WARFARIN & SULFONAMIDES This interaction is one of the most well supported in the literature. The scenario is that a patient presents with signs and symptoms of UTI and the drug of choice appears to be Septra or Bactrim DS (sulfamethoxazole / trimethoprim). However, the sulfa antibiotic is known to displace warfarin from protein binding sites where more than 90% of warfarin is bound, affecting the clotting cascade. It is recommended to use an alternative antibiotic, or test INR soon after the sulfa antibiotic is started. The question then comes as to when to test. We recently experienced that scenario where the INR was tested 4 days after the start of the sulfa antibiotic, and there was a slight dose adjustment downward of the warfarin based on a higher, but not remarkably higher, INR. Several days later the INR was 7.7. The moral of the story: alternative drug therapy may be the safest approach.

WARFARIN & AMIODARONE Warfarin can also interact with amiodarone, (a serious interaction), so when starting amiodarone on someone receiving warfarin, frequent INR monitoring is warranted.

LITERATURE REVIEW An excellent review by Juurlink, DN and Redelmeier, DA, et al in 2003, studied hospital admissions likely due to drug interactions. Most  significant were:

• Those admitted for severe hypoglycemia and taking glyburide were 6 times more likely to have been started on a sulfa antibiotic in the last 7 days.

• Those admitted with digoxin toxicity were 12 times more likely to have been treated with clarithromycin (similar to azithromycin) in the previous week.

• Those admitted for hyperkalemia and treated with an ACE inhibitor were 20 times more likely to have been treated with spironolactone or other potassium-sparing diuretic in the previous week.

The above three suggest either alternative drug therapies, or frequent monitoring with patient education as to what symptoms to look for.

NSAIDs & SSRIs We know that NSAIDs are associated with increased risk of GI bleeding, with odds ratios from 1.5 to about 7.0, depending upon the drug. SSRI’s are also associated with a modest increase in risk for GI bleeding (odds ratio of 1.5-3.0). Of particular concern is that the combination of any NSAID with an SSRI results in a significantly elevated, disproportionate risk for GI bleeding with odds ratios as high as 15.6. The increase in risk is even seen with low-dose aspirin (odds ratio of 7.2).

With upper GI bleeding being a serious ADE leading to harm and hospitalization in older adults, this warrants a review of such combination therapy. It also begs the question, should we be using NSAIDs on a routine basis in older adults?

According to the American Geriatric Association (AGS) Pain Management Guidelines from 2009, chronic use of NSAIDs should be avoided in the older adult population whenever possible. If use is required, then mitigation of that risk, in part, should occur with the use of a proton pump inhibitor.

OTHER INTERACTIONS

• ACE inhibitors plus sulfonamides leading to hyperkalemia

• Warfarin and NSAIDs causing bleeding

• Warfarin and dicloxacillin leading to DECREASED INR

• ACEs or ARBs plus potassium supplements causing hyperkalemia

• Trimethoprim can increase serum digoxin levels by up to 75%

• Opiates (primarily Oxycontin or oxycodone) and SSRIs (Prozac, Zoloft, Lexapro, Celexa, etc) or SNRIs (Effexor) resulting in serotonin syndrome, may be more likely in older adults. Look for this in short-term rehab patients who start on relatively high doses of opiates to control pain from joint replacement surgery.

In closing, you now have a list of some of the most significant drug interactions that lead to ADEs in older adults. However, there are many that are not listed here which can be equally impacting to the lives of older adults. My recommendation is to use a drug interaction screening tool at the point-of-prescribing, or ensure that you encourage your patient to ask the pharmacist to ensure they have checked for all possible interactions. ADE avoidance is the best approach to safety.

However, the problem is that studies that seek to find ways to improve adherence do not equally as well study the probable increase in ADEs in their study population. This is evidenced in the AHRQ analysis on medication adherence solutions. Only three studies looked at ADEs as a possible negative outcome with none documented, but there was lacking homogeneity to conclude anything, most likely because they were not looking in great enough detail with the right screening tools. Not unlike why so many studies don’t match up on ADEs in older adults: “You see what you look for and recognize only what you know”. i.e. If you have a shallow knowledge base and don’t know how to recognize probable ADEs in older adults, you will not find them. What leads me to believe that specific research needs to be done is having searched the literature and finding that in the HIV/AIDS population there is good evidence that increasing adherence increases the incidence of ADEs. Since adherence is so critical in being successful at managing HIV/AIDS, they look at this closely, where as other studies do a cursory review.  So why wouldn’t that apply to other populations? Other drugs are toxic and other populations are not immune from ADEs.

But if you know that improving adherence in older adults can increase the risk for ADEs, then on a case-by-case basis you proceed with caution and monitor closely with the intent of improving adherence to achieve specific, realistic goals, but monitor for the development of ADEs. We not uncommonly find that people become dizzy from blood pressure medications after starting an adherence program, so we adjust doses downward. It then lends wisdom that a practitioner may consider lowering doses before starting an adherence program if there is documented poor adherence. Adhering to a well-designed drug regimen is probably one of the most effective forms of health care, when properly implemented.

The authors then categorize patients based on how they make health care decisions, such as “minimalists”, who believe “less is more” (that’s me!), and “believers”, who believe there is a cure for everything, and then the “doubters”, “who worry that almost any treatment will be worse than the disease”. Then comes the categorization of “experts” being fitted into these categories, such as “the Federal Preventive Services Task force as “minimalists” and “doubters”, and that is why experts can look at the data and have differing opinions. The authors then make their point about the uncertainty of “best-evidence” saying, “No one can say with certainty who will benefit by taking a certain drug and who will not”.

Then there is the discussion about the delivery model of health care and how it’s changed. “Medical care used to be paternalism: A doctor dictated what was to be done and the patient complied. This model has largely been abandoned, but Democrats and Republicans are offering a new form of paternalism.”  I agree with the authors when they say treatement decisions should be made between doctor and patient, since everyone has differing circumstances with which they are confronted, and differing needs which alters their priorities, and a “cookie-cutter” style approach doesn’t always meet the needs of the person. However, something must be done to lower health care costs since we spend the most in the U.S., but don’t get the most in return, so I see clearly that certain restrictions and guidelines need to be applied in order to sustain health care services to all. We just need to determine what is the absolute “best-evidence” to start applying and measure the benefits of its application.

I responded to the question a few years back when someone asked me what I do and I said, “My job is to get as close to  the truth as is possible”, and that is what I have been doing for several years, gathering evidence, analyzing it, and determining when “evidence” is not the “best-evidence”, and sometimes downright false, and what evidence is, without a doubt, solid. What I have come to know to be true is that we are causing harm by misusing medications in older adults, to the degree that adverse drug events is now the 4th or 5th leading cause of death by disease and forecast to race to the top of the charts within 10 years. It is also clear that application of the “best-evidence” regarding the appropriate use of medications in older adults can prevent the majority of them from occurring, and lower costs by lowering health care utilization and avoiding complications. We just need to find a way to collaborate as health care providers, with the person at the center, and try to keep the best-evidence in front of us.

Here is the link to the WSJ article.

http://online.wsj.com/article/SB10001424052702303404704577311641531125820.html

http://www.effectivehealthcare.ahrq.gov/ehc/products/160/1007/CER53_LowBoneDensity_FinalReport_20120329.pdf

The care coordination role involves coordinating community resources, referrals for additional care, and communicating with the PCP and other providers. This is an excellent use of the highly skilled pharmacist who has a vast amount of clinical knowledge, and it also demonstrates where pharmacists can expand their non-traditional roles by adding value to the health care system. The article is available to ASCP members only. Correspondence to Dellara F. Terry, MD, MPH, at Dovetail Health at 781-292-3581, may provide further information, as stated in the article.

http://www.goldenliving.com/healthcare-news-events/gl-press-releases/press.aspx?assetId=bf1f968b-74d4-41eb-8262-04ec546b9e0a

In another study in long term care residents, Dr. Garfinkel was able to apply the same algorithm and discontinue 332 drugs in 119 disabled residents. What was observed was a lower one-year mortality rate than in the control group, 21% vs. 45%, and a lower rate of referrals to acute care facilities, 11.8% vs. 30% in the control group. This study also measured a substantial decrease in medication costs.

Reducing medication use in the old-old population makes perfect sense since many older adults will suffer from duplicate therapy and resultant toxicity; serious drug-drug interactions; lack of monitoring that leads to drug toxicity; cumulative anticholinergic drug burden with functional and cognitive decline; yet many of these drugs are used with lacking evidence of benefit in this population. It may be that we see a major shift, in a relatively short period of time, in how medications are used in long term care facilities, and the extended impact may be a significant reduction in hospitalizations thereby improving quality of life and helping save the Medicare system. You can find Dr. Garfinkel’s work referenced on his home page at: http://www.dr-g.co.il/.

It may prove fruitful for organizations involved in ACOs to strongly consider utilizing the assistance of those who understand the nuances of medication use in this population to take advantage of an opportunity that pays big dividends. A well thought out approach and launch of a formal program in LTC facilities can lead to a significant lowering of hospitalization rates.

Some of the more specific findings from the studies were: patient and healthcare professional reports are of similar quality; there was some evidence that different ADRs are reported; new ADRs are reported; reported symptoms to SSRI antidepressants, e.g. Zoloft, Celexa, Lexapro, Prozac, are not found in health care professional reports, implying a reluctance to report unusual psychiatric symptoms to physicians; some evidence that patients report ADR symptoms quicker; no evidence that reports result in distraction from signal detection, meaning they do not distract from ADR “flags” and overload the ADR reporting system.

In all, patient-reporting of ADRs is a reliable method for detecting probable ADRs which can lead to mitigating them and improving quality of life, and avoid unnecessary utilization of health care resources, in addition to lessening the risk for patient harms. However, in my experience, if people are not educated on what side-effects to monitor for, and if physicians do not acknowledge that ADRs are possibly occurring, especially some of the more unusual ADRs, then a large percentage are not addressed and lead to negative outcomes. Several other studies have validated this finding. Lastly, specific questions asked of the patient that may elucidate if an ADR is occurring should be included in every medication review and office visit. The reliability of this technique is greater than what one might think.

Looking deeper at the possible causes, Dr. Hines ascribes lack of medication reconciliation as one cause, where medications are not reconciled in detail which leads to duplicate drug therapies and drug interactions, thought to cause adverse drug events in about 26% of all ADE cases. He also states that patients may not ascribe adverse effects to medications and will not ask for changes in drug therapy. Yet a well-educated patient is more likely to recognize symptoms as an adverse event, implying that patient education upon discharge is critical. One possible solution is to create a comprehensive medication plan that travels with the patient after discharge so critical lab monitoring occurs, along with key educational points on how to properly take medication, in addition to a list of key symptoms that can be likened to the medication therapies the person is taking. Lastly, designing a drug regimen that the patient can adhere to is also important, especially for people discharged after an episode for congestive heart failure. Poor adherence is strongly associated with readmission to the hospital in those with congestive heart failure.

In summary, medications play a key role in mitigating risk and improving survival and function, but when not well managed they can lead to adverse drug events that cause harm and lead to readmission to the hospital. Some data suggest that the adverse events will show up about 14 to 21 days after hospital discharge implying that a well-designed medication plan can prevent a large number of readmissions.

Here are some other links that refer to adverse drug events after discharge from hospitals.

http://bennet.senate.gov/newsroom/press/release/?id=3d5c9532-be2a-4a3f-baa9-3f9196d394aa

http://hospitalmedicine.ucsf.edu/improve/literature/discharge_committee_literature/reengineering_systems/role_of_pharmacist_counseling_in_prevent_ade_after_hosp_schnipper_ama.pdf

http://psnet.ahrq.gov/primer.aspx?primerID=11

http://www.caretransitions.org/documents/Promoting%20effective%20transitions%20of%20care%20-%20JHM.pdf

http://www.ncbi.nlm.nih.gov/pubmed/15903284

These findings are consistent with a large body of literature that the greater the number of medications we take, the greater the risk or likelihood we will experience an adverse drug reaction that causes us to seek medical attention. Using the number of medications a person takes as a screening tool should prompt all parties involved to review a persons medications to: ensure they are needed, that the doses are appropriate,that monitoring is in place, are free from significant drug interactions, and taken properly, since all of these contribute to the risk for an adverse drug reaction.

In the discussion section, the authors go on to talk about the concept of medication-minimization and how this approach might lead to a lower rate of adverse drug events, albeit how it can be a challenging task to pare down a complicated drug regimen. However, there is a growing body of evidence that suggests how we can pare down the drug regimens of an older adults without causing harm, and in many instances improve how they function. This is an area that will gain greater clarity in everyday practice and drastically change medication utilization in older adults, all for the greater good.