relapse rate discussion

0 views
Skip to first unread message

Robert Schwebel

unread,
Jun 14, 2019, 12:36:04 AM6/14/19
to ham...@googlegroups.com

I keep seeing estimates of relapse rates to be 40-60% and have no idea who is measuring what. Is that relapse to what drug and after what intervention and after what period of time? Does it refer to lapse or full blown relapse? Does it measure only people who chose abstinence? What about people who chose to stay within new limits? What about people who go for treatment and never decide to change or hate it and leave (Right there is a huge percentage)? Are they counted?


Also, how many people resolve their drug problems through "treatment" and how does that compare to the numbers who resolve them through natural recovery? 


I'm interested in stimulating some discussion on this.

 

Supposedly the 40-60% relapse rate is a disappointing figure. If it suggests that the others (not included in the 40-60%) get long term help in our crumby system of care, I think it is highly optimistic.


Robert


Robert Schwebel, Ph.D.

Psychologist

Author and Program Developer

The Seven Challenges

(520) 271-6560

www.sevenchallenges.com



Lara Okoloko, LICSW

unread,
Jun 14, 2019, 2:30:36 AM6/14/19
to ham...@googlegroups.com
Speaking of relapse rates - does anyone have articles pointing to differences in outcomes (6 months or 1 year...) for inpatient versus outpatient treatment for SUDs?

Lara Okoloko, LICSW
(206) 745-2381 confidential voicemail 


Sent from my mobile device, please excuse any typos.
--
You received this message because you are subscribed to the Google Groups "HAMS Harm Reduction Professionals" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hamspro+u...@googlegroups.com.
To post to this group, send email to ham...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hamspro/DM5PR18MB13690055A3951435B5FF9D30BDEE0%40DM5PR18MB1369.namprd18.prod.outlook.com.
For more options, visit https://groups.google.com/d/optout.

Kenneth Anderson

unread,
Jun 14, 2019, 4:08:19 AM6/14/19
to ham...@googlegroups.com
Project MATCH had two arms, an aftercare arm and an outpatient arm. 35% of subjects in the aftercare arm maintained continuous abstinence for one year. 19% in the outpatient arm maintained continuous abstinence for one year.On the other hand Schick Shadel Hospital, which uses Pavlovian counterconditioning reports that 60% of patients maintain continuous abstinence for one year.
References
Project MATCH Research Group. (1998). Matching alcoholism treatments to client heterogeneity: Project MATCH three‐year drinking outcomes. Alcoholism: clinical and experimental research, 22(6), 1300-1311.
Lemere, F., & Voegtlin, W. L. (1950). An evaluation of the aversion treatment of alcoholism. Quarterly Journal of Studies on Alcohol.


Kenneth Anderson

unread,
Jun 14, 2019, 4:19:37 AM6/14/19
to ham...@googlegroups.com
Sorry I sent the wrong reference--this is the correct one
Project MATCH Research Group. (1997). Matching alcoholism treatments to client heterogeneity: Project MATCH posttreatment drinking outcomes. Journal of Studies on Alcohol, 58(1), 7-29. doi:10.15288/jsa.1997.58.7

Kenneth Anderson

unread,
Jun 14, 2019, 4:25:39 AM6/14/19
to ham...@googlegroups.com
Outcomes in terms of percentage of drinking days and drinks per drinking day were similar in project MATCH for bboth arms--out patient and an aftercare arm which followed residential. However, Cutler and Fishbain point out that drop outs improved as much as treatment completers which suggests that the effect is due to the decision to enter treatment, not the treatment itself.
References
Project MATCH Research Group. (1997). Matching alcoholism treatments to client heterogeneity: Project MATCH posttreatment drinking outcomes. Journal of Studies on Alcohol, 58(1), 7-29. doi:10.15288/jsa.1997.58.7
Cutler, R. B., & Fishbain, D. A. (2005). Are alcoholism treatments effective? The Project MATCH data. BMC Public Health, 5(1), 75.

Maia Szalavitz

unread,
Jun 14, 2019, 9:35:18 AM6/14/19
to ham...@googlegroups.com
research shows no difference in outcomes for inpt v outpt (that's why the insurance industry killed the first generation of 28 day rehabs in the 90s, cut them in half after massive expansion).  problem is the standards of the time were that everyone got 28 days regardless of whether smoking pot on weekends (denial of heroin use!) or injecting heroin (obviously, they have a problem).  So we don't know if inpt would have an advantage if the patients were appropriately selected— but if we're talking about opioids, it makes zero sense given that much of it is aimed at getting people abstinent and demonizing meds.


For more options, visit https://groups.google.com/d/optout.


--

(Please note: Short emails likely due to overworked hands— "shortness" of tone definitely not intended!)

Maia Szalavitz

unread,
Jun 14, 2019, 9:36:12 AM6/14/19
to ham...@googlegroups.com
Also, see Tom McLellan's paper that compares relapse rates in addiction to those in diabetes and other chronic conditions.  that will be where that figure comes from and the definition will be in there.

Kenneth Anderson

unread,
Jun 14, 2019, 12:56:49 PM6/14/19
to ham...@googlegroups.com
Selection bias is also often a huge factor. TCs like DayTop boasted high success rates but were so hard to get into only the most motivated clients got in.

Steven Slate

unread,
Jun 14, 2019, 1:43:48 PM6/14/19
to ham...@googlegroups.com, Robert Schwebel
Hi Robert,

Treated vs untreated comparisons are hard to come by. I’m attaching a table from one regarding alcohol here, though it’s not a RCT. It’s from the NESARC 2002-2002 epidemiological survey, which could be better than an RCT in some ways. It gives us a longer term view of outcomes than most trials, but it doesn’t give us the immediate results of quit attempts, nor the exact treatment histories of respondents. 

At 5 years since onset of DSM-IV Alcohol Dependence, about 65% of both the treated and never-treated groups are still dependent. After that point the still dependent rates begin to diverge, so that at 20 years or more since onset 13.6% of the treated group are still dependent, while only 4.3% of the never-treated group are still dependent.




Kenneth Anderson

unread,
Jun 14, 2019, 2:04:34 PM6/14/19
to ham...@googlegroups.com
We also need to bear in mind that the NESARC data showed that those who sought treatment tended to have a chronic and severe disorder whereas those who remitted without treatment tended to have a milder disorder

Reply all
Reply to author
Forward
0 new messages