Finding High Risk Persons with Internet Tests to Manage Risk —A Literature Review with Policy Implications to Avoid Violent Tragedies, Save Lives and Money

The goal is to share policy implications of sensitive, specific internet-based tests in place of current approaches to lowering violence, namely fewer mass murders, suicides, homicides. When used, internet-based tests save lives and money. From 2009-2015, a Chicago field test had 324 fewer homicides (saving $2,089,848,548, ROI=6.42). In 60 yrs., conventional approaches for high risk persons (e.g.,. inappropriately releasing poor, severely mentally ill) led to unnecessary expense including yearly: (a) 300 mass murders (59% demonstrating psychiatric conditions); (b) 1-6% having costly personnel challenges; (c) 2,100,000 “revolving door” Emergency-Room (ER) psychiatric admissions (41,149 suicides, 90% mentally ill); (d) 10,000,000 prisoners (14,146 homicides, 20% psychiatric challenges). Current metrics fail [success rates from 25%-73%: (1) for background checks (25%); (2) interviews (M=46%); (3) physical exams (M=49%); (4) other tests (M=73%)]. Internet-based tests are simultaneously sensitive (97%), specific (97%), non-discriminatory, objective, inexpensive, $100/test, require 2-4 hrs.

wonders if the decrease in violent crime convictions could have been even greater had internet-based tests been employed.
Contrast internet-based tests with inadequate, insensitive (not finding the actually high-risk persons), nonspecific (not avoiding falsely labelling low-risk persons) interviews and physical and unstructured psychiatric examination.Also if these violent psychiatric patients were targeted with jobs, anger management, and mentors, these approaches would result in a greater diversion from violent crime saving more lives and money.

The Benefits in Finding At-Risk
Many think that severely mentally ill people are harmless.Indeed most are.
However, some persons with mental health issues commit mass murders, active shooting incidents, suicides and homicides.Their violent actions are related to personal mental health problems in 20% to 59% of the incidents studied.
See Figure 1.According to the Federal Bureau of Investigation (FBI), from 2000 to 2013, there were 160 active shooters.
An active shooter is a person killing or attempting to murder people in a confined, populated area.In contrast, mass murders have at least four victims (Blair & Schweit, 2014).
Depression, psychosis and bipolar disorder led to active shooter violence (Foxx & Levin, 2011).One wonders how much the deinstitutionalization of the severely mentally ill combined with the inadequate, insensitive (not finding the actually high-risk persons), nonspecific (not avoiding falsely labelling low-risk persons) interviews, physical and psychiatric exams, rather than objective, sensitive, specific, internet-based tests, plays in this rise in mass murderers and active shooters.
In the top curve of Figure 2, in 2013, some 41,149 Americans (.01% of the population) were suicidal (Center for Disease Control, 2014).From 1900 to 2013, suicide rose from 11 to 12/100,000; 90% of suicides are related to severe mental illness issues (Barriera, 1999).
Historically, suicides rose during economic recessions.
Since the American Revolutionary War in the 1700s, there are more homicides on the U.S. streets than battlefield mortalities.Thus, living in some America city neighborhoods is more dangerous than living in a conventional warzone.
Murders decreased during the post-World War II and dot.com era economic booms, when more jobs were consistent with a lowering of homicide rates.
As seen in the bottom curve of Figure 2, the fourth leading cause of death among male workers was homicide.The leading cause of death among women laborers was workplace homicide (U.S.Department of Labor, Occupational Safety and Health Administration (OHSA), 2014).Riedell (2003) compared national and workplace homicide rates.He found the curves were similar.
The national homicide rates were 22 times greater than workplace murders.Over 35 years, annual workplace homicides dropped from 1,000 to 400; 25% of yearly workplace homicides were due to robberies.If one included just off-the-work-site murders, these annual work place homicide rates doubled to 800 (Zagar, Kovach, Basile, et al., 2013).With more jobs, homicides decrease.And 25% of workplace homicides are driven by robbery with the rates increasing when just off-site murders are included.
See Figure 3 for a comparison of the homicidal with and without prior arrest.An average of 16% of the killers did not have an arrest, which makes a good case for internet-based tests to find the at-risk and divert the violent-prone from murder (Zagar, Busch, Grove, & Hughes, 2009c).

Evidence that Internet-Based Tests Are Better than Current Ways
As seen in Figure 4, internet-based tests like the Standard Predictor for Adults or Adolescents have a sensitivity and specificity as measured by Area Under the Curve (AUC) or Receiver Operating Characteristic (ROC)=91 to 99 (Zagar & Grove, 2010).To contrast approaches, AUC, ROC, (sensitivity and specificity), interobserver reliability, and percentile hit or miss are listed as decimals and in this and following figures for comparison although not precisely equivalent mathematically.
The current ways on average have a miss rate of 61%.The current ways are physical-psychiatric exams, interviews and judgment, background and credit checks, and short risk tests.
Figure 4. Internet-based tests are better than conventional ways: .91-99 vs. .25-76(current ways miss 60%) One would be better flipping a coin than using current ways.From this Figure 4, one can see conventional approaches have a hit rate on average of 39% or 1 out of 3 cases and a miss rate of 3 out of 5 cases.
The risk tests have an average of 73% hit rate or 3 out of 4 cases but a miss rate of 1 out of 4 persons.The Standard Predictor in addition to conventional approaches has a hit rate of .91 to .99.
As the reader will note later, because of the conventional use of interviews, background checks, physical and psychiatric exams and short tests, Rice, Harris, and Quinsey (1996) demonstrated that forensic providers seemed most ready to release without supervision those most likely to reoffend violently.These same professionals were retaining or recommending intense supervision for the least dangerous patients.See Figure 5 for the physical examination inter-observer agreement that has range from rs = .20-.85 depending upon the medical specialty, presenting illness, and the complexity of the sickness.The mean interobserver reliability was .49.This could be interpreted as a 51% miss rate.
In an unpublished study from the American Board of Medical Specialties (1986), inter-observer agreement among physicians was from .20 to .70 depending on whether psychiatrists, internists, or surgeons were studied.Loke, Liaw, Tiong, Ling and Chang (2002) reported that among 2,200 nurse-physician observations, interobserver agreement were from .27 to .37 on cardiovascular, respiratory, gastrointestinal, urinary, endocrine, joint/skeletal, cerebrovascular, infectious, muscular, oncology, psychiatry, gynecological, eye-nose-throat, dermatology, trauma, fracture, and burns.Madan and Harley (2003) reviewed 114 segments of 23 limbs with inter-observer agreement were from .39 to .52.Among 694 patients grading breast sample nodes, the inter-observer agreement was from .56 to .85 (Bueno-de-Mesquita et al., 2009).Perez-Stable, Miranda, Munoz and Ying (1990) found that physicians under-recognized and misdiagnosed depression in 64.3% of 70 patients.
In another study, Singh, Meyer and Thomas (2014) estimated 5% of patients are misdiagnosed every year, which is likely an underestimate.In a random sample of 169 adults (41 males, 28 females, Mage = 44.6+11.3)and 31 adolescents (12 boys, 11 girls, Mage = 13.9 + 7.5, 61% had a disagreement on primary, secondary and/or tertiary assessment. After seeing these kinds of inaccuracy, insensitivity and lack of specificity, one wonders why deterministic algorithms are not used more often rather than conventional subjective human analysis (Siegel, 2015).Tossing a coin would be more precise than either conventional approach of background/credit check, judgment/interview or physical/psychiatric exam.
Yet for decades, clinicians, executives, human resources, judges, lawyers, among others rely on current ways in important decision making.This occurs despite the inaccuracy 54% to 61%.
To support deterministic algorithms, in 128 of 131 studies (Grove & Meehl, 1996), risk tests were found to be superior to clinical judgment.For more detail on these 131 studies please read Grove, Zald, Lebow, Snitz, and Nelson (2000).
See Figure 6 for the comparison of how inaccurate clinical judgment is.In 4 studies on the inaccuracy of forensic examiner, the hit rates are .39,.53,.64 and .30with M = .46hit rate.Forensic professionals miss 54% of risky dangerous persons.This is close to the combined inaccuracy of background checks, interviews, physical and psychiatric exams, and short risk tests that have miss rate of 61%. Figure 6.Interview-judgment in violence prediction miss rate 54% in finding at-risk persons Note.Interviews and judgment as a percent converted to decimals (Sepejak, Menzies, Webster, & Jensen, 1983;Lidz, 1993;Monahan, 1996;Rice, Harris, & Quinsey, 1996), Background, credit checks (Quinsey, Harris, Rice, & Courmier, 1998, 2006, 2015).
With clinical judgment, Sepejak, Menzies, Webster, and Jensen (1983) discovered that providers over 2 years were only 39% precise in rating potential violence.Six months post discharge, Lidz (1993) showed that mental health clinicians correctly predicted only 53% of violent patients.
Likewise, 6 months after release, mental health care professionals predicted only 64% of "safe" patients.With dangerous, institutionalized patients, Monahan (1996) found that mental health professionals were only 30% accurate in prediction of violence which is a 70% miss rate.
One of the reasons for misdiagnosis or a lack of sensitive and specificity besides conventional inaccurate approaches, noted above, is deceptive self-presentation.There are seven reliable, valid scales of deceptive self-presentation (lie, infrequent responding, defensiveness, superlative self-presentation, response inconsistency, variable response inconsistency and infrequency at back end of test) on the Minnesota Multiphasic Personality Inventory Second Edition (MMPI-2) (Pope, Butcher, & Seelen, 2004).
Because there is a Poisson's distribution of deception (7 x 6 x 5 x 4 x 3 x 2), there are 42,300 deceptive self-presentations.As seen in Figure 7, when the deceptive self-presentations are multiplied times 1,000 neurological and psychiatric illnesses (Diagnostic Statistical Manual Fifth Edition (DSM-V), American Psychiatric Association, 2013); International Classification of Diseases (ICD-9/10) (World Health Organization, 1977, 2013)

Executive Function Correlates with Other Tests and Test Item
To shorten test time, a series of studies were done to identify only the most essential elements; 4 subsamples received the Trail Making Tests A and B, Zagar Executive Function Test, and 1 Likert format item "I make poor decisions": (a) n=48 randomly selected from 181 abused children later homicidal teens with 181 controls, (Hughes, Zagar, Busch, Grove, & Arbit, 2009); and (b) n=48 randomly selected from 127 homicidal youth with 127 controls (Zagar, Busch, Grove, Hughes, & Arbit, 2009d).
Arrests for violent crime decreased 43% among the 2 treatment groups compared with controls.The $3,000/student ($1,400 for wages and $1,600 for administrative costs) had a yield of $1,700 in benefits from reduced crime with a return on investment or ROI=$0.57(Heller, 2014).
In layman's terms, this means a loss of $0.43 for every 1 U.S. 2012 dollar spent.
In contrast, the "2009 to 2012 program within Chicago," a risk test regression was used to identify violent prone students with evidence-based diversions applied to 4,850 students with ROI=$6.47.The difference in ROIs was $0.57versus $6.42 with the latter targeting approach 11 times more cost beneficial and effective (Zagar, Grove, & Busch, 2013;Zagar et al., 2013).

Targeting of jobs resulted in a positive investment while random selection which is what most providers and decision maker now employ did not.
Public programs can do more with less by shifting from remediation to prevention that is targeted, not randomly assigned (Zagar, Grove, & Busch, 2013;Zagar et al., 2013;Heller, 2014;City of Chicago, 2014).This random assignment of treatment which lost money is the current conventional intervention approach.
Current applications of evidence based diversions do not have benefit or effect, because conventional ways don't involve targeting highest at-risk to save lives of individuals and money.There are savings of: (d) in 2013, 89 lives and $455,600,000; (e) in 2013, n=11 lives and $50,600,000 from targeting adult at-risk with jobs; and (f) in 2014, n=76 lives and $380,000,000 with the current city mayor.

Summary
The total savings from 2009 to 2015 is 324 lives and $2,005,848,548 (Saulny, 2009;Shelton & Banchero, 2009;Kapos, 2014;City of Chicago, 2014;Zagar et al. 2013;Zagar et al., 2013).Chicago Public Safety Fund efforts of targeting jobs to at-risk youth.Critically, there were no recidivists or persons who returned to court among 56% of the more than 29,710 nonviolent offenders released from prison to electronic bracelets that is still going on today.
A similar drop is projected in the federal prisons, the first since 1980, and a fall of 12,000 inmates over 2 years.Being aware of the Midwestern movement of nonviolent to electronic bracelets with no recidivists by the Chicago and Cook County leaders, there was a concerted effort to reduce long mandatory minimum sentences of nonviolent offenders in federal prisons (Grossman, 2014) with United States Presidential commutations.

Application to Various Work Groups
There are ten work groups that could benefit from internet-based tests, namely airlines, corrections, education, energy infrastructure, health facilities, military, nonprofit-religious, police, transportation, and veteran organizations with savings, internet and paper and pencil costs and population in Figure 10.In Table 4, as the reader will clearly see the 10 work groups make up 40% of the U.S. work force and 15% of the U.S. population.By employing internet-based tests, 1 to 10% of the at-risk can be found and targeted with evidence-based diversions.These abovementioned costly risks are mass murder, suicide, homicide, PTSD, mental illness, and substance abuse.
There is a saving sare $104,206,500,000 with Returns On Investment (ROIs) of $2 to $323 for every $1 spent on internet-based tests.Not employing internet-based tests and continuing to use conventional approaches of background checks, interviews, and exams only results in continued lost lives and money due to violence.

Pilots
In the U.S., there are 104,000 commercial pilots and 613,000 private pilots.Internet-based tests are consistent with helping decision makers identify pilot health risks and terror actions (New York Times, 2010).Risks include Air Force pilots with depression, PTSD, substance abuse/dependence, and suicide attempts (Otto & Webber, 2013).
In September 2014, a technician at an air-traffic control center set fire to the work station, which caused airlines to lose hundreds of millions of dollars when the center was closed for weeks (Guy, 2014;Slodysko & Rossi, 2014).There is also the use of intoxicants in the cockpit.
To address this issue, 3 agencies (Federal Aviation Administration (FAA), the International Civil Aviation Organization of the United Nations and the European Aviation Safety Agency) have pilots go through annual physicals, and biannual exams after age 40 years.In the British Commonwealth, namely Canada, there are annual physicals and biannual after age 60 years.

Internet-Based Tests Can Help Lower Military Violence
There were psychiatric, substance abuse and violence issues in the military.In 2006, the U. S. Army had a rejection rate of 1,472 of 43,574 recruits after training; 19% of recruits were discharged for unidentified psychiatric issues; 1.2% of U.S. Marines were discharged because of substance abuse; 1.0% of enlisted recruits dropped out.
Recruitment interviews were consistent with 13.9% soldiers considering suicide, 5.3% with a suicide plan, and 2.4% attempted suicide with between 47-60% of these outcomes existing prior to joining the military.Often, PTSD issues led to suicide (Davidson, Hughes, Blazer, & George, 1991).
Costly offenses are perpetrated by military personnel.In 2004, there were 1,798 sexual assault claims, of which 672 resulted in punitive damages (USDOD Sexual Assault Prevention and Response, 2006); 32% of enlisted women felt sexually harassed (Bostok & Daley, 2007).
This could be explained by the fact that 13% of U.S. Navy enlisted men had committed sexual assault prior to service (Stander, Merrill, Thomsen, Crouch, & Milner, 2008).In Figure 12, the number of risks or incident diagnoses was going up, which was a good reason to use internet-based tests.The top curve has total suicides from 1980 to 2013.There are 8,452 costing $8,452,000,000.The middle curve has the total homicides from 1980 to 2013.
There are 2,341 costing $2,341,000,000.In Figure 13, the mass murders, suicides and homicides in the active duty armed forces from 1980 to 2013 are presented.If one assumes that a life was worth $1,000,000, then multiply each year's death by this amount.
There were a total of 30 mass murders from 1980 to 2013.These cost $30,000,000.Recently, among the general U.S. population, there were 300 mass murders in one year.Risk tests were consistent with more accuracy than drug testing (Non-Instrumented Drug Test) introduced to the U.S. Navy, which reduced positive boot camp drug tests by 30%, saving $3,300,000 annually (Norbert, 2002).A risk test was consonant with detection of a far greater range of potential problems with better sensitivity and specificity.
The cost of finding, recruiting and boot camp training is $46,000, with an attrition rate of 19%.So that was $1,800,000,000 lost every year by training persons with mental health, substance abuse, or violent tendencies. In

Personal Injury Cases
In the U.S. yearly, there are 2,000,000 personal injury cases.The average cost is $16,000, with a 2 year time to trial or a mutually agreed upon settlement, representing $32,000,000,000.
Over half involve automobile accidents, followed by medical malpractice and product liability.Auto accidents settle for a median $16,000, medical malpractice, $679,000, and product liability, $748,000.
Plaintiff success rates are 60% for auto accidents, 40% for product liability, and 20% for medical malpractice (Terry, 2011).The cost of a paper-and-pencil exam is $2,000.
The internet-based test costs $100 resulting in savings to the insurance carriers.In Table 4, Column 5 the internet-based tests cost $40,000,000 (Table 4, Column 6).The savings are $3,600,000,000.There is an ROI of 10 + (Table 4, Column 7) for every $1 spent.

Public Safety and Police
Annually, one of 154 police is accused of inappropriate or illegal violent behavior.One homicide out of 1,549 deaths caused by public safety personnel was inappropriate or illegal.
So roughly 1% of police may be assumed to be at-risk, given that accurate statistics were not collected.With 1,200,000 public safety officers, there are 200 yearly wrongful injury or death claims.
Between January 2009 and November 2011, the City of Chicago treasurer paid $455,000 for 441 lawsuits-a rate of $5.54 annually per city resident.Between 2009 and 2010, that is more than twice as much as in Los Angeles ($2.66), and roughly half as much as in New York City ($9.93) (Caputo, 2012).New York City officials paid out $100,000 yearly, costing $1,000,000,000 in a decade (Fields & Jones, 1999;Fyfe & Kane, 2005).
That is $2,575,000,000 wasted human capital (direct and indirect expenditures) for the police lost in one year.
Chicago has the highest police liability payout per citizen in the U.S.
Since 2004 misconduct legal claims against the Chicago Police Department cost $542,000,000 (Schroeder, 2016) with 450 current outstanding police misconduct lawsuits.In Table 4, Column 4, the cost of screening for all public safety officers in the U.S. is $2,400,000,000.
For every police officer killed or police officer suicide multiply $1,000,000 the minimum cost of training and replacing that public safety officer.
Figure 18.Annual U.S. police officers killed and suicides from 1998 to 2013

Prisons and Courts
One third of the annual cost of crime in the US ($37,000,000,000) is the expense of running the U.S. prison system.Imagine what could be done by moving 50 to 80% of the prisoners to electronic bracelets and the savings that would be generated.
Currently, because the Arnold Risk Test is free, it is being used in 16 states despite the fact that it has a miss rate of 30% of violent prone.To assure the safety of the community, internet-based sensitive and specific tests could be used to lower the risk and manage it.
In Cook County, the president released 56% nonviolent offenders to electronic bracelets with no recidivism and no lowering of community safety (Olson & Taheri, 2012;Zagar, Busch, & Grove, 2013).There is a cost incentive to releasing nonviolent offenders to electronic bracelets.
For example, in New York, the annual prison expense is $3,267,105,290.Nonviolent prisoners make up $2,705,422,825 of that cost.Internet-based testing and electronic bracelets would cost $126,353,500.
From state to state, the ROIs of moving nonviolent offenders to electronic bracelets and assuring the safety of the community by using internet-based tests to screen these prisoners range from $2 to $20 (Table 4, Column 7).This allows states to release a majority of those behind bars thus freeing funds to spend on the decaying infrastructure.
Annual large truck crashes cost $24,000,000,000 in 1977 ($13,000,000,000 for quality of life, $9,000,000,000 for productivity loss, $1,000,000,000 for property loss, $941,000,000 in expenses, and $58,000,000 for emergency service) (Miller, 1997;Zaloshnja, Miller, & Spicer, 2000 Imagine using internet-based tests to address the psychiatric, substance abusing and violence risk of transportation and port workers.Table 4, Column 4 has the $9,000,000,000 cost for paper-and-pencil tests.In Table 4, Column 5, the internet-based tests cost is $450,000,000. The saving is $8,550,000,000 (Table 4, Column 6).The ROIs are $10 to 100 for every $1 spent.In Figure 19, basically 48% of persons could be saved from homicide with empirical diversions because 8 + 8 + 14 + 18 = 48% out of 1,200.The largest effects of targeting evidence based diversions are in teen and adult years.If one converts the total diverted in infancy, 24/145 = .17,in childhood, 25/145 = .17,in adolescence, 42/145 = .29,and in adulthood, 54/145 = .37,then one can compare infancy-childhood, 17 + 17 = 34, adolescence, 29, and adulthood, 37 which add up to 1.00 allowing a clear comparison of the relative effects of targeting evidence based diversions across development.
Perhaps focusing on at-risk adults with the same evidenced based diversions of jobs, mentors, and anger management might be economical.With the increasing slope of the effect of evidence based diversions in adulthood, perhaps there might be a better ROI in adulthood.Also in the future targeting high risk infants-children is likely economical in lives and expenses saved.
Each year, newly discharged military personnel join with the 22,000,000 total veterans (Kessler et al., 2014; Army Study to Assess Risk and Resilience in Service members (Army STARRS), 2004-2009;Schoenbaum et al., 2014;Nock et al., 2014).Veterans have risks for mental illness, substance abuse and violence.

Workers' Compensation
In 2009, there was over $600,000,000 in workers' compensation claims.The medical payments made up 53% of the costs.Indemnity was the remainder.In 2013, there were $106,000,000 in claims, $24,000,000,000 in cash payments with an average weekly off work benefit of $832, $17,000,000,000 employer insurance premiums, 4,000 workplace deaths and 40,000 workplace injuries (Insurance Information Institute, 2015).
To get an idea of the diagnoses among workers' compensation claims one can look at medication.Among 60,000 claimants, the most common medication was a pain killer.Second were anti-depressants (Willingham, Gallogly, & Morrell, 2000).So pain was first and depression second.
In order to focus evidence-based treatments on the depression and to reduce the claims, perhaps internet-based tests could be used.Paper-and-pencil exams cost $1,200,000,000 (Table 4, Column 4).
A case study might be the recent professional football player homicide case.During Hernandez's college career in Florida Tebow tried to intervene in a bar fight.The professional club had personality test results for the perpetrator during pre-employment screening.
These were consistent with an aggressive, angry and assaulting person "living on the edge of acceptable behavior".The football club was cautioned about him "becoming a problem for the team" according to Clegg (2013).
Jones ( 2013) noted that the East Coast professional football club losing $15,037,000 in the signing bonus, and a major hit to their salary cap with his guaranteed salary despite being convicted of homicide and in jail, but Rishe (2013) believed the loss is closer to $12,700,000.The ROIs for this case study are 150,370 and 127,000 respectively for every $1 spent given the cost of $100 internet-based test.

Diverting At-Risk Individuals from ER Hospitalization
At-risk people come to hospital Emergency Rooms (ERs).In 2014, there were 136,300,000 ER visits Over 10% (16,200,000) resulted in hospitalization.Just over 10% (2,100,000) of those were psychiatric hospitalizations.
The average psychiatric hospitalization increased from $5,800 in 2003 to $6,400 in 2011.In 2003, the normal length for a psychiatric stay was 6.9 days, which went up to 7.4 days in 2011.
In 2003, there were 1,800,000 psychiatric discharges which rose to 2,100,000 in 2011.That is nearly a half a million psychiatric hospitalizations or 400,000 to 450,000 per calendar quarter.
Compared with maternal neonatal, medical, surgical, and injury hospital stays, the psychiatric hospitalization was the longest (Weiss, Barrett, & Steiner, 2014).Psychiatric ER visits cost $56 which is more than all other ER visits.
One percent of all health expenditures were for mental health and substance abuse.In 2009, the spending on substance abuse prescription medications was $887,000,000; $171,720,000,000 was paid to providers for mental health and substance abuse.
This can be broken down into $147,381,000,000 for mental health.The remaining was spent on substance abuse.
There was $10,461,000,000 for inpatient, outpatient and residential services (U.S.Department of Health and Human Services, 2009;2014).Clearly sensitive and specific internet-based tests are consistent with a lowering of the 2,100,000 psychiatric hospitalizations.
Not only could these hospitalizations be reduced but those psychiatric hospitalizations that are necessary would be shorter with improved sensitivity and specificity of diagnosis so that evidence based treatments could be applied.Given the inaccuracy and lack of precision of conventional approaches of background checks, interviews, physicals and short tests, the violence from mass murders, suicides and homicides would also decrease significantly.
Tests are also consonant with a lowering in the number of return psychiatric ER visits that do not result in hospitalization.So millions to billions could be saved yearly and money generated for health rather than funneled to the prison system.
Physicians are losing millions to billions of dollars in revenue.Communities are less save because of miss rates for violence.

Moving Severely Mentally Ill from Hospitals to Prisons: 1959-2013
Deinstitutionalization of psychiatric patients began in the 1940s.Beginning in 1955, millions of mentally ill individuals were set free (Stroman, 2003).
What is more, the release of the 94% of severely mentally ill without a concomitant use of effective risk assessment (conventional approaches of interviews and physicals) was directly associated with a rise in homicide, mass murder, active shooting incidents, and suicide rates.In Figure 20, one has the numbers that are consistent with the movement of psychiatric beds to jail cells over time with a significant inverse correlation of r = -0.852,p < .01(U.S. Bureau Justice Statistics, 2014; Statistical Abstract).
Around 1980, poor without support migrated from psychiatric beds to jail cells.Today, the largest asylums are prisons.
Figure 20. 1900-2014 prison 1955-2011 psychiatric bed rate/100,000 with mentally ill moving to jail cells In Figure 21, the increasing size of the U.S. prison population is shown.This is historically the largest ever in the history of mankind (U.S. Bureau of Justice Statistics, 2014) Meanwhile, there are nearly 350,000,000 guns in the U.S.This might be related to the public perception that communities are unsafe because the release of mental patients and the current inaccurate approaches in evaluation has led to an increase in the mass murders, suicides, and homicides and a rise in the prison population.
Economics and science should be applied to the growing violence by using internet tests to assist in assessment to lower the violence rates rationally.By using internet-based tests, prisons would be safer for workers and courts safer for judges and personnel.
Furthermore, jobs would not be lost by prison and court personnel.There would be less spent per prisoner by using electronic bracelets rather than room and board in jail.

Ten Prisoner Characteristics and Empirical Diversions from Violence
In Figure 22, there are ten characteristics of offenders.First, 90% of the prisoners are poor, so jobs would help.1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005   Tenth, 22% have severe mental illness challenges, which can be addressed with multidimensional foster care (ROI = 11.20),teen courts (ROI = 9.83), low risk offender restorative justice (ROI = 8.03 in the community or within the prisons' cognitive behavior therapy (ROI = 98.09) and treatment oriented intensive supervision (ROI = 1.62).This set of ten descriptors is similar to Standard Predictor algorithm for predicting homicide, pedophilia, and violence for both adults and youth (Zagar & Grove, 2010).
This set of ten characteristics is consonant with confirming the hypothesis of the sensitivity and the specificity of the cross validated and thrice replicated Standard Predictor predictive equation in the percentages of the offender population having these characteristics.In Figure 23, there a sample of at-risk first presented in Table 3 of Empirical risk factors for delinquency and bests treatments: Where do we go from here?(Zagar, Busch, & Hughes, 2009).In Chicago applying the teen and adult solutions together, alternate thinking, life skills, quantum opportunity and job corps saved 324 lives and over $2,089,000 in Chicago from 2009 to 2013 with a ROI = 6.42 for every $1 spent.If one converts the total diverted in infancy, 24/145 = .17,in childhood, 25/145 = .17,in adolescence, 42/145 = .29,and in adulthood, 54/145 = .37,then one can compare infancy-childhood, 17 + 17 = 34, adolescence, 29, and adulthood, 37 which add up to 100.
Perhaps focusing on at-risk adults with the same evidenced based diversions of jobs, mentors, and anger management might be economical.With the increasing slope of the effect of evidence based diversions in adulthood, perhaps there might be a better ROI in adulthood.Also in the future targeting high risk infants-children is likely economical in lives and expenses saved.
But what is the larger picture of why tens of thousands die and tens to hundreds of billions are spent, if not a trillion, because of inaccurate conventional approaches of assessment and inadequate monitoring of interventions that are not evidence-based?In Figure 24, there are three triangles, the first small, the second bigger, and the third largest representing risks, assessments, and diversions and interventions.The first triangle has the 7% combined at-risk: (1) abuse or violence; (2) severe mental illness; and (3) substance abuse (with brain damage, cognitive delay-retardation, deceptive self-presentation, and pedophilia-sex offending in the center of the triangle of illness or lowered functioning).
These three triangles are nested in each other with the small risks in the center, followed by the bigger current approaches of assessment, and finally on the outside, the largest for diversions.The three triangles are spinning together around in a circle with the assessment triangle acting as a net to catch the risks and the diversions to intervene and prevent the violence (mass murders, homicides and suicides for escaping) or loss in lives and expense.The present interventions of 34 for infants and children, 27 for teens, and 39 or a combined would thus lower violence or the mass murder, homicide and suicide victims that escape the whiffle ball holes.Current ways of assessment miss 60% who are victims of violence costing lives and funds.
In Figure 25, there are the three triangles for the risks, current ways of assessment, and diversions.
In the first triangle at the corners are costly violence, mental illness, and substance abuse with traumatic brain injury, cognitive delay, lying and sex offense-pedophilia in the center making up 7% of any population.Second are the current ways of assessment interview, background check, and physical-psychiatric exam on the corners of the triangle.The arrows are a way to show that these current ways miss roughly 60%.Third are the diversions and interventions which help divert infants-children, 34, teens, 29, and adults, 37. Fourth, there are three triangles, small with risks, bigger with current ways, and largest with diversions and interventions.With the miss rate of 61% the persons at-risk have the wrong diagnosis or assessment.
In Figure 26, the 97% sensitive 97% specific internet-based tests will result in less violence, namely fewer mass murders, homicides, and suicides or victims escaping out of the whiffle ball.The three triangles are the risks, current ways of assessment plus internet-based tests, and diversions.
In the first triangle at the corners are costly violence, mental illness, and substance abuse with traumatic brain injury, cognitive delay, lying and sex offense-pedophilia in the center making up 7% of any population.Second are the current ways of interview, background check, and physical-psychiatric exam on the corners of the triangle.The arrows are a way to show that these current ways with internet-based tests only miss roughly 3%.
Third are the diversions and interventions which help infants-children, 34, teens, 29, and adults, 37. Fourth, there are three triangles, small with risks, bigger with current ways, and largest with diversions and interventions.Diversions will likely improve with better assessment.
Figure 27.The annual number of U.S., E.U., & U.S. + E.U. victims of violence Violence is rising in the U.S. and E.U. due to 60% miss rate of current ways of assessment.In Figure 27, there are estimates given the vagaries of precision in accounting for violence rates in the U.S. and E.U. that may not be equal.There are 1,200 deaths from 300 annual U.S. mass murders, 14,100 homicides, and 41,100 suicides with a total of 56,400 victims.The E.U. has double the population, so there are the estimates of 2,400 deaths from 600 E.U. mass murders, 28,200 homicides, and 82,200 suicides with a sum of 114,800 victims.There are an estimated total of U.S. and EU mass murders or 3,600 total, 42,300 homicides, and 112,300 suicides with a total of 158,200 victims every year.There are also increasing U.S. and E.U. worker productivity losses likely due to 60% miss rate of current ways.In Figure 28 with the work productivity of a life at $1,000,000 U.S. 2012 dollars, the U.S. yearly loss from mass murders is $1,200,000,000, homicides, $14,100,000,000, and suicides, $42,100,000,000 with a total of $56,400,000,000.With the work productivity of a life at $1,000,000 U.S. 2012 dollars, and the E.U. population double that of the U.S., the E.U. yearly work productivity loss estimate from mass murders is $2,400,000,000, homicides, $28,200,000,000, and suicides, $84,200,000,000 with a total of $112,800,000,000.With the work productivity of a life at $1,000,000 U.S. 2012 dollars, and the E.U. population double that of the U.S., the U.S. combined with E.U. yearly work productivity loss estimate from mass murders is $3,300,000,000, homicides, $32,300,000,000, and suicides, $86,600,000,000 with a total $160,400,000,000.So for those with more risk, the return on investment is greater.For example internet-based tests are used in infancy-childhood, teen years, along the life span to lower costs of risks to agencies, government, individuals, and insurance companies as seen in Figure 29.
As shown in Figure 30, there are important economic and management issues that are consistent with using internet-based tests.Internet-based tests have a legal, financial and defensible position of saving lives and funds.
These internet-based tests are at or above the standard-of-care compared with current ways.Knowing the risks of insured persons allows agency managers to calculate reserves for accidents and other risks.
Internet-based tests are consistent with a lowering of risk and consonant with giving decision makers opportunity to make precise fund allocations for risk.Internet-based tests allow doing something about the unknown risks, violence, substance abuse, and mental illness with traumatic brain injury, cognitive-delay-retardation, deception or lying, and sex offense-pedophilia.In Figure 33 with the estimated work productivity of a life at $1,000,000 U.S. 2012 dollars, the U.S. yearly estimated loss from mass murders is $1,200,000,000, homicides, $14,100,000,000, and suicides, $42,100,000,000 with a total of $56,400,000,000.The estimated 50% reduction with internet-based tests is $6,000,000,000, 7,050,000,000, 21,050,000,000 summing to 28,200,000,000.
With the work productivity of a life at $1,000,000 U.S. 2012 dollars and the E.U. population double that of the U.S., the E.U. yearly estimated work productivity loss from mass murders is $2,400,000,000, homicides, $28,200,000,000, and suicides, $84,200,000,000 with a total of $112,800,000,000.The estimated 50% reduction with internet-based tests is $12,000,000,000, 14,100,000,000, 42,100,000,000 that have a sum 56,400,000,000.
With the estimated work productivity of a life at $1,000,000 U.S. 2012 dollars and the E.U. population double that of the U.S., the estimated U.S. combined with E.U. yearly work productivity loss from mass murders is $3,300,000,000, homicides, $32,300,000,000, and suicides, $86,600,000,000 with a total $160,400,000,000.The estimated 50% reduction with internet-based tests is $1,650,000,000, $16,150,000,000, $43,300,000,000 that has a sum $14,100,000,000. In

Discussion and Barriers to Entry
What's new and what does it all mean?There are 4 questions that were posed as hypotheses.These must be addressed.First, violence, namely mass murders, active shooting incidents, suicides and homicides, is related to economic trends.With fewer jobs, suicides increase.With more jobs, homicides decrease.
Due to deinstitutionalization of psychiatric patients, combined with inaccurate conventional assessment and a lack of monitoring, mass murders and active shooting incidents, suicides and homicides increased.
The 20% to 59% of these violent prone had mental illness.And current approaches had a miss rate of 60%.
Second, mental illness, substance abuse, and violence can be sensitively and specifically found with internet-based tests, which are better than conventional approaches (which have a miss rate of 54% to 61%).In this literature review of internet-based testing, the sensitivity (finding the actually high-risk persons) and specificity (avoiding falsely labelling low-risk persons) are 97%.
Mean test time is 2 to 4 hours compared with the paper-and-pencil 7 to 8 hours.In this review of the literature, there is an example, a study.For illustration if among the 236 there are 207 nonviolent and violent offenders in the sample of 236 had been diverted with the empirical treatment of a job, mentor, and anger management training, there would have been savings of 2 lives and $9,999,992 in resources.This is the cost of offenses prevented for an ROI of $6.64 for every dollar spent (Zagar et al., 2013).Just looking at the violent offenders and excluding the nonviolent, the ROI is $17.14 return for every $1 spent.
Extending these estimated savings of using internet-based tests or predictive analytics to various industries presented in this paper would result in ROIs from $2.33 to $323.00 for every $1 spent.Third, severely mentally ill individuals can be diverted from expensive psychiatric hospitalization or expensive return ER visits using sensitive, specific internet-based tests.
Also, since September 25, 2011, the President of Cook County diverted 56% nonviolent offenders from jail to electronic bracelets and other alternatives without return to court or danger to the community.President Obama copied this approach in granting presidential commutations to nonviolent, drug offending federal prisoners, most recently releasing 6,000 in November, 2015.
So the best way to stop a bullet is with a job while screening nonviolent prisoners for release to electronic bracelets and other alternatives can save funds.And business could save losing workers using predictive analytics.
By applying internet-based tests with current approaches to the various sectors of airlines, corrections, education, energy, health, military, police, religious, transportation, veterans, and understanding the concept of the risk triangle inside of the assessment triangle inside of the intervention triangle, one can see that lowering the escape of half to two-thirds of the high risk persons.
Doing something novel always faces the reluctance of organizations and their leaders to adopt a different strategy.Tests have been used for over 80 years and have excellent reliability, sensitivity, specificity and validity.
On the insurance company side, 10% of current employees must be freed up to teach new ones about processing claims.This is an extra added saving to the payer for psychiatric inpatient hospitalization.There are government regulations and the growing hospital and prison business that may be less likely to change.
The validity of this battery of tests for identification of unique risk factors among adults and adolescents gives administrators more options to save lives and lower costs proactively because the prediction of violence, especially homicide and pedophilia, is practical, reliable, and valid.Analyses comparing this sample with the demographics of the E.U. and U.S. populations are consistent with the suggestion that the cross-validated and replicated risks for violence, homicide, and pedophilia are generalizable to those broad areas (Zagar, Busch, Grove, & Hughes, 2009a).

Limitations
There are of course limitations and threats to validity.There may be issues with the size of the sample of only 236 adults and adolescents in that there was not true random sampling, but neither is there in the standardization of tests.
In any such study, there are validity threats due to history, selection, and expectancy bias.Official records may not accurately represent the amount of abuse, delinquency, crime, or other risks, the criteria upon which the accuracy of results are judged.
There may be some bias in the referral for examination.Perhaps other risks may be observed in larger samples and at different rates.There was heterogeneity of variance on some measures or risks, although for most risks, the assumptions of normality and homogeneity of variance were met.
With over 60 years of successful empirical treatments, some with 27 to 35% diversion rates, perhaps the diversion rate of 52% found in Zagar, Busch, and Grove (2013) in the "Chicago from 2009 to 2012" program might be high, but violence rates were greatly decreased and associated costs were saved; it is hard to argue with those facts.The fact that this was replicated in 2013 and again in 2014 gives this empirical, evidence-based approach to lowering violence more credibility.

Summary and Policy Implications
Because internet-based tests are much more sensitive and specific than current approaches (background checks that have a have a miss rate of rate of 75%, interviews that have a miss rate of 54%, physical exams that have a miss rate of 51% and short tests that have a miss rate of 27%), more widely employed internet-based tests could lower homicides, mass murders and suicides, and thus save lives and money.
In the past sixty years, conventional approaches of inadequate assessments and ineffective monitoring of high risk persons (e.g., inappropriately releasing of the poor, severely mentally ill to the community) have led to unnecessary suffering and expense.This suffering and expense includes: (1) 300 mass murders yearly (with 59% of perpetrators demonstrating psychiatric conditions); (2) 1-6% of personnel having costly human resource challenges in businesses, churches, and schools; (3) 2,100,000 annual "revolving door" ER psychiatric admissions with 41,149 suicides (90% related to severe mental illness); and (4) 10,000,000 prisoners, parolees and probationers with 14,146 homicides (22% associated with mental health challenges).
This failure to identify high risk persons is consonant with about 22 daily suicides among military veterans.While conventional metrics generally fail [with success rates ranging from 20% to 73% such as for background checks (25%), interviews (M = 46%), physical exams (M = 49%), and other short tests (M = 73%)], internet-based tests are simultaneously sensitive (97%) and specific (97%), non-discriminatory, objective, and inexpensive: $100 per test, requiring 2 to 4 hours to complete.
A Chicago field test with high risk youth from 2009-2015 prevented 324 homicides (saving $2,089,848,548).Homicides, suicides, mass murders and active shooting incidents are economic.Mental illness, substance abuse and violence or abuse can be sensitively and specifically predicted, treated and diverted beginning with internet-based tests and evidence-based interventions.
Psychiatric patients can be diverted from expensive psychiatric hospitalization or expensive return ER visits with sensitive, specific internet-based tests.Nonviolent offenders can be screened with internet-based tests and released with electronic bracelets and other alternatives without compromising community safety.
Predictive analytics or internet-based tests are cost beneficial for settings with high expense for personnel failure such as airlines, corrections, education, energy infrastructure, health facilities, military, nonprofit-religious, police, transportation, and veteran organizations.Confidence in the usefulness of risk tests should be high now that the risk test model has been applied three times in a large urban setting (Saulny, 2009;Shelton & Banchero, 2009).
Imagine $49,248,700,000 projected yearly saving nationwide with approximately 1,000 lives annually saved to work efficiently and provide for their families.The optimistic estimate is 42,150 lives and costs @ $1,000,000,000 in 2012 U.S. dollars, an expense of $42,150, 000,000,000, nearly forty-two trillion dollars in worker potential.
Giving jobs, mentors and anger management to nearly 10,000 youth from 2009 to 2016 diverted 27% of them from violence.With adults there is a projected 36% diversion and lowering of homicides.By lowering the gaps in assessment and diversions with science makes a peaceful, safer family, community, job, military, police and religious group area for better economic efficiency and productivity.
From a purely economic view that places a premium on human safety, the alternative of using targeted tests with diversions is not a choice but a necessity for a financially healthy, peaceful,

Figure 9 .
Figure 9. Chicago homicides: Target jobs lowering murders & move nonviolent to electronic bracelets

Figure 13
Figure 13.U.S. Department of Defense (DOD) military 1980-2013 homicides suicides mass murders Figur Church.Internet test for 8,441 priests @ $100 = $844,100 saving billions not to mention the victims scared by abuse.E.U.Polish Roman Catholic Church has similar declines in giving and affiliations.Poland has roughly 10% of the U.S. population.Roman Catholic religion is the dominant affiliation among 95%.Poland provides for 95% of E.U.Roman Catholic priests who are missionaries to other E.U. nations(Catholic Hierarchy, 2011).For simplicity assume the E.U. dollars or Polish zloty = 2012 U.S. dollars.At 10% of the U.S. lost yearly revenue $2,366,000,000 = $236,600,000 is the lost revenue from 2002-2013.Add estimated future cost of 1% of Polish priests = 285 @ $1,130,090 (the U.S. cost/pedophile priest) = $11,300,000 + projected lost revenue from 2014-2023 for 10 years @ $236,600,000 = 2,366,000,000 = $2,377,900,000.The total past and future loss is $4,980,500,000.A study comparing English and Polish language on the tests was done on low and middle SES persons (similar toZagar, Kovach, Basile, Hughes, Grove et al., 2013).There was a random sample of 100 bilingual English-Polish teens (63 boys, 37 girls) M age = 15.7 + 1.9 years and 100 bilingual English-Polish adults (62 males, 38 females) M age = 29.3+ 11.2 years.From the U.S. Department of Labor Dictionary of Occupational

Figure 19 .
Figure 19.Infant, child, teen and adult evidence based diversions compared in lives saved

Figure 22 .
Figure 22.Percent of criminal-delinquent population with characteristic or descriptor

Figure 23 .
Figure 23.Infant, child, teen and adult evidence based diversions compared in lives saved

Figure 24 .Figure 25 .
Figure 24.Three triangles of risks, assessments, and diversions, missing at-risk and lowering diversions

Figure 26 .
Figure 26.Internet-based tests have a 3%miss rate when added to current ways reducing violence

Figure 28 .
Figure 28.The annual work product loss of victims of violence in the U.S., E.U. & combined U.S. + E.U

Figure 31 .
Figure 31.The administrative, insurance, legal and management advantages of internet-based tests

Based Test Are Sensitive and Specific: .94-.98 Figure
, [40,300 x 1,000] there are 40,300,000 possible deceptive self-presentations.With that many deceptive self-presentations, internet-based tests can match the complexity of the person coming before an examiner.No wonder current approaches of judgment, background checks, interviews, physical exams, and short tests are no match for objective, sensitive (finding the actually high-risk persons) and specific (avoiding falsely labelling low-risk persons) internet-based tests.8.Sensitivity and specificity for BSS, MMPI, SP, QT and Ravens (records as the criterion)[N=236]In Figure8for the total test battery (BSS, MMPI, SP, QT, Ravens), the True Positives (TPs) are 245 of 253 (97% sensitivity).The True Negatives (TNs) are 449 of 464 (97% specificity).

.S. Total N = 54,675,000,0 00 40% Workforce 15% 1-10%mass murder homicide, suicide, PTSD, mental illness, substance $109,350,000,000 $5,143,500,000 $104,206,500,0 00 $2 to 323
(Bostok & Daley, 2007)dents recorded from 2003 to 2012 only 8 were suicides, a decrease of 50% from the prior decade with 12(CAMI, 2006; 2014).The 2015 pilot who crashed the flight was seen by 46 physicians which is a demonstration of why internet-based tests are needed(Eddy & Clark, 2015).Tests are consistent with replacement of inadequate conventional approaches of interviews, physical exams including psychiatric assessment and short tests that have a miss rate of 54 to 61% of at-risk.Figure 11.U.S. aircraft assisted pilot suicides and world commercial pilot suicides 1982-201512.Energy Workers 4,000 U.S. nuclear power plant controllers supply 20% of the electricity.The probability of some depressed or suicidal individuals among these controllers is ample reason to consider internet-based tests(Bostok & Daley, 2007).In the past, the senior author personally examined a cluster of 3 completed suicides at a Midwestern nuclear plant.There are about 600,000 workers involved in the various sectors of energy production(Bromet, Dew,  Parkinson, & Schulberg, 1988; U.S. Energy Information Administration, 2009; U.S. Nuclear Regulatory  Commission, 2013).The cost for screening for all energy workers is $1,200,000,000 (Table 4, Column 4) with internet test costs of $60,000,000 (Table 4, Column 5) saving $1,140,000 (Table 4, Column 6) with ROIs of $20 to $90 (Table 4, Column 7) for every $1spent.
Titles for the combined group of 200, there were 101 different jobs collapsed into categories making up 402 positions or student status in the life career.A single person could have more than one position, job, or status.
Fourth, 56% of prisoners had single parents so functional family therapy (ROI = 13.69), and prison family integrated solutions (ROI = 4.20) are evidence-based solutions.Fifth, 48% of jailed are substance abusers and the community cost effective approaches are adolescent diversion (ROI = 21.24) and community drug treatment (ROI = 17.92) and in jail, juvenile drug court (ROI = 1.61), adult drug court (ROI = 1.10), and prison drug treatment (ROI = 4.88).

Table 5 .
The low and high costs and ROIs for current ways vs. current ways + internet-based tests

Table 6 .
Sectors for saving lives and funds by using internet-based testsInternet tests reduce business and government payouts to the opposing side.Internet tests aid in freeing up capital or funds for other use.Internet tests free up management time 90% spent on 10% of personnel with challenges sensitively and specifically identifying what is the issue.Internet tests limit liability settlements and risk to co-workers, co-patients, co-students, and co-prisoners.Internet tests are equation-based, non-discriminatory, and objective.See Table6for the reduction in mass murders, homicides and suicides by the use of internet-based tests in selection, retention, promotion, crisis-debriefing, discharge, sentencing, probation and parole release, fitness for duty, and citizen complaint.The work groups include airlines, corrections, education, health, insurance, military, police-public safety, religious-nonprofit, transportation, and veterans.
The E.U. has double the population so there are estimates of 2,400 deaths from 600 EU mass murders, 28,200 homicides, and 82,200 suicides with a sum of 114,800 victims.By using internet-based tests the 50% reduction in the E.U. is the estimate of 1,200, 14,100, 41,100, and 57,400 lives saved yearly in Figure32.The E.U. has double the population so there are estimates of 2,400 deaths from 600 EU mass murders, 28,200 homicides, and 82,200 suicides with a sum of 114,800 victims.By using internet-based tests the 50% reduction in the E.U. is the estimate of 1,200, 14,100, 41,100, and 57,400 lives saved yearly as seen in Figure32.

Table 7 ,
there is the success August, 2008 meeting with Chicago Mayor and the resulting U.S.D.O.J. $78,000,000 grant that saved lives and funds by targeting high risk teens with evidence based diversions using the internet-based test model that was thrice replicated once independently.Next, was the 2011 meeting with Cook County President sharing the same materials; this resulted in 56% reduction of nonviolent offenders from court to electronic bracelets without return to court.Third, the internet-based model was shared in 2012 with the new Chicago Mayor and the Allstate CEO with the $50,000,000 Chicago Public Safety private fund.Fourth, from 2012, the U.S. President begins releasing federal prisoners by presidential commutation lowering the population for the first time in two decades.Fifth, in 2015, the internet-based model was shared with the new Illinois governor, who used the materials for the prison reform to release prisoners statewide.Sixth, based on the Cook County President's success with prisoner release, the U.S. President began using presidential commutation in 2012 through 2014 to release federal nonviolent offenders.There was a lowering of the prison population by 6,000 in November, 2015.

Table 7 .
Success with internet-based test model to lower violence, save lives and funds Coalition for Peace Initiative in Washington, D.C. that combined the assets of the Arthur, Ford, Koch Industries, and MacCarthur Foundation to reform the justice system.Finally, after Zagar testified in the U.S. House of Representatives, the internet-based test model was shared with the U.S. Supreme Court Justices, who announced potential release of 2,500 juveniles who were sentenced to life in prison for homicide without parole based on Miller vs.Alabama and Graham vs.Florida cases if these are safe to release(Clay,  2016).
and the 2012 to 2015 Chicago Public Safety Fund Programs saving of 324 lives and predictable, safe, and stable global society.://dx.doi.org/10.117/1077801207305232 http