Study designs of evaluations included in the review
Randomised controlled trials (RCTs), controlled clinical trials and uncontrolled studies were eligible for inclusion.
Specific interventions included in the review
Studies examining the effect of medicinal selenium supplementation above the normal dietary intake on potentially environment-associated symptoms and syndromes were eligible. Both monopreparations and complex preparations of selenium were eligible. Supplementary treatment for extreme selenium deficiency-related conditions was excluded. No dosage limit was set. However, the authors stated that dosages between 30 and 200 microg/day were considered to be therapeutically effective, with toxic or adverse effects likely with dosages exceeding 400 microg/day. Three of the included studies appeared to exceed these recommended therapeutic dosages. The majority of the included studies were of complex preparations where selenium was one component.
Participants included in the review
Individuals demonstrating clinical characteristics similar to potentially environment-associated health disorders were eligible. Potential environment-associated syndromes included: idiopathic chronic fatigue, chronic fatigue syndrome, functional memory disorder, multiple chemical sensitivity syndrome, and mucosal irritations and disorders of the eyes and respiratory tract. Confirmed exposure to contaminants was not a prerequisite.
Individuals with affective disorders, defined according to the American Psychiatric Association's DSM-IV criteria, were included in the review, as were studies of healthy or elderly persons where the aim was prevention or treatment. Individuals with psychosomatic disorders were included only where symptoms corresponded to a clinical profile for potentially environment-associated disorders. Patients with psychotic disorders, major depressive disorders with psychotic, catatonic or melancholic features, and bipolar affective disorders were excluded, as were studies investigating the effect of selenium supplementation on the cognitive development of children. The patients in the included studies were from a variety of settings, including: out-patient departments, environmental medicine out-patient departments, medical practices, in-patient settings and nursing homes.
Outcomes assessed in the review
Studies with clinical outcome measures or surrogate parameters were eligible, although these were not specified. The outcomes in the primary studies included: muscular complaints, measures of cognition and affect(including the Hamilton Depression Scale), anxiety, infection rates and antibiotic consumption, clinical deterioration, days absent from school due to illness, and respiratory tract infection rates.
How were decisions on the relevance of primary studies made?
Two reviewers independently assessed the primary studies; any disagreements were resolved by consensus.