Developing acceptable brief messages based on behaviour change techniques to encourage medication adherence in people with Type 2 diabetes (Preprint)
Bartlett YK., Farmer A., Rea R., French DP.
<sec> <title>BACKGROUND</title> <p>Brief messages are a promising way to improve adherence to medication for people with type 2 diabetes. However, it is often unclear how messages have been developed and their precise content, making it difficult to ascertain why certain messages are successful or not.</p> </sec> <sec> <title>OBJECTIVE</title> <p>To develop messages that have high fidelity to specified evidence-derived behaviour change techniques(BCTs) and are acceptable to people with type 2 diabetes.</p> </sec> <sec> <title>METHODS</title> <p>Four studies were conducted;1) A workshop (n=21) where behavioural change researchers and health care professionals developed messages based on specific BCTs or beliefs or concerns related to taking medication,2) a focus group study with people with type 2 diabetes (n=23) to assess acceptability of the approach, 3) a survey to ascertain the acceptability of a sub-set of messages to people with type 2 diabetes(n=52) and,4) a survey with behaviour change experts to assess the fidelity of a sub-set of messages to their intended BCT (n=18).</p> </sec> <sec> <title>RESULTS</title> <p>Study 1;371 messages based on 38 BCTs/ beliefs and concerns were developed. Workshop participants rated BCTs to be relevant to medication adherence (Mean =7.12/10, Standard Deviation (SD)=1.55) and messages to have good fidelity(Mean=7.42/10, SD=1.19). Study 2;the approach of providing medication adherence support through text messages was found to be acceptable. Study 3; mean acceptability of all BCTs was above the mid-point (Mean =3.49/5, SD=0.26). Study 4;mean fidelity for all BCTs was above the mid-point (Mean=7.61/10, SD=1.38).</p> </sec> <sec> <title>CONCLUSIONS</title> <p>A library of brief messages was developed that are acceptable to people with type 2 diabetes and are representative of specific evidence-derived BCTs. This approach allowed brief messages to be developed with known content that can be used to test theory.</p> </sec>