ORIGINAL PAPER
In-patient diabetes care: the impact of diabetes outreach team on long-term glycaemic control H. Siddique,1 M. M. Alam,1 A. W. Safi,1 W. B. Leong,2 K. Crowley,1 A. A. Tahrani2,3
1
SUMMARY
What’s known
Aim: Our aim is to assess the impact of inpatient diabetes services on glycaemic control in patient with diabetes admitted to a secondary care hospital in UK. Methods: We performed a retrospective analysis of all diabetes mellitus (DM) inpatients who were seen by our Diabetes Outreach Team from June 2007 to December 2010. Those with an admission diagnosis of hypoglycaemia were excluded. Blood samples including HbA1c at the initial visit and subsequent outpatient follow-up at 3–6 months were collected. Patients admitted with newly diagnosed diabetes were analysed separately. Results: In total 2002 patient data were captured. 778 patients were eliminated initially because of failure to attend follow-up clinic, lack of follow-up HbA1c data, and because of planned discharge to the community. Complete blood samples were available for 1224 patients. Of this, 235 patients (19.2% of those with complete data) were analysed separately as their primary diagnosis was hypoglycaemia. In the remaining 989 patients, 31 (3.1%) new onset Type 1 DM patients and 91 (9.2%) new onset Type 2 patients were analysed separately. In patients with known DM (n = 867) HbA1c improved from 75 mmol/mol (9.0% 2.39) to 69 mmol/mol (8.46% 2.0) (p < 0.001). In the newly diagnosed Type 1 DM (n = 31) patients HbA1c improved from 114 mmol/mol (12.55% 2.27) to 58 mmol/mol (7.43% 2.05) (p < 0.001). In the newly diagnosed Type 2 DM (n = 91) patients HbA1c improved from 93 mmol/mol (10.70% 3.04) to 56 mol/mol (7.29% 1.74) (p < 0.001). In those who presented with hypoglycaemia (n = 235) HbA1c changed from 58 mmol/mol (7.48% 1.59) to 59 mmol/mol (7.59% 1.57) (p = 0.2). Conclusion: By providing a comprehensive care, structured education and appropriate intervention through our Diabetes Outreach Team, we have shown a significant reduction in HbA1c for recently hospitalised patients.
Introduction The increasing prevalence of diabetes possesses a huge challenge to healthcare providers. Currently 2.9 million are diagnosed with diabetes across UK, which is expected to increase to over 4 million by 2025. The diabetes cost on the NHS is significant accounting up to 12% of the total budget for healthcare spending (1). This cost is further increased in patients with diabetes who require hospital admission for a variety of reasons. Between 10% and 20% of acute trust beds are occupied by patients with diabetes (2). The National Diabetes Audit for the year 2008–2009 suggests that at least 40% of in-patients with diabetes had HbA1c values more than 58 mmol/mol (7.5%) and 14.3% of in-patients with type 2 diabetes mellitus
It is part of good clinical practice to provide specialist diabetes care to hospitalised diabetes individuals. The inpatient diabetes team intervention has been shown to contribute in the reduction of hospital length of stay.
What’s new Our research demonstrated that appropriate inpatient diabetes care not only improve glycaemic control for both type 1 and type 2 DM, in those admitted with hypoglycaemia, glycaemic control remained to be well-controlled in the long term. We recommend structured eduction to empower both allied healthcare professionals and patients in the management of inpatient diabetes for better long term outcomes.
Department of Diabetes and Endocrinology, Dudley Group NHS Foundation Trust, Birmingham, UK 2 Department of Diabetes and Endocrinology, Heart of England NHS Foundation Trust, Birmingham, UK 3 Centre of Endocrinology Diabetes and Metabolism, University of Birmingham, Birmingham, UK Correspondence to: Dr Haroon Siddique, Diabetes Centre, Russells Hall Hospital, Pensnett Road, Dudley,West Midlands, DY1 2HQ, UK Tel.: + 44 1384 244 278 Fax: + 44 1384 244 280 Email: Haroon.Siddique@dgh. nhs.uk
Disclosure None.
(DM) had values more than 85 mmol/mol (10%), representing a group with very high risk in developing complications and prolonged length of stay (3). Despite increase investment, HbA1c values did not improve over the same 10 year period (4). The benefit of improved glycaemic control on diabetes-related complications is well established (5,6). Similarly, inpatient hyperglycaemia is a marker of poor outcome and increased mortality, improved glycaemic control in this cohort had been shown to decrease such complications and healthcare costs (7,8). A Healthcare Commission National Survey of in-patients found that the patient with diabetes stay on an average 2.6 days longer than patients without diabetes despite being admitted for the same procedure or condition and have a higher rate of complications (9).
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337. doi: 10.1111/ijcp.12458
1333
1334
The impact of diabetes outreach team on long-term glycaemic control
Suboptimal diabetes care results in increased bed occupancy by people with diabetes (10). It is a common practice in UK hospitals to provide diabetesrelated support to patients admitted elsewhere following a referral (10). The UK National Framework for Diabetes Standard 8 emphasised that effective care is to be delivered for patients admitted for any reason, with diabetes as comorbidity. Since then many hospitals have set up their own diabetes service to care for the patients who fall under this group. Intervention by a dedicated diabetes inpatient team resulted in decreased length of stay (11). To our knowledge, there are no published data looking at the effect of structured diabetes team on objective marker such as long-term glycaemic control (HbA1c) in hospitalised patients. Hence, the aim of this audit was to examine the impact of our front door diabetes team on glycaemic control in patients with DM admitted to our hospital.
Methods Setting Based in the heart of the Black Country, Russells Hall hospital, a large Teaching hospital of 767 beds serves more than 400,000 people. The Front Door Diabetes Service was established in 2006 consisting of a Specialist Registrar, Diabetes Specialist Nurses, Dietician, and Podiatrist, supported by a Consultant Diabetologist. The aim of the service was to take a proactive approach and support those patients who require specialist intervention, such as those admitted with diabetes-related illness or where diabetes is comorbidity. Acute ward areas such as Emergency Department, Emergency Admission Unit, Coronary Care Unit and Medical High Dependence unit were visited everyday for 5 days a week (which subsequently increased to 6 days a week) to identify these patients with a view to improve their quality of care and formulate a structured discharged plan. The opportunity is utilised to empower patients to manage their own condition with confidence as well as educate staff caring for those patients in the wards. Subsequently, patients are either discharged back to primary care, or followed up in specialist diabetes clinics at the hospital setting until their control has stabilised.
Audit methods We performed a retrospective audit of all patients who were seen by our Front Door Diabetes Service between June 2007 and December 2010. The following data were collected: type of diabetes, reason for referral, age, sex and HbA1c at baseline and at 3–6 months after discharge. Appropriate referral
was defined according to the National standard from the ThinkGlucose Initiative, an initiative to improve the care for people with diabetes when they are admitted to hospital, set up by the NHS Institute for Innovation and Improvement, in the UK.
Inclusion criteria Following patients with diabetes were included
• • • •
Diabetes-related emergencies Poor control Acute coronary syndrome Patients with vomiting, diarrhoea or requiring nasogastric feed • Patients requiring intravenous insulin therapy or with poor oral intake • Sepsis related uncontrolled hyperglycaemia
Exclusion criteria
• •
Transient hyperglycaemia Simple educational needs, good control and good self management skills • Patients requiring routine diabetes care were excluded from the audit.
Statistical analysis Data was analysed using SPSS 21.0 software (IBM, Statistical Package for Social Sciences version 21.0; SPSS, Chicago, IL). Data presented as frequencies or mean standard deviation (SD). Paired sample t-test was performed to compare baseline and follow-up HbA1c. Patients admitted with hypoglycaemia, newly diagnosed type 1 DM, and newly diagnosed type 2 DM and previously known DM were analysed separately. A subgroup analysis based on age was also performed. A p-value < 0.05 was considered significant.
Results In total 2002 patient data were captured. Seven hundred and seventy-eight patients were eliminated initially because of loss of follow-up and/or because of planned discharge to the community diabetes team. Hence, full baseline and follow-up data were available for 1224 patients. Of which, 235 patients (19.2%) had a primary diagnosis of hypoglycaemia and were analysed separately. In the remaining 989 patients, 31 (3.1%) had new onset type 1 DM and 91(9.2%) had new onset type 2 DM which were also analysed separately. Thus, we were left with 867 patients with known DM and without hypoglycaemia.
Patients with known DM (n = 867)
The mean age of this group was 64.28 19.62 years. There was a significant improvement in HbA1c ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
The impact of diabetes outreach team on long-term glycaemic control
Table 1 Summary of HbA1c at baseline and follow-up by type of referral
Known DM (n = 867) Newly diagnosed type 1 DM (n = 31) Newly diagnosed Type 2 DM (n = 91) Hypoglycaemia (n = 235)
Baseline HbA1c % (mmol/mol)
Follow-up HbA1c% (mmol/mol)
9.0 2.39 (75) 12.55 2.27 (114) 10.70 3.04 (93) 7.48 1.59 (58)
8.46 7.43 7.29 7.59
2.02 2.05 1.74 1.57
p-value
< 0.001 < 0.001 < 0.001 0.2
(68) (57) (55) (59)
Data presented as mean SD.
Table 2 The changes in HbA1c between baseline and follow-up in different age groups
Age group
Known DM (n = 867)
Newly diagnosed type 1 DM (n = 31)
Newly diagnosed type 2 DM (n = 91)
1 2 3 1 2 3 1 2 3
(n (n (n (n (n (n (n (n (n
= = = = = = = = =
195) 382) 290) 25) 4) 2) 15) 49) 27)
Baseline HbA1c (%)mmol/mol
Follow-up HbA1c (%) mmol/mol
10.29 2.8 (89) 8.78 2.14 (72) 8.42 2.01 (69) 12.15 1.95 (109) 14.83 2.92 (139) 13.10 3.54 (120) 10.80 2.84 (95) 11.03 2.78 (97) 10.05 3.56 (86)
9.17 8.28 8.24 7.62 6.43 7.15 6.94 7.37 7.35
2.47 1.84 1.81 2.20 0.40 1.91 1.54 1.71 1.94
(77) (67) (67) (60) (46) (55) (52) (57) (57)
p-value
< 0.001 < 0.001 0.16 < 0.001 0.009 0.4 < 0.001 < 0.001 < 0.001
Data presented as mean SD. Age groups: Group 1: ≤ 50 years; Group 2: 51–74 years; Group 3: ≥ 75 years.
between baseline and follow-up (Table 1). The proportion of patients with HbA1c < 53 mmol/mol (7%) at baseline compared with follow-up improved significantly [18.6% (n = 161) vs. 24.5% (n = 212), p < 0.001]. The reduction in HbA1c was observed in patients with age < 75 years, and mostly in those ≤ 50 years (Table 2). Patients aged 75 or older had small non-significant improvement in HbA1c (Table 2).
Patients with newly diagnosed type 1 DM (n = 31)
The mean age of this group was 38.90 17.315 years. HbA1c improved significantly between baseline and follow-up (Table 1) regardless of the patient’s age group (Table 2).
Patients with newly diagnosed type 2 DM (n = 91)
The mean age of this group was 66.14 14.65 years. HbA1c improved significantly between baseline and follow-up (Table 1) regardless of the patient’s age group (Table 2).
Patients admitted with hypoglycaemia (n = 235)
The mean age for this group was 71.98 13.64 years. The mean baseline HbA1c was 58 mmol/mol (7.48% 1.59). The aim of treatment in this group was to avoid ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
hypoglycaemia hence the follow-up HbA1c for this group after intervention by our diabetes outreach team was marginally but not significantly higher 59 mmol/ mol (7.59% 1.57) (p = 0.2). Subgroup analysis revealed 66.8% of patients in this cohort were above 70 years of age. The HbA1c increased mildly and non-significantly between baseline and follow-up in those aged ≤ 70 years 58 mmol/ mol (7.50% 1.75) vs. 60 mmol/mol (7.64% 1.87) (p = 0.4) or > 70 years 58 mmol/mol (7.47% 1.51) vs. 59 mmol/mol (7.56% 1.40) (p = 0.4).
Discussion Our study showed that appropriate intervention and care provided by the front door diabetes team improves HbA1c outcomes in patients with known DM and those with newly diagnosed type 1 and type 2 DM. Managing hypoglycaemia and hyperglycaemia in the in-patient setting is a good opportunity for the hospital diabetologist to positively impact on the mortality and morbidity associated with diabetes. The landmark UKPDS study clearly demonstrated that a 1% reduction in HbA1c resulted in 21% decrement in any diabetes-related end-points (12). Despite this, and the clear recommendation by NSF standard 8, in-patient management of diabetes often
1335
1336
The impact of diabetes outreach team on long-term glycaemic control
remains suboptimal. Financial constraints and the cost of setting up a dedicated diabetes team would offset many practising physicians despite their willingness and enthusiasm. Traditionally, these patients are managed by intensive insulin therapy but such measures alone can pose an increased risk of hypoglycaemia as seen in the NICE-SUGAR study (13). An integrated and dedicated diabetes in-patient management team with a balanced approach had previously shown reduction in the length of stay (11,14), reduction in the avoidable admissions and diabetesrelated bed occupancy (9) thereby reducing the healthcare costs (14). Our simple cross sectional audit focused on those patients with diabetes admitted acutely, showed a significant reduction in HbA1c not only for treatment na€ıve patients but also for those who were known to have diabetes, who were previously established on hypoglycaemic agents, through a structured and patient-centred approach.
Education is the key Hospitalisation provides a unique opportunity to address patient’s educational needs. Simple but very relevant educational gaps such as taking the right medications, injection technique, ability to recognise and correct hypoglycaemia with the right food, appropriate blood glucose testing can be addressed during their stay. We used this opportunity to educate and empower patients as well as nursing staff caring for those patients. Experience from other centres had previously shown staff education resulted in reduction in drug administration error and reduced length of stay (11,14). Our own experience is that educating staff in the general ward about basic diabetes care results in improved quality indicators and patient satisfaction as reflected by our results in the national diabetes audit (2). However the key to improving long-term glycaemic control lies with patient empowerment. With the introduction of ‘ThinkGlucose’, an initiative to improve the care for people with diabetes when they are admitted to hospital in UK, we rolled out a teaching module to all front line clinical staff. Our job plans were revisited to provide a consultant led service and our nursing staff were made available 6 days a week. Once discharged from the hospital patients were either sent back to primary care, or followed up in our specialist diabetes clinics at the hospital setting until their control has stabilised. Patients who were bed bound, unwell because of comorbid state and therefore deemed inappropriate for secondary care follow-up, those who were from out of area and those who have good community support and preferred to be followed up in the
community were discharged to the community diabetes specialist team. For those patients who were followed up in the secondary care, we rationalised their diabetes medication, discussed long-term complications, hypoglycaemia management and provided them with contact numbers for advice and support. These patients were subsequently discharged back to the community depending on the complexity of their condition and improvement in their metabolic parameters including HbA1c. The results showed significant reduction in HbA1c not only for new onset diabetes but also for those who were established on treatment. Interestingly patients with Type 1 diabetes also showed reduction in long-term glycaemic control, reiterating the fact that structured education and support from health professionals is an important factor in managing their diabetes. The greatest reductions in HbA1c were seen in those patients with the poorest control, but significant improvements were seen in all groups. The biggest drop however, was seen in those who were ≤ 50 years of age however it is this group who also had the highest HbA1c on presentation. We did not see a significant drop in HbA1c in patients who were above 75 years of age. This could possibly be explained by the conservative approach taken by our Front door diabetes team to avoid hypoglycaemia as the patients in this age group are particularly at high risk of developing hypoglycaemia. Furthermore the baseline HbA1c in this cohort is something that one would consider ‘reasonable’ (Table 2) for their age. We excluded 235 patients referred with hypoglycaemia from our analysis as the aim of the treatment in this group would be to relax the glycaemic control and allow them to have a higher HbA1c to retain hypoglycaemia awareness. Lowering the HbA1c in elderly patients may sometimes be detrimental. Poor nutritional status, associated comorbidities such as dementia, cerebrovascular accident, poor cognitive status, worsening renal and liver functions and polypharmacy may predispose them to increased risk of hypoglycaemia (15). Management should therefore be tailor-made taking the above factors into consideration as well as their ability to recognise and treat hypoglycaemia, American Diabetes Association provides clear management guidelines on this aspect to the practising physicians (16). A significant finding noted in our audit was that patients over 70 years of age account for more than 60% in this cohort. With treatment relaxation, the HbA1c in this group slightly but non-significantly worsened. One of the limitations of this real life audit is the change in staffing level and service delivery that happens with time. In 2007 when the service was introduced, the front door team was run on a small ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
The impact of diabetes outreach team on long-term glycaemic control
scale and our nursing staff covered mainly the acute areas such as emergency admission unit and coronary care unit. Capturing patients in a robust scale in the initial years were limited by time constraints and poor staffing level, however with service expansion in the subsequent years the staffing level improved allowing more time for inpatient diabetes care. Unlike in research setting, day to day clinical practice encounters such practical difficulties, which
References 1 Morgan CL, Peters JR, Dixon S, Currie CJ. Estimated costs of acute hospital care for people with diabetes in the United Kingdom: a routine record linkage study in a large region. Diabet Med 2010; 27(9): 1066–73. 2 National diabetes inpatient audit. Health & Social Care information centre. http://www.hscic.gov.uk/ diabetesinpatientaudit. 3 Bottomley JM. Cases, complications, comorbidity and costs of diabetes in UK: is treatment falling wide of the mark? Br J Diab Vasc Dis 2010; 10(5): 209–15. 4 Currie CJ, Gale EAM, Poole CD. Estimation of primary care treatment costs and treatment efficacy for people with type 1 and type 2 diabetes in the United Kingdom from 1997 to 2007. Diabet Med 2010; 27(8): 938–48. 5 UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998; 352: 854–65. 6 Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
7
8
9 10
11
12
1337
makes our findings even more realistic. A computerbased referral as well as record of review system may help to resolve such issues. Some of the Trusts across UK already have such system in place. In conclusion, by providing a comprehensive care, education and appropriate intervention through our front door diabetes team, we have shown a significant reduction in the objective marker of long-term glycaemic control for recently hospitalised patients.
on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977–86. Baker ST, Chiang CY, Zajac JD, Bach LA, Jerums G, MacIsaac RJ. Outcomes for general medical inpatients with diabetes mellitus and new hyperglycaemia. Med J Aust 2008; 188(6): 340–3. Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG. Management of diabetes and hyperglycemia in hospitals. American Diabetes Association Diabetes in Hospitals Writing Committee. Diabetes Care 2004; 27(2): 553–9. Diabetes UK. Diabetes in the UK 2011-12: key statistics on diabetes. Mahto R, Venugopal H, Vibhuti VS et al. The effectiveness of a hospital diabetes outreach service in supporting care for acutely admitted patients with diabetes. QJM 2009; 102(3): 203–7. Flanagan D, Moore E, Baker S, Wright D, Lynch P. Diabetes care in hospital–the impact of a dedicated inpatient care team. Diabet Med 2008; 25(2): 147– 51. Stratton IM, Adler AI, Neil HA et al. Association of glycaemia with macrovascular and microvascular
13
14
15
16
complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 2000; 321(7258): 405–12. Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360(13): 1283–97. Korytkowski M, Dinardo M, Donihi AC, Bigi L, Devita M. Evolution of a diabetes inpatient safety committee. Endocr Pract 2006; 12(Suppl. 3): 91–9. Abrass IB, Schwartz RS (1998) Special presentation of endocrine disease in the elderly. In Fitzgerald FT, ed. Current Practice of Medicine, vol, 1, no. 2. Philadelphia, PA: Current Medicine, Inc., 1998, pp. 415–8. Kirkman MS, Briscoe VJ, Clark N et al. Diabetes in older adults. Diabetes Care 2012; 35(12): 2650–64.
Paper received May 2013, accepted April 2014
In-patient diabetes care: the impact of diabetes outreach team on long-term glycaemic control H. Siddique,1 M. M. Alam,1 A. W. Safi,1 W. B. Leong,2 K. Crowley,1 A. A. Tahrani2,3
1
SUMMARY
What’s known
Aim: Our aim is to assess the impact of inpatient diabetes services on glycaemic control in patient with diabetes admitted to a secondary care hospital in UK. Methods: We performed a retrospective analysis of all diabetes mellitus (DM) inpatients who were seen by our Diabetes Outreach Team from June 2007 to December 2010. Those with an admission diagnosis of hypoglycaemia were excluded. Blood samples including HbA1c at the initial visit and subsequent outpatient follow-up at 3–6 months were collected. Patients admitted with newly diagnosed diabetes were analysed separately. Results: In total 2002 patient data were captured. 778 patients were eliminated initially because of failure to attend follow-up clinic, lack of follow-up HbA1c data, and because of planned discharge to the community. Complete blood samples were available for 1224 patients. Of this, 235 patients (19.2% of those with complete data) were analysed separately as their primary diagnosis was hypoglycaemia. In the remaining 989 patients, 31 (3.1%) new onset Type 1 DM patients and 91 (9.2%) new onset Type 2 patients were analysed separately. In patients with known DM (n = 867) HbA1c improved from 75 mmol/mol (9.0% 2.39) to 69 mmol/mol (8.46% 2.0) (p < 0.001). In the newly diagnosed Type 1 DM (n = 31) patients HbA1c improved from 114 mmol/mol (12.55% 2.27) to 58 mmol/mol (7.43% 2.05) (p < 0.001). In the newly diagnosed Type 2 DM (n = 91) patients HbA1c improved from 93 mmol/mol (10.70% 3.04) to 56 mol/mol (7.29% 1.74) (p < 0.001). In those who presented with hypoglycaemia (n = 235) HbA1c changed from 58 mmol/mol (7.48% 1.59) to 59 mmol/mol (7.59% 1.57) (p = 0.2). Conclusion: By providing a comprehensive care, structured education and appropriate intervention through our Diabetes Outreach Team, we have shown a significant reduction in HbA1c for recently hospitalised patients.
Introduction The increasing prevalence of diabetes possesses a huge challenge to healthcare providers. Currently 2.9 million are diagnosed with diabetes across UK, which is expected to increase to over 4 million by 2025. The diabetes cost on the NHS is significant accounting up to 12% of the total budget for healthcare spending (1). This cost is further increased in patients with diabetes who require hospital admission for a variety of reasons. Between 10% and 20% of acute trust beds are occupied by patients with diabetes (2). The National Diabetes Audit for the year 2008–2009 suggests that at least 40% of in-patients with diabetes had HbA1c values more than 58 mmol/mol (7.5%) and 14.3% of in-patients with type 2 diabetes mellitus
It is part of good clinical practice to provide specialist diabetes care to hospitalised diabetes individuals. The inpatient diabetes team intervention has been shown to contribute in the reduction of hospital length of stay.
What’s new Our research demonstrated that appropriate inpatient diabetes care not only improve glycaemic control for both type 1 and type 2 DM, in those admitted with hypoglycaemia, glycaemic control remained to be well-controlled in the long term. We recommend structured eduction to empower both allied healthcare professionals and patients in the management of inpatient diabetes for better long term outcomes.
Department of Diabetes and Endocrinology, Dudley Group NHS Foundation Trust, Birmingham, UK 2 Department of Diabetes and Endocrinology, Heart of England NHS Foundation Trust, Birmingham, UK 3 Centre of Endocrinology Diabetes and Metabolism, University of Birmingham, Birmingham, UK Correspondence to: Dr Haroon Siddique, Diabetes Centre, Russells Hall Hospital, Pensnett Road, Dudley,West Midlands, DY1 2HQ, UK Tel.: + 44 1384 244 278 Fax: + 44 1384 244 280 Email: Haroon.Siddique@dgh. nhs.uk
Disclosure None.
(DM) had values more than 85 mmol/mol (10%), representing a group with very high risk in developing complications and prolonged length of stay (3). Despite increase investment, HbA1c values did not improve over the same 10 year period (4). The benefit of improved glycaemic control on diabetes-related complications is well established (5,6). Similarly, inpatient hyperglycaemia is a marker of poor outcome and increased mortality, improved glycaemic control in this cohort had been shown to decrease such complications and healthcare costs (7,8). A Healthcare Commission National Survey of in-patients found that the patient with diabetes stay on an average 2.6 days longer than patients without diabetes despite being admitted for the same procedure or condition and have a higher rate of complications (9).
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337. doi: 10.1111/ijcp.12458
1333
1334
The impact of diabetes outreach team on long-term glycaemic control
Suboptimal diabetes care results in increased bed occupancy by people with diabetes (10). It is a common practice in UK hospitals to provide diabetesrelated support to patients admitted elsewhere following a referral (10). The UK National Framework for Diabetes Standard 8 emphasised that effective care is to be delivered for patients admitted for any reason, with diabetes as comorbidity. Since then many hospitals have set up their own diabetes service to care for the patients who fall under this group. Intervention by a dedicated diabetes inpatient team resulted in decreased length of stay (11). To our knowledge, there are no published data looking at the effect of structured diabetes team on objective marker such as long-term glycaemic control (HbA1c) in hospitalised patients. Hence, the aim of this audit was to examine the impact of our front door diabetes team on glycaemic control in patients with DM admitted to our hospital.
Methods Setting Based in the heart of the Black Country, Russells Hall hospital, a large Teaching hospital of 767 beds serves more than 400,000 people. The Front Door Diabetes Service was established in 2006 consisting of a Specialist Registrar, Diabetes Specialist Nurses, Dietician, and Podiatrist, supported by a Consultant Diabetologist. The aim of the service was to take a proactive approach and support those patients who require specialist intervention, such as those admitted with diabetes-related illness or where diabetes is comorbidity. Acute ward areas such as Emergency Department, Emergency Admission Unit, Coronary Care Unit and Medical High Dependence unit were visited everyday for 5 days a week (which subsequently increased to 6 days a week) to identify these patients with a view to improve their quality of care and formulate a structured discharged plan. The opportunity is utilised to empower patients to manage their own condition with confidence as well as educate staff caring for those patients in the wards. Subsequently, patients are either discharged back to primary care, or followed up in specialist diabetes clinics at the hospital setting until their control has stabilised.
Audit methods We performed a retrospective audit of all patients who were seen by our Front Door Diabetes Service between June 2007 and December 2010. The following data were collected: type of diabetes, reason for referral, age, sex and HbA1c at baseline and at 3–6 months after discharge. Appropriate referral
was defined according to the National standard from the ThinkGlucose Initiative, an initiative to improve the care for people with diabetes when they are admitted to hospital, set up by the NHS Institute for Innovation and Improvement, in the UK.
Inclusion criteria Following patients with diabetes were included
• • • •
Diabetes-related emergencies Poor control Acute coronary syndrome Patients with vomiting, diarrhoea or requiring nasogastric feed • Patients requiring intravenous insulin therapy or with poor oral intake • Sepsis related uncontrolled hyperglycaemia
Exclusion criteria
• •
Transient hyperglycaemia Simple educational needs, good control and good self management skills • Patients requiring routine diabetes care were excluded from the audit.
Statistical analysis Data was analysed using SPSS 21.0 software (IBM, Statistical Package for Social Sciences version 21.0; SPSS, Chicago, IL). Data presented as frequencies or mean standard deviation (SD). Paired sample t-test was performed to compare baseline and follow-up HbA1c. Patients admitted with hypoglycaemia, newly diagnosed type 1 DM, and newly diagnosed type 2 DM and previously known DM were analysed separately. A subgroup analysis based on age was also performed. A p-value < 0.05 was considered significant.
Results In total 2002 patient data were captured. Seven hundred and seventy-eight patients were eliminated initially because of loss of follow-up and/or because of planned discharge to the community diabetes team. Hence, full baseline and follow-up data were available for 1224 patients. Of which, 235 patients (19.2%) had a primary diagnosis of hypoglycaemia and were analysed separately. In the remaining 989 patients, 31 (3.1%) had new onset type 1 DM and 91(9.2%) had new onset type 2 DM which were also analysed separately. Thus, we were left with 867 patients with known DM and without hypoglycaemia.
Patients with known DM (n = 867)
The mean age of this group was 64.28 19.62 years. There was a significant improvement in HbA1c ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
The impact of diabetes outreach team on long-term glycaemic control
Table 1 Summary of HbA1c at baseline and follow-up by type of referral
Known DM (n = 867) Newly diagnosed type 1 DM (n = 31) Newly diagnosed Type 2 DM (n = 91) Hypoglycaemia (n = 235)
Baseline HbA1c % (mmol/mol)
Follow-up HbA1c% (mmol/mol)
9.0 2.39 (75) 12.55 2.27 (114) 10.70 3.04 (93) 7.48 1.59 (58)
8.46 7.43 7.29 7.59
2.02 2.05 1.74 1.57
p-value
< 0.001 < 0.001 < 0.001 0.2
(68) (57) (55) (59)
Data presented as mean SD.
Table 2 The changes in HbA1c between baseline and follow-up in different age groups
Age group
Known DM (n = 867)
Newly diagnosed type 1 DM (n = 31)
Newly diagnosed type 2 DM (n = 91)
1 2 3 1 2 3 1 2 3
(n (n (n (n (n (n (n (n (n
= = = = = = = = =
195) 382) 290) 25) 4) 2) 15) 49) 27)
Baseline HbA1c (%)mmol/mol
Follow-up HbA1c (%) mmol/mol
10.29 2.8 (89) 8.78 2.14 (72) 8.42 2.01 (69) 12.15 1.95 (109) 14.83 2.92 (139) 13.10 3.54 (120) 10.80 2.84 (95) 11.03 2.78 (97) 10.05 3.56 (86)
9.17 8.28 8.24 7.62 6.43 7.15 6.94 7.37 7.35
2.47 1.84 1.81 2.20 0.40 1.91 1.54 1.71 1.94
(77) (67) (67) (60) (46) (55) (52) (57) (57)
p-value
< 0.001 < 0.001 0.16 < 0.001 0.009 0.4 < 0.001 < 0.001 < 0.001
Data presented as mean SD. Age groups: Group 1: ≤ 50 years; Group 2: 51–74 years; Group 3: ≥ 75 years.
between baseline and follow-up (Table 1). The proportion of patients with HbA1c < 53 mmol/mol (7%) at baseline compared with follow-up improved significantly [18.6% (n = 161) vs. 24.5% (n = 212), p < 0.001]. The reduction in HbA1c was observed in patients with age < 75 years, and mostly in those ≤ 50 years (Table 2). Patients aged 75 or older had small non-significant improvement in HbA1c (Table 2).
Patients with newly diagnosed type 1 DM (n = 31)
The mean age of this group was 38.90 17.315 years. HbA1c improved significantly between baseline and follow-up (Table 1) regardless of the patient’s age group (Table 2).
Patients with newly diagnosed type 2 DM (n = 91)
The mean age of this group was 66.14 14.65 years. HbA1c improved significantly between baseline and follow-up (Table 1) regardless of the patient’s age group (Table 2).
Patients admitted with hypoglycaemia (n = 235)
The mean age for this group was 71.98 13.64 years. The mean baseline HbA1c was 58 mmol/mol (7.48% 1.59). The aim of treatment in this group was to avoid ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
hypoglycaemia hence the follow-up HbA1c for this group after intervention by our diabetes outreach team was marginally but not significantly higher 59 mmol/ mol (7.59% 1.57) (p = 0.2). Subgroup analysis revealed 66.8% of patients in this cohort were above 70 years of age. The HbA1c increased mildly and non-significantly between baseline and follow-up in those aged ≤ 70 years 58 mmol/ mol (7.50% 1.75) vs. 60 mmol/mol (7.64% 1.87) (p = 0.4) or > 70 years 58 mmol/mol (7.47% 1.51) vs. 59 mmol/mol (7.56% 1.40) (p = 0.4).
Discussion Our study showed that appropriate intervention and care provided by the front door diabetes team improves HbA1c outcomes in patients with known DM and those with newly diagnosed type 1 and type 2 DM. Managing hypoglycaemia and hyperglycaemia in the in-patient setting is a good opportunity for the hospital diabetologist to positively impact on the mortality and morbidity associated with diabetes. The landmark UKPDS study clearly demonstrated that a 1% reduction in HbA1c resulted in 21% decrement in any diabetes-related end-points (12). Despite this, and the clear recommendation by NSF standard 8, in-patient management of diabetes often
1335
1336
The impact of diabetes outreach team on long-term glycaemic control
remains suboptimal. Financial constraints and the cost of setting up a dedicated diabetes team would offset many practising physicians despite their willingness and enthusiasm. Traditionally, these patients are managed by intensive insulin therapy but such measures alone can pose an increased risk of hypoglycaemia as seen in the NICE-SUGAR study (13). An integrated and dedicated diabetes in-patient management team with a balanced approach had previously shown reduction in the length of stay (11,14), reduction in the avoidable admissions and diabetesrelated bed occupancy (9) thereby reducing the healthcare costs (14). Our simple cross sectional audit focused on those patients with diabetes admitted acutely, showed a significant reduction in HbA1c not only for treatment na€ıve patients but also for those who were known to have diabetes, who were previously established on hypoglycaemic agents, through a structured and patient-centred approach.
Education is the key Hospitalisation provides a unique opportunity to address patient’s educational needs. Simple but very relevant educational gaps such as taking the right medications, injection technique, ability to recognise and correct hypoglycaemia with the right food, appropriate blood glucose testing can be addressed during their stay. We used this opportunity to educate and empower patients as well as nursing staff caring for those patients. Experience from other centres had previously shown staff education resulted in reduction in drug administration error and reduced length of stay (11,14). Our own experience is that educating staff in the general ward about basic diabetes care results in improved quality indicators and patient satisfaction as reflected by our results in the national diabetes audit (2). However the key to improving long-term glycaemic control lies with patient empowerment. With the introduction of ‘ThinkGlucose’, an initiative to improve the care for people with diabetes when they are admitted to hospital in UK, we rolled out a teaching module to all front line clinical staff. Our job plans were revisited to provide a consultant led service and our nursing staff were made available 6 days a week. Once discharged from the hospital patients were either sent back to primary care, or followed up in our specialist diabetes clinics at the hospital setting until their control has stabilised. Patients who were bed bound, unwell because of comorbid state and therefore deemed inappropriate for secondary care follow-up, those who were from out of area and those who have good community support and preferred to be followed up in the
community were discharged to the community diabetes specialist team. For those patients who were followed up in the secondary care, we rationalised their diabetes medication, discussed long-term complications, hypoglycaemia management and provided them with contact numbers for advice and support. These patients were subsequently discharged back to the community depending on the complexity of their condition and improvement in their metabolic parameters including HbA1c. The results showed significant reduction in HbA1c not only for new onset diabetes but also for those who were established on treatment. Interestingly patients with Type 1 diabetes also showed reduction in long-term glycaemic control, reiterating the fact that structured education and support from health professionals is an important factor in managing their diabetes. The greatest reductions in HbA1c were seen in those patients with the poorest control, but significant improvements were seen in all groups. The biggest drop however, was seen in those who were ≤ 50 years of age however it is this group who also had the highest HbA1c on presentation. We did not see a significant drop in HbA1c in patients who were above 75 years of age. This could possibly be explained by the conservative approach taken by our Front door diabetes team to avoid hypoglycaemia as the patients in this age group are particularly at high risk of developing hypoglycaemia. Furthermore the baseline HbA1c in this cohort is something that one would consider ‘reasonable’ (Table 2) for their age. We excluded 235 patients referred with hypoglycaemia from our analysis as the aim of the treatment in this group would be to relax the glycaemic control and allow them to have a higher HbA1c to retain hypoglycaemia awareness. Lowering the HbA1c in elderly patients may sometimes be detrimental. Poor nutritional status, associated comorbidities such as dementia, cerebrovascular accident, poor cognitive status, worsening renal and liver functions and polypharmacy may predispose them to increased risk of hypoglycaemia (15). Management should therefore be tailor-made taking the above factors into consideration as well as their ability to recognise and treat hypoglycaemia, American Diabetes Association provides clear management guidelines on this aspect to the practising physicians (16). A significant finding noted in our audit was that patients over 70 years of age account for more than 60% in this cohort. With treatment relaxation, the HbA1c in this group slightly but non-significantly worsened. One of the limitations of this real life audit is the change in staffing level and service delivery that happens with time. In 2007 when the service was introduced, the front door team was run on a small ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
The impact of diabetes outreach team on long-term glycaemic control
scale and our nursing staff covered mainly the acute areas such as emergency admission unit and coronary care unit. Capturing patients in a robust scale in the initial years were limited by time constraints and poor staffing level, however with service expansion in the subsequent years the staffing level improved allowing more time for inpatient diabetes care. Unlike in research setting, day to day clinical practice encounters such practical difficulties, which
References 1 Morgan CL, Peters JR, Dixon S, Currie CJ. Estimated costs of acute hospital care for people with diabetes in the United Kingdom: a routine record linkage study in a large region. Diabet Med 2010; 27(9): 1066–73. 2 National diabetes inpatient audit. Health & Social Care information centre. http://www.hscic.gov.uk/ diabetesinpatientaudit. 3 Bottomley JM. Cases, complications, comorbidity and costs of diabetes in UK: is treatment falling wide of the mark? Br J Diab Vasc Dis 2010; 10(5): 209–15. 4 Currie CJ, Gale EAM, Poole CD. Estimation of primary care treatment costs and treatment efficacy for people with type 1 and type 2 diabetes in the United Kingdom from 1997 to 2007. Diabet Med 2010; 27(8): 938–48. 5 UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998; 352: 854–65. 6 Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes
ª 2014 John Wiley & Sons Ltd Int J Clin Pract, November 2014, 68, 11, 1333–1337
7
8
9 10
11
12
1337
makes our findings even more realistic. A computerbased referral as well as record of review system may help to resolve such issues. Some of the Trusts across UK already have such system in place. In conclusion, by providing a comprehensive care, education and appropriate intervention through our front door diabetes team, we have shown a significant reduction in the objective marker of long-term glycaemic control for recently hospitalised patients.
on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977–86. Baker ST, Chiang CY, Zajac JD, Bach LA, Jerums G, MacIsaac RJ. Outcomes for general medical inpatients with diabetes mellitus and new hyperglycaemia. Med J Aust 2008; 188(6): 340–3. Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG. Management of diabetes and hyperglycemia in hospitals. American Diabetes Association Diabetes in Hospitals Writing Committee. Diabetes Care 2004; 27(2): 553–9. Diabetes UK. Diabetes in the UK 2011-12: key statistics on diabetes. Mahto R, Venugopal H, Vibhuti VS et al. The effectiveness of a hospital diabetes outreach service in supporting care for acutely admitted patients with diabetes. QJM 2009; 102(3): 203–7. Flanagan D, Moore E, Baker S, Wright D, Lynch P. Diabetes care in hospital–the impact of a dedicated inpatient care team. Diabet Med 2008; 25(2): 147– 51. Stratton IM, Adler AI, Neil HA et al. Association of glycaemia with macrovascular and microvascular
13
14
15
16
complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 2000; 321(7258): 405–12. Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360(13): 1283–97. Korytkowski M, Dinardo M, Donihi AC, Bigi L, Devita M. Evolution of a diabetes inpatient safety committee. Endocr Pract 2006; 12(Suppl. 3): 91–9. Abrass IB, Schwartz RS (1998) Special presentation of endocrine disease in the elderly. In Fitzgerald FT, ed. Current Practice of Medicine, vol, 1, no. 2. Philadelphia, PA: Current Medicine, Inc., 1998, pp. 415–8. Kirkman MS, Briscoe VJ, Clark N et al. Diabetes in older adults. Diabetes Care 2012; 35(12): 2650–64.
Paper received May 2013, accepted April 2014
