Table 3. Effects of iAIs in terms of weekly SUs, moderating effects, and subgroup analyses (one- and two-stage results).
| Primary outcome (SUs) and outcome moderators | Comparisons | Persons5 | One-Stage Results | Two-Stage Results | ||||
|---|---|---|---|---|---|---|---|---|
| Effect6 | 95% CI | p-value | Effect6 | 95% CI | p-value | |||
| Overall effect | 27 | 8,095 | b = −5.02 SU | (−7.57, −2.48) | <0.001 | b = −4.80 SU | (−6.99, −2.61) | <0.001 |
| With outlier studies removed1 | 23 | 3,129 | b = −4.81 SU | (−6.69, −2.93) | <0.001 | b = −4.36 SU | (−5.93, −2.78) | <0.001 |
| With studies with AUDIT-only estimates removed2 | 24 | 6,909 | b = −5.74 SU | (−8.55, −2.92) | <0.001 | b = −5.54 SU | (−7.99, −3.09) | <0.001 |
| Participant-level characteristics | ||||||||
| Sociodemographics | ||||||||
| Gender (female) | 27 | 8,095 | b = 2.19 SU | (0.52, 3.85) | 0.010 | b = 1.56 SU | (−0.06, 3.19) | 0.059 |
| Age (55 or older) | 26 | 8,071 | b = −1.62 SU | (−3.72, 0.48) | 0.130 | b = −1.56 SU | (−3.10, −0.02) | 0.047 |
| Education (high) | 18 | 6,018 | b = 2.12 SU | (0.18, 4.07) | 0.033 | b = −0.08 SU | (−2.83, 2.68) | 0.955 |
| Employment (yes) | 16 | 3,179 | b = −0.30 SU | (−2.67, 2.07) | 0.804 | b = −0.51 SU | (−2.93, 1.92) | 0.682 |
| Partner relationship (yes) | 13 | 6,496 | b = −0.51 SU | (−2.30, 1.28) | 0.576 | b = −0.45 SU | (−1.92, 1.01) | 0.544 |
| Drinking profiles | ||||||||
| Regular versus binge-only drinking3 | 27 | 8,095 | b = −0.99 SU | (−3.19, 1.21) | 0.376 | b = −0.15 SU | (−1.67, 1.38) | 0.851 |
| Quantity | ||||||||
| Heavy versus non-heavy drinking4 | 27 | 8,095 | b = −1.50 SU | (−3.35, 0.36) | 0.114 | b = −6.03 SU | (−10.23, −1.84) | 0.005 |
| Intervention-level characteristics | ||||||||
| Therapeutic guidance | ||||||||
| Unguided | 19 | 7,366 | b = −3.23 SU | (−5.88, −0.59) | 0.017 | b = −2.55 SU | (−4.06, −1.04) | 0.001 |
| Guided | 8 | 729 | b = −10.01 SU | (−14.64, −5.39) | <0.001 | b = −10.68 SU | (−16.80, −4.57) | 0.001 |
| Contrast: Guided versus Unguided | b = −6.78 SU | (−12.11, −1.45) | 0.013 | b = −8.14 SU | (−14.44, −1.84) | 0.011 | ||
| Intensity | ||||||||
| Single session | 11 | 3,050 | b = −3.73 SU | (−7.71, 0.24) | 0.066 | b = −2.97 SU | (−5.14, −0.79) | 0.008 |
| Multiple sessions | 16 | 5,045 | b = −5.96 SU | (−9.31, −2.60) | 0.001 | b = −6.07 SU | (−9.55, −2.60) | 0.001 |
| Contrast: Multiple versus Single | b = −2.22 SU | (−7.43, 2.98) | 0.402 | b = −3.11 SU | (−7.21, 1.00) | 0.138 | ||
| Therapeutic orientation | ||||||||
| Integrated | 18 | 5,550 | b = −6.53 SU | (−9.57, −3.49) | <0.001 | b = −6.65 SU | (−9.77, −3.53) | <0.001 |
| PNF only | 9 | 2,545 | b = −1.98 SU | (−6.24, 2.28) | 0.363 | b = −1.71 SU | (−3.68, 0.27) | 0.091 |
| Contrast: PNF versus Integrated | b = 4.55 SU | (−0.68, 9.79) | 0.088 | b = 4.94 SU | (1.25, 8.64) | 0.009 | ||
| Intervention setting | ||||||||
| Work | 3 | 973 | b = −1.88 SU | (−9.52, 5.76) | 0.630 | b = −2.69 SU | (−7.61, 2.24) | 0.285 |
| Healthcare | 7 | 612 | b = −9.12 SU | (−14.33, −3.90) | 0.001 | b = −8.08 SU | (−15.88, −0.28) | 0.042 |
| Community | 17 | 6,510 | b = −4.12 SU | (−7.04, −1.19) | 0.006 | b = −3.50 SU | (−5.29, −1.70) | <0.001 |
| Study-level characteristics | ||||||||
| Type of control | ||||||||
| Waitlist (WLC) | 7 | 647 | b = −11.86 SU | (−16.01, −7.71) | <0.001 | b = −11.57 SU | (−17.83, −5.30) | <0.001 |
| Other (AOC or MIC) | 20 | 7,448 | b = −2.59 SU | (−4.81, −0.38) | 0.022 | b = −2.44 SU | (−3.94, −0.95) | 0.001 |
| Contrast: WLC versus Other | b = −9.27 SU | (−13.97, −4.57) | <0.001 | b = −9.12 SU | (−15.56, −2.68) | 0.005 | ||
| Assessment only (AOC) | 10 | 2,811 | b = −0.96 SU | (−4.05, 2.12) | 0.541 | b = −0.88 SU | (−2.16, 0.39) | 0.175 |
| Waitlist (WLC) | 7 | 647 | b = −11.86 SU | (−16.01, −7.71) | <0.001 | b = −11.57 SU | (−17.83, −5.30) | <0.001 |
| Minimal intervention (MIC) | 10 | 4,637 | b = −4.33 SU | (−7.51, −1.15) | 0.008 | b = −4.98 SU | (−7.88, −2.07) | 0.001 |
| Interactions | ||||||||
| Unguided with WLC | 3 | 307 | b = −8.25 SU | (−14.16, −2.33) | 0.006 | b = −6.59 SU | (−11.57, −1.61) | 0.009 |
| Unguided with Other control | 16 | 7,059 | b = −2.24 SU | (−4.53, 0.07) | 0.057 | b = −2.10 SU | (−3.62, −0.57) | 0.007 |
| Contrast: Unguided—WLC versus Other control | b = −6.01 SU | (−12.36, 0.34) | 0.063 | b = −4.49 SU | (−9.70, 0.71) | 0.091 | ||
| Guided with WLC | 4 | 340 | b = −14.96 SU | (−20.47, −9.44) | <0.001 | b = −14.11 SU | (−23.68, −4.55) | 0.004 |
| Guided with Other control | 4 | 389 | b = −4.45 SU | (−10.06, 1.17) | 0.121 | b = −5.68 SU | (−10.51, −0.84) | 0.021 |
| Contrast: Guided—WLC versus Other control | b = −10.51 SU | (−18.38, −2.64) | 0.009 | b = −8.43 SU | (−19.15, 2.28) | 0.123 | ||
Significant results are shown in bold.
Other control = MIC or AOC.
1Studies 36–39 were regarded as outlier studies.
3Regular drinking denotes 14 or more SUs weekly for females or 21 or more for males (thus excluding binge-only drinking). Binge-only drinking denotes more than 4 or 6 SUs (females/males) on at least one occasion per week while still totalling less than 14/21 SUs weekly.
4Heavy drinking denotes 35 or more SUs weekly for females and 50 or more for males; non-heavy drinking denotes 14/21 SUs or more, but less than 35/50 SUs, weekly for females/males.
5The number of persons refers to respondents for whom data were available on post-intervention drinking behaviour (complete cases) and, for the participant level, also on the moderator in question.
6Unstandardised regression coefficients (b) indicate the effect of the iAIs in terms of alcohol reduction in SUs.
Abbreviations: AOC, assessment-only control; AUDIT, The Alcohol Use Disorders Identification Test; iAI, internet-based alcohol intervention; MIC, minimal-intervention control (e.g., information brochure); PNF, personalised normative feedback; SU, standard units of alcohol; WLC, waitlist control; 95% CI, 95% confidence interval.