Intended for healthcare professionals


Analysis of adherence to peak flow monitoring when recording of data is electronic

BMJ 2002; 324 doi: (Published 19 January 2002) Cite this as: BMJ 2002;324:146
  1. Helen K Reddel, research fellow (hkr{at},
  2. Brett G Toelle, research officer,
  3. Guy B Marks, research leader,
  4. Sandra I Ware, research assistant,
  5. Christine R Jenkins, research leader,
  6. Ann J Woolcock, professor
  1. Institute of Respiratory Medicine, Royal Prince Alfred Hospital and University of Sydney, PO Box M77, Missenden Road, Camperdown, NSW 2050, Australia
  1. Correspondence to: H K Reddel
  • Accepted 2 September 2001

Peak flow monitoring is widely recommended in international asthma guidelines. However, suspicions about the accuracy of conventional pen and paper records were confirmed when studies with electronic spirometers showed poor adherence and falsification of data.1 There seems to be a prevailing nihilistic attitude to peak flow monitoring, largely based on the perception that satisfactory adherence cannot be achieved. We aimed to measure long term adherence to electronic peak flow monitoring when participants were aware that data were being stored and used to guide treatment.

Participants, methods, and results

We obtained data from a 72 week randomised study comparing two starting doses of budesonide in patients aged 18-75 with poorly controlled asthma. The design and outcomes of the study are reported elsewhere.2 The study incorporated two novel features: twice daily monitoring with electronic diary spirometers (MicroMedical DiaryCard; MicroMedical, Rochester, UK) and titration of dose of budesonide (weeks 17-72) by using a clinical algorithm based on peak flow and diary data. A cumulative chart of peak flow and forced expiratory volume in one second was discussed with each participant at each eight weekly visit. We assessed adherence to monitoring as the percentage of scheduled sessions recorded.

Median overall adherence to monitoring over weeks 1-72 or until withdrawal was 89% (interquartile range 69-97). Adherence declined gradually from 96% in weeks 1-8 to 89% in weeks 64-72 (Spearman's R=−0.20, P<0.0001 for correlation between eight week period and adherence) (figure). Eight participants were withdrawn because of problems with adherence.


Longitudinal adherence to electronic monitoring over 72 weeks (n=61), calculated as percentage of scheduled sessions performed in each eight week period. Boxes enclose the interquartile range, with the median shown as a thick line and the angled portion of the box indicating the confidence interval of the median. Dashed lines connect observations within 1.5 interquartile ranges of the median. Crosses indicate outliers (more than 1.5 interquartile ranges from the median)


With appropriate use of electronic devices it is possible to achieve high levels of adherence to monitoring, which can be maintained in the long term even when patients have few symptoms. These high levels of adherence are in striking contrast to the findings of studies in which participants, unaware of electronic recording, were also required to complete a pen and paper diary. Verschelden et al found 44% adherence to monitoring over three months.1 However, Chowienczyk et al showed higher adherence over eight weeks for 10 participants using electronic monitoring alone compared with 16 participants who were also required to complete a pen and paper diary (median 91% v 64%, P<0.05).3

In consumer marketing, it is axiomatic that non-user friendly design features—whether in mobile phones, internet software, or preprepared meals—will decrease the frequency of use of a product and hence its market share. Similarly, ease of performance of peak flow monitoring could influence adherence to monitoring. Well designed electronic devices substantially ease the burden of monitoring, and this may have been a factor in the high rates of adherence seen in our study and in the short study by Chowienczyk et al.

Additionally, the context in which monitoring is undertaken may influence adherence. In this study, monitoring was used in managing exacerbations by means of a written action plan, and also in adjusting the dose of corticosteroid. These two features are important components of optimal self management education, which has been shown to result in improved outcomes in asthma.4 The perceived usefulness to patients of peak flow monitoring may have been enhanced by visual presentation and discussion of data at each visit.

Use of an electronic device ensures collection of accurate, rapidly accessible spirometric data.5 This is of little benefit if rates of adherence are low. In the present study, selection of a user friendly electronic device and close integration of this technology into the study design seem to have contributed to adherence rates that challenge current cynicism about the feasibility—and hence the clinical usefulness—of peak flow monitoring. The cost of electronic peak flow devices is currently high, but it may be cost effective to develop cheaper devices, which, by attention to details of design, can also facilitate adherence.


HKR initiated the study of adherence, analysed the data, wrote the manuscript, and with SIW monitored the patients. BGT assisted with interpreting the results and revising the paper. GBM provided statistical advice and analysis. CRJ and AJW (deceased) developed the clinical trial protocol and supervised the study. HKR will act as guarantor.


  • Funding AstraZeneca R&D, Lund, Sweden; AstraZeneca, Ryde, Australia; and National Health and Medical Research Council, Canberra, Australia.

  • Competing interests None declared.

  • Embedded ImageThis article is part of the BMJ's randomised controlled trial of open peer review. Documentation relating to the editorial decision making process is available on the BMJ's website


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.