# A validation study of the Intentional Nonadherence Scale among people with type 2 diabetes in the United Kingdom

## TL;DR

This study validates a scale to measure intentional nonadherence to diabetes treatment in the UK, showing it is reliable and useful for identifying reasons behind nonadherence.

## Contribution

The study validates the Intentional Nonadherence Scale (INAS) for use in type 2 diabetes patients in the UK, demonstrating its psychometric properties.

## Key findings

- The INAS showed high internal reliability with Cronbach's alpha values between 0.92 and 0.96.
- The scale demonstrated significant correlations with HbA1c and diabetes-related perceptions.
- The 'Testing Treatment' factor showed a trend towards predicting nonadherence over time.

## Abstract

To examine the psychometric properties of the Intentional Nonadherence Scale (INAS) among people with type 2 diabetes mellitus (PwT2D) in the United Kingdom.

This validation study recruited 260 PwT2D at diabetes intermediate care team clinics in London. Thirty of them participated in the test–retest reliability analysis in 2–4 weeks, while 124 were followed up in 3–6 months for the predictive validity analysis. The psychometric evaluation also comprised internal reliability, structural validity and construct validity that assessed the relationship between the INAS and other established measures, such as the Medication Adherence Report Scale‐5 (MARS‐5), Beliefs about Medicine Questionnaire (BMQ)‐specific, Brief Illness Perception Questionnaire (BIPQ), Patient Health Questionnaire‐2 (PHQ‐2) and glycated haemoglobin (HbA1c).

Exploratory factor analysis revealed four factors, namely ‘Resisting illness’, ‘Resisting medication’, ‘Testing treatment’ and ‘Sensitivity to medication’. All INAS factors demonstrated high internal reliability (Cronbach's alpha = 0.92–0.96). Their test–retest reliability varied between <0.001 and 0.92. Construct validity was demonstrated by its relationship with other measures, including its negative correlations with medication adherence and positive correlations with medication concerns. Significant correlations were also found with HbA1c, as well as with PwT2D's perceptions of diabetes consequences, treatment control, identity and emotional responses to diabetes. ‘Testing Treatment’ showed a trend towards statistical significance with adherence in 3–6 months (coefficient = −0.34, p = 0.09).

The INAS performed well on a number of psychometric properties in this study. It may be a helpful tool for clinicians in identifying specific drivers of intentional nonadherence among PwT2D.