Improving the quality of discharge summaries

Traditionally communication between primary care and hospitals has been deemed suboptimal. The evolution of psychiatric services towards a community model of care warrants an improvement in the communication between the psychiatric services and primary care providers. There is evidence to suggest that with auditing, the quality of discharge summaries can be improved.

Aim

Our aim was to examine and attempt to improve the recording of information within psychiatric discharge summaries in an adult psychiatry department, by means of audit and feedback. 

Introduction

The discharge summary is an important communication tool which aims to summarise the therapeutic and other significant events during inpatient stay. It provides concise details of reasons leading to admission, diagnosis, investigations, etc, and is also helpful as a record of responses to different therapeutic interventions. It can be referred to years later to provide a quick summary of an inpatient stay. 

The discharge summary is useful for primary care providers to continue on the treatment strategies planned during admission. Comprehensive and timely communication between hospitals and general practitioners (GPs) is essential to ensure safe transition from the hospital back to the community. There is evidence to suggest that this may not happen on many occasions.¹ A high-quality discharge summary provided to the appropriate care provider at the right time can potentially reduce adverse events after discharge, decrease healthcare costs, and promote positive outcomes for patients.^2,3 On the other hand, poor information transfer at discharge does appear to increase the likelihood of readmission.⁴ Research suggests that GPs attribute more than one-third of post-discharge adverse events to unsatisfactory information transfer.⁵

The purpose of the discharge summary is particularly important in the context of content and timing. In a survey of the views of GPs on psychiatric discharge summaries the top five headings identified in terms of importance were: admission and discharge dates, diagnosis, medications on discharge, community key worker and date of follow-up.⁶ This is understandable since GPs are mainly concerned about implementing the post discharge care plans. 

Our study was undertaken in an acute adult psychiatric setting in order to analyse the quality of discharge summaries against local guidelines issued by Kerry Mental Health Services. These guidelines are based on Mental Health Commission Code of Practice on Admission, Transfer and Discharge to and from an Approved Centre.⁷ We also aimed to evaluate the satisfaction of GPs receiving these documents. We intended to improve the quality of these documents and resultant satisfaction gradient of the catchment area GPs by a brief educational intervention aimed at the non-consultant hospital doctors (NCHDs) responsible for completing these documents.

Methods

The study was conducted in an inpatient psychiatric unit in a general hospital that is divided into two wards, which have 19 and 23 beds respectively. Discharge summaries are prepared by the psychiatric registrars and senior house officers, typed by secretaries and sent to the patient’s GP, with copies filed to the respective case notes. 

The aim of the audit was to assess the standards of discharge summaries against local guidelines, and to examine whether feedback and educational intervention resulted in any improvement in same. The local guidelines state that the following information should be included: 

• Date of admission and discharge
• Status of admission
• Presenting complaints
• Mental state and physical examination on admission
• Progress and treatment given
• Results of investigations
• Discharge diagnosis with International Classification of Disease (ICD) code
• Medications on discharge
• Follow up arrangements. 

Moreover, the guidelines also state that letters should be typed and filed with case notes within seven days of discharge. In addition to the analysis of the summaries, the authors also sent out purpose designed, face validated, Likert type questionnaires to GPs, the purpose of which was to elicit their opinion on attributes pertaining to the discharge summaries which they received from the KGH psychiatric unit. GPs were explicitly asked for their opinion on the following attributes of the summaries: 

• Timeliness
• Information provided helpful in patient management
• If more information was needed in the summaries
• If they thought the summaries were too cumbersome
• If they felt relevant information was captured in the summaries 
• If they were happy with the overall quality of the summaries. 

The overall aim of the study (and the aim of contacting GPs) was to examine whether auditing and feedback coupled with an educational intervention would improve discharge summaries and increase GPs’ satisfaction.

Results

The summaries from case notes of 44 patients who were discharged from the acute adult psychiatric wards, over a period of three months, were obtained. Data was collected while leaving enough time for the discharge summaries to be filed in the respective case notes. In the next intake of NCHDs in the psychiatry department, the new registrars and senior house officers were briefed on the audit by the lead author. This included feedback of the results from the first cycle of the audit, focusing mainly on the areas for improvement, along with reiteration of the local guidelines. Clinicians were encouraged to adhere to these guidelines in an attempt to improve quality of discharge summaries. 

The cycle was then repeated for another three months with 48 summaries analysed. All the discharge summaries were evaluated using a questionnaire which was devised following above mentioned guidelines. This questionnaire was face validated by the authors and other consultants working in the hospital. Data was analysed using SPSS version 17. 

Table 1 shows the comparison of the information recorded in the discharge summaries by the registrars between cycle one and two. There were significant changes in ‘Past med history’ Chi Sq χ = 0.00 and ‘Past psych history’ Chi Sq χ = 0.03 between cycle one and cycle two.

In addition, reporting information such as the ‘ICD code’ χ = 0.00, the ‘Mental state’ χ = 0.01, ‘Personal/social history’ χ = 0.04 and ‘Investigation carried out’ χ = 0.02, were significant between both audit cycles. Not surprisingly, no significant differences regarding changes in reporting ‘Consultant details’ χ = 0.29, ‘Progress in hospital’ χ = 0.51, ‘Final diagnosis’ were found between cycles.

Following consultation with the GPs we compared their ratings of discharge summaries between cycles one and two (see Table 2).

Encouragingly there were some significant changes between both cycles. GPs’ ratings of ‘Happiness with time-line’ Z = 0.03 and ‘Relevant information included’ Z = 0.02 yielded significant results between cycles, while, ‘Information helpful in management of patients’ Z = 0.22, ‘Would like more info on summaries’ Z = 0.72, ‘Summaries were too cumbersome’ Z = 0.69, and ‘Happy with overall quality of summaries’ Z = 0.84 were not significant between audit cycles. These results are encouraging because even though GPs’ responses suggest that we did improve the timeliness of our discharge summaries and included more clinically relevant information, they always felt that discharge summaries helped them in the management of their patients. GPs were also happy with the content of the information in the summaries and wouldn’t like them to be more cumbersome. 

The results also suggest that even though a brief educational intervention can improve the quality of the discharge summaries, it does not necessarily translate into improved satisfaction for the GPs with these documents. Another way of looking at this would be that GPs were always happy enough with these documents.  

 (click to enlarge)

Discussion

In this study, an attempt was made to analyse the discharge summaries and compare them to guidelines. An attempt was also made to seek opinions of the GPs who receive these important documents in order to measure their satisfaction with them. The audit cycle was then completed after presenting the results from cycle one and the proposed guidelines to doctors responsible for preparing these documents. The authors hoped that with this educational intervention the quality of the summaries and the GPs’ satisfaction with them would improve. 

The results of this study resonate with previously published work which suggests that communication through discharge summaries is below par. Orel and Greenberg in their study found that only 26% of GPs had received a brief communication about an inpatient stay within two weeks of discharge.⁸ In our study less than half of the GPs surveyed were happy with the timeliness of the summaries, which is better the Orel and Greenberg study but still leaves a lot to be desired. A hypothesis to explain this could be the fact that registrars and senior house officers responsible for completing these documents are not able to do so on time due to already high demands of their work. With the current recruitment problems in the country and the resultant increased workload on the doctors, the situation may be exacerbated. Another speculation could be that the secretarial staff are also over-burdened and are not able to type the letters in a timely manner. In most of the letters, date of dictation was not recorded, which makes it difficult to ascertain at what stage the delay is happening. 

Cochrane et al found that, after discharge, an alarming 90% of elderly patients were receiving different medication regimes at home from those they had been prescribed in hospital.⁹ The psychiatric unit in KGH sends out an immediate discharge summary on the day of discharge to patients’ GPs, which facilitates medication prescribing for the GPs, but these immediate summaries are not exhaustive in the information provided. If the GPs received the typed summaries within a week of discharge, as advised by the guidelines, they will have detailed information on other aspects of treatment and follow-up as well. 

Some studies have identified problems with the quality of discharge summaries. These mainly concern timeliness, accuracy and length.^10,11,12 Our study also showed similar results pertaining to timeliness of summaries but GPs were happy with the information provided in them. The majority of GPs also seemed happy with the length of the summaries and did not believe that they were too cumbersome to read, which was different to a generally perceived notion, including that of the authors of this study. 

There is evidence in the literature to suggest that brief educational interventions are effective in bringing desired changes in the discharge summaries.^13,14 In these studies the interventions were notably more elaborate and time-consuming but our study shows that a less time-consuming educational intervention addressing the components of a good discharge summary, based on the local guidelines, can significantly improve the quality and timeliness of these documents. 

As discharge summaries are also used by admitting registrars and senior house officers in most psychiatric units for reference purposes, efforts should be made to improve areas such as full psychiatric history, investigations done and risk assessment. The authors of this study felt that despite adequate local guidelines about the quality of discharge summaries, these were not being fully adhered to and this could be improved if the doctors were more familiar with these guidelines. Efforts need to be made to educate doctors responsible for preparing discharge summaries in what constitutes a good document, so that discharge summaries fulfil their role as an effective communication tool between hospitals and GPs.

References

1. van Walraven C, Taljaard M, Bell CM et al. Information exchange among physicians caring for the same patient in the community. Canadian Medical Association Journal 2008; 179:1013-8
2. van Walraven C, Seth R, Austin PC, Laupacis A. Effect of discharge summary availability during post-discharge visits on hospital readmission. J Gen Internal Med 2002; 17:186-92
3. Moore C, McGinn T, Halm E. Tying up loose ends: Discharging patients with unresolved medical issues. Archives of Internal Medicine 2007; 167:1305-11
4. Olfson M, Walkup J. Discharge planning in psychiatric units in general hospitals. New Directions for Mental Health Services 1997; 73: 75-85
5. O’Leary KJ, Liebovitz DM, Feinglass J et al. Outpatient physicians’ satisfaction with discharge summaries and perceived need for an electronic discharge summary. J Hosp Medi 2006; 1:317-20
6. Dunn J, Burton S. GPs’ views on discharge summaries. Psychiatr Bull 1999; 23: 355-357
7. Code of Practice on Admission, Transfer and Discharge to and from an approved centre (http://www.mhcirl.ie/Mental_Health_Act_2001/Mental_Health_Commission Codes_of_Practice/Admission,Transfer_Discharge) Accessed September 2014
8. Orrel MS, Greenberg M. What makes psychiatric summaries useful to general practitioners? Psychiatr Bull 1986; 10:107-109
9. Cochrane RA, Mandal AR, Ledger-Scott M, Walker R. Changes in drug treatment after discharge from hospital in geriatric patients. BMJ 1992; 305: 694-696
10. Macauley EM, Cooper GG, Engeset J, Naylor AR. Prospective audit of discharge summary errors. Br J Surg 1996; 83: 788-790
11. Wilson S, Ruscoe W, Chapman M, Miller R. General practitioner-hospital communications: a review of discharge summaries. J Quality in Clinical Practice 2001; 21:104-108
12. Foster DS, Paterson C, Fairfield G. Evaluation of immediate discharge documents - room for improvement? Scottish Med J 2002; 47:77-79
13. Key-Solle M, Paulk E, Bradford K et al. Improving the quality of discharge communication with an educational intervention. Pediatrics 2010 (Oct); 126(4):734-9
14. Myers JS, Jaipaul CK, Kogan JR et al. Are discharge summaries teachable? The effects of a discharge summary curriculum on the quality of discharge summaries in an internal medicine residency program. Academic Medicine 2006; 81(10 suppl):S5-S8

Acknowledgements: The authors would like to thank the secretarial staff of KGH psychiatric unit for their help in this study and also all the GPs in Co Kerry who took part in the study.