Obesity and Substance Misuse: Is There a Relationship?

Abstract

The relationship between obesity and alcohol/drug misuse has been examined through both site studies and large epidemiological studies. In reviewing 19 site studies that have examined alcohol misuse among the obese, in comparison with rates before 1990, rates since 1990 have doubled (6.2% to 14.3%). Of the 7 studies that have examined drug misuse among the obese, rates average 8.0%. Given the potential limitations of varying study populations, methodologies, and prevalence assessments, these rates closely reflect those in the general population, according to data from the National Comorbidity Survey Replication study. As for the findings from 5 large epidemiological studies, Overall, findings do not clearly support the contention that obese individuals are at a higher risk for substance misuse, although specific sub-populations of the obese may be (e.g., those with Cluster B personality disorders). If there is an association between some obese populations and substance abuse, this may be accounted for by overlapping neurohormonal correlates. However, some authorities contend that food and drugs compete for the same reward pathways in the brain, suggesting that obesity may actually be protective against substance misuse.

This ongoing column is dedicated to the challenging clinical interface between psychiatry and primary care—two fields that are inexorably linked.

INTRODUCTION

Obesity is a major health concern in the United States and affects 35.7 percent of adults and approximately 17 percent of children.¹ In turn, according to data from the National Comorbidity Survey Replication study, substance use disorders affect a significant minority of the general population.² According to data from this study, which were collected in a community sample during the years 2001–2003, the prevalence of alcohol abuse is 13.2 percent, alcohol dependence 5.4 percent, drug abuse 7.9 percent, drug dependence 3.0 percent, and any substance use disorder 14.6 percent. In examining relationships between obesity and substance misuse, according to Acosta et al,³ there are numerous parallels, including genetic predisposition, personality factors, environmental risk factors, and common neurobiological pathways in the brain. In this edition of The Interface, we present data on the comorbidity between obesity and substance use disorders and further examine this relationship.

THE LITERATURE REVIEW

Through a literature search of the PubMed and PsycINFO databases, we explored studies examining the prevalence of alcohol and/or drug misuse in various samples of obese individuals. In using these search engines, we entered terms such as “obesity, alcohol, alcohol abuse, drug,” and “drug abuse,” and also examined reference lists within identified articles for additional articles. We did not include articles in which the study population was limited solely to individuals suffering from binge eating disorder. Likewise, we did not include articles that were written in another language or did not provide an explicit percentage with regard to prevalence rates (e.g., one article indicated that the prevalence rate was less than 2.5%). While we may have missed some articles, from our review of the literature, we believe that we have identified a robust number of source articles for this review.

SITE STUDIES

To summarize these data, over the past 50 years, a number of site studies (i.e., investigations in a single institution or several institutions) have examined the prevalence of alcohol and/or drug misuse in various obese populations (Table 1).^4-22 Of the 19 studies that we encountered, 11 (57.9%) consisted of subjects recruited from bariatric surgery settings.

TABLE 1.

Prevalence of alcohol and/or drug misuse in site studies of the obese

FIRST AUTHOR (YEAR OF PUBLICATION)	N	SAMPLE CHARACTERISTICS	PREVALENCE
Mendelson (1961)4	25	Men in a medical/psychiatric clinic	8% alcoholism (PTU)
Reivich (1966)5	33	Referrals to a department of psychiatry	3% alcoholism (lifetime)
Atkinson (1967)6	21	Referrals by nonpsychiatric clinics	5% alcoholism (PTU)
Swanson (1970)7	25	Volunteers for a metabolic study	4% alcoholism (current)
Kalucy (1974)8	20	Bariatric surgery clinic	20% alcoholism (lifetime) 10% amphetamine addiction (lifetime)
Hutzler (1981)9	102	Bariatric surgery clinic	0.9% alcoholism (current) 0.9% drug abuse (current)
Gentry (1984)10	33	Bariatric surgery clinic	6% alcoholism (lifetime)
Dubovsky (1985-86)11	52	Bariatric surgery clinic	2.7% alcoholism (current)
Gertler (1986)12	88	Bariatric surgery clinic	2.3% substance use disorder (lifetime)
Marcus (1990)13	50	Advertisements	32% alcohol abuse/dependence (lifetime) 16% drug abuse/dependence (lifetime)
Black (1992)14	88	Bariatric surgery clinic	22.7% alcohol disorder (lifetime) 5.3% drug disorder (lifetime) 23.7% any substance disorder (lifetime)
Yanovski (1993)15	128	Advertisements	7.0% alcohol abuse (lifetime) 5.5% drug abuse (lifetime) 9.4% any substance abuse (lifetime)
Specker (1994)16	100	Advertisements	16% alcohol abuse/dependence (lifetime) 15% drug abuse/dependence (lifetime)
Ricca (2000)17	344	Outpatient metabolic clinic	1.4% alcohol dependence (lifetime)
Mitchell (2001)18	100	Bariatric surgery clinic	2.6% alcohol abuse (PTU) 10.3% alcohol dependence (PTU) 3.8% drug abuse (PTU)
Rosenberger (2006)19	174	Bariatric surgery clinic	5.2% substance use disorder (lifetime) 0.6% substance use disorder (current)
Kalarchian (2007)20	288	Bariatric surgery clinic	17.7% alcohol abuse (lifetime) 13.2% alcohol dependence (lifetime) 6.6% drug abuse (lifetime) 9.4% drug dependence (lifetime) 32.6% any substance use disorder (lifetime)
Ertelt (2008)21	70	Bariatric surgery clinic	7.1% alcohol dependence (current) 1.4% alcohol abuse (current)
Heinberg (2010)22	413	Bariatric surgery clinic	10.8% substance abuse/dependence (lifetime)

Note: PTU = prevalence timeframe unclear (i.e., current, 12-month, lifetime?)

With regard to alcohol misuse, there appear to be two distinct study eras: one era before 1990 and the other era from 1990 onward. Prior to 1990, the average prevalence of alcohol misuse (using the highest reported percentage for current, lifetime, or prevalence timeframe unclear) in the eight relevant studies is 6.2 percent, with a range from 0.9 to 20.0 percent.^8,9 From 1990 onward, the average prevalence of alcohol misuse in the eight corresponding studies is 14.3 percent, with a range from 1.4 to 32.0 percent.^13,17 In other words, during this later time period, the prevalence rate of alcohol misuse among obese study participants more than doubled.

The explanation for the observed increase in the prevalence rate of alcohol misuse between these two study eras remains unclear. Perhaps the difference reflects a reduced candor of pre-1990 patients with regard to their willingness to disclose personal information (i.e., was there a general patient concern that an unfavorable admission to alcohol or drug misuse might result in surgery refusal?). Also, note that while the studies before 1990 and 1990-after have approximately equal rates of bariatric surgery sites, the total numbers of evaluated bariatric surgery candidates during these two time periods are vastly different (295 vs. 1,133); perhaps the difference in volumes explains the difference in reported rates of alcohol misuse. Another possibility may be that over the past 50 years, the assessment strategy for alcohol and drug misuse has improved, resulting in the illusion of higher prevalence rates during the later time era. A final possibility may be that alcohol and drug misuse has genuinely increased in the general population, resulting in a corresponding inflation of the prevalence rate of misuse among the obese during the second research era.

As for drug misuse, we located seven studies since 1981 reporting prevalence rates. In summarizing these findings, we excluded percentages for “substance use disorders” due to our concern that these might include alcohol misuse, rather than solely represent rates for drug misuse. As in the previous analysis, we used the highest prevalence rate per study if two or more rates were indicated, regardless of current, lifetime, or prevalence timeframe unclear. In doing so, the rate for drug misuse among the obese over the past 30 years is 8.0 percent, with a range from 0.99 to 10. percent.⁸ In comparison with rates for alcohol misuse, the rates for drug misuse are somewhat lower.

In comparing the preceding findings with data from the National Comorbidity Survey Replication study,² which provides community reference rates for various psychiatric syndromes and disorders, there appear to be few differences in prevalence rates (Table 2). However, the difficulty in fully interpreting these data is the lack of specificity in studies of the obese with regard to abuse versus dependence as well as varying prevalence assessments.

TABLE 2.

Comparisons of prevalence rates for alcohol and drug misuse between community and obese samples from site studies

SUBSTANCE MISUSE	COMMUNITY SAMPLESa (LIFETIME)	OBESE SAMPLES FROM SITE STUDIES4-22 (Various Prevalence Points)
Alcohol misuse	Abuse: 13.2%	Misuse pre-1990: 6.2%
	Dependence: 5.4%	Misuse 1990 onward: 14.3%
Drug misuse	Abuse: 7.9%	Misuse: 8.0%
	Dependence: 3.0%

^adata from the National Comorbidity Survey Replication study²

LARGE-SCALE EPIDEMIOLOGICAL STUDIES

In more recent years, investigators have examined large epidemiologic studies to investigate relationships between obesity and alcohol/drug misuse. For example, using a cross-sectional nationally representative sample of 9,125 adults in the United States, Simon et al²³ compared the prevalence rates of substance use disorders in participants with a body mass index of less than 30 to participants with a body mass index of greater than 30; rates were 4.3 percent and 2.9 percent, respectively. Using data from the National Epidemiologic Survey on Alcohol and Related Conditions, Pickering et al²⁴ compared odds ratios for past-year psychiatric diagnoses among participants with healthy weights (n=5,984), and participants who were overweight (n=7,814), obese (n=3,856), and extremely obese (n=354). Rates for alcohol and drug use disorders were not statistically significantly different among the groups, with the exception of drug dependence, which was actually lower in the obese subsample. Scott et al²⁵ examined 12,992 individuals in New Zealand and found no relationships between obesity and alcohol or drug use disorders. In examining data from the World Mental Health Surveys, Scott et al²⁶ found no relationship between obesity and alcohol use disorders. Finally, in a study that examined data from the Collaborative Psychiatric Epidemiology Surveys, Rosen-Reynoso et al²⁷ found odds ratios of less than 1.0 in all ethnic groups with regard to relationships between body mass index and substance use disorders over the past 12-months. In contrast to the preceding findings, through reanalyzed data from the National Epidemiological Survey on Alcohol and Related Conditions, Barry and Petry²⁸ reported that overweight and obesity were associated with an increased risk for lifetime alcohol abuse/dependence in men (not women), but that body mass index was not associated with illicit drug use disorders.

EXPLANATIONS

Our review of the literature, which includes both site studies and large epidemiological studies, did not reveal a consistently evident relationship between obesity and substance use disorders. However, some individual studies/samples did evidence such a relationship. For these latter studies, which appear to be in the minority, there may be several lines of adjuvant support. For example, Grucza et al²⁹ found epidemiological support for a link between familial alcoholism risk and obesity in women as well as possibly men. Using data from the National Epidemiologic Survey on Alcohol and Related Conditions, Gearhardt and Corbin³⁰ suggested a neurobiological link between food consumption and alcohol use. Suter³¹ found that alcohol calories count more in moderation than in heavy use, indicating that periodic excessive alcohol use may directly contribute to excess calories and weight gain. In contrast, Kleiner et al³² found a significant inverse relationship between body mass index and alcohol consumption. Indeed, other authorities emphasize that there are many inconsistencies in the addiction model of eating disorders,³³ and while there may be some shared neurohormonal pathways, there appear to be distinct variations, as well.³⁴

As for the general impression that there is not a relationship between obesity and substance use disorders, Warren and Gold³⁵ report that food and drugs may be competing for the same reward sites in the brain. In support of this perspective, they found among women that as body mass index increased, alcohol consumption in the past year decreased.³⁵

CONCLUSIONS

There have been numerous studies on the relationship between obesity and substance use disorders. For alcohol misuse, comparing studies prior to 1990 with studies from 1990-onward, there appears to be a doubling of prevalence rates in the obese in site studies. However, rough averages of the prevalence rates of both alcohol and drug misuse among the obese in site studies closely parallel rates reported in the National Comorbidity Survey Replication data. In addition, in large epidemiological studies, four of five studies found no relationship between obesity and substance misuse. Still, some studies evidence unexpectedly high rates of substance misuse among the obese. These inconsistencies may reflect specific population characteristics and/or methodological variations. Overall, a conservative conclusion appears to be that while there may be some specific obese populations with higher rates of substance misuse, data generally indicate equal or even lower rates of substance misuse in the obese compared with the general population. Only further research will verify these conclusions.