Methodology of an ongoing, randomized controlled trial to prevent falls through enhanced pharmaceutical care

S. Ferreri et al.

I The American Journal of Geriatric Pharmacotherapy

Methodology of an Ongoing, Randomized Controlled Trial to Prevent Falls Through Enhanced Pharmaceutical Care Stefanie Ferreri, PharmD, CDE1; Mary T. Roth, PharmD, MHS2; Carri Casteel, PhD, MPH3A; Karen B. Demby, PhD3; and Susan J. Blalock, PhD, MPH2 DilJision of Pharmacy Practice and Experiential Education, School of Pharmacy, UnilJersity of North Carolina at Chapel Hill, Chapel Hill, North Carolina; 2DiIJision of Pharmaceutical Outcomes and Policy, School of Pharmacy, UnilJersityof North Carolina at Chapel Hill, Chapel Hill, North Carolina; 3UniIJersity of North Carolina Injury PrelJention Research Center, UnilJersity of North Carolina at Chapel Hill, Chapel Hill, North Carolina; and 4Department of Epidemiology, School of Public Health, UnilJersity of North Carolina at Chapel Hill, Chapel Hill, North Carolina 1

ABSTRACT Background: Falls are the leading cause of both fatal and nonfatal injuries among adults aged ~65 years in the United States. Past research suggests that individuals taking multiple medications are at increased risk offalls. Central nervous system-active drugs in particular have been associated with increased risk. Objective: The goal of this research was to describe the design of a study evaluating the effectiveness of a community pharmacy-based falls prevention program. Also presented are the algorithms used to identifY high-risk patients based on their prescription profile records and to deliver the experimental intervention. Methods: The study is a randomized controlled trial. The target population was community-dwelling older adults (aged ~65 years) at high risk for future falls because: (1) they had experienced ~1 fall within the 12-month period preceding study enrollment; (2) they were currently using ~4 chronic prescription medications; and (3) they were taking ~1 of the high-risk medications targeted by the intervention. Participants were recruited using pharmacy prescription profile records. Individuals in the intervention group received a face-to-face medication consultation provided by a community pharmacy resident. Identification of drug therapy problems and therapeutic recommendations was guided by a series of algorithms developed for this study. All participants were followed up for 24 months. The primary study end points were: (1) time to first fall; and (2) proportion of participants who experienced ~1 fall during the first year of follow-up. Results: Participant enrollment began in September 2005 and was completed in August 2007. A total of 186 individuals were enrolled in the study (mean [SD] age, 74.8 [6.9] years; 132 women, 54 men), and 67 have completed the first year of follow-up. Conclusions: The study is using a rigorous randomized controlled research design, which will enhance the internal validity of its findings. Results of the study, which will be reported after the completion of follow-up data collection activities, will enable us to assess the effects of the intervention on both medication use and the incidence of falls. If the intervention is found to be effective, it will provide a resource for community pharmacists working with older adults at high risk of medication-related falls. (Am] Geriatr Pharmacother. 2008;6:61-81) © 2008 Excerpta Medica Inc. Key words: falls, elderly, medication, pharmacy. The study reported in this article was previously presemed at the National Injury & Violence Prevemion Research Conference sponsored by the Society for Advancement of Violence and Injury Research, October 10-11,2007, Columbus, Ohio.

Accepted for publication March 29, 200B. © 2008 Excerpta Medica Inc. All rights reserved.

doi:1 0.1 016/j.amjopharm.2008.06.005 1543-5946/$32.00

Volume 6 • Number 2

June 2008

61

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

INTRODUCTION Falls are the leading cause of both fatal and nonfatal injuries among adults aged ~65 years in the United States.! Based on data from the National Center for Injury Prevention and Control, falls accounted for 43% of all unintentional, fatal injuries experienced by individuals aged ~65 years in 2005. 2 Although most falls do not result in death, nonfatal falls are especially common among older adults and can have serious consequences. Each year, more than one third of adults aged ~65 years experience a falP Data from the National Electronic Injury Surveillance System-All Injury Program (NEISS-AIP) highlight the magnitude of this problem. In 2006, > 1.8 million falls among individuals aged ~65 years were reported to the NEISS-AIP, accounting for >61% of all unintentional injuries reported. 2 More than 24% of these falls resulted in hospitalization. Experiencing recurrent falls is an established risk factor for nursing home placement. 4 Furthermore, even among those who remain in the community after a fall, fear of falling can cause individuals to restrict their activities, decreasing quality of life and, paradoxically, increasing their risk of future falls. s Clearly, falls represent a major threat to the health and well-being of older adults. Many factors have been associated with an increased risk of falls, including: history offalls; arthritis; depression; gait, balance, or visual deficits; muscle weakness; and functional or cognitive impairment. 6.7 In addition, pharmacoepidemiologic research has found that individuals who use multiple medications are at increased risk of falls,?·8 Typically, individuals taking >3 or 4 medications are identified as a high-risk group. In a 1999 meta-analysis, Leipzig et al7 identified 14 studies examining the association between falls and medication use. Their findings were inconsistent when the end point involved a single fall. However, 4 of the 5 studies examining recurrent falls found significant associations, with the odds of a recurrent fall being 2 to 3 times greater among people taking multiple medications compared with those taking fewer medications. The magnitude of risk associated with specific medications remains controversial. In observational studies, consistent associations have been found between the use ofmedications that have central nervous system (CNS) activity and an increased risk offallS. 9- 11 However, drawing causal inferences from this research is difficult due to a number ofmethodologic issues (eg, the observational nature of most pharmacoepidemiologic studies, confounding by indication, inconsistencies among the medication classification schemes used in different studies).!2 Although information concerning the effect ofspecific medications on the risk of falling continues to emerge,

62

medication review and modification is generally accepted as an important component of multifaceted falls prevention programs. 6 Thus, the "Guideline for the Prevention of Falls in Older Persons," developed as a joint effort by the American Geriatrics Society, the British Geriatrics Society, and the American Academy of Orthopaedic Surgeons,6 highlights the need to review the medications of people who have fallen within the past year and, if possible, reduce the use of high-risk medications, particularly psychotropic drugs. Multifaceted falls prevention programs, which have been found to reduce the risk of falls, usually involve abatement of individual risk factors. 6,l3,l4 The risk factors typically addressed include: home hazards, postural hypotension, use of sedative-hypnotic medications, use of ~4 chronic medications, impairment in tub or toilet transfer, gait or balance impairment, arm or leg weakness, and range-of-motion limitations. After risk identification, targeted intervention strategies are planned and implemented to address each risk factor identified. Although multifaceted falls prevention programs often include a component involving medication review and modification, we are aware of no randomized trials in which this component has been assessed individually. In addition, in most multifaceted falls prevention programs, the medication component is delivered by a registered nurse rather than by a registered pharmacist. Thus, in the current study, we used a randomized clinical trial design to evaluate the effect of a community pharmacy-based medication review and modification program on the risk of falls. Building on the risk abatement intervention approach used in multifaceted interventions, we are targeting individuals at high risk for falls due to their current medication regimens. The interventions are delivered through model community pharmacies offering enhanced clinical services, which include preventive care education and screening. These types of model pharmacies offer a new paradigm for providing clinical pharmacy services in community pharmacy settings. IS In this article, we describe the design of the overall study evaluating the effectiveness of a community pharmacy-based falls prevention program. We also present the algorithms that were developed to identify highrisk patients based on their prescription profile records and to deliver the experimental intervention.

PATIENTS AND METHODS Overview The study is using a randomized controlled research design. After baseline data collection, participants are randomly assigned to either the experimental or control

S. Ferreri et al.

group. Those assigned to the experimental group are offered a free, face-to-face medication consultation provided by a community pharmacy resident. In addition, they are mailed written materials about falls prevention. Control participants receive only the falls prevention written materials. All participants are followed up for 24 months. The primary study end points are: (1) time to first fall; and (2) proportion of participants who experience ~1 fall during the first year of follow-up. The study protocol was approved by the University of North Carolina Institutional Review Board (Chapel Hill, North Carolina). Currently, participant enrollment has been completed. Follow-up procedures are ongoing. Study Sites The study is being conducted in partnership with Kerr Drug, a regional chain of 102 community pharmacies located in North and South Carolina. Kerr Drug currently has 18 community clinical sites. Community pharmacy residents working at 5 of the clinical sites located in North Carolina delivered most of the experimental interventions. Potential study participants were recruited from 32 Kerr Drug pharmacies located within 25 miles of one of the participating clinical community sites. Each Kerr Drug clinical community site has a footprint of -1000 square feet adjacent to the traditional pharmacy where medications are dispensed. Each clinical site is staffed by a full-time clinical pharmacist, and 7 of the sites also provide training for community pharmacy residents. The mission statement of KDI Health Solutions, LLC, a subsidiary of Kerr Drug Inc., is to optimize health care outcomes related to cost, quality, and access, through market leadership in the delivery of innovative community-based clinical services (unpublished data, KDI Health Solutions, LLC, Raleigh, North Carolina, 2008). The clinical sites provide a variety ofservices. These include: drug-therapy monitoring and disease management for patients with asthma, diabetes mellitus, and hypertension; health screening for diabetes mellitus, hypertension, lipid disorders, and osteoporosis; immunizations; smoking cessation; and medication therapy management review clinics. The medication consultations delivered as part of the intervention procedures in this study are similar to those currently offered by clinical pharmacists. Eight residents delivered the intervention: 3 residents in the first year after the initiation of intervention activities and 5 residents in the second year. All residents had recently completed training for their doctorate in pharmacy and were participating in their first postgraduate training experience. In addition, 2 faculty members at

The American Journal of Geriatric Pharmacotherapy

the University of North Carolina at Chapel Hill School of Pharmacy with Doctor of Pharmacy and residency training conducted a total of 8 interventions after the second cohort of residents completed their residencies. Inclusion and Exclusion Criteria Study eligibility criteria were intended to limit the sample to individuals at high risk of falls, particularly falls associated with inappropriate medication use, because these individuals were the most likely to derive the greatest benefit from the intervention. Thus, to be eligible to participate in the study, individuals had to be: (1) aged ~65 years; (2) taking ~4 different chronic prescription medications; and (3) taking ~ 1 of the high-risk medications targeted by the intervention. Although many drugs have been associated with an increased risk of falls, the data are most consistent for CNS-active drugs (eg, benzodiazepines, antidepressants, anticonvulsants, sedative-hypnotics, narcotic analgesics, antipsychotics, skeletal muscle relaxants).7,9,16,17 In addition, use of ~4 medications has been shown to be an independent risk factor for falls.7,9-11,16,17 The specific drugs targeted by the intervention, and incorporated into the study eligibility criteria, are shown in Table I. This list was developed by 2 of the study pharmacists and was based on an extensive review of the literature.7,9,16-22 The list was reviewed by a panel of 3 experienced investigators recognized nationally for their expertise concerning aging. The panel included a pharmacoepidemiologist with expertise in geriatric populations, a geriatrician with expertise in falls and falls prevention, and an epidemiologist with a background in nursing and experience conducting falls prevention programs targeting the frail elderly. To ensure that potential participants were at high risk for medication-related falls, the study: (1) was limited to individuals who had experienced ~1 fall within the 12-month period preceding study enrollment; and (2) excluded individuals who had experienced any falls during that time period which were attributed to syncope. Based on the definition used in the Frailty and Injuries: Cooperative Studies of Intervention Technique trials,23 for the purposes of this study, a fall was defined as "unintentionally coming to rest on the ground, floor, or other lower level." To rule out syncope as the cause ofprevious falls, participants were asked if they passed out or fainted before the fall. Syncope involves the abrupt and transient loss of consciousness. 24 Typically, it is caused by cardiac or neurally mediated factors. Thus, we did not expect the incidence of falls caused by syncope to be affected by the intervention being evaluated in this trial.

63

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

Table I. High-risk medications targeted. Anticonvulsants

Tricyclic antidepressants (cont.)

Benzodiazepi nes

Carbamazepi ne Divalproex sodium Ethosuximide Felbamate Gabapentin Leveti racetam Methsuximide

Clomipramine Desipramine Doxepin Imipramine Maprotiline Nortriptyline Protri ptyli ne Trimipramine

Alprazolam Buspirone Chlordiazepoxide Clonazepam Clorazepate Diazepam Flurazepam Halazepam Lorazepam

Oxcarbazepi ne Phenobarbital Phenytoin Pregabalin Primidone Tiagabine Topiramate Zonisamide

Antidepressants Selective serotonin reuptake inhibitors Citalopram Escitalopram Fluoxetine Fluvoxamine Paroxetine Sertraline

Serotonin-norepinephrine reuptake inhibitors Duloxetine Venlafaxine

Dopamine reuptake blocking agent Bupropion

S-HT2 receptor antagonists Nefazodone Trazodone

Noradrenergic agonist Mirtazapine

Monoamine oxidase inhibitors Isocarboxazid Phenelzine

Tricyclic antidepressants Amitriptyline Amoxapine

64

Antipsychotics and mood stabilizers Chlorpromazine Fluphenazine Haloperidol Loxapine Mesoridazine Molindone Perphenazine Pimozide Thioridazine Thiothixene Trifluoroperazine

Atypical anti psychotics Aripiprazole C10zapine Olanzapine Quetiapine Risperidone Ziprasidone

Skeletal muscle relaxants and antispasmodics Baclofen Carisoprodol Cyclobenzapri ne Methocarbamol Metaxalone Oxybutynin immediate release Tizanidine

Gastrointestinal antispasmodics Clidinium-chlordiazepoxide Dicyclomine Hyoscyamine

Oxazepam Quazepam Temazepam Triazolam Zolpidem

Opioids Codeine Fentanyl Hydrocodone Hydromorphone Levorphanol Meperidine Methadone Morphine Oxycodone Oxymorphone Propoxyphene

Sedative-hypnotics Amobarbital Butabarbital Chloral hydrate Estazolam Mephobarbital Meprobamate Paraldehyde Pentobarbital Secobarbital

Miscellaneous Digoxin Disopyramide

S. Ferreri et al.

In addition, residents of nursing homes and assistedliving facilities were excluded unless they indicated that they managed their own medications. Finally, individuals who were housebound, were unable to read and write in English, or exhibited significant cognitive impairment were excluded because these characteristics would have limited their ability to participate in the intervention.

Patient Recruitment Potential study participants were identified via a central database of prescription records maintained by Kerr Drug. For each participating pharmacy, the database was queried electronically to identifY individuals aged ~65 years who had, within the previous 3 months: (1) filled prescriptions for ~4 different medications; and (2) filled ~1 prescription for one of the high-risk medications targeted by the intervention. A pharmacist then manually reviewed the prescription profiles of the individuals identified to ensure that they appeared to be using ~1 of the high-risk medications targeted. Individuals who met these criteria were sent a letter from Kerr Drug describing the study and soliciting their participation. A brochure that described the study in greater detail accompanied the letter. In addition, a form that individuals could use to indicate their interest in the study and a self-addressed, stamped envelope to return the form were included with the letter. Participants were limited to individuals who either returned the response form indicating an interest in participating in the study or who contacted study personnel in other ways (eg, phone, e-mail) in response to the initial mailing. Individuals interested in participating in the study were interviewed by telephone to assess study eligibility. To assess cognitive function, a 6-item screener derived from the Mini-Mental State Examination was used. 25 Individuals who made >3 errors on the screener were excluded. Before any further data collection, eligible individuals were mailed a consent form and a US Health Insurance Portability and Accountability Act (HIPAA) waiver to sign and return by mail. Data Collection and Randomization Procedures On return of each participant's consent form and HIPAA waiver, baseline data were collected via telephone interview. The interview included questions assessing sociodemographic characteristics, health status, falls history, fear of falling, beliefs concerning current medications (eg, perceived efficacy), medication adherence, medication-related problems (eg, difficulty remembering to take medications), and satisfaction with pharmacy services.

The American Journal of Geriatric Pharmacotherapy

After the interview, participants were randomly assigned to either the intervention or the control group. The allocation sequence used for making group assignments was created using a list of random numbers, generated in blocks of 20 to ensure balancing of group assignment over the duration of participant recruitment. The allocation sequence was concealed from all study personnel except the principal investigator, who had no contact with study participants during the process of data collection or intervention delivery. Immediately after the baseline telephone interview, all participants were mailed a packet of twelve I-month falls calendars and 12 self-addressed, stamped envelopes. This packet also included an "instructional" calendar that provided the definition of a fall used in the study. Participants were asked to record on the calendars each fall and near-fall they experienced and to return the calendars on a monthly basis. Participants were also asked to record dates of health service utilization on the calendar. Finally, the back of the calendar was used to collect information about health status and the occurrence of health problems that could be related to medication changes (eg, difficulty falling asleep or staying asleep). In cases in which a calendar is not returned by day 10 of the month, the participant is contacted, by telephone, to obtain the required information. Data from these calendars are used to create the primary end points for the study (time to first fall and proportion of participants who experience ~1 fall during the 12-month follow-up period). In addition, all individuals who report a fall or near-fall during any given month are contacted by telephone to obtain information concerning the circumstances surrounding the fall; injuries, medical care, or activity restrictions resulting from the fall; and any medication changes that preceded the fall. Participants receive a second set of 12 monthly falls calendars after the 12-month follow-up interview described here. In addition to the monthly falls calendars, follow-up data are collected via quarterly telephone interview at 3, 6,9, and 12 months after the baseline interview. During each interview, the Brief Medication Questionnaire (BMQ) is administered. 26 Using the BMQ, participants are asked to list all of the medications they have taken in the past week. For each medication, participants are asked: (1) the name and strength of the medication; (2) how many days they took the medication; (3) how many times per day they took the medication; and (4) how many pills they took each time. This information will be used to assess changes in medication use over the course of the study. Finally, as a second source of information concerning medication use, we are extracting

6S

The American Journal of Geriatric Pharmacotberapy

S. Ferreri et al.

information from participants' prescription profiles for a 2-year period, beginning 1 year before randomization and continuing for 1 year after randomization. Intervention Procedures

Immediately after the baseline telephone interview, both intervention and control participants were sent a packet of informational materials about the prevention of falls. This packet included 2 brochures available from the Centers for Disease Control and Prevention (What You Can Do to Prevent Falls and Check for Safety: A Home Fall Prevention Checklist for Older Adults).27 The packet also included a refrigerator magnet designed for this project that contained contact information for study personnel. In addition, participants assigned to the intervention group were telephoned and invited to participate in a free, face-to-face medication consultation conducted by a community pharmacy resident at the Kerr Drug clinical site nearest their home. The consultation session was scheduled at a convenient time for the participant, and participants were instructed to bring all of their medications, including over-the-counter drugs and herbal remedies, to the scheduled appointment. During the consultation sessions, which were audiotaped, the resident reviewed the patient's medications and identified drug therapy problems using the model of Strand et al. 28 Special attention was given to medications that have been shown to increase the risk of falls (Table I), with an emphasis on CNS-active medications. 7,9 In addition, the resident solicited information from the patient concerning medical problems associated with an increased risk of falls (eg, orthostatic hypotension, nocturia). To standardize delivery of the intervention, structured algorithms (Appendices A-I) were created by 2 of the study pharmacist investigators (S.P., M.T.R.). The appendices show the algorithms for evaluating the use of the following agents and the risk of falls: anticonvulsants (Appendix A),9,16,19,29 antidepressants (Appendix B),9,16,17,19,30,31 antipsychotics (Appendix C),17,19,29-31 antispasmodics (Appendix D),19,29,31-33 benzodiazepines (Appendix E),8,9,11,16,17,19,29-31,33 opioids (Appendix F),19,31-34 sedative-hypnotics (Appendix G), 8,9,17,29-31,33,34 tricyclic antidepressants (Appendix H),9,11,16,17,19,29-33 and miscellaneous drugs (Appendix 1).8,9,17,29,31-34 The algorithms were reviewed by the same expert panel that reviewed the list of high-risk medications. Feedback from the expert panelists was discussed via in-person meetings and conference calls, and the algorithms were modified to reflect the consensus opinions of both study investigators and expert panelists.

66

During the course of the study, the algorithms were refined when new information about medication risks emerged or new medications were introduced to the market. When conducting a medication consultation, the pharmacy residents used the algorithms to guide the identification of drug therapy problems and the development of therapeutic recommendations. In all cases, however, the residents were instructed to use their clinical judgment and weigh the risk versus benefit of a medication change before making a recommendation. When a drug therapy problem was identified, the resident discussed the problem and potential solutions with the patient. As part of this discussion, the resident assessed patient willingness to make the recommended changes in his or her medication regimen if approved by the physician. As appropriate, residents incorporated patient preferences into their final recommendations. Patients were told that they would be called in a few days to discuss whether the recommended changes had been approved by their physician. Consultation sessions averaged 45 minutes. Follow-up procedures were guided by a written protocol developed to standardize these activities. Specifically, after each medication consultation, the resident prepared a report for the patient's prescribing and primary physician(s) in the standard Subjective Information, Objective Information, Assessment, and Plan note format. Space was provided at the end of the report for physicians to indicate whether they agreed or disagreed with the proposed recommendations (Figure). All reports were reviewed by 1 of the 2 study pharmacist investigators (S.P., M.T.R.) and revised as needed before being faxed by the resident to the prescribing physician and the patient's primary physician. The fax cover letter requested that physicians respond to the recommendations within 48 hours. If a physician failed to respond, the resident made a follow-up contact by telephone. After the physician response was obtained, the resident contacted the patient by telephone to discuss the medication recommendations and implement any authorized changes to the patient's medication regimen. To further standardize delivery of the intervention, each pharmacy resident took part in a half-day training session conducted by study staff. This session included an introduction to useful resources, explanation of the study protocol, discussion ofimportant issues surrounding medication use and falls in the elderly, and review of the algorithms used to guide the medication consultations. Several case examples were presented for discussion. All residents were provided with a copy of Lexi-Comp's Geriatric Dosage Handbook: Including

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

THE FALLS PROJECT UNC Injury Prevention Research Center UNC School of Pharmacy In collaboration with Kerr Drug Patient Name: Jane Doe Date of Birth: 1/1/1921 Dear Dr. Smith: You are receiving this letter because your patient, Jane Doe, has agreed to participate in a randomized controlled trial called The Falls Project. The purpose of the study is to find ways to help older adults decrease their chances of falling. A brochure that describes the study in more detail accompanies this letter. I conducted a comprehensive medication review with Ms. Doe on 3/3/2006 to determine if she is taking any medications that may increase her risk of falling. As part of this review, I asked Ms. Doe to bring with her all of the medications that she is taking, including OTC medications and herbal supplements. I also reviewed the records of the prescription medications that Ms. Doe obtains from Kerr Drug. When I met with Ms. Doe, we discussed how she uses each of her medications. Based on my review and my discussion with Ms. Doe, I would like you to consider the recommendation(s) summarized below. The formulation of these recommendations was guided by a set of algorithms that were developed by clinicians on our research team and reviewed by a clinical pharmacist and a physician who are experts in geriatrics. Copies of the relevant algorithm(s) accompany this letter. Please consider the following recommendations, which may reduce Ms. Doe's chances of falling: Ms. Doe has been using propoxyphene/APAP (Darvocet) since December 2002 for her degenerative disk disease. She takes it regularly, 4 to 5 times each day. As you know, propoxyphene is listed on the Beers criteria and is not recommended in people of advanced age. Propoxyphene offers few analgesic advantages over acetaminophen, yet has the potential to cause significant CNS adverse effects, which may increase Ms. Doe's chance of experiencing another fall. However, she does feel that the medication is working well for her pain and states that Motrin and Tylenol have not helped in the past. Other options for her chronic pain include tramadolSO to 100 mg every 4 to 6 hours if her renal function is adequate or Celebrex 200 to 400 mg twice daily. The use of chronic NSAIDs should be avoided due to the risk of bleeding and ulceration. Ms. Doe has an appointment with you on April 12th, and I informed her that I would let you know of our discussion and that the two of you could decide on a regimen based on risk versus benefit. If you need any additional information, please feel free to call me at XXX-XXX-XXXX. Thank you for your time and consideration. Sincerely,

Deborah Jones, PharmD Kerr Drug

Please indicate below whether or not you agree with the above recommendations. Space is provided for your comments. After you finish, please return this form via fax to XXX-XXX-XXXX. When I receive your reply, I will follow up with Ms. Doe.

COMMENTS: Please sign here if you agree with the recommendations.

Please sign here if you do NOT agree with the recommendations.

Figure. Sample physician report. UNC = University of North Carolina; aTC = over-the-counter; CNS = central nervous system.

67

The American Journal of Geriatric Pharmacotherapy

I S. Ferreri et al.

Monitoring, Clinical Recommendations, and OBRA Guidelines 35 as a reference for more detailed drug information on medications used in the elderly population.

Primary End Points The study has 2 primary end points: (1) time to first fall; and (2) proportion of individuals who experience ~1 fall during the first year of follow-up. Data for creation of both end points will be derived from the monthly falls calendars. Time to first fall will be calculated as the number of days from randomization to the date of the first recorded fall. Secondary End Points Secondary study end points focus on use of the highrisk medications targeted. Data obtained from participant prescription profile records and the BMQ that participants complete will be used to assess changes in the use of these medications that occur after randomization to study group. Specifically, variables will be created to index: (1) change in the number of prescriptions for high-risk medications filled; (2) change in the number of high-risk medications that participants report using; and (3) change in the dosages of high-risk medications used. For each variable, medication use during the year after the date of randomization will be compared with use during the year preceding randomization. Power Analyses Initial power analyses were conducted assuming that 50% of individuals in the control group would experience a fall during the first year of follow-up, with the falls distributed at random (ie, assumption that the hazard is constant). We also assumed that individuals in the experimental group would experience a 25% reduction in the risk of falls, corresponding to a hazard ratio of -0.70. Under these assumptions, and setting a at 0.05, we estimated that a Cox proportional hazards model with time to first fall as the outcome variable would require 262 subjects/group to achieve a power of 0.85. Allowing for a 15% attrition rate over the follow-up period, our initial targeted sample size was set at 310 per group. During the early stages of participant recruitment, the decision was made to add a second year of follow-up to allow more time to observe falls and, thereby, increase the power to detect differences between the experimental and control groups. Approximately 18 months after the initiation of participant recruitment, interim power analyses were conducted when it became apparent that it would be difficult to reach the sample size targeted initially. At the

68

time of this interim analysis, 75 individuals had been randomized and the hazard ratio for experiencing a fall was 0.64 (95% CI, 0.33-1.25; P = NS), favoring the intervention group. This estimate of effect size was then used to update our power analyses. These analyses indicated that 95 subjects per group would provide a power of 0.80 to detect a hazard ratio of 0.64.

RESULTS Participant enrollment began in September 2005 and was completed in August 2007. A total of 186 individuals were enrolled in the study (mean [SD] age, 74.8 [6.9] years; 132 women, 54 men), and 67 have completed the first year of follow-up. Descriptive information concerning the characteristics of study participants is provided in Table II. Primary study findings will be reported after all participants have completed the first year of follow-up in September 2008.

DISCUSSION The study reported in this article is evaluating a falls prevention program delivered via community pharmacies. There is considerable evidence that the active involvement of pharmacists in patient care activities can improve patient health outcomes. 36--40 However, to our knowledge, this is the first study of its type to use community pharmacists to deliver a falls prevention program targeted toward high-risk older adults. A major strength of this study is the use of a randomized controlled design. This design will enhance the internal validity of study findings because it controls for a variety of potential biases. For example, without a control group, any changes in medication use observed among participants in the intervention group could be due to changes in the environment (eg, new practice guidelines issued, publicity concerning a newly recognized medication risk) that are unrelated to the experimental intervention. The use of a control group will allow us to control for these types of potential biases to internal validity. In addition, if participants had been able to select whether they were in the intervention or control group, any differences in the rate of falls or medication use observed could be the result of baseline differences between the groups. For example, it is likely that individuals with greater concerns about their medications and with greater interest in potential medication regimen modifications would be more likely to select the intervention group. Randomization to study group allows us to control for this potential bias. The primary limitations associated with a randomized controlled design involve those regarding generalizability. For example, in this study, we used fairly restrictive

S. Ferreri et al.

Table II. Characteristics of study participants (N

=

186). Age, mean (SO), y Education, no. (%) Less than high school High school only Some college College degree Unknown

74.8 (6.9) 12 (6.5) 33 (17.7) 46 (24.7) 93 (50.0) 2 (1.1)

Sex, no. (%) Female Male

132 (71.0) 54 (29.0)

Race, no. (%) White Black Unknown

165 (88.7) 19 (10.2) 2 (1.1)

Marital status, no. (%) Married Widowed Divorced or separated Never married

110 (59.1) 47 (25.3) 27 (14.5) 2(1.1)

Living alone, no. (%)

65 (34.9)

Health insurance in addition to Medicare, no. (%)

171 (91.9)

Employment status, no. (%) Retired Employed for wages Other

151 (81.2) 15 (8.1) 20 (10.7)

Health problems experienced currently or in the past, no. (%) Back problems Arthritis Dizziness Hypertension Fracture Depression Urinary incontinence Chest pain Hearing problem Cancer Osteoporosis

146 (78.5) 137 (73.7) 135 (72.6) 127 (68.3) 105 (56.4) 85 (45.7) 81 (43.5) 75 (40.3) 70 (37.6) 61 (32.8) 59 (31.7)

inclusion/exclusion criteria. We specifically focused on individuals at high risk of medication-related falls and excluded individuals with cognitive impairment, which

The American Journal of Geriatric Pharmacotherapy

would have made it difficult for them to take part in the intervention. Although we could have included cognitively impaired patients in the study, by delivering the intervention to the primary caregiver of the patient, we decided against this option to minimize sample heterogeneity and, thereby, enhance internal validity. In practice, however, pharmacists routinely work with caregivers when the patient is cognitively impaired. Unfortunately, we will not be able to generalize study findings to these situations. Similarly, study findings may not be generalizable to individuals at lower risk offalls (eg, individuals without a history of falls). The validity ofstudy findings also hinges on our ability to measure the outcomes of interest (ie, occurrence of falls, medication changes). Assessment of falls relies on patient self-report. The falls calendars used in this study represent the state-of-the-art approach to assessing falls. 41 In contrast to intermittent interviews that would require patients to remember falls that had occurred over some past period (eg, falls since the last interview), the calendars allow patients to record any falls they experience immediately after the incident. Thus, recall bias is minimized. Nonetheless, it is likely that patients sometimes forget to record falls they experience, resulting in incomplete data. Although this will result in lower estimates of the incidence of falls, it will not bias the internal validity of the study findings as long as the amount of incomplete data is similar between the intervention and control groups. With respect to our secondary outcomes that focus on the use of high-risk medications, we are using 2 complementary measurement approaches-self-report and review of pharmacy prescription profile records. Although prescription profile records supply objective data, they do not provide a direct assessment of medication use. Rather, they reflect only the possession of medication. Further, prescription profile records may not provide complete information because: (1) patients may use multiple pharmacies; (2) patients may use medications prescribed for family members; and (3) dosage regimens may change without being reflected in the records. Self-report of medication use can partially address these issues. However, self-reports may be prone to social desirability bias. That is, patients may purposefully underreport the use of high-risk medications. Moreover, the magnitude of this bias is likely to be greater in the intervention group, in which patients have been counseled by a pharmacist about medications that increase the risk of falls, and regimen modifications to reduce risk exposure have been discussed. Although both strategies for assessing medication use

69

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

have limitations, by being aware of these limitations, we will be able to compare data collected via the different strategies to increase the validity of the conclusions we draw concerning intervention effects. In conducting the study, we have encountered a number of challenges. First, participant recruitment lagged far behind our projections. As a consequence, we had to revise our power calculations, limiting our ability to detect small intervention effects. Second, we initially intended to collect most data via mailed questionnaire. However, pilot testing revealed that interviewing would be needed to ensure timely completion of data collection instruments. This substantially increased the cost and administrative burden of conducting the study. However, we believe these costs will be offset by increases in data quality (eg, reductions in the amount of missing data, greater consistency in the administration offollow-up data collection instruments). Third, in any study evaluating an intervention, it is important that the intervention be standardized such that all individuals in the intervention arm receive essentially the same intervention. Standardization is relatively easy when the intervention involves an experimental medication. It is much more difficult when delivery of a service is being evaluated, especially when the service is being delivered by multiple individuals at different sites. We addressed this issue by providing training to all residents before they began delivering the intervention. In addition, the residents were closely supervised throughout the study by the 2 study pharmacist investigators, and resident recommendations for medication modifications were reviewed by the supervising investigators before being faxed to the patients' physicians. Thus, although it is not possible to entirely remove variability among residents delivering the intervention, this variability can be minimized by training and supervision.

CONCLUSIONS Falls are a major public health problem among older adults in the United States. The magnitude of this problem is unlikely to decline over the next few decades as the population continues to age. Research suggests that individuals taking multiple medications are at increased risk of falls. CNS-active drugs, in particular, have been associated with increased risk. In this article, we have described the design of a study evaluating the effectiveness of a community pharmacy-based falls prevention program. The study is using a rigorous randomized controlled research design, which will enhance the internal validity of its findings. Results of the study will enable us to assess the effects of the intervention on

70

both medication use and the incidence of falls. These results will be reported after the completion of followup data collection activities. If the intervention is found to be effective, it will provide a resource for community pharmacists working with older adults at high risk of medication-related falls.

ACKNOWLEDGMENTS This project was supported with funds from the National Center for Injury Prevention and Control at the Centers for Disease Control and Prevention to the University of North Carolina Injury Prevention Research Center (R49 CE000196). The authors wish to acknowledge Joseph T. Hanlon, PharmD, MS, and Cathleen S. CoI6n-Emeric, MD, for their assistance with the development and refinement of the algorithms used in this study.

REFERENCES 1. Centers for Disease Control and Prevention (CDC). Falls among older adults: An overview [CDC Web site]. http://www.cdc.gov/ncipc/factsheets/adultfalls.htm. Accessed February 8, 2008. 2. Centers for Disease Control and Prevention (CDC). Web-based Injury Statistics Query and Reporting System (WISQARS) [CDC Web site]. http://www.cdc.gov/ ncipc/wisqars. Accessed July 10, 2007. 3. Hausdorff JM, Rios DA, Edelberg HK. Gait variability and fall risk in community-living older adults: A I-year prospective study. Arch Phys Med Rehabil. 2001;82: 1050-1056. 4. Donald IP, Bulpitt CJ, The prognosis of falls in elderly people living at home. Age Ageing. 1999;28:121-125. 5. Friedman SM, Munoz B, West SI(, et al. Falls and fear of falling: Which comes first? A longitudinal prediction model suggests strategies for primary and secondary prevention. ] Am Geriatr Soc. 2002;50:1329-1335. 6. Rubenstein LZ, Kenny RA, Koval KJ, et al. Guideline for the prevention of falls in older persons. Ann Long-Term Care. 2001;9:42-57. 7. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: A systematic review and metaanalysis: II. Cardiac and analgesic drugs. ] Am Geriatr Soc. 1999;47:40-50. 8. Ziere G, Dieleman JP, Hofman A, et al. Polypharmacy and falls in the middle age and elderly population. Br ] Clin Pharmacol. 2006;61:218-223. 9. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: A systematic review and meta-analysis: I. Psychotropic drugs. ] Am Geriatr Soc. 1999;47:3039.

S. Ferreri et al.

10. Mustard CA, Mayer T. Case-control study of exposure to medication and the risk of injurious falls requiring hospitalization among nursing home residents. Am J Epidemiol. 1997;145:738-745. 11. Ray WA. Psychotropic drugs and injuries among the elderly: A review. J Clin Psychopharmacol. 1992;12:386-396. 12. Agostini TV, Tinetti ME. Drugs and falls: Rethinking the approach to medication risk in older adults. JAm Geriatr Soc. 2002;50:1744-1745. 13. Gillespie LD, Gillespie WJ, Robertson MC, et al. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev. 2003:CD000340. 14. Tinetti ME. Clinical practice. Preventing falls in elderly persons. N Engl J Med. 2003;348:42-49. 15. Christensen DB, Farris KB. Pharmaceutical care in community pharmacies: Practice and research in the US. Ann Pharmacother. 2006;40: 1400-1406. 16. Ensrud KE, Blackwell TL, Mangione CM, et ai, for the Study of Osteoporotic Fractures Research Group. Central nervous system-active medications and risk for falls in older women. JAm Geriatr Soc. 2002;50:16291637. 17. Cumming RG. Epidemiology of medication-related falls and fractures in the elderly. Drugs Aging. 1998;12: 43-53. 18. Evans JG. Drugs and falls in later life. Lancet. 2003; 361:448. 19. French DD, Campbell R, Spehar A, et al. Drugs and falls in community-dwelling older people: A national veterans study. Clin Ther. 2006;28 :619-630. 20. Thapa PB, Gideon P, Cost TW, et al. Antidepressants and the risk of falls among nursing home residents. N Engl J Med. 1998;339:875-882. 21. Smith RG. Fall-contributing adverse effects of the most frequently prescribed drugs. J Am Podiatr Med Assoc. 2003;93:42-50. 22. Kelly KD, Pickett W, Yiannakoulias N, et al. Medication use and falls in community-dwelling older persons [published correction appears in Age Ageing. 2004;33:91]. Age Ageing. 2003;32:503-509. 23. Buchner DM, Hornbrook MC, Kutner NG, et al. Development of the common data base for the FICSIT trials. JAm Geriatr Soc. 1993;41:297-308. 24. Kapoor WN. Current evaluation and management of syncope. Circulation. 2002;106:1606-1609. 25. Callahan CM, Unverzagt FW, Hui SL, et al. Six-item screener to identify cognitive impairment among potential subjects for clinical research. Med Care. 2002; 40:771-781. 26. Svarstad BL, Chewning BA, Sleath BL, Claesson C. The Brief Medication Questionnaire: A tool for screening pa-

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38.

39.

The American Journal of Geriatric Pharmacotherapy

tient adherence and barriers to adherence. Patient Educ Couns. 1999;37:113-124. National Center for Injury Prevention and Control, Division of Unintentional Injury Prevention. What you can do to prevent falls. Check for safety: A home fall prevention checklist for older adults. http://www.cdc.gov/ ncipc/duip/fallsmaterial.htm. Accessed November 16, 2007. Strand LM, Morley PC, Cipolle R], et al. Drugrelated problems: Their structure and function. DICP. 1990;24:1093-1097. Moylan KC, Binder EF. Falls in older adults: Risk assessment, management, and prevention. Am J Med. 2007; 120:493. Landi F, Onder G, Cesari M, et ai, for the Silver Network Home Care Study Group. Psychotropic medications and risk for falls among community-dwelling frail older people: An observational study. J Gerontol A Bioi Sci Med Sci. 2005;60:622-626. Beers MH. Explicit criteria for determining potentially inappropriate medication use by the elderly. An update. Arch Intern Med. 1997;157:1531-1536. Williams CM. Using medications appropriately in older adults. Am Pam Physician. 2002;66:1917-1924. Zhan C, Sangl J, Bierman AS, et al. Potentially inappropriate medication use in the community-dwelling elderly: Findings from the 1996 Medical Expenditure Panel Survey. JAMA. 2001;286:2823-2829. Knight EL, Avorn J. Quality indicators for appropriate medication use in vulnerable elders. Ann Intern Med. 2001;135:703-710. Semla TS, Beizer JL, Higbee MD. Geriatric Dosage Handbook: Including Monitoring, Clinical Recommendations, and OBRA Guidelines. Lexi-Comp's Drug Reference Handbooks. 10th ed. Hudson, Ohio: American Pharmacists Association; 2005. Bunting BA, Cranor CWo The Asheville Project: Longterm clinical, humanistic, and economic outcomes of a community-based medication therapy management program for asthma. JAm Pharm Assoc (2003). 2006;46:133-147. Cranor CW, Bunting BA, Christensen DB. The Asheville Project: Long-term clinical and economic outcomes of a community pharmacy diabetes care program. J Am Pharm Assoc (Wash). 2003;43:173-184. Cranor CW, Christensen DB. The Asheville Project: Shortterm outcomes of a community pharmacy diabetes care program. JAm Pharm Assoc (Wash). 2003;43:149-159. Bluml BM, McKenney JM, Cziraky MJ. Pharmaceutical care services and results in project impact: Hyperlipidemia. JAm Pharm Assoc (Wash). 2000;40:157-165.

71

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

40. Garrett DG, Bluml BM. Patient self-management program for diabetes: First-year clinical, humanistic, and economic outcomes. JAm Pharm Assoc (2003). 2005; 45:130-137.

41. Ganz DA, Higashi T, Rubenstein LZ. Monitoring falls in cohort studies of community-dwelling older people: Effect of the recall interval. J Am Geriatr Soc. 2005;53:2190-2194.

(continued on next page)

Address correspondence to: Susan J. Blalock, PhD, MPH, Division of Pharmaceutical Outcomes and Policy, School of Pharmacy, CB# 7360, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7360. E-mail: [email protected]

72

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix A. Structured algorithm evaluating the use of anticonvulsants and the risk of falls in older adults. The adverse effects associated with anticonvulsants may increase an individual's risk of falling. These agents may cause sedation and dizziness, resulting in the impairment of gait and balance. These effects are more pronounced in the elderly. Therefore, they should be used with caution in this population, especially when an individual is at increased risk of falls. Anticonvulsants as a class have been found to increase the risk of falls and fracture9.16.19.29

Anticonvulsants Carbamazepine Divalproex sodium Ethosuximide Felbamate Gabapentin Levetiracetam Methsuximide Oxcarbazepine Phenobarbital

Phensuximide Phenytoin Pregabalin Primidone Tiagabine Topiramate Trimethadione Zonisamide

I I

I

I

I

I

Concerns

Suggested Approach for Discontinuing Therapy

Suggested Alternatives for Seizures

Suggested Alternatives for Neuropathic Painl Chronic Pain

Suggested Approach for Changing Anticonvulsants

Several anticonvulsants can be problematic for the older population due to their potential for causing adverse effects. The adverse effects associated with the drugs may increase an individual's risk of falling. Phenytoin can cause ataxia, osteopenia, and sedation. Phenobarbital can cause ataxia, memory problems, and sedation. Divalproex sodium can cause tremor, sedation, parkinsonism, and hearing loss. Carbamazepine can cause sedation, neutropenia, and hyponatremia.

Consider slowly tapering patients off the seizure medication if they meet the following criteria.

(1) Patients who have been seizure free with subtherapeutic concentrations. (2) Patients who have been on these drugs for a long time and were placed on anticonvulsants prophylactically or for a few seizures, especially after stroke, neurosurgery, or head trauma.

Alternatives depend on the type of seizure. Use the lowest possible strength for seizure protection. Monitor serum concentrations of medications.

Preferred initial agent for all seizure types in the elderly: Lamotrigine.

Must weigh benefit of treating pain with increased risk of falls. May consider using a tricyclic antidepressant for neuropathic pain.

Refer to Appendix H for tricyclic antidepressants.

The new anticonvulsant should be within therapeutic concentration before tapering the old one. It may take up to 1 year to taper an anticonvulsant during discontinuation or crossover. Use the smallest strength of the medication as necessary. If seizures get worse, you may have to revert to the previous dose and slow down the taper. If adverse effects occur, lowering the dose of the previous medication may help the patient tolerate the new anticonvulsant. Patients should not drive during the taper and for a while after.

73

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

Appendix B. Structured algorithm evaluating the use of antidepressants and the risk of falls in older adults. eNS = central nervous system. It is unclear how anridepressanrs increase an individual's risk of falling. Possible mechanisms include their porenrial co cause sedation and posrural disrurbances, although these effects vary with each agenr and each person. Additionally, anridepressanrs may be indirenly associated with fall risk attributed co faccors such as poor health starus, depression, and weight loss. In studies, anridepressanrs have been found co increase the risk of falls and fracrure 9.16.17.19.30.31 Antidepressants

Selective Serotonin Citalopram Escitalopram Fluoxetine

Reuptake Inhibitors Fluvoxamine Paroxetine Sertraline

Serotonin-Norepinephrine Reuptake Inhibitors Duloxetine Venlafaxine Dopamine Reuptake Blocking Agent Bupropion S-HT2 Receptor Antagonists Nefazodone Trazodone Noradrenergic Agonist Mirtazapine Monoamine Oxidase Inhibitors Isocarboxazid Phenelzine Tranylcypromine Tricyclic Antidepressants Refer to Appendix H.

I

I

I

AVOID

Must weigh benefit of treating depression with increased risk of falls associated with antidepressants.

Paroxetine due co greater anticholinergic properties than other anridepressants, which may increase the risk of falling. Anticholinergic adverse effects include sedation, confusion, dizziness, gait and balance problems, and weakness. AVOID

Fluoxetine due co long half-life, which may be even more pronounced in the elderly, thereby increasing the risk for excessive CNS stimulation, sleep disrurbances, and increasing agitation. AVOID

Fluvoxamine due co drug interactions and availability of effective and safer agents.

Selection of an anridepressant should be individualized, taking into account patient faccors and concomitant medical conditions and medications. Preferred agents include: Citalopram Semaline Escitalopram Bupropion Venlafaxine Duloxetine

AVOID

Mirtazapine due co highly anticholinergic adverse effects. AVOID

Nefazodone, while nor directly linked co falls, is associated with hepacocoxicity and significant drug interactions that limit its use. Alternatives exist that are safer and as effective for treating depression. AVOID

Isocarboxazid, phenelzine, and tranylcypromine in the elderly due to their potential for coxicity and risk of drug-drug and drug-food interactions.

74

Educate patient on the potential for increased sedation, dizziness, and posrural changes from the antidepressant. Monicor closely for adverse effects and falls. Consider switching agent if adverse effects are apparent. There is no one antidepressant or class considered the agent or class of choice in reducing one's risk of falls. The association with antidepressants and fall risk has been attributed to antidepressant agents in general. Consider non pharmacologic approaches, such as behavioral inrerventions, where appropriate.

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix C. Structured algorithm evaluating the use of anti psychotics and the risk of falls in older adults. eNS = central nervous system. Antipsychotics are thought to increase one's risk of falls due to their potential to cause significant adverse effects, including reduced altertness, impaired neuromuscular functioning, sedation, dizziness, postural hypotension, altered gait and balance, and extrapyramidal symptoms. In studies, anti psychotics have been found to increase one's risk of falls. 17,19,29-31

Antipsychotics and Mood Stabilizers Chlorpromazine Fluphenazine Haloperidol Loxapine Mesoridazine Molindone Perphenazine Pimozide Thioridazine Thiothixene Trifiuoroperazine

Atypical Antipsychotics Aripiprazole Clozapine Risperidone Olanzapine Quetiapine Ziprasidone

I AVOID

I

I Suggested Alternatives

Thioridazine due to potential for increased CNS and extrapyramidal adverse effects. This drug has a high incidence of sedation, orthostatic hypotension, and anticholinergic adverse effects, which may increase one's risk of falls.

AVOID

Must weigh risk versus benefit of treating psychoses and the increased risk of falls that has been associated with this class of drugs. If an antipsychotic is needed, consider an atypical antipsychotic at recommended doses for use in the

Mesoridazine due to potential for increased

elderly; however, avoid atypical anti psychotics

CNS and extrapyramidal adverse effects. This drug has

in the elderly individual with dementia. In this case, consider acetylcholinesterase inhibitors.

a high incidence of sedation, orthostatic hypotension, and anticholinergic adverse effects, which may increase one's risk of falls.

Always consider non pharmacologic approaches (eg, behavioral interventions).

AVOID Chlorpromazine due to a high incidence of sedation, orthostatic hypotension, and anticholinergic adverse effects, which may increase one's risk of falls.

AVOID Atypical antipsychotics in elderly individuals with dementia.

75

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

Appendix D. Structured algorithm evaluating the use of skeletal muscle relaxants and antispasmodics and the risk of falls in older adults. Antispasmodics have not been studied in association with increasing fall risk; however, the adverse effects associated with the drugs may increase an individual's risk of falling. These agents are highly anticholinergic and cause sedation, confusion, dizziness, gait and balance problems, and weakness. These effects are more pronounced in the elderly. Therefore, they should be used with caution in this population, especially when an individual is at an increased risk of falls. '9.29.3HJ

Skeletal Muscle Relaxants and Antispasmodics Baclofen Carisoprodol Chlorphenesin Chlorzoxazone Cyclobenzaprine Dantrolene Metaxalone Methocarbamol Orphenadrine Oxybutynin immediate release Tizanidine

Gastrointestinal Antispasmodics Belladonna alkaloids C1idinium-chlordiazepoxide Dicyclomine Hyoscyamine Propantheline

I

I

I

I

AVOID Most Agents

Suggested Alternatives for Treating Spasms or Pain Associated with Muscle Spasms

Suggested Alternatives for Treating Spasms Associated with Neurogenic Bladder or Urinary Incontinence

The benefit of using one of these agents in an elderly individual, especially an individual already at an increased risk of falling, will likely not outweigh the risks and adverse effects associated with these agents.

If the decision is made to use one of these agents in an elderly individual at an increased risk of falls, the following are suggested:

These agents are not recommended to be used in the elderly due to their potential

Use the lowest dose possible. Limit use to 2 to 3 weeks. Document need for

for causing significant adverse effects.

medication in light of fall risk.

Although these agents have not been studied in association with increasing fall risk, the adverse effects associated with the drugs may increase an individual's risk of falling. They are highly anticholinergic and cause sedation,

life than other agents, so it may be preferred if a skeletal muscle relaxant is deemed absolutely necessary.

confusion, dizziness, gait and balance

Consider non pharmacologic approaches,

problems, and weakness. Additionally,

such as exercise and/or physical therapy,

their effectiveness at doses tolerated by

if appropriate.

the elderly is questionable.

76

Other considerations: Methocarbamol has a shorrer half-

If using immediate-release oxybutynin, consider replacing with a longacting formulation, such as extendedrelease oxybutynin or the transdermal oxybutynin formulation. Consider non pharmacologic approaches.

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix E. Structured algorithms evaluating the use of benzodiazepines and the risk of falls in older adults. eNS = central nervous system; SSRls = selective serotonin reuptake inhibitors; SNRls = serotonin norepinephrine reuptake inhibitors. The adverse effects associated with benzodiazepines may increase an individual's risk of falling. These agents are highly anticholinergic and cause sedation, confusion, dizziness, gait and balance problems, and weakness. These effects are more pronounced in the elderly. Therefore, they should be used with caution in this population, especially when an individual is at increased risk of falls. In studies, benzodiazepines as a class have been found to increase the risk of falls and fracture 8 •9,11,16.17,19,29-J1.JJ Benzodiazepines

Alprazolam Buspirone Chlorazepate Chlordiazepoxide Clonazepam Diazepam Flurazepam Halazepam Lorazepam Oxazepam Quazepam Temazepam Triazolam Zolpidem

I

I

I

I

I

Anxiety

Insomnia

Tapering Considerations

Other Indications

I

I

I

I

Suggested Alternatives

Suggested Alternatives

Suggested Alternatives

All suggested alternatives may increase a patient's fall risk. One must determine the risk versus the benefit when selecting an alternative.

The following agents should only be used when all possible reasons for insomnia have been ruled out and behavioral approaches to sleep management (eg, sleep hygiene) have been addressed. The lowest dose possible for a short-term period is recommended. Preferred drug is zaleplon or eszopiclone.

If an indication for the benzodiazepine is not clear, consideration should be given to slowly decreasing the dose. A dose reduction taper should be tailored to the patient's response.

Venlafaxine and duloxetine can be used to treat anxiety, and they are not associated with the same degree of CNS depression as benzodiazepines. SSRls and SNRls can be used to treat anxiety.

Refer to Appendix B for antidepressants.

Consider reevaluating need/indication for the benzodiazepine due to potential for adverse effects, especially falls. It is likely that the risk associated with these agents outweighs any benefit. The lowest dose possible for a short-term period is recommended.

Refer to Appendix G for sedative-hypnotics.

In addition, cognitivebehavioral therapy has been shown to be effective in the management of generalized anxiety disorder.

77

The American Journal of Geriatric Pharmacotherapy

S. Ferreri et al.

Appendix F. Structured algorithms evaluating the use of opioids and the risk of falls in older adults. eNS = central nervous system. The opioids likely increase an individual's risk of falling due to their poremial for causing adverse effeces, including reduced alertness, impaired neuromuscular function, sedation, dizziness, impaired cognition, and unsteadiness or impaired funceioning. In studies, opioids/narcotics have been found co increase one's risk of falls and fracture, although findings are inconsistem. 19.JH4

Opioids Codeine Femanyl Hydrocodone Hydromorphone Levorphanol Meperidine Methadone Morphine Oxycodone Oxymorphone Pemazocine Propoxyphene Nore: The above agems are found in various products and combination produces. I

I

I

AVOID Propoxyphene and combination products. Propoxyphene offers few analgesic advamages over acetaminophen, yet has the poremial to cause significam CNS adverse effeces, which may increase one's risk of falls.

Must weigh benefit of creating pain with one of these agems and the increased risk of adverse effects and falls associated with opioids.

AVOID Levorphanol due co the potemial for significam accumulation of drug and increased risk for eNS adverse effects and CNS coxicity, which may increase one's risk of falls.

Alternatives include, where appropriate: Nonopioid analgesics such as acetaminophen, nonacetylated salicylates, NSAIDs, cramado!, and celecoxib. Note: Avoid indomethacin due to CNS adverse effects; avoid naproxen, oxaprozin, and piroxicam due co increased risk of bleeding, renal failure, high blood pressure, and heart failure. If the opioid is cominued, educate patiem on the poremial for increased sedation, dizziness, unsteadiness, and confusion, and closely monicor for the presence of these adverse effects. Consider the following: Use the lowest dose possible to concrol pain. Limit dose co 1 tablet at a time rather than 1 co 2 tablets. Switch drug if adverse effects are apparent.

78

s.

Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix G. Structured algorithms evaluating the use of sedative-hypnotics and the risk of falls in older adults. eNS = central nervous system; SSRls = selective serotonin reuptake inhibitors; SNRls = serotoninnorepinephrine reuptake inhibitors. The adverse effects associated with sedative-hypnotics may increase an individual's risk of falling. These agents are highly anticholinergic and cause sedation, confusion, dizziness, gait and balance problems, and weakness. These effects are more pronounced in the elderly. Therefore, they should be used with caution in this population, especially when an individual is at increased risk of falls. In studies, sedative-hypnotics as a class have been found to increase the risk of falls and fracture 8 ,9,17,29-31,33,34 Sedative-Hypnotics

Amobarbital Aprobarbital Butabarbital Chloral hydrate Estazolam Ethchlorvynol Glutethimide Mephobarbital Meprobamate Paraldehyde Pentobarbital Secobarbital

I

I

I

I

Anxiety

Insomnia

Other Indications

I

I

I

Suggested Alternatives

Suggested Alternatives

Suggested Alternatives

All suggested alternatives may increase a patient's fall risk. One must determine the risk versus the benefit when selecting an alternative.

The following agents should only be used when all possible reasons for insomnia have been ruled out and behavioral approaches to sleep management (eg, sleep hygiene) have been addressed. The lowest dose possible for a short-term period is recommended. Preferred drug is zaleplon or eszopiclone. Avoid diphenhydramine-containing products (eg, Tylenol PM, Benadryl, Ny to I, Sominex) and doxylamine-containing products (eg, Unisom Nighttime). These agents are highly anticholinergic and cause sedation, confusion, dizziness, gait and balance problems, and weakness. Additionally, their effectiveness for sleep is questionable. Avoid trazodone due to lack of efficacy data and increased risk for adverse effects, such as orthostatic hypotension and anticholinergic effects.

It is likely that the risk associated with this class outweighs any benefit.

Venlafaxine and duloxetine can be used to treat anxiety, and they are not associated with the same degree of CNS depression as sedative-hypnotics. Benzodiazepines and cognitive-behavioral therapy can be used to treat anxiety.

Refer to Appendix Efor benzodiazepines. SSRls and SNRls can be used to treat anxiety.

Refer to Appendix 8 for antidepressants.

Nonpharmacologic measures, including behavioral therapy and lifestyle modifications, are the mainstay for management of insomnia in older adults.

79

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix H. Structured algorithms evaluating the use of tricyclic antidepressants (TeAs) and the risk of falls in older adults. TCAs are associated with a high incidence of anticholinergic adverse effects, including reduced alertness, impaired neuromuscular functioning, sedation, dizziness, postural hypotension, altered gait and balance, and confusion. In studies, TCAs have been associated with an increase risk of falls 9 ,l1.16.17,19,29-33

TCAs Amitriptyline Amoxapine

Clomipramine Desipramine

Doxepin Imipramine

r

Maprotiline Nortriptyline

I

I

I

Depression

Neuropathic Pain! Chronic Pain

Insomnia

Other Indications

I

I

I

I

Suggested Alternatives

Suggested Alternatives

Suggested Alternatives

Suggested Alternatives

Citalopram Sertraline Escitalopram Bupropion Venlafaxine Duloxetine

Must weigh benefit of treating pain with increased risk of falls.

The following agents should only be used when all possible reasons for insomnia have been ruled out and behavioral approaches co sleep management (eg, sleep hygiene) have been addressed. The lowest dose possible for a short-term period is recommended.

Consider reevaluating need/indication for the TCA due to potential for adverse effects, especially falls. It is likely that the risk associated with this class outweighs any benefic.

AVOID Paroxetine Fluoxetine Nefazodone Fluvoxamine Mirtazapine

Refer to Appendix B for antidepressants. AVOID Amitriptyline Amoxapine Doxepin Imipramine Protriptyline Trimipramine due co pocential for strong anticholinergic and sedating properties, which increase fall risk.

If the TCA is used as adjuvant therapy in the treatment of pain, and effectiveness has been demonstrated, ensure that the individual is on the lowest dose possible to control the pain and minimize adverse events. If TCAs are needed, consider nortriptyline or desipramine. Alternatively, consider lowdose gabapentin, tramado!, pregabalin, or duloxetine for neuropathic pain.

Preferred drug is zaleplon or eszopiclone. Avoid diphenhydraminecontaining products (eg, Tylenol PM, Benadry!, Nyco!, Sominex) and doxylamine-containing products (eg, Unisom Nighttime). These agents are highly anticholinergic and cause sedation, confusion, dizziness, gait and balance problems, and weakness. Additionally, their effectiveness for sleep is

AVOID

questionable.

Isocarboxazid, phenelzine, and tranylcypromine.

Avoid trazodone due co

Refer to Appendix B for antidepressants,

lack of efficacy data and increased risk for adverse effects, such as orthostatic hypotension and anticholinergic effects.

Refer to Appendix Gfor sedative-hypnotics.

80

Protriptyline Trimipramine

S. Ferreri et al.

The American Journal of Geriatric Pharmacotherapy

Appendix I. Structured algorithm evaluating the use of miscellaneous drugs and the risk of falls in older adults. The following drugs or classes of medications are associated with either an increased risk of falls, adverse drug effects that may contribute to falls, or drug effects that may lead to more serious adverse consequences if one falls.

Medication or Class

Association with Falls

Comments

Digoxin>0.125 mg/d9.17.29.Jl

Due to the extremely long half-life in the elderly, this agent may increase the risk of confusion and dizziness and may increase the risk of falls.

Doses >0.125 mg/d add risk without improving outcomes. Serum digoxin concentration should be between 0.5 and 0.9 ng/mL to improve patient outcomes without adverse effects. It is likely that the risk associated with doses >0.125 mg/d outweighs any benefit. Consider decreasing the dose of digoxin or selecting an alternative agent.

Disopyramide and all Class IA antiarrhythmic medications9.J1.32

Disopyramide has strong anticholinergic properties and has been linked to increasing fall risk in the elderly.

Antihypertensive agents 8,9,17,29.J4

May contribute to hypotension or orthostatic hypotension, which may result in Iightheadedness, dizziness, and/or unsteadiness on feet. and may increase one's risk of falling

Any medication with strong anticholinergic effects 29.J1-34

For example, antihistamines, antiemetics.

Anticoagulants and anti platelet agents 31.32.J4

These drugs do not increase one's risk of falling, but they may increase one's risk of bleeding secondary to a fall, so extra caution should be used in weighing risk versus benefit of treatment.

Antidiabetic agents 31 ,33,34

Can lead to hypoglycemia, which may result in lightheadedness, dizziness, and/or unsteadiness on feet, and may increase one's risk of falling.

Acetylchol inesterase in hibitors 31J3

Cognitive impairment is associated with an increased risk of falls; therefore, close monitoring of individuals with cognitive impairment (including those treated with acetylcholinesterase inhibitors) is important.

Antiparkinsonian medications 31

Parkinson's disease is associated with an increased risk of falls; therefore, close monitoring of individuals with Parkinson's disease (including those treated with anti parkinsonian medications) is important.

It is likely that the risk associated with this agent outweighs any benefit. Consider selecting an alternative antiarrhythmic agent.

81