Economic evaluation of a mobile phone text-message intervention for Australian adults with type 2 diabetes

Karen A. Waller, Anagha A. Killedar, Susan E. Furber, Eng Joo Tan, Alice A. Gibson, Adrian E. Bauman, Alison J. Hayes

Research output: Contribution to journalArticleResearchpeer-review


Background: The rising prevalence of type 2 diabetes in Australia is a public health concern, contributing to significant disease burden and economic costs. Text-message programs have been shown to improve health outcomes for people with type 2 diabetes, however they remain underutilized, and no evidence exists on their cost-effectiveness or costs of scale up to a population level in Australia. This study aimed to determine the cost-effectiveness and cost-utility of a 6-month text-message intervention (DTEXT) to improve glycated hemoglobin (HbA1c) and self-management behaviors for Australian adults with type 2 diabetes. Methods: A within-trial economic evaluation was conducted on the DTEXT randomized controlled trial. Incremental cost-effectiveness ratios (ICERs) were determined per 11 mmol/mol (1%) reduced HbA1c and per quality adjusted life year (QALY) gained, compared to usual care. Cost-effectiveness acceptability curves (CEAC) determined the probability of the intervention being cost-effective over a range of willingness to pay thresholds. A scenario analysis was conducted to determine how cost-effectiveness was impacted by using current implementation costs. Results: The DTEXT intervention cost AU$36 (INT$24) per participant, with an ICER of AU$311 (INT$211) per 11 mmol/mol (1%) reduced HbA1c. Based on HbA1c outcomes, DTEXT had a 33% probability of being effective and cost-saving. Based on the QALY outcomes, the intervention had only a 24% probability of being cost-effective. Scenario analysis indicated costs per participant of AU$13 (INT$9) to deliver the intervention, with a reduced incremental cost effectiveness ratio of AU$151 (INT$103) per 11 mmol/mol (1%) reduced HbA1c and a 38% probability of being effective and cost-saving. Conclusions: DTEXT was low cost and potentially scalable, but only had a low to moderate probability of being effective and cost saving. Further research should determine more targeted approaches that may improve cost-effectiveness.

Original languageEnglish
Article number12
Number of pages10
Publication statusPublished - Apr 2023


  • cost-effectiveness
  • diabetes
  • Economic evaluation
  • self-management
  • text-messages

Cite this