This study reports a systematic assessment of treatment efficacy for 15 pilot-scale subsurface flow constructed wetlands of different designs for CBOD, TSS, TOC, TN, NH-N, NO-N, NO-N, and E. coli over the course of one year in an outdoor study to evaluate the effects of design and plants. The systems consisted of a range of designs: horizontal flow (HF) with 50 and 25 cm depth, unsaturated vertical flow (VF) with sand or fine gravel, and intensified systems (horizontal and saturated vertical flow with aeration, and reciprocating fill and drain). Each system was built in duplicate: one was planted with Phragmites and one was left unplanted (with the exception of the reciprocating system, of which there was only one and it was unplanted). All systems were fed with the same primary-treated domestic wastewater. Effluent concentrations, areal and volumetric mass removal rates, and percent mass removal for the 15 systems are discussed. HF wetlands removed CBOD, TSS, TN, NH-N and E. coli by 73-83%, 93-95%, 17-41%, 0-27% and 1.5 log units, respectively. Unsaturated VF and aerated VF wetlands removed CBOD, TSS, TN, NH-N and E. coli by 69-99%, 76-99%, 17-40%, 69-99% and 0.9-2.4 log units, respectively. The aerated HF and reciprocating systems removed CBOD, TSS, TN, NH-N and E. coli by 99%, 99%, 43-70%, 94-99% and 3.0-3.8 log units, respectively. The aerated HF and reciprocating systems achieved the highest TN removal rate of all of the designs. Design complexity clearly enhanced treatment efficacy (HF < VF < Intensified, p < 0.001) during the first two years of plant growth while the presence of plants had minor effects on TN and NH-N removal in the shallow HF design only.