Missouri's Not Guilty by Reason of Insanity Acquittees, 1980–2009: Is Gender Important When Comparing Female and Male Insanity Acquittees and Convicted Offenders?

Abstract

Although increasing attention is being paid to differences between female and male offenders in the correctional population, potential differences between female and male insanity acquittees have not been explored. Also unexplored are the potential differences between female insanity acquittees and their correctional counterparts, both subgroups of the mentally ill offender population. This study analyzed 30 years of insanity acquittals in Missouri, finding that females were more likely to have committed murder, more likely to have been diagnosed with major depression, more likely to be older, more likely to have been married, less likely to have a substance abuse diagnosis, and more likely to be diagnosed with borderline personality disorder than male counterparts. Differences emerged when female insanity acquittees were compared to the female correctional population. Race differences were also found within the female insanity acquittee population. The focal concerns perspective is discussed to interpret implications for future practice and research.

KEYWORDS:

• female offenders
• focal concerns perspective
• gender
• insanity acquittees
• psychiatric diagnoses

Notes

^a Age is missing for one male. ^b Marital status is missing for two males.

**p ≤ .01, ***p ≤ .001.

^a On July 2009, 32 females and 303 males were hospitalized; GAF scores are missing for 3 males.

^b Medication is missing for 8 females and 38 males. ^c Prior hospitalizations is missing for 22 females and 126 males.

*p < .05, **p < .01, ***p < .001.

^a Committing county is missing for 2 males. ^b Substance use at time of crime is missing for 24 females and 134 males. ^c Prior felony conviction is missing for 5 females and 16 males. ^d Major offenses are defined in Missouri statutes and have additional testimony requirements for conditional release and unconditional release.

**p < .01, ***p < .001.

^a Includes all NGRI acquittees in the data set except the 20 who exited the system through unusual means; analyses included 161 females and 967 males. Cox regression: gender hazard ratio = 1.21, p = .087. ^b Includes all NGRI acquittees who were granted conditional releases; analyses included 137 females and 765 males. Linear regression: gender β = −.82, p = .088. ^c Includes all NGRI acquittees ever conditionally released except 9 who had a conditional release and exited through unusual means; analysis included 136 females and 760 males. Logistic regression: gender odds ratio = .88, p = .637. ^d Includes all NGRI acquittees in the data set except the 20 who exited the system through unusual means; analyses included 161 females and 967 males. Cox regression: gender hazard ratio = 1.16, p = .539. ^e Includes all NGRI acquittees who were granted unconditional releases; analysis included 46 females and 239 males. Linear regression: gender β = −.34, p = .758.

*p < .05, **p < .01, for bivariate relationships in table.

^a The statistical difference is the 1980s compared to the combined 1990s and 2000s.

**p < .01, ***p < .001.