Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was done for this briefing in accordance with the published process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting medtech@nice.org.uk.
Published evidence
Five randomised controlled studies (RCTs) are summarised in this briefing, including 589 patients. One study was done in the UK, 2 in Australia and 2 in New Zealand.
All 5 RCTs reported that people in the intervention group, which used Smartinhaler reminders (and clinician feedback in some trials), had higher adherence to medication than those in the comparator group. Most trials found that both the intervention and control groups had improved clinical outcomes compared to baseline.
The differences in clinical outcomes between the intervention and comparator groups varied between the studies. Two studies found that the intervention arm showed significant improvement for some clinical outcomes (Chan et al. 2015, Morton et al. 2016). Chan (2015) recruited patients from the emergency department and they may have had poorer baseline adherence to medication than people in the other studies. Three studies found that there were no significant differences in any clinical outcomes.
An additional study considered the acceptability of Smartinhaler to 7 young people aged between 11 and 16 (Howard et al. 2017). Those surveyed felt that the ability to share information with healthcare professional was very useful, but were worried about how the device looked, and whether it would attract additional attention.
Also, 5 studies were identified that looked at the accuracy and reliability of the Smartinhaler device. Benchtop testing on small numbers of earlier device versions was reported by Burgess et al. (2006) and Foster et al. (2011). Foster et al. (2011) also reported a 7 day field test with patients. Pilcher et al. (2015, 2016) reported that in bench testing, accurate recording occurred in 2,796 of 2,800 actuations (99.9%) for the SmartTurbo and 2,558 of 2,560 actuations (99.9%) for the SmartTouch. Patel et al. (2013) reported the reliability of Smartinhaler devices used as monitoring tools during a 24 week trial with 303 patients. Complete data was available from 2,498 of 2,642 dispensed monitors (94.5%) and 2,498 of 2,549 returned monitors (98.0%).
Table 1 summarises the RCTs and their strengths and limitations.
Strengths and limitations of the evidence
Five good-quality RCTs were reported in full, and one of these was based in the UK NHS. The total patient group included a wide age range of adults and children.
Four of the 5 studies were sufficiently powered to investigate adherence, but may have been underpowered to detect a statistically significant difference in clinical outcomes between the intervention and the control. The study duration may also have been too short to show clinical outcomes.
Adherence reporting is based on a count of the number of activations of the inhaler. Some trials used checks to discount rapid sequences of activation indicating "dumping" medication prior to a review. There is no method to confirm that every activation resulted in delivery of medication to the person with asthma.
The different levels of medication being prescribed to people within a study and comparing the studies may have confused the results relating to clinical outcomes. Some authors noted that if prescription doses are very high, then people may be having the optimum level of medication even if they do not adhere to the prescription.
Reviews in the clinical trials were more frequent than in normal UK practice, and this may have influenced adherence levels.
Table 1 Summary of evidence
|
Study size, design and location |
Intervention and comparator(s) |
Outcomes |
Strengths and limitations |
|
26 children with asthma which was not well controlled despite preventative medicine, 6 to 14 years inclusive RCT Single centre, Australia |
Adherence feedback from Smartinhaler data at review (n=14). No feedback at review (n=12). All patients had a personalised plan, preventative medication and a Smartinhaler monitoring device. Monthly review for 6 months. |
Adherence was 21.1% higher in the intervention group (average over 4 months). Asthma control measures improved for both groups, but there was no significant difference between groups. |
Strengths: No conflicts of interest. Limitations: Results not fully reported with confidence intervals. Small sample size. Funding: not stated. The authors report no conflicts of interest. |
|
220 children (6-15 years) attending the emergency department with asthma exacerbation, and were on, or needed, twice daily treatment with corticosteroids RCT Single centre New Zealand |
All patients had preventer and reliever medication that incorporated a Smartinhaler. Followed up every 2 months for 6 months. Intervention group had audiovisual reminders enabled. Control group had audiovisual reminders disabled. |
Median adherence was significantly greater in the group using Smartinhalers (84% vs 30%). There was no significant difference in the number of days absent from school for any reason, or because of asthma. There was a significant improvement in the group using Smartinhalers for the Childhood Asthma Control Test Score and % with 1 or more parental reported exacerbations. |
Control group adherence was lower than in other studies; this may be because recruitment in the emergency department may have biased selection towards participants with poorly controlled asthma. This may explain why differences between groups for clinical outcomes were observed in this study. Funding: Smartinhalers provided by manufacturer (then called Nexus 6). Investigators provided feedback to the manufacturer about device functions, but the manufacturer had no involvement in any part of the study. Authors declared no competing interests. |
|
110 participants (12 to 65 years) with diagnosis of asthma RCT Single centre, New Zealand |
All patients were prescribed 250 micrograms fluticasone propionate twice daily via a Smartinhaler. Follow-up was for 24 weeks after a 2‑week run-in period. Clinics were at 0, 6, 12, 18, and 24 weeks. Intervention group had audiovisual reminders enabled. Control group had audiovisual reminders disabled. |
Adherence levels were significantly higher in the intervention group. The percentage of medication taken during the final 12 weeks was: intervention mean 88% (standard deviation [SD] 16), control 66% (SD 27). There were no significant differences in clinical outcomes between the 2 groups. |
The 24‑week time period was designed to allow participants to revert to normal practice over the duration of the study. Participants were not aware that adherence was the primary outcome. The dose of fluticasone propionate was 500 micrograms per day for all patients to achieve maximum benefit. Authors state that 80 to 90% of benefit can be obtained at a dose of 200 micrograms per day, which would be achieved at even 50% adherence to 500 micrograms per day. This could explain the lack of difference in lung function or quality of life between the 2 groups. Funding: supported by a grant from GlaxoSmithKline, UK. Authors declare no competing interests. |
|
143 patients (14 to 65 years) with suboptimal asthma control, and prescribed twice daily combined corticosteroid and long-acting b2‑agonist medication for 1 month or more Pragmatic cluster 2×2 factorial parallel-group. RCT 43 GPs, Australia 2010–13 GPs randomised to intervention or control |
Inhaler reminders and feedback (IRF): twice daily Smartinhaler reminders for missed doses, graphs available to GPs and patients. Adherence discussed at GP follow-up. Personalised adherence discussion (PAD): Discussion with GP about barriers to adherence, and patient set goals. Usual care (UC): GP training, asthma plan, inhaler education and follow up appointment. Each branch provided separately plus one with PAD plus IRF. All patients had SmartTrack devices, but only the IRF group (patients and GPs) were made aware of the functionality during trial. Data was collected by telephone at 0, 2, 4 and 6 months. |
There was a statistically significant improvement in adherence between the groups with audiovisual reminders and feedback and the groups without. There were no statistically significant differences between groups for other outcomes. In general, patients had a significant improvement in clinical outcomes between baseline measurements and the study period. |
Strengths: Pragmatic design in primary care setting. Study data collected directly by study staff. Limitations: Several GPs were lost to the study meaning fewer patients were recruited than the planned 220 for the sample size calculation for asthma control. Sample sizes were not calculated for other outcomes. Baseline fluticasone prescriptions were high (718 micrograms/day) compared to a suggested maximum efficiency at 500 micrograms/day. Authors suggest that a slight increase in adherence for all groups may have led to improved clinical outcomes, with higher adherence rates unable to produce further improvement. Funding: National Health and Medical Research Council of Australia with support for equipment purchase (but not Smartinhaler) from manufacturers. Conflicts of interest are declared and are stated in the paper. |
|
90 children (6 to16 years) with poorly controlled asthma (asthma control questionnaire [ACQ] at least 1.5). All taking regular inhaled steroids. Open label RCT. 2 hospital clinics in UK. 2013–14. |
All patients were reviewed in routine asthma clinics at 3, 6, 9 and 12 months. Prior to randomisation all patients had a review of inhaler technique and an education session. Intervention: Smartinhaler device with reminders activated. Data reviewed and adherence discussed at clinic visits. Comparator: Smartinhaler device with reminders deactivated. No review of data until the end of study. |
Primary outcome was ACQ score at 3, 6, 9 and 12 months. This decreased in both groups, but there were no significant differences between the groups. The intervention group experienced significantly fewer unplanned attendances at the GP or emergency department for asthma, and significantly fewer courses of oral steroids were needed. FEV, days off school due to asthma and use of beta agonists were not significantly different between groups. There was a high rate of lost or broken devices, particularly in the intervention arm. |
Strengths: UK setting Protocol based on routine clinical practice 12‑month duration. Limitations: Non-blinded. Funding: Not stated. No competing interests declared. |
Recent and ongoing studies
NCT02386722 Intervention to Improve Inhalative Adherence Cantonal Hospital, Baselland, Switzerland has a primary outcome date of March 2016, but the status is recruiting.