Background: This study evaluates the outcomes of an Internet-at-home intervention experiment that intended to decrease loneliness among chronically ill and physically handicapped older adults through introducing them to the use of an electronic communication facility. Method: To determine the effectiveness of the experiment in terms of reducing loneliness, 15 older adults were interviewed three times: shortly before the start, two years later and immediately after termination of the experiment, while their loneliness scores at zero and post-measurement were compared with those of a control group. Results: Both the participants and the control persons experienced a reduction in loneliness over time. However, the reduction was only significant for the intervention participants. Moreover, the changes in loneliness were significantly greater among the participants compared to the control persons. When looking more in detail, the effect of the experiment was only significant regarding emotional loneliness and among the highest educated. Findings of the qualitative research enabled us to understand the mechanisms through which the intervention helped alleviate loneliness. E-mail was found to facilitate social contact. Furthermore, the computer and Internet were often used to pass the time, taking people's minds off their loneliness. Unexpectedly, the intervention also improved people's self-confidence. Conclusion: The decline in loneliness is likely to be greater if persons under more favorable circumstances are selected and if more social functions of the Internet are used.