Table 2 Descriptive statistics of the reasons for reducing alcohol intake.

From: Genome-wide analyses of behavioural traits are subject to bias by misreports and longitudinal changes

Intake level reported

Reason

N

Mean phenotype

AC

BMI

Disease count

EA

Yes

Illness or ill health

8555 (5.6%)

7.33 (0.10)

28.53 (0.06)

2.77 (0.02)

12.81 (0.06)

Doctor’s advice

3331 (2.2%)

14.72 (0.23)

29.49 (0.09)

2.58 (0.03)

12.65 (0.09)

Health precaution

48,483 (31.7%)

9.98 (0.04)

27.44 (0.02)

1.54 (0.01)

14.17 (0.02)

Financial reasons

7323 (4.8%)

10.27 (0.12)

28.54 (0.06)

1.74 (0.02)

11.57 (0.06)

Other reason

68,066 (44.6%)

7.54 (0.03)

27.71 (0.02)

1.28 (0.01)

13.90 (0.02)

Prefer not to answer

268 (0.2%)

5.77 (0.45)

27.69 (0.30)

1.75 (0.10)

10.90 (0.30)

Do not know

16,729 (11.0%)

7.28 (0.06)

27.96 (0.04)

1.35 (0.01)

13.00 (0.04)

No

Illness or ill health

3541 (9.1%)

—

29.28 (0.10)

2.81 (0.03)

12.35 (0.09)

Doctor’s advice

462 (1.2%)

—

29.99 (0.26)

2.77 (0.09)

11.90 (0.24)

Health precaution

6767 (17.5%)

—

28.26 (0.06)

1.87 (0.02)

12.86 (0.06)

Financial reasons

1600 (4.1%)

—

29.05 (0.14)

1.91 (0.05)

11.26 (0.12)

Other reason

21,362 (55.1%)

—

28.62 (0.04)

1.51 (0.01)

12.88 (0.03)

Prefer not to answer

101 (0.3%)

—

28.23 (0.53)

2.10 (0.17)

9.56 (0.41)

Do not know

4934 (12.7%)

—

28.79 (0.08)

1.53 (0.02)

11.89 (0.07)

  1. The proportion in the bracket in N column indicates the proportion of each reason within two categories (reported intake level or not). The numbers in the bracket in Mean phenotype column are standard error of each mean estimate. N, sample size; AC, alcohol consumption (units per week); BMI, body mass index (kg/m2); Disease count, number of common diseases affected; EA, educational attainment (years of schooling). A total of 132 individuals did not have records for their reasons for reducing their alcohol intake.