103 



Epidemiologic Bulletin No. 27 



syndrome (i.e., a cluster of symptoms), at 

 least two symptoms of wtihdrawal wei^ 

 required to meet the duration critenon. It 

 should be noted, however, thai the pres- 

 ence of withdrawal is not required for a 

 DSM-III-R diagnosis of alcohol depen- 

 dence. Second, any reported symptom of 

 tolerance satisfied the duration criterion 

 for alcohol dependence, because toler- 

 ance constitutes a less acute and episodic 

 symptom of dependence relative lo the 

 other dependence criiena. 



Table 1 presents the 1 -year prevalence 

 rales, standard errors, and population es- 

 timates of DSM-lll-R alcohol abuse and 

 dependence by age. sex. and ethnicity. 

 The DSM-lll-R alcohol abuse and de- 

 pendence groups formed from the 1988 

 NHIS were mutually exclusive: that is, 

 respondents classil'ied as alcohol abusers 

 did nni meet criteria for alcohol depen- 

 dence, while those who met criteria for 

 jIcdIioI dependence were classified as lo 



tiKtnchrr F. Chanj. Ph.D.. Ph.D.. is chit/ 

 oj ilw Bioiiii'fry Branch. Divixiwi nf 

 Bionicfty ami EpuU'n)iohi\'\\ National 

 hixiiniic an A/cahttI Ahuxc anti 

 AUahaUsui (NiAAA). RackvUle. 

 Manfand 



TiiiiMAsC. H.\HH>Kn. Ph.D.. i\ director 

 oflhc Divi.sam itj Bamwiry and 

 Epidcnnolafiy. NIAAA. 



pATKtciA Chov. Ph.D.. /.? a .wnry .T/n/f.T- 

 tician in ihc Biometry Branch. Division 

 of Biometry and Epidemiology. NIAAA 



ROCKK PtChEftiNC. M.S.. IS a computer 

 programmer nt the Biometry Branch. 

 Division of Biometry' and Epidemiology. 

 NIAAA. 



Deborah A. Dawson. Ph.D.. is a survev 

 statistician m the Biometry Branch. 

 Division of Biometry and Epidemiology. 

 NIAAA. 



FREDERtCK S. Stinson. Ph.D.. IS project 

 manager for the Alcohol Epidemiologic 

 Data System, operated by CSR. 

 Incorporated. Washington. DC. under 

 contract to NIAAA. 



John Noble, B.A.. is deputy director of 

 the Division of Biometry and 

 Epidemiology. NIAAA. 



18-29 



30^W 45-64 



Age Group in Years 



65 arxl older 



I White male 

 n White female 



Nonwhite male 

 Nonwhite female 



Rgure 1 Prevalence of DSM-lil-R alcohol abuse and dependence by age. sex. and 

 ethnicity: United States. 19B8. 



whether ihey also met the criteria for al- 

 cohol abuse 



The I -year prevalence for total alco- 

 hol abuse and dependence in the NHIS 

 .sample was 8. 63 percent, representing 

 15.295,000 Americans (Table 1) More 

 respondents were classified as alcohol 

 dependent (6.25 percent) than alcohol 

 abusing (2.38 percent). Among those re- 

 spondents meeting the DSM-III-R diag- 

 nostic crilena for dependence, an 

 overwhelming proportion also met the 

 cnlena for alcohol abuse Tlic dominance 

 of the dual abuse-dependence diagnosis 

 was generally consistent across each 

 age. sex. and ethnic subgroup of the 

 population. 



One-year prevalence was even more 

 dramatic for males, with 13.35 percent 

 meetmg cntena for alcohol abuse or de- 

 pendence. The prevalence rate for fe- 

 males was much lower, 4.36 percent. 

 Prevalence also was greater among 

 whites (9 13 percent) than among non- 

 whites (5.60 percent) (Table I). Rales for 

 white males and while females exceeded 

 the rates for their nonwhite counterparts 

 by 33 6 percent and 46,6 percent, 

 respectively. 



Prevalence rales were higher among 

 males and females under the age of 45 

 years than among those aged 45 and old- 



er (Table 1 ). For males, the 1-year preva- 

 lence rate in the youngest age group (18 

 lo 29 years) was 23.50 percent The rale 

 fell lo 14.30 percent among 30- to 44- 

 ycar-olds. and was only 2.77 percent in 

 those aged 65 and older Among women, 

 the highest prevalence rale was also 

 found in the youngest age group ( 10.10 

 percent), and the rates fell steadily to 

 0.37 percent in women aged 65 and old- 

 er Possible reasons for the decline in al- 

 cohol abuse and dependence rates with 

 age mclude the artifact of recall, lower 

 survival rates among alcoholics, and re- 

 sponse styles. Alleraatively. the age gra- 

 dient may reflect a true cohort effect; that 

 is, that alcohol abuse and dependence is 

 more prevalent among the more recent 

 generations of Americans. 



Ethnic groups (i.e.. whites and non- 

 whites) showed striking patterns of age- 

 related I -year prevalence rales of alcohol 

 abuse and dependence (Figure 1). As 

 shown in Table 1, among the youngest 

 males, the prevalence rate in whites 

 (26 14) was 2.5 times greater than in 

 nonwhites (10.50). In the next age group 

 (30-44 years), the rates for whites and 

 nonwhites were more similar, with a 

 slight predominance in whiles. This pre- 

 dommance in whites disappeared in the 

 45-M age group, m which rates of alco- 



Voi.. 15. No. 1.1991 



95 



