ANZCTR - Registration

Registration number

ACTRN12613000218796

Ethics application status

Approved

Date submitted

21/02/2013

Date registered

25/02/2013

Date last updated

13/09/2017

Type of registration

Prospectively registered

Titles & IDs

Public title

Improving care through imbedding general practitioners within residential aged care facilities

Scientific title

What is the impact of introducing residential aged care facility based general practitioners compared to non-residential aged care facility based general practitioners on the rate of unplanned hospital transfers, polypharmacy, and falls amongst aged care facility residents.

Secondary ID [1]

Nil

Universal Trial Number (UTN)

Trial acronym

Linked study record

Health condition

Health condition(s) or problem(s) studied:

Aged care

Condition category

Condition code

Public Health

Health service research

Intervention/exposure

Study type

Interventional

Description of intervention(s) / exposure

The intervention is a change in service delivery model. The key changes include: * Introduction of a facility-based general practitioner employed at each site. * A clinical co-ordinator will be appointed at each site from existing staff if possible, to co-ordinate interactions with the general practitioner. The role of the general practitioner at each site will be established and integrated with existing processes and procedures. Communications will be made with existing general practitioners of residents advising them of the change in model of care at each site. In addition to the GP, the additional staff roles being introduced are: * Care manager—“To manage the delivery of person centred care to residents, ensuring the highest standard of ongoing assessment, care planning, evaluation and clinical governance, helping them to live longer, healthier, happier lives.” They will have responsibility for ~40 beds. * Clinical manager—“To support the GP in managing the medical practice, providing expert health services to approximately 150 residents across one or more BUPA care homes. To provide the link between Care Manager and GP in delivery of collaborative, integrated healthcare including assessment, care planning and clinical intervention.” * Registered Nurse in charge—“To supervise the delivery of person centred care to residents, as delegated by the Care Manager, to ensure the highest standard of ongoing assessment, care planning, evaluation and clinical governance, helping residents to live longer, healthier, happier lives.” * Team leader—“To deliver highest quality person centred care to residents, as delegated by the Care Manager/RN in charge. Ensuring safe medication administration, best practice wound care and clinical support for residents within the unit.” Organisational change facilitators will facilitate change in service delivery model at each site within the trial over the initial 63 days of the intervention period. Support will continue to be provided by change facilitators to the respective facility staff over the time period of the intervention, and over the 365 days of prospective follow-up as required. The new GP role is not full time in all facilities, rather, it is proportional to the number of beds at the facility. The new nursing roles will go to current employees if appropriate. The positions will be full time but may include job sharing arrangements.

The medications will no longer be delivered by a registered nurse, dispensing from a medication trolley, rather, they will be pre-packaged, kept in the resident’s room, and residents will be facilitated to take their medications by a personal care attendant. These Personal Care Attendants will be trained to undertake this role. The timing of medication distribution will transition from scheduled rounds at 8am, 12 midday, 6pm, (+ exceptions for particular medications), and towards scheduled rounds at 8am, 2pm, 8pm (+ exceptions for particular medications) with residents providing input where possible as to the timing of their medications.

Intervention code [1]

Other interventions

Comparator / control treatment

The comparator condition is usual care.
Usual care includes around the clock care, existing roles for the Director of Care (responsibility for ~120 beds), registered nurses, and enrolled nurses.
General Practitioners are called in to the facility. There is no nurse liaison role.

Control group

Active

Outcomes

Primary outcome [1]

Polypharmacy
Measurement approach:
1) Mean number of regular medications prescribed per resident.
2) Proportion of residents taking >5 medications.
These medications include oral and parenteral and exclude topical medications (other than patches)

Timepoint [1]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Primary outcome [2]

Unplanned hospital transfers
Measurement approach:
The number of resident transfers to hospital (excluding planned appointments) at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of unplanned hospital transfers per 1000 resident occupied bed days.

Timepoint [2]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Primary outcome [3]

Falls and Fracture falls
Measurement approach:
The number of resident falls and falls resulting in fracture at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
These data will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of falls and rate of falls resulting in fracture per 1000 resident occupied bed days.

Timepoint [3]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [1]

Mortality
Measurement approach:
The number of resident deaths at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of resident death per 1000 resident occupied bed days.

Timepoint [1]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [2]

Complaints
Measurement approach:
The number of complaints at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of complaints per 1000 resident occupied bed days.

Timepoint [2]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [3]

General practitioner out of hours calls
Measurement approach:
The number of general practitioner out of hours calls at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of out of hours general practitioner calls per 1000 resident occupied bed days.

Timepoint [3]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [4]

New infections
Measurement approach:
The number of new resident infections at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate new infections per 1000 resident occupied bed days.

Timepoint [4]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [5]

Antibiotic use
Measurement approach:
The proportion of residents who use antibiotics as of each 63rd day during the trial.

Timepoint [5]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [6]

Analgesic use
Measurement approach:
The proportion of residents who use analgesics (Non-steroidal anti-inflammatory, opioids, paracetamol, combination analgesics) as of each 63rd day during the trial.

Timepoint [6]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [7]

Psychotropic use
Measurement approach:
The proportion of residents who use psychotropic medications as of each 63rd day during the trial.

Timepoint [7]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [8]

Dementia medication use
Measurement approach:
The proportion of residents who use dementia medications as of each 63rd day during the trial.

Timepoint [8]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [9]

PRN -pro re nata -medication use
Measurement approach:
The proportion of residents who are prescribed PRN medications as of each 63rd day during the trial.

Timepoint [9]

A series of cross-sectional snapshots will be taken every 63rd day during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [10]

Staff turnover
Measurement approach:
The number of staff resignations will be summed within each facility during each 63 day period.
This number will be divided by the total rostered staff hours within each facility during each 63 day period to form a rate of staff resignation per rostered staff hour.

Timepoint [10]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [11]

Staff satisfaction
Measurement approach
Annual surveys of staff satisfaction are routinely conducted.

Timepoint [11]

Data from the immediate pre-trial survey (baseline), mid-trial survey (52 weeks), and immediate post-trial survey (104 weeks) will be used.

Secondary outcome [12]

Pressure areas
Measurement approach:
The number of new resident pressure areas at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of new pressure areas per 1000 resident occupied bed days.

Timepoint [12]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [13]

Staff influenza vaccination
Measurement approach:
Data is collected annually on the number of staff who are vaccinated in each facility. This data will be divided by the total number of staff in each facility to create a proportion.

Timepoint [13]

Annual data collected for the immediate pre-trial year, mid-trial year, and immediate post-trial year will be used

Secondary outcome [14]

Skin tears
Measurement approach:
The number of resident skin tears at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of skin tears per 1000 resident occupied bed days.

Timepoint [14]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [15]

Episodes of resident aggression
Measurement approach:
The number of episodes of resident aggression at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial.
This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial.
The resultant outcome will be the rate of episodes of resident aggression per 1000 resident occupied bed days.

Timepoint [15]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [16]

Medication Errors
Measurement approach: The number of episodes of medication error at each facility will be collected in an ongoing basis during the trial and summed over each 63 day period during the trial. This number will be divided by the number of resident occupied bed days in each facility over each 63 day period during the trial. The resultant outcome will be the rate of medication error per 1000 resident occupied bed days.

Timepoint [16]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [17]

Unplanned resident admission to hospital

Timepoint [17]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Secondary outcome [18]

Number of resident days spent in hospital for an unplanned admission

Timepoint [18]

Data will be collected in an ongoing basis during the trial, and summed within each facility within each 63 day period during the trial. The stepped-wedge design is broken into "steps" of 63 days duration. The stepped-wedge trial will last for 504 days, following which there will be 365 days of prospective follow-up.

Eligibility

Key inclusion criteria

All residents of 15 participating Residential Aged Care Facilities are included in the study

Minimum age

18 Years

Maximum age

No limit

Sex

Both males and females

Can healthy volunteers participate?

Yes

Key exclusion criteria

No resident at participating facilities will be excluded.

Study design

Statistical methods / analysis

Linear mixed model analyses will be conducted for each outcome where assessments (data broken into 7 week blocks) are nested within the residents, residents are nested within sites and sites are nested within clusters and each are treated as random effects indicating that they are representative of a broader population of time points, residents, and nursing facilities. The intervention will be considered as a fixed effect (i.e. that it has not been drawn from a broader population of interventions, rather, that it is a specific intervention of interest). Two styles of analysis will be undertaken. The first will only involve data from the stepped-wedge cluster RCT design. The “roll-out” block for each site and the 9 week block following this will be excluded from the analysis in each case as although the intervention has commenced at each site, it is not yet fully instituted (this time that will be excluded from the analysis is not included in the 504 day time period of the stepped-wedge trial). A total of 18 weeks of "block-out" time is required to ensure that there is sufficient time for staff at each site to be trained, for a suitable general practitioner to be recruited, for them to give notice for leaving their previous employment and commencing at the research site. Intervention by time interaction effects will also be examined. Secondary analyses will be conducted that incorporate 12 months of retrospectively collected, pre-trial data, and 12 months of prospectively collected post-trial data. Economic analysis The health economics analyses will include investigations of reductions in direct medical costs foreseen due to changes in the rates of hospital admissions, infections, ER visits, afterhours consultations, medication usage (all types including antibiotics, psychotropic, analgesic etc.) and supplemental feeds; and reductions in indirect costs due to lost productivity that are envisaged with increased staff satisfaction leading to decreased sick leave and staff turnover, as well as decreased time spent on resolution of care and GP-related complaints. These costs will offset to some extent the salary costs of employing resident GPs. Medicare billing for consultations and procedures will similarly be quantified to calculate additional offset of salary costs. Health economics analyses will include the evolution of direct and indirect costs for intervention versus control over time. Units of resource consumption will be combined with Australian unit costs to calculate direct medical costs. For example published Pharmaceutical Benefits Scheme medication cost data will be combined with medication usage data from patient records to calculate costs of medications. Medical Benefits Scheme item numbers and reimbursement values will be combined medical consultations, procedures and investigations to calculate the total costs of these items. In the case of hospitalizations, several methods will be employed. As a first pass, Australian Diagnosis-Related Group costs will be used, based on the diagnosis of the condition that led to a hospital separation. A weakness of doing this is that for aged care residents, the average length of stay is likely to be higher than the average length of stay used to calculate the A-DRG costs. To address this, the number of days in hospital will be recorded by BUPA for each hospitalization separation. This information is currently already captured in the patient records. This information will be used to up-scale the hoteling and other relevant part of the DRG cost where appropriate to partially account for the higher costs of aged care residents’ hospitalisation costs. The benefits of this approach are: 1. that it will utilize routinely collected information easily available to BUPA, without having to gain patient consent, as would be required if a detailed hospital patient record analysis was performed; 2. and as the information is already currently collected by BUPA, there will be no additional research costs. To further account for uncertainty relating to hospitalization costs, sensitivity analysis will be performed using different hospitalization-related cost assumptions, varying costs within plausible ranges. The impact of doing this on the total costs for the intervention and control arms will be assessed. Changes in indirect costs due to lost staff productivity will be assessed using human resource data routinely collected by BUPA. Days off work will be combined with daily salary to estimate the value of lost productivity from sick leave. Descriptive and comparative statistical analyses will be performed. Cost-consequence analyses will summarize the direct, indirect and total costs per resident in each treatment arm, as well as the differences in these costs between each treatment arm. Further analyses may include cost-benefit ratio (return on investment) and incremental cost-effectiveness ratio. As health utility data have not been and will not be collected, it will not be possible to calculated costs per quality-adjusted life years gained.

Recruitment

Recruitment status

Completed

Date of first participant enrolment

Anticipated

6/05/2013

Actual

8/07/2013

Date of last participant enrolment

Anticipated

Actual

4/10/2015

Date of last data collection

Anticipated

Actual

4/10/2015

Sample size

Target

15

Accrual to date

Final

15

Recruitment in Australia

Recruitment state(s)

NSW,QLD,SA,VIC

Funding & Sponsors

Funding source category [1]

Charities/Societies/Foundations

Name [1]

BUPA Health Foundation

Country [1]

Australia

Primary sponsor type

University

Name

University of Tasmania

Country

Australia

Secondary sponsor category [1]

University

Name [1]

Monash University

Country [1]

Australia

Other collaborator category [1]

Commercial sector/Industry

Name [1]

BUPA Aged Care

Country [1]

Australia

Ethics approval

Ethics application status

Approved

Ethics committee name [1]

Social Sciences Human Research Ethics Committee (Tasmania) Network,University of Tasmania and Department of Health and Human Services

Ethics committee address [1]

Human Research Ethics Committee (Tasmania) network
University of Tasmania
Private Bag 01
Hobart 
Tasmania 7001
Australia

Ethics committee country [1]

Australia

Date submitted for ethics approval [1]

Approval date [1]

12/02/2013

Ethics approval number [1]

H12892

Summary

Brief summary

This study aims to evaluate the effect of changing the current model of care in 15 BUPA nursing homes, to a model that employs GPs directly within the homes, in terms of resident health and healthcare resource use outcomes (primary outcomes: polypharmacy, unplanned hospital transfers, general practitioners’ out of hours calls). Nursing homes in four states and in metropolitan and regional locations will be inducted into the program in a randomised and step-wise order, with seven weeks of preparation in the facility before the GP is employed. 
It is envisaged that employment of GPs will improve several important factors, including medical access and care for residents; increased satisfaction to residents and relatives of residents; reductions in medical costs associated with aged care; improved job satisfaction for employees with subsequent decreased turn-over rates and absenteeism.

Trial website

Public notes

The outcomes that were to be collected and analysed at the participant level can still be analysed at the site level.  Our planned analysis of outcomes at the participant level required linkage of two or more aged-care provider datasystems using a common unique identifier. During data extraction we detected inconsistencies in participant identifiers in one of the datasystems which was migrated from one system to another system during the trial. Consequently, we could only rely on the datasystems that had consistent participant identifiers. This meant that we could only conduct analyses at the site-level.

Contacts

Principal investigator

Name

Prof Andrew Robinson

Address

Professor of Aged Care Nursing and Co-Director Wicking Dementia Research and Education Centre
Private bag 121
Hobart, Tasmania 7001

Country

Australia

Phone

+61 3 62264735

Email

[email protected]

Contact person for public queries

Name

Andrew Robinson

Address

Professor of Aged Care Nursing and Co-Director Wicking Dementia Research and Education Centre
Private bag 121
Hobart, Tasmania 7001

Country

Australia

Phone

+61 3 62264735

Email

[email protected]

Contact person for scientific queries

Name

Terry Haines

Address

Allied Health Research Unit
Kingston Centre
Kingston Rd
Cheltenham
Vic 3192

Country

Australia

Phone

+61 3 92651822

Email

[email protected]

Trial Review

Documents added manually

Documents added automatically