From the time of Hippocrates, physicians have recommended bed rest to relieve symptoms and hasten recovery from illness. The association of bed rest with therapy remains deeply ingrained. Recent evidence of potential adverse effects such as deep venous thrombosis, osteoporosis and skin breakdown have led to a reduction in the recommended duration of bed rest, rather than a serious challenge to its medical benefit. Allen and colleagues used techniques commonly applied to other therapies to evaluate the evidence for benefit from bed rest.

They searched MEDLINE and the Cochrane library for studies of the therapeutic effect of bed rest. Further searching was based on references and citations. The initial search produced over 2,000 abstracts, but only 39 randomized controlled trials could be identified. These trials studied the effect of bed rest on more than 5,700 patients being treated for 15 different diseases or conditions. Bed rest following a medical procedure was studied in 24 trials. The procedures included lumbar puncture, spinal anesthesia, cardiac catheterization, liver biopsy and skin grafting. Each procedure had different outcome measures. For seven outcomes, especially headache following lumbar puncture or spinal anesthesia, the results were better with bed rest, but none was statistically significant. Conversely, for the 26 measured outcomes, the results were worse with bed rest, and nine of these achieved statistical significance. Examples of highly significant adverse effects with bed rest were found in hematoma formation and back and leg pain following cardiac catheterization, headache following spinal anesthesia for obstetric procedures and nausea following lumbar puncture.

In 15 trials, bed rest was studied as a primary treatment for conditions such as low back pain, spontaneous labor, uncomplicated myocardial infarction, acute hepatitis and rheumatoid arthritis. Six outcomes showed improvement with bed rest, but none of these reached statistical significance. Overall, 25 outcomes were worse with bed rest, and nine of these achieved statistical significance. Among outcomes that were significantly worse when bed rest was used were time to recovery from acute hepatitis, venous thrombosis after myocardial infarction, length of first stage of labor and need for analgesia and assisted delivery in spontaneous labor, and early disability scores in acute low back pain.

Although isolating the effect of bed rest from other interventions makes studies difficult to conduct and evaluate, the authors conclude that little evidence could be found to support the use of bed rest. Bed rest was found to be non-contributory or even harmful in several studies covering a diverse range of conditions. The authors call on physicians to distinguish between inactivity forced by symptoms and recommendations of bed rest as part of a therapeutic plan. In general, bed rest is not supported by clinical evidence and may be harmful in the few conditions that have been studied. They quote advice given in 1944 to always consider “bed rest as a highly unphysiologic and definitely hazardous form of therapy, to be ordered only for specific indications and discontinued as early as possible. ”