The importance of medical admissions units (MAU) has been emphasised by the royal colleges and the Society for Acute Medicine. This study looked at the time to treatment of four common medical conditions before and after the establishment of a dedicated MAU. Before the development of the MAU, treatment given in the emergency department (ED; median 111 minutes) was significantly quicker than on the admitting general medical ward (median 262 minutes, p<0.001). Following the establishment of the MAU, treatment given in the ED (median 70 minutes) remained significantly quicker than on the MAU (median 180 minutes, p<0.05). Treatment was given significantly quicker on the MAU compared with the antecedent admitting medical wards (p<0.05). In addition, more patients were treated within protocol-driven time guidelines. In summary, the establishment of a MAU significantly improved time to treatment, compared with admitting directly to general medical wards. This has implications for patients who are boarded directly to medical wards when the MAU is at full capacity.
Statistics from Altmetric.com
In the UK, numbers of emergency medical admissions continue to rise on an annual basis, putting increasing strain on often stretched resources. In response, the royal colleges have published a series of reports including ‘Acute medicine: the physicians role’,1 which outline the importance of dedicated medical admissions units (MAU) with sufficient senior medical cover, supported by a robust infrastructure and bed management policies.
By 2002 all but three of Scotland’s 27 acute receiving hospitals had developed a MAU, leaving just the three smallest district general hospitals (including Queen Margaret Hospital) without one.2
The implementation of a dedicated admissions unit has been shown to reduce both the median length of hospital stay and also the number of bed waits, with a significant cost saving.3 4 However, the potential impact of the establishment of a MAU on time to treatment has not previously been evaluated. We therefore undertook an audit looking at the time to standard treatment for four common medical conditions, both before and after the establishment of a MAU at Queen Margaret Hospital in 2006. Before this, the hospital operated a common front door policy in which all patients, including general practitioner (GP) referrals, were assessed in the emergency department (ED).
The four conditions and standard treatments that were studied included: low molecular weight heparin for potential acute coronary syndrome; corticosteroids for exacerbation of chronic obstructive pulmonary disease; antibiotic therapy for sepsis; antibiotic therapy for pneumonia.
Information was prospectively collected over a 10-week period preceding the establishment of a MAU. New admissions with the index conditions were identified from the admissions board, and case notes were reviewed on a daily basis. Cross-reference was made with bed management to ensure patients were not missed. Data recorded included time of admission to hospital, whether the patient was a GP referral or a self-referral to the ED, time of treatment and where the treatment was administered.
After the MAU had been established for 12 months, the audit was repeated for a further 6 weeks, assessing the same parameters, to determine any impact on time to treatment of the new admissions infrastructure (see table 1).
The median and quartiles of the times to treatment for the four index conditions are displayed in figs 1–4.
Before the establishment of the MAU (n = 141)
Treatment prescribed in the ED but given on the ward was significantly slower than that given in the ED (acute coronary syndrome p<0.001, pneumonia p<0.001, sepsis p<0.05, chronic obstructive pulmonary disease p<0.01 all Mann Whitney U-Test (MWUT)). There was no significant difference in the time to treatment of GP referrals and self-referrals to the ED. There was also no significant difference between times to treat the four conditions (p>0.05, Kruskal–Wallis test).
Following the establishment of the MAU (n = 75)
Time to treatment for all four conditions combined was quicker in the ED (median 70 minutes, quartiles 19–236) compared with direct GP referrals to the MAU (median 148 minutes, quartiles 88–401, p<0.05, MWUT) and prescriptions written in the ED but given in the MAU (median 205 minutes, quartiles 82–340, p<0.05, MWUT). Again there was no overall difference in the time to treat GP referrals compared with self-presentations to the ED. However, self-referrals to the ED were treated significantly more quickly than direct GP referrals to the MAU (p<0.01, MWUT). Once more there were no differences in time to treat the four conditions (p>0.05, Kruskal–Wallis test). There was a significant reduction in time to treatment on the MAU for all four conditions combined, compared with that on the former admitting general medical wards (p<0.05, MWUT).
Queen Margaret Hospital, Dunfermline, is a 367-bed district general hospital in semirural eastern Scotland. Until 2006 all medical patients were initially assessed by the on-call medical team in the ED. If admission was deemed necessary then they were placed in the first available bed in a medical ward. Treatment prescribed and given in the ED was significantly quicker than when prescribed in the ED ‘to be given on the ward’. This was true for all four conditions, adding on a median delay in treatment of 151 minutes. This is no surprise, as it adds several process steps to the patient’s journey before the administration of the prescribed treatment. Indeed, 100% of patients with pneumonia treated in the ED received antibiotics within the guideline 4 h; however this figure fell to 25% if given on the ward.
The 20-bed MAU was opened in September 2006, taking admissions both from the ED and direct GP referrals. It is open 24 h per day and has dedicated medical staff (two middle grade doctors, one foundation year doctor plus a consultant) plus nursing staff at levels comparable to the ED. If no beds are available in the MAU, medical admissions are diverted to the ED. General medical wards no longer receive direct admissions.
Following this, time to treatment was reassessed. Treatment (for all four pathologies combined) given in the ED remained significantly quicker than that given in the MAU. This was true both of prescriptions written in the ED to be given in the MAU (145 minutes slower) and also GP referral direct admissions to the MAU (78 minutes slower). Treatment on the MAU was, however, significantly quicker compared with the admitting general wards that it replaced (p<0.05, MWUT). This is largely due to the admission infrastructures in place, including dedicated staff, on the MAU. This improvement is even more striking considering that the majority of patients treated in the MAU (60%) were de novo GP referrals who required full assessment before initial treatment, rather than transfers from the ED with initial treatment already prescribed. Furthermore, 86% of patients with pneumonia treated in the ED received antibiotics within 4 h, compared with the improved figure of 73% on the MAU. There was no overall difference between time to treatment for all GP referrals compared with ED patients as a proportion of GP referrals (20%) were initially assessed and treated in the ED.
In conclusion, treatment in the recently established MAU is commenced significantly sooner than admitting directly to general wards, and a greater proportion of patients are treated within designated target times. However, times remain significantly slower than treatment given in the ED. The reasons behind this are manifold. Undoubtedly, the 4-h target in the ED is a major driver. Furthermore, the triage system that operates in the ED, but not the MAU, highlights those patients who require treatment most urgently. As treatment in the ED is commenced significantly quicker than elsewhere in the hospital, this highlights a potential risk in ‘fast tracking’ patients out of the department before treatment has been started. One caveat, however, is that enforcement of a ‘time to treatment’ target may increase misdiagnoses and the inappropriate use of antibiotics.8 Therefore, although time to treatment is an important quality indicator, and of clinical relevance, the importance of making a correct diagnosis should not be overshadowed.
Finally, this study has implications for patients who are boarded directly to a medical ward when the MAU is at full capacity (as is the protocol at other centres) potentially being exposed to a delay in initial treatment. This highlights the need for every hospital admission to be assessed in a dedicated unit with a robust admission infrastructure that is fit for purpose.
Competing interests None.
DJB is the guarantor for this paper.
Provenance and Peer review Not commissioned; externally peer reviewed.
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.