Nonurgent patients in the emergency department? A French formula to prevent misuse

A cross-sectional study was conducted in the adult ED of La Conception Hospital in Marseille, France. Data was collected during a week, from 9 am to 8 pm. Each day, two time slots of two hours corresponding to peak ED consultation were randomly selected.

We included all patients aged 18 years and older presenting to the ED during the study period, arriving by their own means and identified as nonurgent by a triage nurse in the triage area. Consent for an interview and medical chart access was obtained from the patients. Confidentiality was ensured. Patients were excluded if they had communication difficulties not related to their presenting complaint.

Data were collected immediately after the triage process and before examination by the medical staff of the ED. Following informed consent, patients were invited to complete a questionnaire with a research assistant in the ED. The standardized questionnaire included 32 items [Additional file 1] The first part of the questionnaire covered socio-demographic data (age, gender, country of origin, marital status) and socio-economic characteristics (level of education, employment status, health insurance status); the second part covered the patient's usual source of care (follawed by a GP, number of GP consultations in the last year); the third part covered the ED visit (day for visiting the ED, principal reason for attending the ED, duration of presenting problem, previous contact with a physician for the same reason prompting the patient to visit the ED, reference to the ED, level of urgency perceived by the patient on a scale from 0 for "no urgency" to 20 for "extremely urgent problem").

A last part was designed to assess the willingness of patients identified as nonurgent to be reoriented to a hypothetical PCU outside the ED and to explore factors associated or not with this reorientation (such as guarantee to be received in consultation, a PCU near the hospital, and request for an ED physician's agreement for the reorientation...).

At the end of the ED visit, a short questionnaire was completed by ED physicians for each patient included [Additional file 2]. The variables covered investigations and treatments performed in the ED, and referral and discharge decisions made (home or hospital admission).

Data were collected on Microsoft Excel software and analyzed on Spss 15.0. We compared all the variables characterizing the patient and the ED visit with the patient's willingness to accept or not the reorientation to a PCU. Data were analyzed by using frequencies and χ² tests to investigate the association between dichotomous and categorical variables. The Mann-Whitney U test was used for ordinal variables and Student's t test for continuous variables.

Moreover, we used a multivariate logistic regression method to quantify the strength of the independent associations between factors linked to the patient and the willingness regarding reorientation to a PCU. Variables obtaining a p value < 0.2 in the univariate analysis were included in the logistic regression analysis.

Our study does not need to be approved by an ethics committee under the criteria of the bioethics law. Indeed, the transposition into French law of Directive 2001/20/CE, which relates to good clinical practice in the conduction of trials on drugs for human use, has required the modification of certain provisions that concern the protection of persons participating in biomedical research, in particular those provisions concerning the conditions for the authorization of biomedical research. The law 2004-806 of August 9, 2004 categorize three research categories: biomedical research, research on standard care, and non interventional research [28]. The non interventional research does not need to be approved by an Committee for the Protection of Persons, because of its observational nature and does not require the authorization of the National Commission for Informatics and Freedom due respect for patient anonymity.

The mean age of nonurgent patients was 36.3 years (SD, 11.7; range 18-70). In this group, 58.8% were men and 69.4% were employed [Table 1]. Most of the nonurgent patients had medical insurance (80%); 23.5% of them were covered by French national health insurance for individuals and families with low income and resources. This insurance is called the "CMU". The majority of nonurgent patients reported being followed by a GP (70.6%) [Table 2].

Presenting problems had lasted less than 24 hours for two third of patients identified as nonurgent by triage nurses. One third of nonurgent patients had tried to contact their GP before presenting to the ED. Most nonurgent patients were self-referred (76%); the others were referred by their GP (17.6%) or referred for medico-legal reasons (5.9%) (employer, police) [Table 3].

The most common reasons provided for attending the ED were pain (65.8%) and a need for diagnostic investigations (37.6%). Half the nonurgent patients were consulting for traumatologic problems. Nearly one quarter (22.3%) came to the ED because of difficulty in accessing their usual sources of care.

The mean level of urgency perceived by the patient was 10.6 ± 5.6 (median = 10; range 1-20). One third received additional investigations; only 6 patients received treatment, and none were hospitalized.

More than two-thirds of patients identified as nonurgent (68.2%) would accept being reoriented to a PCU by the triage nurse. Patients accepting the reorientation were significantly more likely to be employed, benefited less from "CMU" insurance [Table 2], and had a lower mean level of perceived urgency [Table 3]. In the multivariate analysis, willingness to accept reorientation was associated with employment status and the level of urgency perceived by the patient [Table 4]. Employed persons were 4.5 times more likely to accept reorientation (odds ratio (OR), 4.5; 95% confidence interval (CI), 1.6-12.9). Inversely, persons who perceived a high level of urgency were the least likely to accept a reorientation (OR, 0.9; 95% CI, 0.8-0.9).

Thirteen percent of nonurgent patients willing to accept reorientation did not want to go to a PCU located outside the hospital. Several nonurgent patients agreeing to be reoriented to a PCU outside the hospital stipulated certain conditions: a guarantee to be received in consultation (23.5%), a PCU near the hospital (23.5%), and the agreement of an ED physician for the reorientation (17.6%).

Among the nonurgent patients refusing reorientation (31.8%), almost 41% would be willing to pay an extra fee for being treated at the ED. These patients had a higher level of education (p = 0.038).