Is Diversion Swift?

Abstract

Formal diversion programs are increasingly popular options for offenders with mental illness. Diversion is recommended, and often assumed, to be swift in that eligible persons should be quickly identified and enrolled. In this study, the authors examine the length from initial arrest to enrollment into mental health court and compare it to time from arrest to disposition for offenders with and without mental illness traditionally processed. The authors, using medians as the metric and limiting the period to 1 year, found time to mental health court was 70 days, whereas traditional processing for offenders with and without known mental illness was 37 and 76 days, respectively. The authors also found detention status during this period to have a large effect on processing time.

Keywords

mental health courts processing time diversion offenders with mental illness

The disproportionate representation of persons with serious mental illness (schizophrenia spectrum, bipolar, and major depressive disorders) in the criminal justice system, especially in prisons and jails, is well studied and undisputed (e.g., James & Glaze, 2006; Lamb & Weinberger, 1998; Steadman, Osher, Robbins, Case, & Samuels, 2009). To reduce the number of and frequency with which persons with serious mental illness are prosecuted in standard fashion within the criminal justice system, formal diversion programs have recently been established and supported by the federal government (Petrila & Redlich, 2008), including mental health courts (MHCs).

MHCs are specialty criminal courts that mandate and monitor community treatment with the aim of increasing public safety and bettering the lives of offenders with mental illness (Redlich et al., 2006). A major rationale underlying the programs is to reduce the number of days spent in jail before and after disposition. The courts have been in existence for more than a decade, and there is a growing literature indicating that they can be successful in reducing recidivism (e.g., Hiday & Ray, 2010; McNiel & Binder, 2007; Moore & Hiday, 2006; Steadman, Redlich, Callahan, Robbins, & Vesselinov, 2011) and in improving mental health symptoms (e.g., Boothroyd, Poythress, McGaha, & Petrila, 2003). Because the courts as an intervention have shown effectiveness, and because the courts have limited capacities, questions have arisen about the process of diversion, including who gets diverted, whether biases exist, how long diversion takes, and where people are housed while awaiting diversion (Almquist & Dodd, 2009; Seltzer, 2005; Steadman, Redlich, Griffin, Petrila, & Monahan, 2005). For example, some research has found that older White women are overrepresented in diversion programs (e.g., Naples, Morris, & Steadman, 2007; Steadman et al., 2005) in comparison to their presence in the criminal justice system.

In the present study, our primary question is whether diversion is swift. In answering this question, we examine factors (e.g., pretrial detention, offenses, gender) that may influence the length of time between arrest and diversion as compared to time between arrest and disposition outcome for samples of offenders processed traditionally (with and without known mental health problems). What occurs between arrest and diversion is not well studied but may influence who is diverted from jail into treatment and the eventual effectiveness of the diversion itself.

Mental Health Court Processing

Annually, approximately two million adults with serious mental illness are admitted into U.S. jails (Steadman, Robbins, Islam, & Osher, 2007). The jail environment is difficult for inmates without mental illness and is said to be particularly difficult for persons with serious mental illness (Haney, 2001). Although jails are constitutionally required to provide mental health services (F. Cohen & Gerbasi, 2005), most have inadequate mental health staffing to provide even the most basic services of screening and crisis intervention (Anno, 2001; Ditton, 1999). As a result, jail detainees with serious mental illness often have behavioral problems, decompensate, and generally do not fare well. Violations of rules and other disciplinary infractions can lead to longer stays because of increased sanctions. Persons with mental illness are more likely to be detained in jail (as opposed to released on own recognizance or have cases dismissed) and, once jailed, stay incarcerated 2.5 to 8 times longer relative to their non–mentally ill counterparts (Council of State Governments, 2005; but see Draine, Wilson, Metraux, Hadley, & Evans, 2010). MHCs and other diversion programs were developed in direct response to these identified issues of overrepresentation and potentially unconstitutional and disproportionate jail stays. However, such courts are not without controversy (Seltzer, 2005; Stefan & Winick, 2005).

One particular controversy involves the time it takes to enroll people into MHCs. A main goal of diversion is to divert eligible individuals swiftly (Steadman, Morris, & Dennis, 1995). Indeed, in the Council of State Government’s Essential Elements of a Mental Health Court, Element 3 specifies the need to identify, refer, and accept MHC clients “as quickly as possible” (Thompson, Osher, & Tomasini-Joshi, 2008, p. 3). If people experience lengthy processing in the criminal justice system, particularly while in jail, diversion is arguably not occurring. However, in determining whether diversion is swift, it is important to consider how long traditional criminal processing (i.e., from arrest to disposition) typically takes. Kyckelhahn and Cohen (2008) reported that in 2004, about 25% of felony defendants from large urban counties were processed within 30 days of arrest, about 50% were processed in 85 days or less, and 88% were processed within 1 year. T. H. Cohen and Reaves (2006) also reported that violent crimes tend to take longer to process than nonviolent crimes, and a key factor influencing length of processing time is whether the person was detained in jail the entire time or released (or never incarcerated; also see Redlich, 2010).

Little knowledge exists about the length of the MHC diversion process. Although several studies have reported on how MHC clients get referred into the court (e.g., Linhorst et al., 2009; Luskin, 2001), few have examined time of processing. Moreover, to our knowledge, the studies that do exist have focused exclusively on the time from referral to acceptance decision making (Almquist & Dodd, 2009), which does not capture the entire time frame from (a) arrest to (b) referral to diversion program to (c) diversion program decision to (d) diversion program enrollment. Estimates from referral to MHC decision (Steps b to c) have reportedly ranged from a few hours (Petrila, Poythress, McGaha, & Boothroyd, 2001) to 3 months (O’Keefe, 2006). In a study of seven MHCs, Steadman and colleagues (2005) found the median length of time between referral to acceptance decision to be 20 days and the modal length to be less than 1 day. Although Steadman et al. did not go into depth about the reasons for the wide range of time between these decision points, they noted that two of the seven courts accounted for most of the same-day referral-acceptance decisions. In addition, Steadman et al. reported the primary and secondary agents of referral into the court; public defenders emerged as a common denominator, though several other referral pathways were noted (e.g., other judges, competency examination orders). Though not previously examined to our knowledge, the referral agent is likely to influence the time it takes from arrest to referral as well as the time from referral to disposition decision.

Morris and Steadman (2008) investigated the entire period from arrest to enrollment for 34 non-MHC postbooking jail diversion programs and found the median length to be 11 days. The entire process (Steps a to d above) took about 11 days for non-MHC postbooking jail diversion programs; by contrast, Steps b and c alone were found to take 2 to 3 times as long for MHCs. MHCs may have implemented successful procedures to refer potentially eligible persons and then make decisions in a short span of time, but the length of time before referral and after the disposition decision may, however, take much longer. It is this entire period—from initial arrest to eventual enrollment—that is important in determining whether diversion is swift. This is a critical question since, as mentioned, a major rationale for MHCs is the presumed cost savings for fewer jail days by defendants being diverted.

The Present Study

In the present study, we compare the time between arrest and MHC enrollment to the time between arrest and adjudication for offenders with and without mental illness processed through the traditional criminal justice system. We also examine whether pre-enrollment/predisposition detention affects these time periods. Three separate samples are used: a sample of MHC defendants and two distinct samples of defendants processed through the traditional system.

The MHC sample consists of participants from the MacArthur MHC study (see Redlich et al., 2010; Steadman et al., 2011). The first traditional adjudication sample is from data collected by the U.S. Department of Justice (i.e., State Court Processing Statistics [SCPS]; Bureau of Justice Statistics), which tracks the time from arrest to disposition outcome for felony defendants from the largest urban counties (U.S. Department of Justice, 2006). Disposition is a comparable end point to MHC enrollment because the majority of MHCs require guilty pleas as a condition of enrollment (Redlich et al., 2006; Seltzer, 2005). In the United States, about two-thirds of felony defendants plead guilty. Presumably, the majority of the SCPS sample is not seriously mentally ill. We utilize these data to establish a baseline comparison for the time between arrest and adjudication for offenders without mental health problems as well as to investigate other factors. More specifically, within these data, we investigate court processing time in prosecutorial diversion, specialty court, and mental health treatment subsamples.

The second treatment as usual (TAU) sample is the matched comparison sample from the MacArthur MHC study. Considering that offenders with mental illness may have disproportionate jail stays, it is important to include a comparison sample of traditionally processed offenders with mental health problems. Participants in the MacArthur TAU (MacTAU) sample were recent arrestees who met the eligibility criteria for the MHC but were never referred to, or rejected from, the court. Thus, the MacTAU participants had charges and rates of serious mental illnesses comparable to the MHC sample as well as being from the same jurisdictions as the MHC participants. MacArthur study researchers worked with mental health personnel in the jails to select potential TAU participants and enroll them into the study (see Steadman et al., 2011, for more details).

Method

All data analyzed here are from archival records. The demographic factors (age, gender, and race/ethnicity) were obtained from the participating MHC or from county criminal justice records or personnel. The four MacArthur sites were San Francisco, California (n = 254), Santa Clara County, California (n = 334), Hennepin County, Minnesota (n = 248), and Marion County, Indiana (n = 211). Target criminal offenses (charges, relevant dates, and final dispositions), criminal history variables (number of previous arrests), and jail days were also obtained from county or national (Federal Bureau of Assistance) records or personnel.

MacArthur study

The MacArthur study was designed to determine the overall effectiveness of MHCs in reducing recidivism and enhancing treatment access and engagement. This study is a massive undertaking, involving four data collection sites, two samples (MHC sample n = 447; comparison TAU sample n = 600) at each site, two 90-min self-report interviews (baseline and 6-month follow-up), and extensive collection of objective record information 18 months before and after baseline. Data were collected from 2005 to 2008 (for more information about the study and samples, see Redlich et al., 2010; Steadman et al., 2011).

The study here involved utilizing a small portion of the data collected. Specifically, in addition to demographic and criminal history variables (see Table 1), the primary variables we utilized were (a) target arrest date, (b) MHC enrollment or TAU disposition date, (c) detention status in between arrest and outcome (dichotomized as 0 = never detained or detained and released and 1 = detained the entire time), and (d) TAU disposition outcome (case dismissed, pled guilty, went to trial, and other/unknown/pending).

Table 1:

Sample Characteristics for the MacArthur Treatment as Usual (TAU) and Mental Health Court (MHC) Groups

	Total Sample (N = 646)		TAU (n = 336)		MHC (n = 311)
Variable	M	SD	M	SD	M	SD	t	p
Male (yes = 1)	0.61	0.487	0.63	0.484	0.59	0.492	0.912	.362
White (yes = 1)	0.51	0.500	0.54	0.499	0.48	0.500	1.397	.163
Age	37.18	10.101	37.07	9.574	37.29	10.654	−0.281	.779
Detain status (yes = 1)	0.45	0.498	0.50	0.501	0.39	0.488	2.870**	.004
Offense type (yes = 1)
Person	0.30	0.458	0.29	0.455	0.31	0.461	−0.417	.677
Property	0.33	0.471	0.31	0.463	0.36	0.480	−1.308	.191
Drug	0.16	0.364	0.16	0.371	0.15	0.356	0.585	.559
Minor	0.21	0.410	0.24	0.426	0.19	0.393	1.449	.148
Felony Offense (yes = 1)	0.64	0.479	0.66	0.473	0.62	0.486	1.16	.247
Adjudication type (yes = 1)
Dismissal	0.11	0.307	0.20	0.402	—		—
Trial (acquittal or guilty verdict)	0.01	0.117	0.03	0.162	—		—
Divert–MHC	0.48	0.500	—		1.00		—
Guilty plea	0.35	0.479	0.68	0.466	—		—
Other disposition outcomes^a	0.04	0.207	0.09	0.282	—		—
Number of prior arrests	7.83	5.707	8.45	5.573	7.16	5.783	2.832**	.005
Disposition settled after 1 year (yes = 1)	0.11	0.311	0.14	0.345	0.08	0.267	2.465**	.014

Other disposition outcomes include no file/no record/unknown, sentence of stay of imposition, and those recorded as “disposition pending.”

p ≤ .01.

The original MacArthur study included 1,047 total participants. However, only 646 were included here (335 MHC participants and 311 TAU participants). First, relevant data for the TAU sample for the Santa Clara, California, site could not be retrieved (without going to extreme efforts) because of the manner in which these data were stored and categorized. Thus, Santa Clara participants from both the MHC (n = 136) and the TAU sample (n = 198) were not included here. In addition, disposition dates and outcomes for 62 of the 146 cases from the San Francisco TAU sample could not be located by the court personnel there. Finally, 5 cases from the Hennepin TAU sample did not possess valid data on disposition date and disposition outcome and were excluded.

As can be seen in Table 1, there were few significant differences between the MacTAU and MacMHC samples included here. About two thirds of both samples were White; person and property offenses each accounted for about 30% of the target offenses, followed by public order or other minor offense (21%) and drug-related offenses (16%). For the TAU group, the majority of cases (68%) were resolved with guilty pleas, whereas 20% of them were dismissed.

Three statistically significant differences were noted between the MacArthur samples. On average, the TAU sample had slightly more prior arrests (to the target arrest) than the MHC sample (TAU = 8.45, MHC = 7.16; d = .23; see Table 1). In addition, only 39% of the MHC group was detained the entire period between arrest and disposition, whereas 50% of the TAU sample was similarly detained. Finally, the TAU sample had more respondents whose cases took longer than 1 year to resolve (14%) compared to the MHC group (8%). Thus, as described below, we analyze the data using the full samples and with the subsample of those whose cases were disposed within 1 year, which is similar to our baseline, comparison SCPS data.

State Court Processing Statistics

The SCPS data sets are maintained by the Bureau of Justice Statistics, Department of Justice. Since its inception in 1994, the SCPS project has tracked all felony cases in May of even-numbered years from 40 of the 75 most populous counties. Information covers arrest to final disposition, sentencing, and beyond. These data yield a nationally representative sample for the examination of how swiftly felony cases are disposed. Eight waves of data have been collected thus far, but only the data collected in 2002 included the unique item of whether offenders received mental health treatment or counseling, which is why we selected the 2002 data set (which was not released for public use until 2006; http://bjs.ojp.usdoj.gov/index.cfm?ty=pbdetail&iid=896).

Data collection for each wave spans only 1 year for all crimes but murder (which is followed for 2 years). The complete sample size for the 2002 wave is 15,358. However, 2,340 of the sample have data missing on the variable of interest—number of days from arrest and disposition—because of a still-pending decision before data collection ended (n = 154), because the duration was longer than a year (n = 1,858; 12% of the total sample), or because they were otherwise missing (n = 328). To compare on key variables, we focused on the remaining 13,018 cases, which were cases disposed within a year (which, as mentioned above, also was done with the MacArthur samples). As seen in Table 2, about one third of the selected SCPS sample were Caucasian, most (82%) were male, and 41% were detained the entire period between arrest and disposition. The sample had an average of five arrests before the current offense. For the most serious target arrest charge, 23% of the cases were person crimes (e.g., violent and sexual offenses), 31% property related, 36% drug related, and the remainder minor crimes (see Table 2).

Table 2:

Sample Characteristics for the State Court Processing Statistics (SCPS) Sample ( N = 13,018)

Variable	M	SD
Male (yes = 1)	0.82	0.385
White (yes = 1)	0.36	0.481
Age	30.96	10.46
Detain status (yes = 1)	0.406	0.491
Offense type (yes = 1)
Person	0.23	0.423
Property	0.31	0.463
Drug	0.36	0.480
Minor	0.10	0.295
Adjudication type (yes = 1)
Dismissal	0.24	0.427
Trial	0.04	0.207
Diverted^a	0.07	0.261
Guilty plea	0.63	0.484
Other disposition outcomes	0.001	0.023
Number of prior arrests	5.01	4.11

Diversion in SCPS was not specified. Detailed diversion programs are not known, but 396 of the 956 diverted cases were assigned special treatment that includes cases of prosecutorial diversion, cases with special treatment court assignment, and cases in the “other” category.

Results

In addition to calculating the mean and median length of time between arrest and outcome, we examined the factors that can potentially influence this time. Because the dependent variable (i.e., time from arrest to diversion or disposition) was highly skewed to the right, our focus was on the median (though we also report the means). For the multivariate regression analyses, we logged the dependent variable, which produced a more normal distribution. We first examined the time from arrest to case disposition for the general jurisdiction sample (i.e., the SCPS data) as this served as our baseline comparison from which to compare the MacArthur data. Significance for alpha levels was set at p ≤ .01.

SCPS Data—Baseline

Number of days from arrest to adjudication was, on average, 104.44 days (SD = 92.71), with a median of 76 days (see Table 3). Next, we computed a series of bivariate correlations to determine what factors potentially influenced the length of time. We computed Spearman’s rho for dichotomous variables. As shown in Table 4, correlations were generally less than or equal to r = .10, indicating small magnitudes. A notable exception was the impact of detention status, with the correlation indicating that those detained the entire period had significantly shorter times from arrest to disposition than those who had been released. As shown in Figure 1, for those detained, median time from arrest to adjudication was 46 days; for those not detained the entire time, the median length was 106 days.

Table 3:

Mean and Median Time (in Days) From Arrest to Diversion or Disposition Outcome for the State Court Processing Statistics (SCPS) and MacArthur Treatment as Usual (TAU) and Mental Health Court (MHC) Samples

		Length
Sample	n	M	SD	Mdn
SCPS	13,018	104.44	92.71	76
SCPS prosecutorial diversion	288	97.05	96.71	60
SCPS specialty (drug) court	186	112.75	104.53	81
SCPS mental health treatment	101	120.37	78.78	106
MacTAU	336	151.64	225.17	52
MacMHC	311	135.53	193.52	80
MacTAU—within 1 year only	289	75.70	88.05	37
MacMHC—within 1 year only	287	91.60	71.97	70

Table 4:

Bivariate Correlations With Number of Days From Arrest to Disposition

	SCPS (n = 13,018)	MacTAU (n = 336)	MacMHC (n = 311)
Male	0.016	−0.025	−0.091
Age	−0.028	0.034	0.025
White	0.032	0.083	0.019
Detain status	−0.278***	−0.542***	−0.498***
Number of prior arrests	−0.046	−0.072	−0.252***
Offense type
Person	0.124	−0.025	−0.125
Minor	−0.000	−0.192***	0.024
Drug	−0.070	0.082	−0.020
Property	−0.040	0.135	0.116
Disposition–TAU outcomes
Diverted (SCPS)	0.024	—	—
Dismissal	−0.073	0.130	—
Trial	0.130	0.071	—
Guilty plea	−0.007	−0.059	—

Note. SCPS = State Court Processing Statistics; MacTAU = MacArthur treatment as usual; MacMHC = MacArthur mental health court.

***

p < .001.

Figure 1:

Median Days From Time to Diversion or Disposition by Detainment Status

The 2002 SCPS data set also included subsamples of defendants whose disposition outcomes were categorized as “special assignment.” For our purposes, we were interested in those classified under the categories prosecutorial diversion (n = 288) and specialty court (n = 186; note, however, that 176, or 95%, were sentenced to drug courts). These two groups were mutually exclusive. Because of the relevance to our research questions (i.e., swiftness of diversion), we examined the mean and median lengths of time from arrest to outcome for these two groups separately (see Table 3). For the prosecutorial diversion subsample, the mean length of time was 97.05 days (SD = 96.71) and the median length was 60 days, which is shorter than the overall median (by 16 days). For the drug court subsample, the mean length was 112.75 days (SD = 104.53) and the median was 81 days, which is slightly longer than the overall median (by 5 days). We also note that when these two subsamples and other specially assigned cases are removed from the total sample, the mean (M = 104.16, SD = 91.99) and median (76 days) lengths remain virtually unchanged (original M = 104.66 and Mdn = 76 days).

In addition, the 2002 SCPS data set allowed us to examine a proxy subsample of offenders with mental health problems. Specifically, 101 people were said to have received mental health treatment and/or counseling. The mean and median length between arrest and adjudication for this sample of 101 mental health treatment recipients were M = 120.37 days (SD = 78.78), with a median of 106 days. We found that it took about 2 weeks to 1 month longer to process this subsample of offenders who received mental health treatment in comparison to offenders who did not receive such treatment.

Macarthur MHC and TAU Samples

Using a similar pattern to that followed for the SCPS data, we next examined the two MacArthur samples. When using the full sample (i.e., not restricted to 1 year), the key variable, number of days from arrest to adjudication, has an overall mean of 144 days and a median of 65 days. The difference between the two subgroups was not statistically significant (TAU M = 151.64, SD = 225.17; MHC M = 135.53, SD = 195.32), t(644) = 0.97, p = .33, d = .08. In terms of median days, the TAU sample was 52 days, whereas the MHC sample was 80 days.

Next, we examined the means and medians using only those cases resolved within 1 year. First, almost twice as many TAU participants (n = 46, 14%) as MHC participants (n = 24, 8%) took longer than 1 year, and, second, the SCPS data (our baseline comparison) limited their sample to 1 year. In this restricted sample, the means for the MacMHC and MacTAU samples were 91.60 (SD = 71.97) and 75.70 (SD = 88.05), respectively, t(574) = 2.37, p < .02, d = .20. The median lengths were 37 days for TAU and 70 days for MHC (see Table 3).

Bivariate correlations with time to MHC or TAU disposition are shown in Table 4. We can again see the significant and negative influence of detention status on processing time for both MacArthur samples. The median times, displayed in Figure 1, show quite similar patterns to that found in the SCPS data. Persons detained the entire period have significantly shorter processing times than those released or never detained, and vice versa. Generally, correlations for the MacMHC and MacTAU samples showed similar patterns, with two exceptions. First, in the MacMHC but not the MacTAU sample, having more prior arrests was related to faster processing, and vice versa. Second, MacTAU participants (but not MacMHC participants) charged with a minor crime had shorter processing times than those charged with more serious crimes.

Next, to fully answer our research question, we conducted multivariate regressions with the logged dependent variable of time from arrest to diversion or disposition. We used the same set of variables in the bivariate analyses and conducted two separate regressions. The first (Model 1) did not limit the sample to those cases processed within 1 year. However, in the first regression, 29 MacTAU cases were omitted because they had uninterpretable disposition outcomes.¹ The second regression analysis (Model 2) included only those offenders (in both MacArthur samples) whose cases were processed within 1 year; this resulted in a total n of 554 (for mean and median lengths for these subsamples, see Table 3).

Both regressions explained about 30% of the variance (see Table 5). In Model 1, MHC or TAU status did not have a significant impact on processing time (β = .16, p = .13; see Table 5). Detention status remained a robust predictor, and minor charges, in comparison to property crimes, were quicker to process, even in consideration of detention status. Finally, White offenders took longer to process than non-White offenders (see Table 5).

Table 5:

Regression Analyses for the MacArthur Samples

	Model 1 (n = 617)				Model 2 (n = 554)
	β	SD	t	p	β	SD	t	p
Number of prior arrest	0.007	0.009	0.794	.428	0.006	0.009	0.713	.476
Detain status (yes = 1)	−1.474***	0.102	−14.421	.000	−1.206***	0.098	−12.297	.000
Offense type (ref = property)
Person	−0.163	0.119	−1.376	.169	−0.145	0.114	−1.268	.205
Drug	0.148	0.143	1.040	.299	0.086	0.139	0.622	.534
Minor	−0.530***	0.131	−4.033	.000	−0.480***	0.126	−3.826	.000
Age	−0.003	0.005	−0.514	.607	−0.003	0.005	−0.642	.521
Male	0.083	0.100	0.830	.407	0.031	0.096	0.326	.745
White	0.199**	0.098	2.030	.043	0.113	0.093	1.215	.225
Disposition (ref = guilty plea)
Mental health court	0.156	0.102	1.529	.127	0.320**	0.097	3.297	.001
Dismissal	0.186	0.158	1.179	.239	0.041	0.159	0.260	.795
Trial	0.235	0.425	0.552	.581	−0.637	0.509	−1.251	.212
R ²	.326				.302

p < .01 ***p < .001.

In Model 2, MHC status did have a significant impact processing time. Specifically, as seen in Table 5, in comparison to the MacTAU offenders pleading guilty (which was the majority of the sample, 68%), offenders diverted to the MHC took significantly longer to process. Namely, being sentenced to MHC resulted in a 32% increase in processing time compared to those who pleaded guilty. In addition, although the influence of detention status and minor charges (compared to property charges) remained significant and in the same direction as found in Model 1, the effect of being White on processing time did not.

Discussion

In this article, we addressed the length of time between initial arrest and diversion, attempting to determine if diversion is swift. We compared length of MHC enrollment to the time between initial arrest and disposition outcomes for offenders with and without mental health problems traditionally processed. Using median lengths, we found that in a nationally representative sample (in 2002), it took about 76 days from start to finish to process most defendants. For offenders with mental illness diverted to the MHC in three (large) counties, it took 70 days (when similarly limited to 1 year), and for a matched comparison sample of offenders with mental illness from the same three counties conventionally processed, it took 37 days. Thus, we did not find diversion to be swift; indeed, we found diversion to take about twice as long in comparison to traditional processing of offenders with mental illness from the same jurisdictions.

As defined here, “start to finish” was the time between arrest and diversion, or arrest and case outcome. Earlier research on seven MHCs found the median number of days between MHC referral and disposition decisions, two decision points that necessarily occur in between arrest and diversion, to be 20 days (Steadman et al., 2005). Thus, although we did not measure the time between each step or decision point, we found the whole process to take 2.5 times longer than the two intermediary steps (i.e., referral to disposition decisions). Whether more time is spent between arrest and referral, or between disposition decision and diversion enrollment, for example, was not examined here but will be important to determine in future research. It may be that MHC processing is lengthier because of eligibility assessments and community placement issues. As noted by one San Francisco stakeholder, “The screening process and admission to BHC [behavioral health court] is not exactly efficient or effective” (McNiel & Binder, 2010, p. 231).

We also found that time to diversion or disposition was affected by other factors. Most notable was whether the offender was detained in jail the entire time or was out in the community. Presumably, when a defendant is in jail, the motivation exists for alacritous processing compared to when in the community and the potential for failure to appear and delays in processing increases. MHCs are diversion programs. They aim to reduce the number of days spent incarcerated both pre- and postadjudication. On one hand, 39% of the MHC defendants were incarcerated the entire time, compared to 41% and 50% of the SCPS and MacTAU samples, respectively. The difference in detention rates between the MacMHC and MacTAU samples was significant, indicating that the MHCs may be successful in decreasing jail time for potential enrollees during referral and disposition decision making (though this was not specifically examined in our study).

On the other hand, those who were detained the entire time were processed about 2.5 to 7.5 times faster than those who had been released. In the MHC sample, it took about 1.5 months to enroll in the court for those detained. For those MHC defendants who had been released, it took about 4 months (see Figure 1). Incarcerating offenders with mental illness prior to diversion seems inconsistent with the concept of the courts (and cost savings). Almquist and Dodd (2009) discuss due process concerns and the negative effects of jail during this waiting period and advocate for persons not to be detained. However, if detained, MHC clients are processed much faster than those released, jail-related and due process concerns must be weighed against admitting the offenders into the MHC programs earlier, programs that have been shown to have beneficial long-term effects in comparison to traditional processing (e.g., Hiday & Ray, 2010; Steadman et al., 2011).

Another question raised by the present findings is how much “diversion” (from jail) takes place for defendants with mental illness beyond what already occurs in traditional processing. The SCPS data indicated that about one fourth of defendants’ cases were dismissed (see Table 2). In addition, about 60% of defendants were released from jail prior to disposition (90% of whom within 1 month, as reported by T. H. Cohen & Reaves, 2006), and overall 48% of those convicted were not incarcerated after adjudication. In contrast, there is some evidence that MHC clients are not always entirely “diverted” in that some attend court hearings from in custody rather than from the community. Redlich et al. (2010) reported that in one MHC, 45% of clients attended half to all of their status review hearings from in custody. Thus, it is quite probable that had they not entered the court, MHC clients would have avoided incarceration (and possibly a conviction) via other avenues. Thompson et al. (2008) specifically state that “[t]he time required to accept someone into the program [i.e., the MHC] should not exceed the length of the sentence that the defendant would have received had he or she pursued the traditional court process” (p. 3). Given that the average time to MHC was about 3 months and that a majority of offenders are likely to either have their case dismissed or receive probation or time served for their sentence, it is quite probable that the time it takes to accept and enroll eligible MHC clients does exceed analogous times received for traditional sentencing. Of course, it is important to note that diversion is more than reducing jail time. As mentioned earlier, the research studies demonstrating the benefits of MHC are becoming too numerous to ignore. The short-term benefits of getting out of jail must be juxtaposed with the potential for longer-term benefits of MHC participation (i.e., access to treatment, reductions in recidivism, improved quality of life).

Research on MHCs is beginning to establish a better understanding of for whom the courts are most effective and under what circumstances. Using the MacArthur data, Steadman and colleagues (2011) found that criminogenic factors (e.g., pretarget arrests and jail days) rather than personal, clinical, or treatment factors were better predictors of MHC success (i.e., reduced recidivism). Of note, we also did not find personal characteristics such as age, gender, or race to influence time to enrollment or disposition. Even number of prior arrests was not influential after considering other factors. Rather, when limited to those processed within 1 year (which included most, 89%, participants), the only factors to influence processing time were MHC status, detention status, and charge type (minor crime).

Conclusions and Limitations

Although intriguing, the present data can be considered only preliminary. First, the SCPS sample and the MacArthur samples differed in demographic and criminal history factors and when the data were collected. However, we did not find personal characteristics to have a significant impact on our main variable. It is also important to note that the SCPS data set included only offenders charged with felonies, whereas the MacArthur samples included offenders charged with felonies and misdemeanors. As seen in the multivariate analyses, less serious crimes (specifically minor offenses) were found to have shorter processing times (in comparison to more serious property crimes). Thus, if our MacArthur samples were all felony offenders, our finding that time to MHC enrollment took longer than traditional processing would likely have been even more robust.

Second, data from only three MHCs were included here, and thus findings may not generalize to other MHCs that are notoriously individualistic. Furthermore, some data were inaccessible, including 42% of the San Francisco TAU sample from the MacArthur study. We did not address whether time to MHC enrollment is predictive of MHC success; this is an interesting question for future research. Also, determining whether time in detention itself influences who gets diverted (a question not addressed here) would be important. In a legal system with a guarantee of equal protection (the Fourteenth Amendment), diversionary practices—particularly ones demonstrably effective—that are biased would be highly controversial, if not unconstitutional.

The length of time it takes for TAU criminal processing for the nation at large and time to MHC enrollment was approximately the same (76 and 70 days). On one hand, this can perhaps be seen as encouraging because it has been assumed that mental illness is associated with longer periods of criminal justice involvement (but see Draine et al., 2010). Accordingly, the finding that time to MHC enrollment was 6 days shorter than traditional processing for offenders without mental illness is promising. On the other hand, the national sample included only offenders charged with felonies, whereas the MHC sample included offenders charged with misdemeanors and felonies. Although determining the national baseline was illustrative, arguably the more apt comparison reported here was the MacTAU sample given that these offenders had known mental illnesses, were matched to the MHC sample, and were from the same jurisdictions. In this comparison, we found MHC processing to take about twice as long as traditional court processing. Although these findings indicate that diversion is not swift, swiftness in diversion enrollment may be less important than enrollment in diversion itself.

Footnotes

This article is part of a special issue titled “Diversion from Standard Prosecution”, edited by Kirk Heilbrun and David DeMatteo of Drexel University. This research was generously supported by the John D. and Catherine T. MacArthur Foundation. We are especially grateful to John Monahan, the director of the MacArthur Network on Community Mandated Treatment, and to network members. This research was also supported by the University at Albany Faculty Research Award Program. We are indebted to Brian Case, Karli Keator, and Roumen Vessilnov for their invaluable assistance. Finally, we thank the court and jail personnel who participated in this study; without their dedication and assistance, we could not have conducted the study.

Notes

References

Almquist

Dodd

(2009). Mental health courts: A guide to research-informed policy and practice. New York, NY: Council of State Governments Justice Center. Retrieved from http://www.consensusproject.org/jc_publications/mental-health-courts-a-guide-to-research-informed-policy-and-practice/Mental_Health_Court_Research_Guide.pdf

Anno

B. J.

(2001). Correctional health care: Guidelines for the management of an adequate delivery system. Washington, DC: U.S. Department of Justice, National Institute of Corrections.

Boothroyd

Poythress

McGaha

Petrila

(2003). The Broward Mental Health Court: Process, outcomes and service utilization. International Journal of Law and Psychiatry, 26, 55-71.

Cohen

Gerbasi

J. B.

(2005). Legal issues regarding the provision of mental health care in correctional settings. In Scott

C. L.

Gerbasi

J. B.

(Eds.), Handbook of correctional mental health (pp. 260-272). Washington, DC: American Psychiatric Publishing.

Cohen

T. H.

Reaves

B. A.

(2006). Felony defendants in large urban counties, 2002. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Council of State Governments. (2005). Criminal Justice/Mental Health Consensus Project. Retrieved from http://www.consensusproject.org

Ditton

(1999). Mental health and treatment of inmates and probationers. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Draine

Wilson

A. B.

Metraux

Hadley

Evans

A. C.

(2010). The impact of mental illness status on the length of detention and the legal mechanism of jail release. Psychiatric Services, 61, 458-462.

Haney

(2001). The psychological impact of incarceration: Implications for post-prison adjustment. Retrieved from http://aspe.hhs.gov/hsp/prison2home02/Haney.htm#N_18_

10.

Hiday

V. A.

Ray

(2010). Arrests two years after exiting a well-established mental health court. Psychiatric Services, 61, 463-468.

11.

James

D. J.

Glaze

L. E.

(2006, September). Mental health problems of prison and jail inmates (Bureau of Justice Statistics Special Report, NCJ 2136). Washington, DC: U.S. Department of Justice.

12.

Kyckelhahn

Cohen

(2008). Felony defendants in large urban counties, 2004. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/fdluc04.pdf

13.

Lamb

H. R.

Weinberger

L. E.

(1998). Persons with severe mental illness in jails and prisons: A review. Psychiatric Services, 49, 483-492.

14.

Linhorst

D. M.

Dirks-Linhorst

P. A.

Stiffelman

Gianino

Bernsen

H. L.

Kelley

B. J.

(2009). Implementing the essential elements of a mental health court: The experiences of a large multijurisdictional suburban county. Journal of Behavioral Health Services and Research, 37, 427-442.

15.

Luskin

M. L.

(2001). Who is diverted? Case selection for court-monitored mental health treatment. Law and Policy, 23, 217-236

16.

McNiel

Binder

(2007). Effectiveness of a mental health court in reducing criminal recidivism and violence. American Journal of Psychiatry, 164, 1395-1403.

17.

McNiel

Binder

(2010). Stakeholder views of a mental health court. International Journal of Law and Psychiatry, 33, 227-235.

18.

Moore

M. E.

Hiday

V. A.

(2006). Mental health court outcomes: A comparison of re-arrest and re-arrest severity between mental health court and traditional court participants. Law and Human Behavior, 30, 659-674.

19.

Morris

Steadman

H. J.

(2008, March). Time to enrollment in a multi-site jail diversion study. Paper presented at the National GAINS Conference, Washington, DC.

20.

Naples

Morris

L. S.

Steadman

H. J.

(2007). Factors in disproportionate representation among persons recommended by programs accepted by courts for jail diversion. Psychiatric Services, 58, 1095-1101.

21.

O’Keefe

(2006, September). The Brooklyn Mental Health Court evaluation: Planning, implementation, courtroom dynamics, and participant outcomes. New York, NY: Center for Court Innovation.

22.

Petrila

J. P.

Poythress

McGaha

Boothroyd

(2001, Winter). Preliminary observations from an evaluation of the Broward County Mental Health Court. Court Review, pp. 14-22.

23.

Petrila

J. P.

Redlich

A. D.

(2008). Mental illness and the courts: Some reflections on judges as innovators. Court Review, 43, 164-173.

24.

Redlich

A. D.

(2010). The past, present, and future of mental health courts. Manuscript submitted for publication.

25.

Redlich

A. D.

Steadman

H. J.

Callahan

Robbins

P. C.

Vesselinov

Ozdogru

A. A.

(2010). The use of mental health court appearances in supervision. International Journal of Law and Psychiatry, 33, 272-277.

26.

Redlich

A. D.

Steadman

H. J.

Monahan

Robbins

P. C.

Petrila

(2006). Patterns of practice in mental health courts: A national survey. Law and Human Behavior, 30, 347-362.

27.

Seltzer

(2005). Mental health courts: A misguided attempt to address the criminal justice system’s unfair treatment of people with mental illnesses. Psychology, Public Policy, and Law, 11, 570-586.

28.

Steadman

H. J.

Morris

S. M.

Dennis

D. L.

(1995). The diversion of mentally ill persons from jails to community-based services. American Journal of Public Health, 85, 1630-1635.

29.

Steadman

H. J.

Osher

F. C.

Robbins

P. C.

Case

Samuels

(2009). Prevalence of serious mental illness among jail inmates. Psychiatric Services, 60, 761-765.

30.

Steadman

H. J.

Redlich

A. D.

Callahan

Robbins

P. C.

Vesselinov

(2011). Assessing the impact of mental health courts on arrests and jail days: A multi-site study. Archives of General Psychiatry, 68, 167-172.

31.

Steadman

H. J.

Redlich

A. D.

Griffin

Petrila

Monahan

(2005). From referral to disposition: Case processing in seven mental health courts. Behavioral Sciences and the Law, 23, 1-12.

32.

Steadman

H. J.

Robbins

P. C.

Islam

Osher

F. C.

(2007). Re-validating the Brief Jail Mental Health Screen to increase accuracy for women. Psychiatric Services, 58, 1598-1601.

33.

Stefan

Winick

B. J.

(2005). A dialogue on mental health courts. Psychology, Public Policy, and Law, 11, 507-526.

34.

Thompson

Osher

Tomasini-Joshi

(2008). Improving responses to people with mental illnesses: The essential elements of a mental health court. New York, NY: Council of State Governments Justice Center.

35.

U.S. Department of Justice, Bureau of Justice Statistics. (2006). State court processing statistics, 1990–2002: Felony defendants in large urban counties (ICPSR02038-v2). Ann Arbor, MI: Inter-University Consortium for Political and Social Research.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

9.55 MB