Over the last 30 years there has been a significant reduction in inpatient beds in NHS general hospital and mental health settings, with figures ranging from 299,000 in 1987/88 to 141,000 in 2018/19. Large cuts have happened in mental health and learning disability inpatient wards; comprising about 73% and 97% respectively.

A major proportion of the cuts in mental health and learning disability beds are due to shifts in policies and initiatives that emphasise care for people in the community.

There are numerous inter-related factors, such as patient demand, demographics and patient behaviour, national guidelines and local circumstances, but also initiatives such as the model of care, access standards, internal hospital processes and discharge, availability and coordination with other services, and workforce supply, which can all affect the optimum need for inpatient beds or service at a time and setting (King’s Fund, 2017).

There have been a number of solutions and pathways identified nationally to improve the provision of acute psychiatric care that include service assessment and improvement involving commissioners, providers and clinicians. The aim is to simplify structures and find innovative solutions in order to share resources, deliver services and reduce waiting times following assessment to inpatient admission or home treatment (Crisp et al., 2016). Another objective has been to develop trusted assessment schemes to reduce delays in discharge (NHS, 2017).

Previous studies have reported patient and service-related factors such as demographics, severity and chronicity of illness, readmissions and detention under the mental health act impacting on length of stay. A systematic review by Tulloch et al. (2011) suggested a set of patient characteristics having moderate effects on length of the stay. However, the review was restricted to studies carried out in the USA. Recent primary European studies report factors such as affective and psychotic disorder, severity and chronicity of illnesses, readmissions and involuntary detentions to have a significant effect on length of stay (Wolff et al., 2015; Dimitri et al., 2018).

The authors of this paper aimed to explore which patient and service level factors influence the length of the stay in a busy urban mental health trust in the UK (Crossley & Sweeney, 2020).

Methods    

The authors conducted a retrospective cohort study of inpatient admissions in eight inpatient wards during May 2017. Data was collected from electronic inpatient records six weeks previous to admission until 1^st January 2018 on:

• Pre-admission factors, such as services assessed by the patients over a six -week period
• Demographic factors on admission
• Admission factors such as location, timing and mode of assessments, detention under the mental health act; patient-related factors such as diagnosis and observation levels during admission; and service factors such as out of area admissions, inter-ward transfers, locum consultant involvement, change of consultant, polypharmacy, type of the ward and involvement of the patient flow team
• Post-admission or discharge factors such as referrals to the community team, accommodation or rehabilitation.

The study was based in a mental health trust in Manchester (UK) which supports a population of 500,000 people. Of those 66.7% are white British compared to the UK average of 85.4%. A high proportion of social deprivation, unemployment, and high prevalence of severe mental illness had been compared to the national average in the population served.

The trust’s inpatient facility comprises three male acute wards, two female acute wards, one mixed-gender acute ward, one male PICU and one 72-hour short stay assessment ward with a total of around 160 beds. The authors reported that permission for this service evaluation project was obtained from the audit department, hence it did not require ethical approval.

They carried out bivariate analysis initially and utilised multiple regression models to explore the outcomes.

Results

106 patients were admitted during the study period of one month, however, only 97 patients were included as the data collection stopped on 1st January 2018. Patients who were discharged after this period were not included.

The mean age of patients was 36 years old, and the majority of the sample were male (67%). Psychosis comprised 30% of diagnosis, followed by personality disorder (21%), depression and bipolar disorder, each constituting about 8%. The median length of the stay was 22 days; the mean 36.1 days denoting a positive skew, with the longest admission lasting for 226 days.

Reduced length of stay

Patients who had a reduced length of stay included those who were:

• Known to the home treatment team, community mental health team and other community teams, in particular those who had a higher frequency of contact with the home treatment team
• Admitted to and discharged from short stay assessment wards (p <0.001).

Factors that did not impact on the length of the stay included:

• Re-admissions
• Presentation to A&E,
• Assessment under mental health act in the six weeks prior to admission
• The day, timing and location of assessments.

Longer length of stay

Patients who had a longer length of stay included those who had:

• An allocated care coordinator (p=0.033), and who had accessed secondary care outpatients (p=0.034) and admission under mental health act section 2 or 3
• Locum consultant involvement (p<0.001), increased observation levels (p<0.001), poly-pharmacy (p=0.031), increased number of ward transfers (p<0.001) and referral to accommodation on discharge (p=0.018)
• A diagnosis of psychosis or bipolar disorder (p <0.001).

Multiple regression models predicted that:

• Increased observation levels increased the length of stay up to 18 days
• Diagnosis of psychoses or affective disorders at discharge, and referral to rehabilitation increased length of stay from 18 to 72 days approximately.

Conclusions

• The results of the study support the current literature findings on patient and service-level factors that can influence the length of the stay. Patient-related factors such as increased observation levels, diagnosis of psychoses or affective disorders, and detention under the mental health act were all shown to increase the length of the stay.
• However, the study did not measure in specific outcome or severity of illness on admission or discharge.
• Moreover, there is some optimistic evidence that specific service model such as short-stay assessment ward in this Trust has shown some positive impact in reducing the length of stay.

Strengths and limitations

The strength of the study is the breadth of the data the authors used to explore factors including pre-admission, during admission and post-admission that can surely contribute to the length of the stay for people with mental health difficulties. Nevertheless, it is not clear whether a data collection instrument was used, or whether completeness and plausibility of the data were routinely checked, demands for documentation related to missed values were re-checked and clarified. Thus, we need to be cautious with the robustness of the findings.

Additionally, there is a lack of data on the severity of illnesses, re-admissions, comorbidity and risks, outcome measures, staff skills and other resources available to the patients during the time of assessment or admission. Another major limitation of the study is its small and not representative sample, relatively small period of one month of admission and follow-up of data for less than a year. Consequently, the generalisability of the findings can be considered an issue, as the study was conducted on a single site (a Trust in Manchester).

Implications for practice 

This study is a brief descriptive service evaluation audit including how patient- and service-related factors impact or have an effect on the length of inpatient stay. This study builds on existing evidence and is consistent with various factors that play a positive or negative role in the length of stay in acute inpatient wards.

A large study of this nature with a long follow-up period across different NHS Trusts in the UK or regions would be helpful to better understand this area of research. It would be also good to further investigate and evaluate the recent guidelines developed by the Royal College of Psychiatrists’ quality network for acute adult inpatients which mandates best practice (Crisp et al., 2016).

Statement of interests

None.

Links

Primary paper

Crossley, N., Sweeney, B. (2020) Patient and service-level factors affecting length of inpatient stay in an acute mental health service: a retrospective case cohort study. BMC Psychiatry 20, 438 (2020).

Other references

Crisp, N., Smith, G. and Nicholson, K. (Eds.) Old Problems, New Solutions – Improving Acute Psychiatric Care for Adults in England (The Commission on Acute Adult Psychiatric Care, 2016)

Tulloch AD, Fearon P, David AS. Length of stay of general psychiatric inpatients in the United States: systematic review. Adm Policy Ment Health. 2011 May;38(3):155-68. doi: 10.1007/s10488-010-0310-3.

Wolff, J., McCrone, P., Patel, A. et al. Predictors of length of stay in psychiatry: analyses of electronic medical records. BMC Psychiatry 15, 238 (2015). https://doi.org/10.1186/s12888-015-0623-6

Dimitri G, Giacco D, Bauer M, Bird VJ, Greenberg L, Lasalvia A, Lorant V, Moskalewicz J, Nicaise P, Pfennig A, Ruggeri M, Welbel M, Priebe S. Predictors of length of stay in psychiatric inpatient units: Does their effect vary across countries? Eur Psychiatry. 2018 Feb;48:6-12. doi: 10.1016/j.eurpsy.2017.11.001.

Embank L, Thompson J, McKenna H, and Anandaciva S. (2017). NHS hospital bed numbers: past, present and future. The King’s Fund.

NHS Long term implementation plan for mental health (2019). Last accessed: 7^th May 2021.

NHS improvement (2017). Developing trusted assessment schemes: ‘essential elements’.

Turkington, D., Moorhead, S., Turkington, G., King, C., Bell, L., & Pickersgill, D. (2020). Improving patient flow in acute psychiatric wards: Enhanced bed management and trusted assessment. BJPsych Bulletin,44(4), 159-162. doi:10.1192/bjb.2020.12

Photo credits

• Photo by Terry Vlisidis on Unsplash
• Photo by Steinar Engeland on Unsplash
• Photo by Larm Rmah on Unsplash
• Photo by Brooke Lark on Unsplash
• Photo by Daniel Tseng on Unsplash
• Photo by Kevin Hendersen on Unsplash
• Photo by Myriam Zilles on Unsplash