“…51,53 Although not unique to pain studies, an important consideration in designing a pain trial is the extent to which those patients with a strong treatment preference exhibit reluctance or refusal to participate or to continue participating. 56,64,93,100 To overcome this reluctance, rescue medications (eg, acetaminophen, aspirin, or other nonsteroidal anti-inflammatory drugs, or weak opioids such as codeine, tramadol, dihydrocodeine, or dextropropoxyphene) can be tried. 55 A balancing act exists, however, between providing sufficient rescue relief to retain participants in a study, but not so much that data noise invalidates the treatment effect estimate 55 or contributes to falsenegative results.…”