Multicenter Registry for Hyperbaric Oxygen Therapy
Oxygen delivered at pressures higher than atmospheric, hyperbaric oxygen (HBO), is a therapy approved for several existing and emerging indications
What is the Multicenter Registry for Hyperbaric Oxygen Therapy and why do we need it?
Most hyperbaric providers have experienced insurance denials for indications approved by the Undersea and Hyperbaric Medical Society. The only cure for these denials is reliable data. The Multicenter Registry for Hyperbaric Oxygen Therapy is designed to provide these data.
The Registry is a database designed to gather a consistent set of measures about the use, outcomes, and safety of Hyperbaric Oxygen Treatments from hyperbaric centers around the world. The goal is to publish this information for quality improvement, patient care, research, and to document the outcomes from hyperbaric oxygen treatment. The Registry collects data for the 14 UHMS-approved indications for hyperbaric treatments as well as selected other emerging indications. The data from the participating centers are aggregated to provide information that no single center could collect on its own. Currently most of the data available about hyperbaric oxygen treatments are based on small numbers from a limited number of centers.
The database uses REDCap software (from Vanderbilt University). REDCap is a web-based, HIPAA-compliant data collection system that’s easy to use and widely available. The Registry contains a core set of data that needs to be completed for all patients being treated at participating centers. These data include basic demographic data, treatment information, safety, and defined outcome measures. The outcome data are tailored to the indication and are primarily routinely used clinical measures. It only takes a few minutes to enter the required data for each patient. Data do not need to be entered every day on every patient.
Importantly, no identifiable data leave a participating center. Only de-identified data are sent to the Multicenter Registry server. Identifiable demographic data are not shared with other centers and are for use only at the center generating it. Each participating center can use their own data to easily and quickly get reports about their treatments at any time. De-Identified data (no PHI) are sent quarterly to the registry. At least yearly a global report is generated. Centers can then benchmark their data with the aggregated data.
Currently, there is no cost to participate. Governance is through a steering committee made up of a participant from each enrolled center. The Steering Committee sets policies & procedures and oversees the operation of the registry. The steps to joining the Registry are:
- Have your institution read the Consortium Agreement and sign a New Member Agreement
- Obtain IRB approval (either use the Dartmouth IRB as the IRB of record or seek approval at your site)
- Get REDCap access (either use REDCap at your site or use Dartmouth REDCap through a Data Use Agreement)
If your center is interested in being part of this important project to improve the data available about hyperbaric oxygen treatments contact us at: HyperbaricRegistry@groups.dartmouth.edu
New Member Agreement
Dartmouth IRB Approval for Registry
Dartmouth IRB Reliance Agreement and Request to Rely Forms
Research use of information form
Registry Information and Intake Form
Questionnaires Used in Registry
Alfred Health, Melbourne, Australia
Augusta Health, Fishersville VA
Avera McKennan Hospital, Sioux Falls, SD
Beverly Hospital Wound and Hyperbaric Medicine Center, Beverly, MA
Dartmouth Hitchcock Medical Center
DDRC Healthcare, Hyperbaric Medical Centre, Plymouth, UK
Duke University Medical Center, Durham, NC
East of England -Lhm Hyperbaric Unit, James Paget University Hospital, Great Yarmouth
Elliot Health System
Geisel School of Medicine at Dartmouth College
Hyperbaric Medicine Unit, St Richard's Hospital, Chichester, UK
Mayo Clinic, Rochester, MN
Midlands Diving Chamber, Rugby, UK
North England Medical and Hyperbaric Services, Hull, UK
Northwest Recompression Unit, Birkenhead, UK
Saint Alphonsus Health System
Spectrum Health, Grand Rapids, Michigan
St. Lukes Health System, Boise, ID
The Diver Clinic, Poole, Dorset UK
The Hyperbaric Unit, Whipps Cross University Hospital, London, UK
The Prince of Wales Hospital
UC San Diego Health
University of Pennsylvania Health System
University of Maryland Medical Center
University of Rochester Medical Center
Wesley Hyperbaric, Auchenflower, AU
White Plains Hospital, White Plains, NY
All dates are date-shifted during the deidentification process which happens when the data are exported from your site
Some IRBs treat this registry like an intervention trial and apply standards that might not be totally appropriate for a registry. They may ask for a research protocol as if this is a clinical trial or an interventional trial. Note that in the Dartmouth IRB Registry Application, the application is an "APPLICATION FOR REVIEW FOR RESEARCH OR DATABASE DEVELOPMENT INVOLVING ONLY CLINICAL HEALTH INFORMATION OR SPECIMEN USE OR USE OF SECONDARY RESEARCH INFORMATION OR SPECIMENS.” So, it wasn’t reviewed like an intervention trial and shouldn’t be considered that way.
You can choose whether to participate in long-term follow up in addition to the basic registry.
You have options. If you are nonprofit, you can choose to install redcap on a server at your center. If you want to put your data into the redcap at Dartmouth-Hitchcock then you would need a data use agreement for that.
You do not need a data use agreement to export the data from your redcap to the de-identified registry. No identifiable data goes into the multicenter registry.
We have a secure encrypted method to upload the data. In addition, the data are de-identified when exported from your redcap.
In the consortium agreement you are agreeing to collect data on all patients. If your IRB requires a consent form, then you just need to keep in mind the data completeness policy (enclosed and on website). We have a waiver of consent for the main registry and use a consent form for the long-term follow up.
No, but the steering committee has the ability to set a membership fee. If your center does not want to pay the fee you can leave the registry. You will have representation on the steering committee.
All dates are date-shifted during the deidentification process
Many centers have a waiver of consent approved by their IRB for the registry. In other words, at many centers there is not a separate consent form for the registry. Here’s the explanation for why we have for the waiver of consent
The IRB follows the established guidelines in 45CFR46.116. Briefly, the four criteria are:
- Minimal risk: The registry fits this definition because only de-identified data are pooled.
- Would not adversely affect rights and welfare of subject: The outcome measures used in the registry are all clinically useful and are worthwhile measures for monitoring the quality of care at each center, regardless of a center’s participation in the pooling of de-identified data for research. These are measures that arguably should be done in some form at each center as part of clinical care, and so are not new measures or interventions that might affect the patient’s treatment. The REDCap registry at Dartmouth has good data security measures in place and the use of de-identified data minimizes risks to patients/participants. Also, these data are important for future patients and society. We understand and support the Principle of Respect for Persons (Common Rule) which indicates each individual has the right to make decisions about the use of his/her own private information - but there is also a responsibility to future patients to assure people are receiving effective, well studied interventions. We consider the waiver of consent justified as these data will be securely managed, are not particularly sensitive or stigmatizing, and will benefit future patients and society.
- Research could not be practicably carried out without the waiver: Research looking at issues such as outcomes/ morbidity data where not having access to all subjects would affect the statistical outcome is the type of research where obtaining consent can be considered impractical. For this registry, some indications for treatment are rarely or infrequently seen at each center and omission of data, or data from only a subset of subjects, could affect the statistical outcomes and damage the credibility of the registry results. This is also true if cases that did particularly well or particularly poorly are not included. Because there are so little data in this area a 100% sample is needed to characterize actual treatment and remove the bias of only those who actively consent. These data could be collected and used for quality activities - but the need is for generalizable information that can be applied to optimize care in multiple settings.
- Whenever appropriate, subject provided with additional pertinent information after participation: Not appropriate in this case, since results in the combined database are de-identified and so cannot be linked back to a particular person from the combined database. But, each center retains the ability to identify patients after their treatment is complete which allows centers to contact their patients in the event that the study identifies a risk or benefit in the future. We expect research results using registry data to be disseminated to clinicians and hospitals and hope that this knowledge will improve care at the individual level in the future.
When you enter the data each patient is assigned a unique registry identifier (e.g. 50000-101). At your site you will keep a record of name, MRN, etc. for 50000-101 so that you will know which patient that is.
You will be entering data into the Dartmouth-Hitchcock REDCap. This requires a Data Use agreement because you will be entering dates when you are entering data. All identifiers are removed when the data are exported to the Registry at Dartmouth College. In the de-identified registry date of birth is replaced by birth year and all the dates are shifted.
Your local REDCap will have identifiers in it and is very useful for our operations. You can use the identifiers to find additional information about particular patients, run reports, etc. Importantly, none of the identifiers go into the de-identified registry at Dartmouth College.
No, this is already covered in the Consortium agreement that your center already signed. No additional data use agreement is needed with Dartmouth College.