HIV & Aging Clinical Recommendations

Chapter 13

Drug-Drug Interactions and Polypharmacy in HIV and Aging

  • The primary care provider is highly encouraged to perform annual complete medication reconciliation and a medication review at every visit so that a complete and active medication list is available. This process isn’t complete until the prescriber discontinues medications no longer indicated and notifies the dispensing pharmacy and patient thus reducing risk for toxicity and/or drug-drug interactions.
  • To assist with the medication reconciliation and assessment, it is recommended to utilize a validated instrument such as Beers list or STOPP/START in evaluating potentially inappropriate prescribing.
  • It is recommended that patients utilize one pharmacy or a pharmacy with an integrated pharmacy computer network and where possible, utilize an HIV specialty pharmacy.
  • For patients with renal insufficiency, the Cockcroft-Gault derived creatinine clearance calculation should be used to determine the appropriate medication dose or frequency adjustments. While less accurate in older patients, this equation is still widely used in renal dosing charts, by the FDA and within package inserts. However, the renal function estimated by MDRD if unadjusted for body surface area may also be a reasonable substitute although has not been validated in adults >70 years old.
  • In the setting of hepatic dysfunction, certain medications need dose adjustment.

The prevalence of disease and comorbidities increases with advancing age, and along with this process come additional medications to treat comorbidities. Increasing number of medications increases the risk of adverse drug events and drug-drug interactions (Steinman 2014; Beers 1997). The difficulty with determining drug-drug interactions is that the studies are traditionally done in young and healthy volunteers and, in the case of medications intended to be used to treat people living with HIV (PLWH), in HIV-positive volunteers. This is done to minimize any potential confounders due to age, reduced renal and/or hepatic function, concomitant medications and co-morbidities. The true extent of a drug interaction in an older patient may never be able to be fully assessed due to these reasons. Therefore, we must extrapolate from the available data and assume that the extent of an interaction will be at least as great as that observed in the study population. While no significant differences in pharmacokinetics (i.e. absorption, distribution, metabolism, excretion) have yet been found when studies have been conducted in aging individuals, other differences do exist that must be accounted for (Klotz 2009; Herrlinger 2001; Onen 2010).

In the geriatric population as a whole, pharmacodynamic differences predispose patients to more adverse drug reactions. This is often attributable to enhanced sensitivity to centrally and peripherally mediated anticholinergic side effects (e.g. tricyclic antidepressants, diphenhydramine, doxepin, muscle relaxants, antiemetics, antispasmodics, antidepressants), reduced benzodiazepine clearance (e.g. chlordiazepoxide, diazepam), decreased baroreceptor responsiveness (alpha-blockers, beta-blockers, beta-agonists) and increased CNS sensitivity to opioids and sedative-hypnotics (see Table 1). Caution should be used when prescribing and monitoring anticoagulation therapy and antipsychotics should not be used to treat insomnia or other non-approved indications. Additional medications that serve as markers for increased potential for interactions and adverse drug events among PLWH include ritonavir- and cobicistat- boosted protease inhibitors and integrase strand-transferase inhibitors, statins, tenofovir, H2-antagonists and proton pump inhibitors (see Table 2) (Kennerfalk 2002; Pizzuti 2006).

While little evidence exists for medication dosing based on serum concentrations in older PLWH, the clinician must rely on data from the uninfected population for dosing guidance. Additionally, screening for high alert medications on the patient’s medication list is necessary to reduce the likelihood of medication-induced reactions, which are more likely due to altered pharmacokinetic/pharmacodynamic parameters or concomitant medical problems commonly experienced by aging patients.

Assessing and Evaluating Polypharmacy

Treatment of the multiple medical problems often seen in the elderly may require the patient to see providers in specialty clinics, or be hospitalized and medications initiated while hospitalized. For these reasons, the primary care provider or their collaborating pharmacist (when available) is highly encouraged to perform an annual medication reconciliation so that a complete and active medication list is available. This involves having the patient return to clinic with all of their medications and ensuring that the primary care provider has a complete list of meds prescribed by specialty providers. Annual medication reconciliation can also be facilitated by obtaining a 3-month dispensing history from the pharmacy. The accuracy of this list is increased if the patient uses one pharmacy or goes to pharmacies that utilize a computer network that is integrated into the work areas of the rest of the healthcare team. The importance of critically assessing the continued clinical need for each medication and a medication review cannot be overemphasized and should be done at every visit. Medication review/reconciliation provides an opportunity to determine which medications are no longer clinically indicated and may be discontinued to reduce the risk of toxicity, untoward adverse events and drug-drug interactions. Medication review, medication reconciliation and screening for toxicity and drug interactions are best facilitated by a clinically trained pharmacist. Electronic resources such as Epocrates, Lexi-Comp and Tarascon offer up-to-date interaction checking (,, Drug interactions with antiretrovirals may also be assessed through the University of Liverpool’s online tool (

Table 1: Medications Associated with Increased Likelihood of Toxicity

Medication Suggested Management
Antiemetics Use with Caution
Antispasmodics Use with Caution
Antidepressants Use with Caution
Alpha-blockers Use with Caution
Beta-blockers Use with Caution
Benzodiazepines (diazepam, chlordiazepoxide, alprazolam) Should be Avoided
Beta-agonists Should be Avoided
Diphenhydramine Should be Avoided
Doxepin Use with Caution
Fentanyl, oxycodone, morphine, methadone Use with Caution
Meperidine Should be Avoided
Muscle Relaxants (carisoprodol, methocarbamol, baclofen) Use with Caution
Sedative hypnotics (zolpidem, others) Should be Avoided
Temazepam, lorazepam Should be Avoided
Tricyclic Antidepressants Should be Avoided

Both the Beers criteria (Beers 1997) and the STOPP/START (Screening Tool of Older Person’s Prescriptions) (Gallagher 2008) are validated instruments that can be used to reduce potentially inappropriate medications in older patients that have been associated with increased morbidity or mortality in older patients. While both tools are available online with the explicit criteria and are too long to be listed here, the Beers criteria tends to focus on potentially inappropriate medications in older adults while the STOPP criteria focuses on potentially inappropriate medication-disease combinations. Examples of medications on the Beers criteria include use of first-generation antihistamines, benzodiazepines and tricyclic antidepressants to name a few. Examples of STOPP criteria that may be encountered in clinic settings include the use of thiazide diuretics in patients with gout, the use of scheduled opiates without a bowel stimulant or the use of a non-steroidal anti-inflammatory agent in patients with renal disease, moderate to severe hypertension or those with a history of peptic ulcer bleeding or GI bleeds. Out of 600 patients 65 years and older, STOPP criteria identified 329 adverse drug events in 158 patients. Sixty-seven percent were identified as being a factor in hospital admission and 69% were considered avoidable (Hamilton 2011). Much more research needs to be performed in this field and efforts to reduce polypharmacy need to be implemented among PLWH. Clinical pharmacists can serve a key role in optimizing patient care by applying Beers and/or STOPP criteria in their review of patient medication regimens.

Polypharmacy (Hajjar 2017) (as defined by Hajjar, et al. 2017) is the prescribing of more medications than are clinically needed for a given patient, although many clinicians prefer to think of polypharmacy in terms of a threshold number of active prescriptions for a patient (such as >20 prescriptions). In the general population, polypharmacy is estimated to occur in 20-50% of patients, with adverse drug reactions more common and serious in the older patient (Kennerfalk 2002; Pizzuti 2006). It has been recently estimated that 100,000 emergency hospitalizations were due to medications, with 1.5% of these occurring in the elderly (Budnitz 2011). While age alone has little impact on organ reserves or capacity, comorbidities play a larger role (Klotz 2009; Herrlinger 2001). As a result, it is important to recognize that chronological age may not always correlate to biological age. To reduce the risk of polypharmacy it is recommended, where possible, that the patient utilize a single pharmacy, preferably one with experience caring for PLWH or one that has a computer network integrated into their patients’ medical records. Utilizing a specialty pharmacy has been shown to improve HIV care in terms of fewer contraindicated medications and improved adherence (Hirsch 2009). Additional benefits may also include improved pharmacist-prescriber communication regarding clinically significant drug-drug interactions, medication reconciliation needs, monitoring adherence, and providing adherence aids such as Medisets/bubble packs, pillboxes, medication delivery and personalized patient counseling. Patient preference is important in assessing the benefits and risk of additional medications. Some patients are bothered by taking many medications while others are not.

Table 2: Common Medications Interacting with Antiretrovirals

Azole antifungals (esp. itraconazole, posaconazole, voriconazole)
H2-antagonsits (when combined with atazanavir or rilpivirine)
HCVNS/4A inhibitords (boceprevir, telaprevir)
PDE5 inhibitors (esp. tadalafil)
Proton pump inhibitors (when combined with atazanavir or rilpivirine)
Statins (esp. lovastatin and simvastatin)

Recently, more data has become available about polypharmacy related issues among PLWH. In a cross-sectional analysis of the Veterans Administration electronic medical and pharmacy records from October 2009 to September 2010, 16,989 HIV-positive adults receiving antiretroviral therapy and 47,613 HIV-negative adults were identified and assessed for polypharmacy (Edelman 2013; Edelman 2013a). For both the HIV-positive adults and controls, mortality was greater with a higher number of medications, particularly in those who took ≥5 medications. For PLWH patients receiving 3-4 medications, 5-7 medications, and 8 or more medications, hazard ratios for mortality were approximately 1.4, 2.1, and 2.1, respectively, while for controls they were 1.4, 1.5, and 1.9, respectively. The association between polypharmacy and increased mortality is concerning. In addition, data from the HOPS cohort indicated that in PLWH 50 years and older, 54% experienced polypharmacy compared to 34% of PLWH less than 50 years old (Holtzman 2013).

Renal and Hepatic Dysfunction

Despite its shortcomings, the Cockcroft-Gault equation is still primarily used by the FDA, product package inserts and guidelines when drug dosing guidance is needed (Dowling 2010; Lamb 2003). This document recognizes that many labs provide an eGFR rather than a creatinine clearance based on Cockcroft-Gault. Since many aging patients may already have some level of chronic kidney disease, the eGFR derived from the MDRD equation may be used to facilitate renal dosing of medications as long as it is unadjusted for body surface area. Common medications requiring renal dose adjustment include acyclovir, fluconazole, gabapentin, H2-antagonists and most nucleoside RTIs (Table 3). Meperidine should be avoided in patients with renal insufficiency, as should most non-steroidal anti-inflammatory agents. In addition, non-steroidal agents are associated with an increased risk of gastrointestinal bleeding. For HIV-infected patients, preliminary studies suggest that tenofovir requires closer monitoring in an older individual than in a younger cohort (Moore 2010). It is unknown whether the new formulation of tenofovir alafenamide will also require this careful renal monitoring.

Not only does the clinician need to consider renal dysfunction, but also hepatic function in the correct dosing of medications. Hepatic dysfunction, while it can occur in co-infected individuals, is not limited to that population but also in patients with other forms of dysfunction, such as those caused by alcoholic cirrhosis, etc. Medications that are hepatically metabolized may accumulate to supra-therapeutic concentrations in patients with hepatic dysfunction. The risk of toxicity can be reduced by dose adjustment. Rather than utilizing transaminases to determine hepatic dysfunction, hepatic function is best measured by the use of the Child-Pugh score which should be calculated and utilized to determine the proper medication dose.

Antiretrovirals that require dose adjustment based on hepatic function include abacavir, non-nucleoside reverse transcriptase inhibitors and protease inhibitors. Specific dosing information for antiretrovirals as well as details on calculating a Child-Pugh score can be found in the DHHS Guidelines for the Use of Antiretroviral Agents. For the most up-to- date information and for non-antiretroviral medications, the clinician is strongly advised to consult a clinical pharmacist to determine the necessity, and if needed, proper dose of medications for patients with hepatic dysfunction.

For additional information on drug-drug interactions, providers are advised to utilize the tables in the DHHS Guidelines for the Use of Antiretroviral Agents or the CDC/NIH Guidelines for the Prevention and Treatment of Opportunistic Infections. More updated information in an interactive format may be found at University of California San Francisco, HIV In Site Database of Antiretroviral Drug Interactions ( or the Toronto General Hospital Immunodeficiency Clinic Drug Interaction Tables ( For additional information on renal dosing, the DHHS guidelines provide a valuable reference for medication dosing in settings or renal or hepatic dysfunction.

Updated on: 
Monday, October 16, 2017
Updated by: 
James Scott, Pharm D, FCCP


Purpose of this Program: The AAHIVM, ACRIA and AGS (collectively, the “Sponsors,” “we” or “us”) are sponsors of this Website and through it seek to address the unique needs and challenges that older adults of diverse populations living with HIV face as they age. However, the information in this Website is not meant to supplant the advice provided in a doctor-patient relationship.

General Disclaimer: is designed for educational purposes only and is not engaged in rendering medical advice or professional services. The information provided through should not be used for diagnosing or treating a health problem or a disease. It is not a substitute for professional care. If you have or suspect you may have a health problem, consult your health care provider.


Phone: 202-659-0699
Fax: 202-659-0976


AAHIVM National Office
1705 DeSales Street NW, Suite 700
Washington, D.C. 20036


The American Academy of HIV Medicine is a professional organization that supports the HIV practitioner and promotes accessible, quality care for all Americans living with HIV disease. Our membership of HIV practitioners and credentialed HIV Specialists™, HIV Experts™, and HIV Pharmacists™ provide direct care to the majority of HIV patients in the US.