Bidirectional relationships between pain and patterns of cannabis and tobacco use in a US nationally representative sample

One-fifth of US adults experience chronic pain, which is associated with increased tobacco and cannabis use. Although bidirectional relationships between tobacco and pain have been demonstrated, pathways between pain, cannabis use, and co-use of cannabis and tobacco are understudied. We aimed to estimate the effects of (1) substance use (exclusive and co-use of cannabis and tobacco) on later pain intensity, and (2) pain intensity on later substance use. Data were from 31,983 adults in biennial surveys (2015-2021) of the US nationally representative longitudinal Population Assessment of Tobacco and Health Study (n = 71,055 pairs of consecutive surveys; T1 and T2). Past-week pain intensity was dichotomized (≤4/10 no/low pain; >4/10 moderate/severe pain). Mutually exclusive substance use categories (past 30 days) were no cannabis/tobacco use; exclusive cannabis/tobacco use; and co-use. Logistic regression assessed whether T1 substance use affected moderate/severe pain at T2. Multinomial models assessed whether pain status at T1 affected substance use at T2. Compared with no cannabis/tobacco use at T1, co-use (OR: 2.29 [95% CI: 2.09-2.51]), exclusive tobacco use (2.00 [1.86-2.14]), and exclusive cannabis use (1.35 [1.13-1.61]) were all associated with moderate/severe pain at T2. Moderate/severe pain at T1 increased odds of co-use (2.43 [2.22-2.66]), exclusive tobacco (2.12 [1.98-2.28]), and exclusive cannabis use (1.46 [1.29-1.65]) compared with no cannabis/tobacco use at T2, and increased odds of co-use at T2 compared with exclusive cannabis/tobacco use. Findings demonstrated bidirectional relationships between pain and the exclusive use and co-use of cannabis and tobacco and indicate potential synergy in the co-use of cannabis and tobacco with respect to pain.