General Practitioners Recognizing Alcohol Dependence: A Large Cross-Sectional Study in 6 European Countries

Jürgen Rehm; Allaman Allamani; Roberto Della Vedova; Zsuzsanna Elekes; Andrzej Jakubczyk; Inga Landsmane; Jakob Manthey; José Moreno-España; Lars Pieper; Charlotte Probst; Sigita Snikere; Pierluigi Struzzo; Fabio Voller; Hans-Ulrich Wittchen; Antoni Gual; Marcin Wojnar

doi:10.1370/afm.1742

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
Venn diagram of alcohol dependence determined by a general practitioner's assessment or by the CIDI based on self-report.

CIDI = Composite International Diagnostic Interview; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th Edition); GP = general practitioner.

View popup

Table 1

Characteristics of Noninterviewed and Interviewed Patients With a Diagnosis of Alcohol Dependence, by Assessment and Sex

	Noninterviewed (n = 2,012 Men; n = 2,515 Women)		Interviewed^a (n = 3,449 Men; n = 5,027 Women)

	Diagnosis by GP		Diagnosis by GP Only		Diagnosis by GP and CIDI		Diagnosis by CIDI Only

Characteristic	Male	Female	Male	Female	Male	Female	Male	Female	P Value^b	P Value^c
12-mo prevalence, % (95% CI)	6.1 (5.1–7.2)	2.0 (1.4–2.5)	5.6 (4.8–6.3)	1.9 (1.5–2.2)	3.1 (2.5–3.7)	0.6 (0.4–0.8)	5.9 (5.1–6.7)	2.4 (1.9–2.8)	—	—
Age, mean (SD), y	51.0 (10.1)	50.1 (10.2)	50.9 (10.9)	50.1 (12.1)	49.8 (11.7)	48.8 (12.1)	38.3 (13.2)	38.2 (13.5)	<.0001^d	<.0001^d
SES, self-classified, % (95% CI)
Above average	NA	NA	5.3 (2.1–8.4)	7.9 (2.6–13.3)	10.8 (4.7–16.8)	10.5 (0.0–21.8)	9.7 (5.6–13.8)	12.2 (6.3–18.1)	—	—
Average	NA	NA	58.8 (52.1–65.5)	50.6 (40.9–60.2)	48.5 (39.3–57.7)	38.5 (21.4–55.6)	61.0 (54.3–67.8)	66.6 (58.1–75.1)	—	—
Below average	NA	NA	36.0 (29.4–42.5)	41.5 (31.9–51.1)	40.7 (31.7–49.8)	51.0 (33.0–68.9)	29.3 (23.0–35.6)	21.2 (13.9–28.6)	.0219	.0033
Unemployed for health or other reason, % (95% CI)	NA	NA	27.8 (21.9–33.8)	36.6 (27.4–45.8)	34.0 (25.4–42.7)	38.9 (21.4–56.4)	15.6 (10.5–20.6)	17.3 (10.5–24.1)	.0021^e	<.0001^d
Smoking (GP and self), % (95% CI)	70.0 (61.3–78.7)	68.1 (54.5–81.7)	59.9 (53.3–66.5)	54.3 (44.6–63.9)	69.1 (60.5–77.8)	76.9 (61.6–92.2)	58.8 (51.9–65.6)	47.4 (38.3–56.4)	.0010^e	.0546
BMI (GP), mean (SD)	26.2 (5.0)	26.2 (5.9)	27.2 (6.7)	27.0 (5.8)	25.8 (4.8)	24.7 (6.3)	26.0 (4.5)	23.4 (4.8)	.0072	.0697
Hypertension (GP), % (95% CI)	34.2 (25.6–42.7)	54.0 (39.9–68.1)	45.2 (38.6–51.9)	45.0 (35.4–54.6)	43.7 (34.7–52.8)	54.4 (36.5–72.2)	19.1 (13.6–24.6)	12.0 (6.3–17.7)	.0623	.0005^d
Liver problems (GP), % (95% CI)	25.8 (17.9–33.8)	24.0 (12.0–36.0)	19.5 (14.1–24.8)	20.4 (12.4–28.5)	32.2 (23.6–40.8)	20.0 (6.0–33.9)	1.6 (0.0–3.4)	0.8 (0.0–2.2)	<.0001^d	<.0001^d
Depression (GP), % (95% CI)	15.8 (9.2–22.4)	22.0 (10.3–33.7)	10.9 (6.7–15.1)	18.6 (10.8–26.4)	27.3 (19.2–35.5)	19.9 (5.4–34.4)	7.2 (3.6–10.7)	14.2 (7.6–20.8)	.0002^d	.7108
Anxiety (GP), % (95% CI)	24.2 (16.4–31.9)	40.0 (26.2–53.8)	16.8 (11.9–21.8)	20.3 (12.7–27.8)	32.0 (23.4–40.6)	33.1 (16.2–50.0)	10.2 (6.0–14.4)	19.0 (11.7–26.4)	<.0001^d	.4873
K10^f
Above cutoff for serious mental distress, % (95% CI)	NA	NA	8.0 (4.5–11.6)	8.6 (3.2–14.1)	21.5 (14.0–29.0)	21.5 (7.0–36.1)	8.6 (4.7–12.6)	16.3 (9.5–23.2)	.0002^d	.0569
Total score, mean (SD)	NA	NA	8.3 (8.1)	10.3 (8.6)	14.0 (8.6)	14.3 (9.0)	9.4 (7.6)	11.9 (9.0)	<.0001^d	.0084
WHODAS 2.0, mean (SD)^g
Total score	NA	NA	11.8 (16.0)	14.0 (14.7)	19.6 (19.1)	19.3 (18.5)	12.3 (15.3)	12.4 (14.9)	<.0001^d	.2727
Days unable to conduct usual activities or work because of health condition	NA	NA	2.3 (7.3)	2.0 (6.1)	4.8 (9.0)	2.4 (3.4)	2.1 (5.8)	2.3 (6.1)	.0025^e	.8362
Ethanol use by drinkers, mean (SD), g/d^h	NA	NA	50.3 (38.8)	46.9 (36.1)	87.6 (69.9)	62.7 (58.6)	52.6 (50.1)	42.2 (50.3)	<.0001^d	.5451
Chronic heavy drinking (≥100 g/d), % (95% CI)^h	NA	NA	9.3 (4.2–14.5)	10.8 (0.0–22.5)	32.4 (22.5–42.3)	22.2 (4.9–39.4)	13.7 (7.7–19.7)	8.8 (1.4–16.2)	<.0001^d	.3814
Binge drinking (≥200 g/wk), % (95% CI)^h	NA	NA	8.5 (3.4–13.6)	15.0 (1.3–28.8)	28.8 (19.2–38.4)	21.3 (4.5–38.1)	17.1 (10.5–23.7)	7.2 (0.3–14.0)	<.0001^d	.8955

BMI = body-mass-index; CIDI = Composite International Diagnostic Interview; GP = general practitioner; K10 = Kessler Psychological Distress Scale; NA = not applicable because this information has not been assessed by general practitioner; SES = socioeconomic status; WHODAS 2.0 = World Health Organization Disability Assessment Schedule 2.

Notes: Confidence intervals are based on standard error. Data were derived from different sources: GP indicates that the general practitioner has provided the answer, whereas self indicates the information derives from the interviewed patient him/herself. GP and self means that both sources of information were used for the respective sample, ie, the GP response is displayed for the sample assessed by the general practitioner and the self response is displayed for the interviewed sample.
↵a Data were weighted with inverse sampling probabilities.
↵b Compares group means of patients diagnosed by GP and CIDI with patients diagnosed either by CIDI or GP only; controlled for age and sex.
↵c Compares group means of patients diagnosed by GP only with patients diagnosed by CIDI only; controlled for age and sex.
↵d P values significant for Bonferoni-adjusted thresholds: P <.01/16 = .000625.
↵e P values significant for Bonferoni-adjusted thresholds: P <.05/16 = .003125.
↵f K10 cutoff for severe mental distress was 21 points in a total score ranging from 0 to 40, where higher scores indicate more severe distress.
↵g WHODAS 2.0 total score ranges from 0 to 100, where higher scores indicate higher level of disability.
↵h Analyses were computed on patients reporting ethanol use of at least 10 g/d (remaining patients with alcohol dependence were abstaining).

Additional Files

Figures
Tables

Supplemental Appendixes 1-4

Files in this Data Supplement:

Supplemental data: Appendixes 1-4 - PDF file, 9 pages + cover
The Article in Brief

General practitioners recognizing alcohol dependence: a large cross-sectional study in 6 European countries

Jakob Manthey , and colleagues

Background Alcohol consumption is a major risk factor for death and disability. This study examines the rate of identification of patients with alcohol dependence in 6 European countries using 2 different methods: identification by general practitioners and identification by semistructured interview using the Composite International Diagnostic Interview (CIDI) tool for assessing alcohol use disorders.

What This Study Found The 2 methods discover about the same number of alcohol dependent people, but there is little overlap between the people identified. The 12-month prevalence of alcohol dependence was 5.1 percent when assessed by the physician and 5.5 percent when assessed by the CIDI interview. Although physician assessment and the CIDI yielded a similar prevalence, they identified different patient populations, with fewer than one-fifth of the cases identified by both methods. Both methods identified alcohol dependence in 8.7 percent of patients, confirming that in Europe, alcohol dependence is common and disabling among primary care patients. Compared with the CIDI, general practitioners identified more patients with severe alcohol dependence. The CIDI was not as successful as a physician in identifying alcohol dependence in older patents.

Implications

These significant differences, the authors conclude, raise questions about the validity of the CIDI and its status as a reference standard for assessing alcohol use disorders.