Get access

A Surveillance Tool Using Mobile Phone Short Message Service to Reduce Alcohol Consumption Among Alcohol-Dependent Patients




In public health settings, short message service (SMS) appears to be a promising low-cost modality for reducing alcohol consumption. Here, we test a simple interactive SMS-based helpline with detoxified alcohol-dependent patients to extend findings to curative settings.


This controlled, prospective, 2-group before-after block-assignment, open pilot study tested the feasibility and efficacy of an 8-week outpatient interactive mobile phone SMS intervention (n = 42) against treatment as usual (TAU; n = 38) after inpatient detoxification. Patients were asked whether they needed any help via an automatically generated text message twice a week. A therapist called the individual back when notified. Alcohol consumption was assessed using the telephone version of Form-90 4 and 8 weeks after discharge from inpatient detoxification. The primary end point was defined as attaining low-risk consumption (males ≤30 g or 3.75 units per drinking day (DDD); females ≤20 g or 2.5 units per DDD) 8 weeks after discharge. Missing data were replaced by multiple imputation.


Among all messages sent, 20.5% were followed by a phone call. Feasibility and acceptability were good, as indicated by successful implementation of the SMS procedure and the rapid inclusion of patients. Adherence was satisfactory with 57.14% of the participants replying to at least 50% of the prompts. Patients reported a typical preadmission DDD of 281.25 ± 244.61 g. In the SMS group, 55.7% of 42 patients, and 40% of 38 patients in the TAU group, achieved low-risk consumption (risk diff: 0.16; 95% CI −0.06 to 0.37; p = 0.122).


In detoxified alcohol-dependent patients, relapse prevention based on SMS was well received and implemented efficiently and rapidly. An adequately powered multicenter study is currently being conducted to test the nonsignificant but encouraging findings of this exploratory study with more rigorous trial methods (ISRCTN78350716).