Our first hypothesis was that different chronic pain conditions are characterized by a common set of psychosocial factors that can be identified by studying the number of coexisting pain sites. Thus, it is believed that different pain conditions share common risk factors and primary chronic pain is now recognized as a disease on its own rather than a symptom of another disease 28.īuilding on these ideas, we applied machine learning to the UK Biobank dataset to synthesize a wide range of multidimensional pain-agnostic features and develop a predictive model capable of classifying and forecasting different pain conditions and the spreading of pain across body sites. Moreover, coexisting pain conditions, which over one-third of pain patients report experiencing, are associated with higher impact of pain, including lower quality of life and poorer response to treatment 22, 27. Despite differences between these conditions, evidence suggests that pain conditions overlap with one another 22, share a common genetic risk profile 23, 24 and show similar alterations in the central nervous system 15, 25, 26. The Task Force for the Classification of Chronic Pain recommends classifying chronic pain conditions based on their etiology (for example, musculoskeletal pain), underlying pathophysiology (for example, neuropathic pain) or body site (for example, back pain) 20, 21. A data-driven framework with clinical utility for predicting pain conditions is currently missing. Furthermore, these previous prospective studies have rarely been validated in out-of-sample patients and the generalizability of the findings to new patients remains unknown 18, 19. Brain imaging and genetic studies also suggest that biological factors predispose individuals to chronic pain conditions 15 however, these studies are often circular, as pain measurements or attitudes toward pain are used as pain predictors and most candidate brain-imaging markers have been identified in studies with small sample sizes, making them difficult to reproduce in larger and more diverse groups 16, 17. Additionally, factors including pain severity and duration 9, 10, 11, 12, fear of pain 13 and pain catastrophizing 4, 14 have been linked to worsening back pain. Prognostic studies have shown that certain factors, such as maladaptive pain-coping strategies, somatization of pain and history of pain increase the likelihood of developing chronic back pain 4, 7, 8, 9. Access to large cohorts of participants with chronic pain has provided unprecedented opportunities to tackle these problems and better understand the determinants of chronic pain 6. This holistic framework, referred to as the biopsychosocial model for chronic pain 5, can be challenging to define owing to the difficulties of simultaneously measuring and distinguishing multidimensional factors in large groups of patients living with pain. Instead, it is widely accepted that the interactions between biological, psychological and social factors play a greater role in determining chronic pain conditions and patients’ overall functioning 5. Unfortunately, the causes of chronic pain and its prognosis are often unknown, as tissue damage following injury is rarely an accurate predictor of clinical outcomes 4. Pain is the primary reason that individuals seek healthcare and is a leading cause of disability among working adults 1, 2, 3. Our findings show that chronic pain conditions can be predicted from a common set of biopsychosocial factors, which can aid in tailoring research protocols, optimizing patient randomization in clinical trials and improving pain management. The risk of pain spreading was then validated in the Northern Finland Birth Cohort ( n = 5,525) and the PREVENT-AD cohort ( n = 178), obtaining comparable predictive performance.
A simplified version of this score, named the risk of pain spreading, obtained similar predictive performance based on six simple questions with binarized answers. Key risk factors included sleeplessness, feeling ‘fed-up’, tiredness, stressful life events and a body mass index >30. In longitudinal analyses, the risk score predicted the development of widespread chronic pain, the spreading of chronic pain across body sites and high-impact pain about 9 years later (AUC 0.68–0.78). This data-driven model was used to identify a risk score that classified various chronic pain conditions (area under the curve (AUC) 0.70–0.88) and pain-related medical conditions (AUC 0.67–0.86).
Using data from the UK Biobank ( n = 493,211), we showed that pain spreads from proximal to distal sites and developed a biopsychosocial model that predicted the number of coexisting pain sites. Chronic pain is a complex condition influenced by a combination of biological, psychological and social factors.
0 kommentar(er)
