Science You Can Use: “Shoot ’em while you got ’em?” New study raises questions about early Depo-Provera shots and breastfeeding problems.

A hot-off-the-press study is raising questions about the effect of early Depo-Provera contraceptive shots and breastfeeding outcomes.

For some time I’ve been hearing that mothers – often low income women of color – are being given Depo-Provera shots very early postpartum – before they leave the hospital with their new babies.

Why not wait a few weeks or months?  My retired pediatrician father-in-law refers to the rationale as “shoot ’em while you got ’em.”  The concern is that these women are likely to skip their 6 week follow up OB appointments, and so never get contraception which might prevent a new and unintended pregnancy.  Sounds neat and tidy, right?

But what if it’s not?  What if it causes women to be unable to sustain breastfeeding because it either delays their mature milk from coming in, or leads to low production, or both?  People who work with nursing moms have been raising a flag about this for some time.  Now it is the subject of a new study – one which raises some important questions.

Some background:  In 2009 and 2010 The World Health Organization (WHO) and the Centers for Disease Control (CDC) issued conflicting recommendations regarding the early postpartum use of Depo-Provera.  Reviewing the same body of literature, the WHO raised  concerns about the early use of Depo-Provera, while the CDC essentially gave it a green light.  This prompted the president of the Academy of Breastfeeding Medicine to argue in a press release, “[t]he new guidelines ignore basic facts about how breastfeeding works.”  We’ll get to that in a moment.

A later analysis of the literature on breastfeeding and early use of Depo-Provera concluded:

Current evidence is methodologically weak and provides an inadequate basis for inference about a possible causal relationship between early postpartum medroxyprogesterone use and poor breastfeeding outcomes.  However, given the presence of a strong biological model describing the potential deleterious effect of postpartum medroxyprogesterone use on lactation, further research that improves on current literature is warranted.

The “strong biological model” is this:  Depo-Provera is synthetic progesterone.  When a baby is born the mother’s progesterone level plummets, triggering the onset of mature milk production (“milk coming in”).  These shots of progesterone are begin given right as the mother’s levels should be falling.  As the Academy of Breastfeeding Medicine puts it, “An injection of artificial progesterone could completely derail this process.’’

The study out this week looked directly at what happens after women receive the shot.  Epidemiologist Dr. Elizabeth Brownell of Connecticut Children’s Medical Center recruited 183 mothers who had  delivered at low-income-serving obstetric and pediatric clinics to participate in a retrospective cohort study.  68 of these mothers had been given Depo-Provera shots an average of  37 hours after their babies’ birth.  They completed surveys which were merged with birth certificate data and medical records, and these were compared to the other 115 mothers with similar characteristics who had not been given the shot.

The results?  “Consistent with the biologic model, the results raised the possibility of a detrimental effect of [Depo-Provera] on duration of any breastfeeding, but differences in these distributions did not achieve statistical significance.” The difference held even when adjusted for co-founders.  Women who did not receive the shot breastfed for an average of 41 days.  Those who did breastfed for an average of 30 days.  Women receiving the shot were 14% more likely to cease breastfeeding within 6 weeks postpartum than those who were not.

So let’s look at why these differences didn’t achieve statistical significance.  Check out this chart below.  The dotted line represents the women given the shot, the solid line is the comparison group.  Every drop downward represents a woman stopping breastfeeding.

Dr. Brownell points out that the reason the difference didn’t achieve statistical significance may be because the two lines seem to meet in several places.  Do you see how they meet at 2 weeks, 4 weeks, and six weeks?  That’s probably because if you ask a mother who gave birth up to a year ago when she stopped breastfeeding she’s more likely to say “I breastfed for about two weeks,” than “I breastfed for 11 days.”

Otherwise, don’t you see a pretty clear difference in outcome?

I asked Dr. Brownell about this question of statistical significance, and she explained:

The effect we detected (14-22% increased risk) is deemed “statistically” small and clinically irrelevant by some.  Arguably, once you consider the multiple other factors associated with early breastfeeding cessation (hospital factors, home factors, return to work), a 14-22% (or even smaller) increased risk of breastfeeding cessation within six-weeks is indeed clinically relevant.  Additionally, from a public health perspective, once you take into account the amount of births and the prevalence of breastfeeding, this effect is exponentially more substantial.

What are the implications of these findings?  Again, Dr. Brownell:

We agree that these results are not statistically significant, and should not be over-interpreted; we are not suggesting that women should never elect to use Depo-Provera in the immediate postpartum period.

However, given these results are consistent with the biologic model, we believe responsible clinical caution and a standardized consent process (that includes disclosure of the potential negative effect of Depo-Provera on lactation) is needed for clinicians prescribing Depo-Provera in the immediate postpartum period.

Dr. Brownell has another paper coming out on the effect of early Depo-Provera use on exclusive breastfeeding soon, so stay tuned.

Did you have the Depo-Provera shot before you left the hospital?  Do you think it had an impact on your breastfeeding experience?  If you work with nursing moms who receive the shot in the hospital, have you observed an impact?

Image credit:  Wikimedia Commons

7 thoughts on “Science You Can Use: “Shoot ’em while you got ’em?” New study raises questions about early Depo-Provera shots and breastfeeding problems.

  1. It’s a horrible idea. I don’t even think Depo Provera should be on the market… too many side effects. It raised my blood pressure to stroke causing levels as well as bleed heavily every single day for 6 months… of course I didn’t get pregnant, I couldn’t have sex…


  2. I tried the patch after birth but it made me bleed all the time, so the doctor suggested the depo shot. I used that for about two years. Eric breastfed for 14 months, much of that while I was on the shot. Most birth control has side effects, matters not if it’s the pill, the shot or the IUD. There will be side effects since you are messing with the natural order of your body. It just depends on what you are willing to deal with. After my second child I am now on the IUD, the merena, my doctor warned me I may have trouble losing weight when I got it. I guess if I can’t lose weight with it we’ll see about when it’s gone.


  3. Depo-provera injections must be given every 3 months. If a woman is not trusted to use birth control and someone makes the decision to give her an injection before she leaves the hospital, what’s to ensure that she’ll return in 3 months for the next injection?

    “Those women” are encouraged to receive the injection with no full disclosure of risk, “You don’t want to become pregnant again do you?” “No.” “Okay, here’s a shot for you.” That’s commonly the whole conversation.

    Let’s get paternalistic practices out of healthcare.


    1. “Those women” are judged by the providers to be the wrong age, the wrong color, the wrong socioeconomic group to be able to make their own best decisions. They are also deemed incapable of having a conversation about their own lives.

      As the providers aren’t making her pregnant, they shouldn’t be ordering this drug to be given without informed consent.


  4. I think you need to clarify better what “not statistically significant” means in terms of research. It means that there really may be no difference at all between these two groups, and that the observed difference may have been fully due to chance. It means that conclusions should NOT be drawn based on this information alone. This link has more information on that:

    This research is enough to call for more research with a larger sample size, but it isn’t enough to draw any conclusions about Depo.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s