Article

Introduction of New South Wales adult subcutaneous insulin prescribing chart in a tertiary hospital: its impact on in‐patient glycaemic control

Authors:
If you want to read the PDF, try requesting it from the authors.

Abstract

Aims: Erratic blood glucose levels (BGL) are commonly observed amongst patients with diabetes mellitus during hospital admission. Patients on insulin therapy often do not have their doses titrated adequately by their team doctors during admission, and insulin is well-known to be a high risk medication prone to administration error. The aim of this study is to assess the impact of a state-wide Adult Subcutaneous Insulin Prescribing Chart (ASCIPC) on glycaemic control and insulin prescribing pattern in a tertiary hospital. Methods: An audit on the clinical records of in-patients who were on subcutaneous insulin therapy in the first week of July 2014 (prior to ASCIPC, n = 56) and in the first week of July 2015 (10 months after introducing ASCIPC, n = 62) was conducted at Liverpool Hospital. Results: Following introduction of ASCIPC, fewer BGL readings were missed (9.1 vs 11.6%, p = 0.032) and medical officers were more likely to adjust insulin dosage (71.0 vs 42.6%, p = 0.002) when compared to baseline. Glycaemic control improved, with lower mean BGL (9.4 ± 2.0 vs 10.4 ± 2.6 mmol/L, p = 0.021) and greater proportion of BGLs within the normal range of 5-10 mmol/L (56.2 vs 47.7%, p = 0.041). Omission of insulin doses after ASCIPC remained common, with over 40% of patients having at least one dose of insulin omitted during the audit week. Conclusion: Our study showed that introduction of ASCIPC had positive impacts on glycaemic management for patients on subcutaneous insulin therapy during admission. More work is required to reduce the rate of insulin omission and to further improve glycaemic control for in-patients.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... It is well documented that medication errors are the leading case of patient harm in the hospital setting (Roughhead, Semple & Rosenfeld., 2013), with insulin being associated with more errors than any other class of medicines (Wong et al., 2016). The Australian ...
... It is not uncommon for insulin-requiring patients to experience erratic glycaemic control during their admission, depending on metabolic effects of acute illness, pain, infection or invasive procedures (Wong et al., 2016). When appropriate glycaemic management is neglected, patients are at higher risk of poor clinical outcomes, extending hospital length of stay (Taylor et al., 2018). ...
... As discussed in Chapter 2 of this paper, insulin is one of the most common drugs involved in medication errors, with many different brands, sound-alike names and more recently the use of high-dose insulins (Thornton, 2015). Consistent with findings in other study's (Cornish., 2014;Kolanczyk & Dobersztyn., 2016;Wong et al., 2016), examples of incorrect prescribing in the present study included; duplication of rapid and short acting sliding scale insulin (SSI) orders (NovoRapid and Actrapaid, NovoRapid and Humalog active insulin SSI orders), rapid acting insulin being charted instead of a pre-mixed insulin order or vice versa (Humalog instead of Humalog Mix 25, NovoMix30 added to Lantus in a basal bolus insulin regimen instead of NovoRapid), and no basal insulin ordered for patients with type 1 diabetes. ...
... Like paper-based bedside observation charts (Chatterjee et al., 2005;Preece et al., 2013), insulin forms have typically been developed by clinicians for use in their own hospital or regional health service, resulting in substantial design differences between institutions (Christofidis et al., 2012). Nevertheless, only a handful of publications describe the clinician-led development and implementation of a subcutaneous insulin chart (Cheung et al., 2011;Lehnbom et al., 2009;McIver et al., 2009;Rushmer and Voigt, 2008;Wong et al., 2016). In each of these studies, a chart was developed and introduced at one or more clinical sites, with data collected before and after implementation (without the use of a control site). ...
... Collectively, this work has yielded mixed results in terms of post-implementation improvements. For example, across different studies, reported improvements have included: clearer documentation of insulin orders (McIver et al., 2009); improved documentation of insulin administration (McIver et al., 2009;Rushmer and Voight, 2008); increased BGL testing (Cheung et al., 2011;Wong et al., 2016); more BGLs within the normal range (Cheung et al., 2011;Wong et al., 2016); fewer hypoglycaemic events (Cheung et al., 2011); fewer hyperglycaemic events (Wong et al., 2016); and improved compliance with hypoglycaemia protocols (Lenbohm et al., 2009;Rushmer and Voigt, 2008). When interpreting these results, it should also be noted that, in all cases, it is unclear to what extent the documented improvements reflected the design of the insulin chart itself versus the impact of the staff training and other changes that accompanied (or happened to coincide with) its implementation. ...
... Collectively, this work has yielded mixed results in terms of post-implementation improvements. For example, across different studies, reported improvements have included: clearer documentation of insulin orders (McIver et al., 2009); improved documentation of insulin administration (McIver et al., 2009;Rushmer and Voight, 2008); increased BGL testing (Cheung et al., 2011;Wong et al., 2016); more BGLs within the normal range (Cheung et al., 2011;Wong et al., 2016); fewer hypoglycaemic events (Cheung et al., 2011); fewer hyperglycaemic events (Wong et al., 2016); and improved compliance with hypoglycaemia protocols (Lenbohm et al., 2009;Rushmer and Voigt, 2008). When interpreting these results, it should also be noted that, in all cases, it is unclear to what extent the documented improvements reflected the design of the insulin chart itself versus the impact of the staff training and other changes that accompanied (or happened to coincide with) its implementation. ...
Article
Full-text available
Insulin is a high-risk medicine that has been implicated in serious adverse events for hospital inpatients, including medication-error related deaths. Most insulin errors occur during administration, and “wrong dose” is the most common type. A paper-based subcutaneous insulin chart (the “NSIC”) was developed for the Australian Commission on Safety and Quality in Health Care, using a range of human factors methods, with the aim of reducing the opportunity for errors. The present lab-based study empirically assessed whether the NSIC's human factors design translates into improved user-performance in the determination of insulin doses, compared with a pre-existing chart. Forty-one experienced nurses and 48 novice chart-users completed 60 experimental trials (30 per chart), in which they determined doses to administer to patients. Both groups determined insulin doses faster, and made fewer dose errors, when using the NSIC. These results support the utility of the usability heuristics employed in developing the chart.
... Only two studies looked at economic or financial sustainability/ impact of interventions [25,33] and there was a general lack of consideration of the theoretical basis or justification for interventions. Well-reported studies included details of usual care in the control/before group, as well as the organizational characteristics and readiness for change [15,29,33,37,46,57]. ...
... The majority of the studies were conducted in the USA (n=17). Three studies were conducted in Canada [36,37,54], two in Australia [35,57] and two in Spain [41,55]. ...
... ward) level or single institution level. Three studies described large-scale, resource-heavy initiatives involving more than one hospital site [37,39,57]. Most interventions involved the introduction of insulin order sets using Computerized physician order entry or a dedicated insulin order form, often including decision support tools such as guidelines or dosing algorithms. ...
Article
im To conduct a systematic review of literature to identify interventions that are effective in improving insulin prescribing for people with diabetes in the hospital setting. Methods Computerized bibliographic databases were searched for studies published in English that described the effectiveness of interventions to improve insulin prescribing within the hospital setting. Studies were eligible for inclusion if they reported data that compared insulin prescribing practice after an intervention or compared with a control group. Studies were not excluded on the basis of publication date, geographical location or risk of bias assessment. Results We identified 35 studies for inclusion in the review, including two cluster randomized controlled trials, two cohort studies, and 31 uncontrolled before–after studies. Studies reported a variety of interventions that aimed to increase insulin prescribing accuracy or completeness or decrease the use of discouraged subcutaneous sliding scale insulin regimens. Differences in definition of insulin prescribing error, terminology and common practice based on geographical location was evident, and quality issues with respect to study design and reporting somewhat limited the interpretation of conclusions. Conclusions Implementing strategies that are sensitive to local context and designed to increase adherence to insulin prescribing guidelines are associated with a reduction in prescribing errors. Future implementation should build on effective approaches including multifaceted interventions involving multiple stakeholders at various institutional levels. Future studies in insulin prescribing errors would benefit from the use of standardized approaches, terminology and outcome measures to enable greater comparison.
... A Canadian tertiary-care teaching hospital found that the implementation of a standardised, preprinted insulin order set facilitated best practices for insulin therapy, improved patient safety and was highly supported by treating practitioners 17 . A further study in New South Wales found that introduction of an adult subcutaneous insulin-prescribing chart had positive impacts on glycaemic management for patients on subcutaneous insulin therapy during admission 18 . ...
Article
Full-text available
Aim Our aim was to design a new insulin prescribing tool in compliance with the Irish Medicines Safety Network recommendations. Methods In 2015, we undertook a review of the existing paediatric subcutaneous insulin-prescribing sheet introduced to Cork University Hospital in 2013. This involved a retrospective analysis of 15 consecutive in-patient insulin prescribing charts and a questionnaire distributed to health professionals. Following this a new insulin prescribing chart was designed and implemented in 2016 and a re-audit was performed in 2017. Results The 2017 re-audit demonstrated that the new insulin chart was viewed as easier (95% of previous users n=18) and safer (n=16) to use. There was less confusion (2017: 28%, n=11/39 vs 2015: 50%, n=17/34 2015) and the ALERT system helped staff standardise hypo/hyperglycaemia management (71%, n=28). Conclusion The new paediatric insulin prescribing chart has improved safety and ease of prescribing insulin. The colour coded quasi graph and ALERT system has made it easier to appreciate capillary blood glucose trends and manage them safely.
... 10 11 In Australia, a tertiary hospital saw better glycaemic control for inpatients with the introduction of a new insulin prescribing chart. 11 With VRIII it is important to provide an adequate substrate with the intravenous insulin to avoid hypoglycaemia as well as maintain a stable blood glucose through varying the infusion rate rather than the type of substrate. 3 Intravenous fluids act as this glucose substrate and are important to maintain fluid and electrolyte balance as glucose with insulin leads to hypokalaemia; so, additional potassium must be prescribed unless contraindicated. ...
Article
Full-text available
Variable rate intravenous insulin infusions (VRIII) are used to maintain stable blood glucose in hospitalised patients with diabetes who are unable to eat or have a severe illness where good glycaemic control is paramount. With VRIII it is important to prescribe an adequate substrate to avoid hypoglycaemia and maintain electrolyte balance. Traditionally the substrate would have been varied to achieve this; current guidelines advise varying the infusion rate rather than the type of substrate. The local hospital Trust updated their VRIII prescription chart to reflect the Joint British Diabetes Societies’ suggestions for inpatient care in October 2014. A local audit in January 2015 highlighted that 48% of patients on VRIII were prescribed the correct fluid as per the guideline. A questionnaire to assess prescriber knowledge regarding VRIII showed 40.4% of prescribers selected appropriate fluid for a patient with normal renal function and 11.5% of prescribers selected appropriate fluid for a patient with renal failure. An educational podcast was devised to explain the rationale behind appropriate fluid prescription with VRIII; this was shown to prescribers. Following the podcast, 75.8% of prescribers selected appropriate fluids for normal renal function and 54.5% for renal failure. Questionnaires were completed to assess prescriber knowledge prepodcast and postpodcast. Following the podcast, there was a significant increase in questionnaire scores, indicating improved prescriber knowledge surrounding VRIII. A reaudit of prescriptions for VRIII showed improvement in practice, where 63% of patients on VRIII were prescribed correct fluids. The use of a simple audiovisual podcast on VRIII led to a significant improvement in prescriber knowledge. Podcasts are an ideal medium to raise awareness around safety issues, including safe prescription of insulin. Further work will include the follow-up of participants to evaluate sustained knowledge and improvements of prescriptions in practice, with the overall aim of improving patient safety.
Article
Full-text available
Diabetes is common in hospitalised patients and insulin is frequently required for management. Insulin is a high-risk drug, accounting for about 15% of reported medication-related incidents. Despite its complexity, insulin management in hospitals is often undertaken by junior and non-specialist staff. Improving insulin management requires addressing safe prescribing and administration as well as quality use of insulin. Common errors in insulin use are well documented and can be addressed through form design and enhancing decision support. We undertook to improve insulin management using a locally proven improvement methodology. New forms were developed for intravenous and subcutaneous insulin and blood glucose management. Audited pilot studies in four hospitals confirmed improved insulin management without adversely impacting on overall diabetes management as assessed using Glucometrics. Subsequently, the forms have been introduced to 70% of Queensland public hospitals with roll-out to remaining hospitals continuing. Large-scale standardisation of insulin management is feasible.
Article
Background/objectives: Hyperglycaemia is commonly observed in patients with diabetes mellitus (DM) while receiving enteral nutrition (EN) in hospital, and hyperglycaemia has been shown to be associated with poor clinical outcomes. The aim of this study was to assess the glycaemic status of patients with DM who received EN during hospital admission and evaluate the impact of intervention by a specialist diabetes team (SDT) on glycaemic control and clinical outcomes of these patients. Subjects/methods: A retrospective review of patients with DM who required EN during hospital admission was conducted. We compared patient characteristics, glycaemic profile and clinical outcomes between patients who were managed by SDT and those who were managed by the admitting team. Results: Seventy-four patients with DM on EN were included in this study, of whom 27 were managed by SDT while on EN. Compared with patients managed by the admitting team, those who were reviewed by SDT had better glycaemic control during the period of EN as well as during the 24 h after EN was ceased. These patients also had shorter length-of-stay in hospital and lower in-patient mortality. Conclusions: Our findings confirmed that there was a role for SDT in managing patients with DM who received EN during their hospital admission. These patients had improved glycaemic control while receiving EN and had better clinical outcomes. Further prospective studies will be required to validate the findings of this study.
Article
To determined the relationship between perioperative glucose control and postoperative nosocomial infection rate is 100 consecutive diabetic patients undergoing elective surgery. One hundred initially uninfected diabetic patients undergoing elective surgery were prospectively monitored for perioperative glucose control and postoperative nosocomial infection rate. Glucose control was determined by the attending surgeon or diabetologist. A large tertiary care hospital that serves as the in-patient facility for a local diabetes center. All patients were screened for infection preoperatively. Only initially uninfected patients were enrolled, and all patients received perioperative antibiotic coverage. Perioperative glucose control and postoperative nosocomial infection rate were monitored prospectively. APACHE II scores were determined on all patients. Patients were stratified into two groups: those with relatively "good" perioperative glucose control (all values < or = 220 mg/dL) and those with "poor" control (at least one value > 220 mg/dL). Contingency tables were generated, comparing nosocomial infection rates vs perioperative glucose control. Correlation coefficients between APACHE II score and maximum and mean glucose values were also determined. A serum glucose > 220 mg/dL on postoperative day one (POD 1) was a sensitive (87.5%) but relatively nonspecific (33.3%) predictor of the later development of postoperative nosocomial infection. In patients with hyperglycemia (> 220 mg/dL) on POD 1, the infection rate was 2.7 times that observed (31.3% vs 11.5%) in diabetic patients with all serum glucose values < 220 mg/dL. When minor infection of the urinary tract was excluded, the relative risk for "serious" postoperative infection increased to 5.7 when any POD 1 blood glucose level was > 220 mg/dL. On the basis of correlation coefficients between serum glucose values and APACHE II score, only 18% of the variance in the highest serum glucose could be explained by disease severity alone. We conclude that diabetic patients undergoing major cardiovascular or abdominal surgery have an increased risk of infection that is further exacerbated by early postoperative hyperglycemia. The high rate of nosocomial infection observed in diabetic patients with poor glucose control suggests that hyperglycemia itself may be an independent risk factor for the development of infection. Efforts to improve perioperative glucose homeostasis in diabetic patients may reduce the incidence of nosocomial infection and thereby improve outcome.
Article
A dedicated subcutaneous insulin prescription chart incorporating glucose monitoring results, forced functions, and management guidelines was introduced to facilitate better hospital diabetes control. Point of care capillary blood glucose monitoring charts for 99 people with diabetes from the period before the introduction of the new chart, and 106 after its introduction were reviewed. A total of 12,649 blood glucose levels (BGLs) were collected for glucometric analysis. Following the introduction of the chart, there was an increase in the number of BGLs performed daily from 4.5 ± 1.2 to 4.9 ± 1.3 (p = 0.05). There was an increase in the proportion of BGLs within the ideal range of 4-9.9 mmol/L (51.8% vs. 54.1%, p = 0.01). There was a reduction in hypoglycaemic events (proportion of BGLs <4 mmol/L in the whole population decreased from 5.2% to 3.4% (p < 0.001), proportion of BGLs <4 mmol/L for each patient decreased from 5.6 ± 9.2% to 2.9 ± 5.4% (p = 0.01), proportion of days where patient had a BGL <4 mmol/L decreased from 17.6 ± 22.6% to 11.4 ± 18.8% (p = 0.03)), despite an increase in the use of supplemental insulin (14.2 ± 35.7 vs. 29.4 ± 51.4 u nits/patient, p = 0.02). We conclude that the use of a dedicated hospital subcutaneous insulin prescription chart can reduce hypoglycaemia and improve some measures of glycaemic control.
Article
Despite a lack of evidence that sliding scale insulin has any clinical benefit, and some evidence that it may even be detrimental, sliding scale insulin is still commonly prescribed in hospitals today. Adopting a proactive rather than a reactive approach to managing diabetes by the use of 'supplemental insulin', given in conjunction with either considered adjustments to the patient's regular anti-diabetic therapy or the provision of basal insulin, is a more effective and safer means of improving glycaemic control in hospital. There are now randomized trial data to support this approach. These data, together with the recognition that there is no evidence base for the use of sliding scale insulin, coupled with changes to insulin prescribing charts in Australia, should lead to the demise of sliding scale insulin use in hospital.
Article
To identify predictors of hypoglycemic and hyperglycemic episodes in hospitalized patients with diabetes with special attention to the effectiveness of sliding scale insulin regimens. Prospective cohort study. Urban university hospital. One hundred seventy-one adults with diabetes mellitus as a comorbid condition admitted consecutively to medical inpatient services during a 7-week period. Demographic, clinical, and laboratory data from inpatient medical records. Rates of hypoglycemic (capillary blood glucose, < or = 3.3 mmol/L [< or = 60 mg/dL]) and hyperglycemic (capillary blood glucose, > or = 16.5 mmol/L [> or = 300 mg/ dL]) episodes. Of the patients, 23% experienced hypoglycemic episodes, and 40% experienced hyperglycemic episodes. The overall rates of hypoglycemic and hyperglycemic episodes were 3.4 and 9.8 per 100 capillary blood glucose measurements, respectively. Independent predictors of hypoglycemic episodes included African American race (relative risk [RR], 2.13) and low serum albumin level (RR, 1.92 per 100-g/L decrease); corticosteroid use was associated with a reduced risk of hypoglycemic episodes (RR, 0.32; P < .05). Independent predictors of hyperglycemic episodes included female gender (RR, 1.67), severity of illness (RR, 1.22 per 10 Acute Physiology and Chronic Health Evaluation III units), severe diabetic complications (RR, 2.32), high admission glucose level (RR, 1.33 per 5.5 mmol/L), admission for infectious disease (RR, 2.14), and corticosteroid use (RR, 3.74; P < .05). Of 171 patients, 130 (76%) were placed on a sliding scale insulin regimen. When used alone, sliding scale insulin regimens were associated with a 3-fold higher risk of hyperglycemic episodes compared with individuals following no pharmacologic regimen (RRs, 2.85 and 3.25, respectively; P < .05). Suboptimal glycemic control is common in medical inpatients with diabetes mellitus. The risk of suboptimal control is associated with selected demographic and clinical characteristics, which can be ascertained at hospital admission. Although sliding scale insulin regimens are prescribed for the majority of inpatients with diabetes, they appear to provide no benefit; in fact, when used without a standing dose of intermediate-acting insulin, they are associated with an increased rate of hyperglycemic episodes.
Article
High blood glucose concentration may increase risk of death and poor outcome after acute myocardial infarction. We did a systematic review and meta-analysis to assess the risk of in-hospital mortality or congestive heart failure after myocardial infarction in patients with and without diabetes who had stress hyperglycaemia on admission. We did two searches of MEDLINE for English-language articles published from 1966 to October, 1998, a computerised search of Science Citation Index from 1980 to September, 1998, and manual searches of bibliographies. Two searchers identified all cohort studies or clinical trials reporting in-hospital mortality or rates of congestive heart failure after myocardial infarction in relation to glucose concentration on admission. We compared the relative risks of in-hospital mortality and congestive heart failure in hyperglycaemic and normoglycaemic patients with and without diabetes. 14 articles describing 15 studies were identified. Patients without diabetes who had glucose concentrations more than or equal to range 6.1-8.0 mmol/L had a 3.9-fold (95% CI 2.9-5.4) higher risk of death than patients without diabetes who had lower glucose concentrations. Glucose concentrations higher than values in the range of 8.0-10.0 mmol/L on admission were associated with increased risk of congestive heart failure or cardiogenic shock in patients without diabetes. In patients with diabetes who had glucose concentrations more than or equal to range 10.0-11.0 mmol/L the risk of death was moderately increased (relative risk 1.7 [1.2-2.4]). Stress hyperglycaemia with myocardial infarction is associated with an increased risk of in-hospital mortality in patients with and without diabetes; the risk of congestive heart failure or cardiogenic shock is also increased in patients without diabetes.
Article
"Stress" hyperglycemia may be associated with increased mortality and poor recovery in diabetic and nondiabetic patients after stroke. A systematic review and meta-analysis of the literature relating acute poststroke glucose levels to the subsequent course were done to summarize and quantify this relationship. A comprehensive literature search was done for cohort studies reporting mortality and/or functional recovery after stroke in relation to admission glucose level. Relative risks in hyperglycemic compared with normoglycemic patients with and without diabetes were calculated and meta-analyzed when possible. Thirty-two studies were identified; relative risks for prespecified outcomes were reported or could be calculated in 26 studies. After stroke of either subtype (ischemic or hemorrhagic), the unadjusted relative risk of in-hospital or 30-day mortality associated with admission glucose level >6 to 8 mmol/L (108 to 144 mg/dL) was 3.07 (95% CI, 2.50 to 3.79) in nondiabetic patients and 1.30 (95% CI, 0.49 to 3.43) in diabetic patients. After ischemic stroke, admission glucose level >6.1 to 7.0 mmol/L (110 to 126 mg/dL) was associated with increased risk of in-hospital or 30-day mortality in nondiabetic patients only (relative risk=3.28; 95% CI, 2.32 to 4.64). After hemorrhagic stroke, admission hyperglycemia was not associated with higher mortality in either diabetic or nondiabetic patients. Nondiabetic stroke survivors whose admission glucose level was >6.7 to 8 mmol/L (121 to 144 mg/dL) also had a greater risk of poor functional recovery (relative risk=1.41; 95% CI, 1.16 to 1.73). Acute hyperglycemia predicts increased risk of in-hospital mortality after ischemic stroke in nondiabetic patients and increased risk of poor functional recovery in nondiabetic stroke survivors.
Article
Diabetes mellitus is a risk factor for death after coronary artery bypass grafting. Its relative risk may be related to the level of perioperative hyperglycemia. We hypothesized that strict glucose control with a continuous insulin infusion in the perioperative period would reduce hospital mortality. All patients with diabetes undergoing coronary artery bypass grafting (n = 3554) were treated aggressively with either subcutaneous insulin (1987-1991) or with continuous insulin infusion (1992-2001) for hyperglycemia. Predicted and observed hospital mortalities were compared with both internal and external (Society of Thoracic Surgeons 1996) multivariable risk models. Observed mortality with continuous insulin infusion (2.5%, n = 65/2612) was significantly lower than with subcutaneous insulin (5.3%, n = 50/942, P <.0001). Likewise, glucose control was significantly better with continuous insulin infusion (177 +/- 30 mg/dL vs 213 +/- 41 mg/dL, P <.0001). For internal comparison, multivariable analysis showed that continuous insulin infusion was independently protective against death (odds ratio 0.43, P =.001). Conversely, cardiogenic shock, renal failure, reoperation, nonelective operative status, older age, concomitant peripheral or cerebral vascular disease, decreasing ejection fraction, unstable angina, and history of atrial fibrillation increased the risk of death. For external comparison, observed mortality with continuous insulin infusion was significantly less than that predicted by the model (observed/expected ratio 0.63, P <.001). Multivariable analysis revealed that continuous insulin infusion added an independently protective effect against death (odds ratio 0.50, P =.005) to the constellation of risk factors in the Society of Thoracic Surgeons risk model. Continuous insulin infusion eliminates the incremental increase in in-hospital mortality after coronary artery bypass grafting associated with diabetes. The protective effect of continuous insulin infusion may stem from the effective metabolic use of excess glucose to favorably alter pathways of myocardial adenosine triphosphate production. Continuous insulin infusion should become the standard of care for glycometabolic control in patients with diabetes undergoing coronary artery bypass grafting.
Guidelines for routine glucose control in hospital https
  • D Chipps
  • S Cornelius
  • B Depczynski
  • K O ' Heels
  • D Neal
  • J Wong
  • S Zoungas
  • Nw Cheung
Guidelines for routine glucose control in hospital. Sydney: Australian Diabetes Society
  • D Chipps
  • S Cornelius
  • B Depczynski
  • K Heels
  • O' Neal
  • D Wong
  • J Zoungas
  • S Cheung