Barriers to Care for Nursing Home Residents With Substance Use Disorders: A Qualitative Study

Abstract

Objectives:

Over the past decade, the numbers of older adults with opioid and substance use disorders (OUD/SUD) have increased. As this population enters nursing homes (NHs) in increasing numbers, it is crucial to consider their capacity to manage issues related to OUD/SUD. This study aimed to examine current NH protocols for care coordination of residents with OUD/SUD as well as facility-related barriers to providing care to this vulnerable population within the NH.

Methods:

Twenty-four semistructured interviews were conducted with NH staff including directors of nursing, administrators, nurses, and physicians in July 2020. Staff were recruited from 11 different post–acute care and long-term care facilities located in urban and suburban communities of Chicago. Interviews were conducted virtually (via teleconference platform or by telephone) and subsequently coded using ATLAS. ti 8 (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany) using constant comparative method.

Results:

Qualitative analyses identified 3 themes around NH barriers to care for residents with SUD/OUD: (1) staff preparedness, (2) staff perceptions of addiction, and (3) overall lack of resources. Results revealed a strong need for the development of consistent policies, as well as standardized, educational interventions for NH staff that target SUD/OUD management in this vulnerable population.

Conclusions:

The evaluation and impact of persons with SUD/OUD entering NHs are an important topic that requires further study. More resources and staff training are necessary to ensure that residents with SUD/OUD have access to appropriate care within these settings.

Substance use disorders (SUDs) and opioid use disorders (OUDs) in older adults represent one of the fastest-growing public health concerns in the United States.^1,2 There was a 53.5% increase in older adults seeking treatment for OUD from 2013 to 2015,³ and nearly 125,000 US hospitalizations of older adults were opioid-related in 2018.⁴ In 2020, the Centers for Disease Control and Prevention reported drug overdose deaths among older adults increasing more than 6-fold from 1999 to 2019.⁵

The proportion of older adults entering nursing homes (NHs) with OUD/SUD is increasing.⁶ A 2020 study of more than 7 million Medicare beneficiaries found that older adults with OUD-related hospitalizations were discharged to NHs at higher frequencies than non–OUD-related hospitalizations (26.4% vs 22%).⁷ As the aging population enters NHs in increasing numbers, it is crucial to consider facility capacity to manage residents with these disorders.⁸

Nursing home residents face significant barriers to diagnosis and treatment of OUD/SUD.⁹ Older adults are less likely to be screened for substance use in any setting.¹⁰ Symptoms or sequelae of substance use may be misdiagnosed as symptoms of aging.^11,12 Older adults on medications for OUD (MOUDs) may experience medication interruption during care transitions. In addition, residents who are weaned off opioids in the NH and resume use after discharge are at greater risk of overdose or death.¹³ Recent studies have also shown that older adults with SUD may be routinely denied admission to NHs.^14,15 Although declining admission due to an OUD diagnosis or treatment with MOUDs violates the Americans with Disabilities Act,¹⁶ there is little regulation of admission practices, and not much is known of the current admission protocols and standards of care in NHs for residents with OUD/SUD.

The purpose of this study is to understand (1) current NH protocols for the care of residents with OUD/SUD and (2) the knowledge, attitudes, and experiences of NH staff caring for this population to identify barriers to effective care.

METHODS

Setting and Participant Recruitment

Semistructured interviews were conducted with administrators and staff of post–acute care and long-term care nursing facilities within a single Chicago-land care-network affiliated with an urban academic medical center in Chicago, Ill, in July 2020. Facilities ranged in size from 100 to 250 beds with a mix of short-stay and long-term care residents. Volunteers were recruited from 11 facilities via email. Two initial recruitment emails were sent to staff en masse at all 11 locations. Thirty individualized follow-up emails were sent to those who expressed initial interest with 24 interviews confirmed. A second email sent to the 6 staff who did not reply to the first follow-up generated no further responses (response rate = 24/30). Of the 11 facilities solicited, interviews were completed with staff representing all 11 buildings. Participants were entered into a raffle for a $100 gift card.

Qualitative Survey Design

A semistructured interview was developed to explore (1) NH protocols, (2) staff perceptions of this population, and (3) potential gaps in staff training and resource availability through a series of open-ended questions. The interview also included 2 Likert-type questions; the first asked participants to rate their level of agreement with accepting patients with SUD/OUD into their facility (1 = strongly disagree to 5 = strongly agree), and the second asked participants to rate how well-trained they felt to respond to a substance-related overdose. Demographics including age, sex, ethnicity, highest level of education, role at the facility, and number of years in role were also collected.

The script was reviewed by the health literacy office at the affiliated academic medical center.

Data Collection

Interviews were held over a secure Health Insurance Portability and Accountability Act–compliant audio teleconference platform (Zoom Video Communications, Inc) or by telephone. Interviews ranged from 30 to 60 minutes and were conducted by a trained interviewer (M.Y.). Verbal informed consent was obtained from all participants. Interviews were recorded using Zoom and transcribed with Microsoft Word (version 16.48, 2021).

This project was reviewed and approved by the institutional review board of the affiliated academic medical center.

Data Analysis

Interview data were deidentified, analyzed, and coded independently by 2 members of the research team (K.J.B. and M.Y.). An inductive coding method was used to identify and organize the data into general themes. Four transcripts were initially transcribed to establish a coding scheme, after which investigators discussed discrepancies and finalized the coding scheme. Discrepancies were adjudicated by a third reviewer (S.L.). An additional 20 transcripts were coded to further test and refine the categories. A total of 85 codes were used to describe themes present in 755 quotes. Frequency of themes were analyzed by type of facility and whether formal programming was present for SUD/OUD residents. Coding and qualitative analyses were completed using ATLAS.ti 8.4.4 (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany).

Sociodemographic interviewee characteristics, facility payor distribution, and Likert-type responses were analyzed using a Mann-Whitney U test given a small sample size and a nonnormal distribution. Statistical analyses were conducted in Stata version 16 (StataCorp LLC, College Station, TX).

RESULTS

Twenty-four NH staff representing 11 facilities consented to participate (Table 1). At the time of interview, 4 of 24 facilities reported having formal programming for residents with an addiction history. An additional 5 facilities reported accepting residents with SUD. Most participants ranged in age from 31 to 45 years (16 of 24). Almost half of participants were Black (11 of 24); the majority had either a bachelor’s or graduate degree (20 of 24) and were female (19 of 24). Staff included administrators (9 of 24), director of nursing (8 of 24), nurses (2 of 24), physicians (4 of 24), and a certified alcohol and drug counselor (CADC) (1 of 24). Half of respondents (12 of 24) had been in their current role for 1 to 5 years. Only race was statistically significantly different between facilities with formal programming versus no formal programming.

TABLE 1.

Sociodemographic Information of Interview Participants

Variable, n (%)	Total (n = 24)	Formal Programming (n = 9)	No Formal Programming (n = 15)	P
Age range, y				0.21
31–45	16 (66.7)	5 (55.6)	11 (73)
45–54	6 (25)	3 (33.3)	3 (20)
≥55	1 (4.2)	1 (11.1)	0
Unknown	1 (4.2)	0	1 (6.7)
Female sex	19 (79.1)	6 (6.7)	13 (86.7)	0.51
Race				0.009
Black	11 (45.8)	7 (77.8)	4 (26.7)
White	12 (50)	1 (11.1)	11 (73.3)
Asian	1 (4.2)	1 (11.1)	0
Highest level of education				0.55
Associate degree	2 (8.33)	0	2 (13.3)
Bachelor’s degree	10 (41.7)	5 (55.6)	5 (33.3)
Some graduate	1 (4.2)	1 (11.1)	0
school
Graduate degree	10 (41.7)	3 (33.3)	7 (46.7)
Unknown	1 (4.2)	0	1 (6.7)
Role				0.21
Administrator	8 (33.3)	4 (44.4)	4 (26.7)
Director of nursing	8 (33.3)	3 (33.3)	5 (33.3)
Physician	3 (12.5)	0	3 (20.0)
Medical director	1 (4.2)	0	1 (6.7)
Regional director	1 (4.2)	1 (11.1)	0
Addiction	1 (4.2)	1 (11.1)	0
counselor
Licensed practice	1 (4.2)	0	1 (6.7)
nurse
Nurse practitioner	1 (4.2)	0	1 (6.7)
Time in role, y				0.59
<1	5 (20.8)	2 (22.2)	3 (20)
1–5	12 (50)	5 (55.6)	7 (46.6)
6–10	4 (16.7)	1 (11.1)	3 (20)
>10	1 (4.2)	0	1 (6.7)
Unknown	2 (8.33)	1 (11.1)	1 (6.7)

Facility payor-mix distribution was obtained for 8 of 11 facilities. The ranges of NH residents by the primary payor source were as follows: Medicaid, 53% to 97%; Medicare, 1% to 16%; managed care, 0% to 16%; private pay, 0% to 2%; and other payor source, 0% to 4%. Of facilities with formal programming, 95% or more of residents were funded through Medicaid. Payor distribution difference was not statistically significant between facilities with formal programming versus no formal programming (Table 2).

TABLE 2.

Facility Payor-mix Distribution (Percentage) by Facility Type

Primary Payor Source Median (IQR)	Total (n = 8)*	Formal Programming (n = 3)	No Formal Programming (n = 5)	P
Private pay	0.5 (2)	0 (0.5)	2 (2)	0.34
Medicare	3 (4.5)	3 (0.5)	6 (6)	0.65
Managed care	2 (4)	1 (1)	4 (6)	0.13
Medicaid	92.5 (7.25)	95 (0.5)	90 (8)	0.23
Other	0.5 (1)	1 (0.5)	0 (1)	0.87

^*Payor-mix distribution data were provided for only 8 of the 11 facilities.

IQR indicates interquartile range.

Likert-type Interview Responses

Half of the staff interviewed (12 of 24) agreed or strongly agreed to the statement: “I believe that we are well trained in how to respond if a patient overdoses in the nursing home.” Staff at facilities with formal addiction programming reported a statistically significantly higher comfort in responding to an overdose (median Likert = 5) compared with facilities without any programming (median Likert = 3) (P = 0.004) (Table 3). Staff from both types of facilities had neutral or slightly positive attitudes toward accepting patients with OUD (median Likert = 4 in facilities with formal programming vs median Likert = 3 with no programming), with no significant difference between the 2 groups (P = 0.153).

TABLE 3.

Likert-type^* Interview Statements in Staff From Nursing Homes With Formal Addiction Programming Versus No Formal Addiction Programming

Statement	Formal Programming (n = 9)	No Formal Programming (n = 15)	P
“I believe that we are well trained in how to respond if a patient overdoses”
Mean (SD)	4.4 (0.726)	2.9 (1.19)	0.004
Median	5	3
“Overall we should accept patients with a history of SUD/OUD”
Mean (SD)	3.9 (0.60)	3.4 (0.91)	0.15
Median	4	3

^*Likert-type scores (1 = strongly disagree to 5 = strongly agree).

OUD indicates opioid use disorder; SUD, substance use disorder.

Qualitative Interview Results

The qualitative analysis was structured into 3 parts: (1) themes related to the general admission protocol for patients with substance use histories, (2) policies on the management of MOUDs across different facilities, and (3) knowledge, perceptions, and attitudes among staff interviewed.

Admission Protocols: A Range of Criteria

All facilities (regardless of formal programming status) used a range of structured and unstructured criteria to evaluate patients for acceptance as a resident, which varied widely between facilities. In addition to standard criteria requiring the patient to have a skilled need requiring subacute physical, occupational, or speech therapies, facilities completed a history review investigating length of time since substance use, frequency of substance use, and type of substance used. Program adherence for patients receiving MOUDs was reviewed. Some facilities requested toxicology results from the hospital (Table 4, 1b). Inadequate reimbursement of residents with addiction diagnoses was often cited as a potential barrier to admission (Table 4, 1c).

TABLE 4.

Variability in Nursing Home Protocols for Residents With Substance Use Disorders

Theme	Subtheme	Illustrative Quotations
1. Admission criteria	a. Behavioral contracts 24 (19)*	“I’ll initiate a behavioral contract […] contingent upon being compliant with your methadone program. There will be no visitors for the first two weeks. And you have to submit to a weekly drug test for four weeks.” —DON 062901 “We have our hospital liaisons make it clear to the potential resident that they would not be able to go out on pass.” — Administrator 07202
	b. History and drug screen 51 (20)	“When we get the referral […] we ask. Does the patient have a drug history? Is it active? […] we do ask for a drug tox screen.” —DON 070603 “We don’t ask for drug toxicology […] we review the medical record, we have liaisons that help us get an idea of what the person’s behavior is like. […] If we all feel like they’re not going to be able to comply then we as a facility move on from it because we primarily do sub-acute nursing and long term care, we don’t have a specific drug treatment program.” —Administrator 071705
	c. Profit 26 (10)	“Our rate is based on how the patient eats, how the patient walks, so if you have patients that are opiate users that are kind of walking and talking, you’re not making as much money on those type of patients […] you don’t get payment for [group therapy and counseling].” — Administrator 072301
	d. Perceived fit with other NH residents 19 (13)	“If I have […] a 32-year-old schizophrenic with a history of heroin use and you got old ladies who have fallen down the stairs, do I want to mix the two populations?” —DON 062901
	e. Perceived risk to NH 10 (1)	“The other resident had a substance use history, had not been actively using and I guess […] his roommate enticed him.” —DON 070603 “You have to be really careful […] [If the] patient comes inside your facility and overdoses, then you’re responsible […]” —DON 062901
	f. General preferences 96 (41)	“We don’t have a formalized program, so we would typically evaluate [a patient’s] access to [illicit] substances, how mobile are they, would they be able to have a visitor bring it in […] or individually maneuver the wheelchair to an area where they would be able to get access?” —DON 072001
		“You may have five […] substance [referrals], but out of the five, you might have a person on methadone, or you might have that person who is trying to get clean. You have to get the lesser of the evils.” —DON 062901
2. MOUD policies	a. MOUD program policies 73 (34)	“[T]hey have to be on some type of program.” —DON 070801 “We have had residents who have a history of heroin [who] get here and then they say I want to start by going to the methadone clinic because I’m having cravings. We will set them up for that.” —Administrator 07202 “We have a good working relationship with the […] behavioral health clinic [next door, and] their [MOUD] program.” —Administrator 070701
	b. Administration of MOUDs in NH 33 (14)	“We had a guy who was actively withdrawing. There was a delay in the [buprenorphine] being delivered, and then another delay starting it. All those barriers[…] it was difficult to coordinate.” —NP 071401 “I can’t dose [methadone] but everything is labeled and we keep records of what we give and what time. Then we can administer it.” —DON 072202 “As a facility we are not licensed to administer methadone unless it’s for pain. If […] it’s for substance abuse we’re not allowed to.” —DON 070603
	c. Ability to self-administer MOUD 11 (5)	“If you’re not clinically capable of self-dosing […] that’s a barrier. We can’t administer it for them. We Face-Timed the resident and said: ‘Can you tell me how you self-administer your medication? Do you know how much [dose]? Can you hold a pen? I need to see your dexterity.’” —Administrator 070701
3. Services	a. Transport to methadone clinic 23 (14)	“There’s no supervision […] They go to the methadone clinic and then they come back and what happened between there and us, it’s up to them. Sometimes they won’t get back on the bus.” —NP 071401 “We have staff that will take the patient and go to the methadone clinic […] The patients cannot go by themselves.” —Regional Director 070702 “Getting on the shuttle, waiting in line, doing your urine drop, picking up your methadone, bringing it back, giving it to the nurse […] requires a level of functioning that we don’t [see] in the nursing home.” —Physician 072701
	b. Access to MOUDs in NH 11 (1)	“When the patient comes to the facility it’s always a question of who’s going to pay for [the medication]? Can we even obtain it from the pharmacy for suboxone? Is the physician at the facility [X-waivered]?” —Physician 071604 “If I’m getting an appointment [at an MOUD clinic] for a resident on Medicaid, [or] if they don’t take whatever the insurance is, you’ve got to wait 30 days. Substance abuse needs immediate attention.” —DON 062901 “[We] have a psychiatrist [who] prescribes suboxone.” —DON 070801
	c. Support services 52 (43)	“I stay in contact with the patients. I make sure they take their medications. They get lock boxes. For some of them I have their medications in my office. I take it to them myself.” —CADC 070601 “In a normal circumstance we would do two group meetings [everyday], five days a week.” —Administrator 070701 “We have a psychologist that comes in five times a week and a psychiatrist that comes in three times a week.” —DON 070801
	d. Overdose protocol and management 42 (19)	“The main [response] is 911 […] We don’t have a doctor inhouse [so] we would have to order to get the Narcan.” — DON 072202 “Narcan is at their bedside for [MOUD] program [residents…] When they get their [MOUD] from the clinic they also get two doses of Narcan [and we] have our own supply of Narcan that comes from our pharmacy.” — DON 070603
	e. Staff experience with OUD 58 (23)	“A lot of my newer nurses I guarantee you wouldn’t recognize [an overdose] and wouldn’t know what to do with that situation.” —DON 072102 “You know it’s that kind of thing where they say if you worked at the county hospital, you can work anywhere?” — DON 062901

^*N total thematic codes (n thematic codes from facilities with formal programming).

CADC indicates certified alcohol and drug counselor; DON, director of nursing; MOUDs, medications for opioid use disorder; NH, nursing home; NP, nurse practitioner; OUD, opioid use disorder.

Nine of the 24 staff interviewed were from facilities with formal programming for residents with a diagnosis of SUD in conjunction with a skilled need. Interviewees noted that these programs were instituted in response to an increase in the number of referrals with substance use history. Facilities with formal programming accepted referrals who (1) demonstrated a skilled need and (2) were willing to sign a behavioral contract stipulating cessation of any illicit drug use and implementation of random drug testing. Failure to adhere to the contract could result in discharge. Some facilities restricted residents with a substance use history from leaving the facility on pass for an initial period except for medical appointments. Patients who refused to sign the behavioral contract or the initial pass restrictions or who were actively using drugs without intention to stop were generally not accepted for admission (Table 4, 1a).

A common theme emerging from admission review centered around the perceived fit of this population and risk to the facility, staff, and other residents. Perceived fit differed, depending on facility location and type. Many administrators expressed reluctance to “mix” populations. Given the perception that individuals with substance use histories are generally younger, fit in terms of age and sociability was also assessed during admission decisions (Table 4, 1d). Several staff members also expressed concern about mingling persons with active substance use with those with a more remote history, thereby potentially triggering relapse (Table 4, 1d, e).

Perceived risk also varied on location and facility type. Some staff expressed concern for the potential for residents to continue to use or become aggressive and agitated, thus imparting an individual risk to safety of staff and other residents. Conversations about individual risk frequently turned to concerns of legal liability (Table 4, 1e). For some facilities, this risk aversion manifested in admission preferences for patients with a history of substance use who were immobile and more dependent for care needs (Table 4, 1f). For facilities serving a demographic where addiction histories were more common, admissions criteria seemed more flexible and based on which referral was preferable in the context of the other referrals that day (Table 4, 1f).

Policies on the Management of MOUDs Accepting Patients on Methadone

Most respondents commented that patients were rarely accepted without pre-established methadone clinics. There was an overall expectation that the referring hospital should connect the patient to a maintenance program before arrival (Table 4, 2a). However, this was not an all-encompassing policy; 2 staff members from 2 different facilities stated that they regularly set up residents with a treatment center within 24 hours of admission if they did not have one. In one case, the behavioral health treatment offices were housed in the same building as the NH, which facilitated a working relationship supporting residents with a history of substance use (Table 4, 2a).

Accepting Patients on Buprenorphine

Interviewees seldom distinguished between accepting patients on methadone versus buprenorphine-containing medications such as buprenorphine-naloxone. Several interviewees reflected on the difficulty of buprenorphine access for new admissions, particularly when there was no Drug Addiction Treatment Act of 2000 (DATA)–waivered NH physician (Table 4, 3b). Delays in accessing the medication could trigger withdrawal (Table 4, 2b). In most cases, patients were expected to self-administer their MOUDs to be accepted (Table 4, 2c).

Management and Regulation of MOUDs in the NH

Interviews revealed that MOUDs are heavily regulated within each facility, although specific policies were facility-dependent. Some nurses cited facility policy restricting administration of MOUDs; residents must be able to self-administer (Table 4, 2c). Many facilities required that MOUDs be stored in “double lock boxes” inside the resident’s room, where the nurse unlocks the outside box and the resident unlocks the inside box. However, other facilities allowed their nurses to administer MOUDs (Table 4, 2b).

Access to Methadone

Only accredited and certified opioid treatment programs¹⁷ can dispense methadone for the treatment of OUD. Regular travel to methadone clinics was identified by staff as a significant barrier for residents with physical or cognitive impairments (Table 4, 3a). Furthermore, facilities usually require an escort to methadone appointments. However, staff is not always available, and occasionally, residents travel unaccompanied. Staff commented that unsupervised travel presents a perceived added diversion risk (Table 4, 3a).

Access to Buprenorphine

Beyond physically obtaining the medication, respondents commented that access to buprenorphine-containing medications is often limited by both insurance coverage and physician availability (Table 4, 3b). Only 2 of 24 facilities reported having a DATA-waivered physician on staff who prescribed buprenorphine. However, respondents note that unlike methadone buprenorphine-containing medications can be delivered to the NH without requiring a resident to visit a clinic for pickup, which improves access.

Services Provided

Other facility-provided programming varied widely. Some facilities offered counseling groups, but frequency and facilitation were variable, ranging from twice a day every weekday to sporadically. Group sessions might be conducted by an outside licensed clinical social worker, Narcotics Anonymous or Alcoholics Anonymous, or on-site CADCs or psychiatric rehabilitation services coordinators. In 1 case, the CADC supervised MOUD administration (Table 4, 3c).

Overdose Protocol

Despite being in the same network, overdose protocols as well as staff familiarity with them varied across facilities. Thirteen of 24 respondents acknowledged having staff who had administered naloxone and/or called 911 as part of overdose management. Although most respondents said naloxone was readily accessible, this was not universal. One respondent thought a physician order was necessary for naloxone, so calling 911 was their primary overdose response (Table 4, 3d). Administrators noted that overdose management is rarely taught in nursing school so staff learned it on the job (Table 4, 3e).

Knowledge, Perceptions, and Attitudes

Staff in facilities with and without formal programming expressed an overarching belief that NH residents with substance use are a growing population (Table 5, 1a) and are trending younger (Table 5, 1b). Several interviewees noted these residents were often undomiciled with longer NH stay (Table 5, 1c).

TABLE 5.

Knowledge, Perceptions, and Attitudes of Nursing Home Staff

Theme	Subtheme	Illustrative Quotations
1. Population trends	a. Increasing prevalence in history of substance use 25 (8)	“[…]the population that comes to nursing homes has changed. The little old ladies, the little old man…alot of them are going back home, to home health, or to supportive living facilities. In […] the business side of things, you’re more likely now to take someone who has a history of drug use.” —Administrator 072101
	b. Change in demographics 15 (3)	“[…]you don’t have the gray hair ladies anymore, you have the 45-year-olds, 30-year-olds. They are changed and almost a lot of’em have secondary drug addiction.” —Regional Director 070702
	c. Homelessness and substance use 13 (4)	“A lot […] come to us homeless or if they have fractured relationships with their families and they don’t have anywhere to go.” — Administrator 072002
2. Staff preparedness	a. Low preparedness leads to stigma 15 (5)	“[The nurses] will have a panic attack [if] you come upon a patient [with] street narcotics in their room, it’s taboo, it’s scary, and the person that you’re dealing with is unsavory to being with.” —Administrator 071705 “If you weren’t taught about [substance use] or didn’t deal with it directly, or know someone directly, you may be against it.” —DON 070606
		“I was dead set against [a formal program]. I didn’t become a nurse to do drug and rehab counseling. When I got to this building, I realized that it’s pretty selfish of me to not want to deal with them. At the end of the day they are still patients, they still need assistance.” —DON 070603
		“[The nurses] holler, scream at them—they have a thing about people that use drugs. The reason is because they’re not educated on addiction […] But like I said, it’s a whole lot of them that probably got somebody in their family that’s doing something.” —CADC 070601
	b. Learning on the job 37 (14)	“I didn’t go to a class. When you when you have [residents with] substance abuse, it’s painful trial and error.” — DON 062901
		“The more you see the more you learn…I didn’t even know what oxycodone looked like[…]” —DON 072202
		“When I first started I was so straight edge, I couldn’t tell you the difference between two substances.” — Administrator 0707_01
	c. Desired formal education 21 (15)	“Overdose and management of overdose is not something that is typically taught in a geriatrics fellowship.” — Physician 071601
		“[Nursing education] should focus more on the mental health part of [substance use]. I actually sat in on an AA meeting[…] It touched me because of how [one of the speakers] went through life.” —DON 070606
		“[I want more training in the] psychiatric aspect of substance use, training in different types of drugs, temperament exercises, psychological exercises. If you take residents who are developmentally delayed you need proper training to deal with those residents… same thing with any resident who is under the influence of substances.” —DON 062901
3. Staff perception of residents	a. Negative perceptions 107 (42)	“I feel like a lot of those people who have substance abuse problems also can be manipulative […] it’s part of their personality” —DON 072202
		“They’re unpredictable and their anxiety can turn into anger and violence. They can be very aggressive.” — Administrator 072002
	b. Time-intensive 13(1)	“It monopolizes the time in the facility because management is [taking] you away from the routine operational tasks.” —Administrator 071705
		“You’re spending 20 minutes arguing with someone […] when you could [be spending] time with your other residents.” —DON 072202
	c. Risk of visitors 20 (9)	“People’s family members will bring drugs up […] sometimes when they go out for smoke break […] they have had stuff thrown over the gate to them.” —Administrator 072002
		“The new [patient] had a family member come and bring him some stuff and they both overdosed.” —DON 070603
		“Families are big enablers […] They’ll come bring them drugs. I’ve had people bring it in a Dunkin’ Donuts’ bag… rolled up inside a sandwich. They can be creative.” —DON 062901
	d. Substance use disorders as a medical illness 5(3)	“[They] have a disease. that’s what it is. [Their] behavior is not out of the norm. it’s normal for an addict.” —CADC 070601 “Some people handle stress or outside stressors differently, some people eat, some people shop, some people take illicit drugs, some people drink. It is a medical diagnosis, and we have to be more educated on substance abuse in nursing. Im going to work since I got the role. I want to actually educate them because it is a diagnosis.” —DON 070606
	e. Association of mental illness and substance use 32 (13)	“[it’s like] brother and sister, whenever you see a psychiatric diagnosis, there’s usually alcohol and drug involved.” —Regional Director 070702 “You can have hallucinations and delusions, [so] you need to quiet it down…[substances] help to numb them, it gives them the opportunity to remove themselves even if only temporarily […]” —DON 070801
4. Lack of resources	a. Limited staffing 27(6)	“There might be one LPN for 15 to 20 patients [and] other staff [are not] thinking about substance use disorder. it’s really hard to keep as close of an eye on people as everyone would wish.” —Physician 071604 “There is a very large need. I’m not against having people with addiction […] we just don’t have what they would need.” —DON 072001
		“We started the suboxone, but we needed to monitor every 15 minutes. I’m only there for 8 hours and the nurses have 15 patients, or more. We were all just nervous about it, it just never worked, and the patient left AMA.” —NP 071401

N total thematic codes (n thematic codes from facilities with formal programming).

CADC indicates certified alcohol and drug counselor; DON, director of nursing; LPN, licensed practical nurse; NP, nurse practitioner.

In light of these perceived population changes, staff-identified barriers to care can be categorized into 3 major themes: (1) staff preparedness, (2) staff perception of residents, and (3) lack of facility resources.

Staff Preparedness

Staff noted a general unpreparedness in managing this population (Table 5, 2a). Many nurses noted that care for those with SUD is not taught routinely in nursing school, and the NH may be the first time that staff are exposed to this population (Table 5, 2b). Participants noted that this lack of exposure might result in a fear of care of these residents, which could further exacerbate negative perceptions of substance use (Table 5, 2a).

Administrators and staff in facilities without formal substance use programming identified a need for education on substance identification, withdrawal, and overdose (Table 5, 2c). The need extended to physician level as one of the NH physicians commented that addiction care is not routinely included in geriatrics fellowship (Table 5, 2c). Conversely, staff in facilities with formal programming noted they felt more prepared to identify and manage withdrawal and overdose.

Staff Perception of Residents

Staff perceptions of and attitudes toward NH residents with a history of substance use were wide-ranging. Many staff had negative perceptions of residents with a history of substance use. Staff described these residents as time intensive (13 of 24), manipulative (11 of 24), aggressive of violent (10 of 24), drug-seeking (3 of 24), scary of taboo of unsavory (6 of 24), unpredictable (5 of 24), challenging (5 of 24), or deceptive (4 of 24) (Table 5, 2a, 5.3a, b). Furthermore, residents with substance use histories were often associated with having overall “[bad] behaviors” (11 of 24) and were perceived as “taking resources” from other residents (Table 5, 3b). Negative perceptions of residents were exacerbated in environments with low staff-to-resident ratios (Table 5, 4a).

Staff offered anecdotes of illicit drug use in the NH and the possible consequences of accepting patients who might actively use. Specifically, a number of staff perceived family/friends as potential sources for an ongoing use disorder. The level of perceived risk associated with accepting patients was noted to be a function of whom they might be associated with (Table 5, 3c).

Whereas some staff characterized SUD as being part of a resident’s “personality” (Table 5, 3a), other staff characterized SUD as a lifelong medical diagnosis (Table 5, 3d). Furthermore, some staff identified this normalization of use disorders as a medical diagnosis as a potential educational opportunity (Table 5, 3d). Many staff identified a relationship between psychiatric diagnoses and substance use and perceived preexisting mental illness as a risk factor for a SUD (Table 5, 3e).

Lack of Overall Resources

Lastly, many participants commented on the lack of overall resources. Even in facilities with formal programming, some staff felt NHs were often inappropriate settings for persons with substance use because of the lack of resources. Moreover, participants noted that MOUD induction is particularly difficult to arrange in NHs because of inadequate staffing for appropriate monitoring (Table 5, 4a), and low staff-to-resident ratios could exacerbate aggressive behaviors in response to delayed dispensing of MOUDs. Despite this, many staff acknowledged that older adults with SUD are limited when considering post–acute care, and an NH may be the only option.

DISCUSSION

The admission and management of NH residents with SUD/OUD are variable and facility-dependent. Consequently, admissions are highly individualized based on facility-selected factors including history, age, length of substance use, and time since last use. One common thread is the overarching attempt to stratify potential admissions by the risk the resident may bring to the facility, staff, or other residents. This perception of risk differs between facilities and compounds the complex admission criteria, which results in a lack of consistency even within the same network. This can impair or prevent efficient and safe hospital discharges. As NH admissions for those with SUD/OUD or active treatment with MOUDs are protected under the Americans with Disabilities Act,¹⁶ understanding barriers to admission is crucial in supporting facilities to provide standardized and high-quality care to their residents.

Variability in facility policies in other areas of care for admitted residents with SUD may lead to care deficits. For example, in an NH population where residents have cognitive or physical impairments, a policy of self-administration of MOUDs from a double lockbox may prevent admission for patients who cannot self-administer medication. In addition, the daily or weekly travel required to receive MOUDs from methadone clinics is a significant barrier for many NH residents with functional impairments. Appropriate evidence-based management of OUD may be threatened if the facility cannot adapt their MOUD-dispensing protocols to serve this population. In addition, there are no formalized tools in NHs to screen and manage OUD to our knowledge, and the few SUD screening tools validated for the geriatric population predominantly center around alcohol use disorders.^18,19 Each facility with formalized programming developed its own model, which contributes to the variability of OUD/SUD care.

Finally, stigma toward addiction is prevalent among staff who were interviewed. Stigma surrounding residents with SUD/OUD exacerbates health disparities for an already vulnerable group with higher rates of comorbid conditions such as pain, polypharmacy, and geriatric syndromes that necessitate post–acute care.²⁰ Improved education around the management of residents with SUD/OUD in the NH setting may foster greater empathy and understanding of these disorders as a medical diagnosis. Standardized care protocols and staff training will be critical to close gaps and raise capacity to care for this population in NHs.

Limitations

A few potential study limitations are worth noting. It was conducted in a single-care network in one geographic region, which potentially limits generalizability. Given the small sample size, the demographic distribution in the participant sample may not mirror the distribution of the entire facility staff. Therefore, the finding that race was significantly different between staff in facilities with and without formal programming requires exploration in a larger sample size before postulating the reason behind this demographic difference.

Furthermore, the majority of participants interviewed held senior management roles, which excludes perspectives of frontline staff. Future qualitative work should engage more varied roles (eg, bedside staff, case managers, patient navigators, unit managers). Nevertheless, administrators are most familiar with admission policies that impact access to skilled care services for this population. Although administrators interviewed are familiar with admission protocols, the actual admission evaluation and decision process is directed by a facility-specific office based on an internal predetermined protocol. Most participants agreed that NH should accept patients with SUD/OUD. Next steps should include evaluation of facility-specific predetermined protocols to assess whether they mirror staff perspectives. Another question to explore is whether the individual risk concerns cited by interviewed staff are formally reflected in legal risk assessments from a network-level organizational perspective.

Finally, these interviews did not explore effects of reimbursement on admission and care decisions for this population in depth. It is unclear how a use disorder diagnosis affects NH’s reimbursement, as most qualifying diagnoses relate to a recent hospitalization or change in functional status.²¹ Further investigation on how use disorder diagnoses impact reimbursement is needed.

CONCLUSIONS

Patients with SUD/OUD face significant barriers to receiving post–acute care. Admission and management protocols are heterogeneous, even within 1 NH network in 1 state. Many barriers identified in this study could be ameliorated with policy changes such as eliminating the DATA waiver and modifying methadone regulations to allow ease of management within NHs. As the population of older adults with SUD/OUD continues to grow, it is imperative not only to develop standardized staff education and care guidelines to improve quality and access to care for this population, but also to advocate for thoughtful and compassionate policy change that allows NHs to offer best practice care to their residents.