Schizophrenia Research 110 (2009) 28–32

Contents lists available at ScienceDirect

Schizophrenia Research

j ourna l homepage: www.e lsev ie r.com/ locate /schres

Epidemiological trends in psychosis-related Emergency Department visits inthe United States, 1992–2001

Anand Pandya a,b, Gregory Luke Larkin c,⁎, Ryan Randles d, Annette L. Beautrais e, Rebecca P. Smith f

a Department of Psychiatry, Cedars-Sinai Medical Center, United Statesb Department of Psychiatry, NYU School of Medicine, United Statesc Department of Surgery, Division of Emergency Medicine, Yale University School of Medicine, United Statesd Department of Surgery, Division of Emergency Medicine, University of Texas, Southwestern Medical Center, Dallas, United Statese Department of Psychological Medicine, University of Otago, Christchurch, New Zealandf Department of Psychiatry, Mount Sinai School of Medicine, United States

a r t i c l e i n f o

⁎ Corresponding author.E-mail address: gluke.larkin@yale.edu (G.L. Larkin

0920-9964/$ – see front matter © 2008 Elsevier B.V.doi:10.1016/j.schres.2008.12.015

a b s t r a c t

Article history:Received 27 August 2008Received in revised form 12 December 2008Accepted 17 December 2008Available online 20 March 2009

Mental health visits represented an increasing fraction of all Emergency Department (ED) visitsin the U.S. between 1992 and 2001. This study used the National Hospital Ambulatory MedicalCare Survey, a 4-staged probability sample of ED visits from geographically diverse hospitalsaround the U.S., to assess the contribution of all psychosis-related visits to this overall trend.Unlike other mental-health-related ED visits, the rate of psychosis-related visits did notincrease. This lack of change is notable in the context of dramatic changes in both healthcarefinancing and antipsychotic prescribing practices during this period. There was an unexpecteddecrease in Medicare-funded psychosis-related ED visits at a time of increasing Medicareenrollment overall. An important demographic trend over this decade was the increasingurbanization of psychosis-related ED visits coincident with a relative decrement in such visitswithin rural areas.

© 2008 Elsevier B.V. All rights reserved.

Keywords:PsychosisEpidemiologyEmergency DepartmentsNHAMCSMental health

1. Introduction

Mental illness constitutes the second-largest diseaseburden in the United States (Hansen and Elliott, 1993).Changes in financing since 1990 have ushered in an erawheretreatment for these diseases is restricted, fragmented,managed, outpatient, and out-of-pocket: access remains anissue (Appelbaum, 2003; Lamb and Weinberger, 2005; NewFreedom Commission on Mental Health, 2003).

A few population data characterize prevalence or trends inacute service utilization among patients with serious mentalillness (SMI). The Healthcare for Communities Survey showedincreased ED use by SMI patients, but the study was cross-sectional, of small sample size (n=170) and lacked annual,population-based, longitudinal data (Mechanic and Bilder,2004). The only nationally-representative study of trends in

).

All rights reserved.

mental health services in the 1990s reported increasedutilization (12% in 1990–92 to 20% in 2000–02) (Kessleret al., 2005), independent of both socio-demographic factorsand illness severity. However, this survey lacked informationabout ED utilization for psychosis. Understanding ED utiliza-tion is important: In the U.S. only EDs provide this vulnerablepopulation with guaranteed access to medical care (Centersfor Medicare and Medicaid Services: EMTALA; Fields et al.,2001).

From 1992 to 2001 ED mental health visits increasedoverall (38%) (Larkin et al., 2005), as did visits for anxiety(Smith et al., 2008) and suicide attempts (Larkin et al., 2008).However, no studies have examined national trends inpsychosis-related ED visits from 1992–2001. We sought toaddress this gap using a national probability sample.

2. Methods

Conducted annually, the National Hospital AmbulatoryMedical Care Survey's (NHAMCS) ED component measures

29A. Pandya et al. / Schizophrenia Research 110 (2009) 28–32

emergency health care utilization, employing a 4-stageprobability sample of visits to U.S. non-institutional generaland short-stay hospitals, excluding federal, military, andVeterans Affairs facilities (Ciompi, 1987; Cohen, 1993; Cohenet al., 2000, 1996; Cohen and Kochanowicz, 1989; Cohen andTalavera, 2000; Cohler and Beeler, 1996; Cohler and Ferrono,1987). NHAMCS covers geographic primary sampling units(approximately 112), hospitals within primary sampling units(approximately 600 total), EDs within hospitals, and patientswithin EDs (http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm). Data are collected by hospital staff during annual,randomly-assigned, 4-week periods, and coded using ICD-9-CM (International Classification of Diseases, 9th Revision,Clinical Modification, 1991).

ED visits from 1992–2001were categorized as mental-health-related if records met any of the three criteria:

1. Diagnosis-based psychiatric problems (ICD-9-CM diag-noses 290.0–305; 307–310; 311–319.0 or V-codes 61.1–71.02);

2. NCHS-assigned Patient Reason-for-Visit Classificationcodes related to mental health (79), 1100.0–1199.9; or

3. Injury E-codes related to suicide/suicide attempts, E950.0–E959.9.

Visits not meeting at least one of the above criteria weredeemed non-mental-health visits. ICD-9-CM codes 290–319were excluded if they lacked corresponding DSM-baseddiagnoses.

Table 1Average rates for all mental health-related ED visits (MHRV) and psychosis-related

All mental health related visits (M

Total number estimated in 1000's (Nest); 95%confidence interval (CI)

Nest=52,774; 95%CI=49,676, 55,

Mean age (SE) years 39.5 (0.2) years

N (actual sampled visits) %

Overall 16,774 100b15 years old 1451 8.715–29 years old 3951 23.630–49 years old 7212 4350–69 years old 2411 14.470+ years old 1749 4.4Female 8164 48.7Male 8610 51.3White, non-Hispanic 10,346 61.7Black, non-Hispanic 3862 23Hispanic (All) 1990 11.9Other, non-Hispanic 576 3.4Northeast U.S. region 5502 32.8Midwest U.S. region 3473 20.7Southern U.S. Region 4368 26Western U.S. Region 3431 20.5Metropolitan EDs 14,850 88.5Non-metropolitan EDs 1924 11.5

N (actual sampled visits) Sample % (unweighInsurance statusPrivate 4552 27.1%Medicare 2758 21.8%Medicaid 3657 21.8%Self-pay 3747 20.7%Other 2333 13.9%

Injury/poisoning visit 2726 16.3%

Psychotic disorder-related visits were assigned specificDSM-compatible categories: ICD 9 CM: 295–295.9 (Schizo-phrenia), 297.3 (Folie á deux), 298.8 (Brief Psychotic Dis-order), 298.9 (Atypical Psychosis, or Psychosis NOS), and thenon-DSM-based code used by the National Center for HealthStatistics (NCHS) to describe reasons-for-visit 1155.0 (delu-sions or hallucinations).

Absolute numbers of ED visits were estimated usingcensus-based, NCHS-assigned patient weights. ED visit ratesper-population were calculated using denominator estimatesof the civilian, non-institutionalized US population, adjustedfor under-enumeration (Census U Bot).

We analyzed cases by age, sex, race/ethnicity, insurancestatus, location in a metropolitan statistical area (MSA), andregion of the country. We report only estimates with a relativestandard error less than 30% and more than 29 raw datarecords per cell (Hing et al., 2003; McCaig, 2004). Confidenceintervals (95%CI) for visit rates were calculated using therelative standard error of the estimate, controlling forweighting, four-stage sampling, and cluster effects usinggeneralized estimating equations from SUDAAN-8.0(Research Triangle Park, NC). “Least squares” linear regressionwas used for trend analysis (STATA 7.0, StataCorp, CollegeStation, TX). Differences in continuous variables wereassessed using two-tailed independent samples t-tests orrepeated measures ANOVA. Bonferroni corrections (pb0.01)were used to adjust for multiple comparisons betweengroups.

ED visits.

HRV) All psychosis visits

872 Nest=5245; 95% CI=4577, 5913

45.0 (0.7) years

/1000 EDvisits

Rate per 1000 USpopulation (95% CI)

Rate per 1000 USpopulation (95% CI)

% of MHRV

54.2 19.7(18.6,20.9) 2.0(1.7,2.2) 10.021.6 8.1(7.1,9.1) 0.2(0.1,0.3) 2.551.8 21.6(19.7,23.4) 1.8(1.2,2.4) 8.381.7 26.4(24.4,28.3) 2.9(2.4,3.4) 10.958.1 16.5(14.8,18.1) 1.8(1.3,2.3) 10.954.1 25.6(23.0,28.2) 3.6(2.5,4.6) 14.152.1 19.5(18.2,20.8) 1.8(1.5,2.1) 9.256.5 19.9(18.6,21.3) 2.2(1.8,2.5) 11.069.2 18.6(17.4,19.8) 1.7(1.4,2.0) 9.165.5 31.2(27.4,34.9) 4.2(3.2,5.2) 13.556.2 17.6(15.6,19.5) 1.3(0.8,1.8) 7.464.5 12.0(9.4, 14.6) 1.5(0.4,2.6) 12.566.4 24.6(22.5,26.7) 2.5(1.9,3.0) 10.250.4 20.7(18.9,22.6) 2.4(1.8,2.9) 11.647 17.4(16.1,18.7) 1.7(1.3,2.1) 9.860.5 18.1(16.6,19.7) 1.6(1.1,2.0) 8.855.7 19.7(18.5,20.9) 2.1(1.8,2.4) 10.748.9 19.8(18.1,21.6) 1.6(1.1,2.0) 8.1

ted) Population-based % of MHRV % of psychosis visits

29(27,30)% 15(10,19)%17(16,19)% 30(26,35)%20(19,21)% 24(18,30)%21(19,22)% 15(12,20)%13(12,15)% 16(10,20)%30(29,32)% 18(13,23)%

Table 2Trends in rates for all mental health-related ED Visits (MHRV) and psychosis-related ED visits — by gender, age and race/ethnicity.

Trend By year All MHRV Psychosis visits

N⁎ Rate per 1000 Rate per 1000

Overall 1992–1993 3360 16.7(15.0,18.3) 1.9(1.6,2.3)1994–1995 2834 17.7(16.0,19.3) 1.8(1.5,2.1)1996–1997 2907 19.8(17.9,21.6) 2.0(1.6,2.3)1998–1999 3307 21.8(19.6,23.9) 2.0(1.6,2.5)2000–2001 4366 22.3(20.5,24.1) 2.0(1.7,2.3)

p=0.002 p=0.182Female 1992–1993 1642 16.2(14.4,18.1) 1.6(1.2,2.0)

1994–1995 1387 17.1(15.3,19.0) 1.9(1.4,2.3)1996–1997 1401 19.5(17.4,21.6) 1.6(1.2,2.0)1998–1999 1580 21.7(19.3,24.2) 1.9(1.4,2.4)2000–2001 2154 22.4(20.3,24.4) 1.9(1.4,2.3)

p=0.002 p=0.308Male 1992–1993 1718 17.1(15.2,19.0) 2.3(1.8,2.8)

1994–1995 1447 18.2(16.2,20.1) 1.7(1.2,2.2)1996–1997 1506 20.0(17.9,22.2) 2.3(1.8,2.9)1998–1999 1727 21.8(19.3,24.2) 2.2(1.6,2.8)2000–2001 2212 22.3(20.2,24.4) 2.2(1.7,2.6)

p=0.002 p=0.761White, non-Hispanic 1992–1993 2163 15.2(13.6,16.8) 1.7(1.3,2.0)

1994–1995 1766 16.4(14.8,18.0) 1.6(1.2,1.9)1996–1997 1800 18.0(16.2,19.8) 1.6(1.2,2.0)1998–1999 1902 21.0(18.8,23.2) 1.9(1.5,2.3)2000–2001 2715 22.3(20.4,24.3) 1.8(1.4,2.1)

p=0.002 p=0.278Black, non-Hispanic 1992–1993 721 26.1(22.2,30.0) 4.1(2.7,5.4)

1994–1995 650 27.1(22.9,31.3) 3.8(2.5,5.1)1996–1997 696 36.0(31.0,41.0) 4.8(3.2,6.4)1998–1999 888 34.7(29.6,39.9) 4.1(2.6,5.7)2000–2001 907 31.4(27.4,35.3) 4.2(3.0,5.5)

p=0.234 p=0.728Hispanic (all) 1992–1993 372 18.6(14.9,22.2) 1.02(0.4,1.6)

1994–1995 306 18.0(14.5,21.4) 1.2(0.3,2.1)1996–1997 318 17.3(13.8,20.8) 1.2(0.4,1.9)1998–1999 428 17.7(14.1,21.2) 1.3(0.4,2.3)2000–2001 566 19.0(16.0,22.0) 1.8(1.0,2.6)

p=0.853 p=0.095b15 years 1992–1993 308 7.7(5.9,9.5) 0.3

1994–1995 260 6.8(5.1,8.5) 0.11996–1997 263 8.4(6.3,10.5) 0.11998–1999 260 8.3(6.2,10.4) 0.12000–2001 360 9.3(7.3,11.2) 0.2

p=0.101 p=0.55915–29 years 1992–1993 819 18.2(15.1,21.4) 2.1(1.3,2.9)

1994–1995 711 20.4(17.0,23.9) 1.8(0.9,2.8)1996–1997 645 21.4(17.8,25.1) 1.7(0.9,2.5)1998–1999 773 23.4(19.2,27.7) 1.9(0.9,2.9)2000–2001 1003 24.1(20.5,27.6) 1.5(0.8,2.1)

p=0.002 p=0.12130–49 years 1992–1993 1396 21.6(18.4,24.8) 2.5(1.7,3.4)

1994–1995 1239 23.9(20.5,27.3) 2.5(1.6,3.3)1996–1997 1305 26.7(23.0,30.5) 3.3(2.3,4.4)1998–1999 1451 29.5(25.2,33.9) 3.2(2.1,4.3)2000–2001 1821 29.6(26.0,33.2) 2.9(2.1,3.8)

p=0.005 p=0.25550–69 years 1992–1993 478 14.6(11.5,17.6) 1.2(0.5,1.9)

1994–1995 361 13.7(10.6,16.7) 1.7(0.7,2.7)1996–1997 406 17.2(13.8,20.5) 1.6(0.7,2.5)1998–1999 487 17.7(14.0,21.5) 2.2(0.9,3.5)2000–2001 679 18.6(15.5,21.7) 2.2(1.3,3.2)

p=0.036 p=0.024N=70 years 1992–1993 359 20.6(15.9,25.3) 4.7(2.7,6.7)

1994–1995 263 22.3(17.3,27.2) 3.7(1.9,5.4)1996–1997 288 22.4(17.3,27.6) 2.9(1.2,4.6)1998–1999 336 29.5(23.0,36.1) 2.5(1.0,4.0)2000–2001 503 32.0(26.2,37.9) 4.2(2.5,5.8)

p=0.018 p=0.523

N⁎ = Actual sampled visits.

30 A. Pandya et al. / Schizophrenia Research 110 (2009) 28–32

3. Results

Psychosis-related visits accounted for approximately 10%of all mental health ED visits in 1992–2001 (Table 1). Non-Hispanic Black individuals had the highest visit rate (4.2/1000), more than twice that of non-Hispanic Whites (1.7) orHispanics (1.3). Fewer psychosis-related visits were injury oroverdose-related (18%) compared to all mental health visits(30%; Fisher's pb0.001). Medicare (30%) and Medicaid (24%)provided insurance for most cases. Compared to all mentalhealth visits, psychosis-related visits were approximately halfas likely to be privately insured (29% v.15%; Fisher's pb0.001).

While overall mental-health ED visits increased, the rate ofpsychosis-related visits per capita remained stable (Table 2).No increases were observed by gender, racial/ethnic group orgeographic region. Visit rates increased in those aged 50 to 69,by 83%, from 1.2 to 2.2 per 1000 U.S. population across thedecade. Psychosis-related visits in non-metropolitan (rural)areas decreased, accompanied by a reciprocal increase inmetropolitan areas (Table 3). Psychosis-related visits forMedicare patients decreased (Table 4).

4. Discussion

Psychosis-related ED visits remained stable while mentalhealth ED visits increased by more than a third (Larkin et al.,2005). This stability is notable given radical changes intreatment and prescribing practices for psychosis duringthis decade. In 1992, clozapine was the only Second-Generation Antipsychotic (SGA) available, but by 2001,84.5% of all office-based physician visits for antipsychoticsincluded a SGA prescription (Aparasu et al., 2005). Our findingsuggests that, overall, increased prescription of SGAs did notreduce ED visits.

Psychosis-related ED visit rates for Medicare beneficiariesdecreased significantly while ED Medicare visits for mentalhealth, mood, anxiety, substance, and suicide-related pro-blems increased (Larkin et al., 2005, 2008; Smith et al., 2008),and while the number of Medicare enrollees also increased(“Centers for Medicare and Medicaid Services: MedicareEnrollment Reports”; Program Information on Medicaid &State Children's Health on Medicaid & State Children's HealthInsurance Program (SCHIP)). Increased Medicaid enrollmentmay represent greater penetration among populations withlow rates of psychosis and/or low rates of ED utilization,perhaps especially likely where enrollment growth has beenattributable to waiver programs like the California FamilyPlanning Access Care and Treatment program (“FamilyPlanning, Access, Care and Treatment Program”). It is alsopossible that managed care had a greater cost-containmenteffect for psychosis-related visits than other mental-healthvisits, since individuals with higher levels of psychosis aremore likely to have had a period of enrollment in publicmanaged mental health plans (Wingerson et al., 2001).

Diversion of those with SMI to jails and prisons (Green-berg and Rosenheck, 2008; Lamb and Weinberger, 2005)might also help explain this trend.

Visit rates increased only in 50- 69-year-olds. Because wecorrected for population growth by age, this trend cannot beexplained by the general aging of the population, unless theshift to an older age distribution is greater for those with

Table 4Trends in rates for all mental health-related ED visits and psychosis-relatedED visits — by insurance status.

Year % of All MHRV % of psychosis visits

Private 1992–1993 27.1% 12.4%1994–1995 28.5% 20.8%1996–1997 29.2% 16.8%1998–1999 26.7% 9.4%2000–2001 31.1% 15.2%

p=0.331 p=0.766Medicare 1992–1993 18.6% 36.9%

1994–1995 15.3% 21.3%1996–1997 14.9% 27.2%1998–1999 18.1% 31.5%2000–2001 19.7% 34.5%

p=0.535 p=0.046Medicaid 1992–1993 22.9% 30.9%

1994–1995 17.8% 18.9%1996–1997 18.8% 18.5%1998–1999 21.3% 28.0%2000–2001 19.1% 22.3%

p=0.608 p=0.677Self-pay 1992–1993 18.8% 16.5%

1994–1995 19.5% 17.1%1996–1997 23.5% 21.5%1998–1999 21.3% 15.6%2000–2001 19.6% 16.5%

31A. Pandya et al. / Schizophrenia Research 110 (2009) 28–32

psychosis than for other Americans. More plausible explana-tions include that older psychotic patients find it hard tobreak habits of seeking ED care (Parks et al., 2006) or thatolder individuals with late-onset psychosis in the context ofdementia contribute to increased utilization by older people.Regardless of source, this trend should be considered inservice planning, resource allocation and bridging gapsbetween emergency medicine, geriatrics and mental health(Cuffel et al., 1996; Jeste et al., 1999).

We found psychosis-related ED visits increased in metro-politan areas, coincident with a reciprocal decrease in ruralsettings. This trend might be explained by a net migration ofindividuals with psychosis from rural to urban regions, inorder to access urban-based psychiatric hospitals and/orspecialist psychiatric care (Peen and Dekker, 2004; Manders-heid and Henderson, 1999).

We found psychosis-related ED usagewas higher for Black,non-Hispanic individuals, consistent with prior findingsshowing that racial minorities with severe mental illnessexhibit greater use of EDs (Young et al., 2005).

Our study has several limitations: VA hospitals wereexcluded; data do not permit analysis of repeat visitors;psychiatric diagnostic practice moved from use of DSM-III to

Table 3Geographic trends in rates for all mental health-related ED visits (MHRV) andpsychosis-related ED VISITS.

Trend By year All MHRV Psychosis visits

N⁎ Rate per 1000US

Rate per 1000US

Northeast 1992–1993 967 18.1(15.6,20.6) 2.1(1.4,2.9)1994–1995 772 23.3(20.3,26.3) 2.0(1.2,2.8)1996–1997 877 25.3(22.0,28.6) 2.1(1.3,3.0)1998–1999 1330 28.5(24.4,32.5) 3.0(1.9,4.1)2000–2001 1556 27.8(24.6,31.0) 3.0(2.1,3.9)

p=0.022 p=0.059Midwest 1992–1993 753 20.5(17.8,23.2) 3.4(2.5,4.3)

1994–1995 734 18.8(16.3,21.4) 1.9(1.3,2.6)1996–1997 652 21.2(18.4,24.0) 2.5(1.6,3.4)1998–1999 572 21.5(18.6,24.5) 2.0(1.3,2.7)2000–2001 762 21.5(19.0,24.0) 2.0(1.3,2.7)

p=0.232 p=0.206South 1992–1993 878 13.7(11.9,15.4) 1.2(0.8,1.6)

1994–1995 716 15.1(13.2,17.0) 1.7(1.1,2.2)1996–1997 777 16.9(14.8,19.0) 1.7(1.2,2.1)1998–1999 844 20.0(17.5,22.6) 2.0(1.3,2.6)2000–2001 1153 20.8(18.6,23.0) 1.8(1.4,2.3)

p=0.002 p=0.101West 1992–1993 762 16.0(13.6,18.3) 1.3(0.8,1.9)

1994–1995 612 15.4(13.1,17.7) 1.7(1.0,2.4)1996–1997 601 18.0(15.5,20.4) 1.8(1.2,2.4)1998–1999 561 19.0(16.1,21.8) 1.4(0.8,2.0)2000–2001 895 21.8(19.2,24.4) 1.6(1.0,2.1)

p=0.018 p=0.711Metropolitan area 1992–1993 3010 15.1(13.5,16.7) 1.8(1.5,2.2)

1994–1995 2561 17.5(15.9,19.2) 1.8(1.4,2.1)1996–1997 2470 19.6(17.8,21.5) 2.2(1.8,2.6)1998–1999 2903 22.1(19.8,24.3) 2.2(1.8,2.7)2000–2001 3906 24.5(22.5,26.6) 2.3(1.9,2.7)

p=0.000 p=0.027Non-metropolitan

area1992–1993 350 26.9(23.1,30.7) 2.51994–1995 273 18.3(15.5,21.0) 2.01996–1997 437 20.2(17.4,23.0) 1.2(0.7, 1.8)1998–1999 404 20.8(17.8,23.7) 1.4(0.8,2.1)2000–2001 460 16.1(13.8,18.3) 1.2(0.7,1.7)

p=0.035 p=0.047

N⁎ = Actual sampled visits.

p=0.643 p=0.116Other 1992–1993 12.5% 9.3%

1994–1995 18.9% 16.8%1996–1997 13.6% 9.0%1998–1999 12.6% 8.8%2000–2001 10.5% 8.0%

p=0.374 p=0.618

DSM-IV during the study period, incorporating criteriachanges for psychotic disorders. While coders used aconsistent definition of psychosis, their dependence onclinical documentation may have led to some diagnosticdrift as clinicians became increasingly familiar with DSM-IVcriteria.

Against the national backdrop of rising ED oversubscrip-tion, stable visit rates for psychosis warrant further investiga-tion. Future work should explore trends toward increased ageand urbanization of ED-reliant psychotic patients, as thesetrends have important resource implications for both emer-gency and mental health care systems.

Role of funding sourceThe authors of this study received no funds for this research.

Conflict of interestThe authors have no actual or potential conflict of interest including any

financial, personal or other relationships with other people or organizationswithin three (3) years of beginning the work submitted that couldinappropriately influence, or be perceived to influence, their work.

AcknowledgmentNone.

References

Aparasu, R.R., Bhatara, V., Gupta, S., 2005. U.S. national trends in the use ofantipsychotics during office visits, 1998–2002. Ann. Clin. Psychiatry 17 (3),147–152.

Appelbaum, P., 2003. The ‘Quiet’ crisis in mental health services. Health Aff.22 (5), 110–116.

32 A. Pandya et al. / Schizophrenia Research 110 (2009) 28–32

Census U Bot [Electronic Version]. Retrieved February 29, 2008, from www.census.gov.

Centers for Medicare and Medicaid Services: EMTALA [Electronic Version].Retrieved 24 September, 2008 from http://www.cms.hhs.gov/emtala/.

Centers for Medicare and Medicaid Services: Medicare Enrollment Reports.Retrieved February 27, 2008, from www.cms.hhs.gov.

Ciompi, L., 1987. Review of follow-up studies on long-term evolution andaging in schizophrenia. In: Miller, N.E., Cohen, G.D. (Eds.), Schizophreniaand Aging: Schizophrenia, Paranoia, and Schizophreniform Disorders inLater Life. Guilford Press, New York, NY.

Cohen, C., 1993. Age-related correlations in patient symptom managementstrategies in schizophrenia: an exploratory study. Int. J. Geriatr. Psychiatry8, 211–213.

Cohen, C., Kochanowicz, N., 1989. Schizophrenia and social network patterns: asurvey of black inner-city outpatients. Community Ment. Health J. 25 (3),197–207.

Cohen, C., Talavera, N., 2000. Functional impairment in older schizophrenicpersons. Toward a conceptual model. Am. J. Geriatr. Psychiatry 8 (3),237–244.

Cohen, C., Talavera, N., Hartung, R., 1996. Depression among aging personswith schizophrenia who live in the community. Psychiatr. Serv. 47 (6),601–607.

Cohen, C., Cohen, G., Blank, K., Gaitz, C., Katz, I., Leuchter, A., et al., 2000.Schizophrenia and older adults. An overview: directions for research andpolicy. Am. J. Geriatr. Psychiatry 8 (1), 19–28.

Cohler, B., Beeler, J., 1996. Schizophrenia and the life course: implications forfamily relations and caregiving. Psychiatr. Ann. 26, 745–756.

Cohler, B., Ferrono, C., 1987. Schizophrenia and the adult life-course. In: Miller,N., Cohen, G. (Eds.), Schizophrenia and Aging. Guilford Press, NewYork, NY,pp. 189–199.

Cuffel, B.J., Jeste, D.V., Halpain, M., Pratt, C., Tarke, H., Patterson, T.L., 1996.Treatment costs and use of community mental health services forschizophrenia by age cohorts. Am. J. Psychiatry 153 (7), 870–876.

Family Planning, Access, Care and Treatment Program. Retrieved 23 August,2008, from http://www.dhcs.ca.gov/SERVICES/MEDI-CAL/Pages/FamilyPACTMedi-CalWaiver.aspx.

Fields, W., Asplin, B., Larkin, G., Marco, C., Johnson, L., Yeh, C., et al., 2001. TheEmergency Medical Treatment and Labor Act as a federal health caresafety net program (review). Acad. Emerg. Med. 8 (11), 1064–1069.

Greenberg, G.A., Rosenheck, R.A., 2008. Jail incarceration, homelessness andmental health: a national study. Psychiatr. Serv. 59 (2), 170–177.

Hansen, T.E., Elliott, K.D., 1993. Frequent psychiatric visitors to a VeteransAffairs medical center emergency care unit. Hosp. Commun. Psychiatry44 (4), 372–375.

Hing, E., Gousen, S., Shimizu, I., Burt, C., 2003. Guide to using masked designvariables to estimate standard errors in public use files of the National

Ambulatory Medical Care Survey and the National Hospital AmbulatoryMedical Care Survey. Inquiry 40 (4), 401–415.

International Classification of Diseases, 9th Revision, Clinical Modification.(1991).). Geneva, Switzerland: World Health Organization.

Jeste, D.V., Alexopoulos, G.S., Bartels, S.J., Cummings, J.L., Gallo, J.J., Gottlieb, G.L.,et al.,1999. Consensus statement on the upcoming crisis in geriatricmentalhealth: researchagenda for thenext2 decades. Arch. Gen. Psychiatry 56 (9),848–853.

Kessler, R.C., Demler, O., Frank, R.G., Olfson, M., Pincus, H.A., Walters, E.E., et al.,2005. Prevalenceand treatmentofmental disorders,1990 to2003.N. Engl. J.Med. 352 (24), 2515–2523.

Lamb, H.R., Weinberger, L.E., 2005. The shift of psychiatric inpatient care fromhospitals to jails and prisons. J. Am. Acad. Psychiatry Law33 (4), 529–534.

Larkin, G., Claassen, C., Emond, J., Pelletier, A., Camargo, C., 2005. Trends in U.S.emergency department visits for mental health conditions, 1992 to 2001.Psychiatr. Serv. 56 (6), 671–677.

Larkin, G., Smith, R., Beautrais, A., 2008. Trends in US emergency departmentvisits for suicide attempts, 1992–2001. CRISIS 29 (2), 73–80.

Mandersheid, R.W., Henderson, M.J., 1999. Mental Health, United States,1998. US Dept. of Health and Human Services, Rockville, MD.

McCaig, L.F., 2004. Using National Hospital Ambulatory Medical Care Survey(NHAMCS) data for injury analysis. Paper Presented at the User's DataConference July 12–14, Washington, DC.

Mechanic, D., Bilder, S., 2004. Treatmentofpeoplewithmental illness: a decade-long perspective. Health Aff. 23 (4), 84–95.

New Freedom Commission on Mental Health, 2003. Achieving the Promise:Transforming Mental Health Care in America, Final Report. DHHS Pub.no. SMA-03-3832. U.S. Government Printing Office, Rockville, MD.

Parks, J., Svendsen, D., Singer, P., Foti, M.E., 2006. Morbidity and Mortality inPeople with Serious Mental Illness. National Association of State MentalHealth Directors (NASMHPD), Alexandria, VA.

Peen, J., Dekker, J., 2004. Is urbanicity an environmental risk-factor forpsychiatric disorders? Lancet 363 (9426), 2012–2013.

Program Information on Medicaid & State Children's Health on Medicaid &State Children's Health Insurance Program (SCHIP). Retrieved February29, 2008, from www.cms.hhs.gov/TheChartSeries.

Smith, R.P., Larkin, G.L., Southwick, S.M., 2008. Trends in US emergencydepartment visits for anxiety-relatedmental health conditions,1992–2001.J. Clin. Psychiatry 69 (2), 286–294.

Wingerson, D., Russo, J., Ries, R., Dagadakis, C., Roy-Byrne, P., 2001. Use ofpsychiatric emergency services and enrollment status in a publicmanaged mental health care plan. Psychiatr. Serv. 52 (11), 1494–1501.

Young, A.S., Chinman, M.J., Cradock-O'Leary, J.A., Sullivan, G., Murata, D., Mintz,J., et al., 2005. Characteristics of individuals with severe mental illness whouse emergency services. Community Ment. Health J. 41 (2), 159–168.