Symptom networks in patients with substance use disorders

Abstract

Background: Reciprocity between symptoms of psychiatric disorders is increasingly recognized to contribute to their chronicity. In substance use disorders (SUD) little is known on reciprocal interactions between symptoms. We applied network analyses to study these interactions.

Methods: We analyzed 11 DSM-IV / DSM-5 criteria for SUD for the most prevalent substances in addiction care (alcohol, cannabis, cocaine, stimulants, and opioids) in a sample of 10,832 SUD patients in treatment. First, we estimated an overall symptom network. Second, we compared symptom networks between the different substances. Finally, we tested differences in symptom networks between DSM-IV and DSM-5.

Results: In the overall symptom network for SUD patients the most central symptom was: "spending substantial amount of the day obtaining, using, or recovering from substance use". The symptoms "giving up or cutting back on important activities because of use" and "repeated usage causes or contributes to an inability to meet important obligations", were the symptoms that influenced each other the most. Networks differed between substances both in global strength and structure, especially regarding the position of "use despite health or interpersonal problems". Networks based on DSM-5 criteria differed moderately from DSM-IV, mainly because "craving" was more central in the DSM-5 network than "legal problems" in DSM-IV.

Conclusions: Network analyses can identify core symptoms of SUD that could maintain the disease processes in SUD. Future studies should address whether targeting these core symptoms with precedence, might help to break through the addictive process.

Keywords: Clinical population; Network analyses; Psychopathological networks; Substance use disorder; Symptom interaction.