The impact of implementing a psychiatric emergency hotline on the reduction of acute hospitalizations in a Swiss tertiary hospital

Hospitalization in psychiatry is not always beneficial and can even be harmful in the medium/long term [1‐6]. Furthermore, psychiatric patients are usually admitted to hospital on a voluntary basis, but in some situations, they may be committed as an involuntary patient [7]. In addition, hospital treatment and management of psychiatric patients is one of the most expensive care that can be provide to psychiatric patients [8, 9]. Despite highly developed outpatient networks in many developed countries, overcrowding of inpatient psychiatry services remains a common problem [10‐13]. This overcrowding is problematic not only in terms of inadequate conditions with respect to rooms and other facilities, but also in terms of the quality of care. It is now well established that overcrowded units have negative impact on staff who are less able to offer optimal care to patients [14‐16].

Considering a significant increase in psychiatric disorders over the last two decades [17‐20], it is important to assess different strategies to limit avoidable inpatient treatment, in particular the use of triage instruments [21, 22]. Avoidable inpatient treatment refers to situations in which outpatient treatment can be provided to the patient without at least equivalent outcomes. Outpatient treatment refers to a range of services, including mobile crisis intervention teams and acute day clinics for which non-inferiority has been demonstrated [23, 24]. In addition, for some conditions such as borderline personality disorder it is specifically recommended to avoid inpatient treatment whenever possible [25]. The reduction of unnecessary interventions is a broader objective in medicine, which is advocated by the Choosing wisely campaign [26]. Several models exist and have proven their effectiveness for this purpose [27, 28]. By focusing on day-hospitals or intensive treatment in the community, these models reduce the number of hospital admissions [7, 27]. An important question is how to address acutely ill patients to the appropriate care setting, highlighting the need of an appropriate triage. The identification of those patients needing hospital admission as the most intensive intervention is a critical task for triage that needs the development of appropriate instruments.

To date, despite compelling literature on triage instruments in different emergency medical specialties [29‐31]. Triage instruments in psychiatry have mainly been developed for assessing patients arriving at psychiatric emergency departments with a focus on the detection and management of acute life-threatening conditions [32, 33]. A Swiss research group has implemented a more comprehensive triage system to reduce the number of hospitalizations and has recently published positive results [34]. However, in addition to a triage by telephone these authors emphasized face-to-face evaluations and their approach may thus be relatively resource-intensive.

Before the start of the present study, there was no triage instrument in Geneva, Switzerland, and the admission to the psychiatric hospital was left to the discretion of the treating outpatient physician (Fig. 1a). The emergency hotline allowed the treating physician working in private practice, wishing to hospitalize a patient, to discuss the clinical situation with a specialist psychiatrist able to propose alternative solutions whenever it was possible in order to limit inappropriate hospital admissions.

The present study assessed the impact of the implementation of an emergency hotline in psychiatry. The aim was to evaluate the benefit of emergency hotline on the rate of hospitalizations in acute psychiatric wards and the number of days spent in hospital (days per patients and overall days). The hypothesis underlying this second objective is that with better indications for in-hospital treatment, treatment goals could already define before admission and the hospitalization could be limited to attaining these goals.

To our knowledge this study is the first to evaluate the impact of an emergency hotline for admission to acute psychiatric wards using an emergency hotline staffed with a specialist psychiatrist. Our results suggest that implementation of this simple emergency hotline is feasible and can reduce the number of hospitalizations following a referral from a private practitioner, with a highly significant decrease of in hospital admissions. We also observed a trend-level decrease in the average length of stay for patients hospitalized after referral. The time required to manage any situation proved to be short and almost 80% of situations could be managed in less than 30 min. As mentioned above, a Swiss research group has already implemented and tested a triage instrument in the form of a specific unit designed for face-to-face assessment [34]. Other triage tools have been used in psychiatric emergency services [21, 33]. They allow the evaluation of the urgency of the situation taking into account the severity of the disorder, the risk of violence, the socio-demographic factors, etc. Although essential, these models are not completely similar to our intervention. The differences between these triages instruments and the emergency hotline implemented in terms of means deployed and costs incurred are significant, in particular with respect to the limited resources required for the emergency hotline presented here.

Given the clear reduction in hospitalization of referred patients, it is an important question on which basis the psychiatrist on-site was able to make the decision not to hospitalize the patient. One potential decision parameter is the clinical motivation for hospitalization or diagnosis, because inpatient treatment is recommended only in exceptional cases for certain disorders, for example borderline personality disorder. However, our data show no significant effect of diagnosis on admission rate. We did not have quantitative data on illness severity, which could be relevant criterion. Another explanation for the lack of effect of diagnosis may be related to the fact that training of inpatient psychiatrists and the referring private practitioners is very similar, thus resulting in a similar evaluation of the clinical aspects of the case. However, these results are in contrast with the international literature which has shown that the severity of a psychiatric illness is positively correlated with an increase in hospitalizations rates and the length of stay in hospital [39, 40]. Other factors also influence the increase in hospitalizations rates, such as socio-demographic characteristics of individuals (e.g., living alone and being unemployed) [41]. The influence of the socio-demographic background of the subjects observed in our study could not be measured due to the lack of available data on this item. Nevertheless, in almost half of the cases, the indication for hospitalization was eventually considered as not adequate despite comparable training of referring and triaging psychiatrist. This observation raises the question of other unmeasured factors influencing the appropriate referral to hospitalization. One critical issue may be the knowledge of the available psychiatric facilities, in particular for intensive outpatient and assertive community treatment. These programs have been strongly developed in Geneva over the year preceding the present study. Therefore, psychiatrists that have worked in private practice for some time may have lacked knowledge in term of psychiatric facilities in their city. In contrast, the psychiatrist on-site had excellent knowledge, because all the facilities belong to the same division.

Thus, the emergency hotline became a tool for coordinating the patient trajectory or simply advice from experts on the currently existing mental healthcare system. Therefore, this study brings another perspective on the management of the psychiatric care system and hospital admissions. Indeed, with this lack of knowledge of the healthcare system in place and the need to discuss situations that are complex and difficult to manage alone, it seems essential to improve communication between private practitioners and hospital doctors. Here, we only focus on the reduction of admission rates, but the present results suggest that this form of hotline for coordinating patient trajectories could have a broader use including the optimal orientation of patients without referral for hospitalization. We have not received any complaints about the implementation of such a procedure. On the contrary, the possibility to discuss complex clinical situations was well received by the doctors working in private practice.

However, it is important to note that addressing these situations has also been made easier by the mental health services existing in Geneva. Implementing the same hotline in a place where outpatient mental health services are non-existent would probably not have the same effect on the reduction of hospitalizations observed in our study. Thus, this aspect must be taken into account in the generalization of this finding.

As mentioned above, the emergency hotline that has been implemented was not without cost. However, the average time for managing these triage situations proved to be short and we estimated the total cost of the hotline to be approximately 2428 CHF over the three months post-intervention period. Nevertheless, a decrease in healthcare costs was also estimated at 111′840 CHF following our intervention, showing a significant impact on the reduction of healthcare costs. Additionally, the hotline proved to be remarkably simple to use because only one doctor had to be involved in the management of admissions. It was also fairly easy to find people who could take on the emergency hotline responsibility, since the task was shared among psychiatrist residents from the hospital. Thus, no training was necessary, and no new positions had to be created, as these doctors were experts in their field and were perfectly familiar with the healthcare system in place in Geneva. Moreover, in view of the impact of our emergency hotline on the number of hospitalizations, we are certain there is need for a psychiatric triage system. Although this study focused on the clinical aspect and the decisions made when hospitalization requests are processed, we believe it is important to reflect on the impact of our emergency hotline when it is put into an economic and public health perspective.

With hospitalization costs reaching approximately 1000 CHF per day, versus about 400 CHF per day for intensive outpatient clinical care, inappropriate hospitalizations can have considerable effects on health care costs. Reducing treatments and medical interventions are of utmost importance today if we want to maintain and improve the quality of care available. At a time when healthcare costs are constantly rising [42], and in a context of an increasingly unstable global economic situation, it seems imperative that these issues be taken into consideration and that we introduce such instruments into the management of psychiatric hospitalization. Reducing costs by limiting inappropriate hospital admission has become a priority and is one of the preferred themes of the Choosing Wisely campaign across the world that seeks to help physicians and patients engaging in conversations about unnecessary procedures [26]. Sharing evidence around overuse of hospitalization in psychiatry is essential in raising physician awareness and encouraging behavior change.

The use of an emergency hotline psychiatric ward is feasible, simple and seems effective to reduce the proportion of hospitalization. The investment to implement the triage tool seems small compared to the benefit on the management of admissions in acute care psychiatric units. Moreover, beyond the practical aspect and the effect on the number of hospitalizations, this intervention has undoubtedly a high impact on health care costs, making it an essential tool in the management of health care systems when viewed from a public health perspective. Future research should aim to further elucidate the relationship between this triage tool and psychiatric patient management in this specific setting using consistent methodological approaches.