March 2018 CLINIMINDS V. 1 / 20 2018 18 | | ISSUE 1 MOVING TOWARDS A TRANSLATIONAL PHARMACOVIGILANCE SYSTEM INSI SIDE ST E STOR ORY T E C H N O L O G Y R E G U L A T I O N S S A F E T Y Artificial Intelligence Seeing Policy Paper published by PvPI to include vector through the lens of ICMRA on Big Data borne disease as part Pharmacovigilance Analytics Pharmacovigilance for Public health Artificial intelligence may be called The working group of the ICMRA as an ability of a computer system to comprising of subject matter experts Pharmacovigilance programme of perform task that require human from European Medical Agency India (PvPI), soon to include intelligence such as cognition (EMA), Health Canada and diseases caused in tropical climate, through visual acuity, voice Medicines and Healthcare Product vector borne diseases like malaria recognition, language translation Regulatory Agency (MHRA) had and dengue, tuberculosis and HIV- leading to decision execution of a developed the policy paper with a list AIDS as part the Pharmacovigilance certain function of initiatives pertaining to the for public health. The programme implementation of big data analytics would be soon incorporated by all in pharmacovigilance. SEARN member countries . Newsletter 1
March 2018 CLINIMINDS RECOMMENDATIONS INTENDED ON MITIGATING THE ISSUES IN DRUG AE REPORTING Reporting systems must act as a mechanism to document work and share information between care providers to minimize the duplication of work. Artificial Art ial Int ntell llig igenc ence-Seeing t Seeing thro hrough t ugh the lens he lens o of Pharm Pharmac acov ovigi igilanc lance e (Cont.) Pharmacovigilance as we know is a science with a set of time, one does not need to manually execute the program pre-defined functions to collect, analyse, monitor adverse as it would be auto executed in order to accomplish the event reports in understanding the safety profile of drug. task, if presented with the exact same variables as that of the earlier scenario. The set pre-defined functions would include case processing through data entry of adverse event forms into In the second case if the task is not accomplished then too safety database, medical review, aggregate reporting, the procedure would be stored in the program and next signal detection, risk evaluation and mitigation strategies. time when the program is auto executed it would not take the same path thus minimizing error With patient’s awareness and regulatory compliance we may have seen a surge of adverse event data over last few This process self-learning through experience is called years, resulting in the urgent need for the application of machine learning. automation. Pharmacovigilance is the only discipline For example imagine a scenario where in which you have where in which timelines and quality data are evaluated received an email from a patient who has experienced on a benchmark of 100 % and a compromise in these two nausea, followed by headache and bleeding from nose on parameters are considered to be zero tolerance. lisinopril, the patient also mentions that he has a history Automation of the above pre-defined function is possible renal impairment and also that he was a chain smoker for through machine learning, which is an integral component which he took varenicline to quit smoking of Artificial Intelligence. An algorithm created on the principle of machine learning What is Artificial Intelligence? would have the capability to auto recognizes and identify the suspect drug from concomitant therapy, Artificial intelligence may be called as an ability of a adverse event from medical history and not only this, computer system to perform task that require human through robotic process automation it may also integrate intelligence such as cognition through visual acuity, voice an email function with the safety database which would recognition, language translation leading to decision enable auto data entry, preparation of auto case narratives execution of a certain function. and auto sending of emails to patients or physician for further follow up from, the safety database. Machine learning is based on reinforced data, where in which when an algorithm is executed to accomplish a This is ‘Artificial Intelligence’, a capability attained specific task. through self-learning to process thousands of data within seconds. If it accomplishes the algorithm ends and the entire procedure is auto stored in a program, which means next Newsletter 2
March 2018 CLINIMINDS Reference: An Article by on Artificial Intelligence as an Aid to Pharmacovigilance By Adam Sherlock, Christopher Rudolf (Last accessed on 10.02.2018) Automation in Pharmacovigilance Data Processing – Do You Trust Artificial Intelligence? By Dr. Vivek Ahuja, Vice President, Global Pharmacovigilance, Aris Global With automation employees engaged in manual data entry (Last accessed on 10.02.2018) would be upskilled in the execution of AI process. Poli olicy Paper Paper pub publi lished by I hed by ICMRA on A on Big D Big Dat ata Anal a Analytics ics; Experts Experts from rom EM EMA, A, Healt ealth h Canada anada & M & MHRA to A to examine t examine the o he oppor pportunities unities a and li nd limitat ations ons of B of Big Da ig Data and a and Analyti Analytics in Pharm in Pharmac acov ovigi igilan lance e (Cont.) The International Coalition of Medicines Regulatory included that all ICMRA members should be invited to Authorities’ (ICMRA) has released a policy paper which share the results of research and validation studies on big examines strengths and limitations of big data and data sources, along with real-world data (Fig. 2.0), EHR, analytics in pharmacovigilance. EMR and AHD with traditional SRS data when developed. The working group of the ICMRA comprising of subject Reference: Big Data and Pharmacovigilance: ICMRA matter experts from European Medical Agency (EMA), Working Group Looks at Opportunities and Challenges Health Canada and Medicines and Healthcare Product (Last accessed on 16.02.2018) Regulatory Agency (MHRA) have developed a policy paper with a list of initiatives pertaining to the implementation of big data analytics in pharmacovigilance. DID YOU KNOW ? ‘Big Data’ is a sub group of ICMRA working group. In 1962, Kefauver efauver Harr arris s Amen endment dment, also known as Drug ug Efficacy cacy Amen endment dment, was One of key areas in exploring the opportunities with big introduced by the Fe Fede deral al Fo Food od, Drug ug an and d data and analytics would be in the spontaneous reporting Cosmet osmetic c Act ct. This was in response to the systems (SRS) (Fig. 1.0) which currently contains limited Thali Th lido domide de trag aged edy in which thousands of structured / unstructured data. children were born with birth defects due to Data collected through voluntary reporting, often has its the consumption of Thalidomide by the own challenges, with limited information available about pregnant women for reducing their morning patient’s demographics, medical history, past drug history, sickness. This drug was prescribed without concomitant therapies, onset date and time of adverse undergoing a proper trial to determine its event, it becomes difficult to analyze the incident rate of safety. ADR and the total number of ADR occurring in a After its passage, in addition to population with patient exposure. demonstrating safety, manufacturers were The members from the expert committee of ICMRA now required to provide proof of working groups have agreed to share their knowledge in effectiveness of their drugs prior to identifying gaps and thus contributing to regulatory approval. The amendment also required harmonization. them to disclose accurate information about their products side effects Of the many recommendations made from Big-data sub group members to the ICMRA. Key recommendations ZINBR BRYT YTA A to o fac ace res restri rictions ions on on use, e, Multiple sclerosis (MS) is condition in which the immune system attacks myelin sheath, which may eventually foll ollow owing ing a a safet ety rev review iew by by EMA A damage the neurons and leave scar tissue, relapsing show howing s ing seri riou ous li liver dam er damage age multiple sclerosis is a type of sclerosis in which there is flare up following remission, which means during the Newsletter 3
Recommend
More recommend
