To further investigate the reported association between β2-adrenoreceptor (β2AR) agonists and Parkinson's disease (PD) risk. Different subtypes of β2AR agonists were considered together with an improved PD case ascertainment and increased follow-up period.

We used Cox regression with age as the timescale and use of β2AR agonists as a time-dependent exposure variable. Groups of short-, long-, and ultra-long-acting β2AR agonists (SABA, LABA & ultraLABA) were considered separately as exposure variables. We also adjusted for educational level - which in Norway is strongly associated with smoking - as well as for the use of nicotine replacement therapy (NRT). A sensitivity analysis was performed by excluding individuals with chronic obstructive pulmonary disease (COPD) as treatment indication.

A total of 11075 incident PD cases were identified during the follow-up. We found that SABA (HR= 0.85; 95%CI: 0.80,0.90), LABA (HR= 0.87; 95%CI: 0.82,0.92) and Ultra-LABA (HR= 0.59; 95%CI:0.49,0.90) were associated with a markedly lower PD risk. Minor changes in the estimates were found after adjusting for educational level and NRT. The estimates were attenuated when we excluded individuals with COPD but remained strong for ultra-LABA (HR=0.65; 95%CI: 0.29,1.14).

Consistent with our previous study, we found a marked association between the exposure to β2AR agonists and reduced PD, which was strongest for ultra-long-acting β2AR agonists. This association was only slightly attenuated after adjustment for education and NRT, and after excluding individuals with COPD as treatment indication. This suggests that smoking, a prominent confounding variable, cannot fully explain the associations between β2AR agonists and PD risk.