forms of substance abuse rehabilitation has been length of stay in treatment. Virtually all studies of rehabilitation have shown that patients who stay in treatment longer and/or attend more treatment sessions, have better posttreatment outcomes (Ball and Ross, 1991; De Leon, 1984, 1994; Hubbard et al., 1997; Simpson 1981, 1997; Simpson et al., 1997a,b). Specifically, several studies have suggested that outpatient treatments of less than 90 days are more likely to result in early return to drug use and generally poorer response than treatments of longer duration (Ball and Ross, 1991; Simpson, 1981, 1997; Simpson et al., 1997a,b).

Though length of stay is a very robust, positive predictor of treatment outcome, the nature of this relationship is still ambiguous. Clearly, one possibility is that patients who enter treatment gradually acquire new motivation, skills, attitudes, knowledge, and supports over the course of their stay in treatment; that those who stay longer acquire more of these favorable attributes and qualities; and that the gradual acquisition of these qualities or services is the reason for the favorable outcomes. An equally plausible possibility is that "better motivated and better adjusted patients" come into treatment ready and able to change; that the decisions they made to "change their lives" were made in advance of their admission and because of this greater motivation and "treatment readiness" they are likely to stay longer in treatment and to do more of what is recommended. These two interpretations of the same facts have very different implications for treatment practice. If treatment gradually produces positive changes over time, it is obviously clinically sound practice to retain patients longer—perhaps even through coercion—and to provide them with more services during treatment. On the other hand, if well motivated, high functioning, compliant patients enter treatment with the requisite skills and supports necessary to do well, then efforts to provide more services or to coerce patients into longer stays may not add to the effectiveness of more streamlined and less expensive rehabilitation efforts.

Participation in AA/NA—AA is of course recognized as a self-help or mutual support organization and not a formal treatment. For this reason, and because of the anonymous quality of the group, not much research has been done to evaluate this important part of substance abuse rehabilitation until recently (McLatchie and Lomp, 1988; McCrady and Miller, 1993; Nowinsky and Baker, 1992; Project MATCH, 1997). While there has always been consensual validation for the value of AA and other peer support forms of treatment, the past few years have witnessed new evidence showing that patients who have an AA sponsor, or who have participated in the fellowship activities—have much better abstinence records than patients who have received rehabilitation treatments but have not continued in AA. McKay and his colleagues (1997a) found that participation in posttreat-