Accuracy of Physical Activity Monitors in Pregnant Women

Abstract

No physical activity (PA) monitor has been validated in pregnant women. PA monitors may assist pregnant women in the adoption and maintenance of a PA program. PURPOSE: To determine the accuracy of three pedometers and one accelerometer in pregnant women during treadmill walking. METHODS: Subjects were 18 women in the second or third trimester (20-36 weeks) who were screened for pregnancy-related risk factors by the nursing staff at an obstetrics office. Only those free of all risk factors were eligible to participate. Each subject was fitted with a belt below the abdomen containing three PA monitors: Yamax Digiwalker SW-200 (DW), New Lifestyles NL-2000 (NL), and GT3X Actigraph Accelerometer (ACT). The Omron HJ-112 (HJ) was placed in the pants pocket. Subjects walked at 54, 67, 80, and 94 m/min for two minutes each. Actual steps for each speed were hand-tallied and compared with steps recorded by each PA monitor. Percentage of actual steps was calculated for each device at each speed. Data were analyzed using repeated measures ANOVA. Paired t-tests and Bonferroni adjustment were used to further explore interactions. RESULTS: There were significant effects of speed and device as well as a significant interaction. At 54 m/min, the PA monitors were significantly less accurate (80.0+34.3%) compared to 67 (98.7+14.4%), 80 (99.3+14.8%), and 94 (96.5+16.4%) m/min (P<0.05). Overall, the HJ (98.4+7.9%) and the NL (105.6+15.0%) were significantly more accurate than the ACT (89.1+14.9%) and DW (81.0+35.5%) (P<0.05). There was a significant interaction between device and speed (F 9,7)=18.447, P<0.001). At 54 m/min, the HJ (95.8+15.1%) and NL (108.4+19.3%) were more accurate (P<0.05) than the ACT (81.5+19.1%) and DW (34.1+24.4%). The DW was less accurate than the ACT (P<0.05). At 67 m/min, the NL (105.4+24.4%) was significantly more accurate (P<0.05) than the ACT (89.8+19.1%). At 80 m/min there were no significant device differences. At 94 m/min, there were significant device differences (P<0.05), with the ACT (86.6+15.4%) being less accurate than the HJ (98.7+4.6%) and NL (104.4+10.8%). CONCLUSIONS: In pregnant women, the ACT and DW are less accurate than the HJ and NL. The HJ appeared to be the most accurate. These results can be useful in developing further research and PA programs that focus on walking during pregnancy.