Family History of Substance Use Disorder and Likelihood of Prescription Drug Misuse in Adults 50 and Older

Abstract

Objective:

Individuals who are family history positive (FHP) for substance use problems have increased risk for substance use, substance use disorders (SUDs), and psychopathology. Links between FHP status and prescription drug misuse (PDM) have not been well investigated; this study examined PDM in adults 50 and older by FHP status.

Methods:

Data were from the U.S. NESARC-III (n= 14,667). Participants reported their opioid PDM, tranquilizer/sedative PDM, SUD, psychopathology, and family history status (i.e., first- and second-degree relatives with alcohol/substance use problems). Prevalence rates were estimated by FHP status, and logistic regressions compared FHP and family history negative (FHN) groups.

Results:

FHP status was associated with significantly higher rates of PDM (e.g., past-year opioid PDM, FHP: 3.8%, FHN: 1.5%) and SUD from PDM (e.g., past-year SUD, FHP: 1.2%, FHN: 0.2%); also, prevalence varied by family history density, with the highest rates in those with three or more relatives with substance use problems (e.g., past-year opioid PDM: 5.5%). Overall, 32.2% of FHP individuals with past-year PDM had past-year co-occurring SUD and psychopathology diagnoses, versus 11.0% of FHN individuals.

Conclusion:

FHP status could inform treatment decisions in adults 50 and older with conditions for which prescription opioids or tranquilizer/sedatives are indicated.

INTRODUCTION

Having a family history of alcohol or substance use problems (family history positive, or FHP) is an established risk factor for of alcohol and other substance use, heavy or problematic use, and use disorder symptoms (Chartier, Thomas, & Kendler, 2017; Haeny et al., 2020; VanderBroek, Acker, Palmer, de Wit, & MacKillop, 2016). Risk imparted by FHP status covaries with family history density (FHD), as individuals with a greater number of family members with histories of alcohol or substance use problems have greater levels of use, substance use disorder diagnoses (SUD), and other consequences from use (Karriker-Jaffe, Chartier, Bares, Kendler, & Greenfield, 2021; Pandey et al., 2020).

Research in adolescents and young adults suggests a number of higher-order factors differ between FHP individuals and those without a family history (family history negative, or FHN). These include greater impulsivity (Khemiri et al., 2020), greater discounting of delayed rewards (VanderBroek et al., 2016), altered reward processing (Filippi et al., 2019), higher levels of externalizing or internalizing behavior (Kamarajan et al., 2015; Wasserman et al., 2020), such as aggression or depressive symptoms. Neuroimaging studies also indicate significant differences in the structures and pathways thought to underlie the altered behavioral responses found in FHP individuals (Cservenka, 2016; Lees et al., 2021).

In addition to these neurobiological differences versus FHN individuals, FHP individuals have elevated prevalence rates of depressive symptoms and major depression (Dawson & Grant, 1998), conduct disorder or antisocial personality (Barnow, Schuckit, Lucht, John, & Freyberger, 2002), and other psychopathology than FHN individuals (Araujo & Monteiro, 1995). They also encounter greater social and familial disadvantage. FHP individuals experience higher rates and levels of childhood maltreatment (Taplin, Saddichha, Li, & Krausz, 2014) and socioeconomic disadvantage and deprivation (Hill, Blow, Young, & Singer, 1994), and they also report higher levels of perceived parental rejection and lower parental affection than FHN individuals (Barnow et al., 2002).

Nonetheless, gaps in our knowledge about FHP individuals remain. First, no nationally representative studies have examined the influence of FHP status on prescription drug misuse (PDM), which is the use of a controlled prescription medication without a prescription or in ways not intended by the prescriber (e.g., non-orally or more frequently than prescribed). Liebschutz and colleagues (2010) examined a sample of adults with chronic pain in a primary care setting and found that presence of a prescription opioid or sedative use disorder was associated with FHP status, suggesting the importance of further study in nationally representative data. Second, family history studies have not separately examined late middle aged and older adults 50 years and older. Prevalence rates of PDM have increased since the early 2000s in such adults (Schepis & McCabe, 2016; West & Dart, 2016) and PDM from opioid or tranquilizer/sedative medication (e.g., benzodiazepines) is associated with significant consequences that include overdose, unintentional injuries, and memory problems (West & Dart, 2016). PDM also is linked to elevated rates of psychopathology across the population (Martins et al., 2012) and in adults 50 and older (Schepis, Simoni-Wastila, & McCabe, 2019). Finally, some of the familial factors associated with FHP status are also associated with elevated prevalence of PDM, including greater parent-child attachment serving as a protective factor (Cerdá et al., 2014) and parental PDM and other substance use as a risk factor for PDM by offspring (Griesler, Hu, Wall, & Kandel, 2021).

Research investigating potential links between FHP status and PDM in aging adults could address these literature gaps. FHP status is a useful screening tool for both any substance use and substance use severity (Dawson & Grant, 1998; Lee, Lee, Kim, Yi, & Choi, 2013; Warner, White, & Johnson, 2007); it also suggests a higher prevalence rate of psychopathology or comorbid SUD and psychopathology (Dawson & Grant, 1998; Redgrave, Coughlin, Heinberg, & Guarda, 2007). As such, if FHP status is linked to PDM in adults 50 years and older, it could suggest the importance of screening for FHP status in those prescribed opioid or tranquilizer/sedative medication. Such screening could identify individuals at increased risk who might benefit from non-pharmacological interventions and/or careful medication monitoring to prevent initiation and entrenchment of PDM. Furthermore, the links between PDM and psychopathology suggest that psychopathology could be elevated in adults 50 and older who are both FHP and engaged in PDM. If so, this would highlight potential mental health treatment needs and clinical characteristics to target in treatment among FHP aging adults, especially those with a PDM history.

Given these gaps in the literature, we used data from the National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III) to examine links between FHP status and past-year and lifetime opioid or tranquilizer/sedative PDM and substance use disorder (SUD) from such PDM in a nationally representative sample of adults 50 years and older. We also examined differences in relationships between PDM and non-PDM SUD prevalence, and PDM and psychopathology, as functions of FHP status. Finally, all analyses incorporated FHD to understand whether observed relationships varied as a function of the number of family groups (e.g., parent, sibling) with a history of alcohol or substance use problems.

MATERIALS AND METHODS

The NESARC-III was a nationally representative survey of the civilian and non-institutionalized U.S. population 18 years and older. Sampling used a multistage probability design, with oversampling for non-white adults. Weights were used to create a sample representative of the U.S. adult population at the time of the survey, with weighting adjustments for oversampling and non-response. The overall response rate was 60.1% (screening response rate: 70.2%, interview response rate: 84.0%), which is similar to other nationally representative surveys of substance use, like the U.S. National Survey on Drug Use and Health (Center for Behavioral Health Statistics and Quality, 2017). The Westat and NIH IRBs reviewed and approved the NESARC-III protocol, and the first author’s IRB exempted this study from further oversight. Other resources describe the NESARC-III methodology in more detail (Grant et al., 2014; Grant et al., 2015).

Participants

From an initial sample of 36,309 US adults, this research used data solely from adults 50 years and older (n = 14,667). The weighted older adult sample was mainly female (53.3%), white, non-Hispanic (75.5%), married (60.9%), and 50-64 years of age (59.6%). The sample was also 10.0% black, non-Hispanic, and 9.0% Hispanic/Latino.

Measures

Lifetime and Past-Year Prescription Drug Misuse

The NESARC-III captured PDM in a larger module that assessed non-alcohol and non-tobacco drug use. The prescription medication classes of interest for this investigation were opioid and tranquilizer/sedative medications; stimulant PDM was excluded due to its very low prevalence rates in adults 50 and older (Schepis, Ford, Wastila, & McCabe, 2020). The larger drug use module was introduced via:

“Now I’d like to ask you about your experiences with medicines and other kinds of drugs that you may have used on your own – that is, either without a doctor’s prescription; in greater amounts, more often, or longer than prescribed; or for a reason other than a doctor said you should use them.”

Opioid medications were described as: “Painkillers, for example…methadone, codeine, Demerol, Vicodin, Oxy-Contin, opioid, oxy, Percocet, Dilaudid, Percodan, morphine”. Tranquilizer/sedative medications were described as: “Sedatives or tranquilizers, for example…barbs, downers, Ambien, Lunesta, phenobarbital, pentobarbital, Halcion, Tuinal, Nembutal, Seconal, Librium, Valium, Xanax, benzodiazepines, tranks (tranquilizers), Ativan.” Participants endorsing lifetime PDM were asked whether the most recent episode was within the last 12 months or prior to that, capturing past-year PDM.

Family History of Alcohol and/or Drug Problems

The NESARC-III captures first- and second-degree family history of alcohol or drug problems in separate modules, beginning with alcohol. To improve assessment validity, it uses five DSM-5 alcohol use disorder or non-alcohol SUD symptoms to clarify the definition of family history. The family history of alcohol problems module is introduced with: “Now I would like to ask you some questions about whether any of your relatives, regardless of whether or not they are now living, have ever been alcoholics or problem drinkers. By alcoholic or problem drinker, I mean a person who has physical or emotional problems because of drinking; problems with a spouse, family, or friends because of drinking; problems at work or school because of drinking; problems because of driving after drinking or a person who seems to spend a lot of time drinking or being hung over.” The family history of drug use problems is virtually identical (including the DSM-5 SUD symptoms referenced), with “alcoholics or problem drinkers” replaced by “had/having problems with drugs.”

Both assessments query biological parents, siblings, and children (first-degree relatives); biological grandparents and aunts/uncles (second-degree relatives) are also captured, though half-siblings are not assessed. For this study, four variables are captured, with alcohol and/or substance use problems combined. The first three variables are: one, any family history; two, any first-degree family history; and three, any second-degree family history.

The final categorical variable is based on FHD via count of first- and second-degree relative groups (above) with a family history, with categories of zero, one or two, and three or more relative groups with a family history of alcohol and/or substance use problems. In selecting this categorization (i.e., zero, one or two, and three or more relative groups with a family history), sensitivity analyses were performed that examined differences between those with one, two, or three or more family groups in terms of lifetime PDM, past-year PDM and the SUDs. These sensitivity analyses found only one significant difference between those with one and two family groupings (lifetime SUD from PDM) and only one significant difference between those with two and three family groupings (past-year tranquilizer/sedative PDM). These sensitivity analyses supported the decision to split family history groups into the current categories, given the limited significant results specific to having exactly two relative groups with a family history of alcohol and/or substance use problems.

Substance Use Disorders (SUDs)

The Alcohol Use Disorder and Associated Disabilities Interview Schedule-5 (AUDADIS-5) was used to capture SUDs, with strong reliability (Grant et al., 2015). The lifetime and past-year SUDs included in analyses were alcohol use disorder, tobacco use disorder, other drug use disorder, and SUD from PDM. Other drug use disorder was any of cannabis, stimulant, hallucinogen, inhalant, heroin, or other drug (e.g., MDMA) use disorders. SUD from PDM specifically captured SUD from opioid and/or tranquilizer/sedative PDM.

Mental Health

The AUDADIS-5 was also used to capture DSM-5 psychopathology, with eleven lifetime and past-year diagnoses included here: major depression, dysthymia, bipolar I, post-traumatic stress disorder (PTSD), generalized anxiety, social anxiety, panic disorder, specific phobia, agoraphobia, antisocial personality disorder, and borderline personality disorder. For lifetime and past-year diagnoses, the five anxiety diagnoses were collapsed into an “any anxiety disorder” category, given smaller samples with some diagnoses and lifetime or past year PDM (e.g., agoraphobia), which created larger standard errors. Also, all past-year mood disorders were aggregated into an “any mood disorder” category; again, this was due to smaller samples with both a past-year depressive disorder and PDM. Lifetime suicide attempts (yes/no) were also included. The AUDADIS-5 has fair to good reliability for psychopathology (Grant et al., 2015). Mental health-related quality-of-life was captured via the SF-12v2, with T-scores normed to a mean of 50 and standard deviation of 10; it is reliable and valid in older adults (Bentur & King, 2010; Ware, Kosinski, & Keller, 1996).

Data Analyses

Stata 16.1 (StataCorp LLC, College Station, TX) was used for analyses, with the complex survey features and data weights of the NESARC-III incorporated in analyses. Initial analyses estimated prevalence rates and 95% confidence intervals (95% CIs) of lifetime and past-year opioid or tranquilizer/sedative PDM by the three dichotomous family history variables and the FHD variable (i.e., zero, one or two, and three or more relative groups with a family history); this also was done for lifetime and past-year SUD from PDM. Second, prevalence rates and 95% CIs were estimated within those with lifetime PDM for the captured lifetime SUD and mental health diagnoses (including lifetime suicide attempts) by the three dichotomous family history variables: any, first-degree, and second-degree family history.

Controlling for age, race/ethnicity, sex, household income, marital status, employment status, and education, logistic regressions computed adjusted odds ratios (aORs) of the lifetime SUD and mental health diagnoses for FHP adults, with FHN individuals as the reference group. These analyses were repeated within those engaged in past-year PDM for the captured past-year SUD and mental health diagnoses, using the same control variables. Finally, lifetime SUDs and mental health diagnosis prevalence and 95% CIs were estimated within those with lifetime PDM, comparing FHN to FHP individuals, focused on the four-level FHD variable (i.e., zero, one or two, and three or more relative groups with a family history) as a sensitivity analysis to study more nuanced FHP effects. Again, logistic regression estimated aORs, with the zero family member group as the reference, and these analyses controlled for the sociodemographics noted previously.

RESULTS

Among all adults 50 years and older, 8.4% (95% CI = 7.6-9.2%) and 6.4% (95% CI = 5.9-7.0%) engaged in lifetime opioid or tranquilizer/sedative PDM, respectively (see Table 1). For past year PDM, these numbers were 3.1% (95% CI = 2.8-3.6%) for opioids and 1.7% (95% CI = 1.5-2.0%) for tranquilizer/sedative PDM.

Table 1:

Family History of Alcohol and/or Other Drug Problems and Prescription Drug Misuse (50 and older; n = 14,704)

	Lifetime PDM		Past Year PDM		Lifetime SUD from PDM	Past Year SUD from PDM
	Opioid	Tranquilizer/Sedative	Opioid	Tranquilizer/Sedative
	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)
Entire Sample	8.4 (7.6-9.2)	6.4 (5.9-7.0)	3.1 (2.8-3.6)	1.7 (1.5-2.0)	1.8 (1.5-2.1)	0.9 (0.7-1.1)
Any Relative
Yes	10.2 (9.3-11.1)***	7.8 (7.2-8.5)***	3.8 (3.3-4.4)***	2.1 (1.7-2.4)**	2.4 (2.1-2.9)***	1.2 (0.9-1.5)***
No	4.2 (3.4-5.1)	3.2 (2.5-4.0)	1.5 (1.1-2.0)	0.9 (0.6-1.3)	0.3 (0.2-0.5)	0.2 (0.1-0.4)
Any First Degree Relative †
Yes	11.0 (10.0-12.0)***	8.5 (7.9-9.3)***	4.1 (3.6-4.8)***	2.4 (2.0-2.8)***	2.8 (2.4-3.3)***	1.3 (1.0-1.7)***
No	5.1 (4.3-6.1)	3.8 (3.1-4.6)	1.9 (1.5-2.4)	0.9 (0.6-1.2)	0.5 (0.4-0.8)	0.3 (0.2-0.5)
Any Second Degree Relative ‡
Yes	10.7 (9.8-11.7)***	8.5 (7.7-9.3)***	4.0 (3.5-4.7)***	2.1 (1.7-2.6)*	2.9 (2.4-3.4)***	1.4 (1.1-1.8)***
No	6.1 (5.3-7.1)	4.4 (3.8-5.1)	2.3 (1.9-2.7)	1.3 (1.0-1.7)	0.8 (0.6-1.1)	0.4 (0.2-0.6)
Number of Relatives
0	4.2 (3.4-5.1)	3.2 (2.5-4.0)	1.5 (1.1-2.0)	0.9 (0.6-1.3)	0.3 (0.2-0.5)	0.2 (0.1-0.4)
1 or 2	8.1 (7.1-9.3)***	5.8 (5.1-6.6)***	3.0 (2.5-3.7)***	1.5 (1.2-2.0)*	1.6 (1.2-2.0)***	0.7 (0.5-1.0)**
3 or more	14.3 (12.8-15.9)***	11.9 (10.8-13.1)***	5.5 (4.6-6.6)***	3.1 (2.5-3.9)***	4.2 (3.5-5.1)***	2.1 (1.6-2.8)***

^*denotes difference at p < 0.05

^**at p < 0.01

^***at p < 0.001; all differences are from “no” or “zero relatives”

PDM = Prescription Drug Misuse

^†First degree relatives are composed of biological parents, full siblings, and children.

^‡Second degree relatives are composed of maternal and paternal biological grandparents, aunts, and uncles.

All analyses controlled for age, race/ethnicity, sex, household income, and education.

Lifetime and Past-Year PDM and SUD from PDM by Family History Status and Density

Per Table 1, having any FHP family member was associated with higher rates of lifetime and past-year opioid or tranquilizer/sedative PDM. To illustrate, 10.2% and 7.8% of FHP individuals engaged in lifetime opioid or tranquilizer/sedative PDM, respectively, versus 4.2% and 3.2% who were not FHP. Over the past year, 3.8% and 2.1% of FHP individuals engaged in opioid or tranquilizer/sedative PDM, respectively, versus 1.5% and 0.9% of those without a family history. Similar patterns held when examining those with any first- or second-degree family member with an alcohol and/or substance use problem. Strikingly, those with three or more family groups with an alcohol and/or substance use problem had the highest rates of lifetime or past-year PDM (e.g., 14.3% for lifetime and 5.5% for past-year opioid PDM). These rates were significantly higher than those with one or two family groups, which were significantly higher than those who were not FHP.

The pattern of higher rates in FHP individuals, with the highest in those with three or more family groups, held for both lifetime and past-year SUD from PDM. Of those with any family history, 2.4% had a lifetime SUD from PDM, versus 0.3% of those who were not FHP. Those with three or more family groups with an alcohol and/or substance use problem had the highest rates, at 4.2% (see Table 1). Online-only Supplemental Table 1 captures lifetime and past-year rates of PDM or SUD from PDM by specific family member groupings. Again, having a family member with an alcohol and/or substance use problem history in the group was associated with significantly higher rates of PDM or SUD from PDM versus those who were not FHP.

Lifetime SUDs and Psychopathology by Family History Status in those with Lifetime PDM

As captured in Table 2, FHP individuals who engaged in lifetime PDM had higher aORs for SUD from PDM and all other examined SUDs, mental health diagnoses and lifetime suicide attempts than those who were not FHP. Around 20% of FHP individuals with lifetime PDM had a lifetime SUD from PDM, versus less than 10% of those who were FHN. Roughly half of FHP individuals who had engaged in lifetime PDM had an alcohol or tobacco use disorder (48.4% and 51.0%, respectively), nearly one-third had a lifetime major depression or anxiety disorder diagnosis (31.5% and 31.2%, respectively), and 11.6% had made a lifetime suicide attempt (versus 2.6% of those with lifetime PDM but not family history).

Table 2:

Lifetime Correlate Prevalence by Family History Status among Those Engaged in Lifetime Prescription Drug Misuse (PDM)

	Any Family History			Any First-Degree Family History†			Any Second-Degree Family History‡
	Yes	No	OR	Yes	No	OR	Yes	No	OR
SUD from PDM	19.5 (16.5-23.0)	5.1 (2.9-8.8)	4.23*** (2.29-7.81)	21.0 (17.7-24.8)	7.6 (5.2-10.9)	2.89*** (1.85-4.54)	21.7 (18.4-25.4)	9.8 (7.1-13.5)	2.30*** (1.60-3.32)
Alcohol Use Disorder	48.4 (44.3-52.5)	23.6 (17.5-31.1)	2.95*** (1.94-4.48)	50.8 (46.7-54.8)	28.1 (22.7-34.3)	2.61*** (1.85-3.68)	52.3 (48.2-56.5)	31.3 (26.4-36.7)	2.34*** (1.81-3.02)
Tobacco Use Disorder	51.0 (46.7-55.2)	24.2 (18.6-30.9)	2.87*** (1.94-4.25)	53.3 (49.0-57.5)	29.7 (24.5-35.5)	2.31*** (1.70-3.15)	52.7 (47.9-57.4)	36.7 (31.4-42.4)	1.69*** (1.27-2.26)
Other Drug Use Disorder§	26.1 (22.8-29.6)	14.3 (9.6-20.8)	1.82** (1.20-2.77)	27.6 (24.2-31.2)	15.5 (10.9-21.7)	1.81** (1.17-2.80)	29.3 (25.6-33.3)	15.7 (12.1-20.2)	1.98*** (1.48-2.66)
Major Depression	31.5 (27.7-35.5)	18.0 (13.9-22.9)	1.99*** (1.40-2.83)	32.7 (28.6-36.9)	20.7 (16.5-25.7)	1.81*** (1.20-2.54)	37.6 (33.3-42.1)	24.3 (20.6-28.4)	1.66*** (1.23-2.25)
Dysthymia	12.6 (10.7-14.8)	6.0 (3.4-10.5)	1.92 (0.99-3.72)	13.1 (11.2-15.4)	7.5 (5.1-10.9)	1.61* (1.01-2.59)	14.1 (11.6-17.1)	7.4 (5.0-10.7)	1.80* (1.07-3.01)
PTSD	11.5 (9.3-14.1)	3.0 (1.5-5.9)	3.76** (1.69-8.37)	12.4 (10.0-15.3)	4.3 (2.7-6.6)	2.78*** (1.62-4.79)	13.3 (10.7-16.4)	4.8 (3.2-7.1)	2.67*** (1.66-4.28)
Any Anxiety Disorder¶	31.2 (28.1-34.4)	11.9 (7.1-19.3)	3.12*** (1.73-5.63)	32.6 (29.3-36.1)	16.4 (12.5-21.3)	2.22*** (1.58-3.11)	34.5 (30.8-38.4)	17.4 (13.8-21.8)	2.28*** (1.64-3.16)
Borderline Personality	23.8 (21.0-26.9)	8.9 (5.1-15.1)	2.78** (1.50-5.20)	25.9 (22.8-29.3)	9.9 (6.9-14.2)	2.70*** (1.71-4.24)	26.3 (23.0-30.0)	13.4 (9.7-18.2)	1.99*** (1.33-2.96)
Suicide Attempt	11.6 (9.4-14.3)	2.6 (1.2-5.6)	4.09** (1.74-9.59)	12.3 (9.7-15.3)	4.8 (3.0-7.6)	2.24** (1.26-3.98)	13.3 (10.5-16.7)	5.1 (3.5-7.4)	2.46*** (1.52-3.97)

^*denotes difference from “no” at p < 0.05

^**denotes difference from “no” at p < 0.01

^***denotes difference from “no” at p < 0.001

SUD = Substance Use Disorder; PDM = Prescription Drug Misuse

All analyses controlled for age, race/ethnicity, sex, household income, marital status, employment status, and education.

^†First degree relatives are composed of biological parents, full siblings, and children.

^‡Second degree relatives are composed of maternal and paternal biological grandparents, aunts, and uncles.

^§Other Drug Use Disorder is one or more of DSM-5 Cannabis, Stimulant (including both amphetamine-type stimulant and cocaine), Hallucinogen, Inhalant, Heroin, and Other Drug (e.g., MDMA) Use Disorders.

^¶Any Anxiety Disorder is one or more of Generalized Anxiety Disorder, Specific Phobia, Social Anxiety Disorder, Agoraphobia, and Panic Disorder.

Past-Year SUDs and Psychopathology by Family History Status in those with Past-Year PDM

Fewer significant differences by FHP status were found for past-year diagnoses among those with past-year PDM (see Table 3), but prevalence rates of any SUD from PDM, alcohol use disorder, tobacco use disorder, and any anxiety disorder were significantly higher in FHP individuals than in those who were not FHP. When captured as lifetime or past-year co-occurring SUD and mental health diagnoses (Figure 1), 40.4% of FHP individuals with lifetime PDM had co-occurring diagnoses, versus 16.2% of those who were not FHP. For the past year, 32.2% of those with PDM and FHP status had co-occurring diagnoses, versus 11.0% of those who were not FHP.

Table 3:

Past Year Correlate Prevalence by Family History Status among Those Engaged in Past Year Prescription Drug Misuse (PDM)

	Any Family History			Any First-Degree Family History†			Any Second-Degree Family History‡
	Yes	No	OR	Yes	No	OR	Yes	No	OR
SUD from PDM	25.0 (20.2-30.5)	8.8 (4.0-18.4)	3.76** (1.50-9.45)	25.9 (20.7-33.0)	13.0 (7.8-20.9)	2.24* (1.16-4.32)	29.4 (24.2-35.2)	11.8 (7.6-18.1)	2.92*** (1.69-5.07)
Alcohol Use Disorder	18.0 (14.0-22.9)	7.5 (3.7-14.5)	2.42* (1.05-5.60)	18.9 (14.5-24.3)	9.5 (5.8-15.1)	1.90* (1.01-3.57)	22.8 (17.6-29.0)	6.5 (3.9-10.8)	3.35*** (1.80-6.26)
Tobacco Use Disorder	34.5 (29.6-39.7)	18.5 (12.1-27.3)	2.11* (1.18-3.76)	36.0 (30.4-41.9)	21.1 (14.8-29.3)	1.61 (0.92-2.81)	41.7 (35.0-48.7)	29.8 (22.9-37.6)	1.15 (0.73-1.81)
Other Drug Use Disorder§	10.7 (7.5-15.1)	7.0 (3.0-15.7)	1.46 (0.51-4.20)	11.6 (8.2-16.3)	6.1 (2.5-13.9)	1.78 (0.64-4.93)	12.9 (8.2-17.8)	3.6 (1.6-7.8)	2.43* (1.12-5.28)
Any Depressive Disorder¶	20.8 (16.5-25.8)	15.4 (8.8-25.5)	1.38 (0.68-2.82)	22.2 (17.7-27.5)	13.7 (8.7-20.8)	1.71 (0.99-2.97)	22.3 (17.3-28.2)	16.4 (11.8-22.3)	1.30 (0.80-2.10)
PTSD	12.9 (9.6-17.2)	1.7 (0.2-11.2)	8.82* (1.03-75.94)	14.3 (10.6-19.1)	2.5 (0.9-6.9)	6.31** (1.89-21.08)	14.0 (10.2-19.0)	6.8 (3.9-11.6)	1.98* (1.05-3.71)
Any Anxiety Disorder±	31.7 (26.6-37.1)	10.8 (4.8-22.2)	4.02*** (1.79-9.00)	33.7 (28.4-39.4)	13.9 (7.9-23.3)	2.90*** (1.60-5.25)	35.7 (30.3-41.5)	17.2 (11.5-24.9)	2.26*** (1.40-3.62)

^*denotes difference from “no” at p < 0.05

^**denotes difference from “no” at p < 0.01

^***denotes difference from “no” at p < 0.001

SUD = Substance Use Disorder; PDM = Prescription Drug Misuse

All analyses controlled for age, race/ethnicity, sex, household income, marital status, employment status, and education.

^†First degree relatives are composed of biological parents, full siblings, and children.

^‡Second degree relatives are composed of maternal and paternal biological grandparents, aunts, and uncles.

^§Other Drug Use Disorder is one or more of DSM-5 Cannabis, Stimulant (including both amphetamine-type stimulant and cocaine), Hallucinogen, Inhalant, Heroin, and Other Drug (e.g., MDMA) Use Disorders.

^¶Any Depressive Disorder is either of Major Depressive Disorder or Dysthymia.

^±Any Anxiety Disorder is one or more of Generalized Anxiety Disorder, Specific Phobia, Social Anxiety Disorder, Agoraphobia, and Panic Disorder.

Figure 1: Prevalence of Lifetime and Past Year Co-Occurring Psychopathology and SUD by Family History Status.

All comparisons were significant at a p < 0.001 level.

SUD = Substance Use Disorder

Psychopathology was defined as any of Major Depression, Dysthymia, Bipolar I, PTSD, Generalized Anxiety Disorder, Specific Phobia, Social Anxiety Disorder, Agoraphobia, or Panic Disorder.

Lifetime and Past-Year SUDs and Psychopathology by Family History Density in Those with PDM

Finally, analyses of family history status-correlate relationships indicated that the highest prevalence rates of every lifetime SUD and mental health condition (including suicide attempts) among those with lifetime PDM was highest in those with three or more FHP groups (Table 4). Over half had alcohol or tobacco use disorder (60.8% and 59.7%, respectively), and over one-third had major depression (38.0%), any anxiety disorder (38.6%), or a non-alcohol/non-tobacco SUD (34.7%); 16.8% had attempted suicide, and 24.4% had a lifetime SUD from PDM. Adjusted odds of all correlates were significantly higher in those with three or more FHP groups than in those with one or two, and those with one or two had higher aORs than those with no FHP groups for five of nine outcomes. Past-year correlates among those with past-year PDM are captured in online-only Supplemental Table 2. There were fewer significant differences, though those with three or more FHP groups had the highest prevalence rates across correlates.

Table 4:

Lifetime Correlate Prevalence by Family History Density among Those Engaged in Lifetime Prescription Drug Misuse (PDM)

	No Family Members	1-2 Family Members	≥ 3 Family Members	1-2 versus None	≥ 3 versus None	1-2 versus ≥ 3
	% (95% CI)	% (95% CI)	% (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
SUD from PDM	5.1 (2.9-8.8)	15.4 (11.8-19.8)	24.4 (20.4-28.9)	3.46 (1.75-6.83)***	5.20 (2.86-9.47)***	1.50 (1.05-2.16)*
Alcohol Use Disorder	23.6 (17.5-31.1)	37.8 (32.8-67.2)	60.8 (56.1-65.3)	1.93 (1.27-2.94)**	5.25 (3.31-8.33)***	2.72 (2.00-3.70)***
Tobacco Use Disorder	24.2 (18.6-30.9)	43.6 (37.9-49.4)	59.7 (54.6-64.6)	2.25 (1.44-3.52)***	3.91 (2.64-5.78)***	1.74 (1.26-2.40)***
Other Drug Use Disorder†	14.3 (9.6-20.8)	18.7 (15.0-23.2)	34.7 (30.0-39.7)	1.23 (0.79-1.91)	2.80 (1.77-4.42)***	2.27 (1.62-3.20)***
Major Depression	18.0 (13.9-22.9)	26.0 (21.7-30.8)	38.0 (32.1-44.2)	1.58 (1.10-2.28)*	2.61 (1.69-4.03)***	1.65 (1.13-2.40)**
Dysthymia	6.0 (3.4-10.5)	9.1 (7.1-11.5)	16.7 (13.4-20.7)	1.45 (0.76-2.78)	2.52 (1.22-5.20)*	1.74 (1.19-2.54)**
PTSD	3.0 (1.5-5.9)	6.3 (4.4-9.0)	17.5 (14.0-21.8)	2.17 (0.90-5.24)	5.80 (2.63-12.81)***	2.67 (1.68-4.25)***
Any Anxiety Disorder‡	11.9 (7.1-19.3)	24.9 (21.2-29.0)	38.6 (33.9-43.6)	2.48 (1.33-4.63)**	4.02 (2.23-7.26)***	1.62 (1.19-2.21)*
Borderline Personality	8.9 (5.1-15.1)	16.4 (13.1-20.3)	32.6 (28.5-36.9)	1.90 (0.99-3.62)	4.04 (2.15-7.58)***	2.12 (1.55-2.91)***
Suicide Attempt	2.6 (1.2-5.6)	7.2 (5.4-9.6)	16.8 (13.0-21.5)	2.73 (1.11-6.73)*	5.69 (2.41-13.45)***	2.08 (1.37-3.15)***

^*denotes difference at p < 0.05

^**denotes difference at p < 0.01

^***denotes difference at p < 0.001.

All analyses controlled for age, race/ethnicity, sex, household income, marital status, employment status, and education.

OR = Odds Ratio; SUD = Substance Use Disorder; PDM = Prescription Drug Misuse

^†Other Drug Use Disorder is one or more of DSM-5 Cannabis, Stimulant (including both amphetamine-type stimulant and cocaine), Hallucinogen, Inhalant, Heroin, and Other Drug (e.g., MDMA) Use Disorders.

^‡Any Anxiety Disorder is one or more of Generalized Anxiety Disorder, Specific Phobia, Social Anxiety Disorder, Agoraphobia, and Panic Disorder.

DISCUSSION

Consistent with literature on alcohol and non-PDM substance use (Dawson & Grant, 1998; Lee et al., 2013; Warner et al., 2007), FHP status was linked to higher odds of lifetime and past-year PDM, and lifetime and past-year SUD from PDM in U.S. adults 50 years and older. Similarly, FHP status in those with lifetime PDM was linked to higher odds of lifetime SUD, depressive disorders, anxiety disorders, PTSD, borderline personality disorder, and suicide attempts, versus those with lifetime PDM who were FHN; furthermore, odds of lifetime co-occurring SUD and psychopathology were elevated in FHP individuals, versus FHN adults. Again, these findings are consistent with past research suggesting that FHP status is a marker of greater overall dysfunction, including elevated psychopathology (Dawson & Grant, 1998; Redgrave et al., 2007). For PDM, SUD from PDM, and psychopathology among those with PDM, greater FHD was also associated with stepwise increases in odds of these outcomes. Past-year psychopathology and co-occurring disorders were less consistently linked to FHP status in those with past-year PDM, though past-year alcohol use disorder and any anxiety disorder were consistently linked to FHP status.

A wealth of research indicates that FHP individuals have genetic and neurobiological differences from FHN individuals that predate any use of alcohol or other substances (Cservenka, 2016), and other research consistently finds higher levels of psychopathology, socioeconomic and familial disadvantage (Araujo & Monteiro, 1995; Barnow et al., 2002; Dawson & Grant, 1998; Hill et al., 1994; Taplin et al., 2014). Lees et al. (2021) found elevated sleep disturbances, single psychopathology, and multiple psychopathology diagnoses in FHP children who are 9 or 10 years of age; youth from the same sample also displayed altered patterns of neural activation in response to a reward-based learning task (Kwarteng et al., 2021), and FHP individuals consistently display poorer executive functioning (Cservenka, 2016). FHP individuals also are more likely to suffer abuse and neglect, have greater levels of psychopathology that are also associated with substance use, and experience familial and socioeconomic environments that further predispose the FHP individual to substance use (Araujo & Monteiro, 1995; Barnow et al., 2002; Dawson & Grant, 1998; Hill et al., 1994; Taplin et al., 2014). Edenberg et al. (2019) posit that common genetic traits underlie at least part of the relationship between the development of AUD and major depression. Indeed, research has identified potentially common genes for the disorders examined in this study like the GABRA2 gene, which corresponds to the GABA A2 receptor and has been linked to polysubstance use (Engin, Liu, & Rudolph, 2012), or the CHRM2 gene, which is linked to both AUD (Edenberg & Foroud, 2013) and mood disorders (Cannon et al., 2011).

FHP status may represent a higher-order risk factor for a variety of SUDs and psychopathology (e.g., Araujo & Monteiro, 1995), and this study supports that by adding evidence of higher prevalence rates in FHP persons of PDM, SUD from PDM, and both lifetime psychopathology among those with PDM, and co-occurring SUD and psychopathology among those with PDM. Thus, FHP status is also a risk factor for the development of PDM and may be a useful screening tool for the likelihood of PDM in those who are prescribed opioid or tranquilizer/sedative medications. Given the significant risks associated with these medications in older adults, with tranquilizer/sedative prescribing contraindicated to those 65 and older per the Beers Criteria (2019), FHP status may be a key clinical characteristic to include in a multimodal assessment for risks associated with prescribing these medications to those 50 and older. While FHP status alone is not a reason to discontinue effective medication treatment or to avoid prescribing to those 50 and older, it may be one factor to aid clinician decision making and medication monitoring, especially when it is not clear that medication treatment is the best course of action.

Limitations

First, given that respondents were reporting on the alcohol and substance use of family members, some respondents may have misreported their family history status. For instance, respondents may not have directly observed alcohol and/or substance use in a family member who would meet criteria for an SUD, or respondents could have misperceived a single instance of heavier use (e.g., one intoxicated episode) as indicative of an alcohol or substance use problem. Second, family history status necessarily predates PDM engagement, but this temporal ordering does not establish causality. Furthermore, this research does not identify specific underlying and developmentally important traits (e.g., impulsivity, internalizing symptomatology, prescription drug availability) that potentially influenced these outcomes. Longitudinal studies are needed that examine how pre-existing traits that link family history and substance use later influence incidence and progression of PDM. Also, some degree of self-report and self-selection bias is likely. Despite this past studies indicate self-report substance use data are reliable and valid (O’Malley, Bachman, & Johnston, 1983), and the response rates for the NESARC-II are similar to other large, nationally representative studies (Center for Behavioral Health Statistics and Quality, 2017). The NESARC-III had poorer coverage of adults 65 and older, with likely undersampling of controlled access facilities that housed these older adults, like nursing homes (Grant et al., 2014). Finally, the reliability of the AUDADIS-5 to assess psychopathology was only fair, though the substance use assessment had strong reliability (Grant et al., 2015).

Summary

This investigation provides the first evidence of elevated PDM and SUD from PDM in FHP individuals, versus FHN individuals, in a nationally representative sample of U.S. adults 50 years and older. Furthermore, we found that FHD co-varied significantly with PDM and SUD from PDM, such that prevalence rates were highest in those with three or more relatives with histories of substance use problems. These key results agree with the extensive literature linking FHP status to elevated non-PDM substance use and non-PDM SUD, and they suggest the potential utility of screening for family history status when prescribing opioid or tranquilizer/sedative medication to adults 50 and older. FHP status could be one of many factors that discourage use of these medications in aging adults in favor of other effective treatments for pain (e.g., cognitive-behavioral therapy, acceptance and commitment therapy, short-term non-steroidal anti-inflammatory drug use, or physical therapy), sleep (e.g., cognitive-behavioral therapy for insomnia), or anxiety (e.g., selective serotonin reuptake inhibitors or cognitive-behavioral therapy). Finally, among FHP patients who may require these opioid or tranquilizer/sedative medication, these findings indicate such individuals may need vigilant medication monitoring to prevent PDM.

Data Availability Statement:

The NESARC-III data are available upon request from the U.S. National Institute on Alcohol Abuse and Alcoholism (NIAAA) at https://www.niaaa.nih.gov/research/nesarc-iii/nesarc-iii-data-access.