Stroke Risk Increases Risk for Memory Loss
Jan 19th
A recent study in Neurology found a link between the risk of having a stroke and future development of memory loss or cognitive decline. The link was strongly tied to high blood pressure. The study, called REGARDS, involved 23, 752 older adults, average age 64, and followed them for 4 years. In those with a higher risk of stroke at the start of the study they had a higher incidence of cognitive issues. For every 10mmHg increase in systolic blood pressure (the upper number) there was a 4.1% increase in the risk of developing cognitive issues.
The upper number of blood pressure, also called systolic, is very important in older adults as it tends to rise with age, which was another risk factor from this study that was correlated with stroke risk. In my experience with helping older adults it is common to find isolated systolic hypertension, occurring in about 30% over the age of 80. Isolated systolic hypertension is where the upper number is elevated but the lower (diastolic) number is normal. Treatment of this type of high blood pressure is very effective at preventing strokes in which only 13-18 people need to be treated to prevent one stroke. What is also important to note is that “silent strokes”, in which you have very small strokes that go undetectable but contribute to cognitive decline, can be reduced by better management of your high blood pressure.
My advice, pay attention to your blood pressure and if that upper number is higher than 140, see your doctor and start a conversation. Also keep in mind that self-monitoring of your blood pressure and sharing those recordings with your doctor is more likely to catch high blood pressure, and is also known to be more accurate than the occasional blood pressure check in the doctor’s office.
Patient Reported Symptoms from Crestor
Jan 19th
Here’s a link from AskAPatient.com, a consumer-based website that allows people to post what they believe are side-effects from their medications. Although there is no validation that reported side-effects are always from the medication, there is usually a trend that can be followed along with some activity by the person helping to point the finger of suspicion at a particular medication. One way to strengthen an argument that a medication is responsible for a life-altering side-effect is to stop it, and then monitor how you feel and function. This report is on Crestor, not that I’m picking on Crestor, but it speaks well to how the medication is most likely responsible for such severe pain that it prevented this 86 year old women from being able to care for herself. In the final analysis, one would want to consider restarting the Crestor and see if the same side-effect returns. If it does, then we have further strengthened the argument that the drug is responsible. These reports also strengthen the argument that people with changes in how they feel and function, especially after starting a new medication, should report them to their physician and not take no for an answer until there has been a rigorous process to verify if the medication may be responsible. A simple shrug of the shoulders or a “I think not” reply should not suffice.
http://www.askapatient.com/viewrating.asp?drug=21366&name=CRESTOR
The Risk of Determining Risk with Multivariable Models
Jan 12th
Here is an article that reviews the use of statistical applications for determining risk which proves there is a significant misuse of statistics in the medical literature thereby leading to inaccurate conclusions. The importance of this subject matter can not be overstated since there is an abundance of published studies in which practitioners take the authors conclusions for granted and apply those conclusions to their daily practice, yet the stated conclusions may be false.
My quest for a solid explanation started some time ago when reported findings from a couple studies made no sense at all, and I read the statements of “..after adjusting for confounding variables or risk factors..”. I first verified my suspicion in Dr. James DeMuth’s book on pharmaceutical statistics where he states we can not adjust for confounding variables and assume the conclusion is a direct cause and effect. All we can state is a correlation exists, but some authors seem to go much farther in drawing conclusions. In this article, by Dr. John Concato, et al, published in Annals of Internal Medicine, 1993;118:201-210, reviews common problems with the use of multivariable analyses, the first being the over-fitting of data, in which too few sample outcomes are applied to a model, and then correlations drawn from those too few outcomes. Other problems involve non-conformity to a linear gradient, for example, the impact of left ventricular ejection fraction on negative outcomes not being linear, in terms of risk, as it is lowered drastically depending upon where the initial ejection fraction was measured at baseline, e.g. a reduction from 40% to 30% has much lower risk as compared to a reduction from 25% to 15%. Another problem is what the author says is a “violation of proportional hazards”, in which the risk or hazard of an independent variable is assumed to be constantly proportional. This false assumption is probably where I became suspicious, that a simple “adjusting for confounding variables” assumes all variables to be constant and exist in a simple relationship of direct correlation. The main point I make is that many statistical analyses have limitations and become more limited in their usefulness when they are not applied correctly thereby leading to incorrect conclusions.
Isn’t it true, that on any given day, we can read the summary of a research article which states that a treatment is associated with a negative outcome, after having adjusted for confounding factors, etc. And then one month later we read the opposite results from another study. It begs the question, which article is most accurate in their conclusions, if either? The use of statistics in medical research has accelerated in the last twenty years but not without concerns over the quality of its applications. I, like most human beings, tend to take things at face value. But as more findings hit the news I find myself having to employ discipline and not form any opinion until a detailed analysis of the study can be made to determine if, in fact, the purported conclusions have any meaning at all. The other application of this principle is for the lay person, no matter how educated, do not react to any information in the news.
Trusting What Your Patients Tell You: Detecting Adverse Events in the Office Setting
Dec 30th
There are several published articles that have caught my attention that focus on perceived and self-reported adverse events in the patient population. One article, Adverse Drug Reactions in Elderly Patients as Contributing Factor for Hospital Admission: cross sectional study, Mannesse, C., et al, BMJ Vol 315, Oct. 1997. They reported that people were reliable in detecting adverse drug events in themselves, with a correct opinion in 73 out of 93 who answered the question whether they were experiencing an adverse drug event (ADE). However, these were adverse events that were less severe, and the more serious adverse events that caused hospitalization, such as gastrointestinal bleeding, were not well detected. In another study, Patient-reported Medication Symptoms in Primary Care, Weingart, S, MD, et al, ArchIntMed, Vol 165, Jan. 24, 2005, the authors measured how many people thought they were experiencing an adverse event (179 people & 286 events) and what percentage reported them to their physicians, along with what percentage were acted upon by their physicians. Only 69% were reported to their physicians, and only 76% were acted upon. The authors estimated that a failure to report symptoms led to 19 (21%) ameliorable ADEs, and 2(2%) preventable. A failure to change therapy in 48 cases led to 31 (65%) ADEs that authors estimated to be ameliorable. The conclusion is that people are fairly good at knowing when they are experiencing and ADE, a lesson I learned early on in working with older adults in a retirement community, where an older adult convinced me he was experiencing memory loss from one of his medications, but I didn’t believe him at first. So if we listen and believe, kind of like having faith, we should be able to detect an ADE before it affects the function of the older adult. We just need to ask the question.
In a study in progress, we performed comprehensive medication reviews (CMRs) in 69 older adults and within the CMR questionnaire we asked the two questions, 1) What history do you have of adverse drug events (medication side-effects)?, and 2) Do you think you are experiencing an adverse drug event right now? Of the 69 participants, 28 answered both questions positive, yet only 4 said they were having an ADE but had a negative history of an ADE. When applying a 2×2 contingency table with Fischer’s test for exactness, we found a p-value of 0.00001 meaning the correlation was highly significant. That is to say, people were far more likely to say they are experiencing an ADE if they had a history of an ADE, as opposed to those without a history. This aligns with another article titled Adverse Drug Reaction Risk factors in Older Populations, Hajjar, E., Hanlon, JT, et al, AmJGerPharm., Dec. 2003 Vol. 1 (2) 82-89. Risk factors for ADEs were developed by a panel of experts in ADEs and one risk factor identified and agreed upon by the experts was a history of an ADE.
So it stands that it might be wise to put in your office visit routine the questions of “What history of adverse effects do you have from medications?” and “Do you think you are experiencing an adverse effect right now?”, and then listen and act. You might positively affect the life of an older adult by keeping them more functional and/or out of the hospital.
Emergency Hospitalizations for Adverse Drug Events in Older Americans
Dec 20th
A study published recently in the New England Journal of Medicine, by Dr. Daniel Budnitz, et al, states that there are just a few drugs commonly used in older adults that lead to emergency hospitalizations, those drugs being: warfarin (Coumadin), insulin, oral antiplatelets and oral hypoglycemic agents like glipizide and glyburide. These findings are similar to a study Dr. Budnitz published in Annals of Internal Medicine in December 2007 in which these classes of medications were implicated in emergency department visits for ADEs. These two studies do state very well that if we focus on just a few key drugs we can prevent a large percentage of ADEs in older adults that lead to ER visits and/or hospitalization.
In the more recent study, increasing age, number of medications and these four classes of drugs were most strongly associated with ADEs leading to emergency hospitalizations. Other medication categories that were involved in a large number of emergency hospitalizations were: cardiovascular drugs, antibiotics, sedative hypnotics, anticonvulsants, opioid analgesics, and Beer’s list medications. When factoring digoxin out of the Beer’s list admissions the percentage of admissions from Beer’s list drugs dropped nearly in half. It was also interesting that in 47% of cases the dose of digoxin was appropriate, according to Beer’s criteria. This means the dose of digoxin alone is not a completely effective screening criterion, but perhaps more frequent monitoring and drug interactions may play a role.
Speaking of drug interactions, this study did not report on the role that drug interactions have in leading to ADEs in older adults. There are recent data that list just a few critical drug interactions that lead to emergency room visits and hospitalizations in older adults. You can learn about those drug interactions at my post: http://elderdrugs.com/2010/11/adding-a-new-medication-watch-out-for-these-drug-interactions/
With regard to the anti-diabetic medications insulin and oral hypoglycemics, the implication is strong that routine monitoring and engagement of the person with diabetes is critical in catching downward trends of blood glucose in order to prevent severe hypoglycemia from occurring. As a diabetes educator I have always preached that the person with diabetes who is actively engaged in their own health is more likely to be successful, as opposed to the person who does not take an active role in managing their diabetes.
In conclusion, this study confirms previous studies and the role that just a few medication classes play in causing a large percentage of ER visits and hospitalizations in older adults. However, we mustn’t forget the large percentage of ADEs that lead to functional or cognitive impairment that we may just “chalk up” to aging. With the aging body less able to tolerate the insult from multiple medications, and with so many possible ADEs that can occur in those with complex medication regimens, anything less than a comprehensive medication review and development of a structured medication plan will fall short of the goal. So I advise to look at the entire body of literature and make your comprehensive medication review checklist in order to not miss some key contributors to ADEs in the older adult.
Are You Using a Medication to Treat a Side-Effect and Don’t Know It?
Dec 10th
When reviewing medications of an older adult it is not unlikely that some of the medications could have been prescribed to treat side-effects from other medications. This is called a “prescribing cascaded”. Some examples of prescribing cascades are:
- Starting a bladder health drug such as Detrol (tolterodine), Ditropan (oxybutynin), Vesicare (solifenacin) or other, in someone who develops urinary incontinence from Aricept (donepezil), Exelon (rivastigmine) or Razadyne (galantamine). In one study it was shown that people who used this class of memory health drugs were 1.5 times more likely to be prescribed a bladder health agent. The bladder health agent may also lessen the beneficial effects of the memory health drug.
- NSAIDs such as ibuprofen, naproxen, among many others, causing high blood pressure from the effects of the NSAID on the kidneys. What can follow is the addition of a medication for high blood pressure.
- The addition of a diuretic due to the development of fluid retention from a medication for high blood pressure such as amlodipine (Norvasc).
- The development of joint pain from a bone health drug such as Fosamax (alendronate), Actonel (risedronate) or Boniva (ibandronate), and then an analgesic or pain medication is added, which can then lead to side-effects depending upon the type of pain medication started. This prescribing cascade was validated by FDA and alerts were posted in 2008 and 2009.
- One last example is the use of Reglan (metoclopramide) for nausea, which then leads to the side-effect of movement disorders that mimic Parkinson’s disease. Those that present with this Parkinson’s-like side-effect are more likely to be misdiagnosed with Parkinson’s disease and treated with more medications to manage the tremors or slowness in movement that are the side-effects from the metoclopramide.
A key point to keep in mind is that you should always look back to when a medication was started to see if the side-effect, which is being treated as a condition or disease, followed not long after the addition of the primary drug. This is trying to develop a “temporal association” or an association related to the time of starting a new drug and the appearance of a side-effect. This leads to the primary principle in geriatric pharmacy management and that is “Always suspect the drug until proven otherwise”. If you can spot prescribing cascades it may be possible to reduce unnecessary medications that lead to adverse effects. Check with your doctor of pharmacist if you suspect a side-effect from a drug that may be treated with another drug. There is a good article that lists other prescribing cascades and speaks to the concepts more in depth at: http://www.australianprescriber.com/magazine/34/6/162/6
Comparative Effectiveness of Medication Adherence Interventions
Dec 8th
The Agency for Healthcare Research and Quality (AHRQ) has posted their comparative effectiveness review of medication adherence interventions. This can be useful for pharmacists and case managers. Here’s the link: http://www.effectivehealthcare.ahrq.gov/ehc/products/296/877/Medication-Adherence_Draft-Report_20111206.pdf
FDA reviewing reports of major bleeding from Pradaxa
Dec 8th
The FDA just sent a release stating they are looking at results from a clinical trial that included 18,000 people to see if major bleeding events from Pradaxa (dabigatran) are occurring more frequently than first reported. What stimulated this action were post-marketing reports of major bleeding events. FDA will release information regarding their analysis when available. Here’s the FDA link: http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm282820.htm
In the meantime, use Pradaxa with caution in older adults, not with NSAIDs and adjust the dose in those with renal impairment.
What Do Adverse Drug Event Databases Tell Us?
Dec 4th
There are several web sites I have come across that allow for the retrieval of information on specific drugs and their possible association with adverse events, defined as an unintended effect from a medication that causes a person to seek medical attention. The sites are:
- http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.htm
- http://www.drugcite.com/
- http://www.adverseevents.com/
- http://www.fdable.com/aers/fda_adverse_events?gclid=COmq1qf8wqwCFQrHKgodIn3hpg
What I can glean from these sites is that they store data that are commonly found on the FDA AERS website, which is a culmination of reports of suspected adverse drug events, not necessarily confirmed, but listed as possible ADEs from the implicated medications. As stated on the FDA website: “AERS data do have limitations. First, there is no certainty that the reported event was actually due to the product. FDA does not require that a causal relationship between a product and event be proven, and reports do not always contain enough detail to properly evaluate an event.” So what we have is a rather loose set of data without cause and effect having been confirmed. That being the case, I would caution anyone from reading too much into any interpretation from such a database as it can be misleading, and subsequently lead to inappropriate decisions being made regarding medication therapies.
I searched each database for a specific drug, for example alendronate, and came up with different levels of incidence, so I question the usefulness of these databases for that purpose. Also stated by FDA: “Further, FDA does not receive all adverse event reports that occur with a product. Many factors can influence whether or not an event will be reported, such as the time a product has been marketed and publicity about an event. Therefore, AERS cannot be used to calculate the incidence of an adverse event in the U.S. population.” Most likely there are differences in how the administrators of each website manages the database thereby producing differing results.
I also found unexpected ADEs not implied elsewhere. For example, bone health drugs (Forteo, Boniva, Fosmax, etc.) causing falls. I’m not certain what to make of that. I guess it’s possible if these drugs are causing muscle and joint pain, which has been reported by FDA. Perhaps we need to look into these databases further to see if there is a valid statistical correlation which then warrants further research. The site www.adverseevents.com claims to have a filtering process that can draw such associations. I haven’t looked into the specifics, nor am I an expert in the use of statistical modeling, so I can’t make any further comment. What I do know is that this approach has some potential, as already demonstrated by FDA, but there is also the approach of validating ADEs through studies that needs to occur.
The strongest method of validating direct cause and effect between a drug and an ADE is a case-controlled prospective trial. Reviewing databases in a retrospective manner has limitations which limit the strength of any argument of implied cause and effect. Frequently, the most reliable statement when working with a database in a retrospective manner is that one can say there my be a correlation, nothing more. Recall the study published in 2010 that made the claim that opiates are associated with GI bleeding, more so than NSAIDs, just by performing a statistical analysis on the database. However, the authors did list one of the limitations of not having access to OTC use of NSAIDs. In review of pooled data including older adults in a retirement community, we found 13.8% were actively using OTC NSAIDs. (http://elderdrugs.com/2010/12/oxycontin-similar-drugs-up-fracture-heart-attack-risk-cnn-reports/) Add to that the number of NSAID-users who are taking SSRIs, and you can run the odds ratio for GI bleeding up to 15. So there are limitations to this type of ADE tool and they can be best used to screen for possible problems that can affect a large number of people, but then there need to be more specific studies conducted to validate the possible problem.
Adverse Drug Event Causation: A Review of the Literature
Dec 3rd
I have reviewed many studies related to the causative factors of adverse drug events (ADEs) in older adults in order to learn where to apply my efforts in preventing them from happening in those I serve. But I never really put them together into one picture to get a better idea of the most reported and validated risk factors that lead to ADEs. Since I have on file what I consider a very large sampling in the literature (22 studies from 1997 to present) I decided to do just that. My findings are reviewed below. I only included studies in which the intent of the author(s) was to measure risk factors for ADEs and not studies in which specific drug classes were studied, or where authors studied drug interactions leading to ADEs, or medications known to cause falls, and so on. So in effect, this is a compilation of studies where the intent was to uncover ADE risk and/or causation by either retrospectively or prospectively looking at pooled cases in which ADEs were documented to have occurred, in the hospital, emergency room, or ambulatory care. That by itself could lead to a limitation in identifying what are the most significant causes of ADEs, since those that work in emergency rooms, or general practitioners in ambulatory care, may not know where to look for causation, therefore authors looking at charts or databases are working with data that may not necessarily reflect other significant causes of ADEs that lie deeper. One example is with the use of timolol eye drops. There is a body of literature that validates the systemic effects from these eye drops that can lead to CHF, hypotension, bradycardia, dizziness with subsequent falls, depression and other adverse effects. But in none of these studies do the authors ever mention systemic effects from eye drops as a cause. Why? “You see only what you look for; you recognize only what you know” is a quote from Dr. Michael Chisner, a cardiologist who used that quote when teaching students about the assessment process and trying to diagnose cardiovascular illness. So what I’m saying is that many ADEs go unreported or the results are inconsistent, as you will see in these findings, because there is inconsistency in ADE detection methods. We must also consider that no more than 11% of ADEs are reported in health system adverse event reporting systems, so the ADE quagmire most likely runs much deeper. Perhaps we can come to the point where we finally understand all the relevant risk factors for ADEs in order to build the most effective screening tool for prescribers, nurses and pharmacists in order to design the most effective and safest medication regimen for older adults.
Findings
Number of Medications and Co-morbidity
Of 22 studies reviewed the two most commonly cited risk factors for ADEs are number of medications, and number of co-morbid conditions. Most studies (N=5) cite 5 or more medications as being the most reliable risk factor for predicting an ADE, one study said 3, others just reported a rising correlation with more medications, e.g. 10% increase in risk for each additional medication. With regard to co-morbidities, there were several methods of reporting this, three by number of diagnoses, two by the Charlson co-morbidity index, and a couple by disability index or worsening health status. Not all studies had a report of number of medications or chronic conditions, and two said there was no correlation with number of medications and one study cited no correlation with number of diagnoses.
Gender, Age, Transitions in care
There was no congruency with gender, with one saying none, and two correlated with being female and two with being male. Only three studies reported on transitions in care, two saying there was a correlation and one saying no correlation. Age also had inconsistent reporting with four studies saying no correlation and three saying there was a correlation.
Medication Classes
For specific medication classes there is a huge disparity between studies with studies not always reporting the same medication classes as other studies, and some studies contradicting the positive correlation of a particular medication class from other studies. In general, the number of positive correlations from these studies by medication class is as follows: antibiotics (6 studies); antipsychotics (5); antidepressants (8); benzodiazepines (6); cardiovascular meds (11); opiates (5); anti-epileptics (1); cardiac glycosides (6); insulin (5); oral hypoglycemics (5); diuretics (10); warfarin (10); NSAIDs (8); ACE inhibitors (3); beta-blockers (4); anti-platelets (2); electrolytes (2); PIMs (3); steroids (3); and Parkinson’s medications (1).
Other considerations: Drug interactions, ADE history, lack of monitoring
One must consider that a reported ADE by medication class could have actually resulted from a drug interaction that went undetected or not known at the time. For example, SSRIs and NSAIDs are strongly correlated with risk for GI bleeding, yet if not known it would most likely be reported as NSAID-induced. Also, this may tie into another risk factor found in a couple of other studies and that is the recent addition of medications, or medication changes, as a predictor of ADE risk. That would link to the concept that drug interactions are largely responsible for ADEs with one study saying 26% of all ADEs are drug interaction related. Another consideration is that there are other risk factors to take into account such as history of an ADE. One Australian study verified this as a predictor for a repeat ADE. Also, many of the medication classes where ADEs are found to commonly occur imply that more frequent monitoring may have prevented those ADEs. Thus, lack of monitoring is a risk factor for ADEs, as reported in other studies, and may account for up to 40% of all ADEs.
Summary
The most reliable predictors of ADEs are number of medications and co-morbidity and not necessarily specific classes of medications. Yet there are many confounding factors that can lead to ADEs therefore a comprehensive approach to detection is most likely to detect all areas of risk, which then leads to targeting areas for interventions to prevent ADEs. Understanding the nuances of ADE causation in older adults is necessary in order to make the largest impact on ADE prevention, anything less will not necessarily result in a significant reduction of occurrence of this 5th leading cause of death by disease, albeit a man-made disease.
