ANSWER: C

Rationale:

This question integrates the mechanism hierarchy (anovulation as primary, cervical mucus as secondary, endometrial atrophy as tertiary) with the concept that anovulation defines the contraceptive ceiling — meaning the greatest pregnancy risk arises whenever the ovulation-suppression mechanism is compromised. The hormone-free interval is the period of greatest vulnerability because, during those days without active hormone, the falling steroid levels release the hypothalamic-pituitary axis from suppression and FSH begins follicular recruitment. If a woman then extends the hormone-free interval by missing pills at the start of the next pack, the recruited follicle can continue to mature and reach ovulation before exogenous hormone restores suppression. This is why missed-pill guidance places the greatest emphasis on the pills immediately following the hormone-free interval and why lengthening the pill-free interval is the most consequential adherence error.

• Option A: Option A is incorrect because the cervical mucus barrier, while an important secondary mechanism, is not the only mechanism that matters and is not the dominant determinant of pregnancy risk — anovulation is the contraceptive ceiling, and it is loss of ovulation suppression, not cervical mucus, that drives the elevated risk of missed pills after the hormone-free interval.
• Option B: Option B is incorrect because endometrial atrophy is the tertiary, back-up mechanism, not the primary mechanism, and missed pills do not cause clinically meaningful implantation-permitting endometrial regeneration within 24 hours; the dominant risk is breakthrough ovulation, not endometrial recovery.
• Option D: Option D is incorrect because timing does matter — missed pills are not equally risky throughout the cycle; the pills adjacent to the hormone-free interval carry the greatest risk because of the follicular recruitment that occurs during the pill-free days, whereas missed pills mid-pack with intact prior suppression carry comparatively lower risk.
• Option E: Option E is incorrect because the start of the pack, immediately after the hormone-free interval, is actually the most dangerous time to miss pills, not the safest — there is no estrogen reserve that protects against ovulation at this point; on the contrary, follicular recruitment is already underway.

ANSWER: B

Rationale:

This question integrates three clinically important effects of ethinyl estradiol — VTE risk, blood pressure elevation, and anti-androgenic benefit — by tracing all of them to a single underlying property: EE's potent first-pass hepatic estrogenic effect. Because EE resists hepatic metabolism (owing to the 17-alpha-ethinyl substitution) and passes through the liver in high concentration after oral absorption, it produces strong estrogenic stimulation of hepatic protein synthesis. This drives up production of coagulation factors (II, VII, VIII, X) and reduces activated protein C sensitivity, increasing VTE risk; it increases angiotensinogen synthesis, contributing to blood pressure elevation; and it increases sex hormone-binding globulin (SHBG) synthesis, which binds circulating androgens and lowers the free androgen fraction, producing the anti-androgenic benefit. Recognizing this common hepatic origin explains why route of administration matters — transdermal estradiol, which bypasses the high-concentration first-pass hepatic exposure, produces far less of these hepatic protein effects and carries lower VTE and blood pressure risk.

• Option A: Option A is incorrect because the shared origin of these effects is hepatic protein synthesis stimulation, not endothelial nitric oxide synthase activation; nitric oxide effects do not explain the coordinated increases in coagulation factors, angiotensinogen, and SHBG.
• Option C: Option C is incorrect because EE does not produce these three effects by displacing progestins from plasma protein binding — the effects arise from EE-driven hepatic synthesis of specific proteins, not from protein-binding displacement of progestins.
• Option D: Option D is incorrect because the relevant site is the liver, not adipose tissue — the coordinated increases in coagulation factors, angiotensinogen, and SHBG are products of hepatic synthesis, and adipose estrogen receptor activity does not account for these plasma protein changes.
• Option E: Option E is incorrect because EE does not produce these effects by inhibiting renal estrogen clearance — the mechanism is potent first-pass hepatic estrogenic stimulation of protein synthesis, and the route-dependence of these effects (oral EE versus transdermal estradiol) specifically implicates hepatic first-pass exposure rather than renal clearance.

ANSWER: A

Rationale:

This question integrates three separate Category 4 contraindications for combined methods — prior VTE, migraine with aura, and smoking over age 35 — and asks the student to recognize their common pharmacological root. Each of these contraindications is driven by the ethinyl estradiol (EE) component: EE promotes hepatic synthesis of coagulation factors and reduces activated protein C sensitivity (the venous thrombotic mechanism underlying the VTE contraindication), and EE promotes platelet activation and arterial thrombotic risk that compounds the elevated arterial risk in migraine with aura and in older smokers. Because progestin-only methods contain no EE, they do not produce this estrogen-mediated venous and arterial thrombotic risk amplification, and they are therefore classified Category 1 or 2 across these conditions that are Category 3 or 4 for combined methods. This single integrating principle — that the estrogen component drives the major contraindications — explains why progestin-only methods are the default safe choice in women with estrogen-contraindicated conditions.

• Option B: Option B is incorrect because progestins do not function as anticoagulants and do not lower VTE or stroke risk below baseline — their safety in these conditions derives from the absence of the EE-mediated risk amplification, not from any active protective or anticoagulant effect.
• Option C: Option C is incorrect because progestin-only methods do achieve systemic absorption (the progestin-only pill, implant, and DMPA all produce systemic progestin levels) — their safety is not due to a complete absence of systemic drug but to the absence of the estrogen component responsible for the thrombotic risk; even systemically absorbed progestins do not carry the EE-type thrombotic risk.
• Option D: Option D is incorrect because the combined patch and ring contain EE and carry the same estrogen-mediated contraindications as the oral combined pill — they are also Category 4 in these conditions, and progestin-only methods do offer a specific advantage by eliminating the estrogen component.
• Option E: Option E is incorrect because progestins are not structurally identical to ethinyl estradiol (they are distinct steroid classes acting on different receptors) and their safety is not due to rapid clearance preventing coagulation effects — it is the absence of estrogenic hepatic stimulation that accounts for their favorable thrombotic safety profile.

ANSWER: D

Rationale:

This question integrates the concept of enzyme induction (which accelerates systemic metabolism of contraceptive steroids) with the distinct pharmacokinetics of local versus systemic delivery systems. Combined pills, the progestin-only pill, and the etonogestrel implant all depend on maintaining adequate systemic plasma concentrations of their active steroids; when rifampin induces CYP3A4 and accelerates systemic clearance, those plasma levels fall and contraceptive efficacy is compromised. The LNG-IUD is fundamentally different: its contraceptive effect — endometrial decidualization and atrophy plus cervical mucus impermeability — is mediated by extremely high local levonorgestrel concentrations within the uterus, hundreds of times higher than the systemic plasma levels. Even if enzyme induction accelerates the metabolism of the small amount of levonorgestrel that reaches the systemic circulation, the local intrauterine concentrations that actually produce contraception are unaffected. Because the mechanism is local rather than systemic, the LNG-IUD remains Category 1 with enzyme inducers.

• Option A: Option A is incorrect because the LNG-IUD does not work by delivering a high systemic dose that survives induction — its systemic levonorgestrel levels are actually very low; its protection comes from the local site of action, not from a large systemic dose buffer.
• Option B: Option B is incorrect because rifampin's effect on contraception is not about whether the drug physically reaches the uterine cavity — rifampin acts by inducing hepatic and intestinal enzymes that metabolize contraceptive steroids systemically; the LNG-IUD's protection is due to its local mechanism of action, not to any inability of rifampin to reach pelvic tissues.
• Option C: Option C is incorrect because the LNG-IUD is a hormonal device containing levonorgestrel, not a copper device — the copper IUD is the non-hormonal device; the LNG-IUD is protected by its local hormonal mechanism, not by being non-hormonal.
• Option E: Option E is incorrect because enzyme inducers do not selectively spare levonorgestrel — rifampin reduces both ethinyl estradiol (by more than 50%) and levonorgestrel (by approximately 40%) systemically; levonorgestrel-containing oral and implant methods are not automatically protected, and only the LNG-IUD is protected, by virtue of its local mechanism.

ANSWER: E

Rationale:

This question integrates the mechanism of the ethinyl estradiol–lamotrigine interaction (EE induction of UGT1A4 accelerating lamotrigine glucuronidation) with its reversibility to explain a cyclic, bidirectional clinical pattern. During the 21 active-pill days, EE induces UGT1A4 and increases lamotrigine clearance, lowering lamotrigine plasma concentrations and predisposing to breakthrough seizures. During the 7 hormone-free days, EE is withdrawn; UGT1A4 induction reverses toward baseline, lamotrigine clearance slows, and lamotrigine levels rebound upward, producing dose-related toxicity such as dizziness and diplopia. The result is two opposite problems appearing at different points in the same pill cycle — subtherapeutic lamotrigine during active pills and supratherapeutic lamotrigine during the hormone-free interval. This is precisely why EE-containing methods are problematic in lamotrigine users and why a continuous progestin-only method (no EE, no UGT1A4 induction, no cyclic fluctuation) is preferred.

• Option A: Option A is incorrect because the symptoms are not unrelated to the contraceptive — they are a direct and characteristic consequence of the cyclic EE–lamotrigine interaction; simply increasing the baseline lamotrigine dose would worsen the hormone-free-interval toxicity.
• Option B: Option B is incorrect because it inverts the direction of the interaction — it is ethinyl estradiol that induces lamotrigine metabolism (via UGT1A4), not lamotrigine that induces EE metabolism; the cyclic problem is fluctuating lamotrigine levels, not estrogen excess.
• Option C: Option C is incorrect because ethinyl estradiol does not permanently destroy UGT1A4 — the induction is reversible, which is exactly what produces the rebound in lamotrigine levels during the hormone-free interval; the effect is not a progressive irreversible decline producing seizures only.
• Option D: Option D is incorrect because the combined pill does have a substantial, well-documented effect on lamotrigine levels through UGT1A4 induction — the cyclic symptoms are a drug interaction, not simply normal menstrual-cycle variation in seizure threshold.

ANSWER: B

Rationale:

This question integrates the mechanistic difference between levonorgestrel and ulipristal acetate (pre-ovulatory LH-surge suppression versus post-LH-surge follicular rupture inhibition) with the timing of both the intercourse (60 hours ago) and the patient's cycle position (near ovulation). Two factors make ulipristal acetate (UPA) the preferred agent. First, because she is near ovulation, the LH surge may already be underway; levonorgestrel works only by suppressing the LH surge before it peaks and loses efficacy once the surge has begun, whereas UPA, as a selective progesterone receptor modulator, can still inhibit or delay follicular rupture even after the surge has started — exactly the situation most likely to result in imminent ovulation. Second, at 60 hours post-intercourse, both agents are still within their windows (levonorgestrel to 72 hours, UPA to 120 hours), but UPA's efficacy advantage is greatest precisely when ovulation is imminent and when treatment is delayed. The integration of mechanism and timing therefore points to UPA.

• Option A: Option A is incorrect because it misattributes post-LH-surge activity to levonorgestrel — levonorgestrel cannot reliably inhibit follicular rupture once the LH surge has begun; that capability belongs to ulipristal acetate, which is why UPA, not levonorgestrel, is preferred when ovulation is imminent.
• Option C: Option C is incorrect because at 60 hours the levonorgestrel 72-hour window has not closed, and levonorgestrel does not work post-ovulation — its mechanism is pre-ovulatory; the reasoning given is doubly wrong.
• Option D: Option D is incorrect because oral emergency contraception is still effective at 60 hours — both agents remain within their windows, and ulipristal acetate in particular retains efficacy out to 120 hours; the claim that neither oral agent can work this late is incorrect.
• Option E: Option E is incorrect because the two agents are not equally effective regardless of cycle position — their efficacy diverges precisely around the time of the LH surge, where ulipristal acetate retains activity that levonorgestrel lacks, making cycle position a decisive factor in agent selection.

ANSWER: C

Rationale:

This question integrates ulipristal acetate's pharmacology at the progesterone receptor with the mechanism of the copper IUD to resolve a clinical dilemma. Ulipristal acetate (UPA) is a selective progesterone receptor modulator that requires access to the progesterone receptor (PR) to exert its antiovulatory effect. In a woman with a progestin method such as the etonogestrel implant, ambient etonogestrel occupies the PR and competes with UPA, attenuating UPA's efficacy. Adding levonorgestrel emergency contraception faces a related problem and provides no advantage. The copper IUD, by contrast, works through copper ion cytotoxicity to spermatozoa — a mechanism entirely independent of PR status and therefore unaffected by the circulating etonogestrel. This makes the copper IUD the pharmacologically preferred option, and it provides reliable ongoing contraception as well. The integrating principle is that when ambient progestin compromises the receptor-dependent oral agents, a mechanism that does not depend on the progesterone receptor is the rational choice.

• Option A: Option A is incorrect because ambient etonogestrel does not synergistically enhance ulipristal acetate's effect — it competes with UPA at the progesterone receptor and attenuates its efficacy; the interaction is antagonistic interference, not synergy.
• Option B: Option B is incorrect because adding levonorgestrel to a woman already on a progestin method does not reinforce contraceptive efficacy as emergency contraception — systemic levonorgestrel EC faces the same ambient-progestin problem and provides no reliable advantage, which is why the copper IUD is preferred.
• Option D: Option D is incorrect because the presence of the implant does bear directly on ulipristal acetate's receptor pharmacology — ambient etonogestrel at the progesterone receptor attenuates UPA efficacy, so the 120-hour window is not the only relevant consideration.
• Option E: Option E is incorrect because an etonogestrel implant beyond its 3-year approved duration cannot be assumed to maintain reliable ovulation suppression — declining etonogestrel release is precisely why she may be at risk, so emergency contraception is appropriately indicated rather than unnecessary.

ANSWER: A

Rationale:

This question integrates the weight-dependent pharmacokinetics of all three principal emergency contraceptive options into a single reliability hierarchy. Levonorgestrel EC efficacy declines first and most steeply with rising BMI: serum levonorgestrel concentrations after the 1.5 mg dose fall significantly above approximately BMI 26 and become inadequate for reliable ovulation suppression above approximately BMI 35, owing to increased volume of distribution and potentially faster metabolism. Ulipristal acetate is more weight-robust than levonorgestrel and maintains superior efficacy at higher BMI categories, making it the preferred oral agent above BMI 26, though even UPA shows some attenuation at very high BMI. The copper IUD is entirely weight-independent because its mechanism — copper ion cytotoxicity to spermatozoa — does not depend on achieving any systemic drug concentration; it remains maximally effective at any BMI. The resulting hierarchy at high BMI is copper IUD greater than ulipristal acetate greater than levonorgestrel.

• Option B: Option B is incorrect because the three options do not decline equally with rising BMI — they have markedly different weight sensitivities (levonorgestrel most sensitive, copper IUD insensitive), and this difference is precisely what makes BMI a useful basis for agent selection.
• Option C: Option C is incorrect because it inverts the relationships — levonorgestrel is the least weight-robust option, not the most, and the copper IUD does not lose efficacy at higher body weight; the copper IUD is weight-independent.
• Option D: Option D is incorrect because ulipristal acetate does not become completely ineffective above BMI 26 — it is more robust than levonorgestrel at higher BMI and is the preferred oral agent in that range; the claim that levonorgestrel is preferred at high BMI is the opposite of the evidence.
• Option E: Option E is incorrect because the copper IUD is the most reliable, not the least reliable, option at high BMI — its copper ion mechanism is unaffected by pelvic adipose tissue, and there is no interference with copper ion release based on body weight.

ANSWER: D

Rationale:

This question integrates the Pearl Index (failures per 100 woman-years) with the perfect-use versus typical-use distinction to explain why LARCs behave differently from adherence-dependent methods. The gap between perfect-use (approximately 0.3%) and typical-use (approximately 7 to 9%) failure rates for the combined pill is driven almost entirely by adherence errors — missed pills, late pills, vomiting, and inconsistent use — which compromise the ovulation-suppression ceiling that the pill depends on. Long-acting reversible contraceptives (the etonogestrel implant and intrauterine devices) remove adherence from the efficacy equation entirely: once placed, they deliver continuous contraception without requiring any daily or coital action by the user. Because there is no adherence-related failure pathway, their perfect-use and typical-use failure rates converge to nearly the same very low value. This is the pharmacological and behavioral basis for the recommendation that LARCs are preferred whenever consistent adherence is uncertain.

• Option A: Option A is incorrect because LARCs do not have higher perfect-use failure rates than the combined pill — they have very low perfect-use rates, and their typical-use rates are similar to their perfect-use rates because adherence is removed, not because their perfect-use efficacy is poor.
• Option B: Option B is incorrect because the Pearl Index can be and is calculated for LARCs — their perfect-use and typical-use rates are measured, not assumed; the two rates are nearly equal because of the absence of adherence-dependent failure, not because of any inability to compute the index.
• Option C: Option C is incorrect because the combined pill's typical-use rate is higher, not lower, than its perfect-use rate — real-world adherence is worse than trial conditions, which widens rather than narrows the gap; the statement inverts the relationship.
• Option E: Option E is incorrect because LARCs do not require more frequent clinical visits than the combined pill — they require fewer (insertion and removal only) — and the elimination of the perfect-use versus typical-use gap results from the absence of adherence dependence, not from increased monitoring.

ANSWER: E

Rationale:

This question integrates several DMPA features and asks the student to distinguish which arise from depot pharmacokinetics and which from a separate mechanism. Two features arise directly from the depot pharmacokinetics: the robustness against enzyme-inducing drug interactions is conferred by the sustained high medroxyprogesterone acetate (MPA) concentrations released slowly from the intramuscular or subcutaneous depot, which provide a buffer so that even accelerated metabolism is unlikely to drop levels below the contraceptive threshold; and the delayed return to fertility (median 9 to 10 months, occasionally up to 18 months) results from the prolonged tail of MPA release and continued hypothalamic-pituitary-ovarian (HPO) suppression after the last injection. The bone mineral density (BMD) reduction, however, arises from a different mechanism: by suppressing the HPO axis, DMPA lowers endogenous estradiol, and the resulting hypoestrogenic state reduces BMD (a reversible effect that does not increase fracture risk in adults). Recognizing this distinction — depot kinetics for the interaction buffer and fertility delay, hypoestrogenism for the bone effect — reflects integrated understanding of DMPA pharmacology.

• Option A: Option A is incorrect because DMPA contains no copper and does not act through copper ion release — that is the mechanism of the copper IUD; DMPA's features arise from depot progestin pharmacokinetics and HPO suppression.
• Option B: Option B is incorrect because the delayed return to fertility is not caused by permanent ovarian failure — fertility returns in all users, and the delay is a reversible pharmacokinetic phenomenon of the depot, not ovarian failure.
• Option C: Option C is incorrect because these features are not direct toxic effects on ovary, liver, and bone — they reflect depot pharmacokinetics (interaction buffer, fertility delay) and HPO-suppression-mediated hypoestrogenism (bone effect), not direct organ toxicity.
• Option D: Option D is incorrect because it misassigns the mechanisms — the drug-interaction robustness arises from depot pharmacokinetics (not direct hepatic receptor binding), and the bone effect arises from hypoestrogenism secondary to HPO suppression (not direct skeletal receptor binding); only the delayed fertility return is correctly attributed to depot kinetics, making this assignment incorrect overall.

ANSWER: B

Rationale:

This question integrates three concepts — relative versus absolute risk, progestin-generation VTE differences, and the VTE risk of pregnancy — into accurate patient counseling. The relative difference she read about is real: third-generation progestins (such as desogestrel) carry approximately 1.5 to 2 times the VTE risk of levonorgestrel-containing formulations. However, relative risk must be interpreted against the absolute baseline, which is low: the non-pregnant, non-pill baseline VTE rate is approximately 2 per 10,000 woman-years, all modern low-dose combined pills raise this only modestly (an absolute excess on the order of a few extra events per 10,000 woman-years), and even the higher-risk third-generation pills produce a small absolute excess. Critically, this absolute risk remains substantially below the VTE risk of pregnancy itself (approximately 29 per 10,000 pregnancies). The pharmacologically sound counseling is therefore to acknowledge the real relative difference, place it in absolute context, and offer a levonorgestrel-containing pill as the lowest-VTE-risk option without alarm.

• Option A: Option A is incorrect because avoiding all hormonal contraception on the basis of a relative risk doubling ignores the low absolute baseline and the fact that pregnancy — the alternative outcome — carries a higher VTE risk; this counseling would not serve the patient's interests.
• Option C: Option C is incorrect because a doubling of relative risk does not mean VTE occurs in half of users — the absolute rates are on the order of a few per 10,000 woman-years; misinterpreting relative risk as absolute incidence is a fundamental error this question is designed to test.
• Option D: Option D is incorrect because progestin generation is relevant to VTE risk — the 1.5 to 2-fold difference between third- and second-generation progestins is well documented, not fabricated; dismissing it entirely is inaccurate.
• Option E: Option E is incorrect because pregnancy does not carry essentially no VTE risk — pregnancy carries a substantially higher VTE risk (approximately 29 per 10,000 pregnancies) than any modern low-dose combined pill, so remaining off contraception and risking pregnancy is not the lower-VTE-risk path.

ANSWER: A

Rationale:

This question integrates the paradoxical net effect of ritonavir on ethinyl estradiol (EE) with the contraceptive-neutral profile of integrase strand-transfer inhibitors (INSTIs) to guide a regimen-selection decision. Ritonavir is one of the most potent known CYP3A4 inhibitors and is used to boost protease inhibitor levels, yet its net effect on EE is a reduction in exposure (approximately 40 to 50%): ritonavir induces UGT-mediated EE glucuronidation, and this induction of the alternative conjugation pathway outweighs the CYP3A4 inhibition, producing lower EE levels and unreliable combined hormonal contraception. INSTIs (dolutegravir, raltegravir, bictegravir) do not meaningfully alter EE or progestin pharmacokinetics. Therefore, where clinically appropriate from the HIV-treatment standpoint, switching from a ritonavir-boosted protease inhibitor regimen to an INSTI-based regimen would remove the interaction and make combined hormonal contraception reliable. The integrating insight is that the counterintuitive net direction of the ritonavir effect (reduction, not elevation, of EE) drives the recommendation toward a contraceptive-neutral class.

• Option B: Option B is incorrect because ritonavir does not raise ethinyl estradiol levels despite being a CYP3A4 inhibitor — its net effect is a reduction in EE exposure through UGT induction, so the combined pill is rendered less reliable, not more effective.
• Option C: Option C is incorrect because efavirenz does affect ethinyl estradiol — it is a potent CYP3A4 inducer that reduces EE and progestin levels, so switching to efavirenz would not improve contraceptive reliability; the claim that efavirenz has no effect on EE is wrong.
• Option D: Option D is incorrect because adding a second protease inhibitor would not protect ethinyl estradiol — the net EE-lowering effect of ritonavir-boosted regimens operates through UGT induction, and adding more protease inhibitor would not reverse this; the strategy is pharmacologically unsound.
• Option E: Option E is incorrect because antiretroviral regimens do affect hormonal contraception — several classes (ritonavir-boosted protease inhibitors, cobicistat-boosted regimens, efavirenz, nevirapine) reduce EE exposure — so the choice of regimen is far from moot, and her current boosted-protease-inhibitor regimen specifically compromises combined pill reliability.

ANSWER: C

Rationale:

This question integrates the principle that anovulation is the contraceptive ceiling with the concept of adherence-dependence to guide method selection. Both combined and progestin-only pills depend on maintaining adequate drug levels to preserve their contraceptive mechanisms; for the combined pill and the desogestrel progestin-only pill, this means sustaining ovulation suppression, and for the norethindrone minipill it means sustaining continuous cervical mucus impermeability. A patient who predicts frequent missed doses will repeatedly compromise these drug-level-dependent mechanisms — and because anovulation is the contraceptive ceiling, lapses that allow breakthrough ovulation carry the greatest pregnancy risk. A long-acting reversible contraceptive (LARC), such as the etonogestrel implant or an intrauterine device, removes adherence from the efficacy equation entirely, so its perfect-use and typical-use failure rates are nearly identical and very low. The pharmacologically sound approach is to acknowledge her pill preference but counsel that a LARC is superior given her stated adherence pattern.

• Option A: Option A is incorrect because the specific method does matter pharmacologically when adherence is poor — the adherence-dependent methods will underperform their perfect-use efficacy in a patient who forgets doses, while adherence-independent LARCs will not; method choice is not pharmacologically neutral here.
• Option B: Option B is incorrect because the norethindrone 0.35 mg minipill has the least forgiving window of the progestin-only options — a strict 3-hour window — making it a poor choice for someone who forgets pills; it is not the most forgiving option.
• Option D: Option D is incorrect because a higher ethinyl estradiol dose does not eliminate the need for consistent daily dosing — adherence remains essential for any oral method regardless of EE dose, and a higher EE dose would only increase VTE and other estrogen-related risks without solving the adherence problem.
• Option E: Option E is incorrect because typical-use and perfect-use failure rates are not identical for oral contraceptives — the gap between them is driven precisely by adherence errors, so her forgetfulness is highly relevant, and a pill is not as reliable as a long-acting method in her case.