The International Journal of Psychosocial Rehabilitation

Factors Influencing Diversion from
 State Mental Health Hospitals

Douglas A. Marty, MSW
Trevor J. Manthey, MSW, PhD.
Charles A. Rapp, Ph.D.
Mark C. Holter, Ph.D.

Office of Mental Health Research and Training
University of Kansas School of Social Welfare

Twente Hall
1545 Lilac Lane
Lawrence, KS 66044-3184

Marty DA, Manthey TJ, Rapp CA & Holter MC (2014) Factors Influencing Diversion from State
 Mental Health Hospitals
.International Journal of Psychosocial Rehabilitation. Vol 18(2) 141-152

Objective:  While there has been much research on predictors of psychiatric hospitalizations there has been little research on the community resources, supports and processes used to divert a hospital episode. The purpose of this study is to address this gap by studying (1) the community resources available as an alternative to state psychiatric hospitalization; and (2) the practices exhibited when determining whether state hospitalization is necessary. 
Methods: A mixed methods design was developed. The purpose of the first arm was to assess what non-hospital resources were available to mental health centers. The second arm looked at the processes center staff goes through in determining hospitalization or community diversion.
Results:  Differences were noted between centers with high and those with low diversion rates.  Centers that tended to use the state hospital less had more community diversion resources available, had an agency philosophy aimed at diversion, and used processes which included shared decision-making. Further, staff had more experience and established protocols to ensure follow-up services were in place. 
Conclusions:  Agencies that fostered a philosophy and protocol focusing on community diversion, provided alternative resources for consumers in crisis, and had adequate monitoring and training increased diversion rates and avoided unnecessary hospitalizations.

Three decades of deinstitutionalized driven initiatives and policies undertaken between 1970 and 2002 drastically reduced state psychiatric admissions in the U. S.. From a high of 475,000 in 1971, admissions dropped to 156,000 at state psychiatric hospitals in 2002 (Mandersheid, Atay, & Crider, 2009). Further augmentation of this trend for moderating state psychiatric hospitalization admissions remains a significant goal of public mental health services today. The President’s New Freedom Commission on Mental Health (2003) called for “swiftly eliminating unnecessary and inappropriate institutionalization that severely limits integrating adults with serious mental illnesses … into their community” (pg. 53).  However, state psychiatric hospital admissions rose by 21.1% to 189,000 between 2002 and 2005 (SAMHSA, 2004; SAMHSA, 2007). Part of the increase may be attributed to forensic or sexually dangerous persons ordered for rehabilitation at the state hospital (Fisher, Geller, & Pandiani, 2009). However, a report by The National Association of State Mental Health Program Directors in 2004 stated that 28% of beds in state hospitals serve forensic clients (NASMHPD, 2005).  Given that data, a large percentage of persons being admitted to state psychiatric hospitals are under no legal mandate to be admitted. In line with public policy, there has been a wealth of studies on predictors of hospitalization for individuals with psychiatric disabilities (e.g. Pasic, Russo, & Roy-Byrne, 2005; Song, Biegel, & Johnson, 1998).  For instance, variables such as managed care (Moran, Doerfler, Sherz, & Lish, 2000), co-occurring disorders (Min, Biegel, & Johnson, 2005), community characteristics (Fortney, Xu, & Dong, 2009), hospitalization history (Russo, Roy-Byrne, Jaffe, Ries, et al., 1997), medication adherence (Lang, Meyers, Korn, Lee, et al. 2010), family relationships (Sullivan, Wells, Morgenstern, & Leake, 1995) and a wide range of other clinical, demographic, and social factors (Klinkenberg & Calsyn, 1996; Pfeiffer, O’Malley, & Shott, 1996) have all been researched extensively. While there has been much research on predictors of hospitalization there has been little research on the processes and actions taken by the screener or resources available to those who are charged with making the hospitalization decision.  This is surprising given that screeners are the gatekeepers to and at the nexus of the hospitalization process. This paper explores the screeners’ experience during the admission process.

In virtually all states, the decision to hospitalize a person is made by a single person or, in some systems, by a group of clinicians. In this state, a person is considered in need of hospitalization if he/she lacks capacity to make informed decision concerning treatment and either (1) is a danger to self/others/property or (2) is substantially unable to provide for basic needs, such as food, clothing, shelter, health or safety (TAC, 2012). If the person agrees and there is a bed available, an admission occurs. In the case where a person refuses to enter the hospital, the case is referred to the court for disposition. In either case, a clinical decision was made. It’s this decision that often has important sequelae for the person and his or her family, the safety and values of a community, and the resource costs to the state mental health system. Since there are dramatically fewer state hospital beds than in years past, a decision to hospitalize anyone inevitably jeopardizes access to this service to someone who may require it.

Despite the importance of this decision, there has not been research that sought to uncover the decision-making process and influences on the people who are called on to make these important decisions. To begin to address this gap, this qualitative study sought to identify the factors that influenced the decision to hospitalize or not.

Generally, persons admitted to state managed mental health hospitals must be evaluated for necessity by a mental health screener.  Community Mental Health Center’s (CMHC) are the agencies responsible for screening adults for potential state mental health hospitalization. The primary goal of CMHCs is to provide quality care, treatment and rehabilitation to individuals with mental health problems in the least restrictive environment.  The CMHCs strongly endorse treatment at the community level, in order to allow individuals to keep functioning in their own homes and communities. In addition, their role also includes acting as the gatekeeper for admission to the state’s psychiatric hospitals. The primary admission criterion is dangerousness to self or others. For this study we identify screeners as employees of CMHC’s responsible for determining if individuals meet safety, diagnostic, and functional criteria that would warrant state hospitalization.  In order to understand the factors influencing the decision whether a client should be diverted from or admitted to a state hospital, qualitative interviews were held with individual screeners. Screener identification was achieved through maximum variation sampling.  Participant selection was determined by first ordering individual CMHC rates of hospitalization and then the rate of hospital admission for screeners within those centers. The final participants were screeners working at three community mental health centers with high rates and three centers with low rates of state psychiatric hospitalization during fiscal year 2009. For centers with lower rates of hospitalization, two are considered to be in urban areas and one in a semi-urban population center. For the three centers representative of higher hospitalization rates, their population density is considered as urban, semi-urban, and one densely populated rural area. The study was conducted over a one year period. The following table displays the sampling framework:

Table 1- Sampling Framework

Organization- CMHC


Consumer Screens Reviewed

Low Hospitalization Rate (N=3)

Low Diversion Rate (N=6)

Hospitalized (N=6)

Not Hospitalized (Diverted) (N=6)

High Diversion Rate (N=6)

Hospitalized (N=6)

Not Hospitalized (Diverted) (N=6)

High Hospitalization Rate (N=3)

Low Diversion Rate (N=6)

Hospitalized (N=6)

Not Hospitalized (Diverted) (N=6)

High Diversion Rate (N=6)

Hospitalized (N=6)

Not Hospitalized (Diverted) (N=6)







Screener Interview Criteria

Rates of state hospitalization for each of the 26 CMHCs were obtained from Kansas Social and Rehabilitative Services- Disability and Behavioral Health Services (SRS-DBHS). Criteria for center inclusion in the study were based on State Hospital Usage Rates. The number of hospital days used and rates for FY 2008 were obtained from SRS-DBHS. The range of state hospitalization days used varied from 5 to 97 hospital days used for every 1,000 persons in the CMHC’s area (includes persons younger than 18 years of age). Three centers with low hospitalization rates (Group A) and three centers with high hospitalization rates (Group B) were asked to participate in the study. These centers agreed to ask their screeners to be available for interviewing. To determine which individual screeners to interview from the participating centers, the hospitalization rates for each screener were obtained as well from SRS-DBHS. The two screeners with the highest admission rate and the two screeners with the lowest from each center were asked to be interviewed (n=24). Screeners were then assigned to one of two groups: Group A- those screeners working at agencies with lower hospitalization rates and Group B- those screeners from agencies higher hospitalization rates. All identified screeners agreed to participate.   

A semi-structured Screener Interview Guide was developed by investigators that contained a variety of open-ended questions regarding the mental health screening process and seventeen Yes/No questions that were considered to be generic to the screening process. The Guide focused on the screener’s credentialing, role, alternative to hospital services, and decision making processes while conducting hospital screens. The Guide may be obtained by contacting the primary author.  To aid dependability, a draft of the guide was presented to a committee of CMHC program directors for review, editing, comment, and feedback. 

After the participating screeners had been identified, completed state hospital mental health screen forms done by these screeners were obtained from the medical records department at each participating CMHC.  Except for the screener‘s name, the forms were de-identified.  Investigators received 5-10 completed forms for each screener.  Investigators then selected one screen for a person who was hospitalized and one for a person diverted from a state hospital admission from each screener. This process yielded a total of 48 screens used during the interview process. The screening forms were used in the interview to stimulate discussion and add context to the process. The interviews with the screeners lasted approximately 60 minutes and followed semi-standardized interview guidelines (Berg, 1995). All interviews were recorded and later transcribed for analysis. Data analysis was conducted inductively following recommendations from Miles and Huberman (1994) which involves the progressive abstraction of themes from raw data.

Following the interviews, investigators reviewed each transcript as they were made available looking for and recording salient features and patterns of responses given by the screeners as they responded to the questions asked of them from the Interview Guide. These were written down and shared with the other investigators to build consensus and affirm objectivity around broad themes found to run through the screener responses. Once a framework of general themes were recognized through saturation and accepted, investigators revisited the transcripts to look for specific examples of the general themes earlier established. Throughout the process, investigators reviewed the same transcripts, shared notes and findings, and came to consensus on their discoveries. If the theme was determined to be an integral element contained within the interview, it was identified as a pattern helping to inform on the practice of that mental health screener. Once patterns were identified, the next step was to test whether differences existed between those centers and/or screeners with high rates and low rates of state psychiatric hospitalization. Once again the transcripts were reviewed. Investigators identified patterns between groups and noted any differences. To help ensure internal validity throughout the process, investigators independently coded the transcripts and then discussed their findings. Where differences were found, examples were identified and conclusions were drawn contrasting the practice behavior between the two groups.
The study was approved by the institution’s Institutional Review Board and compensation ($100) was given to each of the interviewed screeners for their time. 

For ease of interpretation, the findings are discussed using the following four general themes that emerged during analysis:
•    Screener Characteristics,
•    Agency Structure/Philosophy,
•    Available Diversion Resources, and
•    Client Characteristics.

A summary of findings is presented in Table 2.

Table 2- Summary of Findings by High and Low Diversion



Group A

Group B



Length of Time at Agency

10.8 Years

5.9  Years

Length of Time in Mental Health

16.8  Years

12.7  Years

Length of Time as a Screener

10.5  Years

4.8  Years

State Screening Standards

83% Able to Recall Standards

50% Able to Recall Standards



Philosophical Role

Goal to Divert/ Provide Least Restrictive Setting Possible

Goal to Ensure Client Safety/Prevent Liability

Monitoring of Diversion Plan


Not Available

Agency- Structure


Service Availability

Not Limited to Individuals with SPMI

Limited to CMHC Clients

Attendant Care Services

Not Limited to CMHC Clients

Limited to CMHC Clients



Not limited to CMHC Clients

Limited to CMHC Clients

Substance Abuse Services

Not limited to CMHC Clients

Limited to CMHC Clients

Screener Training Available

Generally, Supervisor Job Shadowing

Generally had None or initially, Coworker Job Shadowing


Diversion Resources

Resources Used Most Often

Crisis Stabilization/ Case Management, Detox Services 

Therapy and Crisis Intervention

Crisis Stabilization


Not available

Respite Care Available within 24 Hours


Not available

Client Characteristics

Clients Diverted Known by CMHC

38% Known

63% Known

Clients Diverted with Suicidal Ideation

58% Suicide Ideation

25% Suicide Ideation

Clients Diverted with Psychotic Symptoms

42% Psychotic Symptoms

17% Psychotic Symptoms


Screener Characteristics
Screeners working at agencies with lower hospitalization rates (Group A) reported they had worked more years at their agency (10.8 years vs. 5.9 years) and slightly more years in the mental health field (16.8 years vs. 12.7 years). A significant difference using the t-test statistic was found for the number of years screeners had been completing mental health screens between the groups. Screeners from agencies in Group A had been conducting them for an average of 10.5 years and those from Group B, 4.8 years (t (22) = 0.031, p<.05). 

During interviews, when asked about any professional conduct standards in place for the screening process, the majority of screeners were aware of the state mandated screening policies, however, this varied by hospitalization rates by CMHC. Ten of the twelve screeners (83%) from Group A were able to report the standards with only six (50%) from higher admitting centers (Group B) being able to do so. A majority of screeners from both groups indicated that the lack of screener training was problematic and that not fully understanding the standards was indicative of this need. Ongoing screener training was identified as being needed in order to help ensure knowledge of diversion resources and skills were established and remained current.

Agency Structure/Philosophy
One of the primary differences between agencies in Group A & B was in the philosophical role identified by screeners.  Two major themes regarding philosophy were found during screener interviews: 
•    Stated goal of diversion, and
•    Need to provide safety/avoid liability.

The first theme noted was an attitude of trying everything possible to divert someone from the hospital largely came from screeners in Group A. Examples of statement content that supported this view included comments with these characteristics:
•    That diversion from a state mental health hospitalization was their goal
•    State hospitalization should be reserved for the most needy
•    The screener would take reasonable risks to divert
•    They looked upon the state hospital unfavorably
•    They felt that persons in crisis were best suited for the least restrictive setting possible.

The majority of screeners within Group B made statements that their role was to help ensure that persons in crisis were safe and “being looked after” with the state hospital being the primary resource. Examples of statement content that supported this view included comments with these characteristics:
•    The need to ensure a person is safe due to liability
•    Did not feel comfortable taking risks needed to support a diversion plan
•    There were few options other than hospitalization for persons who presented for a mental health screen.

In addition to philosophical differences, CMHC’s which had program guidelines that allowed individuals who were not clients of the CMHC to access services had lower hospitalization rates.  For instance, CMHCs in Group A did not limit Psychosocial Rehabilitation Services to only persons with SPMI (x 2(1, N=17) = 4.98, P< 0.05).  CMHCs from Group A did not restrict their Crisis Attendant Care Service to only clients of the CMHC (x2 = (1, N=6) = 6.00, P= 0.05) nor they restrict Crisis Resolution Services to clients of the CMHC (x2 = (1, N=6) = 6.00, P< 0.05).  In addition, consumers of Group A had access to Respite Care Services and Crisis Case Management Services within 24 hours of need (x2 = (1, N=17) = 4.65, P< 0.05).  These services included Outpatient Substance Abuse counseling and Detoxification Services at their CMHC. Finally, screeners from these centers reported that someone from their agency was responsible for monitoring whether the recommendations contained in any diversion plan that was developed in lieu of hospitalization were followed (x2 (1, N=24) = 6.17, p<.05).

When coding of all agency/philosophical statements was complete, these were grouped into whether the statement emphasized state hospitalization or not.  Screeners employed at CMHCs from Group A made a far less percentage of comments supporting state hospitalization (8%) than those screeners from agencies in Group B (50%).  Likewise, screeners from Group A saw their role more in terms of diverting hospitalization and/or made statements about their role as being neutral toward state hospitalization. Table 3 displays the percentage of screener responses when comments were grouped into major themes for both groups.

Table 3- Percent of Comments Reflecting Screener’s Perceived Role and Philosophy toward Hospitalizations 

Screeners (n=24)

Comments Supporting Diversion

Comments Supporting Hospitalization

Neutral Comments

Group A CMHCs (n=12)




Group B CMHCs (n=12)




The following are excerpts from two separate screeners’ discussion of their role and philosophy of screening. 
Example from a CMHC in Group A:

I make assessments about what is effective to help the client with their particular problem. And one of the principles we’re supposed to look at is least restrictive, which means not inpatient if possible, my interpretation.  You’ve got to weigh all the factors. And I honestly don’t know, and I’ve thought about this a lot. I honestly don’t know how you teach somebody to go with their gut and make a call. Because that comes from experience, self-esteem, self-confidence, willingness to take risks; there’s so much involved with that. I’d rather see a person who’s new being overly cautious and making a safe call putting somebody in a psych hospital and gradually learning to take some risks and learning that they can do things differently. …You know some people aren’t willing to take risks, [they] shouldn’t be screeners. I really feel strongly in that way, and probably they won’t be for very long.

Example from a CMHC in Group B:
I see my role as a screener as just that, the gatekeeper between people and the hospitals,…. I’m more likely to hospitalize them for, honestly, for liability reasons.  You know just to make sure they’re going to be safe.  And so that’s probably why there’s an increase in hospitalizations.  … so I am always focused on the liability probably more than most people are at least around here. I want that client to be safe.  And I know they’re going to be safe in the hospital.  If the hospital staff says he doesn’t need to be in the hospital, I’ll take that with a grain of salt, but if a hospital staff says he needs to be in a hospital, to me that’s adding liability, ... if I divert and something happens… then I have to justify that [diversion] even more. 

Screeners were also asked about training specific to screening that they had received or were required to attend in order to be considered a state mental health screener. The overwhelming majority of screeners interviewed reported that there was a lack of any standardized required training. This perception was prevalent for screeners from both groups. 

Overall, the training mechanism screeners talked about the most (62%) was shadowing another screener. Most often this was shadowing a more experienced screener to observe and learn the process.  The length of time shadowing varied widely by CMHC, but generally shadowing occurred with experienced screeners several times over a period of a month or more before doing a screening on their own. The only major difference found in the shadowing structure between agencies was that Group A screeners described shadowing specifically done with their supervisor. Screeners from these agencies also stated supervisors made it clear that they should be committed to diverting persons from the state hospital into other community resources. 

The next mechanism used for training described by all screeners was general training (33%), either provided by the community mental health center or that the screener chose to attend, that would increase their knowledge and skills related to screening.  Most of these trainings were related to working with persons in distress, but not specifically for screeners.  Examples include elective trainings that screeners attended (e.g. conferences and workshops) on such things as diagnosis, suicide prevention, the forensic national risk assessment conference, etc.  Some in-house trainings were offered such as, how to fill out the screening form, crisis intervention, and how to build a case for hospital diversion or admission. Twenty-five percent of all screeners said they were not aware of any specific training related to screening. Three screeners or 13% talked about having their supervisor or another screener available for consulting on screens.  Two screeners mentioned at one time having a manual or handbook on screening, and one screener from a Group A agency talked about having other screeners review screens and giving feedback after reviewing the form.

Available Diversion Resources

Screeners were asked questions about their use of diversion resources including those used most, those used least, which resources were easiest to use, and those most difficult to use. Noteworthy differences were found in resources used most and least between Groups. Overall, screeners from Group A (n=12) tended to use crisis stabilization options (50%), crisis case management (50%) and detoxification (50%) most often.  Crisis stabilization options included the availability of an adequately staffed place where clients can go outside of the hospital was one of the major resources cited as helpful for Group A.  It is a place that has some level of staffing where a client can go short-term to be observed and assisted through the crisis.  Often it was referred to as a safe place or respite care.  In this study, we use crisis stabilization to encompass all of these different arrangements. For the other two crisis intervention services, crisis case management included services provided by a dedicated crisis team at the CMHC and detoxification services (detox) were those where an individual who is currently abusing drugs/alcohol may become sober. Screeners from Group B (n=12) used two resources most often:  Referral to Outpatient Therapy (75%) and crisis intervention (67%). Crisis intervention services are usually limited to attendant care services or completing a risk assessment.
Having a crisis stabilization option is one of the most noteworthy differences between the groups.  Those CMHCs in Group A all had a crisis stabilization option.  Forty-two percent of screeners (5) from Group B reported crisis stabilization options as those used least. At these agencies, the majority had no crisis stabilization option, while two had a limited diversion option (e.g. availability only on weekends).  Further, screeners from Group B said that having a crisis stabilization option would be very helpful for avoiding hospitalizations.  The following are some quotes from screeners from Group A about crisis stabilization options:

If you want to divert people from psychiatric hospitalization, the Crisis Recovery Center (CRC) is the number one resource…that would make the most difference in my opinion.  The crisis stabilization concept- where a person could get observed, get some rest, get some medication, and even get medical attention if they need it- would make the biggest difference in diversion. It’s less expensive than hospitalization and least restrictive. That’s actually my primary way to diverting from the state hospital is our Crisis Recovery Center. 

Yes CRC has been helpful. That’s my number one (option to) divert.  It’s very, very helpful.  If we didn’t have our Crisis Recovery Center… we’d have a lot more difficulties. It’s a great resource and cost savings for the state.

Oh, I think CRC is great over here. … the availability of something like that is fantastic when we see somebody that, you know, they really don’t need in a hospital, but they need a lot of support, and so it’s a nice place to kind of let someone spend some time and kind of come to, kind of get themselves back together. 

Although used successfully, a variety of challenges for using a crisis stabilization option were noted during the interviews of screeners from these CMHCs.   These included:
•    Limited bed space,
•    Fear of sending someone to the crisis unit who might be too high of a risk,
•    Not being able to use the crisis stabilization option for people who are belligerent, uncooperative, disruptive, extremely psychotic and that might upset other people,
•    Not allowing people who have alcohol and drug issues,
•    Having to go through multiple steps to get the person admitted into the crisis stabilization option,
•    If there are major medication issues, the crisis stabilization unit may not be a good option,
•    Not being able to use crisis stabilization for people who are not current clients of the community mental health center.

Client Characteristics
Using a structured open-ended interview format, each screener discussed two recent screenings they completed: one that resulted in a hospitalization and one that resulted in a diversion. The goal was to identify the degree to which the client themselves or community resource factors contributed to either outcome. A total of 48 screens completed by participants from both groups were used to aid context to the interview. The interview transcripts were then analyzed separately to understand the screeners’ reasoning for their decisions and whether the reasoning expressed by screeners varied between groups.

For all those clients screened from CMHCs of Group B, 63% (15) were current clients.  That compares to 38% (9) for consumers from Group A. Using the 24 screens, twelve from each group where consumers were diverted from a hospital admission, Group A had a higher percentage where major psychiatric symptoms were noted at intake. These included suicidal ideation and disorientation.  Fifty-eight percent (7) had suicidal ideation versus 25% (3) of the persons from Group B.  Forty-two percent (5) of the persons screened from Group A had pronounced psychotic symptoms versus 17% (2) from lower performance rate centers.

Other findings for all screens that did not result in a hospital admission (24) include: (1) In 63% (15) the screener felt that it was not a serious attempt to cause harm to self or others (e.g. not life threatening, no risk of imminent harm, not having a specific plan) and (2) In 42%  (10) of the screenings, the screener felt that the person being screened was in a safe place (e.g. jail, group home, treatment center). The rationale for hospital diversion almost exclusively was based on one, and at times both, or these perceptions.

Having personal knowledge of the client seemed to be an important factor in judging the risk factor. Screeners used knowledge of the client to assess risk and in cases where there was no knowledge of a person’s baseline behavioral state, screeners were less able to assess whether the risks were typical and tended to express this lack of knowledge as a reason for hospitalization. In eleven of the twenty-four screenings (46%) reviewed for persons hospitalized, the person was not known to the screener and was not a current client of the center. This rate varied by Group. Of those clients screened from Group B, 73% (8) were current clients of the center, compared with a client rate 27% (3) for the higher performance centers.

The following are perspectives that screeners cited as affecting their decision during the 24 interviews to divert a client from a state hospitalization:
•    In 46% (11) of the screenings, the screener felt the client was cooperative/calm/rational (e.g. able to work on diversion plan, compliant with med shot, verbal commitment to follow through with plan). The same percentage chose diversion from the state hospital because the client could go to a safe environment (families, CRC, friends, local hospital, etc.),

•    In 33% (8) of the screenings, the screener felt the client was not psychotic. The same percentage was true for clients who denied intent of harm or had no plan to do so,

•    In 38% (9) of the screenings, the screener knew the client and felt comfortable with a diversion plan. The same percentage determined the problem is behavioral or situational (e.g. developmental disability, attention seeking, manipulative behavior).

The findings suggest four significant influences on screener decision-making. First, the disparity of diversion resources across the CMHC catchment areas was clearly a factor. As expected, CMHCs in Group A had a wider range of available and accessible services and diverted more persons from the state hospital into safe communities environments. These CMHCs appeared to offer crisis services to a variety of persons in their communities including those persons who may have abrupt suicide ideation or in need of detoxification services not just those with psychiatric disabilities or current clients of the center.  The two most critical resources identified by them were crisis stabilization options and attendant care programs.  But even here, an increased capacity to access psychiatric/medication evaluations and services designed for persons who pose a threat of dangerousness was seen as important.

Substance abuse was commonly mentioned by screeners in a number of contexts as a factor contributing to the need for mental health screens. Although there were noteworthy differences between the groups in the availability of substance/alcohol treatment services, almost all the screeners expressed that substance abuse increases the risk for harm to self or others, thereby making diversion a “harder call.” Most screeners reported that an increase in the availability of options for client detoxification would reduce the amount of state hospital admissions made.  

Second, the study found differences in screener characteristics. Screeners from Group A had statistically more years of experience completing mental health screens. However, it should be noted that the screeners from Group B were not inexperienced having an average of almost 5 years of screening experience.

Third, the role and philosophies were different. Screeners from Group A viewed their roles as screeners as finding ways to divert individuals while screeners from lower CMHCs seemed more concerned with liability.  For instance, screeners from this group made far fewer comments supporting hospitalization and made more comments regarding the positive aspects of diverting individuals or stating a neutral stance.  The reverse was found for screeners from Group B.   

On the one hand this may not be a surprising finding. The screeners from Group B appear to have far less diversion options available to them. If these CMHCs do not have access to effective community diversion resources, it seems reasonable to assume that they would be less willing to take a risk to not hospitalize and rely only on existing non-crisis services in the community. This same dynamic may help explain why screeners at CMHCs form Group A appear more willing to take risks that might be involved with the development of a diversion plan. It seems, however, like it is more than just resources. One center in Group B actually closed a crisis stabilization program due to “lack of use”.

While access to resources may explain some of the differences found, a fourth contributing factor may also be the screener’s level of organizational support for developing diversion plans. For example, one agency from Group A enacted an informal checks and balances policy comprised of the need to contact supervisor staff prior to hospitalization. The other two within this group have access to other crisis workers 24 hours a day, 7 days a week. It does not seem surprising that screeners from agencies where decision making is more routinely shared tend to be more comfortable taking risks to develop diversion plans. Lastly, screeners from these CMHCs also reported that someone from their agency was responsible for following-up on persons with diversion plans.

Given the human and financial costs of state hospitalization, the paucity of screener specific training that was provided to both groups of CMHC’s was surprising. At the time of the study, there were no state standards for screener training nor was there any state sponsored training. In fact, most screeners did not mention use of any standardized manual for screeners that would have included guidelines, policies, and suggested methods. If screeners were using resource and/or procedure guidelines, these were individually developed and then adopted by the screener or center.

Limitations and Conclusions
Firm conclusions are not warranted from the results of this initial qualitative study of screener decision-making. It does reinforce the assumption that certain resources (i.e. crisis stabilization, crisis case management, attendant care, detox) would reduce admission to state psychiatric hospitals. In current times, resources for such services have been difficult to find. Beyond resources, this study suggests that agency commitment and philosophy concerning diversion with supports such as training, shared decision-making by organizational professionals, and protocols for follow-up monitoring of diversion plans may make a difference. These actions would not necessarily require additional funding and are largely under the control of the mental health agency. The need for additional research with more representative samples seems warranted.


Berg, B. (1995). Qualitative Research Methods for the Social Sciences (2nd ed.). Needham Heights, MA: Simon & Shuster Co..

Fortney, J. C., Xu, S., & Dong, F. (2009).  Community-level correlates of hospitalizations for persons with schizophrenia.  Psychiatric Services, 60(6), 772-778.

KDADS (2010). Kansas Department of Aging and Disabilities. Topeka, Kansas. Information may be found at:

Klinkenberg, W. D., & Calsyn, R. J. (1996). Predictors of receipt of aftercare and recidivism among persons with severe mental illness: A review. Psychiatric Services, 47, 487–496.

Lang, K., Meyers, J. L., Korn, J. R., Lee, S., Sikirica, M., Crivera, C., Dirani, R., & Menzin, J.(2010). Medication adherence and hospitalization among patients with schizophrenia treated with antipsychotics. Psychiatric Services, 61(12), 1239-1247.

Miles, M. B., & Huberman, A. M. (1994). An expanded sourcebook: Qualitative data analysis (2nd ed.). Thousand Oaks, CA: Sage.

Min, M. O., Biegel, D. E., & Johnson, J. A. (2005). Predictors of psychiatric hospitalization for adults with co-occurring substance and mental disorders as compared to adults with mental illness only. Psychiatric Rehabilitation Journal, 29(2), 114-121.

Moran, P. W., Doerfler, L. A., Sherz, J., & Lish, J. D. (2000). Rehospitalization of psychiatric patients in a managed care environment. Mental Health Services Research, 2(4), 191-199.

NASMHPD, 2005. National Association of State Mental Health Program Directors. Information may be found @:
New Freedom Commission on Mental Health, (2003).  Achieving the Promise: Transforming Mental Health Care in America. Final Report. DHHS Pub. No. SMA-03-3832.  Rockville, MD.

Pasic, J., Russo, J., & Roy-Byrne, P. (2005). High utilizers of psychiatric emergency services.Psychiatric Services, 56(6), 678-684.

Pfeiffer, S. L., O’Malley, D. S., & Shott, S. (1996). Factors associated with the outcome of adults treated in psychiatric hospitals: A synthesis of findings. Psychiatric Services, 47, 263– 269.
Russo, J., Roy-Byrne, P., Jaffe, C., Ries, R., Dagadakis, C., & Avery, D. (1997). Psychiatric status, quality of life, and level of care as predictors of outcomes of acute inpatient treatment. Psychiatric Services, 48, 1427–143

Song, L., Biegel, D. E., & Johnson, J. A. (1998). Predictors of Psychiatric Rehospitalization for Persons with Serious and Persistent Mental Illness. Psychiatric Rehabilitation Journal, 22(2), 155-166.

SRS-DBHS, (2007). Kansas Health Policy Authority (KHPA)-Social and Rehabilitative Services, Disability and Behavioral Health Services (SRS-DBHS). Topeka, Kansas.

Sullivan, G., Wells, K. B., Morgenstern, H., & Leake, B. (1995). Identifying modifiable risk factors for rehospitalization: A case-control study of seriously mentally ill persons in Mississippi. American Journal of Psychiatry, 152, 1749–175.

TAC (2012). Treatment Advocacy Center. Information may be found at:


Copyright 2014 ADG, SA. All Rights Reserved.  
A Private Non-Profit Agency for the good of all, 
published in the UK & Honduras