Ignite Creation Date:
2024-05-06 @ 4:18 PM
Last Modification Date:
2024-10-26 @ 2:08 PM
Study NCT ID:
NCT04944225
Status:
NOT_YET_RECRUITING
Last Update Posted:
2024-02-28
First Post:
2021-06-21
Brief Title:
Opioid Reduction Strategy South Western Ontario
Sponsor:
Lawson Health Research Institute
Organization:
Lawson Health Research Institute
Study Overview
Official Title:
Implementation of a Multi-faceted Opioid-Use Reduction Strategy for South Western Ontario A Pragmatic Stepped-Wedge Cluster Randomized Trial
Status:
NOT_YET_RECRUITING
Status Verified Date:
2024-02
Last Known Status:
None
Delayed Posting:
No
If Stopped, Why?:
Not Stopped
Has Expanded Access:
False
If Expanded Access, NCT#:
N/A
Has Expanded Access, NCT# Status:
N/A
Acronym:
None
Brief Summary:
Pain is a major risk factor for chronic postoperative pain Adequate perioperative pain relief is an important metric for patient satisfaction and to achieve good recovery outcomes Opioids remain the primary systemic pharmacotherapy for intraoperative and postoperative analgesia particularly for moderate to severe pain When used judiciously opioids are effective in reducing suffering and helping patients cope with postoperative pain However there are challenges - a side effects can result in harm like respiratory depression b over-reliance on opioids can increase drug dependency c over-prescription can encourage addiction overdose and death leading to a human and financial burden from both an individual and public health standpoint Over-prescription of opioids for acute pain is strongly linked to patient morbidity and mortality For example a new opioid prescription raises the risk of lethal or non-lethal overdose as well as the conversion from opioid-naive to chronic user Canadian Institute of Health Information CIHI and Public Health Agency of Canada PHAC data emphasize the public health need to reduce reliance on opioids From January 2016 to June 2018 more than 9000 Canadians died from apparent opioid related harms In 2017 an average of 17 Canadians were hospitalized for opioid poisonings each day - an increase from 16 per day in 2016 Prescription opioid use appears to be an early driver of the current crisis Given the local and national severity of the opioid crisis there is need for a pragmatic timely and scalable intervention to reduce reliance on opioids as we strive to improve healthcare for patients and alleviate the economic burden on the medical system This proposal for a stepped-wedge randomized trial of a multi-faceted opioid-use reduction strategy addresses key drivers of the opioid crisis and has the potential to reduce patient exposure to opioids and thereby improve morbidity and mortality Hospitals involved in this study will all eventually participate in an opioid reduction strategy that will limit the access and prescription of opioids to surgical patients and will incorporate various opioid reduction strategies at both a patient and hospital level
Detailed Description:
Proposed Trial Design The proposed prospective randomized trial will employ a stepped-wedge design which is a variant of a cluster randomized trial This is a pragmatic two-armed parallel group registry based cluster RCT This SWAHN associated hospitals trial will be embedded into routine care new standard of care with implementation of the intervention delivered by perioperative personnel rather than research staff Patient characteristics and outcomes will be obtained from ICES administrative health care database This pragmatic design allows broad inclusion of the south west LHIN hospitals and a large representative sample of perioperative patients that should yield highly generalizable findings The hospitals will be randomized 11 with concealed allocation methods and hospitals will be notified of their group allocation by the study team 2 months before the intervention start date
In this trial the unit of randomization is the cluster ie each of the hospital and the unit of analysis is the patient The investigators chose a cluster randomized design to enhance intervention uptake and adherence logistical convenience and to minimize cross-group contamination Perioperative patients typically receive all their intervention at the same centre making this population suitable for cluster-level interventions Delivery of the opioid reduction strategy intervention in this cluster trial follows what occurs in routine care where the perioperative team in each hospital will be trained to follow the same protocol or policy for patients under their care
This will entail the roll out of the intervention in a randomized fashion and sequential crossover of clusters from control to intervention until all clusters are exposed This type of trial is a pragmatic approach to investigate and implement a service delivery change given that the intervention will occur at the institutional level and individual patients will not need to be randomized
This multi-centre stepped-wedge cluster- RCT is designed to evaluate the benefits of an opioid reduction strategy compared to standard practice as per local hospital protocol in patients undergoing elective surgical procedures who are prescribed opioids to control acute post- operative pain In order to overcome data collection challenges the investigators will collect the outcomes from all elective surgical patients no exclusions from electronic databases ICES for the pre- and post-intervention Following a 2-month baseline period in which all sites use their usual standard-of-care opioid prescription one of the randomly selected sites will begin implementing and following the opioid reduction strategy other sites will be randomly added to the intervention group every two months until the strategy is in place at all the sites total 12 months of data collection One month before crossover to the phase of opioid reduction strategy at each site all anesthesia and research staff at the hospital site will be trained by the study PI Mahesh Nagappa on the various intervention components on implementing the opioid reduction strategy and on collection of data
Twelve hospitals within the Southwestern Ontario Academic Hospital Network SWAHN will be invited to participate London Health Sciences Centre and St Josephs Healthcare London will be the lead sites For this study we have selected the perioperative setting in which opioid analgesia prescribing is common during the hospital discharge
Participants Recruited patients will be those receiving postoperative analgesia at participating study sites after undergoing an elective surgical procedure Given that the intervention will be implemented on an institutional-wide level eg via changing hospital order sets and protocols there will be no need to randomize individual patients Since the intervention is focused on opioid use post-discharge intraoperative and post-operative pain management will not be specified in the opioid reduction strategy In-hospital pain management will be at the discretion of the attending anesthesiologist as described elsewhere in the literature
Control group Pain management during the standard of care phase In the standard of care phase anesthetic and surgical care will be as per standard practice according to local hospital protocol for both the control and the intervention group Generally this means patients will be maintained on a more liberal opioid regime than in the opioid reduction strategy phase and will receive opioid and other medications for the acute postoperative pain The choice of opioids will be at the discretion of the managing team There will be a minimum 2-month baseline period before entry of the first randomized cluster to the intervention arm
Intervention group Pain management in the opioid reduction strategy phase The opioid reduction strategy including the tools and locally contextualized approach to implementation will be co-designed by a multi-disciplinary team of surgeons anesthesiologists pharmacists and researchers with expertise in knowledge translation together with patient representatives To ensure indigenous perspectives are integrated one of the patient representatives will be from the local indigenous community
The intervention will involve a multi-faceted 3 component please see the supplementary file approach involving 1 opioid prescription caps default maximum number of tablets for discharge prescriptions as defined by evidence-based guidelines such as httpsmichigan-openorgprescribing-recommendations and httpswwwhqontariocaevidence-to-improve-carequality-standardsview-all-quality-standardsopioid-prescribing-for-acute-pain 2 patient education tools eg What is a normal pain trajectory How to manage the pain Benefits and potential harms of pharmacologic analgesia Non-pharmacologic analgesia management What to do if pain is excessive 3 provider education tools eg including procedure-specific evidence-based recommendations for multi-modal analgesia comparison of local baseline prescribing patterns with exemplary prescribing patterns defining targeted reduction if baseline prescribing is at odds with best evidence review of best evidence about optimal analgesia perioperatively and 4 bi-weekly cumulative prescriber feedback on opioid prescribing patterns post-intervention and until end-of-study
The strategy for implementing each of these components will be contextualized in collaboration with the multidisciplinary team from each hospital prior to entry to the active intervention phase through meetings with the SWAHN Opioid Choosing Wisely Committee and with local meetings intensified during the first weeks of the active intervention
Practical arrangements for allocating participants to trial groups Each hospital will be randomly assigned to one of the crossover dates prior to the start of the study by a statistician who is blinded to hospital identities using a computer-generated list of random numbers
Proposed methods for protecting against sources of bias It is not possible to blind clinicians or study staff to hospital allocation because the appropriate painopioid management must be transparently applied upon entry to the intervention phase of the study and therefore the unit of randomization is at the hospital level Outcome adjudicators will be blinded and patients will remain blinded We anticipate low risk of selection bias at the patient level because all eligible patients will be enrolled in the study From the investigators previous experience the investigators expect few 5 participants to refuse consent to use their data
Clinicians will abide by the appropriate opioid reduction strategy algorithms The duration of the opioid reduction strategy phase will vary at each hospital based on the randomization schedule ranging from 4 to 12 months
Proposed frequency and duration of follow up The investigators will collect data on patients from PODs 1 or hospital discharge whichever comes first and POD 30 or death whichever comes first
Sample size and power calculations As expected the investigators found that the sample size calculation is very tricky for this prospective stepped wedge trial The investigators were originally aiming for sample size calculations based on the Hussey Hughes approach for analysis with fixed time effects and random cluster effects However the investigators settled on estimating the outcome based on the available power as the available sample size was very large 8000 per month The investigators will co-opt the ICES statistical team for the analysis along with one of local statistical experts This way the investigators will navigate the layers of complexity to overcome the analysis plan The methods for adjusting for confounders has evolved significantly in recent years If the investigators consider that there are multiple levels of potential confounders uneven contamination over timecluster sizes larger intracluster correlations hospital level clustering and physician level clustering then the large number of patients across 12 hospitals 7000-8000month the investigators will likely have greater than 90 power to detect a 25 reduction in morphine equivalents even with worst case scenario estimates for above mentioned clustering and confounders Using hospital statistics data we estimate that 8000 eligible patients would be registered across 12 SWAHN hospitals over 4 weeks with a 25 effect size in the MME and a between hospital coefficient of variation of 015 Assuming a constant case-load during the pandemic independent hospital effects and a 5 significance level the trial would have 90 power to detect a 25 reduction in MME If the assumption of independent hospital effects was not met and the 12 SWAHN hospital clusters functioned effectively as 12 large hospitals power would be still 80
Analyses will be carried out using intention-to-treat based on the date each hospital is assigned to cross over rather than the actual crossover date With the patient as the unit of analysis we will determine the effect of the intervention on primary and secondary outcomes The investigators will use generalized linear mixed models GLMM with logit link for the binary outcome and random intercept to account for the clustering of patients within hospitals The model will adjust for patient characteristics eg age obesity OSA etc and for time secular trends A time by treatment interaction will be tested The effect of the intervention will be estimated by odds ratios OR and 95 confidence intervals CI in the final model Secondary outcomes will be analyzed using GLMM for binary outcomes linear mixed model for continuous outcome assuming normal distribution and survival analysis for length of hospital stay Rate of missing data should be low 5 as outcomes are routinely collected to guide clinical care Statistical significance will be defined as P005 or 95 confidence intervals that exclude the null effect All analyses will be conducted using the intention-to-treat approach We will analyze the primary outcome and all other continuous variables using mixed-effects linear regression with the intervention and time as fixed effects in the model
A random intercept and slope for time defined at the cluster level will account for within-period and between-period intracluster correlations Considering the relatively small number of clusters the investigators will use the Kenward-Roger correction to avoid a potentially inflated type I error rate The binary secondary outcomes will be analyzed using a similar approach but with mixed-effects logistic regression models Outcomes will also be stratified according major types of surgery Statistical analyses will be performed using SAS and R statistical package
In this trial the unit of randomization is the cluster ie each of the hospital and the unit of analysis is the patient The investigators chose a cluster randomized design to enhance intervention uptake and adherence logistical convenience and to minimize cross-group contamination Perioperative patients typically receive all their intervention at the same centre making this population suitable for cluster-level interventions Delivery of the opioid reduction strategy intervention in this cluster trial follows what occurs in routine care where the perioperative team in each hospital will be trained to follow the same protocol or policy for patients under their care
This will entail the roll out of the intervention in a randomized fashion and sequential crossover of clusters from control to intervention until all clusters are exposed This type of trial is a pragmatic approach to investigate and implement a service delivery change given that the intervention will occur at the institutional level and individual patients will not need to be randomized
This multi-centre stepped-wedge cluster- RCT is designed to evaluate the benefits of an opioid reduction strategy compared to standard practice as per local hospital protocol in patients undergoing elective surgical procedures who are prescribed opioids to control acute post- operative pain In order to overcome data collection challenges the investigators will collect the outcomes from all elective surgical patients no exclusions from electronic databases ICES for the pre- and post-intervention Following a 2-month baseline period in which all sites use their usual standard-of-care opioid prescription one of the randomly selected sites will begin implementing and following the opioid reduction strategy other sites will be randomly added to the intervention group every two months until the strategy is in place at all the sites total 12 months of data collection One month before crossover to the phase of opioid reduction strategy at each site all anesthesia and research staff at the hospital site will be trained by the study PI Mahesh Nagappa on the various intervention components on implementing the opioid reduction strategy and on collection of data
Twelve hospitals within the Southwestern Ontario Academic Hospital Network SWAHN will be invited to participate London Health Sciences Centre and St Josephs Healthcare London will be the lead sites For this study we have selected the perioperative setting in which opioid analgesia prescribing is common during the hospital discharge
Participants Recruited patients will be those receiving postoperative analgesia at participating study sites after undergoing an elective surgical procedure Given that the intervention will be implemented on an institutional-wide level eg via changing hospital order sets and protocols there will be no need to randomize individual patients Since the intervention is focused on opioid use post-discharge intraoperative and post-operative pain management will not be specified in the opioid reduction strategy In-hospital pain management will be at the discretion of the attending anesthesiologist as described elsewhere in the literature
Control group Pain management during the standard of care phase In the standard of care phase anesthetic and surgical care will be as per standard practice according to local hospital protocol for both the control and the intervention group Generally this means patients will be maintained on a more liberal opioid regime than in the opioid reduction strategy phase and will receive opioid and other medications for the acute postoperative pain The choice of opioids will be at the discretion of the managing team There will be a minimum 2-month baseline period before entry of the first randomized cluster to the intervention arm
Intervention group Pain management in the opioid reduction strategy phase The opioid reduction strategy including the tools and locally contextualized approach to implementation will be co-designed by a multi-disciplinary team of surgeons anesthesiologists pharmacists and researchers with expertise in knowledge translation together with patient representatives To ensure indigenous perspectives are integrated one of the patient representatives will be from the local indigenous community
The intervention will involve a multi-faceted 3 component please see the supplementary file approach involving 1 opioid prescription caps default maximum number of tablets for discharge prescriptions as defined by evidence-based guidelines such as httpsmichigan-openorgprescribing-recommendations and httpswwwhqontariocaevidence-to-improve-carequality-standardsview-all-quality-standardsopioid-prescribing-for-acute-pain 2 patient education tools eg What is a normal pain trajectory How to manage the pain Benefits and potential harms of pharmacologic analgesia Non-pharmacologic analgesia management What to do if pain is excessive 3 provider education tools eg including procedure-specific evidence-based recommendations for multi-modal analgesia comparison of local baseline prescribing patterns with exemplary prescribing patterns defining targeted reduction if baseline prescribing is at odds with best evidence review of best evidence about optimal analgesia perioperatively and 4 bi-weekly cumulative prescriber feedback on opioid prescribing patterns post-intervention and until end-of-study
The strategy for implementing each of these components will be contextualized in collaboration with the multidisciplinary team from each hospital prior to entry to the active intervention phase through meetings with the SWAHN Opioid Choosing Wisely Committee and with local meetings intensified during the first weeks of the active intervention
Practical arrangements for allocating participants to trial groups Each hospital will be randomly assigned to one of the crossover dates prior to the start of the study by a statistician who is blinded to hospital identities using a computer-generated list of random numbers
Proposed methods for protecting against sources of bias It is not possible to blind clinicians or study staff to hospital allocation because the appropriate painopioid management must be transparently applied upon entry to the intervention phase of the study and therefore the unit of randomization is at the hospital level Outcome adjudicators will be blinded and patients will remain blinded We anticipate low risk of selection bias at the patient level because all eligible patients will be enrolled in the study From the investigators previous experience the investigators expect few 5 participants to refuse consent to use their data
Clinicians will abide by the appropriate opioid reduction strategy algorithms The duration of the opioid reduction strategy phase will vary at each hospital based on the randomization schedule ranging from 4 to 12 months
Proposed frequency and duration of follow up The investigators will collect data on patients from PODs 1 or hospital discharge whichever comes first and POD 30 or death whichever comes first
Sample size and power calculations As expected the investigators found that the sample size calculation is very tricky for this prospective stepped wedge trial The investigators were originally aiming for sample size calculations based on the Hussey Hughes approach for analysis with fixed time effects and random cluster effects However the investigators settled on estimating the outcome based on the available power as the available sample size was very large 8000 per month The investigators will co-opt the ICES statistical team for the analysis along with one of local statistical experts This way the investigators will navigate the layers of complexity to overcome the analysis plan The methods for adjusting for confounders has evolved significantly in recent years If the investigators consider that there are multiple levels of potential confounders uneven contamination over timecluster sizes larger intracluster correlations hospital level clustering and physician level clustering then the large number of patients across 12 hospitals 7000-8000month the investigators will likely have greater than 90 power to detect a 25 reduction in morphine equivalents even with worst case scenario estimates for above mentioned clustering and confounders Using hospital statistics data we estimate that 8000 eligible patients would be registered across 12 SWAHN hospitals over 4 weeks with a 25 effect size in the MME and a between hospital coefficient of variation of 015 Assuming a constant case-load during the pandemic independent hospital effects and a 5 significance level the trial would have 90 power to detect a 25 reduction in MME If the assumption of independent hospital effects was not met and the 12 SWAHN hospital clusters functioned effectively as 12 large hospitals power would be still 80
Analyses will be carried out using intention-to-treat based on the date each hospital is assigned to cross over rather than the actual crossover date With the patient as the unit of analysis we will determine the effect of the intervention on primary and secondary outcomes The investigators will use generalized linear mixed models GLMM with logit link for the binary outcome and random intercept to account for the clustering of patients within hospitals The model will adjust for patient characteristics eg age obesity OSA etc and for time secular trends A time by treatment interaction will be tested The effect of the intervention will be estimated by odds ratios OR and 95 confidence intervals CI in the final model Secondary outcomes will be analyzed using GLMM for binary outcomes linear mixed model for continuous outcome assuming normal distribution and survival analysis for length of hospital stay Rate of missing data should be low 5 as outcomes are routinely collected to guide clinical care Statistical significance will be defined as P005 or 95 confidence intervals that exclude the null effect All analyses will be conducted using the intention-to-treat approach We will analyze the primary outcome and all other continuous variables using mixed-effects linear regression with the intervention and time as fixed effects in the model
A random intercept and slope for time defined at the cluster level will account for within-period and between-period intracluster correlations Considering the relatively small number of clusters the investigators will use the Kenward-Roger correction to avoid a potentially inflated type I error rate The binary secondary outcomes will be analyzed using a similar approach but with mixed-effects logistic regression models Outcomes will also be stratified according major types of surgery Statistical analyses will be performed using SAS and R statistical package
Study Oversight
Has Oversight DMC:
None
Is a FDA Regulated Drug?:
False
Is a FDA Regulated Device?:
False
Is an Unapproved Device?:
None
Is a PPSD?:
None
Is a US Export?:
None
Is an FDA AA801 Violation?:
None