In a recent study published in the BMC Medicine, a group of researchers identified the factors influencing the variability in depression prevalence among chronic pain sufferers and developed clinical prediction models for estimating depression likelihood in this group.
Study: Variability in the prevalence of depression among adults with chronic pain: UK Biobank analysis through clinical prediction models. Image Credit: fizkes/Shutterstock.com
Background
Chronic pain is a major global disability cause, affecting over 30% of the population and often coexisting with depression, which disables roughly 5% of adults worldwide. The relationship between chronic pain and depression is well-established; each condition has the potential to worsen the other.
Despite this, the prevalence of depression among those with chronic pain is variable, with estimates ranging from 15% to 85%, influenced by differences in depression definitions, pain severity, and demographic factors such as gender, additional health conditions, and socioeconomic status.
Further research is needed to refine the understanding of the complex relationship between chronic pain and depression and to enhance the accuracy and applicability of clinical prediction models across diverse populations.
About the study
The present study utilized data from the United Kingdom (UK) Biobank. It focused on participants who completed the "online mental health self-assessment" between 2016 and 2017 and the "experience of pain" questionnaire from 2019 to 2020.
The UK Biobank's large dataset, combined with detailed surveys on pain and mental health, provided a unique platform for exploring chronic pain and its association with depression.
The "experience of pain" questionnaire was selected over the baseline data due to its more extensive array of pain types and additional variables related to pain characteristics.
Chronic pain was defined using criteria from the International Classification of Diseases 11th Revision, categorizing it as either widespread or regional based on participant responses. This distinction was important because the nature and location of pain are significant factors in the prevalence of depression among those affected.
Additionally, the study considered multisite pain and its impact on mood disorders, integrating questions about the most bothersome pain areas and the nature of the pain (neuropathic or not).
Depression was defined using a dual approach: a professional diagnosis linked from healthcare records and self-reported symptoms through a validated short form of the Composite International Diagnostic Interview.
This method aimed to capture a comprehensive view of participants' lifetime mental health history, which is crucial for understanding fluctuating conditions like depression.
The study also used the Patient Health Questionnaire to assess current depression among participants, adding another layer to the analysis. Statistical analyses included logistic regression models developed to estimate depression probability among chronic pain sufferers.
The models integrated a range of predictors, including demographic details, pain characteristics, and lifestyle factors, highlighting the complexity of chronic pain's impact on mental health.
Study results
The present comprehensive analysis involved 24,405 UK Biobank participants with chronic pain. Among these individuals, 3.7% reported present depression, 32.6% had a lifetime history of depression, 21.8% exhibited subthreshold depressive symptoms throughout their lives, and 45.6% had no lifetime history of depression.
The cohort predominantly comprised white individuals (97.1%) with an average age of 64.1 years, highlighting the need to consider a variety of demographic factors in understanding depression among those with chronic pain.
For those experiencing chronic widespread pain, 45.7% reported a lifetime history of depression, with prevalence rates varying significantly from 25.0% to 66.7% based on individual characteristics.
A prediction model incorporating variables such as age, body mass index (BMI), smoking status, physical activity, and medical history showed moderate discrimination and good calibration, suggesting its utility in clinical settings. Notably, age, gender, and BMI emerged as significant predictors of a lifetime history of depression.
Similarly, among those with chronic regional pain, 30.2% had a lifetime history of depression. The model for this group included predictors like the nature of pain and regular opioid use, and it demonstrated similar levels of discrimination and calibration.
Key predictors again included age, gender, and the specific characteristics of pain, which significantly influenced depression outcomes.
The study also assessed present depression, finding that 10.5% of individuals with chronic widespread pain and 2.5% of those with chronic regional pain were currently depressed.
Different predictors were relevant for these outcomes, with smoking status, physical activity, and comorbid conditions like chronic kidney disease playing significant roles. Models developed for current depression demonstrated moderate to high levels of discrimination and good calibration, indicating their potential reliability.
Additional analyses confirmed that the prediction models were generally robust across different types of regional pain, although some categories, like stomach and chest pain, showed slightly lower predictive accuracy.
AI model combines speech and brain activity to diagnose depression