Clinical Trial of a New Software ENgine for the Assessment & Optimization of Drug and Non-drug Therapy in Older peRsons (SENATOR)

Clinical Trial of a New Software ENgine for the Assessment & Optimization of Drug and Non-drug Therapy in Older peRsons

A Prospective, Multinational, Randomized, Open Label Parallel Arm Trial With Blinded Outcome Adjudication Quantifying the Efficacy of SENATOR in Reducing Adverse Drug Reactions in Older Hospitalized Subjects

Clanwilliam Health, University of East Anglia, ARTTIC International Management Services, Clininfo S.A., NHS Grampian, University Ghent, Hospital Universitario Ramon y Cajal, Istituto Nazionale di Ricovero e Cura per Anziani, Landspitali University Hospital

Primary Objective: To quantify the benefits of the SENATOR decision support software on the reduction of ADR rates in older hospitalized patients. Secondary Objectives: To evaluate the effect of SENATOR with regard to use of appropriate non-pharmacological therapies in subjects with one core geriatric syndrome. Tertiary Objectives: to examine the association of SENATOR use with subject survival, morbidity and health related quality of life. Health Economic Objective: To examine the potential health economic consequences of using SENATOR. There are two study phases: Phase I: Prospective multinational, multicentre observational study to estimate the baseline adjudicated medical and surgical ADR rates by clinical subspeciality in 6 international sites. Phase II: Prospective multinational, multicentre, block randomized, two parallel arm, open label, controlled trial, with blinded outcome ascertainment, of the efficacy of SENATOR software in reducing ADRs in older hospitalized subjects.

Phase I is designed to test the electronic case report form (eCRF) and the ADR ascertainment method in the six clinical sites in advance of Phase II (randomization phase). In Phase I, we recruited 644 older multi-morbid patients from the 6 clinical sites. After obtaining written informed consent, patients' demographic, clinical and medication details were entered to the eCRF. In the event of one a 12 item Trigger List of adverse clinical events occurring, the eCRF automatically generated a Trigger List assessment proforma. The 12 items in the Trigger List included: New onset falls New onset unsteady gait Acute kidney injury Symptomatic orthostatic hypotension Serum electrolyte disturbance Symptomatic bradycardia New onset major constipation Acute bleeding Acute dyspepsia/nausea/vomiting Acute diarrhea Delirium Symptomatic hypoglycemia In addition, we have included 'Unspecified adverse event' in order to capture the wide range of well recognized ADRs associated with various medications. For example, the rapid onset of a generalized maculopapular rash in a patient with penicillin hypersensitivity would be identified as an ADR under the 'Unspecified adverse event' category. ADR adjudication in Phase I was blinded and no ADR adjudications were undertaken by the site principal investigator (PI). ADRs were defined as 'definite', probable', 'possible', 'unlikely' or 'indeterminate' according to WHO-UMC ADR causality critria. ADR severity was defined according to a modified Hartwig ADR severity scale ranging from Level 1 (trivial) to Level 7 (fatal). Consensus on ADR causality was achieved through a potential endpoint adjudication committee (PEPAC), whose members were the 6 clinical site PI's. A matrix for achieving consensus was devised, such that there was a final decision on the causality of all potential ADRs.

For outcome data, details were extracted from patients' case records to determine if trigger list adverse clinical events had occurred following randomization. These trigger list events represented the great majority of adverse drug reactions (ADRs) and were independently adjudicated by a blinded end-point committee comprised of the co-PI's, such that no co-PI adjudicated potential ADRs at his own site.

Physicians attending multi-morbid older patients i.e. with 3 or more chronic medical conditions receive a SENATOR software-generated report with advice details on potentially inappropriate pharmacotherapy and/or potentially inappropriate prescribing omissions.

Incident adverse drug reactions (ADRs). at least one likely or certain, non-trivial hospital acquired ADR.

Subjects adjudicated by the Potential Endpoint Committee as having experienced one or more probable or certain adverse drug reactions (ADRs).

Inclusion Criteria: Provision of informed consent by the patient or legal guardian/next-of-kin Age ≥ 65 years Arrival to hospital within previous 72 hours Admitted as a general medical or surgical on call patient Anticipated in-hospital stay of > 48 hours, ≥ 3 active (requiring current medication) chronic medical disorders Exclusion Criteria: Admitted under: Geriatric Medicine Clinical Pharmacology Palliative Medicine Clinical Oncology Hematology Intention of primary team at the time of subject admission to seek a Geriatric Medicine, Clinical Pharmacology or Palliative Medicine in-patient consultation Life expectancy in the opinion of the admitting clinician of < 3 months Admission directly to an intensive care unit, Admission with primary acute psychiatric illness (excluding delirium) Admission with non-accidental overdose/self-harm Anticipated immediate transfer to alternative non-participating clinical service/hospital Clinical diagnosis of acute Liver failure estimated Glomerular Filtration Rate <10 ml/min per 1.73 m2 Solid organ transplant recipients Patients with malignancy receiving systemic chemotherapy Hospitalized for elective procedure Patient was more than 24 hours in the Emergency Department under the care of a different team to that which finally is in charge of them Patients who are actively participating in another clinical trial

Dalton K, Curtin D, O'Mahony D, Byrne S. Computer-generated STOPP/START recommendations for hospitalised older adults: evaluation of the relationship between clinical relevance and rate of implementation in the SENATOR trial. Age Ageing. 2020 Jul 1;49(4):615-621. doi: 10.1093/ageing/afaa062.

Lavan AH, O'Mahony D, Gallagher P, Fordham R, Flanagan E, Dahly D, Byrne S, Petrovic M, Gudmundsson A, Samuelsson O, Cherubini A, J Cruz-Jentoft A, Soiza RL, Eustace JA. The effect of SENATOR (Software ENgine for the Assessment and optimisation of drug and non-drug Therapy in Older peRsons) on incident adverse drug reactions (ADRs) in an older hospital cohort - Trial Protocol. BMC Geriatr. 2019 Feb 13;19(1):40. doi: 10.1186/s12877-019-1047-9.

Development and clinical trial of a new Software ENgine for the Assessment & Optimization of drug and non-drug Therapy in Older peRsons (SENATOR).