Design of a clinical decision support system to predict nonadherence to insulin therapy in people with type 2 diabetes mellitus
Student thesis: Master Thesis and HD Thesis
- Lea Kirstine Lewerenz Skriver
- Simone Walther
- Malene Winther Nielsen
4. term, Clinical Science and Technology, Master (Master Programme)
Background and aim: In 2021, 537 million people worldwide had diabetes mellitus, of which 80-90 % have type 2 diabetes mellitus (DM2). One of the biggest challenges of DM2 is that patients are nonadherent to treatment. This involves especially insulin users because they experience the greatest consequences of nonadherence. Nonadherence is usually only acted upon when the patient is already affected by their nonadherence in the form of complications. Therefore, there is a need for predicting nonadherence when insulin therapy is initiated. The aim of this project is to investigate how a prototype of a clinical decision support system to predict nonadherence to insulin therapy in patients with DM2 can be designed.
Method: Based on the spiral model, several iterative processes were carried out. After every iteration a new version of the prototype was designed and evaluated. A scoping review was conducted to identify factors in relation to nonadherence to insulin therapy. These factors formed the basis for the design of the prototype. Evaluation of the design consisted of four heuristic evaluations and four think-aloud tests. In total, three versions of the prototype were designed.
Results: A total of 36 factors in relation to nonadherence to insulin therapy were identified, of which 15 factors were selected for the prototype. In the heuristic evaluations and in the think-aloud tests, respectively, 30 problems were identified with suggestions for improvement. Based on the suggestions, the third version of the prototype consisted of two parts: a questionnaire part and a result part. The questionnaire part consisted of 15 yes/no questions based on the 15 factors. The result part consisted of a risk assessment with an associated color code and a list of identified optimization points.
Conclusion: The prototype is based on 15 factors. The main focus in the newest prototype is simplicity in both design and the amount of information to support the health professionals' decision-making process. The design contributes to the prototype being useful, simple, and intuitive to use.
Method: Based on the spiral model, several iterative processes were carried out. After every iteration a new version of the prototype was designed and evaluated. A scoping review was conducted to identify factors in relation to nonadherence to insulin therapy. These factors formed the basis for the design of the prototype. Evaluation of the design consisted of four heuristic evaluations and four think-aloud tests. In total, three versions of the prototype were designed.
Results: A total of 36 factors in relation to nonadherence to insulin therapy were identified, of which 15 factors were selected for the prototype. In the heuristic evaluations and in the think-aloud tests, respectively, 30 problems were identified with suggestions for improvement. Based on the suggestions, the third version of the prototype consisted of two parts: a questionnaire part and a result part. The questionnaire part consisted of 15 yes/no questions based on the 15 factors. The result part consisted of a risk assessment with an associated color code and a list of identified optimization points.
Conclusion: The prototype is based on 15 factors. The main focus in the newest prototype is simplicity in both design and the amount of information to support the health professionals' decision-making process. The design contributes to the prototype being useful, simple, and intuitive to use.
| Language | Danish |
|---|---|
| Publication date | 2023 |
| Number of pages | 96 |