OBJECTIVE To fabricate a stemless plastic scintillation detector (SPSD) and characterize its linearity and reproducibility, and its dependence on energy and dose per pulse; and to apply it to clinical PDD and output factor measurements. METHODS An organic bulk heterojunction photodiode was fabricated by spin coating a blend of P3HT and PCBM onto an ITO-coated glass substrate and depositing aluminum top contacts. Eljen scintillators (~5 × 5 × 5 mm3 ; EJ-204, EJ-208, and EJ-260) or Saint-Gobain scintillators (~3 × 3 × 2 mm3 ; BC-400 and BC-412) were placed on the opposite side of the glass using a silicone grease (optical coupling agent) creating the SPSD. Energy dependence was measured by using 100, 180, and 300 kVp photon beams from an orthovoltage treatment unit (Xstrahl 300) and 6 and 10 MV photons from a Varian TrueBeam linear accelerator. Linearity, dose per pulse dependence, output factors, and PDDs were measured using a 6 MV photon beam. PDDs and output factors were compared to ion chamber measurements. A control device was fabricated by substituting polystyrene (PS) for the P3HT/PCBM layer. No photocurrent should be generated in the control device and so any current measured is due to Compton current in the electrodes, wires, and surroundings from the irradiation. Output factors were corrected by subtracting the signal measured using the control device from the photodiode measured signal to yield the photocurrent. RESULTS Each SPSD had excellent linearity with dose having an r2 of 1 and sensitivities of 1.07 nC/cGy, 1.04 nC/cGy, 1.00 nC/cGy and 0.10 nC/cGy, and 0.10 nC/cGy for EJ-204, EJ-208, EJ-260 (5 × 5 × 5 mm3 volumes), BC-400, and BC-412 (3 × 3 × 2 mm3 volumes), respectively. No significant dose per pulse dependence was measured. Output factors matched within 1% for the large scintillators for field sizes of 5 × 5 cm2 to 25 × 25 cm2 , but there was a large under-response at field sizes below 3 × 3 cm2 . After correcting the signal of the small scintillators by subtracting the current measured using the PS control, the output factors agreed with the ion chamber measurements within 1% from field sizes 1 × 1 cm2 to 20 × 20 cm2 . The impact of Cerenkov emissions in the scintillator was effectively corrected with a simple reflective coating on the scintillator. In comparison to a 6 MV photon beam, the large scintillator SPSDs exhibited 37%, 52%, and 73% of the response at energies 100 kVp, 180 kVp and 300 kVp, respectively. CONCLUSIONS The principle of the SPSD was demonstrated. Devices had excellent linearity, reproducibility, and no significant dose per pulse dependence, and a simple reflective coating was sufficient to correct for Cerenkov emissions from within the scintillator. The devices demonstrated similar energy dependence to other scintillator detectors used in a radiotherapy setting.