Trial Review

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12620001303932
Ethics application status
Approved
Date submitted
3/09/2020
Date registered
2/12/2020
Date last updated
21/04/2022
Date data sharing statement initially provided
2/12/2020
Type of registration
Prospectively registered

Titles & IDs
Public title
Specialist Treatment of Inpatients: Caring for Diabetes (STOIC-D) - A single centre, randomised, controlled trial assessing the impact of a proactive specialist inpatient diabetes service on outcomes in inpatients with diabetes or hyperglycaemia.
Scientific title
Specialist Treatment of Inpatients: Caring for Diabetes (STOIC-D) - A single centre, randomised, controlled trial assessing the impact of a proactive specialist inpatient diabetes service on outcomes including glucometrics, adverse events, and processes of care, in inpatients with diabetes or hyperglycaemia.
Secondary ID [1] 302232 0
None
Universal Trial Number (UTN)
U1111-1257-8929
Trial acronym
STOIC-D
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Diabetes mellitus 318932 0
Hyperglycaemia 318933 0
Condition category
Condition code
Metabolic and Endocrine 316908 316908 0 0
Diabetes
Public Health 317426 317426 0 0
Health service research

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
The complex process-of-care intervention is described in detail:

Trial entry
All patients who are admitted to the Royal Melbourne Hospital (RMH) during the trial period will be assessed for eligibility for the trial. If they meet the inclusion criteria they will be randomised at the time of admission to the RMH to either the intervention or the control arm of the trial. Randomisation will occur automatically and electronically through the Electronic Medical Record (EMR) software.

Proactive specialist care (intervention arm)

First electronic review + Home team management
Within 24 hours of admission to hospital for business day admissions (Monday to Friday, excluding public holidays) and within 48 hours of admission to hospital for non-business day admissions (Saturday, Sunday, and public holidays) after entering the trial, a participant’s EMR record for the admission will be reviewed electronically and remotely by a member of the Inpatient Diabetes Service (IDS). A standardised Electronic Diabetes Review (EDR) document will be completed and entered into the clinical record. The participant’s home team junior doctor will also be alerted of this review directly through the RMH’s internal communications systems. No direct changes to the participant’s inpatient management will be made during the first electronic review. The participant’s home team will manage their diabetes and glucose as per their clinical judgement.

Ongoing electronic review + Home team management
Following the first electronic review a participant will remain on the remote review list. Their glucose and medication record will be reviewed at least every second business day for the duration of their participation in the trial. If a change in glycaemic management is indicated, the IDS team member will complete a follow-up EDR document enter this into the clinical record. If this occurs the participant’s home team junior doctor will also be alerted of this review directly. No direct changes to the participant’s inpatient management will be made during ongoing electronic reviews. The participant’s home team will manage their diabetes and glucose as per their clinical judgement.
Indications for a change in management include glucose outside of the target range for that patient or a change in clinical circumstances that are likely to cause glucose to move outside of the target range e.g. glucocorticoid commencement, patient fasting for surgery, and thus a proactive change in glycaemic management is required. The specific glycaemic range targeted for most individuals will be between 5.0 – 10.0 mmol/L as recommended by national and international inpatient diabetes guidelines but may differ for some individuals based on the clinical context.

Escalation to Bedside Diabetes Review (BDR)
If any of the following criteria are met the participant will be escalated by an IDS team member for a Bedside Diabetes Review (BDR) by the IDS using a direct review EMR list:
• Request from the home team that the participant be reviewed directly by the IDS.
• A diabetes diagnosis type that is anything other than type 2 diabetes (T2DM) e.g. type 1 diabetes (T1DM), type 3c diabetes (T3cDM), Latent Autoimmune Diabetes in Adults (LADA).
• Commencement of a significant dose of oral glucocorticoids i.e. equal to or greater than prednisolone 5mg total daily dose or equivalent.
• If on oral glucocorticoids at the time of admission, an increase in dose above the regular outpatient dose.
• Commencement of supplemental or replacement therapeutic nutrition i.e. parenteral nutrition or enteral feeds administered via an enteral access tube e.g. nasogastric tube (NGT), percutaneous endoscopic gastrostomy (PEG) tube. This does not include supplemental nutrition that a participant ingests orally e.g. Fortisip® nutritional supplement, regardless of whether this is prescribed by a dietitian.
• A major hypoglycaemic event i.e. BGL less than or equal to 3.0 mmol/L.
• A major hyperglycaemic event i.e. BGL greater than or equal to 20.0 mmol/L.
• Persistent hyperglycaemia i.e. 2 days on which there is any BGL greater than or equal to 15.0 mmol/L or 3 days on which there is any BGL greater than or equal to 12.0 mmol/L.
• HbA1c greater than or equal to 8.0%.

Bedside Diabetes Review (BDR)
A BDR involves a member of the IDS performing an inpatient consultation at the participant’s bedside. This may involve modification of the participant’s inpatient management, communication with their home team, and communication with the treating nurse unit. The participant will be seen directly as frequently and as long as they continue to derive benefit from this intensity of management. This will be determined clinically by the IDS according to each individual’s clinical context based on the criteria described below.

De-escalation to ongoing electronic review
At any point following the initial direct review an IDS team member may determine that the participant’s diabetes and glucose management are stable and that they no longer require ongoing bedside review by the IDS. They will then de-escalate the participant to electronic review, which shall occur at least every second business day. Criteria for this de-escalation include:
• No major hypoglycaemic events i.e. BGL less than or equal to 3.0 mmol/L and fewer episodes of hypoglycaemia.
• No major hypoglycaemic events i.e. BGL greater than or equal to 20.0 mmol/L and fewer episodes of hyperglycaemia.
• A greater proportion of glucose levels within the patient’s individualised target range.

If following this the participant again meets criteria for escalation to direct IDS review, this re-escalation to bedside review will occur at that time.

Duration of intervention administration
The trial will continue for 12 months such that the intervention will be administered for a total of 12 months.

Inpatient Diabetes Service (IDS) members
The IDS will be comprised of a endocrinology consultant, endocrinology registrar, endocrinology resident, diabetes nurse practitioner, dietitian, and on occasion a credentialled diabetes educator. While all members of the IDS will be able to collect information and consult upon patients, only the endocrinology consultant, endocrinology registrar, endocrinology resident, and diabetes nurse practitioner will be able to change medication prescriptions as per standard scopes of practice.
Intervention code [1] 318522 0
Treatment: Other
Comparator / control treatment
Standard care is described below for the purposes of this study.

Standard care (control arm)
None of the following procedures described pertaining to standard care represent a departure from the management of diabetes in hospital inpatients as currently occurs at the Royal Melbourne Hospital (RMH).

Home team management
The participant’s home team will manage their diabetes and glucose as per their clinical judgement.

Escalation to IDS
If the home team believes they need additional advice or support in managing a participant’s diabetes or glucose they can refer the participant to the Inpatient Diabetes Service (IDS) for review.

Bedside Diabetes Review (BDR)
A BDR involves a member of the IDS taking a referral from the home team and providing immediate phone advice. Depending on the clinical context the IDS will then perform an inpatient consultation at the participant’s bedside at an appropriate time. This may involve modification of the participant’s inpatient management, communication with their home team, and communication with the treating nurse unit. The participant will be seen directly as frequently and as long as they continue to derive benefit from this intensity of management. This will be determined clinically by the IDS according to each individual’s clinical context.

De-escalation to home team management
At any point following initial IDS review an IDS team member may determine that the participant’s diabetes and glucose management is stable and that they no longer require ongoing IDS review. They will then de-escalate the participant to home team management. Criteria for this de-escalation include:
• No major hypoglycaemic events i.e. BGL less than or equal to 3.0 mmol/L and fewer episodes of hypoglycaemia.
• No major hypoglycaemic events i.e. BGL greater than or equal to 20.0 mmol/L and fewer episodes of hyperglycaemia.
• A greater proportion of glucose levels within the patient’s individualised target range.
Control group
Active

Outcomes
Primary outcome [1] 325015 0
To determine if proactive specialist care of hospital inpatients with diabetes or new hyperglycaemia results in a change in patient-day mean glucose. Patient-day mean glucose is calculated by grouping all glucose results by patient-day and returning a mean value for each patient-day of all hospital admissions that comprise the study population.
Timepoint [1] 325015 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [1] 386545 0
• Healthcare-associated infection (HAI) incidence
Defined as a sterile-site positive culture or clinician action on a suspected infection (antibiotic prescription) where both the positive culture was taken and the antibiotic prescription commenced at least 48 hours following admission.
Timepoint [1] 386545 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for the duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [2] 386546 0
• Proportion of participants with patient-day weighted mean glucose < 12.0 mmol/L. This outcome is assessed using capillary blood glucose test results from the networked blood glucose meters being employed in this trial.
Timepoint [2] 386546 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for the duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [3] 386547 0
• Mean patient-day weighted mean glucose
Defined as the mean of glucose results for a particular patient on a particular calendar day. This outcome is assessed using capillary blood glucose test results from the networked blood glucose meters being employed in this trial.
Timepoint [3] 386547 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for the duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [4] 386548 0
• Proportion of Adverse Glycaemic Days (AGD)
An AGD is defined as a day on which a participant experiences either hypoglycaemia (glucose less than or equal to 4.0 mmol/L) or hyperglycaemia (glucose greater than or equal to 15.0 mmol/L). This outcome is assessed using capillary blood glucose test results from the networked blood glucose meters being employed in this trial.
Timepoint [4] 386548 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [5] 386549 0
• Ward transfers to a high-dependency unit incidence
Defined as participants who were transferred to a high-dependency unit, including the Intensive Care Unit (ICU), Coronary Care Unit (CCU), and Respiratory Care Unit (RCU) from a hospital ward at least once in the course of their admission. This does not include participants who were admitted to a high-dependency unit from the emergency department or another hospital. This will be assessed from information captured in patient medical records.
Timepoint [5] 386549 0
Duration of inpatient period in trial, i.e. variable depending on when trial exit criteria are met. This outcome is assessed daily for duration of the inpatient period while in trial until exit criteria are met, which includes hospital discharge.
Secondary outcome [6] 388039 0
Length of inpatient stay.
This will be assessed from information captured in patient medical records.
Timepoint [6] 388039 0
This will be assessed at the time of hospital discharge, regardless of whether other trial exit criteria are met.

Eligibility
Key inclusion criteria
• Admission to the Royal Melbourne Hospital City Campus during the study period
• Either:
o a diagnosis of diabetes recorded in the clinical record at the time of hospital admission, or
o a random glucose result greater than or equal to 11.1 mmol/L recorded during the admission
• Age greater than or equal to 18 years
• Non-pregnant
• Admission to a study ward, which includes:
o C 2B Cardiology
o C 2West
o C 3S Surgery
o C 3SW General Surgery
o C 4S NSURG
o C 4SE
o C 5 North
o C 5SE General Medicine
o C 5SW General Medicine
o C 6 South East
o C 6B ICU
o C 6SW Nephrology
o C 7B Haematology and BMT
o C 7SE Plastics
o C 7SW Orthopaedic
o C 8B NEUR
o C 9E VIDS
o C 9W Surgery
o C AMU
• Admission under a study admitting unit, which includes:
o AMU – Medical
o ANPM – Medical
o ANST – Medical
o ASSM – Medical
o BMTX – Medical
o BOE – Surgical
o CARD – Medical
o CHNP – Surgical
o CP – Medical
o CPEU – Medical
o CR – Surgical
o CSUR – Surgical
o EGS – Surgical
o EMER – Medical
o EPIL – Medical
o FLXU – Surgical
o GAST – Medical
o HAEM – Medical
o HB – Surgical
o HNOE – Surgical
o IMMU – Medical
o MU1 – Medical
o MU2 – Medical
o MU3 – Medical
o NEPH – Medical
o NEPS – Surgical
o NEUR – Medical
o NSRI – Surgical
o NSUR – Surgical
o OMFS – Surgical
o OPHT – Surgical
o ORTH – Surgical
o ORTM – Surgical
o ORTS – Surgical
o PLSA – Surgical
o PSR – Surgical
o PSRE – Surgical
o PSRI – Surgical
o RESP – Medical
o RHEU – Medical
o SCU – Medical
o TSUR – Surgical
o TT – Surgical
o UROI – Surgical
o UROL – Surgical
o VASC – Surgical
o VASI – Surgical
o VIDS – Medical
Wards and admitting units are periodically modified by the hospital administration for operational and administrative reasons. If this occurs during the study period, appropriate modification of the study protocol will be performed by the investigators.
All of the above criteria must be met for a potential participant to be eligible to enter the trial.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Length of stay < 48 hours

Study design
Purpose of the study
Treatment
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Dynamic block randomisation with a single stratification factor (type 1 diabetes, medical admitting unit, surgical admitting unit) will be used. The 3 subgroups are:
1. Participants with a diagnosis of type 1 diabetes mellitus (T1DM)
2a. Participants admitted under a surgical admitting unit
2b. Participants admitted under a medical admitting unit

Membership of subgroup 1 takes precedence, i.e. participants with a diagnosis of type 1 diabetes mellitus will be block-randomised independent of their admitting unit. Membership of subgroups 2a and 2b is mutually exclusive. If a patient is admitted under a medical unit and then transferred to a surgical unit, they remain part of group 2a and the data they contribute be analysed as such.
Masking / blinding
Blinded (masking used)
Who is / are masked / blinded?


The people assessing the outcomes
Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
10.1 SAMPLE SIZE ESTIMATION, JUSTIFICATION & POWER CALCULATIONS
a. Infection rate estimates and mean patient-day glucose mean and sd estimates were obtained from Kyi et. al.
b. To detect a difference in patient-day mean glucose, group sample sizes of 1102 (551 patients per treatment arm) achieve 80.037% power to reject the null hypothesis of equal means when the population mean difference is µ1 - µ2 = 9.0 - 9.5 = -0.5 with standard deviations of 2.7 for group 1 and 3.2 for group 2, and with a significance level (alpha) of 0.050 using a two-sided two-sample unequal-variance t-test. Assuming a 10% drop-out rate, 1226 (613 patients per treatment arm) will need to be recruited.
c. Additionally, to detect a difference in infection rates, group sample sizes of 658 (329 in intervention and 329 in control) would achieve 80.055% power to detect a statistically significant difference between the group proportions of 4.6% (2.7% vs. 7% in the control). The proportion the intervention group is assumed to be 2.4% under the null hypothesis and 7% under the alternative hypothesis. The proportion in the control group is 7%. The test statistic used is the two-sided Z-Test with un-pooled variance. The significance level of the test is 0.0500. Assuming a 10% dropout rate, 732 patients would need to be recruited to achieve 80% power.
d. However, if we assume a conservative infection rate of 3% in the intervention group and 7% infection rate in the control group, by recruiting 1376 patients (accounting for a 10% drop-out rate), we would have 90% power to detect a different in infection rates between the intervention groups. Therefore this study aims to recruit 1376 patients, giving us sufficient power to detect a difference in mean patient-day glucose (the primary outcome) in addition to a difference in infection rates.’

10.2 STATISTICAL METHODS TO BE UNDERTAKEN
Primary analysis will be performed on an intention-to-treat basis with sensitivity analysis performed on a per-protocol basis to confirm the robustness of the findings.
Patient-day mean glucose is calculated by grouping all glucose results by patient-day and returning a mean value for each patient-day of all hospital admissions that comprise the study population.
A difference in patient-day mean glucose between treatment groups at a patient’s exit from the study will be tested using a 2-sided t-test allowing unequal variance.
Patient day mean glucose will also be modelled using a Generalised Estimating Equation (GEE) with an autoregressive correlation structure - a model choice that assumes that the correlations between patient day mean glucose measurements taken on the same patient will decrease as the number of days between different measurements increases. The model will test whether a time-treatment interaction effect is present and include it in the primary analysis if the interaction effect is statistically significant. The primary outcome will otherwise be unadjusted. An adjusted model controlling for potential confounders including gender, baseline age, relative socioeconomic deprivation etc. will also be performed.”
A difference in infection rates between the 2 groups will be tested using a two-sided test for proportions with un-pooled variance.
Continuous measures will be assessed for normality and log transformed where appropriate.
In case of participants missing the follow up assessment, appropriate imputation techniques will be employed in consultation with the study statistician.
Descriptive statistics and tables will be used to summarize relevant patient demographics and outcomes overall, by intervention groups. Continuous data will be reported as means (with standard deviations) if approximately normally distributed, and medians (inter-quartile range) and full [range] otherwise. Categorical data will be reported as frequencies and percentages and summarized as proportions with 95% confidence intervals as appropriate.
The baseline differences between study arms will be examined using a two-tailed T-test for continuous, normally distributed data, while the Wilcoxon rank-sum test will be used for skewed or ordinal data and either chi-2 or Fisher’s exact for categorical data.
Multiplicity of testing will be acknowledged but for the primary and secondary endpoints, multiple statistical tests will be otherwise unadjusted.
To avoid overestimating hyperglycemia and hypoglycemia, multiple blood glucose measurements taken during the same hour will be handled according to Weinberg et. al.
Adverse glycaemic days (AGDs), defined as the proportion of patient-days for which the BG level was below 4.0 mmol/L or above 15.0 mmol/L will be compared between the intervention and control group with a two-sided proportion test.
Data analyses will be carried out using the Statistical Package for the Social Sciences (IBM® SPSS® Statistics Premium Grad Pack Version 25.0) or Stata Corporation (Stata Statistical Software: Release 16. College Station, TX: StataCorp LP; 2019).”

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
1/02/2021
Actual
15/02/2021
Date of last participant enrolment
Anticipated
30/11/2022
Actual
Date of last data collection
Anticipated
31/12/2022
Actual
Sample size
Target
1700
Accrual to date
1417
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 17432 0
Royal Melbourne Hospital - City campus - Parkville
Recruitment postcode(s) [1] 31161 0
3050 - Parkville

Funding & Sponsors
Funding source category [1] 306653 0
Hospital
Name [1] 306653 0
The Royal Melbourne Hospital
Country [1] 306653 0
Australia
Primary sponsor type
Hospital
Name
Melbourne Health
Address
300 Grattan Street
Parkville
Victoria 3050
Country
Australia
Secondary sponsor category [1] 307202 0
None
Name [1] 307202 0
Address [1] 307202 0
Country [1] 307202 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 306835 0
Melbourne Health Human Research Ethics Committee
Ethics committee address [1] 306835 0
The Royal Melbourne Hospital
300 Grattan Street
Parkville
Victoria 3050
Ethics committee country [1] 306835 0
Australia
Date submitted for ethics approval [1] 306835 0
27/05/2020
Approval date [1] 306835 0
06/07/2020
Ethics approval number [1] 306835 0
HREC: 2020.075

Summary
Brief summary
This study aims to answer the question: Does proactive (early intervention) specialist care for hospital inpatients with diabetes or new hyperglycaemia improve glucometric and clinical outcomes? All patients admitted to The Royal Melbourne Hospital during the trial period who have diabetes or hyperglycaemia will be randomised to usual care or proactive specialist care. Proactive specialist care involves the inpatient diabetes service (IDS) identifying patients with diabetes or hyperglycaemia proactively without the team looking after the patient necessarily referring to the IDS for their input. The IDS will then make management suggestions with the effect of these suggestions on outcomes such as glucose levels, in-hospital complications, and length of stay assessed.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 105178 0
A/Prof Spiros Fourlanos
Address 105178 0
Department of Diabetes & Endocrinology
The Royal Melbourne Hospital
300 Grattan Street
Parkville
Victoria 3050
Country 105178 0
Australia
Phone 105178 0
+61 3 9342 7365
Fax 105178 0
+61 3 9342 8933
Email 105178 0
[email protected]
Contact person for public queries
Name 105179 0
Spiros Fourlanos
Address 105179 0
Department of Diabetes & Endocrinology
The Royal Melbourne Hospital
300 Grattan Street
Parkville
Victoria 3050
Country 105179 0
Australia
Phone 105179 0
+61 3 9342 7365
Fax 105179 0
+61 3 9342 8933
Email 105179 0
[email protected]
Contact person for scientific queries
Name 105180 0
Spiros Fourlanos
Address 105180 0
Department of Diabetes & Endocrinology
The Royal Melbourne Hospital
300 Grattan Street
Parkville
Victoria 3050
Country 105180 0
Australia
Phone 105180 0
+61 3 9342 7365
Fax 105180 0
+61 3 9342 8933
Email 105180 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
The nature of this trial requires, for practical purposes, patients to be entered into the trial prior to or potentially in the absence of any contact with the trial team. While the NHMRC guidelines for waiver of individual consent are met for the purpose of conducting the trial, release of potentially identifying individual patient data is not covered by this waiver and cannot be reliably obtained from the entire study population.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
9052Study protocol  [email protected]



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.