Main outcome
For the period 1st of September 2016 until the 31st of December 2018, a total of 136 eligible patients were asked to participate in the study and 88 were included. The median age among the participants in the PAP-group was 54 years, in the combination-group it was 54 years and in the mindfulness-group it was 53 years. Among those who declined participation, the average age was 56 years (29 women and 19 men) (fig. 1).
The patients were randomised into the three groups, PAP (n = 29), Combination (n = 29) or Mindfulness (n = 30). In the PAP-group, there were 24.1% (2 men, 5 women) dropouts with an average age of 53 years. In the Mindfulness group there were 20% (3 men, 3 women) dropouts with an average age of 50 years. The Combination group had 17.2% (5 women) of dropouts with an average age of 53 years (figure 1). Two dropouts were due to illness, five persons did not show up at follow-ups even after two reminders. Six individuals did not want to continue without giving any reason, four people cited lack of time and one person moved and could not continue participating in the study (fig. 1). After exclusion of those with fewer than four valid activity monitor wear days, n = 26 in the PAP-, n = 26 in the combination- and n = 25 in the mindfulness group remained. The wear time with activity monitors differed between 0-12 days. There were no significant baseline differences between the dropouts and the remaining participants (see Additional file 1). There were over 80 percent of the participants at each time-point who wore the accelerometer for four days or more (see Additional file 2)
In the sensitivity analyses, we analysed the data in several different ways, both with one valid activity monitor day (see Additional file 3), and four valid activity monitor days (tab 2), with similar results.
Fulfilment of feasibility criteria
We monitored several feasibility criteria to evaluate the suitability of the study design (27). If all the feasibility criteria were fulfilled, the main study was considered possible to conduct without further changes in the protocol. If the criteria were not fulfilled, the protocol was considered to need adjustment, and if the criteria were fulfilled to less than 70% it was considered not possible to carry on with a full-scale study in the current form.
A recruitment rate of 30% was considered to be successful: 88 (64.7%) of all the 136 patients eligible for the study, who were asked to participate, accepted.
A dropout rate of less than 30% was considered successful: During the study there were a total of 20.4% (5 men, 13 women) dropouts with an average age of 52.
A successful attendance rate to the mindfulness course was set to ≥ 70%, of those randomised to any of the groups containing mindfulness should participate in at least 75% of the mindfulness meetings. There were 52% (n=15 in the mindfulness group, n=16 in the combination) who attended six times or more to the meetings.
Seventy percent of patients randomised to any group containing mindfulness should practice mindfulness for at least 20 minutes with the web-based application at least five days a week (a total of 800 minutes or more). The mean time spent in web-based training, during the study, was 184.69 minutes with a standard deviation of 330.93 minutes (minimum 0 and max 1300 minutes). Only 8% (seven persons) did 800 minutes or more (n= 4 in the mindfulness group, n=3 in the combination).
Intervention outcomes
Differences in change between intervention groups
Regarding group differences in alteration over time, percentage sedentary time showed no significant differences between the three groups (p = 0.26; tab. 2; fig. 2). Neither were there any significant differences regarding change in mean percentage of time in LIPA (p = 0.17; tab. 2; fig 3) between the three groups, nor in mean percentage of time in MVPA (p = 0.13; tab 2; fig. 4).
Self-reported leisure time activity showed a significant overall difference in change (p < 0.01) between the groups (tab. 2; fig. 6). Self-reported daily activity, on the other hand, showed no significant difference in change between groups over time (p = 0.43; tab. 2; fig. 7).
Changes over time within groups
The combination group significantly decreased their percentage sedentary time from baseline to six months (-2.8, p = 0.04; fig 2) and increased the percentage of time in LIPA (2.4, p = 0.05 fig. 3). There was also a significant increase of one percentage unit of time in MVPA (p = 0.01 fig 4.) between baseline and three months, although the increase did not last over time. Within the other groups (PAP and mindfulness) there were no significant changes regarding percentage of time in sedentary, LIPA or MVPA.
Self-reported leisure time activity (tab. 2; fig. 5) showed significant increases both in PAP (1.11, units p < 0.01 at three months, 1.17 units, p < 0.01 at six months) and the Combination (1.62 units, p < 0.01 at three months, 2.00 units p < 0.01 at six months). There was no significant difference between groups regarding self-reported daily activity (tab. 2; fig. 6) but there was a significant increase within both the PAP group (1.21 units p < 0.01 at three months, 1.09 units p < 0.01 at six months) and the Combination group (1.16 units p < 0.01 at three months, 1.14 units p < 0.01 at six months) and in the Mindfulness group at six months (0.77 units p = 0.02).
Secondary intervention outcomes
The analysis did not show any significant differences in change between groups regarding SRH (p= 0.86; tab. 2; fig. 7), ISI (p= 0.56; see Additional file 4) or FFMQ (p= 0.38; see Additional file 4). Neither were there significant differences in change over time in blood pressure (see Additional file 4), weight (tab.2; figure see Additional file 5) nor in blood lipids (see Additional file 4) between the groups.
Within all groups, there were small increases in units of SRH, significant increase in the combination-group (0.4 p = 0.03 at three months, 0.5 p = 0.01 at six months) (tab. 2; fig. 7). There was a significant decrease regarding weight within the PAP-group (tab.2; see Additional file 5 for figure) at three months (-1.8 kilograms; p < 0.01) and at six months (-2.4 kilograms; p < 0.01), and within the Combination group a decrease at six months (-1.6 kilograms; p = 0.01). Subsequently, there was a tendency towards decreased BMI as well (see Additional file 4 for table, see Additional file 6 for figure).