CoChat
Sign Up
Cited by 2Psychiatric care and mental health servicesEmergency and Acute Care StudiesHealthcare Decision-Making and Restraints

Referral of patients to psychiatric emergency departments by police: A systematic review and meta-analysis

Open access
Goldschmidt, Thomas, Bermpohl, Felix, Schreiter, Stefanie, van Bergen, Annet H., Kupka, Ralph W., Schouler-Ocak, Meryam, Gutwinski, Stefan, Deutscher, Karl·2025·Psychiatric Services
DOI 10.1176/appi.ps.20240323
Open this paper in CoChat — free accountOpen PDF

Abstract

OBJECTIVE: A significant number of patients in psychiatric emergency departments (PEDs) are brought in by police. The authors sought to identify the weighted global rate of police referrals to PEDs, to compare the characteristics of police-referred patients with those of patients not referred by police, and to assess possible effects of the COVID-19 pandemic on referral rates. METHODS: A systematic review was conducted by searching medical databases according to PRISMA guidelines. Multivariate and pairwise meta-regression analyses were performed. RESULTS: Overall, 1,374 articles were screened, and 34 samples comprising 542,143 patients (61,647 of whom were referred by police) from 28 articles were included. On average, 13.7% of PED admissions were referred by police, a rate that was higher in the United States (22.8%). Compared with patients not referred by police, police-referred patients were significantly more likely to show aggressive behavior (relative risk [RR]=4.73), to be homeless (RR=1.84), and to have a diagnosis of a substance use (RR=1.33) or psychotic (RR=1.45) disorder but were less likely to have depressive (RR=0.31) or adjustment (RR=0.58) disorders. Police-referred patients were also more likely to be admitted to inpatient care (RR=1.35) and involuntarily (RR=4.38). The pandemic had no apparent effect on police-referral rates to PEDs. CONCLUSIONS: A substantial proportion of psychiatric patients is referred to PEDs by police. Police-referred patients had characteristics that were similar to those of incarcerated individuals with mental disorders. Early treatment may prevent later incarceration of police-referred PED patients. Interventions to reduce homelessness among psychiatric patients might help reduce police referrals to PEDs.

Integrity signals

Registry

Crossref · 10.1176/appi.ps.20240323

Cites

66 of 66 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

17 of 75 cited sources checked · 7 supported · 0 flagged · accuracy not yet measured

  • Findingown data

    The weighted mean police-referral rate within the nine U.S.-based study samples was 22.8% (95% CI=16.8%-29.3%).

    “The weighted mean of police-referral rates within the nine U.S.-based study samples (7, 8, 10, 11, 30, 33, 40, 44, 47) was 22.8% (95% CI=16.8%-29.3%).”
    Quote in the paperSubstantial claim1 of 9 sources checked
    • cites [7] source
    • cites [8] Sales GN, 1991, Hosp Community Psychiatry

      Not checked in the free run.

    • cites [10] source
    • cites [11] Reinish LW, 1995, Bull Am Acad Psychiatry Law
    • cites [30] Downey LVA, 2010, Prim Psychiatry

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [33] Gillig P, 1990, Hosp Community Psychiatry source
    • cites [40] McNiel DE, 1991, Hosp Community Psychiatry source
    • cites [44] source
    • cites [47] Watson MA, 1993, Hosp Community Psychiatry source
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.
  • Interpretationcitation

    The elevated involuntary admission rate among police-referred patients may result from a summed effect of multiple risk factors for involuntary admission.

    “The high rates of involuntary inpatient admissions (RR=4.38)among police-referred patients compared with patients admitted via other routes may be due to a summed effect due to multiple risk factors for involuntary admissions, given evidence that aggression, psychotic and substance use disorders, as well as police referral itself are risk factors for involuntary admission (53).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [53] source

      The Results (pp. 10–11) associate psychotic disorders, risk to others or aggression, and police involvement with involuntary admission. But substance misuse was not identified as an increased-risk factor (OR 0.81, 95% CI 0.66–1.00), and the Discussion (p. 13) says the review could not examine how the factors interacted. The proposed summed effect is therefore a plausible inference by the citing authors, not a finding established by this source. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
  • Findingown data

    Using data from four studies, the review could not establish a significant impact of the COVID-19 pandemic on police referrals to PEDs.

    “Using data from four studies (9, 13, 27, 48), we could not establish a significant impact of the COVID-19 pandemic on police referrals to PEDs.”
    Quote in the paperSubstantial claim4 of 4 sources checked
    • cites [9] source

      The Methods and Results describe a comparison of mental health ED attendances in Western Australia in 2019 and 2020, including mode of arrival. The passages available from our copy do not show police referrals or their statistical result, so they cannot establish whether this study contributed data to the review’s finding (searched: police referrals, police mode of arrival). (LLM, conf=0.67)

      Experimental — accuracy not yet measured
    • cites [13] source
    • cites [27] source

      Table 1 (indexed under 5. Conclusions; page unavailable) compares police referrals at the Swiss psychiatric ED: 82 (11.7%) in 2016 and 60 (10.4%) in 2020. This makes the paper a relevant source of data for the review, but it does not report a police-specific significance test or establish the review’s conclusion from four studies; the table’s p-value applies to referral sources as a group. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [48] source

      The Methods (Study design and sample; page unavailable) say patients could arrive via law enforcement, but list no police-referral measure among the data extracted. The Results report changes in PED visits and hospitalizations, not police referrals. Searching for police and law-enforcement referrals found no comparison that could verify this study’s contribution to the review’s finding. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
  • Findingown data

    No significant differences in police-referred PED admission rates were found for age.

    “No significant differences in police-referred PED admission rates were found for age and prevalence of suicidal symptoms or of organic mental, bipolar, and personality disorders.”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.
  • Findingown data
    § Referral of Patients to Psychiatric Emergency Departments by Police: A Systematic Review and Meta-Analysis

    Police-referred PED admissions were more likely to involve individuals diagnosed with psychotic or schizophrenia spectrum disorders (RR=1.45) than non-police-referred admissions.

    “and to have a diagnosis of a substance use (RR=1.33)or psychotic (RR=1.45)disorder but were less likely to have depressive (RR=0.31)or adjustment (RR=0.58)disorders.”
    • Also stated in: § Comparisons Between Police-Referred and Other PED Admissions · plain
      “Compared with PED admissions not involving police, police-referred PED admissions were significantly less likely to involve female individuals, substantially more likely to involve individuals showing aggressive behavior, more likely to involve individuals who were homeless or diagnosed as having substance use or psychotic or schizophrenia spectrum disorders, and less likely to involve individuals”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.
  • Classified as probably not a substantial claim. 24
  • May combine two claims. 25
  • The quote only partly states this claim. 3

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing), S4 (based). May combine two claims.

    Study design and conductcitation

    The observation period 'during the pandemic' was not clearly defined beforehand because the beginning and end of the pandemic differed substantially between geographic regions.

    “The observation period”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [24] Sanyaolu A, 2021, Infect Dis source

      The Discussion says that geographic location and the timing of spread varied in their effects across countries, and the Introduction describes cases emerging in different regions at different times. Neither passage says that the pandemic’s beginning and end differed substantially by region or that an observation period could not be defined beforehand. Page numbers are not available in our copy. (LLM, conf=0.78)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The review followed PRISMA reporting guidelines.

    “We followed PRISMA reporting guidelines (14) (see the online supplement to this review for ad- ditional details).”
    Quote in the paperNot a substantial claim1 of 1 sources checked

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Eligible studies had a data collection period of at least 7 days.

    “Eligible studies included observational studies and retrospective chart reviews that had a data collection period of at least 7 days because of the possibility of differences between weekday and weekend PED admission rates (15).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [15] source

      The Results (Distribution of in-patient deaths, admissions, discharges, seclusions and violent incidents by day of week) report significantly fewer psychiatric hospital admissions on Saturdays and Sundays than on weekdays, supporting the rationale for comparing days of the week. But the Discussion (Strengths and limitations) says the authors lacked data on emergency-department presentations and subsequent admissions, so the paper does not establish a difference in PED admission rates. The seven-day eligibility rule is the citing authors’ method, not a criterion reported by this study. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    The most common study locations were the United States (N=9) and Australia (N=6).

    “The most common study locations were the United States (N=9) (7,8,10,11,30,33,40,44,47) and Australia (N=6) (9,32,(37)(38)(39)46).”
    Quote in the paperNot a substantial claim1 of 10 sources checked
    • cites [7] source
    • cites [8] Sales GN, 1991, Hosp Community Psychiatry

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [10] source
    • cites [11] Reinish LW, 1995, Bull Am Acad Psychiatry Law

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [30] Downey LVA, 2010, Prim Psychiatry

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [33] Gillig P, 1990, Hosp Community Psychiatry source
    • cites [40] McNiel DE, 1991, Hosp Community Psychiatry source
    • cites [44] source
    • cites [47] Watson MA, 1993, Hosp Community Psychiatry source
    • cites 9,32,(37)(38)(39)46

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingcitation

    One recent Australian study included 446,815 psychiatric emergency admissions, representing 82.4% of cumulative admissions of all included studies.

    “One recent study from Australia included 446,815 psychiatric emergency admissions (46), this being by far the largest study, representing 82.4% of the cumulative admissions of all studies included in this review.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [46] source

      The Results report 446,815 mental health-related emergency department presentations by adults in Queensland, Australia, over six years. The source counts presentations, not psychiatric admissions; some patients were discharged from the emergency department. The 82.4% figure is the citing review’s calculation across its included studies, not a finding attributed to this source. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    Four studies, representing five samples for each pandemic period, reported on police-referral rates before and during COVID-19 (prepandemic N=10,326; during pandemic N=9,396).

    “Four studies, representing five samples for each of the two pandemic periods analyzed (prepandemic, N=10,326 patients; during pandemic, N=9,396 patients), reported on policereferral rates to PEDs before and during the COVID-19 pandemic (9,13,27,48).”
    Quote in the paperSubstantial claim4 of 4 sources checked
    • cites [9] source

      The Methods and Results describe mental-health ED presentations in 2019 and 2020, so this paper fits the before-and-during comparison. Our copy does not report police-referral rates: the Results discuss private transport and ambulance use, while the available Table 1 passage stops at the 'Mode of arrival' heading. The four-study count and combined patient totals are aggregate figures, not figures this individual study reports. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [13] source
    • cites [27] source
    • cites [48] source

      The Methods and Results compare psychiatric emergency visits before and during the pandemic, but the passages on study variables and visit outcomes do not report police-referral rates. A search for police referrals found no such measure in our copy. The pooled patient totals and count of four studies also cannot be verified from this individual paper. (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationreasoning

    The findings are consistent with the earlier review, even though the two reviews shared only nine studies.

    “We note that our findings are consistent with those of the earlier review, even though our review and the previous review shared only nine studies (7,8,11,32,36,37,40,45,47).”
    Quote in the paperNot a substantial claim1 of 9 sources checked
    • cites [7] source
    • cites [8] Sales GN, 1991, Hosp Community Psychiatry

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [11] Reinish LW, 1995, Bull Am Acad Psychiatry Law

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [32] source
    • cites [36] source
    • cites [37] source
    • cites [40] McNiel DE, 1991, Hosp Community Psychiatry source
    • cites [45] source
    • cites [47] Watson MA, 1993, Hosp Community Psychiatry source

      The Methods and Discussion describe a study of police referrals at nine psychiatric emergency services, not a review comparing sets of studies. Searches for an earlier review and nine shared studies found no passage addressing the claimed overlap or consistency of findings. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Some studies have suggested that local differences in patient characteristics and different health policies could explain between-study heterogeneity.

    “Some studies have suggested that local differences in patient characteristics and different health policies could explain between-study heterogeneity (5,12).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [5] source
    • cites [12] source

  • The quote only partly states this claim.

    Findingown data

    Police-referred PED admissions were significantly more likely to lead to inpatient treatment (RR=1.35) than admissions not involving police.

    “Separate meta-analytic models revealed that police-referred PED admissions were significantly more likely to lead to inpatient treatment and involuntary inpatient admission (Table 1).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.

  • May combine two claims.

    Findingown data
    § Referral of Patients to Psychiatric Emergency Departments by Police: A Systematic Review and Meta-Analysis

    After screening 1,374 articles, 34 study samples from 28 articles were included, representing 542,143 psychiatric emergency admissions, of which 61,647 involved law enforcement.

    “Overall, 1,374 articles were screened, and 34 samples comprising 542,143 patients (61,647 of whom were referred by police) from 28 articles were included.”
    • Also stated in: § RESULTS · plain
      “After screening 1,374 articles found in a systematic database search conducted on February 25, 2024, we included 34 study samples from 28 articles (7-11, 13, 27-48), representing 542,143 psychiatric emergency admissions (of which 61,647 involved law enforcement).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites 7-11, 13, 27-48

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    The included articles were published between 1990 and 2023 and reported data collected between 1986 and 2021 in 11 countries.

    “Published between 1990 and 2023, these articles reported on data collected between 1986 and 2021 in 11 countries.”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    Police-referred PED admissions were significantly more likely to lead to involuntary inpatient admission (RR=4.38) than admissions not involving police.

    “Separate meta-analytic models revealed that police-referred PED admissions were significantly more likely to lead to inpatient treatment and involuntary inpatient admission (Table 1).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.

  • May combine two claims.

    Findingown data

    Bipolar disorder was associated with an increased risk of police referral, with a relative risk of 1.41, but this did not reach statistical significance.

    “Patients with bipolar disorders had an overall increased risk (RR=1.41)for being referred by police that did not reach statistical significance.”
    Quote in the paperNot a substantial claim

  • May combine two claims. The quote only partly states this claim.

    Findingown data

    Female gender was a negative predictor of admission among police-referred PED patients, with a relative risk of 0.72; equivalently, police-referred patients were more likely to be male than other PED patients.

    “The most important predictor of police-referred PED admission was aggressive behavior (RR=4.73),followed by homelessness (RR=1.84)and a diagnosis of a psychotic (RR=1.45)or substance use (RR=1.33)disorder, whereas female gender was a negative predictor (RR=0.72)(Table 1).”
    • Also stated in: § CONCLUSIONS · plain
      “police-referred patients were more likely to be homeless, to be male, and to have a diagnosis of substance use or psychotic disorders”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § Referral of Patients to Psychiatric Emergency Departments by Police: A Systematic Review and Meta-Analysis

    There was no significant difference in police-referral rates to PEDs between pre-pandemic and during-pandemic periods; both had referral rates of 9.6%, with a weighted mean relative risk of 0.91 (95% CI=0.70-1.19, I2=86%).

    “The pandemic had no apparent effect on police-referral rates to PEDs.”
    • Also stated in: § Proportions of Police-Referred PED Admissions Before Versus During the COVID-19 Pandemic · hedged
      “A random-effects meta-analysis did not indicate a significant difference in these rates between the two periods (both had referral rates of 9.6%), with a weighted mean of relative risk rates of 0.91 (95% CI=0.70-1.19,I 2 =86%) for police-referred PED admissions during the pandemic compared with prepandemic periods.”
    • Also stated in: § DISCUSSION · plain
      “We did not observe a significant effect of the COVID-19 pandemic on policereferred PED admission rates.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.

  • May combine two claims.

    Premisecitation

    Individuals referred to general emergency services by police show high rates of mental health problems, frequently including substance use disorders, and a high proportion comprises men in their early 30s.

    “Individuals who are referred to general emergency services by police show high rates of mental health problems, frequently including substance use disorders, and a high proportion comprises men in their early 30s (2).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [2] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Between one in 10 and one in 100 police encounters involve individuals with mental health problems.

    “Between one in 10 and one in 100 police encounters involve individuals with mental health problems (4,5).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [4] source
    • cites [5] source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Study selection, data extraction, and quality appraisal were carried out independently by two authors, with differences resolved via consensus.

    “The multistage study selection process, data extraction, and quality appraisal were carried out independently by two authors (T.G., K.D.), the results were compared between these two authors, and differences were resolved via consensus.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    In a large Seattle study, police officers made health care referrals rather than criminal legal system referrals in cases of perceived severe mental illness, suicidal individuals, and persons under the influence of a substance; fewer health care referrals were made for individuals not cooperating with law enforcement.

    “In a large study conducted in Seattle (4), police officers made health care referrals (rather than referrals to the criminal legal system) in cases of perceived severe mental illness, suicidal individuals, and persons under the influence of a substance; fewer health care referrals were made for individuals not cooperating with law enforcement.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] source

  • Classified as probably not a substantial claim.

    Premisecitation

    The 2016 review was based mainly on studies published before 2000 (63%).

    “Yet, this review relied on rather broad inclusion criteria, included heterogeneous settings of psychiatric health care, lacked statistical rigor based on meta-analytical methods, was based mainly on studies published before 2000 (63%), and reported a trend toward lower police referral rates in more recent studies.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5] source

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    The literature search was systematic and performed in MEDLINE, Embase, PsycInfo, and CINAHL.

    “We systematically searched the literature databases MEDLINE (via PubMed), Embase (via OvidSP), PsycInfo, and CINAHL (via EBSCOhost) by entering free- text and controlled vocabulary terms into search windows (see the online supplement).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Only 11 of the 28 articles met all quality criteria, leading to an overall assessment of moderate study quality.

    “Only 11 of the 28 articles met all quality criteria, leading to an overall assessment of moderate study quality.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    Study quality was assessed with a modified version of the JBI Checklist for Prevalence Studies, omitting criteria 6 and 7.

    “Study quality was assessed with a modified version of the JBI Checklist for Prevalence Studies (16), with criteria 6 and 7 omitted because of a lack of applicability to the abovementioned study aims and design (a filled-in checklist for all included studies is available in the online supplement).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [16] source

  • Classified as probably not a substantial claim.

    Premisecitation

    The 2016 review lacked statistical rigor based on meta-analytical methods.

    “Yet, this review relied on rather broad inclusion criteria, included heterogeneous settings of psychiatric health care, lacked statistical rigor based on meta-analytical methods, was based mainly on studies published before 2000 (63%), and reported a trend toward lower police referral rates in more recent studies.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The 2016 review reported a trend toward lower police referral rates in more recent studies.

    “Yet, this review relied on rather broad inclusion criteria, included heterogeneous settings of psychiatric health care, lacked statistical rigor based on meta-analytical methods, was based mainly on studies published before 2000 (63%), and reported a trend toward lower police referral rates in more recent studies.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The model was checked for small-study effects by funnel plot and Egger's regression-based test.

    “We checked the model for possible small-study effects by visually examining a funnel plot (20), supplemented by Egger's regression-based test for funnel plot asymmetry (21).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [20] source
    • cites [21] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The rate of patients brought to general emergency services by police ranges from 0.8% to 1.3%.

    “The rate of patients brought to general emergency services by police ranges from 0.8% to 1.3% (2).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [2] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The systematic review and meta-analysis was registered with PROSPERO on March 3, 2023 (registration no. CRD42023401044).

    “This systematic review and meta-analysis was registered with PROSPERO on March 3, 2023 (registration no.CRD42023401044).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    All searches were performed by two authors.

    “All searches were performed by two authors (T.G., K.D.).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    The 2016 review relied on broad inclusion criteria and included heterogeneous settings of psychiatric health care.

    “Yet, this review relied on rather broad inclusion criteria, included heterogeneous settings of psychiatric health care, lacked statistical rigor based on meta-analytical methods, was based mainly on studies published before 2000 (63%), and reported a trend toward lower police referral rates in more recent studies.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5] source

  • May combine two claims.

    Interpretationreasoning

    The decrease in admissions reported in those studies was not seen in the Berlin study, suggesting differences in local characteristics.

    “The decrease in admissions reported in these studies was not seen in the Berlin study, suggesting differences in local characteristics.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    Police officers who bring individuals with mental problems to a PED may be frustrated when those individuals are not admitted to a psychiatric ward.

    “Police officers who bring individuals with mental problems to a PED may be frustrated when these individuals are not admitted to a psychiatric ward (54).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [54] source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The review followed international guidelines in developing its methods.

    “We could include a large number of PED admission cases in this study and followed international guidelines in developing the methods for this review.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The review's results were more robust than the 2016 review because meta-analytical methods were used.

    “Our results were more robust, however, because of the use of meta-analytical methods.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The included studies did not represent all parts of the world, omitting African and South American countries, which limits the generalizability of the results.

    “The included studies did not represent all parts of the world, omitting, for example, African and South American countries, limiting the generalizability of our results.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The nine U.S. studies in the review were small, with 2,787 patients total versus 539,356 in all other studies, and less recent, with 22% conducted after 2000 versus 74% of other studies.

    “Note, however, that the nine U.S. studies (7,8,10,11,30,33,40,44,47) were rather small (N=2,787 patients in all U.S. studies vs. N=539,356 in all other studies) and less up to date: only 22% of the U.S. studies were performed after 2000 versus 74% of all other studies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductnone given

    The included studies had overall moderate quality, which appeared sufficient for the statistical approaches used.

    “An assessment of study quality revealed that the included studies had overall moderate quality, which appeared to be sufficient for the statistical approaches we used.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Incarcerated individuals show a preponderance of males and high prevalence of homelessness and substance use.

    “Interestingly, a preponderance of male individuals and a high prevalence of homelessness and substance use are also seen among incarcerated individuals (50, 57, 58).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [50] source
    • cites [57] source
    • cites [58] source

  • May combine two claims.

    Premisecitation

    The COVID-19 studies focused only on the first months of the pandemic, when admission rates in most PEDs were lower than before the pandemic.

    “However, these studies focused only on the first months of the pandemic, during which admission rates in most PEDs were lower than before the pandemic (9,27,48,(65)(66)(67).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites 9,27,48,(65)(66)(67)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Restrictive inclusion criteria limited potential between-study heterogeneity due to individual study characteristics.

    “By applying restrictive inclusion criteria, we could limit potential between-study heterogeneity due to individual study characteristics.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The higher overall inpatient admission rate among police-referred patients may also indicate a police-referral-specific factor that increases the probability of inpatient care.

    “but also hints at a potential specific factor for police referral that may increase the probability of inpatient care.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

    Interpretationreasoning

    Because police-referred patients share many characteristics with incarcerated psychiatric patients, early psychiatric treatment may help reduce the risk for incarceration of police-referred patients.

    “Because police-referred patients share many characteristics with incarcerated psychiatric patients, early psychiatric treatment may help reduce the risk for incarceration of police-referred patients.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    This is the first systematic review using meta-analytical methods to investigate rates of patients brought to PEDs by police.

    “To our knowledge, this is the first systematic review using meta-analytical methods to investigate rates of patients brought to PEDs by police.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    ACT and Housing First may also help reduce police referrals to PEDs.

    “ACT and Housing First may also help reduce police referrals to PEDs.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    In order to reduce heterogeneity, the review defined 'police involvement' more narrowly as individuals being brought to the PED by police and excluded all psychiatric services outside of hospitals.

    “An important difference between our review and the 2016 review (5) was that in order to reduce heterogeneity between studies, we defined”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [5] source
45 of 45

  • The authors state this as an assumption.

    The random-effects meta-analytical model assumes that each study sampled from a distinct and possibly unique population rather than that all studies sampled from the same population with different degrees of error.

    “We used a random-effects meta-analytical model, which, unlike a fixed-effects model, rests on the assumption that each study had sampled from a distinct and possibly unique population rather than that all studies had sampled from the same population with different degrees of error (i.e., variation in sampling from the target population) (17).”

  • The authors state this as an assumption.

    Prediction intervals assume the model variance to be an estimate.

    “We calculated prediction intervals with the method proposed by Higgins et al. (26), assuming the model variance to be an estimate.”

  • The authors state this as an assumption.

    There is substantial between-study variation in sample size.

    “As described in Methods, we assumed substantial between-study variation in sample size, and by applying a random-effects model, we tried to ensure that the influence of smaller studies on the outcomes of interest was not dwarfed by that of larger ones (see the forest plot in the online supplement).”

  • The paper relies on this without support.

    The included studies are representative enough of PED admissions worldwide to support a global estimate of the police referral rate.

    “We found that 13.7% of PED admissions were referred by police worldwide”
    “The included studies did not represent all parts of the world, omitting, for example, African and South American countries, limiting the generalizability of our results.”
4 of 4

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Included article falls outside the stated publication range

    Minor

    The paper says the included articles were published through 2023, but reference 13 is dated 2024 and is identified elsewhere in the results as one of the included studies.

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 13

  1. UN Human Rights Council, 2017, Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest Attainable Standard of Physical and Mental Health
  2. Sales GN, 1991, Hosp Community Psychiatry, 42, 950
  3. Reinish LW, 1995, Bull Am Acad Psychiatry Law, 23, 289
  4. van Buuren S, 2011, J Stat Softw, 45, 1
  5. Sanyaolu A, 2021, Infect Dis, 14, 117863372199126
  6. Downey LVA, 2010, Prim Psychiatry, 17, 53
  7. Fähndrich E, 1999, Psychiatr Prax, 26, 242
  8. Gillig P, 1990, Hosp Community Psychiatry, 41, 804
  9. Karlovic D, 2003, Alcoholism, 39, 93
  10. McNiel DE, 1991, Hosp Community Psychiatry, 42, 425
  11. Watson MA, 1993, Hosp Community Psychiatry, 44, 1085
  12. Walmsley R, 2003, World Prison Population List
  13. Nusselder KJ, 2020, Tijdschr Psychiatr, 62, 530

Cited by

Sorted by citations

Related

Authors
Goldschmidt, T., Bermpohl, F., Schreiter, S., van Bergen, A. H., Kupka, R. W., Schouler-Ocak, M., Gutwinski, S., & Deutscher, K.
Affiliations
  • Humboldt-Universität zu Berlin (DE)
  • Department of Psychiatry and Neurosciences, Charité–Universitätsmedizin Berlin, Freie Universität Berlin and Humboldt Universität Berlin, Berlin (Goldschmidt, Bermpohl, Schreiter, Schouler-Ocak, Gutwinski, Deutscher); Department of Psychiatry, Amsterdam University Medical Center, Amsterdam Public Health Research Institute, Amsterdam (Goldschmidt, van Bergen, Kupka).
ORCID
Venue
Psychiatric Services
Publisher
American Psychiatric Association Publishing
Published
2025-06-01
Volume
76
Issue
6
Pages
581-588
ISSN
  • 0022-1597 (ISSN-L)
  • 1075-2730
  • 1557-9700
DOI
10.1176/appi.ps.20240323
OpenAlex
W4408881779
Cited by
2
References
55
Citations by year
2026: 1 · 2025: 1
FWCI
0.44
Type
journal article
Language
en
Open access
Locations
  • Psychiatric Services (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Social Sciences › Psychology › Clinical Psychology › Psychiatric care and mental health services
  • Health Sciences › Medicine › Emergency Medicine › Emergency and Acute Care Studies
  • Social Sciences › Psychology › Clinical Psychology › Healthcare Decision-Making and Restraints
Keywords
police referrals, psychiatric emergency department, police use of force, aggression, homelessness, substance use disorders, psychosis, involuntary psychiatric admission, COVID-19 pandemic, mental health service utilization
MeSH terms
COVID-19, Mental Disorders, Emergency Service, Hospital, Emergency Services, Psychiatric, Ill-Housed Persons, Humans, Referral and Consultation, United States, Police

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read under definition version 3 (extract-v3) on 2026-09-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-28.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review