Perceived Need for Care, Help Seeking, and Perceived Barriers to Care for Alcohol Use Disorders in a National Sample

The high rates of alcohol abuse and dependence found in many countries worldwide indicate that alcohol misuse is a large public health problem. Epidemiological studies estimating the lifetime prevalence rates of alcohol use disorders within the population report rates ranging from 12% to 30% ( 1 , 2 ). A useful summary of studies assessing the prevalence rates of alcohol use disorders between the years 1980 and 2000 was done by Somers and colleagues ( 3 ).

Several studies have shown that help seeking and perceived need for treatment of an alcohol use disorder may be influenced by the occurrence of a comorbid mental condition ( 4 , 5 , 6 , 7 ). Individuals with an alcohol use disorder and a concurrent mental disorder have increased odds of experiencing more severe symptoms of alcohol use disorders ( 8 ), and one could hypothesize that this would make them more likely to seek help, because help seeking is shown to be correlated with severity of alcohol use disorder symptoms ( 9 ). Because most individuals with an alcohol use disorder also meet criteria for an additional axis I or axis II diagnosis ( 10 ), there is a need for further investigation of how comorbidity influences help seeking for alcohol use disorders.

A small body of literature on the role of self-perceived barriers to care has accumulated over the past decade. Self-perceived barriers to care can be categorized into two groups, attitudinal barriers and structural barriers ( 11 ). Barriers are considered to be attitudinal if they are based on a person's perception of the problem and are within the person's control—for example, thinking that the problem will get better on its own. In contrast, structural barriers to treatment are outside of a person's control and may act to influence help seeking—for example, the high cost of treatment. Research on self-perceived barriers to care has shown that attitudinal barriers to treatment are endorsed more frequently than structural barriers ( 12 , 13 , 14 , 15 ).

There are several limitations of the previous research on self-perceived barriers and perceived need for treatment of an alcohol use disorder. First, there is a lack of literature on the relationship between the comorbidity of mental disorders and perceived need for treatment of alcohol use disorders within the population. Second, most studies of self-perceived barriers to care examine emotional problems and alcohol problems simultaneously; thus it can be unclear whether responses reflect barriers to care for alcohol use or for some other problem (such as depression).

The first objective of this study was to determine prevalence rates of help seeking and perceived need for care in a nationally representative sample of persons with an alcohol use disorder and to compare these rates with rates of no help seeking and no perceived need for care. Our second objective was to examine the most frequent types of help seeking and self-perceived barriers to care, and our third was to identify the sociodemographic and mental disorder correlates of help seeking, perceived need for care, and no perceived need for care for an alcohol use disorder. This study was conducted with data from the largest addiction and mental health survey to date, which was funded by the National Institute on Alcohol Abuse and Alcoholism and was designed to assess the rates and correlates of alcohol use disorders at the national level. It is the first survey to assess the rates of both axis I and axis II disorders in the population and to ask questions about perceived barriers to care and help seeking in the context of alcohol abuse and dependence.

Methods

Sample

Data were drawn from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) ( 16 ), a nationally representative sample of the U.S. population. The survey is cross-sectional in design and was conducted between August 2001 and May 2002 using computer-assisted personal interviewing. The sample included 43,093 respondents 18 years of age and older. Hispanic and non-Hispanic black households and persons aged 18–24 were oversampled to increase the number of these respondents. The survey had an overall response rate of 81%. Ethical approval was obtained from the U.S. Census Bureau and the U.S. Office of Management and Budget ( 17 ), and respondents gave informed consent.

Dependent variables

We obtained a subsample of individuals who met criteria for a lifetime diagnosis of alcohol abuse, dependence, or both (N=11,843). This sample was further divided into those who met criteria for lifetime alcohol abuse (N=7,062), and those who met criteria for lifetime alcohol dependence (N=4,781). Alcohol abuse and alcohol dependence were analyzed separately to clarify attribution of any patterns of correlates. Alcohol abuse and dependence are both characterized by harmful patterns of drinking, but alcohol dependence is thought to be a more severe form of alcohol use disorder and is characterized by symptoms such as physiologic tolerance and withdrawal ( 18 ).

Three main groups were defined. First, respondents who answered yes to the question "Have you ever gone anywhere or seen [a health care professional] for a reason that was related in any way to your drinking?" were included in the help-seeking group. Types of help seeking included Alcoholics Anonymous (AA) or 12-step meetings; family services and social services; detoxification clinic or ward; inpatient ward; rehabilitation program; outpatient, outreach, or partial-day treatment program; emergency services; halfway house or therapeutic community; crisis center; employee assistance program; clergy, priest, or rabbi; physician, psychiatrist, psychologist, or social worker; and other agency or professional.

Second, respondents who answered yes to the question "Was there ever a time when you thought you should see a doctor, counselor, or other health professional or seek any other help for your drinking, but didn't go?" were included in the perceived need group. Respondents in the perceived need group could choose from a list of 26 barriers to care, which we divided into two categories, structural barriers and attitudinal barriers ( 11 ).

Third, we created a category of individuals who did not perceive a need for help and did not seek it. This category included individuals who responded "no" to both of the questions "Have you ever gone anywhere or seen [a health care professional] for a reason that was related in any way to your drinking?" and "Was there ever a time when you thought you should see a doctor, counselor, or other health professional or seek any other help for your drinking, but didn't go?" Participants with missing data on these questions were excluded from the analyses. Respondents with missing data answered "unknown" to either both of the above questions or answered no to one question and "unknown" to the other.

Independent variables

Sociodemographic and physical conditions. The following sociodemographic variables were included in the analysis: age, race-ethnicity, marital status, income, and education level. On the basis of previous work suggesting that number of health conditions increases the likelihood of perceived need and help seeking, we created a trichotomous variable to represent no general medical conditions, one general medical condition, and two or more general medical conditions ( 7 ).

Mental disorders. Axis I and axis II mental disorders were defined according to DSM-IV criteria using the Alcohol Use Disorders and Associated Disorders Interview Schedule (AUDADIS-IV). Past research ( 19 , 20 ) has documented the reliability of the AUDADIS-IV as a diagnostic tool. Information on the criteria used to assess for alcohol abuse or dependence in the NESARC has been articulated elsewhere ( 21 ). We included individuals who met lifetime criteria for any of the following axis I or axis II disorders: mania, dysthymia, major depression, panic disorder, agoraphobia, social phobia, specific phobia, and generalized anxiety disorder. A general category of drug use disorders (abuse and dependence) was also analyzed. All axis II personality disorders were included in the analysis with the exception of borderline, schizotypal, and narcissistic personality disorders, because these disorders were not included in the survey.

Analysis

All analyses were completed using the appropriate statistical weights to ensure that the data were representative of the national population. We used SUDAAN software ( 22 ) to estimate variances of the odds ratio estimates, and we used stratification and sample weighting information provided in the NESARC data set.

First, rates of lifetime alcohol abuse and alcohol dependence within the entire sample were calculated. Second, the prevalence of each type of help seeking, stratified by gender, was calculated. Previous research has documented gender differences in types of help seeking ( 23 ). A chi square analysis was performed to determine significant differences between males and females. Third, multinomial logistic regression analyses were used to compare help seeking, perceived need, no help seeking or perceived need, and sociodemographic characteristics. No help seeking or perceived need was used as the reference group, and odds ratios were used to compare the different sociodemographic correlates associated with the three groups. Because the study required multiple comparisons, we used a more conservative alpha of .01. Fourth, multiple logistic regression analyses were used to compare help seeking, perceived need, no help seeking or perceived need, and mental disorders. Odds ratios were adjusted for all significant sociodemographic variables. The group that did not seek help or perceive a need for it remained as the reference group, and significance was again tested at p<.01. Fifth, we calculated the prevalence of each perceived barrier to care stratified across gender. A chi square analysis was performed to determine significant differences between men and women.

Results

Of the 43,093 survey respondents, 11,843 (28%) met criteria for a lifetime diagnosis of alcohol abuse or dependence. Of the total respondents, 7,062 (16%) met criteria for lifetime alcohol abuse, and 4,781 (11%) met criteria for lifetime alcohol dependence. Of the respondents who met criteria for a lifetime alcohol use disorder, 9,535 (81%) did not perceive a need for help or seek it.

Help seeking

Table 1 shows the prevalence of help seeking among individuals with lifetime alcohol abuse or dependence. Rates of help seeking were low among individuals with alcohol abuse (7.5%) and higher for individuals with alcohol dependence (24.7%). Several types of help seeking differed significantly between men and women with alcohol abuse and dependence. Among those with alcohol abuse, these included AA or 12-step meetings, detoxification clinic, rehabilitation program, and other agency or professional. Individuals with alcohol dependence also had significant results for those types of help seeking, plus outpatient, outreach, or partial-day treatment program; emergency care; and employee assistance program.

Enlarge table

Sociodemographic correlates of help seeking and perceived need

Table 2 displays the results of a multinomial logistic regression analysis illustrating the differences in sociodemographic variables among individuals who perceived a need for help, sought help, or did neither. Among individuals with lifetime alcohol abuse, men were significantly less likely than women to seek help. Significantly more likely than others to seek help were individuals aged 45–64; those who were widowed, separated, divorced, or never married; those with a lower income; those with less education; those who were black or Hispanic-Latino; and those with two or more general medical conditions.

Enlarge table

Among individuals with lifetime alcohol dependence, men, individuals aged 18–29, those who had never been married, and those who were Asian, Hawaiian, or Pacific Islander were less likely to seek help. Significantly more likely to seek help were those who were widowed, separated, or divorced; who had a lower income or less education; or who were American Indian or Alaskan Native were. Individuals with a general medical condition were significantly more likely to have perceived a need and to have sought help for it.

Mental disorder correlates of help seeking and perceived need

The mental disorder correlates of perceived need for care, help seeking, and no perceived need for care or help seeking are presented in Table 3 . Among individuals with alcohol abuse, those with mania, dysthymia, and major depression were significantly more likely to seek help. Respondents with panic disorder were significantly more likely to have perceived need but were not more likely to seek help. Among the axis II personality disorders, those with antisocial personality disorder were significantly more likely to seek help, and those with histrionic personality disorder were significantly more likely to have perceived a need for help but not sought it. Individuals with a drug use disorder were significantly more likely to seek help for an alcohol use disorder.

Enlarge table

Among respondents with alcohol dependence, those who had mania, dysthymia, major depression, and panic disorder were more likely to have perceived a need for care and to have sought help. Respondents with generalized anxiety disorder were more likely to have perceived a need for care but not to have sought help. Most axis II disorders were associated with an increased likelihood of help seeking. Last, individuals with a drug use disorder were more likely to have both perceived a need for care and to have sought help.

Self-perceived barriers to care

Table 4 shows the prevalence of self-perceived barriers to care among individuals with an alcohol use disorder. Three percent (N=195) of respondents with alcohol abuse had perceived a need for help for an alcohol use disorder but had not sought help, and 18% (N=858) of those with alcohol dependence had perceived a need for help without seeking it. Among individuals with alcohol dependence, men and women differed significantly on several self-perceived barriers. Perceived barriers to care with the highest levels of endorsement were "thought the problem would get better by itself" and "thought I should be strong enough to handle alone." Attitudinal barriers to care were endorsed more frequently than structural barriers to care among respondents with alcohol abuse or dependence.

Enlarge table

Discussion

With this study, we sought to determine the prevalence and correlates of perceived need for care and help seeking among respondents with lifetime alcohol abuse or dependence from a nationally representative sample.

In this study, respondents who perceived a need for care but did not seek help endorsed attitudinal barriers to care more frequently than structural barriers, and this finding is consistent with previous research ( 12 , 13 , 14 , 15 ). Given that attitudinal barriers are consistently reported as the most common barriers to care, public perceptions about alcohol use disorders should be targeted as an area of intervention for increasing rates of help seeking. One study of the effectiveness of a community-level mailing of pamphlets containing information on the effects of alcohol and guidelines for monitoring drinking showed that alcohol abusers reduced their drinking and were more likely to seek help over a one-year period ( 24 ). In this study, we found that men were less likely than women to seek help for alcohol use disorders. As documented in previous studies, men were more likely to have an alcohol use disorder but were less likely to seek help for it ( 25 ). Regular screening for alcohol use disorders, along with changed public perceptions toward help seeking for alcohol use disorders would likely lead to a decrease in the gender gap.

Our findings show that individuals with an alcohol use disorder who were at an advantage in other areas (such as having a higher socioeconomic status, more education, or an absence of any comorbid general medical conditions) were less likely to notice that their drinking was a problem. This finding, along with the finding that 81% of individuals with a lifetime alcohol use disorder did not perceive a need for care or engage in seeking help, suggests that it may be beneficial to regularly screen for alcohol use disorders in primary health care settings. Primary care settings commonly have low rates of screening for alcohol use disorders. However, several studies have documented the benefits of regular screening for alcohol use disorders in these settings ( 26 , 27 , 28 , 29 ). Proper screening in primary health care settings could also lead to interventions geared toward increasing education about levels of healthy drinking, reducing stigma associated wtih help seeking, and providing motivation for seeking help.

Persons with an alcohol use disorder and a comorbid mood disorder or personality disorder had significantly increased odds of perceived need for care and help seeking, even after adjustment for sociodemographic differences. Conversely, those with an alcohol use disorder and a concurrent anxiety disorder did not have increased odds of perceiving a need for care or seeking help. It is not clear why the anxiety disorders did not fit this pattern. Perhaps many individuals with anxiety disorders perceived that alcohol use was an effective way of coping with their anxiety symptoms.

As has been discussed in previous research, individuals with co-occurring disorders tend to experience greater clinical severity ( 8 ), and individuals with an alcohol use disorder and an additional axis I or II mental disorder also have a greater likelihood of experiencing a more severe alcohol use disorder ( 20 ). The severity of an alcohol use disorder also strongly predicts an individual's perceived need for help and motivation to seek help ( 9 ). Individuals with an alcohol use disorder and additional psychopathology have decreased functioning and increased symptoms, which results in alcohol use that is more maladaptive and increases the odds of having perceived need for care or help seeking. Those with an alcohol use disorder who do not have a concurrent mental disorder have decreased odds of perceived need or help seeking and may not have the decrease in functioning necessary to encourage help seeking. Because individuals without a co-occurring disorder are less likely to perceive that care for an alcohol use disorder is needed, regular screening for alcohol use disorders should be broader and be conducted in primary health care settings, not just mental health care settings.

The results of this study must be interpreted in the context of the following limitations. First, the survey questions are retrospective and analyses focused on lifetime alcohol use disorders; therefore, responses are subject to possible recall bias. Second, the NESARC design is cross-sectional, which allows investigation of the prevalence and correlates of help seeking and alcohol use disorders at only one specific point in time. Third, diagnoses were made by lay interviewers and not by clinicians, which may affect the rates of reported mental illness. Fourth, the time frame for incidence of alcohol use disorders and for help seeking and perceived need was lifetime, which does not account for individuals with a lifetime alcohol use disorder who may have recovered without formal treatment at the time of the survey. Past research has demonstrated the rates and correlates of natural recovery from alcohol use disorders within the population ( 21 ).

Conclusions

In summary, only a small percentage of individuals who met criteria for an alcohol use disorder perceived that they needed care. People were more likely to perceive their drinking as problematic if they were older, of a lower socioeconomic status, had less education, had one or more general medical conditions, had never married, were widowed or divorced, or met criteria for some additional axis I or axis II disorder. Regular screening for alcohol use disorders in primary health care settings is recommended to target the large proportion of individuals who do not perceive a need for care.

Acknowledgments and disclosures

Preparation of this article was supported by a graduate scholarship from the Social Sciences and Humanities Research Council Canada, by a New Investigator Award from the Canadian Institute of Health Research, and by the Canada Research Chairs program from the Government of Canada.

The authors report no competing interests.