Are subthreshold alcohol dependence symptoms a risk factor for developing DSM-IV alcohol use disorders? A three-year prospective study of ‘diagnostic orphans’ in a national sample

Abstract

Aims

Research suggests that diagnostic orphans (i.e., individuals experiencing only 1–2 criteria for DSM-IV alcohol dependence) may be at increased risk for developing more severe alcohol problems. This study aimed to: (i) investigate the course of diagnostic orphans in the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), and (ii) explore whether a specific symptom endorsement pattern(s) could identify diagnostic orphans at Wave 1 who remitted or progressed to alcohol dependence at Wave 2.

Methods

Current drinkers (n = 15,751) were divided into diagnostic groups at Waves 1 and 2: no-alcohol use disorder (AUD); one-criterion orphans, two-criterion orphans, alcohol abuse, or alcohol dependence. Multinomial logistic regression analysis explored the association between diagnostic status at Wave 1 and Wave 2. Chi-square tests investigated differences in the criteria endorsement patterns of diagnostic orphans.

Results

Compared to the no-AUD group, one-criterion orphans at Wave 1 were twice as likely to be in the abuse group and four times more likely to be dependent at Wave 2. Two-criterion orphans were three times more likely to be in the abuse group and eight times more likely to have progressed to dependence. Criterion endorsement patterns of diagnostic orphans at baseline did not significantly differentiate between those who remitted and those who progressed to dependence at follow-up.

Conclusions

Like previous research, diagnostic orphans are at increased for developing to more severe alcohol problems. Relying solely on the DSM-IV AUD diagnostic criteria, however, may not be sufficient to identify those diagnostic orphans who are at risk for progressing to dependence.

Introduction

The DSM-IV (American Psychiatric Association, 1994) diagnostic criteria for alcohol abuse and dependence isolate a group of individuals who do not meet the diagnostic criteria for either disorder but who experience one or two diagnostic criteria for alcohol dependence – the so-called ‘diagnostic orphans’ (Hasin & Paykin, 1998). Diagnostic orphans have been identified in adolescent and young adult samples (Pollock & Martin, 1999, Rohde et al., 2001, Chung et al., 2002, Eng et al., 2003, Wells et al., 2006, Schuckit et al., 2008), in genetic, clinical and community studies (Olfson et al., 1996, Hasin & Paykin, 1999, Ray et al., 2008); Sarr et al., 2000, Shankman et al., 2008), and in the general population (Hasin & Paykin, 1999, Harford et al., 2005, Harford & Yi, 2008, McBride et al., 2009a, McBride et al., 2009b). It is estimated that between 10.9% and 14.3% of all alcohol users in the general population meet the requirements for diagnostic orphan status (Hasin & Paykin, 1999, Harford & Yi, 2008, McBride et al., 2009b).

The identification of diagnostic orphans has been met with a degree of scepticism in the literature. It has been suggested that the symptoms experienced by this group of alcohol users are too mild to warrant clinical attention or diagnosis (cf. Narrow et al., 2002, Hoffmann & Hoffmann, 2003). Findings from a growing body of cross-sectional studies, however, have revealed that compared to individuals without an alcohol use disorder (AUD), diagnostic orphans are more likely to frequently engage in hazardous drinking patterns (Hasin & Paykin, 1998, McBride et al., 2009b), are at increased risk for experiencing other mental disorders including drug use disorders (Ray et al., 2008, McBride et al., 2009a) and report moderate levels of functional impairment (Olfson et al., 1996, McBride et al., 2009c). Recent research has also demonstrated that diagnostic orphans are at greater risk for experiencing more adverse life events and poorer physical health compared to those with alcohol abuse (McBride et al., 2009b, McBride et al., 2009c).

Although it has been proposed that diagnostic orphan status is a transient phenomenon (Hasin & Paykin, 1998), findings from longitudinal studies suggest that this sub-threshold condition has the potential to be persistent and increases the likelihood for the transition to AUD. For example, Schuckit et al. (2008) explored the five-year course of DSM-IV AUD in a sample of 12–19 years olds (n = 616), who were offspring of individuals involved in the Collaborative Study on the Genetics of Alcoholism (COGA; Begleiter et al., 1995). Alcohol and drug use intake and problems were highest for the alcohol abuse and dependence groups at the follow-up; however, diagnostic orphans were significantly more likely to experience higher rates of conduct disorder and alcohol or drug use compared to the no-AUD group. Regression analysis revealed that diagnostic orphans at baseline were almost twice as likely to meet the criteria for an AUD at the follow-up, compared to those with no-AUD. No clear outcome pattern for diagnostic orphans emerged at the follow-up, with equal percentages (23.5%) remaining in the same category or returning to no-AUD, respectively. Approximately 38.2% of diagnostic orphans progressed to alcohol abuse and 14.8% developed alcohol dependence. In another five-year study of participants involved in the San Diego Prospective Study (Schuckit & Smith, 1996), Eng et al. (2003) demonstrated that 17% of diagnostic orphans at the baseline assessment progressed to either alcohol abuse or dependence at the follow-up, compared to only 3.1% of the baseline no-AUD group. Recent analysis of data from the National Longitudinal Survey of Youth (NLYS) also revealed that although 61.7% of diagnostic orphans experiencing one dependence criterion at baseline returned to no-AUD at the follow-up, 14.8% had developed an AUD (Harford & Yi, 2008). Only half of the diagnostic orphans experiencing two dependence criteria were categorized as no-AUD five years later (48.7%) whereas 27.4% had progressed to an AUD.

The weight of evidence from the aforementioned studies suggests that this substantial group of alcohol users may not be experiencing temporary symptoms of alcohol dependence and may be at increased risk for progressing to more serious alcohol problems. Analysis of longitudinal data from the general population would strengthen this existing body of research and would be particularly useful for determining how diagnostic orphans might best be handled in the future (cf. Cottler & Grant, 2006, Schuckit & Saunders, 2006). The current study, which analyses data from the longitudinal National Epidemiologic Survey on Alcohol and Related Conditions (NESARC; Grant & Kaplan, 2005), was devised to address two main aims: (i) explore the three-year course of diagnostic orphans in the general population, and (ii) investigate whether a specific alcohol-related symptom, or patterns of symptoms, can help predict which diagnostic orphans may be at increased risk for progressing to DSM-IV alcohol dependence.