Authors Participants Method Measurements Outcome/Conclusion Peck (2003) 35 Police department and 6 sheriff's department Participant survey Author-created survey: assessing police interaction with mentally ill CIT-like programs are effective Hails and Borum (2003) 84 Medium and large law enforcement agencies Survey None Varied responses regarding training and use of specialized response Finn and Stalans (2002) 257 Police officers from North Georgia in-service training Survey––responses to scripts: analyzed using ANOVA None Officers are significant more likely to use civil commitment (p < .05) based on perceptions Teplin (2000) and Teplin and Pruett (1992) 283 Police officers and 85 mentally disturbed persons Observational, in-field study Symptom checklist and "incident coding form" Mentally ill suspects were arrested more often (46.7%) than nonmentally ill suspects (27.9%) Engel and Silver (2001) 322 Police officers and 1,849 nontraffic suspects Systematic observational in-field study Incident coding form Police are not more likely to arrest mentally ill civilians, 7.6% vs. nonmentally disordered civilians, 18.2% Catalano et al., 2005 State of Florida's archived data on involuntary psychiatric hospitalizations initiated by law enforcement in specified time periods between 1999 and 2001. Interrupted time-series design None Law enforcement initiated increased involuntary psychiatric hospitalizations in the weeks following the attacks of September 11, 2001. Concluding that perceived community risk may increase an officer's judgment that a person with mental illness is dangerous. Brakel and South (1968) Two police counties Observational, in-field study Records review of Police are able to discern mental status but lack community collaboration and referral support.     1. Hospital records     2. Police records Sheridan and Teplin (1981) 838 Police referred Pre/postuncontrolled Records review of Reduced recidivism and days as inpatients at CMHC as opposed to state hospital. (CMHC mean days as inpatient = 33.2 vs. state hospital mean days as inpatient = 137.5.)     1. Demographics     2. Pathology of patient     3. Type: police contact     4. Treatment and recidivism Green (1997) One Law Enforcement Officer (LEO) agency with one consulting agency and one in-house program Case study evaluation using both qualitative and quantitative analysis Quantitative: author-created incident forms The more senior of an officer, the less likely they will arrest or refer (–.26, p > .05). Police need more options. Qualitative: interviews Borum et al. (1998) Three LEO agencies with three different specialized programs; total n = 452 Case study evaluation with cross analysis within and between each case. Patrol officer survey Memphis CIT model was effective in maintaining safety (94.4%) and meeting needs of mentally ill (88.8).     1. 4-point Likert scale     2. Open-ended questions Strauss et al. (2005) 485 Patients from Louisville University Hospital brought in by Louisville CIT over period of 1 month Chi-squared used to analyze results from this time-series study Records review interview Data support effectiveness of CIT in identifying psychiatric emergencies Deane et. al (1999) 174 Police departments from 194 U.S. cities with a Population of 100,000 or more Survey analysis comparison Department survey 55% departments had no specialized and 3% had a Memphis CIT-like model.     1. 5-point Likert scale response.     2. Open-ended questions Steadman et al. (2000) Same participant sample from Borum et al. (1998) regarding response to Retrospective: to cross analyze each program regarding response to mental health emergencies Records review Memphis CIT responded to 95% of mental disturbance calls (p < .001) and arrested only 6% (p < .001). Watson and Angell (2007) None Analytical review None Support use of procedural justice in training officers to respond Dupont and Cochran (2000) Outcome data on Memphis CIT Literature review None Supports Memphis CIT model