Skip to main content

Outpatient opioid prescribing by Alzheimer’s diagnosis among older adults with pain in United States



To examine opioid prescribing practices for pain in older adults with and without Alzheimer’s Disease and Related Dementias (ADRD).


This cross-sectional study used National Ambulatory Medical Care Survey data (2014–2016, and 2018). Adults aged ≥ 50 years with pain were analyzed. Prescribing of opioid and concomitant sedative prescriptions (including benzodiazepines, Z-drugs, and barbiturates) were identified by the Multum lexicon code. Multivariable logistic regression evaluated the risk of opioid prescribing or co-prescribing of opioid and sedative associated with ADRD in older adults with pain.


There were 13,299 office visits in older adults with pain, representing 451.75 million visits. Opioid prescribing occurred in 27.19%; 30% involved co-prescribing of opioids and sedatives. ADRD was not associated with opioid prescribing or co-prescribing of opioid and sedative therapy.


Opioid and sedatives are commonly prescribed in older adults with pain. Longitudinal studies need to understand the etiology and chronicity of opioid use in older patients, specifically with ADRD.


This national study examined the opioid prescribing practices for pain in older adults with and without Alzheimer’s Disease and Related Dementias (ADRD).

The study found about 30% of visits for older adults with pain resulted in opioid prescriptions, and the co-prescribing of sedatives were noted in 30% of these visits made by older adults with pain where an opioid were prescribed.

No significant difference exists between ADRD vs. non-ADRD groups in receiving opioid prescriptions or co-prescription of opioids and sedatives among the visits with pain.

Future research should aim to understand the health outcomes associated with opioid prescribing and concomitant receipt of sedatives in ADRD.

Peer Review reports


Opioids are commonly, although somewhat inappropriately, used to manage pain in older adults. Approximately 20% of US population filled at least one opioid prescription in 2018, with the highest opioid prescribing rate in older adults [1]. Opioid use in older adults may increase risk of delirium, falls, and fracture. The American Geriatrics Society (AGS) Beers Criteria strongly recommends against the use of opioids in older adults. Moreover, benzodiazepines and related hypnotics (Z drugs), often referred to as BZDRs, should also be used cautiously in older people, especially those with underlying Alzheimer’s Disease and Related Dementias (ADRD) as they may exacerbate cognitive impairment and increase risk of serious adverse events, including risk of fractures/falls, and hospitalizations [2].

The AGS Beers Criteria suggests to avoid the use of BZDRs in older patients and strongly recommends against the concomitant use of Benzodiazepines and opioids [3]. Despite this consensus recommendation and the risk, opioids and BZDRs are prescribed concomitantly in as many as nearly 1 in 3 older people with ADRD [4]. The objective of this study, was therefore, to examine patterns of opioid prescribing among older adults with pain and to compare the patterns of opioid prescribing among those with ADRD vs. those without. The study also evaluated the concomitant use of sedative prescriptions and opioids among those older adults with and without ADRD.


Data source and study design

The retrospective cross-sectional study was conducted using National Ambulatory Medical Care Survey (NAMCS) data from 2014-2016, and 2018, a publicly available dataset containing records of visits to outpatient facilities in the United States. NAMCS is a national, annual survey, administered by the CDC, that samples outpatient visits to non-federal, office-based providers at community health centers across the US [5]. NAMCS employs a multi-stage probability sampling design to identify a sample of providers capable of representing all providers in the US. At each sampled visit of healthcare professionals, standardized survey collected data about the office visits related to patient care, such as the reason for encounter visits, medical diagnoses coded in ICD-9 or ICD-10 CM codes, medications ordered, patients’ demographics, and providers’ characteristics and practice. Each study record captures up to thirty medications prescribed or used at baseline at each visit. These are classified using the Multum Lexicon system to enable researchers to characterize medication use. After data is collected, complex sampling weights based on patterns of healthcare utilization in the US are used to upsample visits with characteristics that are under-represented and to downsample those with characteristics that are over-represented in the unadjusted dataset. The study was approved under the exempt category by the University of Houston Institutional Review Board.

Study sample and outcome measures

This study identified all office visits involving adults at least 50 years of age with a condition likely to cause pain. These were defined using a list of ICD-9 and ICD-10 codes for painful conditions including (a) Abdominal pain, (b) Back, head & neck pain, (c) Fibromyalgia, neuropathy & systemic disorders related pain, (d) Fractures, contusions, sprains and strains, and (e) Limb/extremity pain, joint pain and non-systemic, non-inflammatory arthritic disorders. Visits made by patients with these conditions were identified based on a series of ICD-9-CM and ICD-10-CM codes. The primary measure of exposure was ADRD status. ADRD diagnosis was operationally defined using a binary indicator variable for clinicians who reported Alzheimer’s disease from the visit questionnaire. Additional ADRD visits were captured through ICD-9-CM code (290, 291.2, 294, 331, 797) and ICD-10-CM codes (G30.0, G30.1, G30.8, G309, F01.50, F01.51, F02.80, F02.81, F03.90, F03.91, F10.27, F10.97, F13.97, F18.27, F18.97, F19.27, F19.97) [6, 7].

The outcome of interest was receipt of an opioid prescription. All opioid prescriptions were operationally classified as narcotic analgesics and narcotic analgesic combinations, and identified using Multum database Lexicon Plus in terms of therapeutic classes drug category codes, in which the narcotic analgesics and narcotic analgesic combinations coded as ‘060’ and ‘191’, respectively [8, 9]. Use of benzodiazepines, Z-drugs and barbiturates, jointly classified as use of sedative medications, was also identified for each visit using the Multum therapeutic classification system. Visits were classified as opioid alone or concomitant opioid and BZDR visits.

Statistical analysis

Descriptive analyses, adjusted for the complex survey design, were conducted to characterize outpatient visits by older people with pain. The weighted analyses were also performed to estimate the overall opioid use as well as concomitant use of sedative prescriptions (benzodiazepines, Z-drugs and barbiturates) and opioids, considering the complex survey design and adjusting for the variables of the cluster, strata and weight. Two weighted, multivariable logistic regression models, adjusted for pain diagnosis, comorbidities, and patient demographics were used to measure the association between ADRD diagnosis and receipt of opioid monotherapy and opioid and BZDR combination therapy, respectively. All analyses were conducted in SAS 9.4 (SAS Institute, Cary, North Carolina).

The Andersen Behavioral Model, a commonly used conceptual framework in health services research, was employed to guide the selection of covariates and to conceptually classify them as predisposing, enabling, and need characteristics that are known to contributed to health services utilization. [10] Variables categorized into these domains were selected based on the published literature and available data in the NAMCS.[11,12,13] Predisposing characteristics included age, sex, race, ethnicity, metropolitan statistical areas. Enabling characteristics included US Census region of residence, payer type, physician specialty and insurance typ. Need characteristics included year of visit, Elixhauser comorbidity score, and substance use disorders &psychiatric conditions, reasons for visit [14].


There were 13,299 unweighted visits for older adults with a painful condition, nationally representing a total number of 451.75 million (SE: 23.31 million) outpatient visits in the US during the study period for an annual mean (SD) of 112.94 (5.83) million visits. Of the national sample of older adults with pain, 1.42% of visits were also related to ADRD, nationally representing 1.60 million (SE: 0.43 million) visits each year.

Table 1 shows the characteristics of outpatient visits among these older adults with pain, by AD status. Some baseline characteristics were comparable between ADRD and non-ADRD groups, including demographics of sex, race, ethnicity, region, and clinical characteristics of reasons for visits and certain type of painful conditions (such as limb/extremity pain, back pain, fibromyalgia). However, significant differences were noted between those with ADRD versus those without ADRD in age group, metropolitan location, payment source, provider specialty, and survey year. Also, compared to non-ADRD visits, those with ADRD were more likely to have fractures and less likely to have abdominal pain. In addition, those with ADRD were more likely to have substance use disorders than their non-ADRD counterparts.

Table 1 Study Characteristics. Outpatient Visits Among older adults with pain, by ADRD status From the US NAMCS

Patterns of opioid prescribing and co-prescribing of opioids and sedatives

Table 2 provides prescribing practice related to opioid prescription or co-prescribing of opioids and sedatives during office visits by older adults with pain. Overall, of all these sampled outpatient visits, 27.19% resulted in an opioid prescription, with an estimated 30.70 million (SE: 2.36 million) opioid prescription annually. Sedative prescriptions were co-prescribed in almost 30% of these visits made by older adults with pain receiving an opioid prescription.

Table 2 Multivariable logistic regression of adjusted association results. Opioid Prescribing & Concomitant Opioid and Sedatives Use, by AD Status Among Elderly Patients Involving a Painful Condition in NAMCS, 2014–2016, 2018

Association of ADRD status with opioid prescription

Table 2 shows the effect of interest from two, adjusted multiple logistic models evaluating the f association between ADRD status and receipt of opioids and concomitant opioid and BZRD medications, respectively. The first model examining the association between ADRD status and receipt of opioid monotherapy found no association between ADRD status and medication receipt. (aOR: 1.356 (0.798–2.302)) (Table 2). Similarly, ADRD status was not associated with receipt of combination opioid and BZRD treatment(aOR: 1.836 (0.831–4.056)).


We found that adults with ADRD are neither more nor less likely than those without ADRD to be prescribed opioids or opioid and sedative combination therapy at US outpatient visits. In this multi-year, nationally representative sample, more than one-quarter of all outpatient office visits for older adults resulted in an opioid prescription and over 30% of these visits ended in sedative co-prescription. The fact that opioid prescribing was not clearly associated with ADRD status suggests that ADRD is not an independent predictor of opioid prescribing. These findings contrast with an earlier study in Medicare data saying that patients with ADRD were less likely than those without to receive an opioid prescription [11]. Yet another US study using Medicare claims and another European study, however, found that ADRD patients were more likely to receive opioid prescriptions [12, 13]. The lack of consensus in this area suggests that more, rigorous longitudinal work is still needed to understand predictors and outcomes of the common use of this potentially dangerous combination of medication in older adults with ADRD. We limited our analysis to those with a comorbid painful condition-those with an increased baseline likelihood of receiving an opioid prescription. Our results demonstrate that providers often look past potential cognitive decline and increased fall risk and still freely prescribe opioids to individuals with ADRD.

The lack of association between ADRD status and opioid prescribing is highly concerning and raises questions about the quality of pain management in individuals with ADRD. In a recent matched cohort study, Taipale and colleagues found that incident opioid use doubles the risk (aHR: 1.96) of fall and fracture in older adults with Alzheimer’s disease [15]. Risk of fall was highest within the first two-months of opioid use (aHR: 2.27) and varied with opioid potency [15]. Therefore, more work is needed to identify differences in trajectory of opioid pharmacotherapy and the incidence of adverse, clinically relevant outcomes in older adults with and without ADRD. There is also a need for continuous evaluation of the extent of co-prescribing of these high-risk medications in the vulnerable ADRD patient group. Future research needs to focus on understanding the sequelae of opioid and sedative combination use in older adults with ADRD. Overall, there is a strong need for concerted efforts to optimize opioid prescribing in ADRD group.

Overall, this study found no variation in the use of opioid medications between the group with or without ADRD. Tailored, pain management guidelines for patients with ADRD are lacking and our findings suggest that providers manage pain in patients with memory loss not differently in prescribing opioids than they do for other older adults. The lack of differences in opioid prescribing speaks fundamentally to the well document challenges providers face in assessing pain among older adults with ADRD [16]. Although many patient reported pain measurement tools exist, these are poorly validated in older adults with ADRD [17]. In addition, there are organizational shortcomings in the care setting often hamper the quality of care, including pain management in the ADRD group [16]. For example, in a survey of nurses and other health workers in European long-term care settings, a lack of education in pain management among the nurses and healthcare workers was reported [18]. In a recent semi-structured interview of nurses in 12 nursing homes in Sweden, nurses described the communicative and organizational challenges when they provided care for the advanced dementia persons with pain, highlighting a demanding need for training and educating skilled nursing facilities to improve the quality of care among people with dementia and pain [19]. Therefore, to provide effective care in pain management in the ADRD population, concerted efforts and practical insights from a multidisciplinary team are needed. Preventing adverse sequelae of opioid pharmacotherapy in older adults with ADRD requires that caregivers at every point in the process of care, including family caregivers, must be trained to identify and appropriately manage pain alongside other conditions, especially in the setting of BZRD pharmacotherapy.

Strengths and limitations

The use of a recent, representative, national database allowed us to characterize current use of opioid and BZRD agents in office based visits for older adults with and without ADRD. In spite of these strengths, this current analyses had several limitations. As NAMCS surveys involve physician practice in nonfederal office-based patient care, current findings are not generalizable to institutionalized patients. Furthermore, visits to hospital-based clinics may be under-represented in NAMCS.

In addition, the NAMCS is a series of cross-sectional surveys which is representative of visit-level data, not patients. This limits the ability to make population based inferences and makes it impossible to establish causality. Furthermore, NAMCS does not collect detailed information on pain severity making it impossible to determine if opioid use was associated with more severe pain. Likewise, there was lack of information of the duration of utilization of opioid pain medications. NAMCS does not contain measures of socioeconomic status, prescriber preference, or prescriber training outside of specialty certification meaning that it is difficult to fully operationalization all constructs of the Andersen Behavioral Model. Lastly, residual confounding may have resulted in statistical imbalance between ADRD vs. non-ADRD group. The cross-sectional nature of the data source makes it difficult to examine this through mediation and moderation modeling. This may have increased the risk of type 1 error, leading us to conclude that there was no association between ADRD status and opioid prescribing. Simply put, the unbalance distribution between the ADRD vs. non-ADRD group may explain the insignificance observed in the results: since there are not many people in the ADRD group, the number of patients with the opioid outcome is also less in the ADRD group, causing a large standard deviation and hence leading to a wide confidence interval of the adjusted odds of opioids associated with ADRD group. Future longitudinal studies using a large database are needed to validate the results of this analysis. In light of these limitations, these results still contribute to our collective understanding of the types of healthcare visits that lead to opioid prescribing in ADRD and non-ADRD patients.

Data availability

The datasets generated and/or analysed during the current study are available in the [Centers for Disease Control and Prevention], [National Center for Health Statistics]. For more information, check:


  1. Moriya AS, Miller GE. STATISTICAL BRIEF #515 Any Use and Frequent Use of Opioids among Elderly Adults in 2015–2016, by Socioeconomic Characteristics.; 2015.

  2. Brandt J, Leong C. Benzodiazepines and Z-Drugs: An Updated Review of Major Adverse Outcomes Reported on in Epidemiologic Research. Drugs in R and D. 2017;17(4):493–507.

    Article  PubMed  Google Scholar 

  3. Samuel MJ. American Geriatrics Society 2015 updated beers criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2015;63(11):2227–2246.

  4. Karttunen N, Taipale H, Hamina A, et al. Concomitant use of benzodiazepines and opioids in community-dwelling older people with or without Alzheimer’s disease—A nationwide register- based study in Finland. Int J Geriatr Psychiatry. 2019;34(2):280–288.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Center for Health Statistics N. National Ambulatory Medical Care Survey: 2018 National Summary Tables.; 2018.

  6. Vickers LE, Martinez AI, Wallem AM, Johnson C, Moga DC. Potentially inappropriate medication use in older adults with Alzheimer’s Disease and related Dementias living in the community: a cross-sectional analysis. Drugs Real World Outcomes. 2021;8(4):519–26.

    Article  PubMed  Google Scholar 

  7. Davis MA, Chang CH, Simonton S, Bynum JPW. Trends in US Medicare Decedents’ Diagnosis of Dementia From 2004 to 2017. JAMA Health Forum. 2022;3(4):e220346.

  8. Soprano SE, Soprano SE, Hennessy S, et al. Assessment of Physician Prescribing of Muscle Relaxants in the United States, 2005–2016. JAMA Netw Open. 2020;3(6).

    Article  PubMed  PubMed Central  Google Scholar 

  9. Neprash HT, Barnett ML. Association of Primary Care Clinic Appointment Time With Opioid Prescribing. JAMA Netw Open. 2019;2(8):e1910373.

    Article  Google Scholar 

  10. Babitsch, B., Gohl, D., & von Lengerke, T. (2012). Re-revisiting Andersen’s Behavioral Model of Health Services Use: A systematic review of studies from 1998–2011. GMS Psycho-Social-Medicine, 9, Article Doc11.

    Article  PubMed  Google Scholar 

  11. Shen C, Zhao X, Dwibedi N, Wiener RC, Findley PA, Sambamoorthi U. Opioid use and the presence of Alzheimer’s disease and related dementias among elderly Medicare beneficiaries diagnosed with chronic pain conditions. Alzheimer’s and Dementia: Translational Research and Clinical Interventions. 2018;4:661–8.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Hamina A, Taipale H, Tanskanen A, Tolppanen AM, Tiihonen J, Hartikainen S. Differences in analgesic use in community-dwelling persons with and without Alzheimer’s disease. Eur J Pain (United Kingdom). 2017;21(4):658–67.

    Article  CAS  Google Scholar 

  13. Wei YJJ, Schmidt S, Chen C, et al. Quality of opioid prescribing in older adults with or without Alzheimer disease and related dementia. Alzheimers Res Ther. 2021;13(1).

  14. van Walraven C, Austin PC, Jennings A, Quan H, Forster AJ. A Modification of the Elixhauser Comorbidity Measures into a Point System for Hospital Death Using Administrative Data. Vol 47.; 2009.

  15. Taipale H, Hamina A, Karttunen N, Koponen M, Tanskanen A, Tiihonen J, Hartikainen S, Tolppanen AM. Incident opioid use and risk of hip fracture among persons with Alzheimer disease: a nationwide matched cohort study. Pain. 2019 Feb;160(2):417–423. PMID: 30325873.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Achterberg WP, Pieper MJ, van Dalen-Kok AH, de Waal MW, Husebo BS, Lautenbacher S, Kunz M, Scherder EJ, Corbett A. Pain management in patients with dementia. Clin Interv Aging. 2013;8:1471–82. Epub 2013 Nov 1. PMID: 24204133; PMCID: PMC3817007.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Husebo BS, Achterberg W, Flo E. Identifying and Managing Pain in People with Alzheimer’s Disease and Other Types of Dementia: A Systematic Review. CNS Drugs. 2016;30(6):481–497.

  18. JENNINGS A, LINEHAN M, FOLEY T. The knowledge and attitudes of general practitioners to the assessment and management of pain in people with dementia. BMC Fam Pract. 2018;19(1):166.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Lundin E, Godskesen TE. End-of-life care for people with advanced dementia and pain: a qualitative study in Swedish nursing homes. BMC Nurs. 2021 Mar 20;20(1):48. PMID: 33743691; PMCID: PMC7981921.

Download references

Acknowledgements and disclosure



This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author information

Authors and Affiliations



YH takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: RRA, TJV and YH. Acquisition, analysis, or interpretation of data: YH and TJV. Drafting of the manuscript: YH and TJV. Critical revision of the manuscript for important intellectual content: RRA, TJV and YH. Statistical analysis: YH. Administrative, technical, or material support: TJV. Supervision: TJV.

Corresponding author

Correspondence to Yinan Huang.

Ethics declarations

Ethics approval and consent to participate

This study received IRB approval from University of Houston.

Consent for publication

Not Applicable (NA).

Competing interests

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Dr. Aparasu reports grants from Astellas, Incyte, and Novartis, outside the submitted work. Dr. Varisco receives research funding from the Foundation for Opioid Response Efforts and provides research consulting services to HEALIX Infusion Therapy. The remaining authors report no conflicts of interest.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Huang, Y., Aparasu, R.R. & Varisco, T.J. Outpatient opioid prescribing by Alzheimer’s diagnosis among older adults with pain in United States. BMC Geriatr 23, 465 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: