Although the "opioid contract" is widely used in the administration of chro
nic opioid therapy, its use has not been well defined and there are few gui
delines for developing or revising such tools. We reviewed opioid contracts
from 39 major academic pain centers and analyzed every statement for its c
ore meaning. These statements were grouped into general categories and then
into specific statement groups. Substantial diversity in the content of th
e 39 contracts was found. Statements could be grouped into 12 general categ
ories, 43 statement groups, and 125 individual statements. Each of the 39 c
ontracts reviewed contained 22.5% +/- 10.9% of the entire list of 125 state
ments and 32.6% +/- 11.2% of the 43 statement categories. Contract length a
ve-aged less than 3 pages (range: 1 to 1 mean 2.2). We describe frequent an
d infrequent themes that may be well suited for inclusion in any given cont
ract. While there are many significant issues related to the usage of a for
mal contract in chronic opioid therapy there was substantial consistency am
ong the contracts in their universal attempts to improve care through disse
mination of information, facilitate a mutually agreed-upon course, or enhan
ce compliance. This study serves as an initial step in considering the risk
s and benefits of an opioid contract as well as its ideal content and prese
ntation J Pain Symptom Manage 1999;18:27-37. (C) U.S. Cancer Pain Relief Co
mmittee, 1999.