Sweatin’ with the Oldies

Abstract & Commentary

Synopsis: Night sweats, day sweats, and hot flashes occur frequently in the elderly and are more closely associated with common ailments than serious illnesses.

Source: Mold JW, et al. Ann Fam Med. 2004;2:391-397.

Night sweats, day sweats, and hot flashes are commonly reported symptoms in primary care. Mold and colleagues previously examined the prevalence of these symptoms in an adult population that included perimenopausal women.1 Since hot flashes and night sweats are closely associated with menopause, they studied a population 64 years and older.

Physicians in the Oklahoma Physicians Resources/Research Network compiled lists of active, elderly patients, totaling close to 4000. After excluding patients who had died, were in nursing homes, were too confused to provide reliable information, had switched physicians, could not be reached by telephone, declined to participate, or could not understand telephone instructions, 799 were left to enroll. Four more were excluded because they did not respond to the night sweats question. The study population was 57% female, 88% white, and highly educated (59% with college or graduate degree). Cigarette use and excessive alcohol use were low (7.5% and 6.3%, respectively).

Night sweats, day sweats, and hot flashes were present in 10.3%, 8.7%, and 7.8%, respectively. Some participants reported more than one symptom; overall, 17.7% (140) had at least one symptom. In multivariate analysis (P < 0.01), night sweats were associated with age, fever, muscle cramps, visual problems, and numbness in hands and feet. Day sweats were associated with fever, restless legs, lightheadedness, and diabetes. Hot flashes were associated with non-white race, fever, nervous spells, and bone pain. In univariate analysis, there was no association with any of the three symptoms and autoimmune diseases, sarcoidosis, tuberculosis, chronic hepatitis, cancer, or thyroid disease. Individual symptoms were grouped into subgroups (for instance, impaired vision, impaired hearing, impaired smell, or numbness of hands or feet into "sensory deficit"). Subjects with night sweats, day sweats, or hot flashes were more likely to have a sensory deficit, bodily pain, visceral pain, or a mood disturbance than those without the three cardinal symptoms. They also scored lower on various quality-of-life scales.

Comment by Allan J. Wilke, MD

One problem with this study is that 80% of the original group were excluded for one reason or another. Mold et al state that the study participants were younger, better educated, more likely male, and had better self-rated health than those people who were excluded. This raises the question of selection bias. Mold et al believe that, if anything, this would tend to underestimate the true prevalence of the primary symptoms. Another potential bias is that the illnesses were all self-reported and not confirmed by chart review.

The most striking finding is the lack of association with any of the "serious" illnesses (autoimmune diseases, TB, cancer, sarcoidosis, hyperthyroidism, etc.) that we were taught to look for. This is primarily the result of conducting the study in a primary care setting.

This raises an interesting dilemma. If one of your elderly patients complains of night sweats, day sweats, or hot flashes, how far and in what direction do you take your investigation? While keeping in mind the classical, but rare, illnesses associated with these symptoms, you would do well to look for the other, more common, illnesses outlined above.

Dr. Wilke, Associate Professor of Family Medicine, Medical College of Ohio, Toledo, OH, is Associate Editor of Internal Medicine Alert.


1. Mold JW, et al. J Fam Pract. 2002;51:452-456.