonstrated the potential for objective measurements of brain anatomical structures and activity patterns that underlie perceptual experiences (Davis et al., , ). Compelling structural MRI (sMRI) studies indicate that gray matter (GM) and white matter (WM) morphology are altered across various pain conditions, including migraine, low back pain, joint pain, and other chronic disorders. These alterations occur in multiple brain regions involved in sensory processing, affective modulation, and cognitive control, such as the somatosensory cortex, prefrontal cortex, cingulate cortex, insula, amygdala, hippocampus, and thalamus (Bhatt et al., ; Seifert et al., ; Selvarajah et al., ). Additionally, functional MRI (fMRI) analyses, a technique to capture changes in functional activity and connectivity within brain regions, have indicated that intrinsic connectivity within and between brain networks involved in pain processing and regulation is disrupted (Galambos et al., ; Mayr et al., ). As such, brain imaging-derived phenotypes (IDPs) provide a valuable framework for elucidating whether alterations in brain structure and function arise as a result of pain or contribute to its development (Messina, Rocca, Goadsby, & Filippi, ; Wang et al., ). This understanding could advance knowledge of the underlying neurobiological mechanisms and facilitate the application of neuroimaging in early prediction, diagnostic categorization, and therapeutic intervention strategies for pain. However, observational studies often face challenges with unmeasured and residual confounding factors that impede the establishment of causal relationships. First, most neuroimaging studies in pain research rely on relatively small, clinically ascertained samples, limiting generalizability and precluding causal inference. Second, conclusions across studies are often heterogeneous, with region-specific alterations varying by pain subtype, duration, and comorbidity profile (Kollenburg et al., ). Moreover, structural and functional imaging markers are frequently examined in isolation, and few studies have systematically integrated multimodal imaging phenotypes within a unified genetic framework. These uncertainties underscore the need for genetically informed approaches capable of disentangling shared genetic liability from directional causal effects across multimodal brain imaging traits and pain phenotypes. To address these challenges, Linkage Disequilibrium Score (LDSC) regression and Mendelian randomization (MR) leverage genome-wide association study (GWAS) summary data to investigate associations between phenotypes. LDSC regression provides an estimate of shared genetic architecture between two phenotypes (B. K. Bulik-Sullivan et al., ). MR analyses strengthen causal inference regarding exposure-outcome associations by minimizing confounding and reverse causality (Burgess, Butterworth, & Thompson, ). Leveraging data from large-scale IDPs GWAS, researchers have uncovered substantial evidence of shared genetic architecture and potential causal relationships between brain imaging features and neuropsychiatric, neurodegenerative, and cardiovascular diseases (Guo et al., ; May-Wilson et al., ; Seyedsalehi et al., ; Yu et al., ). However, such work is still absent for IDPs and pain-related phenotypes.