Both subthalamic nucleus (STN) and caudal zona incerta (cZI) have been implicated as the optimal locus for deep brain stimulation (DBS) in Parkinson's disease (PD). We present a retrospective clinico-anatomical analysis of outcomes from DBS targeting both STN and cZI. Forty patients underwent bilateral DBS using an image-verified implantable guide tube/stylette technique. Contacts on the same quadripolar lead were placed in both STN and cZI. After pulse generator programming, contacts yielding the best clinical effect were selected for chronic stimulation. OFF-medication unified PD rating scale (UPDRS) part III scores pre-operatively and ON-stimulation at 1-2 year follow up were compared. Active contacts at follow-up were anatomically localised from peri-operative imaging. Overall, mean UPDRS part III score improvement was 55 ± 9% (95% confidence interval), with improvement in subscores for rigidity (59 ± 13%), bradykinesia (58 ± 13%), tremor (71 ± 24%) and axial features (36 ± 19%). Active contacts were distributed in the following locations: (1) within posterior/dorsal STN (50%); (2) dorsal to STN (24%); (3) in cZI (21%); and (4) lateral to STN (5%). When contacts were grouped by location, no significant differences between groups were seen in baseline or post-operative improvement in contralateral UPDRS part III subscores. We conclude that when both STN and cZI are targeted, active contacts are distributed most commonly within and immediately dorsal to STN. In a subgroup of cases, cZI contacts were selected for chronic stimulation in preference. Dual targeting of STN and cZI is feasible and may provide extra benefit compared with conventional STN DBS is some patients.