Wednesday, July 15, 2015

Some TDCJ treatment programs increase recidivism

A friend forwarded me a copy of this recidivism analysis from Texas Department of Criminal Justice prison rehabilitation programs, lamenting that "some of the TDCJ rehabilitation programs demonstrably make people worse."

Which ones? Four of nine programs showed participants' recidivism increased after two years in the free world, though after three years only two programs - specifically the Sex Offender Treatment Program and the Pre-Release Substance Abuse Program, the latter of which has consistently resulted in increased recidivism since the agency began studying it - displayed higher recidivism rates.

The two programs with worse outcomes after two years that came out slightly better after three were the Sex Offender Education Program and the Serious and Violent Offender Reentry Initiative.

The SAFP program is the TDCJ rehab program with the best results and was the only one to make a double-digit difference.

11 comments:

A note from Mary... said...

This study is very informative and nicely done. It leads to further questions that other researchers should be interested in. For example, I would like to see a comparison of ALL programs, not just those required prior to release. Also, length of the programs jumps out at me as a possible mediating variable. All in all, this is great information.
I find it Interesting that SVORI is the only program that releases directly from Ad Seg. That's a study in and of itself!
Thank you for this information, Grits!
Truly,
Mary

Anonymous said...

Approximately, one year ago, my son refused to take the SOTP [as did one of the SA 4 before they were released]. He had always questioned the efficacy of the program. Mental health professionals he consulted doubted the program as well. Appears as if they had valid concerns.

He was jailed as a non-violent SO; charged with on-line solicitation of a minor [14 yo who claimed she was 18]; sentenced to 9 years in TDC [no Justice in that place as far as I have been able to ascertain. A week later, the same judge sentenced an adult female who had had a 6 month-long affair with her 13 year old neighbor to probation, but that is another story] When required to take the program, he refused: since to take the program you must admit deviancy. He rejected signing the refusal form as well; since signing the form, you waive any remaining chance of parole.

As retaliation, he was moved from a low security unit [Rosharon], to a high security unit for violent offenders [Hughes] in Gatesville. Where he will most likely; serve out the remainder of his 9 year sentence.

Had he taken the program, he would have most likely been paroled after 3 years of incarceration. I personally expect he will have serve the remainder of his 9 years for his stance, unless the statue regarding his conviction is declared unconstitutional. A portion of the statue has already been declared unconstitutional this past year.

My only expectation now, is that some day he will be returned to the general population.

So it is as if he is between the proverbial rock and hard place, he has been punished to serve the entire length of his sentence for not taking a course that probably increases the chance for recidivism and is most likely does more harm than good.

I generally forward copies of Grits to him, but I am unsure if his finding out about this would be a positive or negative thing. He will grasp the implications of these results immediately.

Anonymous said...

Wait, if he signed the form he would have "waive[d] any remaining chance of parole" but "had he taken the program, he would have most likely been paroled after 3 years of incarceration." Which is it? I'm not meaning to be pedantic, but I truly want to know.

Considering the circumstances of his charge, I wouldn't have signed the form either. SOTP sounds like your typical "one size fits all" program, which doesn't really work for anyone. I wish him, and you, the best of luck on the passage of time until parole.

Gritsforbreakfast said...

It's both, 4:52, they make participation in the program a CONDITION of parole.

Anonymous said...

Scott, your friend needs to stop lamenting. His statement “some of the TDCJ rehabilitation programs demonstrably make people worse” in far from accurate. There are too many other factors that are not included in this study to make such a broad conclusion. Let’s take sex crimes and name some factors.

1. Length of incarceration? (1 year, 25 years)

2. Age of offender? (19, 60)

3. First time felony or multiple felonies? (And what were the prior felonies)

4. Parole or probation violator? (Technical violation like missing a meeting or was it a new crime)

5. Type and sub-type of crime? (Child porn, serial rapist, familial incest, urinating in public)

6. Violent or non-violent offender? (Because of the codification of the law all 3G sex offenders are deemed violent even if there was no weapon or physical force used. Strange huh?)

7. Was the offender released to parole or has the offender completed their sentence?

8. Does the sample include females? If not, why?

9. Freely Admitted guilt, program required admittance of guilt, feigned innocence, or actual innocence? (See Anon @ 9:11)

10. Drug or alcohol problems?

11. Family support? (Living with a supportive and stable family, or no family)

12. Pre and post incarceration occupation? (CEO or migrant worker)

13. Money in the bank or poor?

14. Residential location? (Affluent neighborhood or homeless)

15. Type of TDCJ unit? (Medium or maximum custody)

16. Unit housing and custody level? (Solitary, Ad Seg, Cell block, or Dormitory)

17. Unit mentality? (Rock and roll with frequent lockdowns and major violence, or quiet like the Price Daniels Unit)

18. Mental and physical health?

19. Accomplishments while incarcerated? (OJT, education)

20. Jobs while incarcerated (Clerk, skilled maintenance worker, hoe squad, or medically unassigned)

21. Effects of the registry? (Unable to find a job or a place to live)

22. What was the mentality and skill of the treatment providers that year? (There are some providers with a personal agenda)

23. Did the offender participate in post-incarceration SOTP treatment?

I also have concerns about the comparison group (page 5 in the report): I’m assuming this group is comprised of offenders who did not take the program? Do they include offenders who were non-completers who refused the program or were they kicked out for non-compliance? If they did not take the program, why not? Have all the above factors been applied to the comparison group? Were any of these offenders receiving post-incarceration treatment? I can’t visual a comparison group.

I could go on but hopefully you understand my objections. You can’t take a college graduating class and judge where they are in two and three years based on a single class. Compete individual psychosocial and economic data must be factored in.

Gritsforbreakfast said...

1:13, no recidivism study factors in "Compete individual psychosocial and economic data," the programs treat thousands of people. To claim they can't be evaluated unless every case is assessed individually begs credulity.

Yes there are many other factors, which presumably apply equally to the comparison group as the cohort in treatment. And yet, a judgment may still be made whether the treatment itself is helping or hurting, and if (because of the factors you mention, and others) the program has worse outcomes than a similar cohort that didn't go through it, the expense is hard to justify.

I have less trouble than you envisioning a comparison group: E.g., people charged with the same offense who didn't go through the program. Especially for the high volume programs, if they're not selecting on those criteria the demographics will all come out in the wash.

PRSAP should be ditched, it's never worked. They should double down on investments in SAFP. And the others appear to have at most a modest effect, inviting a more detailed cost-benefit analysis.

Anonymous said...

Grits @ 5:08 I agree. My comments were to counterbalance what appeared on the surface as rush to judgement, crediting a program as the reason for an increase in recidivism. I do not believe that was the intention but not all would understand that. I agree, no study could cover all what I listed, especially the effects of incarceration and release as seen only through eyes of an offender. I would be loathe to discover some draconian minded legislator using the comment to spark an agenda. My desire was to illustrate how many factors affect incarcerated and released offenders.

I believe a supervised post-incarceration SOTP (or other program) to be more effective than the incarcerated version. In prison the offender has to live 24/7 with the others in group, a tough row to hoe! Additionally there is the added stress of all the other offenders outside of group knowing their crime, especially if the offender is returned to their home unit after treatment. That type of knowledge is a dangerous power to posses in the penitentiary.

Post-incarceration supervised treatment programs have many benefits not available in prison. Time is a strong factor. It can take years to iron out psychological kinks. In prison being forced in a short time frame to admit guilt or fail the program is a bitter pill. Good post-incarceration therapists monitor stressors that can lead an offender to recidivate, schooling them back to center line. The offender has a better chance of dealing with the stress of reintegrating as a felon, which generally gets tougher every legislative year. It is an overall win-win for public safety and reintegration.

My thinking is that the study would be more indicative if it was divided into three groups. Programs in prison only, programs in prison followed by programs after release, and programs after release only. No programs at all would be the control group. My gut feel is that the post-release programs only would win, freeing up program bed space for those released to the streets.

By the way, how’s the new workload? I see you’re up and running early. :)

Gritsforbreakfast said...

"I do not believe that was the intention"

Admittedly, I was engaging in a moment of snark, as I'm sure was my correspondent, who is a TX community corrections professional with decades of experience who probably knows the ins and outs of recidivism causes better than either of us. I mainly wanted to record the link so a) I could find it again and b) others could see it. My comments weren't intended as much more than "Hey, look at this!"

Good point on post-incarceration therapy vs. inside, especially for SOs. To me, the substance abuse programming data are more probative, in part because they're much larger and have more people so the data are more robust.

tiapa said...

The Static-2002 is standard used by professional sex offender treatment providers (SOTP) in sex offense cases to assign risk levels and to predict recidivism. It has been well researched. I believe it is the most accurate predictor in use today. This is where TDCJ obtains an Individual sex offender’s risk level, i.e. Low, Moderate, or High. The Static-2002 generates a score of 0 to 14 by analyzing five major historical factors: Relationship to Victims, General Criminality, Age at Release, Persistence of Sexual Offending, and Deviant Sexual Interests.

The TDCJ Rehabilitation Study shows historical recidivism rates for to sex offender treatment programs, SOEP and SOTP-18.

The study states SOEP is for “sex offenders assessed (Static-2002) to pose a low sexual re-offense risk OR who may have an extended period of supervision during which they may participate in treatment”. It does not indicate if the latter group that may participate in treatment during an extended period of supervision are low, medium, or high risk of re-offense. The two groups are separated by an “or” indicating a mixture of risk levels. This might skew their analysis.

The study states SOTP is for “sex offenders assessed (Static-2002) to pose a high sexual re-offense risk”. Is that a Static-2002 High Risk, or is there again a mix or risk levels.

The study makes no reference of medium risk offenders. The Static-2002 does broadly categorize offenders as Low, Moderate, or High risk but does so in fifteen incremental steps (0 to 14). It is notable that the recidivism rates predicted by the Static-2002 are similar generally lower than those calculated in the study. This might be easily attributed to the use of technical violations in the TDCJ studies’ recidivism counts.

I agree with Anonymous @1:13 am. There are far too many factors to consider to simply claim that the pre-release TDCJ programs “demonstrably make people worse”. While these programs should be under constant positive revision, more attention should be given to the factors assailing all offenders upon release. To give an offender some old clothes, a bus ticket, and $100 upon release is near to useless. Just imagine.

Gritsforbreakfast said...

I can agree with all that, 11:05, except that you must then also acknowledge that, except for SAFP, there's no evidence that most of these programs are benefiting anyone, either. If you don't want them to accept blame when recidivism increases at the margins, you don't simultaneously get to claim credit for tiny marginal improvements.

And while the debate here has been about SOs, really it's PRSAP that's most indicted by the study. Consistently worthless, especially by comparison to SAFP for a somewhat similar cohort of offenders.

tiapa said...

I agree Grits@7:35. I would acknowledge that any program that demonstratively increases recidivism or is deemed useless need revision or decapitation.

As for the SOs I'd like to see the stats of SOTP-18 with and without aftercare.

I had to chuckle on a reread of the report.

Page 8 at the end of the ITPC section it reads "For program completers, defined as those who successfully complete both the in-prison and aftercare phases of treatment, this program has the greatest effect on recidivism rates of all the tier rehabilitation programs applicable to prison offenders."

Page 15 at the end of the SAFP section it read "This program has the greatest effect on recidivism rates of all the tier rehabilitation programs for those who successfully complete both the incarceration and aftercare phases of treatment."

I believe the latter is the only correct statement unless their is something hidden in the TDCJ doublespeak.