The duration of post-traumatic amnesia (PTA) following traumatic brain injury (TBI) is a key diagnostic and outcome indicator. However, concerningly, different PTA paradigms record different PTA durations: some over-estimate, others under-estimate, PTA. Thus, a compromise is implied. The potential effect of in-hospital confounders including opioids is unknown. Three clinical groups were prospectively recruited. Group-1: in-patients with moderate-severe-TBI (MS-TBI), considered likely ‘in-PTA’. Group-2: patients rehabilitating after recent MS-TBI, considered ‘out-of-PTA’. Group-3: orthopaedic in-patients without TBI undergoing elective surgery. Only Groups 1&3 were taking opioids. All were administered the Westmead Post-traumatic Amnesia Scale (WPTAS) and the Galveston Orientation and Amnesia Test (GOAT). Results were obtained in n = 56 (Group-1:n = 18, Group-2:n = 13 and Group-3:n = 25). On WPTAS, Groups 1&3 scored similarly, but significantly lower than, Group-2 (χ2 = 8.2, P = 0.017). Contrariwise, on GOAT, Group-1 scored significantly lower than Groups 2&3 (χ2 = 23.99, P < 0.001): however, no patient scored GOAT <75. WPTAS showed moderate sensitivity (72%) but poor specificity (40%) in distinguishing Group-1 from Groups 2&3. Contrariwise, GOAT showed 100% specificity but 0% sensitivity. WPTAS ‘day of week’ and ‘pictures’ combined with GOAT ‘transport medium to hospital’, ‘anterograde amnesia’ and ‘retrograde amnesia’ maximized sensitivity (100%), specificity (85–88%), PPV (77–83%) and NPV (100%) in distinguishing Group-1 from Groups 2&3. Conclusions: Confounders including opioids likely affected WPTAS overall, but not GOAT specificity. A merger, whereby WPTAS sensitivity augmented GOAT specificity, was therefore sought. Favourable items from WPTAS (4/12) and GOAT (3/10) together optimized, and yet simplified, PTA testing; despite prevalent clinical confounders. Less, not more, ‘PTA’ items would benefit both patients and staff alike.