Development, Validation, and Application of a Diabetes Drug Discontinuation Questionnaire: A Multicentric Observational Study

Abstract

Medication adherence refers to patients taking their prescribed treatments at the correct dose, timing, and frequency. In contrast, medication discontinuation involves sustained interruption of the entire regimen, with greater risks of metabolic decompensation and complications. Despite its clinical significance, anti-diabetes medication discontinuation remains poorly understood, with no standardized definitions or validated assessment tools. To develop and validate a culturally adapted, self-administered questionnaire to assess the prevalence, duration, and factors linked to discontinuation of anti-diabetes medication. This multicenter observational study was conducted across 11 diabetes care centers in Northern India. A structured questionnaire was developed through expert consensus and patient pre-testing, followed by psychometric validation. The validated instrument was then administered to a large cohort of adults with diabetes to assess discontinuation patterns and associated factors. Among 747 participants (mean age: 55.98 years; 54.2% male), 189 individuals (25.3%) reported at least one episode of medication discontinuation lasting more than seven consecutive days. Discontinuation was frequently prolonged: 36.3% of affected participants discontinued therapy for one to six months, and 29.4% for more than one year. Compared with those who remained on therapy, individuals who discontinued had higher average glycemic levels, including a mean HbA1c approximately 0.5% higher, along with higher fasting and post-prandial glucose values. The most commonly reported reasons for discontinuation were medication unavailability (34/197; 17.3%), fear of hypoglycemia (33/197; 16.8%), perceived good glycemic control (29/197; 14.7%), and belief that medication was unnecessary (27/197; 13.7%). Discontinuation was more common among individuals with behavioral risk factors and microvascular complications, while concurrent use of cardiovascular medications was associated with greater treatment persistence. Most participants considered discontinuation unjustified and expressed strong support for digital reminder systems. Medication discontinuation affects approximately one in four diabetes patients, often for extended periods. Unlike adherence, discontinuation is rarely measured. This validated questionnaire provides the first reliable tool to assess this behavior in India, highlighting knowledge gaps and access barriers as primary drivers, rather than socioeconomic factors. Interventions should focus on patient education and digital adherence support systems.