Automated insulin delivery (AID) systems substantially improve diabetes management in people with type 1 diabetes but do not account for menstrual cycle related changes in insulin sensitivity. As a result, many menstruating individuals rely on manual adjustments, such as entering fake carbohydrates, modifying insulin settings, or changing basal rates, to maintain glycemic control. This study aims to systematically characterize these manual adjustment strategies, estimate how frequently they are used across the menstrual cycle, and explore their perceived burden, effectiveness, and implications for the development of future AID algorithms that better address menstrual cycle related insulin variability.

Background Automated insulin delivery (AID) systems are expected to substantially improve diabetes management in people with type 1 diabetes but do not account for menstrual cycle related changes in insulin sensitivity, leading many menstruating individuals to rely on manual adjustments such as modifying insulin settings or entering "fake carbohydrates" to maintain glycemic control. This study will address the limited evidence in this area by systematically characterizing the range of manual adjustment strategies used across the menstrual cycle to inform the development of future AID systems.

Methods A mixed methods study design will be used, comprising an online survey followed by semi structured online focus groups with a subset of survey participants. The survey will capture self reported manual adjustment practices, while the focus groups will explore these strategies in greater depth through participant discussions to better understand the range, rationale, and variation of approaches used across the menstrual cycle.

Objectives The primary objective will be to identify and characterize manual adjustment strategies used by menstruating individuals with type 1 diabetes to compensate for menstrual cycle related changes in insulin requirements while using automated insulin delivery systems. Secondary objectives will include assessing the frequency, perceived burden and effectiveness of these strategies, their variation across the menstrual cycle, and identifying user needs to inform the development of future AID systems.

Outcomes Primary endpoints will be user reported manual adjustment strategies and perceptions of AID system use across the menstrual cycle. Secondary endpoints will include the frequency and perceived effectiveness of these strategies, user needs and preferences for AID systems, satisfaction with current system responsiveness, and the impact of menstrual cycle related adjustments on daily life.

Procedures Participants will be recruited through social media and partner networks and will complete an online REDCap survey that includes eligibility screening, informed consent, and collection of baseline characteristics, diabetes management, and menstrual cycle information. Eligible and interested participants will be invited to semi structured online focus groups (approximately five participants per group) to discuss manual adjustment strategies used during the menstrual cycle.

Statistical analysis No formal sample size calculation will be performed for the quantitative survey. Focus group data will be collected until thematic saturation is reached and will be analysed using thematic content analysis in Atlas.ti, involving transcription, coding, development of themes, and exploration of recurring patterns and relationships in manual adjustment strategies. Quantitative data will be analysed using R and Python.

Inclusion Criteria:

• At least 18 years old
• Good knowledge of English or German
• Type 1 diabetes since >3 years
• Regular menstrual cycle (24-35 days) since >2 years
• Using one of the following AID systems: Tandem t-slim:X2 with Control-IQ; Medtronic 780G; DBLG1 (Kaleido, AkkuChek Insight); CAMAPS FX (Ypsopump, Dana); Omnipod 5; AAPS DIY; IAPS DIY; Loop DIY

People with a menstrual cycle

Women with type 1 diabetes and menstrual cycle using AID system

All data extracted from the recordings can be found in the manuscript on the study.