RESULTS
Study flow diagram
The flow chart is shown in Figure 1.
Demographic characteristics of the study population
Demographic characteristics of the study population are shown in Table 1.
Main objective: Brouwer’s asthma profile
After evaluating each patient’s profile independently and blindly, based on the criteria reported by Brouwer et al., the overall concordance between the 3 observers was 64%. Randolph’s kappa coefficient was 0.55 [0.39; 0.71]. When assessed in pairs, Cohen’s kappa coefficient ranged from 0.33 to 0.61 (Figure 2, E-Table 1).
After the Delphi approach (consensus among observers), the final overall agreement was 97% with a Randolph’s kappa coefficient of 0.97 [0.91; 1.00]. The percentage of patients that could be classifiable into a specific profile was 88% (23/26). 38% were anarchic perceivers (10/26), 27% were poor perceivers (7/26), 15% were good perceivers (4/26) and 8% were excessive perceivers (2/26). Two patients were defined as ”unclassifiable” because the first patient did not provide enough data, and the quality of the spirometry recordings of the second patient was considered to be poor. There was no consensus for one child (E-Figure 1 ; E-Table 2).
Secondary objectives
Asthma control
There was a non-significant trend towards improvement in the ACT score between baseline (median 16 [14; 20]) and the end of the study (median 20 [15; 23]) (Figure 3). The change in the ACT scores according to the patients’ perception profile is represented in E-Figure 2.
Therapeutic optimization
The distribution of treatment steps from the beginning to the end of the study is shown in Figure 4. The distribution according to the patients’ perception profile is represented in E- Figure 3.
FEV1 and PEF variability
The mean FEV1 for the first 15 days (1.49 L/s + 0.64) did not differ significantly from the mean for the last 15 days (1.48 L/s+ 0.66). The mean PEF for the first 15 days (3.15 L/s +1.42) did not differ significantly from the mean for the last 15 days (3.22 L/s + 1.61).
FEV1 variability decreased from a median of 75.6% [42.6; 87.9] at the beginning of the study to 35.6% [22.7; 73.4] at the end of the study (p=0.006) (Figure 5a). PEF variability decreased from a median of 90.2% [49.6; 112.7] at baseline to 44.4% [19.3; 97.5] at the end of the study (p=0.03) (Figure 5b).
Tool observance and acceptability
During the first 10 days, 73% of patients (19/26) recorded at least half of the 20 expected measurements (E-Figure 4). After these 10 days, 73% of patients (19/26) achieved the expected minimum of 2 weekly recordings.
Children and their parents were generally very satisfied with Spirobank Smart® follow-up according to the ASQ (After Scenario Questionnaire) (E-Figure 5).
Qualitative analysis
A total of 15 children and 17 parents were included in the qualitative analysis at the end of the study. When asked, ”Were you satisfied with the Spirobank Smart®?”, the entire sample was satisfied with the tool and the follow-up. Regarding the positive points reported by children and parents, the system was particularly appreciated for its playful and intuitive aspect. In addition, some children and parents reported a better perception of the severity of asthma exacerbations thanks to the FEV1 and PEF values displayed on the mobile application. Parents expressed a feeling of comfort and reassurance thanks to the telemonitoring. They felt that the monitoring was close without appearing over-medicalized. To the question: ”Did you find the device too medicalized?” all but one of the parents answered in the negative. Nevertheless, some parents expressed their anxiety whenever the medical team did not respond quickly enough or if the child left home without his portable spirometer. Half of the parents surveyed did not find the device constraining, but the need to perform the measurement daily was experienced as a constraint by some children. The length of the expired breaths required to record technically good spirometric tests also discouraged some children.