Skip to main content

P0458 - Penicillin allergy status and its effects (ALABAMA)

Penicillin allergy status and its effect on antibiotic prescribing, patient outcomes, and antimicrobial resistance (ALABAMA)

Study Overview

The study will evaluate the cost and health outcomes of pre-emptive penicillin allergy testing in patients with a penicillin allergy label. This study is motivated by reports from previous studies that about 9 out of 10 people who have a record of penicillin allergy are found to not have a true allergy when thoroughly tested. This means they could safely take penicillins. The aim of ALABAMA is to find out if people with a penicillin allergy label in their GP health records really do have an allergy by carrying out specialist testing. The results of our study would allow us to know how much providing this service in routine primary care practice would cost the NHS and improves health related quality of life of patients.

We are recruiting approximately 800 patients to be randomised between two treatment options,  one is to receive a penicillin allergy test and the other is to receive usual care, which does not include penicillin allergy testing. Patients in participating GP surgeries are contacted by their GP practice after a preliminary check of their GP electronic health record and identification as being potentially eligible. That is, as being aged 18 years or above, having a current penicillin allergy (or sensitivity) record of any kind in their electronic health record and been prescribed systemic antibiotics (either: penicillin, cephalosporin, tetracycline, quinolone, macrolide, glycopeptide, aminoglycoside, oxazolidinone, monobactam or carbapenem class antibiotic or fosfomycin, nitrofurantoin, trimethoprim, clindamycin, rifampicin, colistin, metronidazole) in the previous 24 months. If the patient is willing to take part, or wishes to find out more information, they notify the ALABAMA research team at the University of Oxford Primary Care Clinical Trials Unit. The research team then arranges either a face to face or telephone appointment for the patient to talk to their GP, or a delegated member of staff, at a time that is convenient to the patient. At this appointment, the GP, or delegated member of staff, checks that the patient is eligible to take part in the trial (including a review of their penicillin allergy history), and check that the patient understands what being in the trial involves. Patients who confirmed as eligible and want to proceed are then asked to give their consent to participate in the trial.

Following this, a member of the research team calls the patient to ask a series of questions, review their medical history (including penicillin allergy history), check understanding of their penicillin allergy, and complete a quality of life questionnaire. Patients are then randomly allocated to either usual care or the Penicillin Allergy Assessment Pathway (PAAP) trial group.

Patients randomised to the usual care group, are not required to attend any further visits for the purpose of the trial. However, if they are prescribed antibiotics for a list of pre-defined infections during their participation in the trial, they are asked to complete a symptom diary for up to 28 days for their first prescription of antibiotics since trial enrolment.

Patients randomised to the PAAP trial group are asked to attend one clinic appointment at a hospital allergy clinic. During the clinic appointment, they receive penicillin allergy testing in a three-stage process. At the first stage, the doctor or nurse asks the patient about their penicillin allergy history. Depending on their history they will either have a skin test or go straight to the ‘oral challenge’ test. The skin test involves a tiny amount of penicillin being injected into the patient’s arm, just under the surface of the skin which takes about 5 minutes. After 15 minutes, the skin is examined to see if the test is positive. Sometimes a second skin test is needed which takes about 45 minutes. If there is no reaction, the patient is moved on to the third stage, the oral challenge test. The oral challenge involves taking a syrup solution containing penicillin. The patient is asked by a nurse or pharmacist (with oversight by doctor, either in clinic or by phone) to take either three increasing doses of penicillin or take one full dose (dependent on allergy history). After each dose the patient is monitored closely and, if they have no reaction at this time, they are asked to take penicillin at home for 3 days to check there is no delayed reaction. A research nurse tells the patient what to do if they experience any symptoms and  calls them 4-6 days after their appointment, to check for delayed reactions.

Tested patients are receiving the results of the PAAP test via letter. In addition, patients who test negative for penicillin allergy receive a leaflet explaining the result. Their GP surgery is informed of the test result and updates the patients’ electronic health records if needed.

The first time trial participants are prescribed an antibiotic for a list of pre-defined infections by their GP during the trial period, they are asked to complete a daily symptom diary for up to 28 days when they start taking their antibiotic. They can stop completing the diary before 28 days if they feel better and have completed their antibiotic course. In addition, every time patients are prescribed an antibiotic for a list of pre-defined infections by their GP during the trial period, they are contacted by the research team to support their completion of some questionnaires at 2-4 days and 28-30 days after their GP appointment.

After 12 months a member of the research team is calling all patients in the trial to ask them to complete a quality of life questionnaire.

The active trial period is from the time patients are randomised until the expected trial assessment end date of December 2023. During the active trial period the research team is reviewing the health records of trial patient participants annually for any information on infections, antibiotic prescriptions and hospital admissions.

Participating patients have consented to our collection of their data on any infections, antibiotic prescriptions for a list of pre-defined infections and hospital admissions they may have for an additional 10 years beyond the end of trial assessment period. To do this we are accessing their medical records and routine health data held by NHS England. For this period of extended follow-up period we are not contacting patients any more.

We are using the data from trial participants on their use of primary care services and health related quality of life outcomes (recorded in the trial), and their hospital attendance and any mortality data (accessed through a data request to NHS England) to compare the outcomes of PAAP testing with usual care over 12 months. We are also using a mathematical model to predict their costs and health outcomes after 12 months and up to 5 years and evaluate the cost-effectiveness of PAAP testing relative to usual care.

Lawful basis & access to data

The lawful basis for obtaining the data is consent, according to GDPR article 9(2)(j) and 6(1)(e). Dr Ruben Mujica-Mota, Dr Miaoqing Yang, Dr Dan Howden, Dr. Rebecca Bestwick will have access to the data. The data will be accessed for approximately 9 months until 30/09/2024, when it will be archived and retained as such until being destroyed on the 30/03/2025.

Data to be obtained

Hospital Episode Statistics (HES), mortality and NHSBSA data will be obtained from NHS England for patients enrolled in the ALABAMA trial, from January 2021 to December 2023 (inclusive). Data collected in the trial for the same patients will also be obtained for the period January 2021 and until December 2023, from the research unit (Primary Care Clinical Trials Unit at the University of Oxford) conducting ALABAMA.

No directly identifiable data will be obtained.

The following special category personal data will be obtained:

  • Penicillin allergy testing receipt (Skin testing and/or direct oral challenge with penicillin) and outcome
  • Adverse reactions to penicillin allergy testing requiring medical treatment and investigations performed
  • Number of antibiotic prescriptions up to 24 months prior to randomisation in ALABAMA
  • Name, dose and duration of antibiotics prescribed in primary and secondary care (from note reviews) after patient joined the study (i.e. randomisation) and up to December 2023 (end of study date)
  • Number of registered diseases in the Quality and Outcomes Framework at randomisation
  • Patient health related quality of life at and 12 months after randomisation
  • For patients with an infection who are prescribed an antibiotic in primary care, patient health related quality of life information at days 2-4 and 28-30 after antibiotic administration
  • Hospital inpatient admissions (including diagnosis and procedures codes) after randomisation and up to end of study date
  • Attendances to Accident & Emergency after randomisation and up to end of study date
  • Outpatient attendances after randomisation and up to end of study date
  • Survival (number of days between randomisation and death, or end of study period date, whichever is earlier)
  • For patients testing negative to penicillin allergy, whether the GP record has been amended at 3 and 12 months after randomisation

Non special category data:

  • Age in years when enrolled in ALABAMA trial
  • Gender
  • General practice code
  • Measure of deprivation for area where resided at the time of enrolment in ALABAMA trial (categories 1-5 or missing)
  • Random allocation to penicillin allergy testing (‘penicillin allergy assessment pathway’) or usual care in ALABAMA trial

Application of data protection principles

  • The lawful basis for the analysis of the data is public task, which is consistent with the collection of the Hospital Episode Statistics data by NHS England. We wish to use the data to address a question which has important implications for healthcare services and which could lead to improvements in public health policy.
  • We have satisfied the requirements of the NHS England and of a NHS Research Ethics Committee and the data will be used fairly and transparently. A protocol has been prepared before starting our research project and the types of data which will be obtained have been declared in this privacy notice. We intend to publish our findings in open access peer reviewed literature.
  • We have made particular efforts to obtain only the data which we require. Therefore no directly identifiable patient data is included, for example we will not have access to names or addresses of patients or their GP practices.
  • We have only requested data from the University of Oxford Primary Care Clinical Trials Unit and NHS England which we need in order to undertake the planned analyses.
  • The accuracy of the data will be ensured through the quality assurance processes at the University of Oxford Primary Care Clinical Trials Unit and NHS England
  • Data will be deleted by 30 March 2025 by the University of Leeds Institute of Data Analytics team
  • The data will be held securely on the University of Leeds’ LASER platform with password protected access available for the health economics research team (Ruben Mujica-Mota, Miaoqing Yang, Dan Howdon and Rebecca Bestwick) only
  • Appropriate records will be kept of the analyses undertaken and a plan has been made for how iterations of the data files will be named

We do not know of the identities of any individuals on whom data will be obtained so it will not be possible either to inform individuals that their data is being processed or for such data to be removed. However individuals who have opted not to have their data shared with the University of Oxford Primary Care Clinical Trials Unit and with NHS England will not be included in the data we have obtained.

For further information on data protection regulations please refer to the Information Commissioner’s website https://ico.org.uk/

If you wish to contact the study team or make a complaint please email Dr Miaoqing Yang m.yang3@leeds.ac.uk or Ruben Mujica-Mota (lead researcher) r.e.mujica-mota@leeds.ac.uk