Understanding the interaction between clinical, emotional and psychophysical outcomes underlying tension-type headache: a network analysis approach

Abstract

Objective: Evidence supports that tension-type headache (TTH) involves complex underlying mechanisms. The current study aimed to quantify potential multivariate relationships between headache-related, psychophysical, psychological and health-related variables in patients with TTH using network analysis.

Methods: Demographic (age, height, weight), headache-related (intensity, frequency, duration, and headache-related disability), psychological and emotional (Hospital Anxiety and Depression Scale, Pittsburgh Sleep Quality Index), psycho-physical (pressure pain thresholds [PPTs] and myofascial trigger points) and health-related variables (SF-36 questionnaire) were collected in 169 TTH patients. Network connectivity analysis was unsupervised conducted to quantify the adjusted correlations between the modelled variables and to assess their centrality indices (i.e., the connectivity with other symptoms in the network and the importance in the modelled network).

Results: The connectivity network showed local associations between psychophysical and headache-related variables. Multiple significant local positive correlations between PPTs were observed, being the strongest weight between PPTs over the cervical spine and temporalis area ([Formula: see text]: 0.41). The node with the highest strength, closeness and betweenness centrality was depressive levels. Other nodes with high centrality were vitality and headache intensity.

Discussion: This is the first study applying a network analysis to understand the connections between headache-related, psychophysical, psychological and health-related variables in TTH. Current findings support a model on how the variables are connected, albeit in separate clusters. The role of emotional aspects, such as depression, is supported by the network. Clinical implications of the findings, such as developing TTH treatments strategies targeting these most important variables, are discussed.