Medical Review of Individual Case Safety Reports

An issue that has been, and continues to be, an area of some controversy is what sort of medical review of individual case safety reports (ICSRs) should be done?  Some of the comments heard around the industry and during audits and inspections I have heard include:

“There’s no requirement for a medical review in the regulations”.

“We have a nurse or pharmacist look at all the cases”.

“Only expedited reports should be seen by a physician”.

“Generic drug ICSRs don’t need a medical review”.

“Non-serious cases are not important enough to require any sort of medical review”.

“The case is processed by a medical professional who is fully capable of doing the medical review while handling the case”.

“Physicians need only look at clinical trial cases”.

“We review the cases at the time of the preparation of the PADER, PSUR”.

“Non-serious expected cases don’t need medical review.  In fact they aren’t even submitted to FDA”.

So let’s have a look at the situation.

To my knowledge there is no actual regulation in the US, EMA or MHRA regulations that requires a physician to review some or all cases that are received by companies in the post-marketing setting.  Although someone in the company may be required to be responsible for safety and safety reporting in the eyes of the law (e.g. the QPPV in the EU) this person need not necessarily be a physician and need not necessarily review each case.  Similarly, CIOMS (particularly CIOMS V) does not seem to address this question.  The issue here revolves primarily around spontaneously reported cases.  Other adverse events, such as those from clinical trials, solicited post-marketing cases, patient support cases etc. usually are reviewed by medical personnel.

Having said that, let’s look at the need for a medical review of a post-marketing case. In the post-marketing setting, cases may come in from a variety of sources: patients, consumers, family of patients, health agencies, lawyers, poison control centers, literature, physicians (“medics” as they are sometimes known in the UK), nurses, pharmacists, dentists, podiatrists and other health care professionals (HCPs).  The bulk of the cases come in from health care professionals and patients in most countries.

For cases reported by HCPs there is, obviously, a “medical review” done by the reporter.  For cases not coming in from HCPs, for many years countries discounted these cases unless they were medically confirmed (which is defined as being corroborated to some degree by an HCP).  Nowadays, non-HCP reports are usually databased and handled like other spontaneously reported cases.

In my experience in the industry as a drug safety professional responsible for safety and, now, as a consultant doing safety and PV audits and gap analyses around the industry, there is a surprisingly large variation in the actual day to day medical review practices.  Here are a few generalities:

• Big pharma companies usually have one or more physicians reviewing cases and many (dozens to hundreds) of medical professionals around the world handling ICSR case processing, medical information and related areas.  In these companies, all cases are seen by medical professionals.  In some companies all cases (serious, non-serious, expected, unexpected, clinical trial, spontaneous, stimulated etc.) are also reviewed by a physician.
• Small and mid-size pharma companies either out-source the safety function and may or may not do a medical review in the company on the out-sourced cases.  When it is done, it may or may not be done by a physician.
• Generic companies may or may not have a physician or medical professional do a medical review.  In fact, some generic houses do not even have a physician on staff.
• Companies that do not have physicians on staff may or may not have a physician as a consultant to review selective cases or handle one-off issues.
• CROs usually have a physician and almost always have HCPs on staff.  The review that is done depends upon what the contracting sponsor/pharma company’s wishes.  Sometimes a medical review is done; sometimes not.  Sometimes by an MD; sometimes not.  It is often decided by cost as the MD review increases the cost per case.
• Health authorities in all cases in my experience have HCPs handling drug safety cases.  However, especially smaller countries with very small Ministries of Health and even smaller pharmacovigilance units, often have no physicians on staff and relay on pharmacists primarily and sometimes nurses.  A physician may or may not be available on a consultative basis.

In other words, pretty much every situation is possible and occurs.

Before commenting further on whether a physician should do a medical review, let’s take a look at what the medical review of an ICSR should include, no matter who does it:

• Accurate transcription of the data from the source documents if there is no direct electronic, computer to computer data transfer.  That is, are there databasing or keying errors, misspellings (particularly of drug names) etc. Are the dates of drug administration correct etc.
• Are the company judgments on seriousness and expectedness correct?
• When required and if done at this stage of case processing, is there a causality assessment and does it make sense?  Is it medically plausible? Is there a temporal relationship? Are there confounders, intercurrent conditions, comedications etc.
• Does the ICSR accurately reflect the medical information in the source documents?  This determination requires medical knowledge.
• Is the coding (MedDRA and drug names) correct?  For MedDRA codes some level of medical knowledge is required for accurate coding as well as familiarity with MedDRA conventions, hierarchy etc.
• How are ambiguity and missing data handled?  Again a medical judgment may be needed here.
• Does the narrative read well and make medical sense “on its face”.  That is, without looking at the source documents, does the narrative convey the essence of the case correctly?
• Has the appropriate follow-up been done? Have the right medical questions been asked and are the answers adequate?

Clearly, for complex or equivocal cases, a physician with experience with the drug and with the disease/indication is necessary.  Sometimes several physicians must weigh in and even then a clear judgment may not be possible.

And, to be very fair, some physicians are not very good and do a poor job and some nurses and pharmacists are absolutely superb at this.

So what should be done? What are the possibilities?

• MD review of all cases: In practice, one of the key issues is the avoidance of late expedited 15-day reports submitted to FDA and other agencies.  Late reports can occur from a variety of reasons but the one that touches on medical review is primarily that of an ICSR that is classified as non-serious or serious and expected and that, in fact, is serious and unexpected.  If these cases are misclassified and not picked up as expeditable until the time of the PSUR/PADER preparation, months or up to a year or more later, there will be a very late expedited report to a health agency.Thus some companies, just to be sure, have a physician review all post-marketing (or at least all spontaneous cases) for seriousness and expectedness.  This often does not include a review of the source documents but simply of the MedWatch (or E2B electronic file) or line listing.  Clearly it depends on how conservative and cautious the company wishes to be.  At the most conservative end, every case has a complete review including source documents.  At the other end, a line listing review monthly or quarterly is done.
• HCP review of all cases, MD review of selected cases:  Similar to the previous bullet point, some companies have all cases reviewed by an HCP – either the HCP in drug safety processing the case or another HCP who did not handle the case (a quality review) and only selective cases are reviewed by a physician.  These selective cases are often “Cases of Interest” or cases on the FDA “Always Expedited List” or AEs/signals for that particular drug etc. Similarly the level of review can be thorough or a simple line listing review.
• HCP review of selected cases, no MD review:  Already getting into a dangerous zone whereby something could be missed.
• No HCP review: As noted, this is sometimes the case in very small companies or generic houses.  It is not a wise avenue to take.  In particular, with the upcoming likely change in the requirements for generic drug labeling in the US, whereby the generic labeling no longer is necessarily identical to the branded labeling, a safety and pharmacovigilance review in the generic companies will be a wise, if not obligatory, requirement.
• CRO review and “hoping for the best” or no review at all: Some companies do not pay attention to what is done by the CRO they have out-sourced the pharmacovigilance function to.  This is most unfortunate and rare but has been seen.  Other companies do minimal pharmacovigilance and no particular medical review.  Obviously not advised.

Final thoughts and recommendations:

For those who read these postings regularly, you know that my views are quite conservative.  My view in this area is a simple one: All cases should be reviewed by a physician, usually looking at the source documents and doing a serious quality review after processing and review by an HCP.

One step down would have this thorough MD review done for all serious cases and all non-serious unexpected cases.  I would still be nervous and would want the non-serious expected cases reviewed periodically (at least as a line listing).

Finally, an HCP review of all cases with selected cases going to an MD for review would be my third and “least bad” choice.