Predictors of emergency department referral in patients using out-of-hours primary care services

Predictors of emergency department referral in patients using out-of-hours primary care services

Health Policy 120 (2016) 1001–1007 Contents lists available at ScienceDirect Health Policy journal homepage: www.elsevier.com/locate/healthpol Pred...

535KB Sizes 9 Downloads 28 Views

Health Policy 120 (2016) 1001–1007

Contents lists available at ScienceDirect

Health Policy journal homepage: www.elsevier.com/locate/healthpol

Predictors of emergency department referral in patients using out-of-hours primary care services Maria Paola Scapinello a , Andrea Posocco a,e , Irene De Ronch a , Francesco Castrogiovanni a , Gianluca Lollo a , Guglielmo Sergi a , Iginio Tomaselli a , Loris Tonon a , Marco Solmi b,c,f , Stefano Traversa a , Vincenzo Zambianco a , Nicola Veronese a,d,f,∗ a

Unità Locale Socio Sanitaria 7, Pieve di Soligo (TV), Continuità Assistenziale Vittorio Veneto (TV), Italy Department of Neurosciences, University of Padova, Padova, Italy c Unità Locale Socio Sanitaria 17, Monselice, Italy d University of Padova, Department of Medicine, Geriatrics Section, Italy e Scuola di Formazione Specifica in Medicina Generale, Regione Veneto, Venice, Italy f Institute of Clinical Research and Education in Medicine (IREM), Padova, Italy b

a r t i c l e

i n f o

Article history: Received 11 January 2016 Received in revised form 21 July 2016 Accepted 25 July 2016 Keywords: Out-of-hours primary care Emergency department Gate-keeping

a b s t r a c t Background: Out of hours (OOH) doctors could manage many cases limiting the inappropriate accesses to ED. However the possible determinants of referral to ED by OOH doctors are poorly studied. We aimed to characterize patients referred from the OOH to ED service in order to explore the gate-keeping role of OOH service for hospital emergency care and to facilitate future research in improving its cost-effectiveness. Methods: A retrospective study was made through data collection of 5217 contacts in a local OOH service in the North-East of Italy (from 10/01/2012 to 03/31/2013). Results: Only 8.7% (=454 people) of the total contacts were referred to ED. In the multivariate analysis, the significant predictors of being sent to ED were: age; residence in nursing home (odds ratios (OR) = 2.00, 95%CI: 1.30–3.10); being visited by a OOH physician (OR = 2.64, 95%CI: 2.09–3.34). Taking infections as reference, cardiovascular diseases (OR = 18.31, 95%CI: 12.01–27.90), traumas (OR = 8.75, 95%CI: 5.36–14.26) and gastrointestinal conditions (OR = 7.69, 95%CI: 4.70–11.91) increased the probability to be referred to ED. Conclusions: OOH service addresses several common medical conditions in communitydwelling and in nursing home context, supporting its filtering function for the ED access. The main reasons of ED access could be a crucial aspect in general population education in order to avoid the overcrowding of the ED. © 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction ∗ Corresponding author at: Department of Medicine—DIMED, Geriatrics Division, University of Padova, Via Giustiniani 2, 35128 Padova, Italy. Fax: +39 0498211218. E-mail addresses: [email protected], [email protected] (N. Veronese). http://dx.doi.org/10.1016/j.healthpol.2016.07.018 0168-8510/© 2016 Elsevier Ireland Ltd. All rights reserved.

The overcrowding of emergency departments (ED), mainly due to its inappropriate use, is a common problem in different countries. Some studies [5,11] investigated

1002

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007

the prevalence and determinants of inappropriate access to ED in order to find possible solutions. Among the most important risk factors for inappropriate access to ED, difficulty in accessing primary health care seems to be relevant [5]. Out of hours (OOH) service in Italy is an important part of primary health care of the National Health System (NHS) providing support to patients who need general practitioner (GP) consultation before the next working day. The NHS, founded in 1978 on the model of the British NHS, is a mainly public system financed by general taxation. GPs work for Local Health Units as independent contractors and they act as gate-keepers for higher levels of care. The nationally shared agreement signed in 2005 defines the tasks, activities and salaries of GPs working in hours and OOH primary care doctors. Recent evidence shows that interventions aiming to improve primary care accessibility may reduce the use of emergency services [8,19]. In this perspective also an uninterrupted medical availability during night time, holidays and week-ends seems to be important [10,14,15,21]. From the literature we know that OOH doctors could safely and effectively manage many cases otherwise referring to ED [7,20] and that in some countries the OOH physicians already manage the large majority of the afterhours problems presented in general population [13]. On the other side non urgent patients frequently go directly to ED for several reasons, such as immediate accessibility, self-perceived urgency [1], perceived need to diagnostic facilities and the belief that hospital specialists are more competent than GPs [12], alongside with the typical selfreferred patient to ED (young man accessing for traumatic injuries) [6,9,17]. In this context, medical referral could be a protective factor for inappropriate accesses to ED [2]. Decision making whether referring or not a patient to the ED is a crucial point in OOH doctors work. To our knowledge there is only one study [3] analyzing the possible determinants of referral of ED by OOH doctors, which suggests that some demographic (e.g. age and gender), context-related (e.g. hour of phone call and distance from OOH service) and some clinical variables (e.g. kind of disease and being residents nursing home) could influence this decision [3]. Our aim is to characterize patients sent from OOH service to ED and to identify the most relevant predictors of referral to ED. With our study we explore whether this service could serve as a filter to hospital emergency care, in order to facilitate future research in improving its costeffectiveness. 2. Material and methods This retrospective study was approved by the Local Institutional review board on July 2015. 2.1. Context and setting The OOH service is part of the primary care system of the INHS (Italian National Health System). In Italy, OOH services are now provided by about 12,000 physicians working in 2952 OOH service delivery points [3]

that work every day of the week from 8 pm to 8 am (night), and during weekends from 10 am on Saturdays to 8 am on Mondays, plus bank/religious holidays, and from 8 am to 8 pm on days when GPs are attending continuing education courses (GPs substitution). The OOH service of Vittorio Veneto (“Servizio di Continuità Assistenziale Vittorio Veneto”) belongs to a greater area of Local Health Unit 7 (in Italian “U.L.S.S. 7”). OOH service of Vittorio Veneto is located in the town of Vittorio Veneto (province of Treviso) and covers the territory of other 7 smaller villages (Cappella Maggiore, Colle Umberto, Cordignano, Fregona, Revine Lago, Sarmede and Tarzo). This area includes land, hills and mountains. Three physicians during day shifts and two during night shifts guarantee medical attention availability for about 53,742 inhabitants. In Italy the OOH service is organized by physicians working as independent contractors for Local Health Units, being remunerated on a fixed hourly rate. However, they work on the same patients and they usually belong to the same geographical area of the GPs. OOH physicians have no access to medical data of patients recorder in GPs’ software and OOH physicians can not prescribe any complementary exams. Similarly to other OOH services in Italy, Vittorio Veneto OOH doctors directly receive telephone requests from patients or caregivers, either providing advice over the phone or offering domiciliary visits (at home or in a nursing home) or asking patients to come to OOH service for medical examination. A free access to OOH doctors’ office is given to all the patients every night from 8 to 10 pm and Sunday from 10 to 12 am and from 4 to 6 pm. The nearest ED is about 3 km far from the OOH service, and is located in the same town.

2.2. Extraction and categorization of the data This study was based on data collected from October, 1st 2012 to March, 31th 2013 in Vittorio Veneto OOH service. Any ambulatory visit, home (or nursing home) visit, or telephone consultation (made only by OOH physicians) was defined as a contact with OOH service. The data of the patients accessing the OOH service were recorded in a paper register containing information about the hour of intervention, name, surname, age, town/village, gender, reason of the contact and type of contact. All these data (except name and surname) were extracted in a standardized Excel sheet by two independent Reviewers (MPS and AP). A third Reviewer checked the quality of the extraction and was available in case of disagreement (NV). Age was categorized in three groups using 18 and 65 years indicating the adult and older age in Italy, respectively. Provenience was categorized as Vittorio Veneto vs. other towns (Cappella Maggiore, Colle Umberto, Cordignano, Fregona, Revine Lago, Sarmede and Tarzo). A variable was identified to define vulnerable patients, i.e. nursing home residents and palliative care needing patients. Type of contact could be phone calls, domiciliary or ambulatory visits. The types of shift included: Saturday, Sunday, Night, Holidays (e.g. Christmas, Easter, New Year day), GP substitution and shifts before holydays.

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007

1003

Table 1 Baseline characteristics of subjects according to the referral to the emergency department (ED) or not. ED (n = 454)

No ED (n = 4763)

p-Value

Age (mean ± SD), years Age <18 years (%) Age ≥65 years (%)

60.8 ± 29.9 14.6 58.1

51.1 ± 25.1 12.9 38.8

<0.0001 0.45 <0.0001

Females (%) Nursing home (%) Palliative care (%)

56.4 11.0 0.7

49.3 3.1 1.1

0.01 <0.0001 0.24

Type of visit Domiciliary visit (%) Ambulatory visit (%)

40.1 17.9

9.6 11.6

<0.0001 0.002

Type of shift Saturday (%) Before holidays (%) Night (%) Holidays (%) Sunday (%) General practitioner substitution (%)

15.0 5.5 52.6 7.7 18.1 1.1

16.4 6.9 47.9 6.9 19.0 2.9

Provenience Vittorio Veneto (%)

54.4

57.1

0.11

0.28

Notes: Unless otherwise specified, data were presented with mean ± standard deviations (SD) for continuous and percentages for categorical variables, respectively. P-Values were calculated using an independent t-test for continuous and chi-square test for categorical variables, respectively.

The reasons for encounter at OOH service were summarized using the Ninth Revision of the International Classification of Diseases (ICD-9). Two additional categories were created, named 0 for certification of death and 17 for information about therapy or practical/logistic information about health system. The classification was based on the reason of the contact (symptoms or the supposed diagnosis) which the OOH physicians wrote on the register after the first contact with the patient (usually by phone). 2.3. Definition of ED referral All patients were screened for any access to ED within 24 h after OOH service intervention. The referral to ED was double-checked through the use of a software (NF Star version 12.0.6/2015) that records medical data of all patients accessing the ED, and through the consultation of OOH register in which doctors were supposed to write all the referrals. The 39 patients who were sent to ED, but did not access were included in analysis anyway as referrals. 2.4. Statistical analysis The dataset was divided in two groups by referral or not to ED. The normal distribution of the continuous variables was checked using a Shapiro–Wilk test and continuous variables were expressed as means ± standard deviations if normal distribution was satisfied. Differences between the groups were analyzed using an independent t-test for continuous variables, and a chi-square test for categorical variables. A multivariate analysis was ran using all factors resulting significantly different (p < 0.05) in the univariate analysis in order to find the most significant predictors of ED referral. Collinearity was estimated with the VIF (variance inflation factor), taking a cutoff of 2 as an exclusion criterion. However, no variable was excluded for this

reason. To overcome convergence problems in the multivariate analysis, medical conditions with less than 100 cases and/or less than 10 patients sent to ED (congenital anomalies and conditions originating in the perinatal period, endocrine, hematological and pregnancy diseases) as well as deceased patients and generic information requests were not included. All the statistical tests were two-sided, considering a p value <0.05 as statistically significant. All analyses were performed using the SPSS for Windows, rel. 21.0 (SPSS Inc. Chicago). 3. Results The 5217 patients included in this study aged 52.8 ± 25.3 years [range: 0–101], with a slight majority of women (50.3%) and prevalently accessed to OOH service from Vittorio Veneto (56.3%). About half of the whole shifts were night shifts (48.5%) and about two thirds of the contacts were phone calls. The residents in nursing home accounted for the 4.5% of the total sample, while 1.0% were in palliative care. In the same period 5511 patients accessed the ED of Vittorio Veneto. Fig. 1 represents, as percentages, the prevalence of medical conditions seen by OOH service. The most frequent conditions were infections (23%), followed by psychiatric diseases (16%), generic information about therapy (10%) and cardiovascular diseases (CVD) (8%). As shown in Table 1, the patients seen by OOH service and sent to the ED were 454 accounting for the 8.7% of the total sample. Those referred to ED were significantly older (p < 0.0001), more often women (p = 0.01) and residents in nursing home (p < 0.0001) than those not referred. Patients were sent to ED mainly after a domiciliary (p < 0.0001) or ambulatory examination (p = 0.002) (Table 1). Conversely, no differences according to type of shift (p = 0.11) or provenience (p = 0.28) were evident.

1004

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007

Fig. 1. Prevalence of conditions seen by emergency medical service (OOH).

Fig. 2. Prevalence of conditions in those referred to emergency department. Notes: Conditions with less than 100 cases (endocrine, hematological and pregnancy diseases), deceased and information about therapy were not considered in this analysis.

Fig. 2 shows the prevalence of medical conditions of those sent to ED. CVD were the most important cause to referral to ED (35%), followed by gastrointestinal diseases (16%) and both traumas and respiratory diseases (10%). Table 2 shows the multivariate analysis made through a logistic regression analysis, taken the referral to ED as outcome. Subjects with age lower than 18 years (OR = 2.67, 95%CI: 1.86–3.83, p < 0.0001) and those aging more than 65 years (OR = 2.45, 95%CI: 1.87–3.22, p < 0.0001) were at increased probability to be sent to ED. A similar pattern emerged for those residents in nursing home (OR = 2.00, 95%CI: 1.30–3.10, p = 0.002) and those visited by a OOH physician (OR = 2.64, 95%CI: 2.09–3.34, p < 0.0001). Taking those with infectious diseases as reference, several medical conditions seemed

to be important for referral to ED, being cardio-vascular diseases (OR = 18.31, 95%CI: 12.01–27.90, p < 0.0001), traumas (OR = 8.75, 95%CI: 5.36–14.26, p < 0.0001) and gastrointestinal conditions (OR = 7.69, 95%CI: 4.70–11.91, p < 0.0001) the most strongly related, while psychiatric conditions were mainly addressed by OOH service (OR = 0.26, 95%CI: 0.11–0.61, p < 0.0001). 4. Discussion In this study, we examined the predictors of ED referral of patients firstly accessing OOH service. Other than age, the most significant predictors seem to be residence in nursing home, the type of contact with OOH service and certain medical conditions.

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007

1005

Table 2 Significant predictors of emergency department (ED) referral. Odds ratio

Age Age 18–65 Age <18 years Age ≥65 years Other characteristics Females Nursing home Ambulatory or domiciliary visit

Reference 2.67 2.45 0.86 2.00 2.64

95% Confidence intervals Lower

Higher

p-Value

1.86 1.87

3.83 3.22

<0.0001 <0.0001

0.68 1.30 2.09

1.08 3.10 3.34

0.20 0.002 <0.0001

Type of shift Saturday (%) Before holidays (%) Night (%) Holidays (%) Sunday (%) General practitioner substitution (%)

Reference 0.70 1.24 0.87 0.94 0.65

0.41 0.89 0.53 0.64 0.24

1.21 1.73 1.43 1.38 1.78

Medical conditions Infective disease Cancer Psychiatric conditions Central nervous system diseases Cardiovascular diseases Respiratory diseases Gastrointestinal disease Genitourinary diseases Dermatological diseases Musculoskeletal diseases Allergies Traumas

Reference 5.70 0.26 2.36 18.31 5.34 7.69 3.75 1.34 0.39 2.14 8.75

2.84 0.11 1.37 12.01 3.27 4.70 2.19 0.51 0.12 0.91 5.36

11.45 0.61 4.08 27.90 8.72 11.91 6.43 3.54 1.29 5.00 14.26

0.20 0.20 0.58 0.75 0.40

<0.0001 0.002 0.002 <0.0001 <0.0001 <0.0001 <0.0001 0.55 0.12 0.08 <0.0001

Notes: Unless otherwise specified, data are presented as odds ratios and 95% confidence intervals. Conditions with less than 100 cases (endocrine, hematological and pregnancy diseases), deceased and information about therapy were not considered in this analysis.

Only 8.7% of the patients accessing OOH service were sent to ED, suggesting that this service is an important filter to avoid inappropriate access to ED, in agreement with previous studies [7,13,20]. Anyway most people (5511) chose to go to ED during after-hours. Inappropriate accesses to ED seem to be a relevant topic in terms of costs, quality of life of patients and of physicians [5,11]. Therefore, the interest in other services mitigating inappropriate accesses to ED is increasing. Among studies strategies to solve this issue, interventions aiming to improve primary care accessibility seem to be among the most useful [8], in particular through an efficient telephonic triage [4,11]. A recent work showed that the quality of telephone triage in some OOH delivery points was low, with the need for a more proper selection of those referred to ED [16]. In this context, our findings add to previous evidence, contributing to education of physicians working in OOH service and their triage activity. In this perspective, OOH Italian service could be a model of uninterruptedly accessible primary care service, being a doctor always available for phone consultations and medical examinations. Although only a little percentage of people encountered at the OOH service is sent to ED, our work can be important in the early identification of those few cases, in order to eventually develop specific “ED referral adequacy measures” in future studies, or specific interventions for those groups.

In this context our study shows that age is a relevant risk factor. Patients younger than 18 years and older than 65 were more likely to be sent to ED by OOH doctors probably because children in Italy are normally seen in primary care by pediatricians and older people have generally more co-morbidities. This pattern is the opposite of the classic ED self-referred non urgent patient, typically an adult aged between 18 and 65 [12]. Moreover, Buja et al. already found a significant association between older age and access to ED [3], consistently with our results. In addition residents of nursing home were more likely to be referred to ED. A possible explanation is that people admitted in these structures are monitored by nurses, who already filter less severe conditions and that call OOH doctors only for more complicated cases. However this phenomenon, already found by Buja et al. [3], should be better studied in other specifically focused works. Patients were sent to ED more often after a visit than a simple phone contact, especially after a domiciliary intervention. This could be explained considering that people examined at home or invited to come to ambulatory visits were selected after the first telephonic contact, while less urgent cases were solved just by phone advice. Therefore, in our study, medical telephonic triage seems to be adequately performed. Taking as reference the infectious diseases, suspected cardio-vascular diseases were the most common reasons

1006

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007

for the referral to ED, since they often need urgent investigations to exclude life-threatening conditions. Traumas appeared to be a risk factor too, maybe because they frequently need urgent diagnostic radiology and orthopedic interventions not feasible in the outpatients OOH service. The prevalence of respiratory diseases was also relevant among referrals, including cases of respiratory insufficiency. Traumas, respiratory and cardiovascular conditions were found significantly related to referral to ED also in the work of Buja et al. [3]. Our study also shows that gastrointestinal diseases were associated to increased probability for referral to ED, probably because this category includes pathologies that require an urgent surgeon consultation, such as several causes of acute abdomen. Conversely psychiatric diseases seem to be statistically a protective factor, indicating that most cases of this kind can be managed just by phone call. In our work, gender, provenience and type of shift did not appear to be independent risk factors for referral to ED, differently from results of the study of Buja et al. [3]. This divergence, however, is probably due to differences between areas, practices and individual doctors as suggested in a review about this topic [18]. Our findings could help to improve the efficiency of referral to ED, focusing the attention on the most problematic situations. Our results suggest that OOH physicians need to be familiar with the evidence based medicine guidelines about some particular conditions (e.g. cardio-vascular diseases) and especially for pediatric and geriatric/nursing home patients. We think that future research is needed to tailor specific guidelines on how to manage these vulnerable patients in the OOH primary care setting, particularly without complementary exams. On the other side, we think that some simple and cost-effective diagnostic tools could probably improve the work of OOH physicians in some common conditions (e.g. ECG for suspected cardiovascular diseases). Moreover, increased referral after a face-to-face assessment, underlines the great importance of medical telephonic triage in filtering less severe conditions. We believe that the use of checklists, protocols, or decision-supporting software packages could further improve this process to appropriately decide if the patient needs or not to be visited in the ambulatory or at home. The findings of our study should be interpreted within its limitations. The main is that this work does not consider whether people invited by OOH physicians to ED were appropriately sent or not and we do not know who sent to ED the patients. Other limitations are the lack of a severity of complaints score and of some important confounders (e.g. ethnicity, formal education or number of medications) in our analysis, that could affect our results. Moreover, we used the ICD-9 categories to classify the reason for encounter, that is a classification proposed for diseases and not for complaints or symptoms as usually seen by OOH physicians. However, since ICD-9 classification is widely used by hospital’s specialists, including those working in ED, we believe that this choice could be appropriate for the interpretation of our results. Finally, this study was retrospective and with a short follow-up period (only 6 months), possibly introducing a bias since the recorded information

were not tailored for our primary outcome, i.e. the referral to ED. In conclusion, our study demonstrates that OOH service addresses several common medical conditions in community-dwelling and in nursing home context, supporting its filtering function for the ED access. Our findings could be useful to define the areas in which OOH service should be improved in order to further limit the access to ED. Additional studies are needed to evaluate the appropriateness of the referral to ED. Since most patients, mainly with non-urgent complaints, go directly to ED in afterhours time in our local context, identifying the main reasons of this choice and informing people about proper healthcare service referring should be a crucial aspect in general population education. Future research is of course needed in this sense. Conflict of interest The authors report no conflict of interest. Source of funding None. References [1] Afilalo J, Marinovich A, Afilalo M, Colacone A, Léger R, Unger B, et al. Nonurgent emergency department patient characteristics and barriers to primary care. Academic Emergency Medicine 2004;11(12):1302–10, http://dx.doi.org/10.1197/j.aem.2004.08.032. [2] Bianco A, Pileggi C, Angelillo IF. Non-urgent visits to a hospital emergency department in Italy. Public Health 2003;117(4):250–5, http://dx.doi.org/10.1016/S0033-3506(03)00069-6. [3] Buja A, Toffanin R, Rigon S, Sandonà P, Carraro D, Damiani G, et al. Out-of-hours primary care services: demands and patient referral patterns in a Veneto region (Italy) Local Health Authority. Health Policy 2015:1–10, http://dx.doi.org/10.1016/j.healthpol.2015.01.001. [4] Campbell JL, Fletcher E, Britten N, Green C, Holt TA, Lattimer V, et al. Telephone triage for management of same-day consultation requests in general practice (the ESTEEM trial): a cluster-randomised controlled trial and costconsequence analysis. The Lancet 2014;384(9957):1859–68, http://dx.doi.org/10.1016/S0140-6736(14)61058-8. [5] Carret MLV, Fassa AG, Domingues MR. Inappropriate use of emergency services: a systematic review of prevalence and associated factors. Cadernos de Saúde Pública 2009;25(1):7–28. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19180283. [6] Chmiel C, Huber CA, Rosemann T, Zoller M, Eichler K, Sidler P, et al. Walk-ins seeking treatment at an emergency department or general practitioner out-of-hours service: a crosssectional comparison. BMC Health Services Research 2011;11:94, http://dx.doi.org/10.1186/1472-6963-11-94. [7] Dale J, Green J, Reid F, Glucksman E, Higgs R. Primary care in the accident and emergency department: II. Comparison of general practitioners and hospital doctors. BMJ (Clinical Research Ed.) 1995;311(7002):427–30. Retrieved from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid= 2550494&tool=pmcentrez&rendertype=abstract. [8] Flores-Mateo G, Violan-Fors C, Carrillo-Santisteve P, Peiró S, Argimon J-M. Effectiveness of organizational interventions to reduce emergency department utilization: a systematic review. PLoS One 2012;7(5):e35903, http://dx.doi.org/10.1371/journal.pone.0035903. [9] Giesen P, Franssen E, Mokkink H, van den Bosch W, van Vugt A, Grol R. Patients either contacting a general practice cooperative or accident and emergency department out of hours: a comparison. Emergency Medicine Journal 2006;23(9):731–4, http://dx.doi.org/10.1136/emj.2005.032359.

M.P. Scapinello et al. / Health Policy 120 (2016) 1001–1007 [10] Grol R, Giesen P, van Uden C. After-hours care in the United Kingdom, Denmark, and the Netherlands: new models. Health Affairs 2006;25(6):1733–7, http://dx.doi.org/10.1377/hlthaff.25.6.1733. [11] Ismail SA, Gibbons DC, Gnani S. Reducing inappropriate accident and emergency department attendances: a systematic review of primary care service interventions. The British Journal of General Practice: The Journal of the Royal College of General Practitioners 2013;63(617):e813–20, http://dx.doi.org/10.3399/bjgp13X675395. [12] Moll van Charante EP, ter Riet G, Bindels P. Self-referrals to the A&E department during out-of-hours: patients’ motives and characteristics. Patient Education and Counseling 2008;70(2):256–65, http://dx.doi.org/10.1016/j.pec.2007.10.012. [13] Moll van Charante EP, van Steenwijk-Opdam PC, Bindels PJ. Out-of-hours demand for GP care and emergency services: patients’ choices and referrals by general practitioners and ambulance services. BMC Family Practice 2007;8(46):8–46, http://dx.doi.org/10.1186/1471-2296-8-46. [14] O’Kelly FD, Teljeur C, Carter I, Plunkett PK. Impact of a GP cooperative on lower acuity emergency department attendances. Emergency Medicine Journal 2010;27(10):770–3, http://dx.doi.org/10.1136/emj.2009.072686. [15] O’Malley AS. After-hours access to primary care practices linked with lower emergency department use and less unmet medical need. Health Affairs 2013;32(1):175–83, http://dx.doi.org/10.1377/hlthaff.2012.0494.

1007

[16] Pasini AGR, Vaona A. A cross-sectional study of the quality of telephone triage in a primary care out-of-hours service. Journal of Telemedicine and Telecare 2015;21(2):68–72, http://dx.doi.org/10.1177/135763X14566573. [17] Philips H, Remmen R, De Paepe P, Buylaert W, Van Royen P. Out of hours care: a profile analysis of patients attending the emergency department and the general practitioner on call. BMC Family Practice 2010;11:88, http://dx.doi.org/10.1186/1471-2296-11-88. [18] Salisbury C. The demand for out-of-hours care from GPs: a review. Family Practice 2000;17(4):340–7, http://dx.doi.org/10.1093/fampra/17.4.340. [19] van den Berg MJ, van Loenen T, Westert GP. Accessible and continuous primary care may help reduce rates of emergency department use. An international survey in 34 countries. Family Practice 2016;33(1):42–50, http://dx.doi.org/10.1093/fampra/cmv082. [20] van der Straten LM, van Stel HF, Spee FJM, Vreeburg ME, Schrijvers AJP, Sturms LM. Safety and efficiency of triaging low urgent selfreferred patients to a general practitioner at an acute care post: an observational study. Emergency Medicine Journal 2012;29:877–81, http://dx.doi.org/10.1136/emermed-2011-200539. [21] Wong WB, Edgar G, Liddy C, Vaillancourt C. Can after-hours family medicine clinics represent an alternative to emergency departments? Survey of ambulatory patients seeking afterhours care. Canadian Family Physician 2009;55:1106–7, e4. http://doi.org/55/11/1106 [pii].