Remarks y The terms “alcohol use disorder” and “hazardous alcohol use” refer to diagnosis according to the ICD-10 or DSM-IV criteria or equivalent ICD-11 and DSM-5 diagnosis. y The included digital interventions encompass both unguided and guided digital interventions. In unguided digital interventions, psychoeducation and psychotherapeutic techniques are provided for the individual to self-manage their symptoms without the help of a health professional. In guided digital interventions, additional guidance is provided from health professionals who can assist participants with technical or health-related questions via chat, email or telephone. y The studies assessing alcohol reduction via digital interventions compared with active comparators usually recruited individuals from specialized treatment facilities. The majority of those interventions combined the digital component with face-to-face treatments, such as TAU or CBT, and last 8–12 weeks. y Differences in findings between active/non-active comparators should be interpreted with caution. For the comparisons involving non-active comparators, the majority of individuals were recruited based on self-reported use patterns and not assessed for an alcohol use disorder. This is contrary to the studies involving active comparators, which recruited participants after the diagnosis of an alcohol use disorder. For this reason, it is important to stress that different findings for active/non-active comparators are likely due to the different characteristics (such as severity) of the target group and the intensity of the treatment provided. y Digital interventions should not be used to replace or detract from provision of other forms of interventions. y There are concerns regarding potentially sensitive content and data privacy while using digitally delivered health interventions. Measures should be taken to ensure that digital interventions are provided under conditions of safety/security, confidentiality, informed consent and privacy of data. This can include the establishment of standard operating procedures that describe protocols for ensuring consent, data protection and storage, and verifying provider licensing and credentials. Further guidance can be found in the 2019 WHO guideline: recommendations on digital interventions for health system strengthening (31).

Implementation considerations y Digitally delivered interventions can provide benefits to people with an alcohol use disorder and those with hazardous alcohol use, especially when provided in addition to TAU. Digitally delivered interventions should not substitute provision of other types of conventional treatment (psychosocial or pharmacological) to people with alcohol use disorders. When face-to-face treatment is not available or acceptable, then self-help digital interventions can be a feasible option to provide support. y Setting up and sustaining digital health solutions can be costly, while costs for individual users are usually not very high, making implementation feasible for the end user. y There is a need to consider the potential digital divide across population groups with some having unequal access to and skills to use digital technologies. Access might be particularly difficult for certain population groups with poor access to network services, mobile devices or electricity, and/or with low literacy and digital literacy skills. Measures should be taken to address inequities in access to mobile devices so that further inequity is not perpetuated in accessing health information and services, including mechanisms to ensure individuals who do not have access to mobile devices can still receive appropriate services. y Country adaptation and translation of digital interventions tools with subsequent evaluation is essential.