Medroxyprogesterone acetate

What it is

Medroxyprogesterone acetate is a synthetic progestin, a laboratory-made relative of the hormone progesterone. It is used in two quite different ways, and the evidence and the harms differ with them. As a 150 mg intramuscular or 104 mg subcutaneous depot injection given every three months it is a contraceptive. As a 2.5 mg, 5 mg or 10 mg tablet it is used for secondary amenorrhoea, for abnormal uterine bleeding, and to protect the lining of the womb in postmenopausal women taking oestrogen. It is broken down in the liver, mainly by hydroxylation through CYP3A4.

What the research says

The depot injection prevents pregnancy very reliably, with a 12-month failure rate of zero to 0.7 in five company studies, by stopping the hormone surge that releases an egg. The oral tablet's best-evidenced job is protecting the endometrium: in a 3-year randomised trial, 95% of women on oestrogen plus 10 mg cyclic medroxyprogesterone had a normal biopsy versus 38% on oestrogen alone, and a second 1-year trial found 5 mg worked too. The harms are substantial and form two separate stories. The injection is associated with significant bone mineral density loss that is only partly reversible, and that is the subject of its boxed warning; the label's own table shows the femoral neck barely recovering two years after the last injection. The oral tablet given with conjugated oestrogens was the arm of the Women's Health Initiative that was stopped early, after a mean 5.2 of a planned 8.5 years, for excess breast cancer and cardiovascular events, and that is the subject of its different boxed warning. A large randomised trial, ECHO, found no substantial difference in HIV acquisition between the injection, a copper coil and an implant; its confidence interval still allows a modest difference, so it answers the observational signal without excluding a small effect.

Evidence grade: Well established.

How it works

Drug class: Synthetic progestin (progestogen); long-acting injectable contraceptive as the depot suspension, and an oral progestin for endometrial protection and menstrual disorders

Medroxyprogesterone acts like the body's own progesterone on the womb and on the hormone control centre in the brain. Given as a depot injection it damps down the pituitary hormones that ripen and release an egg, and it thickens cervical mucus, so pregnancy is prevented. Taken as a tablet alongside oestrogen it keeps the lining of the womb from building up, which is how it reduces the risk of endometrial overgrowth and cancer in women taking oestrogen. (Source 1)

Boxed warning

Women who use Depo-Provera Contraceptive Injection (Depo-Provera CI) may lose significant bone mineral density. Bone loss is greater with increasing duration of use and may not be completely reversible [see Warnings and Precautions (5.1)].

(Source 2)

What it is used for

• Five company studies gave a 12-month failure rate of zero to 0.7 by life-table method, conditional on returning every 13 weeks. The label itself limits use beyond 2 years because of bone loss. Evidence: established. (Source 3)
• In a 3-year randomised placebo-controlled study, adding 10 mg cyclic medroxyprogesterone to conjugated oestrogen 0.625 mg kept 95% of biopsies normal against 38% on oestrogen alone, and eliminated atypia. Evidence: established. (Source 4)
• Both are labelled indications in the absence of organic pathology. No outcome trial for either is quoted on the label read here and none was found in the searches recorded in notes, so the strength of the evidence for these two uses is not established from the sources in this write-up. Evidence: unknown. (Source 5)
• The subcutaneous 104 mg product's boxed warning refers to its use both as birth control and as medical therapy for endometriosis-associated pain, and limits that use to 2 years unless other options are inadequate. No efficacy trial for this indication was reached in this run. Evidence: limited. (Source 6)
• The tablet is not licensed for treating menopausal symptoms; its labelled menopausal role is endometrial protection. Cochrane found combined continuous hormone therapy raises coronary events, venous thromboembolism, stroke, breast cancer and gallbladder disease in relatively healthy postmenopausal women, with fracture reduction the only outcome showing strong evidence of clinical benefit. Evidence: disputed. (Source 7)
• Not a separate indication but a question that mattered. The ECHO randomised trial of 7,829 African women found HIV hazard ratios of 1.04 for the injection versus a copper coil and 1.23 versus a levonorgestrel implant, with confidence intervals including no difference. This is a null result and it supports continued access. Evidence: established. (Source 8)

Interactions

• St John's wort (Hypericum perforatum) (label): St John's wort induces the enzymes that break down contraceptive hormones, so it can lower the hormone level and reduce contraceptive effectiveness. The label names it alongside the enzyme-inducing drugs and tells the prescriber to counsel the woman to use additional or different contraception. (Source 9)
• Enzyme-inducing medicines: barbiturates, bosentan, carbamazepine, felbamate, griseofulvin, oxcarbazepine, phenytoin, rifampin, topiramate (label): All of these speed up the breakdown of contraceptive hormones and may make the injection less reliable. (Source 9)
• HIV protease inhibitors and non-nucleoside reverse transcriptase inhibitors (pharmacokinetic study): Progestin blood levels have gone both up and down when these are taken together, so the direction of the interaction is not predictable. (Source 9)
• Antibiotics (pharmacokinetic study): This is the interaction that did not hold up. Pregnancies have been reported on hormonal contraceptives and antibiotics together, but pharmacokinetic studies have not shown a consistent effect of antibiotics on the hormone levels. (Source 10)
• Food, for the oral tablet (pharmacokinetic study): Taking the tablet with a meal raises how much medroxyprogesterone gets into the blood: peak level by 50 to 70% and total exposure by 18 to 33%. The half-life does not change. This is a real pharmacokinetic effect, so taking the tablet consistently with or without food matters more than which one is chosen. (Source 11)
• CYP3A4 inducers and inhibitors, for the oral tablet (theoretical): Oral medroxyprogesterone is broken down mainly by CYP3A4, so drugs that speed up or slow down that enzyme can change its level. The label is candid that no specific interaction study has been done. (Source 12)
• Alcohol and tobacco, in relation to bone (label): No pharmacokinetic interaction with alcohol is reported on either label read here. What the label does say is that chronic alcohol or tobacco use is one of the osteoporosis risk factors that makes the injection's bone effect more of a concern, and in adolescents smoking was associated with less bone recovery after stopping. (Source 13)
• Anticonvulsants and corticosteroids, in relation to bone (label): These are named on the label as drugs that can reduce bone mass, so taking them alongside the depot injection compounds the bone concern that the boxed warning is about. (Source 13)

Stopping it

• Bone density recovers only partly after the depot injection is stopped, and recovery is less complete the longer it was used. In adults there was partial recovery toward baseline over the two years after the last injection. (Source 14)
• In adolescents treated for more than two years, hip and femoral neck density had still not returned to baseline up to five years after the last injection, and smoking made recovery worse. (Source 15)
• A 2025 review of recent evidence takes a more hopeful view, finding that bone health can be partly or completely restored after stopping the injection, while saying this has not been adequately researched. (Source 16)
• Fertility takes time to come back after the injection stops, with a median of 10 months to conception and a range of 4 to 31 months, independent of how long it was used. There is no way to speed this up. (Source 17)
• For the oral tablet used with oestrogen, the label's instruction is periodic re-evaluation, for example every 3 to 6 months, to see whether treatment is still needed, and the lowest effective dose has not been determined. (Source 18)
• Stopping oral medroxyprogesterone given for secondary amenorrhoea characteristically produces a withdrawal bleed within three to seven days. That is the intended effect rather than a withdrawal syndrome. (Source 5)
• The excess breast cancer risk from oestrogen plus medroxyprogesterone did not disappear when the drug was stopped: it was still present in the randomised groups more than 20 years after the intervention ended. (Source 19)
• Where thrombosis occurs on the injection, the label's instruction is to stop rather than reduce, unless there is no other acceptable contraceptive option. (Source 20)

What goes wrong

The depot injection carries a boxed warning for loss of bone mineral density, which may not be completely reversible. (Source 2)

• Official position, Certainty not rated.
• Size: Not applicable.
• Who: Women using the 150 mg intramuscular depot.
• How long: Risk rises with duration of use.
• Result: The regulator-approved warning text; the label also says use beyond 2 years is not recommended unless other options are considered inadequate.
• Funding: industry-funded (sponsor's label)

Women who use Depo-Provera Contraceptive Injection (Depo-Provera CI) may lose significant bone mineral density. Bone loss is greater with increasing duration of use and may not be completely reversible [see Warnings and Precautions (5.1)]. • It is unknown if use of Depo-Provera CI during adolescence or early adulthood, a critical period of bone accretion, will reduce peak bone mass and increase the risk for osteoporotic fracture in later life [see Warnings and Precautions (5.1)]. • Depo-Provera CI is not recommended as a long-term (i.e., longer than 2 years) birth control method unless other options are considered inadequate [see Indications and Usage (1) and Warnings and Precautions (5.1)].

In adult women, five years of the depot injection reduced lumbar spine bone mineral density by about 5%, with only partial recovery two years after stopping. (Source 14)

• Randomized trial, Moderate certainty.
• Size: Controlled clinical study; 33 treated and 105 control women contributed 5-year spine data, falling to 12 and 60 at 7 years.
• Who: Adult women using Depo-Provera CI 150 mg, compared with women using no hormonal contraception.
• How long: 5 years of treatment plus 2 years of follow-up.
• Result: Mean lumbar spine BMD change -2.86%, -4.11%, -4.89%, -4.93% and -5.38% after 1 to 5 years; spine and hip mean decreases of 5-6% versus no significant change in controls; partial recovery toward baseline over the 2-year post-therapy period, less complete with longer treatment.
• Funding: industry-funded (sponsor's label)

Limit of this finding: The phrase 'partial recovery toward baseline' is doing a lot of work, and the label's own Table 4, carried at the end of this block, shows what it means. Two years after the last injection the spine had gone from -5.38% to -3.13% and the total hip from -5.16% to -1.34%, but the femoral neck had barely moved, from -6.12% to -5.38%. The follow-up numbers are also very small, 12, 7 and 13 women, so those recovery figures are fragile.

In a controlled, clinical study, adult women using Depo-Provera CI (150mg) for up to 5 years showed spine and hip bone mineral density (BMD) mean decreases of 5–6%, compared to no significant change in BMD in the control group. The decline in BMD was more pronounced during the first two years of use, with smaller declines in subsequent years. Mean changes in lumbar spine BMD of ‑2.86%, ‑4.11%, ‑4.89%, ‑4.93% and ‑5.38% after 1, 2, 3, 4, and 5 years, respectively, were observed. Mean decreases in BMD of the total hip and femoral neck were similar. After stopping use of Depo-Provera CI, there was partial recovery of BMD toward baseline values during the 2-year post-therapy period. Longer duration of treatment was associated with less complete recovery during this 2-year period following the last injection.

In adolescent girls the hip bone loss was larger, reaching -6.4% at the total hip by 4.6 years, and the comparison cohort was not matched at baseline. (Source 21)

• Cohort study, Low certainty.
• Size: 389 adolescent females aged 12 to 18 in an open-label non-randomised study; 28 contributed week-240 total hip data.
• Who: Adolescent females aged 12 to 18 using Depo-Provera CI 150 mg, versus an unmatched untreated cohort.
• How long: Up to 240 weeks (4.6 years), with dosing stopped at 120 weeks.
• Result: Mean BMD decrease at 240 weeks -6.4% at total hip, -5.4% at femoral neck, -2.1% at lumbar spine; mean 9.3 injections per user; the untreated cohort gained BMD during growth after menarche but was not matched for age, gynaecologic age, race or baseline BMD.
• Funding: industry-funded (sponsor's label)

Limit of this finding: The comparison group here was not matched to the treated group at the start, which the label states itself, so this is an observational comparison and not a randomised one. Table 5, carried in this block, shows the size of the gap: at Week 120 total hip density had fallen 5.40% in treated adolescents while rising 2.19% in the untreated cohort, and by Week 240 only 28 treated adolescents remained.

The mean decrease in BMD at 240 weeks was more pronounced at total hip (-6.4%) and femoral neck (-5.4%) compared to lumbar spine (-2.1%). Adolescents in the untreated cohort had an increase in BMD during the period of growth following menarche. However, the two cohorts were not matched at baseline for age, gynecologic age, race, BMD and other factors that influence the rate of acquisition of BMD.

In adolescents treated for more than two years, hip and femoral neck bone density had still not returned to baseline five years after the last injection. (Source 15)

• Cohort study, Low certainty.
• Size: 49 adolescents in each duration group at end of treatment, falling to 2 to 3 at 60 months post-treatment.
• Who: Adolescents who used Depo-Provera CI for 2 years or less versus more than 2 years.
• How long: Up to 60 months after the last injection.
• Result: Only lumbar spine BMD recovered to baseline in those treated more than two years; femoral neck and total hip did not recover even up to 60 months post-treatment. Longer treatment and smoking were associated with less recovery. The post-treatment cell sizes are very small, which is a serious limitation.
• Funding: industry-funded (sponsor's label)

Limit of this finding: Table 6, in the sibling block, puts numbers to the non-recovery: among adolescents treated for more than two years the total hip was -6.2% at the end of treatment and still -4.6% at 12 months, -3.6% at 24 months, -4.6% at 36 months and -2.5% at 48 months afterwards, on numbers falling to single figures. Only the lumbar spine came back to baseline.

Post‑treatment follow-up showed that, in women treated for more than two years, only lumbar spine BMD recovered to baseline levels after treatment was discontinued. Adolescents treated with Depo-Provera CI for more than two years did not recover to their baseline BMD level at femoral neck and total hip even up to 60 months post-treatment.

Post-marketing reports for the depot injection include osteoporosis and osteoporotic fractures, as well as deep vein thrombosis and pulmonary embolus, with no reliable frequency. (Source 22)

• Case series, Very low certainty.
• Size: Spontaneous reports from a population of uncertain size.
• Who: Depot medroxyprogesterone users after approval.
• How long: Not applicable.
• Result: No rates can be estimated; the label says so explicitly.
• Funding: industry-funded (sponsor's label)

Limit of this finding: Post-marketing reports have no denominator: the label says plainly that it cannot reliably estimate their frequency or establish that the drug caused them. The two footnote markers printed in the table header, an asterisk and a dagger, are the label's own; the footnotes they point to are carried in the sibling block, and the dagger one is an instruction about aseptic injection technique rather than a side effect.

There have been cases of osteoporosis including osteoporotic fractures reported post-marketing in patients taking Depo-Provera CI. Table 3. Adverse Reactions Reported during Post-Marketing Experience Body System* Adverse Reactions Body as a Whole Chest pain, Allergic reactions including angioedema, Fever, Injection site abscess†, Injection site infection†, Injection site nodule/lump, Injection site pain/tenderness, Injection site persistent atrophy/indentation/dimpling, Injection-site reaction, Lipodystrophy acquired, Chills, Axillary swelling Cardiovascular Syncope, Tachycardia, Thrombophlebitis, Deep vein thrombosis, Pulmonary embolus, Varicose veins

Weight gain on the depot injection averaged 5.4 lb at one year and 16.5 lb at six years in the sponsor's data, and 2% of women left a large trial because of it. (Source 23)

• Cohort study, Low certainty.
• Size: Not stated; initial average body weight 136 lb.
• Who: Women completing 1, 2, 4 and 6 years of therapy.
• How long: Up to 6 years.
• Result: Average gain 5.4 lb at 1 year, 8.1 lb at 2 years, 13.8 lb at 4 years and 16.5 lb at 6 years; 2% withdrew from a large-scale clinical trial because of excessive weight gain. These are completers' averages with no control group, which overstates attributable gain.
• Funding: industry-funded (sponsor's label)

From an initial average body weight of 136 lb, women who completed 1 year of therapy with Depo-Provera CI gained an average of 5.4 lb. Women who completed 2 years of therapy gained an average of 8.1 lb. Women who completed 4 years gained an average of 13.8 lb. Women who completed 6 years gained an average of 16.5 lb. Two percent of women withdrew from a large-scale clinical trial because of excessive weight gain.

Most women on the depot injection have their periods disrupted, and after two years about two thirds have no periods at all. (Source 24)

• Cohort study, Moderate certainty.
• Size: Over 3,900 women in the two clinical trials.
• Who: Women receiving 150 mg Depo-Provera CI every 3 months, aged 15 to 51.
• How long: Median 13 months, range 1 to 84 months.
• Result: Bleeding reported by 57.3% at 12 months and 32.1% at 24 months; amenorrhoea by 55% at 12 months and 68% at 24 months; headache 16.5%, abdominal pain or discomfort 11.2%, weight increase of more than 10 lb at 24 months 37.7%, nervousness 10.8%, dizziness 5.6%, decreased libido 5.5%.
• Funding: industry-funded (sponsor's label)

Limit of this finding: These percentages come from the sponsor's own two trials and have no placebo arm, so they are what happened to women on the injection, not what the injection added. The label's own preamble, carried at the head of this block, says rates from one drug's trials cannot be compared with another's and may not match practice. The label's race distribution for these trials adds up to 100.9% ('46% were White, 50% Non-White, and 4.9% Unknown race'); that is the label's arithmetic, quoted as printed, and it means the breakdown cannot be relied on. The heading 'Reproductive (Urogenital )' with its stray space and the COSTART footnote running into the cell text are the label's own.

Increased weight >10 lb at 24 months (37.7%) Nervous Nervousness (10.8%) Dizziness (5.6%) Libido decreased (5.5%) Reproductive (Urogenital*) Menstrual irregularities: bleeding (57.3% at 12 months, 32.1% at 24 months) amenorrhea (55% at 12 months, 68% at 24 months) * Body System represented from COSTART medical dictionary.

Bleeding pattern changes affect most users of the depot injection, and amenorrhoea becomes more common with time rather than less. (Source 25)

• Cohort study, Moderate certainty.
• Size: Clinical studies of Depo-Provera CI; the label does not restate the denominator in this section.
• Who: Women using the 150 mg intramuscular depot.
• How long: Up to 24 months.
• Result: Amenorrhoea reported by 55% of women by month 12 and 68% by month 24; altered patterns include irregular or unpredictable bleeding or spotting, prolonged spotting or bleeding, and heavy bleeding.
• Funding: industry-funded (sponsor's label)

As women continue using Depo-Provera CI, fewer experience irregular bleeding and more experience amenorrhea. In clinical studies of Depo-Provera CI, by month 12 amenorrhea was reported by 55% of women, and by month 24, amenorrhea was reported by 68% of women using Depo-Provera CI.

The depot injection's own indication section carries a limitation of use: it is not recommended as a long-term method beyond 2 years unless other options are inadequate. (Source 26)

• Official position, Certainty not rated.
• Size: Not applicable.
• Who: Females of reproductive potential using the injection for contraception.
• How long: Beyond 2 years.
• Result: A duration limit written into the indication itself, cross-referenced to the bone mineral density warning.
• Funding: industry-funded (sponsor's label)

Limitations of Use: The use of Depo-Provera CI is not recommended as a long-term (i.e., longer than 2 years) birth control method unless other options are considered inadequate [see Dosage and Administration (2.1) and Warnings and Precautions (5.1)].

Return of fertility after the depot injection is delayed: in a large US study the median time to conception was 10 months from the last injection. (Source 17)

• Cohort study, Low certainty.
• Size: 188 women who discontinued to become pregnant; 114 conceived; data available for 61% of those who stopped to conceive.
• Who: US women who stopped Depo-Provera CI in order to become pregnant.
• How long: Up to 18 months and beyond.
• Result: Of those who do conceive, 68% within 12 months, 83% within 15 months and 93% within 18 months of the last injection; median 10 months, range 4 to 31 months, unrelated to duration of use. No data for 39% of those who stopped to conceive, which is a large loss to follow-up.
• Funding: industry-funded (sponsor's label)

Return to ovulation and fertility is likely to be delayed after stopping Depo-Provera CI. In a large US study of women who discontinued use of Depo-Provera CI to become pregnant, data are available for 61% of them. Of the 188 women who discontinued the study to become pregnant, 114 became pregnant. Based on Life-Table analysis of these data, it is expected that 68% of women who do become pregnant may conceive within 12 months, 83% may conceive within 15 months, and 93% may conceive within 18 months from the last injection. The median time to conception for those who do conceive is 10 months following the last injection with a range of 4 to 31 months, and is unrelated to the duration of use.

Case-control evidence on breast cancer and the depot injection is mixed, with a statistically significant excess in recent users of 12 months or longer in one recent US study. (Source 27)

• Case-control study, Low certainty.
• Size: Five large case-control studies summarised on the label.
• Who: Depot medroxyprogesterone users.
• How long: Varies by study.
• Result: Three of the five studies suggest a slightly increased risk overall, statistically significant in one. One US study found a statistically significant increase in recent users (last use within the past five years) who used depot for 12 months or longer. On the SEER-18 2011 rates quoted, a doubling of risk would move incidence from about 72 to about 144 cases per 100,000 women aged 20 to 49.
• Funding: industry-funded (summarised in the sponsor's label)

Limit of this finding: This is a summary of five case-control studies, which are observational: they can show an association, not that the injection causes breast cancer. Three of the five suggested a slightly raised risk and only one of those reached statistical significance. The label prints 'study1' and 'study2' with the figures as superscript reference markers pointing to the studies listed under its Figure 1; they are carried here as the render produces them, which is why the digits appear pushed against the preceding word. 'study1' is Li and colleagues (2012) and 'study2' the earlier study, both named in the figure's covariate list.

The results of five large case-control studies assessing the association between depo-medroxyprogesterone acetate (DMPA) use and the risk of breast cancer are summarized in Figure 1. Three of the studies suggest a slightly increased risk of breast cancer in the overall population of users; these increased risks were statistically significant in one study. One recent US study1 evaluated the recency and duration of use and found a statistically significantly increased risk of breast cancer in recent users (defined as last use within the past five years) who used DMPA for 12 months or longer; this is consistent with results of a previous study2.

The depot injection is contraindicated in current or past thromboembolic disease, known or suspected breast cancer, significant liver disease and undiagnosed vaginal bleeding. (Source 20)

• Official position, Certainty not rated.
• Size: Not applicable.
• Who: Women considered for the depot injection.
• How long: Not applicable.
• Result: Absolute contraindications on the label; no rates given.
• Funding: industry-funded (sponsor's label)

Active thrombophlebitis, or current or history of thromboembolic disorders, or cerebral vascular disease [see Warnings and Precautions (5.2)]. • Known or suspected malignancy of breast [see Warnings and Precautions (5.3)]. • Known hypersensitivity to Depo-Provera CI (medroxyprogesterone acetate) or any of its other ingredients [see Warnings and Precautions (5.6)]. • Significant liver disease [see Warnings and Precautions (5.8)]. • Undiagnosed vaginal bleeding [see Warnings and Precautions (5.11)].

The oral tablet taken with conjugated oestrogens carries a different boxed warning: cardiovascular disorders, breast cancer and probable dementia. (Source 28)

• Official position, Certainty not rated.
• Size: Not applicable.
• Who: Postmenopausal women taking oestrogen plus progestin.
• How long: Not applicable.
• Result: The regulator-approved warning text; it states estrogen plus progestin therapy should not be used to prevent cardiovascular disease or dementia.
• Funding: industry-funded (sponsor's label)

Estrogen plus progestin therapy should not be used for the prevention of cardiovascular disease or dementia. (See CLINICAL STUDIES and WARNINGS, Cardiovascular Disorders and Probable Dementia.) The Women's Health Initiative (WHI) estrogen plus progestin substudy reported an increased risk of deep vein thrombosis (DVT), pulmonary embolism (PE), stroke and myocardial infarction (MI) in postmenopausal women (50 to 79 years of age) during 5.6 years of treatment with daily oral conjugated estrogens (CE) [0.625 mg] combined with medroxyprogesterone acetate (MPA) [2.5 mg], relative to placebo.

The Women's Health Initiative estrogen-plus-progestin trial was STOPPED EARLY, after a mean 5.2 of a planned 8.5 years, because invasive breast cancer crossed the stopping boundary and the global index showed risks exceeding benefits; coronary events, stroke, pulmonary embolism and breast cancer were all raised. (Source 29)

• Randomized trial, High certainty.
• Size: 16,608 postmenopausal women with an intact uterus: 8,506 on conjugated equine oestrogens 0.625 mg/d plus medroxyprogesterone 2.5 mg/d, 8,102 on placebo, 40 US centres.
• Who: Postmenopausal women aged 50 to 79 at recruitment, mean age 63.
• How long: Mean 5.2 years of follow-up against a planned duration of 8.5 years; the trial was stopped on 31 May 2002 and this report runs to 30 April 2002.
• Result: Hazard ratios: CHD 1.29 (1.02-1.63); breast cancer 1.26 (1.00-1.59); stroke 1.41 (1.07-1.85); PE 2.13 (1.39-3.25); colorectal cancer 0.63 (0.43-0.92); hip fracture 0.66 (0.45-0.98). Absolute excess per 10,000 person-years: 7 more CHD events, 8 more strokes, 8 more PEs, 8 more invasive breast cancers; absolute reductions 6 fewer colorectal cancers, 5 fewer hip fractures; global index excess 19 per 10,000 person-years.
• Funding: independent (Research Support, U.S. Gov't, P.H.S. per PubMed; National Institutes of Health)

Limit of this finding: These hazard ratios are not the result of a completed trial. The trial was stopped on 31 May 2002, after a mean of 5.2 years of a planned 8.5, because the breast cancer result crossed a pre-set stopping boundary. A trial stopped at the moment a harm signal is strongest tends to show that harm at its largest, so the figures should be read as the point at which the monitoring board judged the risks to outweigh the benefits, not as a stable long-run estimate. The absolute numbers are the ones to hold on to: about 7 to 8 extra events per 10,000 women per year for each of the harms.

Absolute excess risks per 10 000 person-years attributable to estrogen plus progestin were 7 more CHD events, 8 more strokes, 8 more PEs, and 8 more invasive breast cancers, while absolute risk reductions per 10 000 person-years were 6 fewer colorectal cancers and 5 fewer hip fractures. The absolute excess risk of events included in the global index was 19 per 10 000 person-years.

The trial's own conclusion was that overall health risks exceeded benefits and that the regimen should not be used for primary prevention. (Source 30)

• Randomized trial, High certainty.
• Size: Same 16,608 women.
• Who: Healthy postmenopausal US women.
• How long: Mean 5.2 years.
• Result: All-cause mortality was not affected during the trial; the authors state the risk-benefit profile is not consistent with a viable primary-prevention intervention.
• Funding: independent (National Institutes of Health)

Limit of this finding: These hazard ratios are not the result of a completed trial. The trial was stopped on 31 May 2002, after a mean of 5.2 years of a planned 8.5, because the breast cancer result crossed a pre-set stopping boundary. A trial stopped at the moment a harm signal is strongest tends to show that harm at its largest, so the figures should be read as the point at which the monitoring board judged the risks to outweigh the benefits, not as a stable long-run estimate. The absolute numbers are the ones to hold on to: about 7 to 8 extra events per 10,000 women per year for each of the harms.

Overall health risks exceeded benefits from use of combined estrogen plus progestin for an average 5.2-year follow-up among healthy postmenopausal US women. All-cause mortality was not affected during the trial. The risk-benefit profile found in this trial is not consistent with the requirements for a viable intervention for primary prevention of chronic diseases, and the results indicate that this regimen should not be initiated or continued for primary prevention of CHD.

The label's version of the WHI table gives the absolute risks side by side, including a venous thromboembolism signal that was not part of the global index. (Source 31)

• Randomized trial, High certainty.
• Size: 16,608 women, 8,506 on CE plus MPA and 8,102 on placebo.
• Who: Postmenopausal women, average 63 years.
• How long: Average 5.6 years of centrally adjudicated follow-up.
• Result: Absolute excess per 10,000 women-years: 7 more CHD events, 8 more strokes, 10 more PEs, 8 more invasive breast cancers; reductions of 6 colorectal cancers and 5 hip fractures; global index excess 19 per 10,000 women-years.
• Funding: independent trial, reported in the sponsor's label.

For those outcomes included in the WHI "global index" that reached statistical significance after 5.6 years of follow-up, the absolute excess risks per 10,000 women-years in the group treated with CE plus MPA were 7 more CHD events, 8 more strokes, 10 more PEs, and 8 more invasive breast cancers, while the absolute risk reduction per 10,000 women-years were 6 fewer colorectal cancers and 5 fewer hip fractures.

Venous thromboembolism roughly doubled on oestrogen plus medroxyprogesterone, from 17 to 35 events per 10,000 women-years, appearing in the first year and persisting. (Source 32)

• Randomized trial, High certainty.
• Size: The WHI estrogen plus progestin substudy.
• Who: Postmenopausal women on daily CE 0.625 mg plus MPA 2.5 mg versus placebo.
• How long: Average 5.6 years.
• Result: VTE 35 versus 17 per 10,000 women-years; deep vein thrombosis 26 versus 13; pulmonary embolism 18 versus 8.
• Funding: independent trial, reported in the sponsor's label.

In the WHI estrogen plus progestin substudy, a statistically significant 2-fold greater rate of VTE (DVT and PE) was reported in women receiving daily CE (0.625 mg) plus MPA (2.5 mg) compared to women receiving placebo (35 versus 17 per 10,000 women-years). Statistically significant increases in risk for both DVT (26 versus 13 per 10,000 women-years) and PE (18 versus 8 per 10,000 women-years) were also demonstrated. The increase in VTE risk was demonstrated during the first year and persisted.

The invasive breast cancer excess was concentrated in women who had used hormone therapy before, where the absolute risk rose from 25 to 46 cases per 10,000 women-years. (Source 33)

• Randomized trial, High certainty.
• Size: The WHI estrogen plus progestin substudy; 26% reported prior hormone use.
• Who: Postmenopausal women on CE plus MPA versus placebo.
• How long: Mean 5.6 years.
• Result: Overall relative risk 1.24, absolute risk 41 versus 33 cases per 10,000 women-years. With prior hormone use: relative risk 1.86, absolute 46 versus 25. With no prior use: relative risk 1.09, absolute 40 versus 36. Cancers in the treated group were larger, more often node positive and diagnosed at a more advanced stage.
• Funding: independent trial, reported in the sponsor's label.

Among women who reported prior use of hormone therapy, the relative risk of invasive breast cancer was 1.86, and the absolute risk was 46 versus 25 cases per 10,000 women-years, for CE plus MPA compared with placebo.

Followed for more than 20 years after interventions lasting a median of only 5.6 and 7.2 years, oestrogen plus medroxyprogesterone remained associated with higher breast cancer incidence, while oestrogen alone was associated with lower incidence and lower breast cancer mortality. (Source 19)

• Randomized trial, High certainty.
• Size: 27,347 postmenopausal women across both WHI trials: 16,608 with a uterus and 10,739 with prior hysterectomy.
• Who: Postmenopausal women aged 50 to 79 at entry, baseline mean age 63.4 years, no prior breast cancer and a negative baseline mammogram.
• How long: Intervention 5.6 years (CE plus MPA) and 7.2 years (CE alone); more than 20 years median cumulative follow-up to 31 December 2017.
• Result: CE plus MPA versus placebo: 584 versus 447 breast cancers, annualised 0.45% versus 0.36%, HR 1.28 (95% CI 1.13-1.45; P < .001), and no significant difference in breast cancer mortality, HR 1.35 (0.94-1.95; P = .11). CE alone versus placebo: HR 0.78 (0.65-0.93; P = .005) for incidence and 0.60 (0.37-0.97; P = .04) for mortality.
• Funding: independent (Research Support, U.S. Gov't, P.H.S. per PubMed)

Limit of this finding: 'More than 20 years of median cumulative follow-up' is the length of the watching, not the length of the treatment. The combined-therapy trial was stopped after a median 5.6 years of treatment and the oestrogen-only trial after 7.2 years; the rest of the time was observation after the drug had stopped. The oestrogen-alone result applies only to women who had already had a hysterectomy and were randomised to oestrogen on its own, and it is not a reason to take oestrogen: it is one arm of one pair of trials. The source consistently says 'was associated with', not 'caused'. The numbers 27 347, 10 739 and 16 608 are printed in the paper with narrow no-break spaces and 'P= .11' with the space after the equals sign; both are the journal's own typography.

In contrast, CEE plus MPA compared with placebo among 16 608 women with a uterus was associated with statistically significantly higher breast cancer incidence with 584 cases (annualized rate, 0.45%) vs 447 cases (annualized rate, 0.36%; HR, 1.28; 95% CI, 1.13-1.45; P < .001) and no significant difference in breast cancer mortality with 71 deaths

In women aged 65 and over, oestrogen plus medroxyprogesterone doubled the rate of probable dementia, from 33 to 45 cases per 10,000 women-years. (Source 34)

• Randomized trial, Moderate certainty.
• Size: 4,532 predominantly healthy postmenopausal women aged 65 and older in the WHI Memory Study.
• Who: Postmenopausal women aged 65 to 79 taking daily CE 0.625 mg plus MPA 2.5 mg versus placebo.
• How long: Average 4 years.
• Result: Relative risk of probable dementia 2.05 (95% CI 1.21-3.48); absolute risk 45 versus 33 per 10,000 women-years. Whether this applies to younger postmenopausal women is unknown.
• Funding: independent trial, reported in the sponsor's label.

Limit of this finding: The PROVERA label contradicts itself on this figure. Its CLINICAL STUDIES section, quoted here, gives the absolute risk of probable dementia as 45 versus 33 per 10,000 women-years. Its WARNINGS section gives 45 versus 22 per 10,000 women-years for the same comparison, and adds raw counts of 40 women on the combination and 21 on placebo. Both figures are printed in the same label, and this entry quotes the clinical studies section accurately; neither has been 'corrected' to match the other. A reader should take the direction of the finding, which both sections agree on, and treat the exact absolute numbers as uncertain. The study was done only in women aged 65 to 79, and the label itself says it is unknown whether the finding applies to younger postmenopausal women. The PROVERA label is quoted exactly as it prints, including its own typographic slips: 'Geriatric Use .)', 'Breast Cancer . )', '(see WARNINGS )' with a space before the bracket, 'metabolized in-vitro', 'via the CYP3A4', 'Reproductive (Urogenital )' and, in its Table 4 heading, '(95%nCI'. None of these is a transcription error.

After an average follow-up of 4 years, the relative risk of probable dementia for CE plus MPA versus placebo was 2.05 (95 percent CI, 1.21–3.48). The absolute risk of probable dementia for CE plus MPA versus placebo was 45 versus 33 per 10,000 women-years. Probable dementia as defined in this study included Alzheimer's disease (AD), vascular dementia (VaD) and mixed type (having features of both AD and VaD). The most common classification of probable dementia in the treatment group and the placebo group was AD. Since the ancillary study was conducted in women 65 to 79 years of age, it is unknown whether these findings apply to younger postmenopausal women.

A Cochrane review puts the harms of combined continuous hormone therapy in absolute terms per 1000 women, across coronary events, clots, stroke, breast cancer, gallbladder disease and lung cancer death. (Source 7)

• Systematic review, Moderate certainty.
• Size: 22 studies, 43,637 women; nearly 70% of the data from HERS 1998 and WHI 1998.
• Who: Mostly postmenopausal American women with some comorbidity, mean age over 60; no study focused on perimenopausal women.
• How long: At least 1 year of hormone therapy; follow-up to 5.6 years plus extensions.
• Result: Combined continuous hormone therapy raised coronary events after 1 year from 2 per 1000 to between 3 and 7 per 1000; venous thromboembolism from 2 to between 4 and 11 per 1000; stroke after 3 years from 6 to between 6 and 12 per 1000; breast cancer after 5.6 years from 19 to between 20 and 30 per 1000; gallbladder disease from 27 to between 38 and 60 per 1000; lung cancer death from 5 to between 6 and 13 per 1000.
• Funding: independent (Research Support, Non-U.S. Gov't per PubMed)

Limit of this finding: These figures are for combined continuous hormone therapy, the kind that includes medroxyprogesterone, and they are not the whole of the review's result. The same paragraph reports the opposite direction for oestrogen-only therapy, which REDUCED breast cancer and clinical fracture and did not raise coronary events at any follow-up time; that sentence is carried in this block so the two can be read together. The evidence base is also narrower than the review's title suggests: nearly 70% of the data came from two trials, most participants were American women over 60 with some existing illness, and no included study focused on perimenopausal women. This is version pub5 of the review; two later versions exist and were not read here.

combined continuous HT increased the risk of a coronary event (after 1 year's use: from 2 per 1000 to between 3 and 7 per 1000), venous thromboembolism (after 1 year's use: from 2 per 1000 to between 4 and 11 per 1000), stroke (after 3 years' use: from 6 per 1000 to between 6 and 12 per 1000), breast cancer (after 5.6 years' use: from 19 per 1000 to between 20 and 30 per 1000), gallbladder disease (after 5.6 years' use: from 27 per 1000 to between 38 and 60 per 1000) and death from lung cancer (after 5.6 years' use plus 2.4 years' additional follow-up: from 5 per 1000 to between 6 and 13 per 1000).

The breast cancers that appeared on oestrogen plus medroxyprogesterone were larger, more often node positive and at a more advanced stage than those in the placebo group. (Source 35)

• Randomized trial, High certainty.
• Size: The WHI estrogen plus progestin substudy, 16,608 women.
• Who: Postmenopausal women taking daily conjugated oestrogens 0.625 mg plus medroxyprogesterone 2.5 mg versus placebo.
• How long: Mean 5.6 years of treatment.
• Result: Tumours larger, more likely to be node positive and diagnosed at a more advanced stage in the combination group; metastatic disease rare with no apparent difference; histological subtype, grade and hormone receptor status did not differ. The label's instruction is yearly breast examinations by a healthcare provider, monthly self-examination, and mammography scheduled on age, risk factors and prior results.
• Funding: independent trial, reported in the sponsor's label.

Limit of this finding: This is about the character of the cancers found, not about how many. It matters because a bigger, node-positive tumour is harder to treat than a small one, so the harm is not only the extra diagnoses. The accompanying observational and meta-analytic statements in the same subsection are not randomised evidence. The PROVERA label is quoted exactly as it prints, including its own typographic slips: 'Geriatric Use .)', 'Breast Cancer . )', '(see WARNINGS )' with a space before the bracket, 'metabolized in-vitro', 'via the CYP3A4', 'Reproductive (Urogenital )' and, in its Table 4 heading, '(95%nCI'. None of these is a transcription error.

In the same substudy, invasive breast cancers were larger, were more likely to be node positive, and were diagnosed at a more advanced stage in the CE (0.625 mg) plus MPA (2.5 mg) group compared with the placebo group.

The only absolute figure the depot label gives for breast cancer is that a doubling of risk would move incidence from about 72 to about 144 cases per 100,000 women aged 20 to 49. (Source 36)

• Case-control study, Low certainty.
• Size: Based on the published SEER-18 2011 incidence rate for US women, all races, aged 20 to 49 years.
• Who: Women aged 20 to 49 using Depo-Provera CI.
• How long: Not applicable.
• Result: About 72 rising to about 144 cases per 100,000 women if risk doubled. Cervical cancer: a statistically non-significant rise, RR 1.22 to 1.28 with 95% CI 0.93 to 1.70 for first exposure before age 35, and RR 1.11 (95% CI 0.96 to 1.29) for ever-use. No overall increased risk of ovarian or liver cancer.
• Funding: industry-funded (summarised in the sponsor's label)

Limit of this finding: The 72 to 144 figure is a worked example, not a measured result: it shows what a doubling of risk would mean against a background rate, and the case-control studies did not find a doubling. The cervical cancer intervals both cross 1, which means no effect was demonstrated, and the label calls them non-significant. Reading these numbers as the risk of using the injection would overstate them.

Based on the published SEER-18 2011 incidence rate (age-adjusted to the 2000 US Standard Population) of breast cancer for US women, all races, age 20 to 49 years, a doubling of risk would increase the incidence of breast cancer in women who use Depo-Provera CI from about 72 to about 144 cases per 100,000 women.

Between 1 and 5 percent of women in the depot injection's trials reported fatigue, backache, nausea, leg cramps, depression and other effects, and bleeding was the commonest reason for leaving a study. (Source 37)

• Cohort study, Moderate certainty.
• Size: The same two clinical trials, over 3,900 women treated for up to 7 years.
• Who: Women receiving 150 mg Depo-Provera CI every 3 months.
• How long: Up to 7 years.
• Result: Asthenia or fatigue 4.2%, leg cramps 3.7%, nausea 3.3%, leukorrhoea 2.9%, breast pain 2.8%, bloating 2.3%, backache 2.2%, oedema 2.2%, dysmenorrhoea 1.7%, depression 1.5%, vaginitis 1.2%, acne 1.2%, alopecia 1.1%, rash 1.1%, hot flushes 1.0%, insomnia 1.0%, arthralgia 1.0%. Adverse reactions leading to discontinuation in 2% or more: bleeding 8.2%, amenorrhoea 2.1%, weight gain 2.0%.
• Funding: industry-funded (sponsor's label)

Limit of this finding: There is no placebo arm behind any of these percentages, so they describe what women on the injection reported, not what the injection added. The discontinuation line is the most useful number here: about one woman in twelve left a study because of bleeding.

Adverse reactions leading to study discontinuation in ≥2% of subjects: bleeding (8.2%), amenorrhea (2.1%), weight gain (2.0%).

The depot injection's post-marketing table also lists osteoporosis, cervical cancer, breast cancer, lack of return to fertility and unexpected pregnancy, and its footnote tells prescribers to use strict aseptic technique. (Source 38)

• Case series, Very low certainty.
• Size: Spontaneous reports from a population of uncertain size; no counts given.
• Who: Depot medroxyprogesterone users after approval.
• How long: Not applicable.
• Result: No rates can be estimated. The table lists, among others, osteoporosis; cervical cancer and breast cancer; lack of return to fertility and unexpected pregnancy; jaundice; anaemia and blood dyscrasia; paralysis and facial palsy; scleroderma. The dagger footnote records reported injection site abscess and infections and instructs strict aseptic injection technique.
• Funding: industry-funded (sponsor's label)

Limit of this finding: A name on a post-marketing list means somebody reported it, nothing more: there is no denominator, no comparison group and no finding that the drug caused it. The label says so itself. 'Unexpected pregnancy' and 'lack of return to fertility' on the same list are worth noticing, because they are failures of the drug's purpose rather than side effects.

† Injection site abscess and injection site infections have been reported; therefore, strict aseptic injection technique should be followed when administering Depo‑Provera CI in order to avoid injection site infections [see Dosage and Administration (2.1)].

Bone density in adolescents treated for more than two years had not returned to baseline at the hip or femoral neck up to 60 months after the last injection, with only the lumbar spine recovering. (Source 39)

• Cohort study, Low certainty.
• Size: 49 adolescents in each duration group at the end of treatment, falling to 2 or 3 by 60 months after treatment.
• Who: Adolescent females treated with Depo-Provera CI for two years or less versus more than two years.
• How long: Up to 60 months after the last injection.
• Result: More than two years of use: total hip -6.2% at the end of treatment, -4.6% at 12 months, -3.6% at 24 months, -4.6% at 36 months, -2.5% at 48 months; femoral neck -5.8%, -4.3%, -3.8%, -3.8%, -1.7%; lumbar spine -3.5% at the end of treatment and back above baseline by 24 months. Two years or less: total hip -1.5% at the end of treatment and 0.3% by 24 months.
• Funding: industry-funded (sponsor's label)

Limit of this finding: The later columns rest on very few people: by 48 and 60 months after treatment there were 9, then 2, adolescents left in the longer-use group. Numbers that small can move a lot by chance, so the shape of the recovery is more reliable than any single figure. The comparison is between two durations of use, not against untreated adolescents, who were gaining bone over the same period.

Total Hip BMD End of Treatment 49 -1.5% 49 -6.2% 12 M post-treatment 33 -1.4% 24 -4.6%

In the Cochrane weight review, the three studies that did show a difference all involved the depot injection, with about 3 kg more weight gain than a copper coil by year three. (Source 40)

• Systematic review, Low certainty.
• Size: 22 studies, 11,450 women; 17 non-randomised studies and 5 randomised trials; 16 of the studies were of depot medroxyprogesterone acetate.
• Who: Users of progestin-only contraceptives compared with another method or no contraceptive.
• How long: 6 months to 10 years.
• Result: Weight gain with the depot injection versus a copper intrauterine device was greater by 2.28 kg at one year (95% CI 1.79 to 2.77), 2.71 kg at two years (2.12 to 3.30) and 3.17 kg at three years (2.51 to 3.83). Adolescents on the injection gained more body fat than a non-hormonal group (11.00%, 95% CI 2.64 to 19.36) and lost more lean mass (-4.00%, -6.93 to -1.07). A later retrospective study reported mean weight gain of 1.3 kg versus 0.2 kg at one year, 3.5 versus 1.9 at four years and 6.6 versus 4.9 at ten years.
• Funding: independent (Research Support, N.I.H., Extramural per PubMed)

Limit of this finding: These are the five studies out of 22 that found a difference; in 15 of them the groups did not differ significantly. The three depot studies named here were retrospective or prospective observational, not randomised, so women who chose the injection may have differed from women who chose a coil in ways that affect weight. The review rated its overall evidence low quality.

In a retrospective study, weight gain (kg) was greater for DMPA versus copper (Cu) IUC in years one (MD 2.28, 95% CI 1.79 to 2.77), two (MD 2.71, 95% CI 2.12 to 3.30), and three (MD 3.17, 95% CI 2.51 to 3.83).

What the evidence supports

The depot injection prevented pregnancy with a 12-month failure rate of zero to 0.7 in the manufacturer's five studies, conditional on reinjection every 13 weeks. (Source 3)

• Randomized trial, Moderate certainty.
• Size: Five clinical studies; sizes not given on the label.
• Who: Women using Depo-Provera CI for contraception.
• How long: 12 months.
• Result: 12-month failure rate zero (no pregnancies reported) to 0.7 by life-table method.
• Funding: industry-funded (sponsor's label)

In five clinical studies using Depo-Provera CI, the 12-month failure rate for the group of women treated with Depo-Provera CI was zero (no pregnancies reported) to 0.7 by Life-Table method. The effectiveness of Depo‑Provera CI is dependent on the patient returning every 3 months (13 weeks) for reinjection.

Adding cyclic oral medroxyprogesterone to conjugated oestrogen almost abolished endometrial hyperplasia and atypia over 3 years. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 356 nonhysterectomised postmenopausal women: placebo n=119, conjugated oestrogen only n=119, oestrogen plus cyclic PROVERA n=118.
• Who: Postmenopausal women aged 45 to 64 with a uterus.
• How long: 3 years, double-blind, placebo-controlled.
• Result: Normal biopsy or no hyperplasia 97% on placebo, 38% on oestrogen alone, 95% on oestrogen plus 10 mg medroxyprogesterone; atypia 0, 12% and 0 respectively; adenocarcinoma 1 (1%), 0 and 0.
• Funding: industry-funded (sponsor's label)

Normal/No hyperplasia (%) 116 (97) 45 (38) 112 (95) Simple (cystic) hyperplasia (%) 1 (1) 33 (28) 4 (3) Complex (adenomatous) hyperplasia (%) 1 (1) 27 (22) 2 (2) Atypia (%) 0 14 (12) 0 Adenocarcinoma (%) 1 (1) 0 0

In the same review, fracture reduction was the only outcome where strong evidence showed a clinical benefit from hormone therapy. (Source 41)

• Systematic review, Moderate certainty.
• Size: Same 22 studies, 43,637 women.
• Who: Postmenopausal women.
• How long: 5.6 years of combined therapy; 7.1 years of oestrogen-only.
• Result: Fractures after 5.6 years of combined hormone therapy fell from 111 per 1000 to between 79 and 96 per 1000. Dementia after 4 years rose from 9 per 1000 to 11 to 30 per 1000 in women over 65. No strong evidence of an effect on colorectal cancer.
• Funding: independent (Research Support, Non-U.S. Gov't per PubMed)

Limit of this finding: Two qualifications belong with this benefit. The review rated the overall quality of evidence for its main comparisons as moderate, and its main stated limitation is that only about 30% of the women were aged 50 to 59 at the start, which is the age at which women actually consider hormone therapy for menopausal symptoms. In the one trial that looked at that age group directly, the only significantly raised risk was venous thromboembolism on combined therapy, with an absolute risk still under 1 in 500, and the study was not powered to detect other differences. This is version pub5 of the review; Crossref shows two later versions, issued in November 2025 and September 2026, which were not fetched in this run.

Risk of fracture was the only outcome for which strong evidence showed clinical benefit derived from HT (after 5.6 years' use of combined HT: from 111 per 1000 to between 79 and 96 per 1000; after 7.1 years' use of oestrogen-only HT: from 141 per 1000 to between 92 and 113 per 1000). Researchers found no strong evidence that HT has a clinically meaningful impact on the incidence of colorectal cancer.

Adding cyclic medroxyprogesterone at 5 mg, not only 10 mg, significantly lowered endometrial hyperplasia compared with conjugated oestrogen alone in a second 1-year trial of 832 women. (Source 42)

• Randomized trial, Moderate certainty.
• Size: 832 postmenopausal women aged 45 to 65; conjugated oestrogen alone n=283, plus cyclic 5 mg n=277, plus cyclic 10 mg n=272.
• Who: Postmenopausal women with a uterus.
• How long: 1 year.
• Result: Cystic hyperplasia 19% on conjugated oestrogen alone, 1% with cyclic 5 mg and 0% with cyclic 10 mg; adenomatous hyperplasia without atypia 1%, 0% and 0%.
• Funding: industry-funded (sponsor's label)

Limit of this finding: This trial compared adding medroxyprogesterone against oestrogen alone, not against no treatment, so it shows how well the progestin protects the womb lining, not whether taking either is a good idea. The trial ran one year, and hyperplasia is a tissue change, not cancer.

The treatment groups receiving 5 or 10 mg cyclic PROVERA (days 15–28) plus daily conjugated estrogens showed a significantly lower rate of hyperplasia as compared to the conjugated estrogens only group.

The depot injection works by stopping the pituitary hormones that ripen and release an egg and by thickening cervical mucus, and it clears very slowly, with a half-life of about 50 days after the injection. (Source 1)

• Blood level study, Moderate certainty.
• Size: Eight women aged 28 to 36 in the single-dose pharmacokinetic study.
• Who: Women given a single 150 mg intramuscular dose.
• How long: Concentrations followed until undetectable, 120 to 200 days after injection.
• Result: Peak plasma concentrations of 1 to 7 ng/mL reached over about 3 weeks; plasma protein binding averages 86%, primarily to albumin, with no binding to sex-hormone-binding globulin; apparent half-life about 50 days; metabolites mainly excreted in urine as glucuronide conjugates.
• Funding: industry-funded (sponsor's label)

Limit of this finding: A half-life of about 50 days, with the drug detectable for 120 to 200 days, is why the effects of this injection cannot be stopped once it has been given, and why fertility takes months to return. The pharmacokinetic study was in eight women, which is small.

Depo-Provera CI (medroxyprogesterone acetate [MPA]) inhibits the secretion of gonadotropins which primarily prevents follicular maturation and ovulation and causes thickening of cervical mucus.

What the evidence does not support

The same study found no excess of osteoporotic fractures specifically, and could not say anything about fracture risk later in life. (Source 43)

• Cohort study, Low certainty.
• Size: Same 312,395 women.
• Who: Same.
• How long: Same.
• Result: Very few osteoporotic fractures overall; incidence not found to be higher in depot users than non-users; the study could not determine an effect on fracture rate later in life.
• Funding: industry-funded (reported in the sponsor's label)

There were very few osteoporotic fractures (fracture sites known to be related to low BMD) in the study overall, and the incidence of osteoporotic fractures was not found to be higher in Depo-Provera CI users compared to non-users. Importantly, this study could not determine whether use of Depo-Provera CI has an effect on fracture rate later in life.

The same Cochrane review's own conclusion is that the fracture question cannot be settled from the existing evidence. (Source 44)

• Systematic review, Moderate certainty.
• Size: Same 19 trials.
• Who: Same.
• How long: Same.
• Result: Evidence quality moderate overall, largely because of the DMPA, implant and patch-versus-ring trials; combined oral contraceptive evidence low overall; many trials small with large losses.
• Funding: independent (Research Support, N.I.H., Extramural per PubMed)

Limit of this finding: The authors' conclusion has two halves and both belong together. They could not determine whether these contraceptives change fracture risk, and they rated the combined-pill evidence low quality overall. They then set that against the benefits, noting that injectables and implants give effective long-term contraception without a daily regimen and that progestin-only methods suit women who must avoid oestrogen. Quoting only the uncertainty makes the review read more negative than the authors left it.

Whether steroidal contraceptives influence fracture risk cannot be determined from existing information. The evidence quality was considered moderate overall, largely due to the trials of DMPA, implants, and the patch versus ring. The COC evidence varied in quality but was low overall. Many trials had small numbers of participants and some had large losses.

The Cochrane authors' overall position is that hormone therapy is not indicated for preventing cardiovascular disease, dementia or cognitive decline, and that data in perimenopausal women are insufficient. (Source 45)

• Systematic review, Moderate certainty.
• Size: Same 22 studies.
• Who: Perimenopausal and postmenopausal women.
• How long: At least 1 year.
• Result: Authors' conclusion; the main quality limitation was that only about 30% of women were aged 50 to 59 at baseline.
• Funding: independent (Research Support, Non-U.S. Gov't per PubMed)

Limit of this finding: Two qualifications belong with this benefit. The review rated the overall quality of evidence for its main comparisons as moderate, and its main stated limitation is that only about 30% of the women were aged 50 to 59 at the start, which is the age at which women actually consider hormone therapy for menopausal symptoms. In the one trial that looked at that age group directly, the only significantly raised risk was venous thromboembolism on combined therapy, with an absolute risk still under 1 in 500, and the study was not powered to detect other differences. This is version pub5 of the review; Crossref shows two later versions, issued in November 2025 and September 2026, which were not fetched in this run.

HT is not indicated for primary or secondary prevention of cardiovascular disease or dementia, nor for prevention of deterioration of cognitive function in postmenopausal women. Although HT is considered effective for the prevention of postmenopausal osteoporosis, it is generally recommended as an option only for women at significant risk for whom non-oestrogen therapies are unsuitable. Data are insufficient for assessment of the risk of long-term HT use in perimenopausal women and in postmenopausal women younger than 50 years of age.

A randomised trial of 7,829 African women found no substantial difference in HIV acquisition between the depot injection, a copper coil and a levonorgestrel implant. This is a null result on a question the observational literature had raised. (Source 8)

• Randomized trial, High certainty.
• Size: 7,830 women enrolled and 7,829 randomised across 12 sites in eSwatini, Kenya, South Africa and Zambia; 7,715 (99%) in the modified intention-to-treat population; 10,409 woman-years of follow-up.
• Who: HIV-seronegative women aged 16 to 35 seeking effective contraception in areas of high HIV incidence.
• How long: 18 months.
• Result: 7830 women enrolled and 7829 randomised, 2609 to the injection, 2607 to the copper coil and 2613 to the implant; 7715 (99%) in the modified intention-to-treat population, with 10 409 woman-years of follow-up. 397 HIV infections, overall incidence 3.81 per 100 woman-years (95% CI 3.45-4.21): 4.19 per 100 woman-years on the injection, 3.94 on the coil, 3.31 on the implant. Hazard ratios 1.04 (96% CI 0.82-1.33, p=0.72) for the injection versus the coil, 1.23 (0.95-1.59, p=0.097) versus the implant, and 1.18 (0.91-1.53, p=0.19) for the coil versus the implant.
• Funding: independent (Bill & Melinda Gates Foundation, USAID and PEPFAR, Sida, South African Medical Research Council, UNFPA; contraceptive supplies donated by the Government of South Africa and USAID)

Limit of this finding: This trial reports its primary hazard ratios with a 96% confidence interval, not the usual 95%. That is not a misprint and must not be read as one: the investigators spent part of their allowable error on interim looks at the data, which widens the interval for the final comparison. A 96% interval is slightly wider than a 95% one, so it is the more cautious of the two. The interval for the comparison with the implant is printed without a percentage label and follows the same convention.

In the modified intention-to-treat analysis, the hazard ratios for HIV acquisition were 1·04 (96% CI 0·82-1·33, p=0·72) for DMPA-IM compared with copper IUD, 1·23 (0·95-1·59, p=0·097) for DMPA-IM compared with LNG implant, and 1·18 (0·91-1·53, p=0·19) for copper IUD compared with LNG implant.

The ECHO investigators' own interpretation is that all three methods were safe and highly effective and that access should continue and increase. (Source 46)

• Randomized trial, High certainty.
• Size: Same 7,829 women.
• Who: Same.
• How long: 18 months.
• Result: Authors' interpretation; they also note HIV incidence was high in this population regardless of method.
• Funding: independent (see the preceding finding)

Limit of this finding: This trial reports its primary hazard ratios with a 96% confidence interval, not the usual 95%. That is not a misprint and must not be read as one: the investigators spent part of their allowable error on interim looks at the data, which widens the interval for the final comparison. A 96% interval is slightly wider than a 95% one, so it is the more cautious of the two. The interval for the comparison with the implant is printed without a percentage label and follows the same convention.

We did not find a substantial difference in HIV risk among the methods evaluated, and all methods were safe and highly effective. HIV incidence was high in this population of women seeking pregnancy prevention, emphasising the need for integration of HIV prevention within contraceptive services for African women. These results support continued and increased access to these three contraceptive methods.

Conjugated oestrogens on their own were NOT associated with an increased risk of invasive breast cancer in the WHI oestrogen-alone substudy, relative risk 0.80. (Source 33)

• Randomized trial, High certainty.
• Size: The WHI estrogen-alone substudy.
• Who: Postmenopausal women with a prior hysterectomy taking daily conjugated oestrogens 0.625 mg alone versus placebo.
• How long: Average follow-up 7.1 years.
• Result: Relative risk of invasive breast cancer 0.80 compared with placebo; no increase was found.
• Funding: independent trial, reported in the sponsor's label.

Limit of this finding: This is the null side of the same question, and it is about oestrogen WITHOUT medroxyprogesterone, in women who no longer have a uterus. It does not mean the combination is safe; the same label reports the combination raising invasive breast cancer in the next paragraph. It is recorded here because quoting only the combination result would make the label look one-sided.

In the WHI estrogen-alone substudy, after an average follow-up of 7.1 years, daily CE-alone was not associated with an increased risk of invasive breast cancer [relative risk (RR) 0.80] compared to placebo (see CLINICAL STUDIES ).

Oestrogen-only hormone therapy, without medroxyprogesterone, ran the other way in the same Cochrane review: it reduced breast cancer and fracture and did not raise coronary events. (Source 7)

• Systematic review, Moderate certainty.
• Size: Same 22 studies, 43,637 women.
• Who: Relatively healthy postmenopausal women taking oestrogen-only hormone therapy.
• How long: 1 to 7 years of use.
• Result: Oestrogen-only therapy raised venous thromboembolism (after 1 to 2 years from 2 per 1000 to 2 to 10 per 1000; after 7 years from 16 to 16 to 28 per 1000), stroke (after 7 years from 24 to between 25 and 40 per 1000) and gallbladder disease (from 27 to between 38 and 60 per 1000), but reduced breast cancer (after 7 years from 25 per 1000 to between 15 and 25 per 1000) and clinical fracture (from 141 to between 92 and 113 per 1000) and did not increase coronary events at any follow-up time.
• Funding: independent (Research Support, Non-U.S. Gov't per PubMed)

Limit of this finding: This is about oestrogen WITHOUT a progestin, so it does not describe medroxyprogesterone. It is recorded because the combined-therapy harms in this entry come from the same paragraph, and quoting only those would make the review look one-sided. Oestrogen alone is not an option for a woman with a uterus, because unopposed oestrogen raises endometrial cancer risk, which is the whole reason medroxyprogesterone is added. The reduced breast cancer interval reaches up to 25 per 1000, the same as the baseline, so the benefit's lower bound is no benefit at all.

but reduced the risk of breast cancer (after 7 years' use: from 25 per 1000 to between 15 and 25 per 1000) and clinical fracture (after 7 years' use: from 141 per 1000 to between 92 and 113 per 1000) and did not increase the risk of coronary events at any follow-up time.

Where the evidence is mixed

A 312,395-woman cohort study found more fractures of any kind in depot users, but no excess of the osteoporotic fractures that low bone density would be expected to cause. (Source 43)

• Cohort study, Low certainty.
• Size: 312,395 female contraceptive users in the United Kingdom.
• Who: Depo-Provera CI users versus contraceptive users with no recorded depot use.
• How long: Mean follow-up 5.5 years.
• Result: Incidence rate ratio for any fracture 1.41 (95% CI 1.35, 1.47). Fracture rate was higher in users who received fewer than 8 injections than in those who received 8 or more. Osteoporotic fractures were very few overall and were not higher in users than non-users.
• Funding: industry-funded (reported in the sponsor's label)

The Incident Rate Ratio (IRR) for any fracture during the follow-up period (mean=5.5 years) was 1.41 (95% CI 1.35, 1.47). It is not known if this is due to Depo-Provera CI use or to other related lifestyle factors that have a bearing on fracture rate.

A Cochrane review of randomised trials confirmed the bone density loss with the depot injection but could not answer whether contraceptives change fracture risk, because no trial measured fractures. (Source 47)

• Systematic review, Moderate certainty.
• Size: 19 randomised controlled trials; 5 compared an injectable with another injectable, implant or IUD.
• Who: Women using hormonal contraception before menopause.
• How long: Varied; search to April 2014.
• Result: Depot medroxyprogesterone acetate was associated with decreased bone mineral density; placebo-controlled trials showed BMD increases for DMPA plus an oestrogen supplement and decreases for DMPA plus placebo supplement; no trial had fracture as an outcome.
• Funding: independent (Research Support, N.I.H., Extramural per PubMed)

Limit of this finding: This review found no trial that measured fractures at all, so nothing here says whether contraceptives change the risk of breaking a bone. What it measured was bone density and bone turnover markers, which are stand-ins. The authors' own sensitivity analysis, restricted to the 11 trials with moderate or high quality evidence, is more mixed than the headline: it found some positive effects of an oestrogen supplement on density, a NEGATIVE effect of subcutaneous depot medroxyprogesterone on lumbar spine density, a negative effect on a bone formation marker, and among combined-pill trials a density decrease with gestodene plus ethinylestradiol 15 micrograms.

No trial had fracture as an outcome. BMD was measured in 17 studies and 12 trials assessed biochemical markers of bone turnover. Depot medroxyprogesterone acetate (DMPA) was associated with decreased bone mineral density (BMD). The placebo-controlled trials showed BMD increases for DMPA plus estrogen supplement and decreases for DMPA plus placebo supplement.

A 2025 rapid review of 32 recent items found a clear association between depot medroxyprogesterone and bone loss, and evidence that the loss can be partly or completely restored after stopping. (Source 16)

• Systematic review, Low certainty.
• Size: 32 items, most on intramuscular depot medroxyprogesterone acetate 150 mg.
• Who: Users of progestin-only contraceptives.
• How long: Publications from 5 May 2012 to 31 August 2023.
• Result: A clear association between any depot use and bone mineral density loss, more common in younger women and women on antiretrovirals; evidence that bone loss can be restored after discontinuation; no apparent loss with hormonal intrauterine devices or implants. The review states restoration after discontinuation is not adequately researched.
• Funding: not stated in the abstract; the paper's funding statement was not reached in this run.

Limit of this finding: This is a rapid review, not a full systematic review or a meta-analysis: it gives no pooled estimate, and by the authors' own account one reviewer read all the abstracts and full papers with only a 5% sample checked by a second. Its restoration finding is a reading of a mixed literature, and the same block says restoration after stopping is 'not adequately researched'. The two sentences must travel together. The source prints '150mg' without a space and a stray 'implants. implementation.'; both are as published.

We found a clear association between any use of depot medroxyprogesterone acetate and bone mineral density loss. This negative effect seems to be more common among younger women and women on antiretrovirals. There is, however, evidence to suggest that bone loss can be restored after depot medroxyprogesterone acetate discontinuation. Hormonal intrauterine device and implant users do not seem to experience bone mineral density loss.

A Cochrane review of 22 studies found only limited evidence of weight change with progestin-only contraceptives, and judged the overall evidence quality low. (Source 48)

• Systematic review, Low certainty.
• Size: 22 eligible studies, 11,450 women; 17 non-randomised studies and 5 randomised trials; 16 studies of depot medroxyprogesterone acetate.
• Who: Users of progestin-only contraceptives versus another method or no contraceptive.
• How long: 6 months to 10 years; search to 4 August 2016.
• Result: Mean weight gain at 6 or 12 months was less than 2 kg (4.4 lb) for most studies; multiyear data showed about twice as much change at two to four years as at one year, but groups generally did not differ significantly. Comparison groups did not differ significantly in 15 of the 22 studies.
• Funding: independent (Research Support, N.I.H., Extramural per PubMed)

Limit of this finding: 'Limited evidence of change' is not the same as evidence of no change. Seventeen of the 22 studies were not randomised, loss to follow-up was high, and the authors rated the overall quality of evidence low, which means a real difference could have been missed. Three studies did show more weight gain with the depot injection than in women using no hormonal method. The source prints 'discontinuation.These' with the space missing, as published.

These 22 studies showed limited evidence of change in weight or body composition with use of POCs. Mean weight gain at 6 or 12 months was less than 2 kg (4.4 lb) for most studies. Those with multiyear data showed mean weight change was approximately twice as much at two to four years than at one year, but generally the study groups did not differ significantly.

In ECHO, fewer women stopped the depot injection for adverse events than stopped the coil or the implant, but more women on the injection died and there were six deaths in that arm. (Source 49)

• Randomized trial, High certainty.
• Size: 2,609 in the depot arm, 2,607 copper coil, 2,613 implant.
• Who: Same trial.
• How long: 18 months.
• Result: Adverse events leading to discontinuation in 109 (4%) on the depot injection versus 218 (8%) on the coil and 226 (9%) on the implant (p<0.0001 for both comparisons). Serious adverse events in 49 (2%), 92 (4%) and 78 (3%) respectively. 12 deaths in total: six in the depot arm, five in the coil arm, one in the implant arm; the trial was not powered for mortality.
• Funding: independent (see above)

Limit of this finding: The trial was not designed or sized to compare deaths. Twelve deaths in 7,829 women over 18 months, six of them in the injection arm, is far too few to show a difference in mortality between the methods, and the paper makes no such claim. The lower discontinuation rate on the injection is a real finding; the death counts are not evidence of harm.

12 women died during the study: six in the DMPA-IM group, five in the copper IUD group, and one in the LNG implant group. Serious adverse events occurred in 49 (2%) of 2609 participants in the DMPA-IM group, 92 (4%) of 2607 participants in the copper IUD group, and 78 (3%) of 2613 participants in the LNG implant group.

The Women's Health Initiative was designed as an 8.5-year primary-prevention trial in 16,608 postmenopausal women with a uterus, recruited at 40 US centres between 1993 and 1998. (Source 50)

• Randomized trial, High certainty.
• Size: 16,608 postmenopausal women aged 50 to 79 with an intact uterus at baseline; 8,506 randomised to the combination tablet and 8,102 to placebo.
• Who: Postmenopausal women aged 50 to 79, recruited by 40 US clinical centres in 1993 to 1998.
• How long: Planned duration 8.5 years; the trial was stopped after a mean of 5.2 years.
• Result: Conjugated equine estrogens 0.625 mg/d plus medroxyprogesterone acetate 2.5 mg/d in one tablet versus placebo. The primary outcome was coronary heart disease with invasive breast cancer as the primary adverse outcome.
• Funding: independent (National Institutes of Health; PubMed records Research Support, U.S. Gov't, P.H.S.)

Limit of this finding: This is the frame for every hazard ratio from this trial. It was a prevention trial in women who were, on average, 63 years old and about a decade past menopause, taking one fixed combination at one fixed dose, and it ran for about three fifths of its planned length. It does not test other progestins, other doses, other routes, or hormone therapy started at the menopause for symptoms.

DESIGN: Estrogen plus progestin component of the Women's Health Initiative, a randomized controlled primary prevention trial (planned duration, 8.5 years) in which 16608 postmenopausal women aged 50-79 years with an intact uterus at baseline were recruited by 40 US clinical centers in 1993-1998.

The label's Table 4 gives the whole WHI result side by side, including the fracture reductions and a venous thromboembolism signal the label flags as outside its global index. (Source 51)

• Randomized trial, High certainty.
• Size: 8,506 on conjugated oestrogens plus medroxyprogesterone and 8,102 on placebo.
• Who: Postmenopausal women, average age 63.
• How long: Average 5.6 years, centrally adjudicated.
• Result: Relative risks with absolute rates per 10,000 women-years: CHD events 1.23 (0.99-1.53), 41 versus 34; all strokes 1.31 (1.03-1.68), 33 versus 25; deep vein thrombosis 1.95 (1.43-2.67), 26 versus 13; pulmonary embolism 2.13 (1.45-3.11), 18 versus 8; invasive breast cancer 1.24 (1.01-1.54), 41 versus 33; colorectal cancer 0.61 (0.42-0.87), 10 versus 16; hip fracture 0.67 (0.47-0.96), 11 versus 16; total fractures 0.76 (0.69-0.83), 152 versus 199; overall mortality 1.00 (0.83-1.19), 52 versus 52; global index 1.13 (1.02-1.25), 184 versus 165.
• Funding: independent trial, reported in the sponsor's label.

Limit of this finding: Three things in this table are easy to misread. The confidence intervals are described by the label itself as nominal and unadjusted for multiple looks and multiple comparisons, so several that just exclude 1 would not survive adjustment. Deep vein thrombosis is marked as not included in the global index, so the headline balance leaves it out. And overall mortality was exactly equal, 52 against 52, which is why this finding is recorded as mixed rather than as harm. The PROVERA label is quoted exactly as it prints, including its own typographic slips: 'Geriatric Use .)', 'Breast Cancer . )', '(see WARNINGS )' with a space before the bracket, 'metabolized in-vitro', 'via the CYP3A4', 'Reproductive (Urogenital )' and, in its Table 4 heading, '(95%nCI'. None of these is a transcription error.

Invasive breast cancerIncludes metastatic and non-metastatic breast cancer, with the exception of in situ breast cancer. 1.24 (1.01–1.54) 41 33 Colorectal cancer 0.61 (0.42–0.87) 10 16

Where the research disagrees

What the boxed warning on medroxyprogesterone is about. The injectable depot and the oral tablet carry different boxed warnings, determined here by parsing LOINC code 34066-1 in each product's SPL. The depot warning is about bone; the oral warning is about cardiovascular disease, breast cancer and dementia with oestrogen plus progestin therapy. Neither warning mentions the other's subject

• DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension 150 mg/mL, Pharmacia & Upjohn, SPL effective 2026-08-25, boxed warning section, position: WARNING: LOSS OF BONE MINERAL DENSITY • Women who use Depo-Provera Contraceptive Injection (Depo-Provera CI) may lose significant bone mineral density. Bone loss is greater with increasing duration of use and may not be completely reversible [see Warnings and Precautions (5.1)]. (Source 2)
• PROVERA (medroxyprogesterone acetate) tablets, Pharmacia & Upjohn, SPL effective 2026-06-29, boxed warning section, position: WARNING: CARDIOVASCULAR DISORDERS, BREAST CANCER AND PROBABLE DEMENTIA FOR ESTROGEN PLUS PROGESTIN THERAPY Cardiovascular Disorders and Probable Dementia Estrogen plus progestin therapy should not be used for the prevention of cardiovascular disease or dementia. (Source 28)
• depo-subQ provera 104 (medroxyprogesterone acetate) injectable suspension, Pharmacia & Upjohn, SPL effective 2025-12-18, boxed warning section, which extends the same bone warning to its endometriosis indication, position: Depo-subQ provera 104 is not recommended as a long-term (i.e., longer than 2 years) birth control method or medical therapy for endometriosis-associated pain unless other options are considered inadequate [see Indications and Usage (1) and Warnings and Precautions (5.1)]. (Source 6)

How much weight the depot injection actually causes people to gain

• The Depo-Provera CI label, section 5.12, reporting the sponsor's uncontrolled completer data, cohort: Women who completed 2 years of therapy gained an average of 8.1 lb. Women who completed 4 years gained an average of 13.8 lb. Women who completed 6 years gained an average of 16.5 lb. (Source 23)
• Lopez and colleagues, Cochrane review of 22 comparative studies in 11,450 women, 2016, systematic-review: Mean weight gain at 6 or 12 months was less than 2 kg (4.4 lb) for most studies. Those with multiyear data showed mean weight change was approximately twice as much at two to four years than at one year, but generally the study groups did not differ significantly. (Source 48)

Whether the bone density lost on the depot injection comes back

• The Depo-Provera CI label, section 14.3, on adolescents treated for more than two years, cohort: Adolescents treated with Depo-Provera CI for more than two years did not recover to their baseline BMD level at femoral neck and total hip even up to 60 months post-treatment. (Source 15)
• Erhardt-Ohren and colleagues, rapid review of 32 items published 2012-2023, AJOG Global Reports 2025, systematic-review: There is, however, evidence to suggest that bone loss can be restored after depot medroxyprogesterone acetate discontinuation. (Source 16)

Whether the depot injection raises the risk of acquiring HIV, which observational and laboratory work had suggested

• The ECHO trial consortium, randomised multicentre open-label trial of 7,829 African women, Lancet 2019, rct: We did not find a substantial difference in HIV risk among the methods evaluated, and all methods were safe and highly effective. (Source 46)
• The prior observational and laboratory literature, as the ECHO investigators themselves describe it in their background, rct: Serious adverse events occurred in 49 (2%) of 2609 participants in the DMPA-IM group, 92 (4%) of 2607 participants in the copper IUD group, and 78 (3%) of 2613 participants in the LNG implant group. (Source 49)

How much

• Reference intake: Dosing is set by the prescriber and the injection is given by a clinician. The label's position for contraception is 150 mg of the intramuscular depot every 3 months (13 weeks) by deep intramuscular injection, not adjusted for body weight. For the oral tablet used to protect the endometrium, the label's position is 5 or 10 mg daily for 12 to 14 consecutive days a month alongside daily conjugated oestrogens 0.625 mg. (Source 52)
• Upper limit: There is no conventional upper limit; what the label sets instead is a limit on duration and a set of timing rules. Use of the depot injection for longer than 2 years is not recommended unless other birth control methods are considered inadequate, because of the effect on bone mineral density. The label also restricts when the first injection may be given, requires an interval of no more than 13 weeks, and requires a pregnancy check if that interval is exceeded. For the oral tablet the label says patients should be started at the lowest dose and that the lowest effective dose has not been determined. (Source 52)
• Studied: The Women's Health Initiative gave conjugated equine oestrogens 0.625 mg/d plus medroxyprogesterone acetate 2.5 mg/d in one tablet to 8,506 women, against placebo in 8,102, recruited at 40 US clinical centres from 1993 to 1998. (Source 50)
• Studied: The endometrial protection trial gave 10 mg of medroxyprogesterone daily for 12 days each cycle plus conjugated oestrogen 0.625 mg/day for 3 years; a second 1-year study used 5 mg or 10 mg cyclically. (Source 4)
• Studied: ECHO randomised women to an injection of 150 mg/mL intramuscular depot medroxyprogesterone every 3 months, a copper intrauterine device or a levonorgestrel implant, for 18 months. (Source 46)
• Studied: The bone density studies used 150 mg intramuscular depot for up to 5 years in adults and up to 240 weeks in adolescents, with a mean of 9.3 injections per adolescent user. (Source 21)

A common belief, and what the research shows

The belief: Medroxyprogesterone is one drug with one safety profile, so what is said about the contraceptive shot applies to the tablet and vice versa.

What the research shows: The two forms carry different boxed warnings, and parsing LOINC code 34066-1 in each label shows it. The injection's warning opens: “Women who use Depo-Provera Contraceptive Injection (Depo-Provera CI) may lose significant bone mineral density.” The tablet's warning opens on a different subject entirely: “Estrogen plus progestin therapy should not be used for the prevention of cardiovascular disease or dementia.” The bone warning does not appear on the tablet and the cardiovascular and dementia warning does not appear on the injection. The second misconception is the opposite one: that because the Women's Health Initiative found harm with oestrogen plus medroxyprogesterone, the contraceptive injection must carry those same cardiovascular and breast cancer risks. The WHI studied postmenopausal women with a mean age of 63 taking 2.5 mg a day by mouth with conjugated oestrogens, which is a different drug level, a different co-medication and a different population from a 150 mg injection in a woman of reproductive age.

Questions and answers

What is it?

Medroxyprogesterone acetate is a synthetic version of the hormone progesterone. It comes in two forms that do different jobs: a long-acting injection into muscle or under the skin, given every three months as contraception, and a tablet used for absent or irregular periods and to protect the lining of the womb in women taking oestrogen after menopause. It is a prescription medicine in both forms, with no dietary source. (Source 5)

What does it do in the body?

As an injection it damps down the pituitary hormones that ripen and release an egg, and thickens cervical mucus, which is how it prevents pregnancy. As a tablet it acts on the lining of the womb, keeping it from building up when oestrogen is being taken, and it is also used to bring on a period that has stopped. Because it lowers oestrogen levels, long use of the injection also reduces bone density. (Source 53)

Is it good or bad for you?

Both, and which one depends on the form and the setting. As contraception the injection is highly effective, with a 12-month failure rate of zero to 0.7, and the ECHO trial showed it does not raise HIV risk. Against that it reduces bone density in a way that may not fully reverse, which is the subject of its boxed warning. As a tablet alongside oestrogen it protects the womb lining, but that same combination in the Women's Health Initiative produced 8 more invasive breast cancers, 8 more strokes, 7 more coronary events and 10 more pulmonary emboli per 10,000 women-years. (Source 31)

How do you get more of it?

Only by prescription, and for the injection only from a clinician. The label's position for contraception is 150 mg of the intramuscular depot every 3 months, shaken before use and given deep into the gluteal or deltoid muscle, with the dose not adjusted for body weight. Taking more is not a thing that happens with a three-monthly injection; the question that does arise is whether to keep having it, and the label limits routine use to 2 years. (Source 52)

If it is harmful, what reduces it?

There is no way to clear a depot injection early; it is designed to release over three months, which is also why its effects, including the delay in fertility returning, cannot be reversed on demand. What the evidence addresses is recovery afterwards: bone density partly recovers over the two years after the last injection, less completely the longer the drug was used, and a 2025 review found bone health can be partly or completely restored after stopping. For the oral tablet, it is simply stopped, and the label asks for periodic review of whether it is still needed. (Source 14)

Why might someone be low in it or missing it?

Nobody is naturally short of medroxyprogesterone; it is a manufactured hormone. Someone may not be able to have it because of the label's contraindications: active or past blood clots or cerebrovascular disease, known or suspected breast cancer, significant liver disease, undiagnosed vaginal bleeding, or known hypersensitivity. Others stop because of the bleeding changes, which lead 8.2% of women to discontinue in the trials. (Source 20)

Which whole foods contain it or feed it?

No food contains medroxyprogesterone and no food feeds it. Food matters only for the tablet's absorption: taking a 10 mg tablet immediately before or after a meal raised the peak level by 50 to 70% and total exposure by 18 to 33%, without changing the half-life. For the injection, food is irrelevant. The label's patient information does raise a food-related concern indirectly, by explaining that the injection makes the body lose calcium stored in bone. (Source 11)

What happens if you do not have it?

Without the injection, there is no contraceptive effect, and the pregnancy rate rises to whatever the alternative gives: in ECHO, 255 pregnancies occurred across the three arms, 71% of them after the assigned method had been stopped. Without the tablet in a woman taking oestrogen and having a uterus, the womb lining is left unprotected. The 3-year trial shows the size of that: 38% of women on oestrogen alone had a normal biopsy against 95% on oestrogen plus medroxyprogesterone, and 12% developed atypia against none. (Source 4)

How can you test for it?

There is no routine blood level test. What is monitored is the consequence: the label says bone mineral density should be evaluated when a woman needs to continue the injection long-term, and that in adolescents the result has to be read against age and skeletal maturity, because a growing skeleton is a moving baseline. For a woman on the tablet with oestrogen, the monitoring is endometrial sampling where there is undiagnosed persistent or recurring abnormal bleeding. Medroxyprogesterone can also interfere with some laboratory tests. (Source 53)

References

1. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 12.1 Mechanism of Action, 12.2 Pharmacodynamics and 12.3 Pharmacokinetics. 2026. Read the source
2. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension, 150 mg/mL - BOXED WARNING section (LOINC 34066-1). 2026. Read the source
3. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.1 Contraception. 2026. Read the source
4. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL STUDIES section, Effects on the Endometrium. 2026. Read the source
5. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - INDICATIONS AND USAGE section. 2026. Read the source
6. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2025-12-18. depo-subQ provera 104 (medroxyprogesterone acetate) injectable suspension - BOXED WARNING section (LOINC 34066-1). 2025. Read the source
7. Cochrane Database of Systematic Reviews; this is version pub5 (2017). Crossref shows two later versions of the same review, 10.1002/14651858.CD004143.pub6 issued 2025-11-27 and 10.1002/14651858.CD004143.pub7 issued 2026-09-04, both with Bofill Rodriguez M first. Neither later version was fetched in this run, so pub5 is what is quoted here and it is superseded. Long-term hormone therapy for perimenopausal and postmenopausal women. [Main results section]. 2017. PMID 28093732, DOI 10.1002/14651858.CD004143.pub5. Read the source
8. The Lancet; PubMed records no individual authors for this record, only the collective name Evidence for Contraceptive Options and HIV Outcomes (ECHO) Trial Consortium, while Crossref lists 34 named authors with Ahmed, Khatija first, so the citation should read Ahmed K and colleagues for the ECHO Trial Consortium. HIV incidence among women using intramuscular depot medroxyprogesterone acetate, a copper intrauterine device, or a levonorgestrel implant for contraception: a randomised, multicentre, open-label trial.. 2019. PMID 31204114, DOI 10.1016/S0140-6736(19)31288-7. Read the source
9. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 7.1 Changes in Contraceptive Effectiveness Associated with Co-Administration of Other Products. 2026. Read the source
10. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 7.1, antibiotics. 2026. Read the source
11. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL PHARMACOLOGY, Pharmacokinetics, A. Absorption. 2026. Read the source
12. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL PHARMACOLOGY, Pharmacokinetics, E. Specific Populations and F. Drug Interactions. 2026. Read the source
13. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.1, osteoporosis risk factors. 2026. Read the source
14. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.2 Bone Mineral Density Changes in Women Treated with Depo-Provera CI. 2026. Read the source
15. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.3, BMD Recovery Post-Treatment in Adolescents. 2026. Read the source
16. AJOG Global Reports. Bone mineral density changes during use of progestin-only contraceptives: a rapid review of recent evidence.. 2025. PMID 40607241, DOI 10.1016/j.xagr.2025.100509. Read the source
17. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.15 Return of Fertility. 2026. Read the source
18. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - DOSAGE AND ADMINISTRATION, Reduction of Endometrial Hyperplasia in Postmenopausal Women Receiving Daily 0.625 mg Conjugated Estrogens. 2026. Read the source
19. JAMA. Association of Menopausal Hormone Therapy With Breast Cancer Incidence and Mortality During Long-term Follow-up of the Women's Health Initiative Randomized Clinical Trials.. 2020. PMID 32721007, DOI 10.1001/jama.2020.9482. Read the source
20. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 4 CONTRAINDICATIONS. 2026. Read the source
21. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.3 Bone Mineral Density Changes in Adolescent Females (12 to 18 Years of Age). 2026. Read the source
22. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 6.2 Post-Marketing Experience. 2026. Read the source
23. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.12 Weight Gain. 2026. Read the source
24. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 6.1 Clinical Trials Experience. 2026. Read the source
25. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.11 Bleeding Irregularities. 2026. Read the source
26. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 1 INDICATIONS AND USAGE. 2026. Read the source
27. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.3 Cancer Risks. 2026. Read the source
28. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - BOXED WARNING section (LOINC 34066-1). 2026. Read the source
29. JAMA; the byline differs between indexes - Crossref and OpenAlex both give a single author, the Writing Group for the Women's Health Initiative Investigators, while PubMed lists 13 named authors with Rossouw JE first. The key follows the two indexes that agree. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.. 2002. PMID 12117397, DOI 10.1001/jama.288.3.321. Read the source
30. JAMA; the byline differs between indexes - Crossref and OpenAlex both give a single author, the Writing Group for the Women's Health Initiative Investigators, while PubMed lists 13 named authors with Rossouw JE first. The key follows the two indexes that agree. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.. 2002. PMID 12117397, DOI 10.1001/jama.288.3.321. Read the source
31. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL STUDIES section, WHI Estrogen Plus Progestin Substudy. 2026. Read the source
32. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - WARNINGS section, c. Venous Thromboembolism. 2026. Read the source
33. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - WARNINGS section, a. Breast Cancer. 2026. Read the source
34. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL STUDIES section, Women's Health Initiative Memory Study. 2026. Read the source
35. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - WARNINGS section, a. Breast Cancer, tumour characteristics, observational and meta-analytic data, and the breast monitoring instruction. 2026. Read the source
36. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.3 Cancer Risks, the absolute breast-cancer figure, Cervical Cancer and Other Cancers. 2026. Read the source
37. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 6.1 Clinical Trials Experience, Table 2 (reactions in 1 to 5% of subjects) and the discontinuation line. 2026. Read the source
38. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 6.2 Post-Marketing Experience, remainder of Table 3 including osteoporosis, cervical and breast cancer and unexpected pregnancy, with both of the table's footnotes. 2026. Read the source
39. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.3, Table 6: bone mineral density recovery up to 60 months after stopping, by two years or less versus more than two years. 2026. Read the source
40. Cochrane Database of Systematic Reviews. Progestin-only contraceptives: effects on weight.. 2016. PMID 27567593, DOI 10.1002/14651858.CD008815.pub4. Read the source
41. Cochrane Database of Systematic Reviews; this is version pub5 (2017). Crossref shows two later versions of the same review, 10.1002/14651858.CD004143.pub6 issued 2025-11-27 and 10.1002/14651858.CD004143.pub7 issued 2026-09-04, both with Bofill Rodriguez M first. Neither later version was fetched in this run, so pub5 is what is quoted here and it is superseded. Long-term hormone therapy for perimenopausal and postmenopausal women. [Main results section, continuing directly on from marjoribanks-2017-hormone-therapy; the label 'MAIN RESULTS:' stands at the head of that block]. 2017. PMID 28093732, DOI 10.1002/14651858.CD004143.pub5. Read the source
42. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL STUDIES section, the second 1-year endometrial study in 832 women comparing cyclic 5 mg and 10 mg with conjugated oestrogen alone. 2026. Read the source
43. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 14.4 Bone Fracture Incidence in Women Treated with Depo-Provera CI. 2026. Read the source
44. Cochrane Database of Systematic Reviews. Steroidal contraceptives: effect on bone fractures in women.. 2014. PMID 24960023, DOI 10.1002/14651858.CD006033.pub5. Read the source
45. Cochrane Database of Systematic Reviews; this is version pub5 (2017). Crossref shows two later versions of the same review, 10.1002/14651858.CD004143.pub6 issued 2025-11-27 and 10.1002/14651858.CD004143.pub7 issued 2026-09-04, both with Bofill Rodriguez M first. Neither later version was fetched in this run, so pub5 is what is quoted here and it is superseded. Long-term hormone therapy for perimenopausal and postmenopausal women.. 2017. PMID 28093732, DOI 10.1002/14651858.CD004143.pub5. Read the source
46. The Lancet; PubMed records no individual authors for this record, only the collective name Evidence for Contraceptive Options and HIV Outcomes (ECHO) Trial Consortium, while Crossref lists 34 named authors with Ahmed, Khatija first, so the citation should read Ahmed K and colleagues for the ECHO Trial Consortium. HIV incidence among women using intramuscular depot medroxyprogesterone acetate, a copper intrauterine device, or a levonorgestrel implant for contraception: a randomised, multicentre, open-label trial.. 2019. PMID 31204114, DOI 10.1016/S0140-6736(19)31288-7. Read the source
47. Cochrane Database of Systematic Reviews. Steroidal contraceptives: effect on bone fractures in women.. 2014. PMID 24960023, DOI 10.1002/14651858.CD006033.pub5. Read the source
48. Cochrane Database of Systematic Reviews. Progestin-only contraceptives: effects on weight.. 2016. PMID 27567593, DOI 10.1002/14651858.CD008815.pub4. Read the source
49. The Lancet; PubMed records no individual authors for this record, only the collective name Evidence for Contraceptive Options and HIV Outcomes (ECHO) Trial Consortium, while Crossref lists 34 named authors with Ahmed, Khatija first, so the citation should read Ahmed K and colleagues for the ECHO Trial Consortium. HIV incidence among women using intramuscular depot medroxyprogesterone acetate, a copper intrauterine device, or a levonorgestrel implant for contraception: a randomised, multicentre, open-label trial. [continuation of the FINDINGS section, following directly on from ahmed-2019-echo-hiv]. 2019. PMID 31204114, DOI 10.1016/S0140-6736(19)31288-7. Read the source
50. JAMA; the byline differs between indexes - Crossref and OpenAlex both give a single author, the Writing Group for the Women's Health Initiative Investigators, while PubMed lists 13 named authors with Rossouw JE first. The key follows the two indexes that agree. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.. 2002. PMID 12117397, DOI 10.1001/jama.288.3.321. Read the source
51. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-06-29. PROVERA (medroxyprogesterone acetate) tablets - CLINICAL STUDIES section, Table 4: relative and absolute risk in the WHI estrogen plus progestin substudy at an average of 5.6 years. 2026. Read the source
52. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 2.1 Prevention of Pregnancy. 2026. Read the source
53. DailyMed (U.S. National Library of Medicine), labeler Pharmacia & Upjohn Company LLC; SPL version effective 2026-08-25. DEPO-PROVERA (medroxyprogesterone acetate) injectable suspension - section 5.1 Loss of Bone Mineral Density. 2026. Read the source