Journal of Gerontological Nursing

Feature Article 

Factors Associated With Potentially Inappropriate Medication Use in Rural, Community-Dwelling Older Adults

Marcia Y. Shade, PhD, MSN, RN; Ann M. Berger, PhD, APRN, AOCNS, FAAN; Claudia Chaperon, PhD, APRN, GNP-BC; Gleb Haynatzki, PhD, DSc; Linda Sobeski, PharmD; Bernice Yates, PhD, RN

Abstract

Potentially inappropriate medication (PIM) use is a serious public health problem in older adults because it may lead to adverse events. The purpose of the current study was to explore PIM use in rural, community-dwelling older adults. Participants (N = 138) underwent one-on-one medication reviews. Approximately one half (49%) of the sample used prescribed and over-the-counter (OTC) PIM. Prescribed and OTC nonsteroidal anti-inflammatory drugs (33%) and anticholinergic medications (28%) were the most frequently used PIM. Use of PIM was associated with a higher number of medications (r = 0.331, p < 0.01), more medical providers (r = 0.223, p < 0.001), and poor physical health (r = −0.193, p < 0.05). Higher number of medications increased the probability of PIM use by 85% (odds ratio: 1.8; 95% confidence interval [1.19, 2.84]). Findings highlight the importance of re-evaluating the monitoring of medications in rural, community-dwelling older adults and the need for sustainable interventions to reduce prescribing and OTC PIM use. [Journal of Gerontological Nursing, 43(9), 21–30.]

Abstract

Potentially inappropriate medication (PIM) use is a serious public health problem in older adults because it may lead to adverse events. The purpose of the current study was to explore PIM use in rural, community-dwelling older adults. Participants (N = 138) underwent one-on-one medication reviews. Approximately one half (49%) of the sample used prescribed and over-the-counter (OTC) PIM. Prescribed and OTC nonsteroidal anti-inflammatory drugs (33%) and anticholinergic medications (28%) were the most frequently used PIM. Use of PIM was associated with a higher number of medications (r = 0.331, p < 0.01), more medical providers (r = 0.223, p < 0.001), and poor physical health (r = −0.193, p < 0.05). Higher number of medications increased the probability of PIM use by 85% (odds ratio: 1.8; 95% confidence interval [1.19, 2.84]). Findings highlight the importance of re-evaluating the monitoring of medications in rural, community-dwelling older adults and the need for sustainable interventions to reduce prescribing and OTC PIM use. [Journal of Gerontological Nursing, 43(9), 21–30.]

Older adults are at risk for using potentially inappropriate medications (PIM). PIMs are any pharmaceutical products that have been determined to have risks that outweigh benefits (Beers, 1997). The regularly updated Beers Criteria is a resource derived from scientific evidence associating PIM use with the risk of negative health outcomes (American Geriatrics Society [AGS] 2012 Beers Criteria Update Expert Panel, 2012; AGS 2015 Beers Criteria Update Expert Panel, 2015; Fick et al., 2003). PIMs have been studied over two decades in terms of health provider prescribing patterns and have been associated with adverse drug events (ADEs), defined as any harm or injury resulting from pharmacological medical intervention. ADEs can comprise medical errors, adverse drug reactions, allergic reactions, and overdoses (Institute of Medicine, 2007). An estimated 42.6% of community-dwelling older adults are prescribed PIMs, such as non-steroidal anti-inflammatory drugs (NSAIDs), anticholinergic agents, nonselective alpha-1 blockers, non-benzodiazepine hypnotic agents, estrogens, and skeletal-muscle relaxants (Davidoff et al., 2015; Kachru, Carnahan, Johnson, & Aparasu, 2015).

Special attention should be given to PIM use among older adults in rural areas. In the United States, individuals 65 and older comprise 18% of residents in rural counties versus 12% in urban settings (Meit et al., 2014). In Nebraska, one third of residents live in rural counties and an estimated 34% are 60 and older (U.S. Department of Health and Human Services, 2015). In comparison to urban settings, older adults in rural settings have higher rates of chronic illness, poorer overall health, and less access to care (Meit et al., 2014).

Poor health experiences of rural, community-dwelling older adults may be manifested by multiple comorbidities, managing symptoms of disease, and seeking health care from multiple providers (Arcury et al., 2012). Chronic illness in independent older adults may lead to changes in health status, presenting as poor physical and mental health, poor sleep, and pain (Lee et al., 2013; Shade, Berger, Dizona, Pozehl, & Pullen, 2016; Zhan et al., 2001). Sleep is often overlooked in association with health status and older adults do not routinely report sleep problems to their providers (Ancoli-Israel, 2009; Vaughn McCall, 2004). Sleep problems have been associated with chronic illnesses, such as depression, changes in circadian rhythm, and use of medications (AGS 2012 Beers Criteria Update Expert Panel, 2012; Ancoli-Israel, 2009). Individuals may seek potentially inappropriate sedative-hypnotic agents from primary care providers, and others may self-medicate with alcohol or over-the-counter (OTC) sleep aids, with or without the knowledge of a health care provider (Basu, Dodge, Stoehr, & Ganguli, 2003).

Use of prescribed or OTC NSAIDs can cause cardiac, renal, and other adverse events (e.g., falls), especially when combined with multiple medications (Barkin et al., 2010; Findley & Bulloch, 2015). Use of NSAIDs is potentially inappropriate and must be carefully considered and monitored (Gnjidic et al., 2014). Concepts such as anticholinergic burden and sedative load in older adults must be considered when assessing medication and PIM use (Fox et al., 2011; Taipale, Bell, Gnjidic, Sulkava, & Hartikainen, 2012). Adults in rural areas may rest, reduce activities, and take OTC medicine to manage comorbidities and symptoms of disease (Arcury et al., 2012). Older adults have reported taking four OTC medications in addition to prescribed medications, and providers may not be aware of all drug use (Chui, Stone, Martin, Croes, & Thorpe, 2014). These factors may lead to inappropriate management of medications, exacerbation of chronic diseases, and ADEs in older adults in rural settings.

PIM use in U.S. community-dwelling older adults has been associated with older age, use of higher numbers of medications, and poorer physical and/or mental health status (Guaraldo, Cano, Damasceno, & Rozenfeld, 2011; Lee et al., 2013; Maher, Hanlon, & Hajjar, 2014; Zhan et al., 2001). One study previously examined health care use factors in association with PIM use in rural, community-dwelling older adults (Blalock et al., 2005). Greater medication needs, lower social support, higher levels of disability, and being diagnosed with chronic conditions have been associated with greater PIM use in rural older adults. PIM use in rural communities (population <50,000 individuals) has also been associated with lack of geriatric care providers (Blalock et al., 2005).

Studies have primarily focused on prescribed PIMs and may have overlooked OTC medications. Patient medical records may not include OTC or current prescriptions. One study reported medication lists in medical charts were only accurate 6.5% of the time (Sarzynski, Luz, Rios-Bedoya, & Zhou, 2014). Research on PIM has been limited from the nursing perspective, in rural areas, and among independent, community-dwelling older adults (Shade, Berger, & Chaperon, 2014). Therefore, the purpose of the current study was to explore the use of and factors associated with PIM use in rural, community-dwelling older adults. The specific aims of the study were to (1) describe characteristics of PIM to avoid and (2) explore the relationships between client singularity and PIM to avoid.

Method

Design and Conceptual Framework

The current study used a descriptive design. An adapted version of Cox's (1982) Interaction Model of Client Health Behavior (IMCHB) guided the study. The IMCHB is a middle-range nursing theory that characterizes interactions that occur between client singularity, client–professional interaction, and health outcomes. Client singularity comprises extrinsic and intrinsic variables. Concepts may be transformed into variables to represent a health care issue.

The Figure displays how the IMCHB was adapted for the current study. The model examined relationships between client singularity variables and a single health outcome. Client singularity included extrinsic background variables of demographic characteristics and health experience. Intrinsic variables included older adults' self-reported health status (physical and mental health and sleep behaviors). The health outcome was a health behavior of PIM use.

Research model of factors associated with potentially inappropriate medication use in community-dwelling older adults, adapted from Cox's (1982) Interaction Model of Client Health Behavior.Note. PROMIS = Patient-Reported Outcomes Measurement Information System.

Figure.

Research model of factors associated with potentially inappropriate medication use in community-dwelling older adults, adapted from Cox's (1982) Interaction Model of Client Health Behavior.

Note. PROMIS = Patient-Reported Outcomes Measurement Information System.

Sample/Setting

A convenience sample of older adults was recruited from a rural family practice clinic in the Midwestern United States from January through October 2015. The rural setting was determined according to rural urban commuting area codes 7 (defined as a small rural town) and 10 (defined as an isolated small rural town). Participants were screened by the researcher (M.Y.S.) according to inclusion and exclusion criteria. Inclusion criteria were: males and females 65 and older, taking three or more prescribed medications, living independently in the community, responsible for personal medication regimen, able to physically complete questionnaires, read and speak English, not taking medications for cognitive impairment, and scoring ≥26 on the Montreal Cognitive Assessment (Nasreddine et al., 2005).

Measurements

Demographic Characteristics. Demographic data included age, gender, education, zip code, marital status, and household income.

Health Experience. Participants were asked about the number of medications taken as well as their current number of health providers. Medications taken in the past month included: prescribed, OTC, herbal, vitamins, and supplements. Names of drugs, dosage, frequency, and length of use were verified with patient records. Routine for taking medications, rationale, length of use, and pharmacies were recorded. Medical diagnoses were recorded from participants' medical records. The Charlson Comorbidity Index was used to assess comorbidity. The index is weighted and scores are assigned for each condition and individual's age and then summed to predict mortality. The 19 conditions were scored from 1 to 6 depending on the risk of dying associated with each one, with higher scores indicating greater risk of dying from the condition; an additional 1 point was added for each decade over age 40. Higher scores indicated higher severity of comorbidity (Charlson, Szatrowski, Peterson, & Gold, 1994).

Health Status. The 12-Item Medical Outcomes Survey Short-Form describes patient-reported physical and mental health status and measures health-related quality of life. The 12-item short-form summary of physical component scores (PCS) and mental component scores (MCS) range from 0 to 100, with higher scores indicating better health. In individuals 65 and older, norm scores have been reported as a PCS of 43.4 and MCS of 55.2 (Ware, Kosinski, & Keller, 1996). Cronbach's alpha was 0.77 for PCS and 0.73 for MCS in the current study.

Sleep was measured using subjective and objective instruments. The Pittsburgh Sleep Quality Index (PSQI) measures perceived sleep quality using 19 self-response items. Five questions related to bed partner were not included in analysis. The PSQI includes seven component scores (i.e., sleep quality, sleep duration, sleep latency, sleep disturbances, daytime dysfunction, habitual sleep dysfunction, and sleep medication use). Each is weighted equally on a scale from 0 to 3 and summed for a global score ranging from 0 to 21, with high scores indicating poor sleep quality. The global scale has a cutoff score >5 to indicate poor sleepers (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Cronbach's alpha was 0.74 in the current study.

The Patient-Reported Outcomes Measurement Information System (PROMIS)-Sleep Disturbance Short Form-8A is a patient-reported sleep measure (Cella et al., 2010). The questionnaire contains eight items for sleep disturbance and sleep-related impairment. A score of 50 (SD = 10) is considered average for the general U.S. population. T-scores >50 indicate poor sleep. Cronbach's alpha was 0.88 in the current study.

Sleep–wake patterns were measured by wrist actigraphy. The ActigraphGT3X® wrist accelerometer is a noninvasive method of indirectly estimating sleep–wake and activity–rest patterns. The device is small and worn on the non-dominant wrist. It records activity–rest patterns, physical activity, or sleep.

Total sleep time (TST), wake after sleep onset (WASO), and wake percentage were analyzed from the actigraph. TST is the number of minutes of actual sleep while in bed, and WASO minutes are the amount of time spent awake after the initiation of sleep (Schutte-Rodin, Broch, Buysse, Dorsey, & Sateia, 2008). Wake percentage is the ratio of TST to the amount of time spent in bed divided by 100. Normal values are considered 7 to 9 hours for TST, <30 minutes or 10% of time in bed for WASO, and a wake percentage of ≤15%. Values not within normal limits may indicate sleep disturbance (Berger et al., 2008). Participants wore the actigraph for seven 24-hour cycles and participants with four or more nights of data were included. Sleep data were analyzed using the Cole Kripke algorithm in the Actilife software by Actigraph®. The sleep diary in-and-out-of-bed times were used to assist processing and analysis of the actigraph data.

Potentially Inappropriate Medications. To identify PIM, prescribed and OTC medications taken in the previous month were compared to the 2012 Beers Criteria that were derived from a framework and rigorous systematic literature review. The review was conducted by an expert consensus panel that graded the quality and strength of the evidence. The 2012 Beers Criteria lists 53 medications or medication classes divided into three categories: (a) recommended medications and classes to avoid in older adults, (b) medications to avoid in older adults with certain diseases, and (c) medications to be used with caution in older adults (AGS 2012 Beers Criteria Update Expert Panel, 2012). The current study focused on medications to avoid in older adults.

Procedures

The University of Nebraska Medical Center's Institutional Review Board approved the study. Each enrolled participant was taught how to complete the questionnaires and sleep diary and how to wear the wrist actigraph. Participants were instructed to bring all prescribed, OTC medications, vitamins, and herbal supplements taken in the past month from home in red bag(s) to the follow-up visit at the clinic.

After 1 week, participants returned all completed questionnaires, the sleep diary, wrist actigraph, and red bag(s). The researcher (M.Y.S.) and participant met in a private space at the clinic. Containers of prescribed, OTC, and supplement home medications taken within the past month were reviewed by interview with the participant and recorded on a medication history form by the researcher. Review questions asked how often and long the medication/vitamin/supplement was taken, and for what diagnosis. Participants were specifically asked about the duration of use for pain medications and if they use substances such as caffeine or alcohol that may interfere with pharmacokinetics. Allergies and history of ADEs were also assessed. Chronic use was defined as the weekly use of a medication for the past 3 months. All medications/vitamins/supplements were returned to the participant along with information about medication safety and a $25 gift card.

The researcher categorized each medication as PIM or non-PIM on an investigator-derived PIM log that listed the 2012 Beers Criteria. Each medication (PIM versus non-PIM) was reviewed with a pharmacist with expertise in geropharmacology (L.S.). Medications were discussed until consensus was achieved at 100% agreement. Participants' primary care providers were updated on results of the review.

Statistical Analysis

An a priori conservative sample size and power analysis was performed based on the logistic regression for Aim 2 using the 12-item short-form PCS. The conclusion was that a sample 106 individuals would achieve 80% power at a significance level (alpha) of 0.05. For Aim 1, PIMs were categorized and coded by dummy variables (yes = 1, no = 0). Descriptive statistics were calculated to determine the proportion of participants and category of PIM.

For Aim 2, univariate descriptive statistics were calculated to analyze variables of demographic characteristics, health experience, and health status factors. Demographic characteristics were coded as dummy variables (gender [male = 1, female = 2]; marital status [married or cohabitating = 1, single or widowed = 2]; highest education [high school graduate = 1, college and above or trade school = 2]; income [below poverty level = 1, above poverty level = 2]; and home location [small rural town = 7, rural area = 10]). The number of medications and health providers were counted, and values were used from subjective questionnaires and actigraphy instruments. Chi-square tests and point biserial correlations were performed to explore relationships between variables.

A binary logistic regression analysis was performed to determine which statistically significant variables predicted the probability of PIM to avoid. The dependent variable was coded as PIM use (yes = 1 and no = 0). Numerical predictor variables were transformed into standardized form due to unit variations. Overall model evaluation was performed using the Hosmer–Lemeshow test for logistic regression to test the goodness of fit (Lemeshow & Hosmer, 1982). Statistical analyses were performed using SPSS version 23.

Results

Sample Characteristics

Recruitment occurred from January through October 2015, and 141 older adults enrolled in the study. A total of three participants withdrew (two for health reasons and one after completing the questionnaires). Table 1 describes the sample. Table 2 displays the mean PCS of 38.7 and MCS of 53.2, indicating poorer self-reported physical and mental health than expected for adults 65 and older. Several participants reported pain (84%) that interfered with normal work, and several reported pain interferences at moderate (27%) to extreme (28%) levels. A PROMIS score of 46.9 reflected sleep disturbance no different than the norm, but the mean PSQI score of 6.5 indicated poor sleep quality. Per actigraphy, TST and wake percentages were within normal limits, and mean WASO minutes per night (50.6 minutes) were slightly more than 10% of TST (456 minutes).

Participant Characteristics (N = 139)

Table 1:

Participant Characteristics (N = 139)

Sample Characteristics of Health Experience and Health Status Variables

Table 2:

Sample Characteristics of Health Experience and Health Status Variables

Potentially Inappropriate Medication Use

Prevalence of the use of PIMs to avoid was 49%; 32% of the sample used at least one and 17% used two or more PIMs. Of participants who took PIMs, 51% used prescribed PIMs, 24% used OTC PIMs, and 25% used prescribed and OTC PIMs. Table 3 lists the frequency of the common classifications of PIMs. NSAIDs, medications with anticholinergic properties, short-acting benzodiazepines, and non-benzodiazepine receptor agonists were the most frequently used PIMs. Of anticholinergic agents, antihistamines (singular or in combination with pain relievers) were used most often OTC to assist sleep. Most NSAIDs were acquired OTC for pain relief.

Characteristics of PIMs to Avoid

Table 3:

Characteristics of PIMs to Avoid

Table 4 shows several significant relationships between PIMs and client singularity variables. Higher use of PIMs was associated with higher number of medications, seeing more providers, poor physical health, and poor sleep quality (p < 0.05). Use of PIMs was not associated with comorbidity, mental health, self-reported sleep disturbances, or objective sleep or wake parameters (p > 0.05).

Correlations Between Demographic, Health Experience, and Health Status Variables and PIMs Use (N = 138)

Table 4:

Correlations Between Demographic, Health Experience, and Health Status Variables and PIMs Use (N = 138)

Predictors of Potentially Inappropriate Medication Use

Table 5 shows results of a logistic regression analysis conducted to predict the probability of PIM use. Client singularity variables that were significantly related to higher use of PIM were entered as predictors. The overall model was examined for fit and 17% of the variance in PIM use was explained by this model. Higher number of medications was the only significant predictor of the increased probability of PIM use. Every increase in 1 count of medication (prescribed and OTC) increased the odds of PIM use by 85% (95% confidence interval [1.19, 2.84]).

Summary of Binary Logistic Regression Analysis Predicting Probability of PIMs Use

Table 5:

Summary of Binary Logistic Regression Analysis Predicting Probability of PIMs Use

Discussion

PIM use was prevalent among the current sample of rural, community-dwelling older adults. Forty-nine percent of participants took at least one PIM, which is higher than the national estimated prevalence of 42%. Consistent with national results, participants frequently used NSAIDs and medications with anticholinergic properties (Davidoff et al., 2015; Kachru et al., 2015). PIM use was associated with use of a higher number of medications, use of a higher number of health providers, poor self-reported physical health, and a trend of poor sleep quality. The only significant predictor of an increased probability of PIM use was a higher number of medications.

Prior studies of PIM use have focused on the prescribing patterns of health care providers. One half of older adults in the current study took prescribed PIM. Prescriptions are written based on patient status and provider expertise. The benefit may outweigh the risk, and the prescribing of PIM may be the only viable treatment option. Prevalence of prescription PIM use may reflect the pharmaceutical management of chronic illness in rural areas (Meit et al., 2014). Client singularity appears to contribute to prevalence rates, as users took prescribed and OTC PIM. Consistent with the findings from Arcury et al. (2012), independent older adults in the current sample took OTC medicine to manage symptoms of disease.

Pain was a common symptom; it was not surprising to find participants frequently using PIM with anti-inflammatory properties. Most NSAIDs were obtained OTC, highlighting the importance of provider pain assessment and prevention of long-term NSAID use. Unsupervised NSAID use may lead to gastrointestinal injury, fluid retention, renal failure, and increased bleeding time in older adults (Gnjidic et al., 2014). Safer alternatives must be used to manage pain in older adults.

Actigraphy showed sleep quantity to be within normal limits, but self-reported sleep quality was poor. Consistent with findings from Lee et al. (2013) and Basu et al. (2003), medications with benzodiazepine properties were prescribed to promote sleep. Participants' use of OTC antihistamines as sleep aids and prescribed benzodiazepines were of concern. Medications with sedative and anticholinergic properties cause falls, dry mouth, dry eyes, dizziness, confusion, and constipation, and increase the risk of delirium and dementia. Sleep problems may be perceived as an inevitable result of aging, and patients may have ambivalent feelings toward taking medications to promote sleep. Nevertheless, readily available OTC medications are being acquired to self-manage poor sleep quality and pain symptoms.

PIM use was associated with a higher number of medications and treatment from more health providers. Unlike previous reports, no association was found between PIM use with older age (Guaraldo et al., 2011). Participants in the current study had a higher mean severity of comorbidity and could not be compared to the modified Charlson Comorbidity Index used by Blalock et al. (2005).

In the current study, PIM use was associated with poorer physical health, but not mental health, similar to findings from Zhan et al.'s (2001) study. Compared to urban areas, individuals in rural communities are less likely to have access to mental and behavioral health specialists (Thomas, Macdowell, & Glasser, 2012). Older adults may not be routinely prescribed PIMs for the treatment of mental or behavioral health. In addition, few OTC medications exist for behavioral health, and older adults may seek supplements that are not included on the Beers Criteria.

Higher number of medications was the only significant predictor for an increased probability of PIM use. Consistent with the study by Maher et al. (2014), polypharmacy increased the chance of using PIM. Community-dwelling older adults now have access to OTC medications that were once only available through provider prescription. Medications are marketed to promote pharmaceutical independence in a society that encourages self-management of disease. The current findings require health professionals to identify and use new strategies to partner with older adults to prevent inappropriate polypharmacy, deprescribe PIM, and optimize medication use (Brandt, 2016).

Strengths and Limitations

The current study was the first to use a nursing model to describe factors associated with PIM use. The model was beneficial for demonstrating the role of client singularity in PIM use and may be helpful in the development of nursing interventions to reduce PIM use. Prior studies of PIMs have usually been conducted in acute or long-term care settings, with minimal information about rural community settings. A modest sample of rural older adults was obtained to explore the concepts. An additional strength is that medications were recorded from interviews and participants' medication and supplement containers instead of the medical record or patient lists (Sarzynski et al., 2014). PIMs were reviewed with a pharmacist (L.S.) with expertise in geropharmacology. Indirect objective measures of sleep–wake patterns were recorded in addition to self-reported sleep. Actigraph measures were complete for analysis because 94% of participants wore the actigraph for the duration of the study.

The current study also had limitations. Generalizability is limited because the convenience sample was mostly Caucasian. Participants were from a Midwest farming community, from one clinic in one rural small town, and selection bias may have occurred. In addition, participants may not have brought all medications taken in the past month in the red bags.

Implications for Nursing Research and Practice

PIM use must be evaluated using a standard method such as the Beers Criteria. Prospective work could explore PIMs in multiple outpatient settings and medical specialties. Studies must be conducted in growing diverse samples of rural older adults. Interventions must be initiated by nurse-led or multi-disciplinary teams, including physicians and pharmacists. A rigorous research design can longitudinally examine and compare health outcomes based on duration of PIM use. In the rural primary care setting, nurses can collaborate with the health care team to develop interventions to reduce and monitor use of prescribed and OTC PIMs (National Conference of State Legislatures, n.d.).

Patient-centered interventions may be community-based or outpatient. Screening tools that include the updated version of the Beers Criteria and patient-reported ADEs can be tested and used by health care staff in primary care clinics (AGS 2015 Beers Criteria Update Expert Panel, 2015). Researchers can conduct randomized controlled trials that promote team interventions to reduce commonly used PIMs. Future studies can compare nonpharmacological interventions to the use or tapering of a pharmacological intervention so researchers can examine the influence of interventions on health outcomes.

Initial nursing assessments can encourage alternative strategies for sleep problems and pain in older adults. Medication reconciliation has been encouraged, but medications must be critically appraised, focusing on the duration of NSAID use and medications with central nervous system and anticholinergic effects. A primary care protocol could promote:

  • regular interactive medication reviews with deprescribing (nurse initiated with pharmacist available and updated Beers Criteria);
  • assessment of patient pharmaceutical literacy and provision of tailored patient education and resources (PIM and ADEs);
  • mandatory follow up to evaluate patients taking new medications or changes to current regimen; and
  • technology to monitor ADE emergency department visits and outcomes of reducing PIM.

Conclusion

PIM use is a serious public health concern. The current study confirmed that despite evidence of risk, use of prescribed and OTC PIMs is prevalent in rural settings. PIM use in rural, community-dwelling older adults impacts the patient/family and health care team. Health care professionals in rural primary care settings must collaborate with community-dwelling older adults to reduce PIM use.

References

  • American Geriatrics Society 2012 Beers Criteria Update Expert Panel. (2012). American Geriatrics Society updated Beers criteria for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society, 60, 616–631. doi:10.1111/j.1532-5415.2012.03923.x [CrossRef]
  • American Geriatrics Society 2015 Beers Criteria Update Expert Panel. (2015). American Geriatrics Society 2015 updated Beers criteria for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society, 63, 2227–2246. doi:10.1111/jgs.13702 [CrossRef]
  • Ancoli-Israel, S. (2009). Sleep and its disorders in aging populations. Sleep Medicine, 10(Suppl. 1), S7–S11. doi:10.1016/j.sleep.2009.07.004 [CrossRef]
  • Arcury, T.A., Grzywacz, J.G., Neiberg, R.H., Lang, W., Nguyen, H., Altizer, K. & Quandt, S.A. (2012). Older adults' self-management of daily symptoms: Complementary therapies, self-care, and medical care. Journal of Aging and Health, 24, 569–597. doi:10.1177/0898264311428168 [CrossRef]
  • Barkin, R.L., Beckerman, M., Blum, S.L., Clark, F.M., Koh, E. & Wu, D.S. (2010). Should nonsteroidal anti-inflammatory drugs (NSAIDs) be prescribed to the older adult?Drugs & Aging, 27, 775–789. doi:10.2165/11539430-000000000-00000 [CrossRef]
  • Basu, R., Dodge, H., Stoehr, G.P. & Ganguli, M. (2003). Sedative-hypnotic use of diphenhydramine in a rural, older adult, community-based cohort: Effects on cognition. American Journal of Geriatric Psychiatry, 11, 205–213. doi:10.1097/00019442-200303000-00011 [CrossRef]
  • Beers, M.H. (1997). Explicit criteria for determining potentially inappropriate medication use by the elderly: An update. Archives of Internal Medicine, 157, 1531–1536. doi:10.1001/archinte.1997.00440350031003 [CrossRef]
  • Berger, A.M., Wielgus, K.K., Young-McCaughan, S., Fischer, P., Farr, L. & Lee, K.A. (2008). Methodological challenges when using actigraphy in research. Journal of Pain and Symptom Management, 36, 191–199. doi:10.1016/j.jpainsymman.2007.10.008 [CrossRef]
  • Blalock, S.J., Byrd, J.E., Hansen, R.A., Yamanis, T.J., McMullin, K., DeVellis, B.M. & Jordan, J.M. (2005). Factors associated with potentially inappropriate drug utilization in a sample of rural community-dwelling older adults. American Journal of Geriatric Pharmacotherapy, 3, 168–179. doi:10.1016/S1543-5946(05)80023-6 [CrossRef]
  • Brandt, N.J. (2016). Optimizing medication use through deprescribing: Tactics for this approach. Journal of Gerontological Nursing, 42(1), 10–14. doi:10.3928/00989134-20151218-08 [CrossRef]
  • Buysse, D.J., Reynolds, C.F. III. , Monk, T.H., Berman, S.R. & Kupfer, D.J. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28, 193–213. doi:10.1016/0165-1781(89)90047-4 [CrossRef]
  • Cella, D., Riley, W., Stone, A., Rothrock, N., Reeve, B., Yount, S. & Hays, R. (2010). The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. Journal of Clinical Epidemiology, 63, 1179–1194. doi:10.1016/j.jclinepi.2010.04.011 [CrossRef]
  • Charlson, M., Szatrowski, T.P., Peterson, J. & Gold, J. (1994). Validation of a combined comorbidity index. Journal of Clinical Epidemiology, 47, 1245–1251. doi:10.1016/0895-4356(94)90129-5 [CrossRef]
  • Chui, M.A., Stone, J.A., Martin, B.A., Croes, K.D. & Thorpe, J.M. (2014). Safeguarding older adults from inappropriate over-the-counter medications: The role of community pharmacists. The Gerontologist, 54, 989–1000. doi:10.1093/geront/gnt130 [CrossRef]
  • Cox, C.L. (1982). An interaction model of client health behavior: Theoretical prescription for nursing. Advances in Nursing Science, 5, 41–56. doi:10.1097/00012272-198210000-00007 [CrossRef]
  • Davidoff, A.J., Miller, G.E., Sarpong, E.M., Yang, E., Brandt, N. & Fick, D.M. (2015). Prevalence of potentially inappropriate medication use in older adults using the 2012 Beers criteria. Journal of the American Geriatrics Society, 63, 486–500. doi:10.1111/jgs.13320 [CrossRef]
  • Fick, D.M., Cooper, J.W., Wade, W.E., Waller, J.L., Maclean, J.R. & Beers, M.H. (2003). Updating the Beers criteria for potentially inappropriate medication use in older adults: Results of a US consensus panel of experts. Archives of Internal Medicine, 163, 2716–2724. doi:10.1001/archinte.163.22.2716 [CrossRef]
  • Findley, L.R. & Bulloch, M.N. (2015). Relationship between nonsteroidal anti-inflammatory drugs and fall risk in older adults. Consultant Pharmacist, 30, 346–351. doi:10.4140/TCP.n.2015.346 [CrossRef]
  • Fox, C., Richardson, K., Maidment, I.D., Savva, G.M., Matthews, F.E., Smithard, D. & Brayne, C. (2011). Anticholinergic medication use and cognitive impairment in the older population: The medical research council cognitive function and ageing study. Journal of the American Geriatrics Society, 59, 1477–1483. doi:10.1111/j.1532-5415.2011.03491.x [CrossRef]
  • Gnjidic, D., Blyth, F.M., Le Couteur, D.G., Cumming, R.G., McLachlan, A.J., Handelsman, D.J. & Naganathan, V. (2014). Nonsteroidal anti-inflammatory drugs (NSAIDs) in older people: Prescribing patterns according to pain prevalence and adherence to clinical guidelines. Pain, 155, 1814–1820. doi:10.1016/j.pain.2014.06.009 [CrossRef]
  • Guaraldo, L., Cano, F.G., Damasceno, G.S. & Rozenfeld, S. (2011). Inappropriate medication use among the elderly: A systematic review of administrative databases. BMC Geriatrics, 11, 1–10. doi:10.1186/1471-2318-11-79 [CrossRef]
  • Institute of Medicine. (2007). Preventing medication errors. Retrieved from www.nationalacademies.org/hmd/~/media/Files/Report%20Files/2006/Preventing-Medication-Errors-Quality-Chasm-Series/medicationerrorsnew.pdf
  • Kachru, N., Carnahan, R.M., Johnson, M.L. & Aparasu, R.R. (2015). Potentially inappropriate anticholinergic medication use in community-dwelling older adults: A national cross-sectional study. Drugs & Aging, 32, 379–389. doi:10.1007/s40266-015-0257-x [CrossRef]
  • Lee, E., Cho, H., Olmstead, R., Levin, M., Oxman, M. & Irwin, M. (2013). Persistent sleep disturbance: A risk factor for persistent or recurrent depression in community-dwelling older adults. Sleep, 36, 1685–1691. doi:10.5665/sleep.3128 [CrossRef]
  • Lemeshow, S. & Hosmer, D.W. Jr.. (1982). A review of goodness of fit statistics for use in the development of logistic regression models. American Journal of Epidemiology, 115, 92–106. doi:10.1093/oxfordjournals.aje.a113284 [CrossRef]
  • Maher, R.L., Hanlon, J. & Hajjar, E.R. (2014). Clinical consequences of polypharmacy in elderly. Expert Opinion on Drug Safety, 13, 57–65. doi:10.1517/14740338.2013.827660 [CrossRef]
  • Meit, M., Knudson, A., Gilbert, T., Yu, A.T.-C., Tanenbaum, E., Ormson, E. & Popat, S. (2014). The 2014 update of the rural-urban chart book. Retrieved from https://ruralhealth.und.edu/projects/health-reform-policy-research-center/pdf/2014-rural-urban-chartbook-update.pdf
  • Nasreddine, Z.S., Phillips, N.A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I. & Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53, 695–699. doi:10.1111/j.1532-5415.2005.53221.x [CrossRef]
  • National Conference of State Legislatures. (n.d.). Meeting the primary care needs of rural America: Examining the role of non-physician providers. Retrieved from http://www.ncsl.org/research/health/meeting-the-primary-care-needs-of-rural-america.aspx
  • Sarzynski, E.M., Luz, C.C., Rios-Bedoya, C.F. & Zhou, S. (2014). Considerations for using the ‘brown bag’ strategy to reconcile medications during routine outpatient office visits. Quality in Primary Care, 22, 177–187.
  • Schutte-Rodin, S., Broch, L., Buysse, D., Dorsey, C. & Sateia, M. (2008). Clinical guideline for the evaluation and management of chronic insomnia in adults. Journal of Clinical Sleep Medicine, 4, 487–504.
  • Shade, M.Y., Berger, A.M. & Chaperon, C. (2014). Potentially inappropriate medications in community-dwelling older adults. Research in Gerontological Nursing, 7, 178–192. doi:10.3928/19404921-20140210-01 [CrossRef]
  • Shade, M.Y., Berger, A.M., Dizona, P.J., Pozehl, B.J. & Pullen, C.H. (2016). Sleep and health-related factors in overweight and obese rural women in a randomized controlled trial. Journal of Behavioral Medicine, 39, 386–397. doi:10.1007/s10865-015-9701-y [CrossRef]
  • Taipale, H.T., Bell, J.S., Gnjidic, D., Sulkava, R. & Hartikainen, S. (2012). Sedative load among community-dwelling people aged 75 years or older: Association with balance and mobility. Journal of Clinical Psychopharmacology, 32, 218–224. doi:10.1097/JCP.0b013e3182485802 [CrossRef]
  • Thomas, D., Macdowell, M. & Glasser, M. (2012). Rural mental health workforce needs assessment—A national survey. Rural and Remote Health, 12, 2176–2187.
  • U.S. Department of Health and Human Services. (2015). Division of Medicaid & Long-Term Care State Unit on Aging 2012–2015 plan for aging services. Retrieved from http://dhhs.ne.gov/medicaid/Documents/AgingServicesStatePlanFY2012-FY2015.pdf
  • Vaughn McCall, W. (2004). Sleep in the elderly: Burden, diagnosis, and treatment. Primary Care Companion to the Journal of Clinical Psychiatry, 6, 9–20. doi:10.4088/PCC.v06n0104 [CrossRef]
  • Ware, J. Jr.. , Kosinski, M. & Keller, S.D. (1996). A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34, 220–233. doi:10.1097/00005650-199603000-00003 [CrossRef]
  • Zhan, C., Sangl, J., Bierman, A.S., Miller, M.R., Friedman, B., Wickizer, S.W. & Meyer, G.S. (2001). Potentially inappropriate medication use in the community-dwelling elderly: Findings from the 1996 Medical Expenditure Panel Survey. Journal of the American Medical Association, 286, 2823–2829. doi:10.1001/jama.286.22.2823 [CrossRef]

Participant Characteristics (N = 139)

Variablen(%)
Gender
  Male78 (56)
  Female61 (44)
Home location
  Small rural town83 (60)
  Rural area56 (40)
Age (years)
  Young-old (65 to 78)70 (50)
  Old-old (79 to 99)69 (50)
Race/ethnicity
  Caucasian137 (98)
  Hispanic1 (1)
  Native American/Alaskan American1 (1)
Marital status
  Married/cohabitating101 (74)
  Single/widowed36 (26)
Incomea
  Below poverty level (≤$10,000)4 (3)
  Above poverty level123 (97)
Education
  Grade 12 and under95 (69)
  College and above or trade school43 (31)
Mean (SD)
Body mass index (kg/m2)30.5 (6.3)
Age (years)77.7 (7.3)

Sample Characteristics of Health Experience and Health Status Variables

VariableMean (SD)
Health experience
  No. of medications11.6 (4.8)
  Charlson Comorbidity Index4.6 (1.5)
  No. of providers2 (0.8)
Health status (N = 139)a
  Physical component score (SF-12)38.7 (9.9)
  Mental component score (SF-12)53.2 (8.6)
Subjective sleep (N = 139)a
  Sleep disturbance (PROMIS SF-8A)46.9 (7.8)
  Sleep quality (PSQI total)6.5 (3.5)
Actigraphy (N = 133)b
  TST hours7.2 (0.9)
  WASO minutes50.6 (27.8)
  Wake percentage12.4 (8.6)

Characteristics of PIMs to Avoid

VariableTotal NParticipants Using Medications (n, %)
One PIM13844 (32)
Two or more PIM13823 (17)
Total PIM use67 (49)
Prescribed PIM6734 (51)
OTC PIM6716 (24)
Taking prescribed and OTC PIM6717 (25)
Classification of Most Common PIM to Avoida
Anticholinergic agents
  Antihistamines (84% of anticholinergics)6716 (24)
    Prescribed (11% of antihistamines)672 (3)
    OTC (94% of antihistamines)6715 (22)
  Skeletal muscle relaxant agents672 (3)
  Tricyclic antidepressant agents673 (4)
Alpha 1 blockers6712 (18)
Short-acting benzodiazepines6712 (18)
Non-benzodiazepine receptor agonists674 (6)
Estrogen677 (10)
NSAIDs6722 (33)
  Prescribed (32% of NSAIDs)677 (11)
  OTC (68% of NSAIDs)6715 (22)
History of falls with benzodiazepine drugs676 (9)

Correlations Between Demographic, Health Experience, and Health Status Variables and PIMs Use (N = 138)

VariablePIMs to Avoid (r)
Age0.004
Charlson Comorbidity Index0.081
No. of medications0.331**
No. of providers0.223**
Physical component score (S)−0.193*
Mental component score (S)−0.031
Pittsburgh Sleep Quality Index (S)0.158a
PROMIS (S)0.073
Total sleep time (A)−0.102
Wake after sleep onset (A)−0.027
Wake percentage (A)−0.059

Summary of Binary Logistic Regression Analysis Predicting Probability of PIMs Use

Predictor VariableUnstandardized βStandard ErrorChi-SquareExponential β Odds Ratiop Value
No. of medications0.6190.2247.621.858*0.006
No. of providers0.220.2061.151.2470.284
Physical component score−0.1540.2020.580.8570.446
Pittsburgh Sleep Quality Index0.180.1870.921.190.336
Authors

Dr. Shade is Assistant Professor, Dr. Berger is Associate Dean for Research and Professor and Dorothy Hodges Olson Endowed Chair in Nursing, Dr. Chaperon is Associate Professor, Dr. Yates is Professor, College of Nursing, Dr. Haynatzki is Professor, College of Public Health, and Dr. Sobeski is Assistant Professor, College of Pharmacy, University of Nebraska Medical Center, Omaha, Nebraska.

Dr. Sobeski is a paid consultant for Nebraska Medicaid Pharmacy and Therapeutics Committee (Nebraska Department of Health and Human Services), and paid faculty for the Dorothy B. Davis & Richard Brook Foundation Grant. The remaining authors have disclosed no potential conflicts of interest, financial or otherwise. The study was funded by the Jonas Center for Nursing Excellence.

Address correspondence to Marcia Y. Shade, PhD, MSN, RN, Assistant Professor, College of Nursing, University of Nebraska Medical Center, 985330 Nebraska Medical Center, Omaha, NE 68198-5330; e-mail: marcia.shade@unmc.edu.

Received: September 22, 2016
Accepted: February 10, 2017
Posted Online: April 11, 2017

10.3928/00989134-20170406-01

Sign up to receive

Journal E-contents