The cycle, contraception and the years before pregnancy

The cycle in the data, the cycle in the body

The adult cycle is the recurring pattern the reader has often had a decade or more to learn. The late-luteal sensitivity drop named in Part 1 of this guide carries forward into adult life; the published cohort evidence (Gamarra 2023, with Tatulashvili 2022 and Brown 2015 as cross-references) names the same direction, with the same wide inter-individual variation. The typical published figure for the late-luteal sensitivity drop is around 5 to 20 percent; the reader’s own personal version of the figure is often more useful than the population average.

The pattern in the CGM data is the strongest single signal. Read across three or four recent cycles, marked from the day each bleed started, the late-luteal drift in glucose often shows itself as a slow rise across the days before the bleed, sometimes with a sharper morning rise, and a return toward baseline within a day or two of the bleed starting. The pattern is real, the variation is wide, and the personal shape of it is what the team will use.

When AHCL helps with the cycle, and what is still on the user

Automated insulin delivery absorbs a portion of the late-luteal sensitivity shift through the algorithm’s own ICR tuning; the user does not need to anticipate every drift to keep the trace inside the target band (Mesa 2024 on AHCL menstrual hypo; lighter cross-reference to Monroy 2025 on 780G across cycle). What AHCL does not absorb is the recognition layer: noticing the pattern in the trace, naming the days that ran differently to the team, and asking the team to read the trace alongside any new symptoms or any new household pattern.

The AID Optimiser is a Grade D educational synthesis on a Grade A and B evidence base, reviewed by manufacturer global medical leads at CamAPS, MiniMed, Tandem, and Insulet; it is never endorsed by, validated by, or co-developed with them. It is not a substitute for the conversation with your diabetes team. The vocabulary GNL uses for the trade-off is “shorter AIT” and “lower target”; the words “aggressive” and “responsive” are out of the canon.

Contraception choices, the diabetes layer

Contraception is, first and last, a personal decision made with the practice nurse, GP, or sexual health team. The diabetes layer is the question of whether the option chosen may shift insulin sensitivity in a direction that warrants a settings review with the diabetes team. The published evidence (C20 anchor; ADA 2026 §15 women’s health) is that combined hormonal contraception can shift glycaemia modestly in some users; a settings review at the start of a new contraceptive and again at three months is appropriate, the same pattern as for any other named sensitivity shift.

The three main classes each carry a different shape of question. Combined hormonal contraception (the combined pill, the patch, the vaginal ring) is the class most consistently linked to a modest glycaemic shift in published cohorts; a settings review at start and at three months is the conservative pattern. Progestogen-only contraception (the mini-pill, the implant, the injectable) is less consistently linked to a shift; the variation across users is wide. Long-acting reversible contraception (the IUD or coil, hormonal or copper) shifts the line less, on average; the hormonal coils may carry a small local-effect-only profile.

None of those one-liners is a prescription. The choice is the personal one; the diabetes layer is the question to bring to the next diabetes review.

PCOS and T1D, the short version

Polycystic ovary syndrome is more common in women with type 1 diabetes than in the general population. The diagnostic framework and the operational management of PCOS live with the gynaecology team, not with the diabetes team (Teede 2023 international PCOS guideline). The diabetes layer of PCOS is the named tendency for baseline insulin sensitivity to sit lower in some women with PCOS, and the consequent question of whether the diabetes settings review may need to step up alongside any PCOS-led conversation.

If PCOS is on the conversation, the practical step on the diabetes side is to name it at the next diabetes review and to ask whether any cycle-pattern data should be read with the diagnosis in mind. PCOS treatments that move insulin sensitivity in named directions (metformin, hormonal contraception used as PCOS management, some weight-management interventions) each carry a settings-review question for the diabetes team. The two conversations sit alongside each other; the diabetes team does not own the PCOS conversation, and the gynaecology team does not own the settings one.

The years before a planned pregnancy, the foundations to hold

For the reader who is planning, or might be planning, a pregnancy in the coming year or two, the foundations to hold are the ones already in the adult guide: a stable cycle pattern that the reader can name, a CGM trace the team has been reading with her, a settled set of foundations (TDD, ICR, ISF, IOB visibility, hypo recognition, hypo treatment), and a relationship with the diabetes team that already includes the question of what planning a pregnancy would look like.

The pre-conception HbA1c target, the folic acid conversation, and the contraception-until-target framework all live on Part 3 of this guide. The Part 2 reader who is planning is well placed to start that conversation at the next adult diabetes review; the team will route the conversation to the joint antenatal-diabetes clinic when the planning becomes active.

Questions to bring to the next adult diabetes review

The next adult diabetes review is the conversation that ties the cycle pattern, the contraceptive choices, the PCOS question (if relevant), and the planning question (if relevant) into the wider work the team already does with the reader. Bring the last three or four cycles marked on the CGM trace. Bring any new contraceptive change or any new conversation about contraception. Bring the PCOS question if it has been raised by the GP or the gynaecology team. Bring the planning question if a pregnancy is on the horizon. Bring the trace, name the pattern, ask the team to read it with you; the settings changes that follow are the team’s, not the reader’s alone.