The Moving is Improving! study evaluated the stimulating effect of a mobilization poster on early mobilization after heart surgery, as measured with the ACSM and TCT scores.

ACSM score was sensitive to measure daily improvement of patient functional activities after heart surgery. A stabilization pattern was observed 4 to 5 days after ICU discharge. This is in line with accelerometer data obtained at our surgical ward [8]. The mobilization poster did not increase ACSM score in the overall group (p = 0.27) or CABG subgroup (p = 0.15).

We are the first to report results of an “aggressive phase I mobilization” strategy [31]. Our study shows that early mobilization is safe and results in increased activity levels without compromising safety. The early mobilization protocol is comparable in physical activity to other published protocols [32]. Afxonidis et al. intensified early postoperative physiotherapy sessions after cardiac surgery starting at the ICU, and reached milestones during hospital stay compared to our Day 2 protocol [15]. Physical therapists in Australia and New Zealand reported in a national survey that all patients are sitting out of bed at first postoperative day and 20% completed one flight of stairs on day 4 after uncomplicated CABG [14]. In our study, 88 to 96% sat on a chair at first postoperative day, 36 to 48% cycled at a cycle ergometer on day 3 and 45 to 75% walked the stairs on day 4 (see Additional file 1: File S7).

The ACSM score increased drastically in this study from day 1 to day 4. That means that implementation of this poster in more conservative mobilization programs focusing on lower metabolic equivalents of physical activity [15] are likely to observe an increase in ACSM score during hospital stay as well.

Higher ACSM scores were observed in men who might have a more competitive attitude to mobilization or overestimate self-reported activities. Self-reported patient activities were aligned with nursing records and with activities performed during physical therapy sessions. Most of these activities were unsupervised, and self-reported activities could therefore shift a borderline score towards a higher level. Also, women might be more hesitant to mobilize after surgery. This finding is in line with a study on postoperative mobilization after total hip arthroplasty [33]. No differences in in-hospital mobilization after cardiac surgery were observed between men and women in another study where accelerometers were used [8]. Further research should quantify this difference using objective data from activity trackers in a larger study. If the difference between men and women persists in future studies, physical therapists should develop motivational interviewing strategies to reduce women patients’ hesitance to mobilization.

In the Netherlands, receiving CABG is an absolute indication for cardiac rehabilitation, while valve surgery is a relative indication [29]. Therefore, a subgroup analysis for CABG was planned. No differences were found in CABG mobilization activities compared to the overall group, with a similar non-significant effect on ACSM score for the PMG (p = 0.15).

The mobilization poster had a significant impact on sitting in a chair, walking to the toilet and along the corridor and cycling at a cycle ergometer. No difference in TCT score for bed and stairs was found. This effect was persistent in the CABG subgroup. Patients generally only lay in bed at day 1 after ICU discharge and this did not change between UCG and PMG. Walking the stairs is a discharge target and thus did not change. These TCT scores might be removed in future studies to reduce administrative load for physical therapists.

Classification of patient mobility with the TCT scores had an overall high agreement with the ACSM score, with ICCs in the good to excellent range for bed, chair, toilet and corridor. There was poor agreement for cycle ergometer and stairs, which was no surprise because these activities are not described in the ACSM score at all. Last, TCT score was planned as an individual score per activity. Future studies might include a cumulative day score for comparison and for potentially clinical cut-off selection.

Scoring was based on self-reported activities and was not blinded, potentially introducing bias. A study in our center used accelerometer measurements for objective qualification of activities, and found similar results [8].

Next, our prospective sequential-group study might include bias compared to a randomized controlled trial, and did not allow for extension of the control group, resulting in unequal group sizes. After the initial PMG group and extensive analysis, poster mobilization was continued as new standard. In this study, baseline characteristics were balanced between both groups (Table 2). Furthermore, a parallel randomized design would allow for informal cross-over as patients might see or discuss the poster at the surgical ward. Last, type I error might have been introduced with 7 primary endpoints (1 ACSM score and 6 TCT scores). With a Holm-Bonferroni correction for our primary endpoints, we still find significant results in our mixed model analysis (Additional file 1: Table S5.3) for chair (p = 0.007), toilet (p = 0.012), corridor (p = 0.025), and not cycle ergometer (p = 0.08).

A linear mixed model was used as length of stay differed between patients and repeated ACSM or TCT measures from the same patient are more similar than responses from other patients. A sample size of 32 patients in the control group might be too small to find a significant effect of a mobilization poster for a composite measure such as ACSM score. Contrarily, the TCT scores focusing on one activity only were able to find an improved effect of the mobilization poster in this study population.

ACSM and TCT scores sitting, and walking to the toilet and corridor were higher in the PMG at the first postoperative day. Preoperative ACSM and TCT scores were not collected the day before surgery. All patients were independent in daily life mobilization, as determined with the Katz Index of Independence in Activities of Daily Living [21]. It is unclear if the poster benefits patients as early as day one, or alternatively, that the patient groups differed already at ICU discharge. As there was no patient selection, and both cohorts were in a short time span, we do not expect preoperative differences and address this increase to be the poster effect.

Other patient groups with intermediate to long hospital stay might benefit from early mobilization. At the cardiology ward and other wards early mobilization for patients with or without a mobilization poster should be evaluated using the ACSM or TCT score. Also congenital, heart transplant and heart assist device patients might benefit from a mobilization poster. Effects of prehabilitation for frail patients [34], patients with overweight or preoperatively expected long cardiopulmonary bypass times [35] for complex cardiac surgery can also be studied with these measures.

The poster group used an A1 paper size poster during ward stay. Functional activities might increase faster with a comprehensive approach including lifestyle changes [36], and using digital persuasive technology (telerehabilitation) focusing on the current activity level and milestones. Future work should focus on patient-specific information and exercises that match the current functional level of patients recovering from cardiac surgery based on self-reported or wearable device measurements [6, 8].