- Email: [email protected]
Hospital referral of detainees during police custody in Amsterdam, The Netherlands
Journal of Forensic and Legal Medicine xxx (2016) 1e4
Contents lists available at ScienceDirect
Journal of Forensic and Legal Medicine j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j fl m
Hospital referral of detainees during police custody in Amsterdam, The Netherlands T. Dorn a, *, A. Janssen a, J.C. de Keijzer b, G.L. van Rijk-Zwikker b, U.J.L. Reijnders b, J.S.K. Luitse c, E. Vandewalle d, M.P. Gorzeman e, R.C. van Nieuwenhuizen e, M. Ceelen a, C. Das b a
Department of Epidemiology & Health Promotion, Public Health Service Amsterdam, P.O. Box 2200, 1000 CE Amsterdam, The Netherlands Department of Forensic Medicine, Public Health Service, Amsterdam, The Netherlands Department of Emergency Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands d Department of Emergency Medicine, VU University Medical Center, Amsterdam, The Netherlands e Department of Emergency Medicine, OLVG Hospital, Amsterdam, The Netherlands b c
a r t i c l e i n f o
a b s t r a c t
Article history: Available online xxx
This study describes how many detainees have been referred to emergency departments for further evaluation or emergency care while in police custody in Amsterdam (years 2012/2013). It provides insights into the diagnoses assigned by forensic doctors and hospital specialists and the appropriateness of the referrals. We made use of the electronic registration system of the Forensic Medicine Department of the Public Health Service Amsterdam. This department is in charge of the medical care for detainees in the Amsterdam region. Hospital diagnoses were obtained through collaboration with several Amsterdam-based hospitals. According to our results, in 1.5% of all consultations performed, the detainee was referred to hospital. The most frequent reasons for referral were injuries (66%), intoxication/withdrawal (11%) and cardiac problems (7%). In 18% of all referrals, hospital admission (defined as at least one night in the hospital) was the consequence. After review of hospital files, the indication for referral as stated by the forensic physician was confirmed in 77% of all cases. A minority of referrals was considered unnecessary (7%). The identified cases allow for a discussion of cases of over-referral. Future research should focus on the problem of under-referral and associated health risks. © 2016 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
Keywords: Care in custody Forensic physicians Referral Detainees Hospital treatment
1. Introduction Annually, almost 40.000 persons are arrested by the Amsterdam police force. Many are addicted to drugs, mentally ill, suicidal, aggressive or intoxicated.1e4 Most of them are suspects of crime, but some persons are taken to the police station because they are endangering themselves or others by their behavior, e.g. psychiatric patients in crisis. In the Amsterdam region, the medical care at the police station is provided by a primary care team consisting of forensic physicians and nurses employed by the Amsterdam Public Health Service. The catchment area of the area covers the city of Amsterdam and a number of neighboring communities
* Corresponding author. Public Health Service Amsterdam, Department of Epidemiology and Health Promotion, P.O. Box 2200, 1000 CE, Amsterdam, The Netherlands. Tel.: þ31 (0) 20 555 5495; fax: þ31 (0) 20 555 5160. E-mail address: [email protected] (T. Dorn).
(Amstelveen, Uithoorn, Ouder-Amstel, Diemen, Aalsmeer, Beemster, Edam-Volendam, Landsmeer, Oostzaan, Purmerend, Waterland, Wormerland, Zaanstad and Zeevang). Almost 30% of all police detainees in Amsterdam are seen by a forensic doctor or forensic nurse.2 Care is provided at police stations and at cell blocks. At police stations, a forensic doctor can be called by the police 24/7. In contrast to police stations, the cell blocks are especially equipped for overnight stays. At these sites, a forensic nurse is present during day-time. The nurse is in charge of delivering the ‘first contact care’ and refers to the forensic doctor if necessary. At night-time, a forensic doctor is on duty. In contrast to Dutch prisons, there is no standard medical intake of all incoming police detainees. Instead, medical assessment may follow at the detainees' request or at the request of the police. Potential reasons for hospital referral are injury, an impaired level of consciousness, (suspected) drug poisoning, behavioral disturbance and any other medical condition causing serious concern. In case of somatic health problems, these patients will be referred to a hospital for further evaluation or
http://dx.doi.org/10.1016/j.jflm.2016.01.025 1752-928X/© 2016 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
2
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4
treatment. In case of psychiatric crisis, psychiatric emergency care will be organized. This study was performed to shed more light on hospital referrals. In the literature on detainees of the police, information on hospital referrals is scarce. In total, we identified five studies reporting information on this topic. Payne-James et al. describe a sample of 150 detainees who required documentation of alleged assault and associated injuries.5 23% of the sample required hospital transfer. Greenberg et al. performed research among mentally disordered detainees of which 32% were admitted to hospital.6 These detainees were held by the police in the context of legislation that gives police the power to remove these persons from a public place to a place of safety. Carter et al. report a hospital transfer rate of 1.6%, with injury, overdose and poisoning, chest pain and collapse as most frequent reasons for referral.7 In a study performed by Heide et al.,8 56 out of 1017 persons who had been re examined by a doctor were referred to hospital (5.5%). Beaufre et al. studied differences between older and younger detainees, demonstrating that older detainees (>60 years) were more often referred to hospital (3% versus 0.7%)9 The current study adds to this literature by assessing whether referrals made by health care staff are appropriate. In primary care, there is a long-lasting discussion on when a referral to hospital care is appropriate.10,11 Whereas under-referral may cause damage to the patient, over-referral may lead to unnecessary costs. We therefore attempted to assess whether or not referral was appropriate by reviewing the medical files of referred arrestees. Altogether, the current research attempts to provide answers to the following questions: 1) What percentage of all consultations delivered by the primary care team of the forensic medical department result in hospital referral? 2) What are the indications associated with referral? 3) What percentage of referrals leads to hospital admission? 4) How often is the indication for referral confirmed by hospital files? 5) What percentage of referrals can be considered inappropriate?
2. Methods In principle, when a detainee is in need of hospital treatment, three pathways are possible: (1) The forensic physician or nurse may refer to a hospital, (2) if mentally disordered, the patient is examined by a psychiatric nurse and referred to psychiatric emergency care, or (3) an ambulance is called by the police, without health care staff being involved in the decision. In this study, we attempted to establish the number of hospital referrals initiated by forensic physicians and nurses (years 2012/2013). Information on the other two pathways was not available. For this study, we made use of the electronic registration of the Forensic Medical Department of the Amsterdam Public Health Service. In total, we identified n ¼ 400 referrals initiated by forensic physicians and nurses. In order to obtain more information on these referrals, we asked all emergency departments in the catchment area of the service to provide us with the diagnosis and treatment of referred patients. Seven out of eight hospitals agreed to participate in the study. In 6 out of 400 cases, it was impossible to trace the patients in the hospital registrations because the name and/or date of birth of the patients was lacking. For the remaining 394 referrals, we were able to obtain hospital information in 244 cases (62%). Reasons for a lack of hospital information were diverse. Partly, it was unclear from our registration to which hospital the patient had been transferred (n ¼ 79). Some patients were referred to non-
participating hospitals or to hospitals outside the region (n ¼ 13). In another 56 cases, no information was available at the hospital, although it was registered as receiving hospital. Possible explanations are a wrong date of birth, misspelled patient names, a missing match between the referral date in our registration and the date according to the hospital consultation registration. After the hospital data was gathered, two experienced forensic physicians assessed whether the referral indication was confirmed (yes/no) and whether or not the referral was appropriate. The physicians assessed the referral independently from each other, based on their individual training and experience. In case of disagreement, consensus was established by discussing the case together. The appropriateness of the referral was performed using a 5-point scale (5 ¼ referral very appropriate; 1 ¼ referral inappropriate). For the purpose of data analysis, we defined referrals scored below 4 as unnecessary. This cut-off was chosen pragmatically to allow a further examination of a group of potential cases of overreferral. For this category of referrals, the forensic physicians were asked to indicate why they assigned a low score. 3. Results In total, during the study period (years 2012 and 2013), forensic physicians referred 400 times to a hospital (psychiatric emergency care excluded). This constitutes 1.5% of all consultations delivered to arrestees of the police (including both telephonic and face-toface consultations) and results from dividing the number of referrals in 2012 and 2013 (n ¼ 400) by the number of consultations in 2012 and 2013 (n ¼ 26,875). Since hospital information was not available for all referrals, the final study sample comprised 244 referrals concerning 243 patients. The patients were predominantly male (92%) and aged 35 years on average (SD ¼ 13.2). In 187 cases (77%), the referral indication as stated by the forensic physician was confirmed by the review of hospital data. In the other 57 cases, the referral indication was not fully confirmed. In 44 cases, hospital admission followed after referral, defined as at least one night in the hospital (18% of all referrals). The reasons for referral are listed in Table 1. The most frequent reasons for referral were injuries (66%), intoxication (11%) and cardiac problems (7%). In the case of injuries, the indication was confirmed in 77%, in the case of intoxication in 85% of cases. As injuries were the most frequent reason for hospital referral, we further studied the characteristics of these referrals (Table 2). Injuries requiring referral were most often located at the extremities (including arms, legs and knees), followed by injuries to the head and/or face (28%). We also examined whether it was made notice of alcohol or drug intoxication in the medical file, demonstrating that in 16% of all injury-related referrals alcohol played a
Table 1 Indications for hospital referral during police custody.
Injury Acute intoxication/withdrawal Cardiac Neurological Gastrointestinal Diabetes Lung Urinary Tuberculosis Siccelcel crisis Total
N
% of all indications
Indication confirmed n
%
160 27 16 15 11 4 3 3 3 2 244
65.6 11.1 6.6 6.1 4.5 1.6 1.2 1.2 1.2 0.8 100
123 23 11 14 5 4 2 2 2 1
76.9 85.2 68.8 93.3 45.5 100 66.7 66.7 66.7 50.0
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4 Table 2 Characteristics of referrals with injury as indication (n ¼ 160).
Location
Patient intoxicated
Inflicted by
Self-inflicted
Extremities Head and/or face Shoulder Internal Back Hip … with alcohol … with drugs Total Police Police dog Police horse
N
%
105 44 7 5 3 2 26 7 33 35 7 2 6
65.6 27.5 4.4 3.1 1.9 1.3 16.3 4.4 20.7 21.9 4.4 1.3 3.8
role and drugs in 4%. In 22% of all injuries requiring referral, the doctor mentioned that they were inflicted by the police, another 4% were self-inflicted (e.g. smashing head against the wall, hitting the wall with fist). Injuries inflicted by the police were defined as injuries directly caused by the police (e.g. hand-cuff injury). Injuries resulting from trying to escape the police were not considered police-inflicted (e.g. fractures due to jumping out of a window). We also tested whether there was a significant age difference in those with injuries as referral indication when compared to those with other indications. It turned out that those with injuries as indication were almost ten years older than those with other indications (t-test for independent samples, p < .001). Among men, the referral indication was injury in 67% of the cases. In women, it was 47%, (difference statistically non-significant; ChieSquare test, p ¼ .08). After review, 17 referrals were labeled inappropriate (7%), suggesting that it was not necessary to refer the person to hospital. Weak indications for referral mainly concerned injuries (n ¼ 12 out of 17). The remaining 5 referrals were related to intoxication (n ¼ 1), TBC (n ¼ 1), neurological (n ¼ 1), gastrointestinal (n ¼ 1) and cardiac problems (n ¼ 1). Reasons for a weak indication were that treatment in the police cell would have been possible (e.g. wound care; n ¼ 4) or that watchful waiting would have been sufficient (n ¼ 7). In four cases, the referral indication was considered weak because there was reason to suspect that the patient was faking or exaggerating symptoms.
4. Discussion This study demonstrates that 1.5% of all consultations delivered by the primary care team of the Amsterdam Public Health Service resulted in referral of the detainee to hospital. The indications associated with referral were diverse, with injuries as leading reason (66%), followed by intoxication/withdrawal (11%) and cardiac problems (7%). In 18% of all referrals, hospital admission was the consequence. After review of hospital files, the indication for referral as stated by the forensic physician was confirmed in 77% of all cases. A minority of referrals was considered unnecessary (7%). These study results are in concordance with Carter et al.7 who report a comparable hospital referral rate and a similar ranking for reasons for referral, with injuries, intoxication/withdrawal and chest pain in the lead. In the study by Heide et al.,8 indications for hospital treatment also concur, but the overall referral rate was higher than in our study (5.5%). From the study conducted by re et al.9 an overall hospital referral rate of 0.7% can be Beaufre calculated (irrespective of age), which is lower than the referral rate re et al.9 conducted research in found in our study. Whereas Beaufre Paris, France, Carter et al.7 presented figures for Sussex, England, and Heide et al.8 for two large German towns. The referral rates
3
reported by Greenberg et al.6 and Payne-James et al.5 constitute subsamples of the population of police detainees and are not directly comparable to our population, with Greenberg et al. focusing on mentally disordered detainees of the police and PayneJames and colleagues presenting figures on detainees requesting documentation of alleged assault. Given these different rates for referrals provided by different studies, the question arises whether referral rates can be compared across studies. The organization of health services for detainees differs from site to site and from country to country. Moreover, the population described in our study consists of detainees who already have been triaged for psychiatric conditions by other health care providers. In addition, our study describes referrals made by forensic physicians, with the omission of hospital transfers initiated by the police who directly call an ambulance in serious and acute cases. Our referral rate therefore should not be mistaken for the ‘total’ rate of persons transported from the police cell to hospital. The referred patients were identified using a database containing all consultations in 2012 and 2013. We attempted to assess all patients referred in 2012 and 2013. The fact that one out of seven hospitals refused to participate in the study should not have affected the representativeness of the sample for the wider population of arrestees in Amsterdam, as there are no reasons to expect that there are differences in case-mix between hospitals. It is clear that research based on medical records has many advantages, but also certain pitfalls. In general, if a referral was made, sufficient information was provided in the medical file on the reasons why the patient was referred. Nevertheless, the assessment of the appropriateness of referrals may have been affected by information bias. Those files with elaborate information are more probably labeled as appropriate after review, in contrast to files which are less elaborate. This might imply that our study provides an overestimation of the rate of over-referrals. Information bias might also account for a mismatch between the original reason for referral and what was considered to be the case at the hospital. In addition, in our study, the medical records have been screened for the occurrence of a list of conditions not directly associated with the justification of a referral, e.g. whether the injury was inflicted by the police. When no mention is made in the record of a certain circumstance, this can mean two things: (1) it was present, but not considered important enough to record or (2) it was not recorded because it was not present. Percentages on injury-related referrals occurring under the influence of drugs or alcohol (4% and 16% respectively), or injury-related referrals due to violence used by the police (22%) reported here therefore provide an underestimation of the true numbers. For research purposes, it would be preferable to record this information systematically. A good example for such a systematic medical examination in the police cell that allows for epidemiological research is provided Chariot et al.12,13 It is hard to determine whether the number of inappropriate referrals is undesirable or not. In our setting, forensic physicians provide care to patients who cannot turn to their own health care provider, giving the forensic physician a special responsibility. In this light, over-referral might be seen a good practice which is in the interest of the patient. Nevertheless, there are strategies to reduce the number of inappropriate referrals by providing systematic feedback to forensic physicians. To date, information on the outcomes of hospital referral is sent to the general practitioner of the patient only. The physician who referred the patient during custody, the forensic physician, does not automatically receive this information, but has to make an active effort to obtain it. In our view, from the point of view of quality improvement, this practice should be changed. The efforts undertaken here to establish the rate of inappropriate referrals constitute an important step as our numbers provide a benchmark for further research. Moreover, the identified
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
4
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4
cases allow for a discussion of cases of over-referral. Further research should answer the question how many of the non-referred patients should have been referred, but weren't. As already noted, there is no standard medical intake of all incoming police detainees in the Netherlands. It is therefore possible that a medical issue is not correctly identified, and consequently a detainee does not get to see a healthcare professional and therefore does not get referred to hospital. Future research could shed more light on this problem of under-referral of detainees and the associated health risks. It also would be worthwhile to perform research on near-miss events in police custody, following the examples in other health care settings.14 This would represent an opportunity to identify and correct flaws that jeopardize patient safety in the police cell. Also the decision-making process of forensic physicians should be studied more in depth. Ideally, this research should be informed by the efforts undertaken in primary care research.10,15The knowledge base on health issues among detainees of the police is still growing. Undoubtedly, this will lead to further professionalization of the work of forensic physicians. Acknowledgments This study was supported by institutional funds provided by the Amsterdam Public Health Service. We also would like to express our gratitude to the participating emergency departments for enabling this research. Conflict of interest None declared. Funding None declared.
Ethical approval None declared.
References 1. Das C, Ceelen M, Dorn T, de Jong JT. Cocaine use and sudden death: excited delirium syndrome. Ned Tijdschr Geneeskd 2009;153:1014e7. 2. Ceelen M, Dorn T, Buster MCA, Stirbu-Wagner I, Donker G, Das C. Health care issues and health care use among detainees in police custody. J Forensic Leg Med 2012;19:324e31. 3. Dorn T, Ceelen M, Buster MCA, Das C. Screening for mental illness in Amsterdam police custody. Psychiatr Serv 2013;64:1047e50. 4. Rekrut-Lapa T, Lapa A. Health needs of detainees in police custody in England and Wales. Literature review. J Forensic Leg Med 2014;27:69e75. 5. Payne-James JJ, Dean PJ. Assault and injury in clinical forensic medical practice. Med Sci Law 1994;34:202e6. 6. Greenberg N, Haines N. The use of Section 136 of the Mental Health Act 1983 in a family of rural English police forces. Med Sci Law 2003;43:75e9. 7. Carter J, Mayhew J. Audit of hospital transfers January to March 2006 from Sussex police custody. J Forensic Leg Med 2010;17:38e40. 8. Heide S, Stiller D, Lessig R, Lautenschlager C, Birkholz M, Fruchtnicht W. Medical examination of fitness for police custody in two large German towns. Int J Leg Med 2012;126:27e35. re A, Chariot P. The health of older arrestees in police cells. Age Ageing 9. Beaufre 2015;44:662e7. 10. O'Donnell CA. Variation in GP referral rates: what can we learn from the literature? Fam Pract 2000;17:462e71. 11. Davies P, Pool R, Smelt G. What do we actually know about the referral process? Br J Gen Pract 2011;61:752e3. 12. Vincent R, Beaufrere A, Chariot P. Detainees arrested for the first time in French police stations. J Forensic Leg Med 2015;31:1e6. 13. Chariot P, Briffa H, Lepresle A, Lefevre T, Boraud C. Fitness for detention in police custody: a practical proposal for improving the format of medical opinion. J Forensic Leg Med 2013;20:980e5. 14. Crane S, Sloane PD, Elder N, Cohen L, Laughtenschlaeger N, Walsh K, et al. Reporting and using near-miss events to improve patient safety in diverse primary care practices: a collaborative approach to learning from our mistakes. J Am Board Fam Med 2015;28:452e60. 15. Ringberg U, Fleten N, Forde OH. Examining the variation in GPs' referral practice: a cross-sectional study of GPs' reasons for referral. Br J Gen Pract 2014;64:e426e33.
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
Contents lists available at ScienceDirect
Journal of Forensic and Legal Medicine j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j fl m
Hospital referral of detainees during police custody in Amsterdam, The Netherlands T. Dorn a, *, A. Janssen a, J.C. de Keijzer b, G.L. van Rijk-Zwikker b, U.J.L. Reijnders b, J.S.K. Luitse c, E. Vandewalle d, M.P. Gorzeman e, R.C. van Nieuwenhuizen e, M. Ceelen a, C. Das b a
Department of Epidemiology & Health Promotion, Public Health Service Amsterdam, P.O. Box 2200, 1000 CE Amsterdam, The Netherlands Department of Forensic Medicine, Public Health Service, Amsterdam, The Netherlands Department of Emergency Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands d Department of Emergency Medicine, VU University Medical Center, Amsterdam, The Netherlands e Department of Emergency Medicine, OLVG Hospital, Amsterdam, The Netherlands b c
a r t i c l e i n f o
a b s t r a c t
Article history: Available online xxx
This study describes how many detainees have been referred to emergency departments for further evaluation or emergency care while in police custody in Amsterdam (years 2012/2013). It provides insights into the diagnoses assigned by forensic doctors and hospital specialists and the appropriateness of the referrals. We made use of the electronic registration system of the Forensic Medicine Department of the Public Health Service Amsterdam. This department is in charge of the medical care for detainees in the Amsterdam region. Hospital diagnoses were obtained through collaboration with several Amsterdam-based hospitals. According to our results, in 1.5% of all consultations performed, the detainee was referred to hospital. The most frequent reasons for referral were injuries (66%), intoxication/withdrawal (11%) and cardiac problems (7%). In 18% of all referrals, hospital admission (defined as at least one night in the hospital) was the consequence. After review of hospital files, the indication for referral as stated by the forensic physician was confirmed in 77% of all cases. A minority of referrals was considered unnecessary (7%). The identified cases allow for a discussion of cases of over-referral. Future research should focus on the problem of under-referral and associated health risks. © 2016 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
Keywords: Care in custody Forensic physicians Referral Detainees Hospital treatment
1. Introduction Annually, almost 40.000 persons are arrested by the Amsterdam police force. Many are addicted to drugs, mentally ill, suicidal, aggressive or intoxicated.1e4 Most of them are suspects of crime, but some persons are taken to the police station because they are endangering themselves or others by their behavior, e.g. psychiatric patients in crisis. In the Amsterdam region, the medical care at the police station is provided by a primary care team consisting of forensic physicians and nurses employed by the Amsterdam Public Health Service. The catchment area of the area covers the city of Amsterdam and a number of neighboring communities
* Corresponding author. Public Health Service Amsterdam, Department of Epidemiology and Health Promotion, P.O. Box 2200, 1000 CE, Amsterdam, The Netherlands. Tel.: þ31 (0) 20 555 5495; fax: þ31 (0) 20 555 5160. E-mail address: [email protected] (T. Dorn).
(Amstelveen, Uithoorn, Ouder-Amstel, Diemen, Aalsmeer, Beemster, Edam-Volendam, Landsmeer, Oostzaan, Purmerend, Waterland, Wormerland, Zaanstad and Zeevang). Almost 30% of all police detainees in Amsterdam are seen by a forensic doctor or forensic nurse.2 Care is provided at police stations and at cell blocks. At police stations, a forensic doctor can be called by the police 24/7. In contrast to police stations, the cell blocks are especially equipped for overnight stays. At these sites, a forensic nurse is present during day-time. The nurse is in charge of delivering the ‘first contact care’ and refers to the forensic doctor if necessary. At night-time, a forensic doctor is on duty. In contrast to Dutch prisons, there is no standard medical intake of all incoming police detainees. Instead, medical assessment may follow at the detainees' request or at the request of the police. Potential reasons for hospital referral are injury, an impaired level of consciousness, (suspected) drug poisoning, behavioral disturbance and any other medical condition causing serious concern. In case of somatic health problems, these patients will be referred to a hospital for further evaluation or
http://dx.doi.org/10.1016/j.jflm.2016.01.025 1752-928X/© 2016 Elsevier Ltd and Faculty of Forensic and Legal Medicine. All rights reserved.
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
2
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4
treatment. In case of psychiatric crisis, psychiatric emergency care will be organized. This study was performed to shed more light on hospital referrals. In the literature on detainees of the police, information on hospital referrals is scarce. In total, we identified five studies reporting information on this topic. Payne-James et al. describe a sample of 150 detainees who required documentation of alleged assault and associated injuries.5 23% of the sample required hospital transfer. Greenberg et al. performed research among mentally disordered detainees of which 32% were admitted to hospital.6 These detainees were held by the police in the context of legislation that gives police the power to remove these persons from a public place to a place of safety. Carter et al. report a hospital transfer rate of 1.6%, with injury, overdose and poisoning, chest pain and collapse as most frequent reasons for referral.7 In a study performed by Heide et al.,8 56 out of 1017 persons who had been re examined by a doctor were referred to hospital (5.5%). Beaufre et al. studied differences between older and younger detainees, demonstrating that older detainees (>60 years) were more often referred to hospital (3% versus 0.7%)9 The current study adds to this literature by assessing whether referrals made by health care staff are appropriate. In primary care, there is a long-lasting discussion on when a referral to hospital care is appropriate.10,11 Whereas under-referral may cause damage to the patient, over-referral may lead to unnecessary costs. We therefore attempted to assess whether or not referral was appropriate by reviewing the medical files of referred arrestees. Altogether, the current research attempts to provide answers to the following questions: 1) What percentage of all consultations delivered by the primary care team of the forensic medical department result in hospital referral? 2) What are the indications associated with referral? 3) What percentage of referrals leads to hospital admission? 4) How often is the indication for referral confirmed by hospital files? 5) What percentage of referrals can be considered inappropriate?
2. Methods In principle, when a detainee is in need of hospital treatment, three pathways are possible: (1) The forensic physician or nurse may refer to a hospital, (2) if mentally disordered, the patient is examined by a psychiatric nurse and referred to psychiatric emergency care, or (3) an ambulance is called by the police, without health care staff being involved in the decision. In this study, we attempted to establish the number of hospital referrals initiated by forensic physicians and nurses (years 2012/2013). Information on the other two pathways was not available. For this study, we made use of the electronic registration of the Forensic Medical Department of the Amsterdam Public Health Service. In total, we identified n ¼ 400 referrals initiated by forensic physicians and nurses. In order to obtain more information on these referrals, we asked all emergency departments in the catchment area of the service to provide us with the diagnosis and treatment of referred patients. Seven out of eight hospitals agreed to participate in the study. In 6 out of 400 cases, it was impossible to trace the patients in the hospital registrations because the name and/or date of birth of the patients was lacking. For the remaining 394 referrals, we were able to obtain hospital information in 244 cases (62%). Reasons for a lack of hospital information were diverse. Partly, it was unclear from our registration to which hospital the patient had been transferred (n ¼ 79). Some patients were referred to non-
participating hospitals or to hospitals outside the region (n ¼ 13). In another 56 cases, no information was available at the hospital, although it was registered as receiving hospital. Possible explanations are a wrong date of birth, misspelled patient names, a missing match between the referral date in our registration and the date according to the hospital consultation registration. After the hospital data was gathered, two experienced forensic physicians assessed whether the referral indication was confirmed (yes/no) and whether or not the referral was appropriate. The physicians assessed the referral independently from each other, based on their individual training and experience. In case of disagreement, consensus was established by discussing the case together. The appropriateness of the referral was performed using a 5-point scale (5 ¼ referral very appropriate; 1 ¼ referral inappropriate). For the purpose of data analysis, we defined referrals scored below 4 as unnecessary. This cut-off was chosen pragmatically to allow a further examination of a group of potential cases of overreferral. For this category of referrals, the forensic physicians were asked to indicate why they assigned a low score. 3. Results In total, during the study period (years 2012 and 2013), forensic physicians referred 400 times to a hospital (psychiatric emergency care excluded). This constitutes 1.5% of all consultations delivered to arrestees of the police (including both telephonic and face-toface consultations) and results from dividing the number of referrals in 2012 and 2013 (n ¼ 400) by the number of consultations in 2012 and 2013 (n ¼ 26,875). Since hospital information was not available for all referrals, the final study sample comprised 244 referrals concerning 243 patients. The patients were predominantly male (92%) and aged 35 years on average (SD ¼ 13.2). In 187 cases (77%), the referral indication as stated by the forensic physician was confirmed by the review of hospital data. In the other 57 cases, the referral indication was not fully confirmed. In 44 cases, hospital admission followed after referral, defined as at least one night in the hospital (18% of all referrals). The reasons for referral are listed in Table 1. The most frequent reasons for referral were injuries (66%), intoxication (11%) and cardiac problems (7%). In the case of injuries, the indication was confirmed in 77%, in the case of intoxication in 85% of cases. As injuries were the most frequent reason for hospital referral, we further studied the characteristics of these referrals (Table 2). Injuries requiring referral were most often located at the extremities (including arms, legs and knees), followed by injuries to the head and/or face (28%). We also examined whether it was made notice of alcohol or drug intoxication in the medical file, demonstrating that in 16% of all injury-related referrals alcohol played a
Table 1 Indications for hospital referral during police custody.
Injury Acute intoxication/withdrawal Cardiac Neurological Gastrointestinal Diabetes Lung Urinary Tuberculosis Siccelcel crisis Total
N
% of all indications
Indication confirmed n
%
160 27 16 15 11 4 3 3 3 2 244
65.6 11.1 6.6 6.1 4.5 1.6 1.2 1.2 1.2 0.8 100
123 23 11 14 5 4 2 2 2 1
76.9 85.2 68.8 93.3 45.5 100 66.7 66.7 66.7 50.0
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4 Table 2 Characteristics of referrals with injury as indication (n ¼ 160).
Location
Patient intoxicated
Inflicted by
Self-inflicted
Extremities Head and/or face Shoulder Internal Back Hip … with alcohol … with drugs Total Police Police dog Police horse
N
%
105 44 7 5 3 2 26 7 33 35 7 2 6
65.6 27.5 4.4 3.1 1.9 1.3 16.3 4.4 20.7 21.9 4.4 1.3 3.8
role and drugs in 4%. In 22% of all injuries requiring referral, the doctor mentioned that they were inflicted by the police, another 4% were self-inflicted (e.g. smashing head against the wall, hitting the wall with fist). Injuries inflicted by the police were defined as injuries directly caused by the police (e.g. hand-cuff injury). Injuries resulting from trying to escape the police were not considered police-inflicted (e.g. fractures due to jumping out of a window). We also tested whether there was a significant age difference in those with injuries as referral indication when compared to those with other indications. It turned out that those with injuries as indication were almost ten years older than those with other indications (t-test for independent samples, p < .001). Among men, the referral indication was injury in 67% of the cases. In women, it was 47%, (difference statistically non-significant; ChieSquare test, p ¼ .08). After review, 17 referrals were labeled inappropriate (7%), suggesting that it was not necessary to refer the person to hospital. Weak indications for referral mainly concerned injuries (n ¼ 12 out of 17). The remaining 5 referrals were related to intoxication (n ¼ 1), TBC (n ¼ 1), neurological (n ¼ 1), gastrointestinal (n ¼ 1) and cardiac problems (n ¼ 1). Reasons for a weak indication were that treatment in the police cell would have been possible (e.g. wound care; n ¼ 4) or that watchful waiting would have been sufficient (n ¼ 7). In four cases, the referral indication was considered weak because there was reason to suspect that the patient was faking or exaggerating symptoms.
4. Discussion This study demonstrates that 1.5% of all consultations delivered by the primary care team of the Amsterdam Public Health Service resulted in referral of the detainee to hospital. The indications associated with referral were diverse, with injuries as leading reason (66%), followed by intoxication/withdrawal (11%) and cardiac problems (7%). In 18% of all referrals, hospital admission was the consequence. After review of hospital files, the indication for referral as stated by the forensic physician was confirmed in 77% of all cases. A minority of referrals was considered unnecessary (7%). These study results are in concordance with Carter et al.7 who report a comparable hospital referral rate and a similar ranking for reasons for referral, with injuries, intoxication/withdrawal and chest pain in the lead. In the study by Heide et al.,8 indications for hospital treatment also concur, but the overall referral rate was higher than in our study (5.5%). From the study conducted by re et al.9 an overall hospital referral rate of 0.7% can be Beaufre calculated (irrespective of age), which is lower than the referral rate re et al.9 conducted research in found in our study. Whereas Beaufre Paris, France, Carter et al.7 presented figures for Sussex, England, and Heide et al.8 for two large German towns. The referral rates
3
reported by Greenberg et al.6 and Payne-James et al.5 constitute subsamples of the population of police detainees and are not directly comparable to our population, with Greenberg et al. focusing on mentally disordered detainees of the police and PayneJames and colleagues presenting figures on detainees requesting documentation of alleged assault. Given these different rates for referrals provided by different studies, the question arises whether referral rates can be compared across studies. The organization of health services for detainees differs from site to site and from country to country. Moreover, the population described in our study consists of detainees who already have been triaged for psychiatric conditions by other health care providers. In addition, our study describes referrals made by forensic physicians, with the omission of hospital transfers initiated by the police who directly call an ambulance in serious and acute cases. Our referral rate therefore should not be mistaken for the ‘total’ rate of persons transported from the police cell to hospital. The referred patients were identified using a database containing all consultations in 2012 and 2013. We attempted to assess all patients referred in 2012 and 2013. The fact that one out of seven hospitals refused to participate in the study should not have affected the representativeness of the sample for the wider population of arrestees in Amsterdam, as there are no reasons to expect that there are differences in case-mix between hospitals. It is clear that research based on medical records has many advantages, but also certain pitfalls. In general, if a referral was made, sufficient information was provided in the medical file on the reasons why the patient was referred. Nevertheless, the assessment of the appropriateness of referrals may have been affected by information bias. Those files with elaborate information are more probably labeled as appropriate after review, in contrast to files which are less elaborate. This might imply that our study provides an overestimation of the rate of over-referrals. Information bias might also account for a mismatch between the original reason for referral and what was considered to be the case at the hospital. In addition, in our study, the medical records have been screened for the occurrence of a list of conditions not directly associated with the justification of a referral, e.g. whether the injury was inflicted by the police. When no mention is made in the record of a certain circumstance, this can mean two things: (1) it was present, but not considered important enough to record or (2) it was not recorded because it was not present. Percentages on injury-related referrals occurring under the influence of drugs or alcohol (4% and 16% respectively), or injury-related referrals due to violence used by the police (22%) reported here therefore provide an underestimation of the true numbers. For research purposes, it would be preferable to record this information systematically. A good example for such a systematic medical examination in the police cell that allows for epidemiological research is provided Chariot et al.12,13 It is hard to determine whether the number of inappropriate referrals is undesirable or not. In our setting, forensic physicians provide care to patients who cannot turn to their own health care provider, giving the forensic physician a special responsibility. In this light, over-referral might be seen a good practice which is in the interest of the patient. Nevertheless, there are strategies to reduce the number of inappropriate referrals by providing systematic feedback to forensic physicians. To date, information on the outcomes of hospital referral is sent to the general practitioner of the patient only. The physician who referred the patient during custody, the forensic physician, does not automatically receive this information, but has to make an active effort to obtain it. In our view, from the point of view of quality improvement, this practice should be changed. The efforts undertaken here to establish the rate of inappropriate referrals constitute an important step as our numbers provide a benchmark for further research. Moreover, the identified
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025
4
T. Dorn et al. / Journal of Forensic and Legal Medicine xxx (2016) 1e4
cases allow for a discussion of cases of over-referral. Further research should answer the question how many of the non-referred patients should have been referred, but weren't. As already noted, there is no standard medical intake of all incoming police detainees in the Netherlands. It is therefore possible that a medical issue is not correctly identified, and consequently a detainee does not get to see a healthcare professional and therefore does not get referred to hospital. Future research could shed more light on this problem of under-referral of detainees and the associated health risks. It also would be worthwhile to perform research on near-miss events in police custody, following the examples in other health care settings.14 This would represent an opportunity to identify and correct flaws that jeopardize patient safety in the police cell. Also the decision-making process of forensic physicians should be studied more in depth. Ideally, this research should be informed by the efforts undertaken in primary care research.10,15The knowledge base on health issues among detainees of the police is still growing. Undoubtedly, this will lead to further professionalization of the work of forensic physicians. Acknowledgments This study was supported by institutional funds provided by the Amsterdam Public Health Service. We also would like to express our gratitude to the participating emergency departments for enabling this research. Conflict of interest None declared. Funding None declared.
Ethical approval None declared.
References 1. Das C, Ceelen M, Dorn T, de Jong JT. Cocaine use and sudden death: excited delirium syndrome. Ned Tijdschr Geneeskd 2009;153:1014e7. 2. Ceelen M, Dorn T, Buster MCA, Stirbu-Wagner I, Donker G, Das C. Health care issues and health care use among detainees in police custody. J Forensic Leg Med 2012;19:324e31. 3. Dorn T, Ceelen M, Buster MCA, Das C. Screening for mental illness in Amsterdam police custody. Psychiatr Serv 2013;64:1047e50. 4. Rekrut-Lapa T, Lapa A. Health needs of detainees in police custody in England and Wales. Literature review. J Forensic Leg Med 2014;27:69e75. 5. Payne-James JJ, Dean PJ. Assault and injury in clinical forensic medical practice. Med Sci Law 1994;34:202e6. 6. Greenberg N, Haines N. The use of Section 136 of the Mental Health Act 1983 in a family of rural English police forces. Med Sci Law 2003;43:75e9. 7. Carter J, Mayhew J. Audit of hospital transfers January to March 2006 from Sussex police custody. J Forensic Leg Med 2010;17:38e40. 8. Heide S, Stiller D, Lessig R, Lautenschlager C, Birkholz M, Fruchtnicht W. Medical examination of fitness for police custody in two large German towns. Int J Leg Med 2012;126:27e35. re A, Chariot P. The health of older arrestees in police cells. Age Ageing 9. Beaufre 2015;44:662e7. 10. O'Donnell CA. Variation in GP referral rates: what can we learn from the literature? Fam Pract 2000;17:462e71. 11. Davies P, Pool R, Smelt G. What do we actually know about the referral process? Br J Gen Pract 2011;61:752e3. 12. Vincent R, Beaufrere A, Chariot P. Detainees arrested for the first time in French police stations. J Forensic Leg Med 2015;31:1e6. 13. Chariot P, Briffa H, Lepresle A, Lefevre T, Boraud C. Fitness for detention in police custody: a practical proposal for improving the format of medical opinion. J Forensic Leg Med 2013;20:980e5. 14. Crane S, Sloane PD, Elder N, Cohen L, Laughtenschlaeger N, Walsh K, et al. Reporting and using near-miss events to improve patient safety in diverse primary care practices: a collaborative approach to learning from our mistakes. J Am Board Fam Med 2015;28:452e60. 15. Ringberg U, Fleten N, Forde OH. Examining the variation in GPs' referral practice: a cross-sectional study of GPs' reasons for referral. Br J Gen Pract 2014;64:e426e33.
Please cite this article in press as: Dorn T, et al., Hospital referral of detainees during police custody in Amsterdam, The Netherlands, Journal of Forensic and Legal Medicine (2016), http://dx.doi.org/10.1016/j.jflm.2016.01.025