Most people determine their weight using a scale at a physician's office, fitness center, or weight-loss center as scales found at these sites are generally considered to be more precise than their bathroom scales at home.
Although the National Institute of Standards and Technology (NIST) has established acceptable tolerance levels for Legal-for-Trade scales used for commercial purposes (diet centers offering pay-by-weight weight-loss programs), no standards have been established by NIST or the Joint Commission on Accreditation of Hospitals for scales used clinically in the care of patients. Despite the lack of scale regulation for patient care, a survey of physicians suggested, for clinical purposes, scales should be precise to 1 lb per 150 lb of body weight.
One study that examined scales used by patients at a hospital concluded that many were "too inaccurate for optimal patient care."
Of the 97 scales examined in the hospital, 62 percent would have failed to meet the NIST criteria, and 22 percent would have failed to meet the physician-imposed standards.
Another study was conducted to compare the precision of scales in physicians' offices, fitness centers and weight-loss centers. Assessments were completed on a total of 223 scales: 94 from primary care clinics, 32 from diabetology/endocrinology clinics, 39 from weight-loss centers, and 58 from fitness centers. Scales were assessed for condition, location in facility, resting surface, and calibration history.
Relatively large inaccuracies were found at all weight levels and at all locations in the sites. For example, at one primary care office using a 250 lb test weight, a 17.5 lb imprecision was noted. Scales found in women's locker rooms provided the only consistent under-estimations of weight.
A significant proportion of the scales were highly imprecise, particularly at higher weight levels. For example, at the 250 lb test weight, nearly 21 percent of the scales were off by more than 6 lb, or approximately 1 body mass index unit. Even for the 100 lb test, more than 25 percent of the measurements were off by 2 lb or more.
Scale precision was significantly related to scale location. Scales located in men's locker rooms (typically in fitness centers) were more inexact than scales in other locations. The resting surface of the scale was significantly related to the precision of the scale. Tile surfaces produced the greatest imprecision, while carpeted surfaces produced the least measurement error, on average.
Not surprisingly, calibration of the scale in the past year resulted in scales with significantly more precise readings. Finally, weight-loss centers with a "pay-by-weight" distinction produced readings that were significantly more precise than other scales.
Physicians who are unaware of this source of variability in weight assessment would most likely assume that a patient's weight had changed significantly. If physicians were to assume, as in the case of heart failure, that the weight change was due to acute changes in volume status, diuretics might be inappropriately adjusted, potentially resulting in patient harm. Similar scenarios are possible in a range of settings and situations including obstetrics, pediatrics, and renal failure-all because patients were weighed on different scales on different occasions and physicians did not take into account this variability.
