MR1452 - The Invasive Dentistry – Endocarditis Association (IDEA) Study: A study of the link between invasive dental procedures and critical medical events including infective endocarditis, myocardial infarction, stroke, pulmonary embolus and spontaneous pre-term birth.

The University of Sheffield previously requested Hospital Episode Statistics Admitted Patient Care Data (HES APC) data and patient-identifiable to facilitate linkage for use in the Invasive Dentistry - Endocarditis Association (IDEA) Study: A study that aims to investigate the link between invasive dental procedures and critical medical events including infective endocarditis, myocardial infarction, stroke, pulmonary embolus and spontaneous pre-term birth. The IDEA study is a National Institute for Health Research (NIHR) funded project.

There is concern that bacteria entering circulation during invasive dental procedures (IDP) could precipitate critical medical conditions including infective endocarditis (IE), myocardial infarction (MI), stroke (ST), pulmonary embolus (PE) and spontaneous pre-term-birth (PTB). Most concern has centred on IE, a heart infection with 30% first-year mortality where oral bacteria are the causal organism in 35-45% of cases. Indeed, before 2008 it was standard care for people at risk of IE to receive antibiotics before IDP (termed antibiotic prophylaxis). The effectiveness of antibiotic prophylaxis had never been proven, and in 2008 NICE recommended its use should stop. However, the UK is the only country where antibiotic prophylaxis is not recommended for those at high-risk of IE. A recent study found that UK IE incidence has risen since 2008. Much less is known about any causal link between IDP and MI, ST, PE or PTB, but these are serious conditions with high mortality/morbidity and where there is public and professional concern about the possibility that IDP could precipitate MI, ST, PE or PTB. It is important to know if IDP precipitates them or not. The purpose of this study, therefore, is to determine if there is a link between IDP and IE, MI, ST, PE or PTB.

By looking back in time in the HES database (as far back as 2000) for IE patients only, the study team will stratify IE patients into risk-groups based on previous inpatient medical care. This will enable the team to, additionally, determine if there is an increased risk of IE following invasive dental procedures in individuals at high risk of IE, compared to individuals at lower risk for IE. Some of the diagnosis or procedure codes used to identify patients at high risk of IE may appear within the data relating to admissions several years prior to the main IE admission falling within the study period, so it is important to retrieve historical data from as far back as possible, where full data exists, in order to capture all the necessary prior admissions to be able to identify high-risk individuals (without missing any due to poor data completion). HES data has improved in completeness over time; the year 1999/2000 was selected as the earliest point at which the team could be confident that the historical data would contain all the relevant codes. This data was disseminated under a previous Agreement.

The study used HES data to identify patients who developed IE, MI, ST, PE or PTB, and routinely collected dental data to identify those patients who had an IDP in the period preceding their medical event. For patients with IE only, additional HES data from the period prior to the IE event was also retrieved, in order to identify patients who were at high risk for IE before developing the condition. Patient identifiable data has been linked by NHS Digital, and NHS Business Services Authority will continue to process this data until it is no longer deemed necessary. The University of Sheffield have not received patient identifiable data, and have only received pseudonymised HES and dental data

The study is important because if IE is linked to IDP there is potential to reduce the number of IE cases by using antibiotic prophylaxis (AP). This could improve patient safety and reduce costs to the NHS. Identifying if IDP precipitates MI, ST, PE or PTB is also important for patient safety and could allow preventative action to be taken. Alternatively, if there is no causal link between IDP and any of these conditions, patients and their clinicians (doctors and dentists) can be reassured about the safety of dental procedures, and unnecessary prevention measures could be stopped outside the UK.

However, analysis of data provided under the previous iteration of this Agreement suggests that invasive dental procedures alone are unlikely to account for the large increase in the incidence of IE that has occurred since the 2008 NICE guidelines recommended the cessation of antibiotic prophylaxis (AP). Prior to this recommendation, national guidelines recommended AP not just for invasive dental procedures but also for a number of specific medical/surgical procedures as well e.g. endoscopy, bronchoscopy, tonsillectomy, endoscopic prostate procedures, some obstetric procedures etc. It may be that the increased incidence of IE may relate to AP not being used in relation to these procedures for individuals.

Using the same study design described above the University of Sheffield would like to expand the analysis to look for any temporal association between the above listed medical/surgical procedures and the development of IE.

The data needed to fulfil this purpose was disseminated under the previous Agreement, and no further data is being requested.

The expansion of the scope of the study is important because if IE is linked to invasive dental procedures and/or any of the other procedures that were previously recommended for AP cover e.g. endoscopy, there is the potential to reduce the incidence of IE by reintroducing AP cover for those procedures where a link exists. Alternatively, if there is no causal association between any of these procedures and IE, patients and their clinicians (doctors and dentists) can be reassured about the safety of these procedures and any unnecessary prevention measures could be stopped.

To address the GDPR principle of data minimisation The University of Sheffield have only requested data for a select number of ICD-10 codes. Further to this, while identifiable data has been requested to facilitate linkage by NHS BSA, The University of Sheffield will only directly process pseudonymised data.

The University of Sheffield is the sole data controller and also processes the data. NHS BSA is a data processor who have been contracted by the University of Sheffield, and they do not make decisions in regards to how the data is being processed.

The team's research collaborators at Taunton and Somerset NHS Foundation Trust, Guy’s and St Thomas’ NHS Foundation Trust, and The Carolinas Healthcare System (North Carolina USA) fulfil an advisory role only, providing valuable input during the selection of ICD-10 codes to use for condition identification, advice on the analysis of causal organism for infective endocarditis cases, (including identification of likely oral organisms and their associated ICD-10 codes), and additional clinical perspective on the study. Research collaborators do not have an active decision-making role, and will not have any access to the data.

In relation to GDPR, special category personal data are being processed under Article 6(1)(e) and Article 9(2)(j) as a task in the public interest as the evidence will inform guidelines and thus medical- and dental- practice around antibiotic prophylaxis ahead of medical and dental procedures, potentially improving patient safety and reducing costs to the NHS.

In relation to the sharing of identifiable patient health data, without consent, by NHS Digital to be processed by NHS BSA, this was addressed by NHS HRA’s approval under Section 251 of the National Health Service Act 2006 (as amended), NHS HRA CAG ref: 17/CAG/0076.

NHS BSA and NHS Digital have facilitated the linkage of national data on courses of dental treatment (NHS Business Services Authority Dental Information Services database) and hospital admissions for infective endocarditis (IE), myocardial infarction (MI), stroke (ST), pulmonary embolus (PE) and spontaneous pre-term birth (PTB) (Hospital Episode Statistics (HES) database) to investigate if there is a link between invasive dental procedures and the development of these conditions. Under this Agreement the University of Sheffield wish to extend this investigation to assess whether there is a relationship between IE and other medical or surgical procedures where AP was previously recommended.

NHS Digital identified within the HES APC data dataset all admissions with a diagnosis (primary or secondary) of IE, MI, ST, PE, or PTB between 1st April 2010 and 31st March 2016. NHS Digital then undertook an internal linkage exercise with data from PDS, to produce a full set of patient identifying information for these admitted patients (NHS number, surname, forenames, date of birth, gender, full address, postcode). From the full HES and patient identifiable data, NHS Digital created two datasets, linked by a unique study ID (encrypted HES ID) for each patient. The HES Identifier (ID) is a pseudonymised identifier which is consistent across the HES products for each patient and can be used to replace the identifiable fields. The two datasets are described below.

Dataset 1: Contains a full set of patient identifiers for all study patients, and their encrypted HES IDs, but no HES clinical data. This dataset has been sent to NHS Business Services Authority (NHSBSA) Dental Information Services in Eastbourne, England. NHSBSA used the supplied patient identifiers to identify study patients within their own database, and retrieve all their dental treatment records from 1st April 2009 - 31st March 2016. Included in these records will be the date of any course of dental treatment and whether the treatment included any extraction, endodontic treatment or a scale and polish. NHSBSA created Dataset 3 by removing the patient identifiers from the retrieved dental treatment records (but retaining the encrypted HES ID provided by NHS Digital). NHSBSA have now sent Dataset 3 (pseudonymised dental treatment records) to University of Sheffield research team, and will securely destroy their copy of Dataset 1 once work has been complete and will notify NHS Digital once this has been done.

Dataset 2: Contains clinical and operational HES data for all inpatient admissions between 1st April 2010 and 31st March 2016, where at least one of the diagnosis fields contains one of a list of specified codes indicating IE, MI, ST, PE, or PTB. It will also contain clinical and operational HES data for all inpatient admissions (with any diagnosis), between 1st January 2000 and 31st March 2016, relating to those patients from within the group identified above whose diagnosis code indicated IE; this is for the purpose of identifying patients who were at higher risk for IE. All records will include the patient's encrypted HES ID, but no patient identifying information. This dataset has now been disseminated by NHS Digital and is now in the possession of the University of Sheffield research team. The University of Sheffield research team linked Dataset 2 (pseudonymised HES inpatient data) to Dataset 3, (pseudonymised dental treatment data provided by NHSBSA for the patients identified in Dataset 1), using the encrypted HES ID for each patient. This data, has been, and will continue to be used for the study analysis.

A case-crossover study design has been used allowing the researchers to compare the frequency of invasive dental procedures in the 3 months immediately preceding an IE (MI, ST, PE or PTB) diagnosis (cases) with the frequency of invasive dental procedures in earlier 3-6, 6-9 and 9-12 month periods (matched control periods).

The research team at the University of Sheffield will continue to conduct the following analyses, these are similar to the analyses used to address the original purpose:

~Similarly, a case-crossover study comparing the frequency of courses of dental treatment NOT involving an invasive dental procedure (cases) in the first 3 months preceding an IE (MI, ST, PE or PTB) diagnosis with their frequency in earlier 3-6, 6-9 and 9-12 month periods (matched control periods).

~ The University of Sheffield will also perform a case-control study comparing the frequency of courses of dental treatment involving (cases) and NOT involving (controls) an invasive dental procedure in each 3 month period preceding an IE (MI, ST, PE or PTB) diagnosis.

~The researchers will also conduct statistical analysis comparing monthly disease free status following a course of dental treatment involving an invasive dental procedure with disease free status following a course of dental treatment NOT involving an invasive dental procedure.

~If a link exists, researchers will also stratify the data by type of invasive dental procedure to assess the relative risk of IE (MI, ST, PE or PTB) associated with each type of invasive dental procedure (extractions, scale and polish or endodontic treatment).

~The research team will repeat the above studies restricting our analysis to those IE cases that arose in individuals who would have been considered at ‘high-risk’

Identifiable patient data sent by NHS Digital to NHS BSA has only been used to identify dental records in their dental records database belonging to the cohort identified by NHS Digital. The research team at the University of Sheffield have not received any patient identifiable data; and have only received Datasets 2 and 3 (pseudonymised datasets).

No other data will be linked at patient level. There will be no requirement or attempt to re-identify individuals within Datasets 2 & 3.

Dataset 2 and 3 are provisioned on a secure virtual environment running on University of Sheffield owned and managed infrastructure located solely in England. Analysts will conduct all data processing including the statistical analyses to fulfil the objectives of the research on this secure virtual environment. Only aggregated data / outputs, with small numbers suppressed in line with the HES analysis guide, will leave the secure virtual environment.

The data will be accessible only to those directly involved in the data processing, all of whom will be substantive employees of the University of Sheffield and have been appropriately trained in data protection and confidentiality.

The identifiable patient information sent to NHS BSA to facilitate linkage is stored in accordance to the NHS BSA Data Handling and Storage Policy. The identifiable data received from NHS Digital is stored in an environment suited to its format and security classification. Physical access to the data will be restricted, and access will require mechanisms such as password-protection. All those accessing the data are substantive employees of NHS BSA, and have received adequate training in data protection and confidentiality.

The data needed to investigate any association between non-dental invasive procedures, e.g. endoscopy, bronchoscopy, tonsillectomy, endoscopic prostate procedures some obstetric procedures etc. and IE, resides within Dataset 2 (already described in Section 5b). The University of Sheffield intend to repeat the analyses described above but use the frequency of endoscopy, bronchoscopy, tonsillectomy, endoscopic prostate procedures in place of the frequency of invasive dental procedures.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract ie: employees, agents and contractors of the Data Recipient who may have access to that data).

All outputs will only contain results in highly aggregated format and as statistical summaries and measures of association. Small numbers will be suppressed in line with the HES Analysis Guide. Record level information will not be released to any third party.

The data will be presented at appropriate Cardiology and Dental scientific meetings and conferences, and in peer reviewed journals and a final report.

Published outputs will enable the data generated by this study to add to the body of evidence which is used to inform policy decisions, (e.g. of NICE), guidelines and care practice around the use of antibiotic prophylaxis for invasive dental procedures. NICE and other international guideline committees will only evaluate peer reviewed, published data when setting or revising their guidance. Hence the team's priority is publication in high impact peer reviewed journals. Nonetheless, if the team's findings are of urgent or critical importance, they will contact relevant guideline committees to give them advance notice of publication, and offer access to the data.

The final report to NIHR is mainly for the benefit of the funder, (NIHR), who will normally make the final report publicly accessible and usually bring important findings to the attention of interested public, patient and professional and groups.

Presentation of data at scientific meetings and conferences will increase awareness and dissemination of the results amongst the relevant professional and scientific communities, enable the team to test their results against peer opinion, and provide an opportunity for contesting of the results by interested clinical and scientific colleagues.

The data relating to the link between invasive dental procedures and IE the team will aim to publish in a major general medical journal such as the Lancet, New England Journal of Medicine, JAMA or BMJ alternatively, if that doesn’t prove possible, then a major international cardiology journal e.g. Circulation or the European Heart Journal. The anticipated date for first manuscript submission was September 2020. However, publications have now been delayed due to delays in receiving the data caused by the impact of COVID-19. Submission is now expected in 2021 and publication in 2022.

The team had also intended to target presentation of the data at a major cardiology meeting (American Heart Association (AHA) meeting in November each year or the European Society for Cardiology (ESC) meeting in August each year) in late 2019 or 2020. The team also considered presentation at the International Association for Dental Research (IADR) meeting in 2020. However, these meetings were delayed due to the COVID-19 pandemic, and these outputs are now expected to be produced in 2021.

The team will present the data on the link between invasive dental procedures and MI or stroke at an IADR meeting and publish in either a dental (possibly the Journal of the American Dental Association (JADA)) or cardiology journal (Circulation of European Heart Journal).

Data on any link between invasive dental procedures and pulmonary embolus or spontaneous pre-term birth will be presented by the University of Sheffield at an IADR meeting, and will most likely publish in JADA. These data will be analysed after the main IE study and so presentations will likely follow in 2021-2 and publication in 2022-3.

For this amendment the publication strategy remains unchanged although delays in receiving the data from NHS BSA and further, and continuing delays caused by the impact of Covid-19 mean that the target dates for outputs have been delayed by around 18 months.

The study has benefitted from patient and public involvement. In order to determine whether the study question was sufficiently important, the study was taken to two well established PPI groups jointly organised by the University of Sheffield and Sheffield Teaching Hospitals NHS Trust for evaluation and comment:

(i) The Cardiovascular Research Patient Panel. This group consists mainly of patients who have suffered from a spectrum of cardiovascular diseases or undergone cardiothoracic surgery. This group has been established for several years, meets approximately once every two months and has considerable expertise in assessing research projects.

(ii) The Dental Research PPI group. This group has been in existence for approximately 6 months and therefore has less experience than the Cardiac group. It comprises patients who mainly attend the Charles Clifford Dental Hospital in Sheffield for dental treatment of one sort of another.

Both groups fed back their view of the importance of the study, the importance of the research questions being asked and the potential value of the study to the NHS. The PPI groups were specifically asked about any potential ethical or other concerns they might have about the study - particularly relating to the linkage of individual data from two national data bases. However, the patients were very happy with the arrangements for NHS Digital to link the two datasets and provide only pseudonymised data.

The study also intend to set up a Project Advisory Group who will have oversight of the project and independently ensure that it is achieving its aims, making good progress and meeting its deadlines.

If the research confirmed an association between invasive dental procedures and IE, it is anticipated that this would be seen as further validation of the recommendation of the European Society for Cardiology, American Heart Association, and other international guideline committees advising that dentists give Antibiotic Prophylaxis (AP) before performing invasive dental procedures on individuals considered at high risk of developing IE. It would also provide further evidence that the current NICE guidelines, recommending no AP, may be wrong, could be putting patients at risk, and should be reconsidered.

Alternatively, if the research demonstrated no association between invasive dental procedures and IE, it would support the recommendations of NICE and provide evidence that the recommendation of other guideline committees around the world - to give AP - could be wrong and should be reconsidered. Such an outcome would also suggest that rather than focussing on AP to prevent the ~45% of IE cases caused by oral bacteria, prevention strategies should instead re-focus on improving oral hygiene in those at risk of IE in order to reduce the small but continuous risk that likely results from daily activities such as tooth brushing, flossing and chewing food, particularly in those with poor oral hygiene.

Whatever the outcome, therefore, the results of this study are likely to impact the guidance given by international guideline committees and therefore the care given by Dentists and Cardiologists for individuals at risk of developing IE.

Currently, although concern has been raised about the possibility that invasive dental procedures could precipitate MI, ST, PE or PTB, the real risk of this is not known and no preventative measures are advocated. A clear demonstration that there is no link between invasive dental procedures and MI, ST, PE or PTB would provide reassurance to clinicians and patients about the safety of dental procedures. Alternatively, if a link were demonstrated between any of these conditions and invasive dental procedures, research could be initiated to identify ways of reducing or eliminating this risk.

The expansion of the scope of the study is important because if IE is linked to invasive dental procedures and/or any of the other procedures that were previously recommended for AP cover e.g. endoscopy, there is the potential to reduce the incidence of IE by reintroducing AP cover for those procedures where a link exists. Alternatively, if there is no causal association between any of these procedures and IE, patients and their clinicians (doctors and dentists) can be reassured about the safety of these procedures and any unnecessary prevention measures could be stopped. As previously mentioned with regard to any findings about the association between invasive dental procedures and IE, any findings regarding any association between invasive medical/surgical procedures e.g. endoscopy, bronchoscopy, tonsillectomy, endoscopic prostate procedures some obstetric procedures etc. and IE would likely have an important influence on the recommendations of national and international guideline committees. The impact of this data on clinical practice would most likely come into effect sometime in 2022/3. Health and Social Care in England will benefit from greater evidence for informing guidelines and thus medical- and dental- practice around antibiotic prophylaxis ahead of medical and dental procedures, potentially improving patient safety and reducing costs to the NHS.

There are no yielded benefits thus far. The receipt of dental data from NHS BSA was delayed, with further disruption caused by the COVID-19 pandemic. For these reasons the analyses are still ongoing.