Article Text

Download PDFPDF

Scant justification for interhospital transfers: a cause of reduced efficiency in the emergency department
  1. G Bertazzoni1,
  2. M Cristofani1,
  3. A Ponzanetti1,
  4. A Trabalzini1,
  5. H Attalla1,
  6. C De Vito2,
  7. P Villari2
  1. 1
    Emergency Department, Policlinico “Umberto I”, Rome, Italy
  2. 2
    Department of Experimental Medicine, “Sapienza” University of Rome, Rome, Italy
  1. Professor G Bertazzoni, Emergency Department, Policlinico “Umberto I” Rome, Via del Policlinico 155, 00161 Rome, Italy; giuliano.bertazzoni{at}


Introduction: Interhospital transfers are one of the critical points of the emergency system, which often cause overcrowding of the emergency department (ED) and limit its effectiveness.

Methods: A retrospective study was carried out, analyzing the clinical case files concerning the ED of the Policlinico “Umberto I” in Rome (Latium region, Italy) with the aim of establishing the reasons for the numerous unjustified transfers.

Results: From 1 January to 30 June 2006, 77 597 admissions to the ED occurred, and 861 patients (1.1%) were sent from other hospitals. 361 patients out of 861 (41.9%) were transferred with critical clinical conditions. The remaining 500 patients (58.1%) were transferred requiring specialised care. The need for specialised care was confirmed in 230 cases (46.0%) and therefore these transfers could be considered justified. The other 270 transfers (54.0%) were unjustified: 138 patients remained in the hospital to which they had been sent, contributing to crowding of the ED; 132 patients were returned, thereby placing them at additional risk.

Conclusion: Unfamiliarity with the regulations governing interhospital transfers is the main cause of scantly justified transfers and the consequent reduction in efficiency of the ED in the receiving hospital.

View Full Text

Statistics from

The arrival by ambulance of patients coming from another emergency department (ED) or from first-aid posts is a critical point in the emergency system, in Italy as well as in all developed countries. This can result in a lessened efficiency of the recipient ED, particularly when the transfer is not justified. The magnitude of this problem is underestimated, given the difficulty of gathering homogenous data, as a result of different legislation in every country and the different procedures to regulate interhospital transfers implemented at the local level.

A number of articles have recently been published on this topic. Kellermann1 identified unnecessary or unjustified transfer of patients as one of the causes of the “crisis in the emergency department”. In his view, overcrowding of ED in the USA results from an increase in ED visits (+26% in the past 10 years) and the reduction in ED (−9%) and their numbers of beds.1 2 Burt et al3 further described the extent of this phenomenon, suggesting the lack of beds, the high number of admissions to ED (14% by ambulance) and the complexity of cases as the reasons behind the transfers. In 2007, Dunn et al4 published a comprehensive review of interhospital transfers of critical patients in the UK, highlighting all the difficulties concerning the different phases of transfer organisation.

During the past 5 years managerial policies in Italy have been implemented to reduce the number of transfers due to lack of beds. Interhospital transfers are, however, often needed for two reasons: (1) the patient’s critical condition (need for urgent and/or highly complex interventions) and (2) the need for specialised care (given the inability of the admitting structure to provide human and equipment resources to complete the processes of diagnosis and therapy—eg, cardiac surgery, neurosurgery, neonatal pathologies that need ad hoc specialists and structures).

Interhospital transfers are one of the critical points in the management of the emergency system in Italy,5 and may cause overcrowding of ED, particularly in the metropolitan area.6 7 Therefore, a retrospective study was carried out analysing the justification for transfers to one of the largest ED in Italy.


The retrospective study was carried out at the second level ED of the Policlinico “Umberto I” in Rome (Latium region, Italy), the main teaching hospital of “Sapienza” University. This ED is the centre of a net that includes approximately 10 hospitals in Rome and 20 hospitals outside Rome, serving a basin of nearly 1.5 million users in an area of approximately 10 000 sq km. The case files are collected on a computerised system that manages the admissions and the route of patients within the structure, from triage until hospitalisation or discharge (GIPSE system).8 This system provides information on the timing of every activity.

From 1 January to 30 June 2006, 77 597 admissions to the ED were registered; 861 (1.1%) were sent from other hospitals and were selected for our survey. For all patients arriving by ambulance from another hospital the following information was obtained: (1) personal data and sex; (2) location and typology of the originating hospital or emergency unit (Italian law calls for four levels of emergency structures at increasing levels of complexity: non-hospital first-aid posts, hospital first-aid posts, first and second level ED, with the second level ED representing the highest in complexity and competence); (3) reason for the transfer (trauma; acute cerebrovascular illnesses; acute cardiovascular illnesses; internal, surgical, oculistic, neonatal and otorhino-facial pathologies); (4) evaluation of vital signs or file from the transferring hospital in order to define the gravity of the patient’s condition (Kellermann criteria);9 (5) request for specialised care (which should be unavailable in the originating hospital); (6) outcome of the emergency unit path (hospitalisation or return to the originating hospital); (7) date and hour of transfer.

Justification for a transfer results from these data (based on the congruity of at least points 4 or 5), as identified by Italian law.10

Statistical analysis

The χ2 test and multiple logistic regression analysis were performed to determine variables that were significantly associated with the probability of an unjustified patient transfer. The explanatory variables included the following: patient age (0, ⩽60 years; 1, >60 years); sex (0, female; 1, male); structure of provenance (0, other structures; 1, second level ED); cause of transfer (1, trauma; 0, other causes); hours of patient transfer (0, 08:01–15:00; 1, 15:01–08:00) and day of the week of patient transfer (0, weekdays; 1, weekend). Adjusted odds ratios (OR) and 95% CI were calculated. A p value of less than 0.05 was considered significant for all analyses, which were performed using STATA statistical software, version 8.0.


During the 6-month study period, 861 patients were sent to the second level ED of the Policlinico “Umberto I” from 29 smaller and/or less complex hospitals, 10 in Rome and the rest outside Rome. Concerning the typology of these structures, 44 patients (5.1%) were transferred from a second level ED, 310 (36.0%) from a first level ED, 471 (54.7%) from hospital first-aid posts, 11 (1.3%) from non-hospital first-aid posts and 25 (2.9%) from hospitals without first-aid posts or ED.

Trauma was the most frequent pathology involved (33%), followed by cerebrovascular illnesses (31%) and cardiovascular illnesses (16%) (fig 1). The average patient age was 53.7 years (SD 26.1); 535 (62.1%) patients were male and 326 (37.9%) female. Trauma was particularly common in the age range 15–60 years (table 1).

Figure 1 Transfers to the second level emergency department of Policlinico “Umberto I”, by type of pathology.
Table 1 Transfers to the second level emergency department of Policlinico “Umberto I”, by patient age and type of pathology

Most of the interhospital transfers occurred between 13:00 and 01:00 hours (fig 2), quite uniformly during the week (table 2). Trauma was a particularly common cause of transfer during the weekend (table 2).

Figure 2 Transfers to the second level emergency department of the Policlinico “Umberto I”, by hour of day.
Table 2 Transfers to the second level emergency department of the Policlinico “Umberto I”, by day of the week and type of pathology

A total of 361 of 861 patients (41.9%) were transferred for critical clinical conditions documented by compromised vital signs and/or by the need for life support. In these cases the transfer was considered justified. Of these 361 patients, 344 (95.3%) were admitted to specialised units of the Policlinico “Umberto I”, whereas 17 (4.7%) returned within 12 h to the originating hospital after laboratory and diagnostic analysis as well as stabilisation of vital functions.

The remaining 500 patients (58.1%) were transferred in clinical conditions that were not critical, but with the request for specialised care by the doctor arranging the transfer. The need for a specialist was confirmed in 230 cases (46.0%) and, therefore, these transfers were considered justified. A total of 188 of these 230 patients (81.7%) were admitted to the Policlinico “Umberto I”, whereas 42 (18.3%) were returned to the originating hospital. By contrast, the specialist deemed the examination not to be urgent and/or the therapy necessary or not such as to warrant the risk of transfer for 270 patients (54.0%); in these cases the transfer was considered unjustified either because these cases were not complex or difficult, or because the hospital that sent the patients had sufficient human resources and/or infrastructure to manage them.

Of the 270 patients transferred who were not in critical condition and had no need of a specialist, 51.1% were admitted to Policlinico departments, thereby adding a hospital admission to an unjustified transfer. The remaining patients (48.9%) were sent back to the originating first-aid post or ED.

The lack of justification for patient transfers was significantly more frequent at the univariate analysis if patients were transferred from a second level ED (p = 0.002) and if the patient transfers occurred in the period from 03:01 to 08:00 hours (p = 0.030). Unjustified patient transfers were more common if the reason for transfer was trauma (p<0.001). The probability of a lack of justification tended to be higher, albeit not significantly, if the patient was female (p = 0.078) and aged 60 years or less (p = 0.125). Results of the multiple logistic regression analysis confirmed almost entirely the results of the univariate analysis (table 3). The probability of a lack of justification was higher in the case of transfer from a second level ED (OR 2.44; 95% CI 1.29 to 4.63) and if the reason for transfer was trauma (OR 2.48; 95% CI 1.81 to 3.40) and lower if the patient was male (OR 0.66; 95% CI 0.48 to 0.90). The association between lack of justification and hour of patient transfer did not reach statistical significance at the multivariate analysis.

Table 3 Multiple logistic regression model assessing the influence of different explanatory variables on the lack of justification for patient transfers


Interhospital transfers during the study period that concerned the second level ED of the Policlinico “Umberto I” accounted for less than 11% of all the transfers occurring during the same period in the Latium region. A total of 193 (23.1%) of 861 transferred patients returned to the originating hospital. This percentage was higher than that reported in the entire region, in which 13% of more than 7000 transfers were sent back to originating hospitals.8

Most transfers originated from a first level ED and hospital first-aid posts (approximately 90%). Some of our data (the day of the week and the hour of the day featuring the highest number of transfers, the average age of patients and the prevalence of male patients, the most common pathologies involved) closely resemble those reported in the American3 and European11 literature as well as in previous Italian data.12

The interhospital transfer is a particularly complex situation, both for the doctor who has to organise it according to objective criteria justifying the critical condition and the complexity of the case,1315 and for the patient who has to be granted safe conditions in order to overcome the stress that might aggravate his or her clinical condition.10 16 Therefore, the doctor must take into account both the risks and the benefits of a transfer, which should be prepared according to well-defined criteria:4 1719 (1) information to be given to the patient and his or her relatives; (2) communication between the transferring and the receiving doctors; (3) stabilisation of the patient’s condition; (4) transport safety; (5) reception by ambulance crew; (6) identification of transfer-admitting personnel and assurance of bed availability in the specialist unit.

The evaluation of transfer justification should be considered for every stage of the system. According to our experience, item nos 2 and 6 appear to be the most important aspects. Failing to comply with these points denotes poor communication between the parties involved, each aimed at guaranteeing their own decision-making privilege. These points might differ in the Italian system from those of other countries. In Italy and particularly in the Latium region transfers generally occur from hospitals of minor complexity to those of major complexity, and are also driven by the major availability of technological and human resources of large hospitals, which, in any case, is a decisive factor determining the reception and the hospitalisation of patients coming from hospitals of minor complexity. This is a major cause of overcrowding in Italian second level ED.

Interhospital transfer guidelines have been elaborated over time by various international scientific bodies.1315 Similarly, Latium region health authorities have formulated “clinical–organisational criteria for transfers of critical patients within the emergency network” through the Public Health Agency.10 However, these recommendations are not always followed by doctors, who often carry out transfers without verifying their actual need and without an adequate exchange of information with their colleagues at transfer-admitting hospitals, which, if of a superior level, according to Italian law, are obliged to admit the patient even if no specialist bed is available.

Lack of health operator training in the process of decision-making and organisation of transfers has been considered by the literature to be one of the causes of unjustified transfers.4 11 16 20 21 Statistical analysis of our data was aimed at identifying factors leading to unjustified transfers: a few factors concerning the patient have turned out to be statistically significant. One of them seems especially relevant: a second level ED that engages in an aid-continuity transfer to another ED of equal complexity is a department that does not implement all its specialist competence. In this sense, the transfer is unjustified and this ED should not be considered a top-quality ED (political issue). Interestingly, while preparing this paper, the second level ED of the hospital with unjustified transfers was declassified to a first level ED.

In conclusion, this study emphasises that unjustified interhospital transfers are a major healthcare issue in Italy, as a consequence of poor training in the laws and recommendations that govern interhospital transfers and inadequate organisation in effecting them. There is a clear need for training courses on this issue, because the disadvantages of an unjustified interhospital transfer involve not only the patient and his or her family (ethical issue), but also the whole community (social issue) and the system of health planning in terms of resource allocation (health policy issue). Paraphrasing a statement by Simpson and Smith,22 we may say that “a patient’s transfer system, as all healthcare systems, requires the best management”.


View Abstract


  • Competing interests: None declared.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.