BMC Medicine (Jul 2019)

A multi-centre, UK-based, non-inferiority randomised controlled trial of 4 follow-up assessment methods in stroke survivors

  • Jonathan Hewitt,
  • Anna Pennington,
  • Alexander Smith,
  • Stephanie Gething,
  • Michelle Price,
  • James White,
  • Richard Dewar,
  • Ben Carter

DOI
https://doi.org/10.1186/s12916-019-1350-5
Journal volume & issue
Vol. 17, no. 1
pp. 1 – 8

Abstract

Read online

Abstract Background Recovery following a stroke is a long and ongoing process. Post-stroke follow-up after leaving the hospital is recommended. Methods for follow-up patients include face-to-face, via the telephone, post or online (internet). However, there is a debate which method is preferred by patients. This study aimed to determine whether telephone interview, online questionnaire and postal questionnaire were as acceptable as face-to-face follow-up. Methods In a blinded, UK-wide, multi-centre, Zelen’s designed, 4-arm (postal, online, telephone, compared to face-to-face), pragmatic non-inferiority randomised controlled trial of the mode of administration, stroke survivors were randomised to postal, online, telephone and face-to-face assessment, in an equal ratio (1:1:1:1). The primary outcome was the proportion of participants that responded to the three allocation groups, compared to the face-to-face group. Subgroup analyses for age, aphasia and type and severity of stroke were carried out. A non-inferiority margin of 0.025 was used, and Holm-Bonferroni multiplicity adjustment was made. Results Of the 2074 eligible patients randomised, 55% were male (1142/2074), with an average age of 73.0 years old (SD = 13.2). Of those randomised, 22% (116/525), 9% (47/515) and 20% (101/513) responded in postal, online and telephone, respectively, compared to 17% (89/521) in the face-to-face group. The reduction in the online response rate compared to face-to-face was found to be both inferior and not non-inferior and estimated as an 8% reduction (95% CI 3.9 to 12.0%; p < 0.001). The association with lower online completion was present regardless of age, stroke type (haemorrhage or infarct) and stroke severity. In haemorrhagic stroke, the reduction in response online, compared to face-to-face, was 21% (95% CI 10 to 32%; p value = 0.002). A secondary analysis found non-aphasic stroke survivors preferred postal completion adjusted odds ratio of 1.43 (95% CI 1.04 to 1.95; p = 0.026). Conclusions The study found that fewer stroke survivors completed follow-up assessment using an online method, compared to face-to-face. This finding was present in all age groups. Caution should be employed when considering online follow-up methods in stroke survivors, particularly in those who have experienced a cerebrovascular haemorrhage. Trials registration ClinicalTrials.gov, NCT03177161. Registered on 6 June 2017.

Keywords