Background: Threatened preterm delivery is a serious obstetrical complication and has for decades been prescribed physical activity restrictions (AR). Adherence to the recommended level of physical AR is however unknown. This study aimed to assess the objectively measured different physical positions and activities of pregnant women recommended AR due to threatened preterm delivery complications, compared to a reference group of uncomplicated pregnant women without restrictions, and to explore if admission status influenced adherence to AR. Methods: A Danish descriptive, clinical multi-center study included singleton pregnancies between 22–33 gestational weeks admitted to an antenatal ward or during midwife consultations either prescribed AR due to threatened preterm delivery or uncomplicated controls without restrictions. For seven days participants wore two tri-axial accelerometric SENS® monitors. Accelerometric data included time spent in five different positions, activities, and step counts. At inclusion demographic and obstetric information was collected. Results: Seventy-two pregnant women participated; 31% were prescribed strict AR, 15% moderate, 3% light, 8% unspecified, and 43% had no AR. Strict AR participants rested in the supine/lateral position for 17.7 median hours/day (range:9.6–24.0); sat upright 4.9 h/day (0.11–11.7); took 1,520steps/day (20–5,482), and 64% were inpatients. Moderate AR participants rested in the supine/lateral position for 15.1 h/day (11.5–21.6); sat upright 5.6 h/day (2.0–9.3); took 3,310steps/day (467–6,968), and 64% were outpatients. Participants with no AR rested 10.5 h/day (6.3–15.4) in supine/lateral position; sat upright 7.6 h/day (0.1–11.4) and took 9,235steps/day (3,225–20,818). Compared to no restrictions, participants with strict or moderate AR spent significant more time in physical resting positions and took significant fewer mean steps. Among strict AR admission status did not alter time spent in the physical positions, nor the step count. Conclusions: Overall, participants adhered highly to the recommended AR. However, discriminating between strict and moderate AR recommendations did not alter how physical resting positions and activities were carried out. The admission status did not influence how participants adhered to strict AR.