Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Baseline pain medication is associated with longer duration of high adherence in a three-month digital treatment program for hip and knee osteoarthritis

Dahlberg, Leif E. LU ; Rowland, Simon P. ; Pearson, Jack T. ; Lohmander, L. Stefan LU orcid and Kiadaliri, Ali LU orcid (2026) In Osteoarthritis and Cartilage Open 8(1).
Abstract

Objective: To investigate whether baseline use of pain medication is associated with program adherence during a three-month digital treatment program for individuals with hip or knee osteoarthritis (OA). Design: An observational cohort study using registry data on weekly participant adherence from 33078 participants enrolled in a digital education and exercise therapy program. Poor adherence was defined as completing less than 80 % of the approximately 20 prescribed weekly activities for two consecutive weeks during the 13-week treatment period. Baseline analgesic use was categorized into six groups: no medication, paracetamol (with/without dietary supplements), NSAIDs (with/without supplements), paracetamol combined with NSAIDs... (More)

Objective: To investigate whether baseline use of pain medication is associated with program adherence during a three-month digital treatment program for individuals with hip or knee osteoarthritis (OA). Design: An observational cohort study using registry data on weekly participant adherence from 33078 participants enrolled in a digital education and exercise therapy program. Poor adherence was defined as completing less than 80 % of the approximately 20 prescribed weekly activities for two consecutive weeks during the 13-week treatment period. Baseline analgesic use was categorized into six groups: no medication, paracetamol (with/without dietary supplements), NSAIDs (with/without supplements), paracetamol combined with NSAIDs (with/without dietary supplements), dietary supplements only, and opioids (with/without other medications). Interval-censored parametric survival models adjusted for baseline characteristics were used for statistical analysis. In sensitivity analyses, alternative definitions of poor adherence were used. A complete case analysis was conducted as a subgroup analysis. Results: Compared to the no-medication group, individuals taking paracetamol—either alone (hazard ratio 0.94, 95 % CI 0.91, 0.98) or in combination with NSAIDs (0.91, 0.87, 0.94) reached poor adherence later. In contrast, opioid users (hazard ratio 1.12, 95 % CI 1.06, 1.19) reached poor adherence earlier. Adjusted median days to reach poor adherence ranged from 39.2 (95 % CI 36.9, 41.4) for opioids to 49.2 (47.7, 50.7) for paracetamol + NSAIDs users. Alternative definitions of poor adherence and a complete-case analysis generally yielded similar findings. Conclusion: Baseline use of paracetamol, alone or with NSAIDs, was associated with longer time to reach poor adherence, whereas opioid use predicted poor adherence earlier.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
eHealth, Osteoarthritis, Pain medication, Physiotherapy
in
Osteoarthritis and Cartilage Open
volume
8
issue
1
article number
100727
publisher
Elsevier
external identifiers
  • pmid:41536420
  • scopus:105025044305
ISSN
2665-9131
DOI
10.1016/j.ocarto.2025.100727
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2025 The Authors
id
ccbf27f0-b1a2-43eb-b1db-f2c25edd7d45
date added to LUP
2026-03-09 11:16:49
date last changed
2026-03-10 03:00:07
@article{ccbf27f0-b1a2-43eb-b1db-f2c25edd7d45,
  abstract     = {{<p>Objective: To investigate whether baseline use of pain medication is associated with program adherence during a three-month digital treatment program for individuals with hip or knee osteoarthritis (OA). Design: An observational cohort study using registry data on weekly participant adherence from 33078 participants enrolled in a digital education and exercise therapy program. Poor adherence was defined as completing less than 80 % of the approximately 20 prescribed weekly activities for two consecutive weeks during the 13-week treatment period. Baseline analgesic use was categorized into six groups: no medication, paracetamol (with/without dietary supplements), NSAIDs (with/without supplements), paracetamol combined with NSAIDs (with/without dietary supplements), dietary supplements only, and opioids (with/without other medications). Interval-censored parametric survival models adjusted for baseline characteristics were used for statistical analysis. In sensitivity analyses, alternative definitions of poor adherence were used. A complete case analysis was conducted as a subgroup analysis. Results: Compared to the no-medication group, individuals taking paracetamol—either alone (hazard ratio 0.94, 95 % CI 0.91, 0.98) or in combination with NSAIDs (0.91, 0.87, 0.94) reached poor adherence later. In contrast, opioid users (hazard ratio 1.12, 95 % CI 1.06, 1.19) reached poor adherence earlier. Adjusted median days to reach poor adherence ranged from 39.2 (95 % CI 36.9, 41.4) for opioids to 49.2 (47.7, 50.7) for paracetamol + NSAIDs users. Alternative definitions of poor adherence and a complete-case analysis generally yielded similar findings. Conclusion: Baseline use of paracetamol, alone or with NSAIDs, was associated with longer time to reach poor adherence, whereas opioid use predicted poor adherence earlier.</p>}},
  author       = {{Dahlberg, Leif E. and Rowland, Simon P. and Pearson, Jack T. and Lohmander, L. Stefan and Kiadaliri, Ali}},
  issn         = {{2665-9131}},
  keywords     = {{eHealth; Osteoarthritis; Pain medication; Physiotherapy}},
  language     = {{eng}},
  number       = {{1}},
  publisher    = {{Elsevier}},
  series       = {{Osteoarthritis and Cartilage Open}},
  title        = {{Baseline pain medication is associated with longer duration of high adherence in a three-month digital treatment program for hip and knee osteoarthritis}},
  url          = {{http://dx.doi.org/10.1016/j.ocarto.2025.100727}},
  doi          = {{10.1016/j.ocarto.2025.100727}},
  volume       = {{8}},
  year         = {{2026}},
}