CIO, sleep training, and evidence or the lack thereof

The sleep training debate has, to no-one's great surprise, popped up again in Parentland.  In the red corner, Rosa Brooks: hell, yeah, stick in those earplugs, sling 'em in the cot and let 'em howl!  What harm could it possibly do?  In the green corner, Hathor, the Cow Goddess Of Attachment Parenting: heresy!  Don't you realise this will traumatise your child and damage his or her trust?  What caring mother could ever do such a thing?

I've commented previously on my opinions on both sleep training in particular and OneTrueWayism in parenting in general, but, as it happens, what drew me into the debate this time was another favourite bugbear of mine – the spot-the-difference game between what the evidence on a contentious topic says and what people with strong opinions on the topic claim it says.  What Hathor claimed, you see, is that her anti-CIO stance had been proved right by scientific research.  Years of study and reams of inquiry, she assured us, all consistently maintain that it is harmful to force your child to cry it out.  Indeed, Ferber himself had been proved wrong on the subject and had recanted his claims as a result.

Now, I can totally understand being anti-CIO – even its strongest proponents admit that it can be a pretty unpleasant experience for everyone concerned.  I'm a lot more sceptical about the belief that it's likely to cause long-term emotional damage – personally, I think babies are a lot more resilient than some of us give them credit for, and I don't think a child who's getting plenty of affection in his life overall is going to suffer permanent trauma as a result of a few bedtimes and naptimes crying alone – but it's a big old world and there's room for a lot of different opinions out there.  But claiming that there's scientific evidence for the supposed harmfulness of CIO – well, that's where things leave the realm of opinion and get into the realm of ascertainable fact.  Or, as it may be, fiction.

I've spent a lot of time looking at what different parenting forums and websites have to say about CIO, including a lot of the CIO-is-the-work-of-the-devil sites, and I've often come across this claim before.  Invariably, the 'evidence' presented (when the person making the claim actually does present any evidence instead of just assuming that the existence of evidence is so obvious as to need no further comment) falls into one or more of three categories:

1. Opinion.

2. Anecdote (often of cases where a number of other things were changed in a child's life at the same time.  "This two-month-old baby was left to cry herself to sleep and her parents stopped spending as much time with her during the day and she was fed less often and, guess what, she didn't thrive.  Obviously the sleep training!")

3. Actual research that isn't actually into CIO. There is a huge amount of research out there to show that regular positive attention and affection is crucially important for children's emotional development, and one of the few issues in parenting that just about anyone with any glimmer of a clue can actually agree on is that prolonged, regular neglect during childhood is liable to cause children problems; sometimes huge problems.  However, sleep training isn't prolonged, regular neglect.  It involves leaving children for short periods at specific times, while giving them just as much loving care as normal at other times (possibly more, since responding lovingly and affectionately to another person tends to be rather easier if you're not going insane with sleep deprivation).  Pointing to studies on the desperate harm suffered by Romanian orphans left abandoned in their cribs all day and every day as evidence of what a Bad Thing sleep training is is about as valid as pointing to studies on starving, malnourished children in the Third World and using them as support for a claim that you're doing your child terrible damage by expecting her to wait an extra twenty minutes for her dinner now and again.

Since no-one from the anti-CIO-for-sleep-training brigade ever seemed to cite any actual studies on the use of CIO for sleep training, I searched Medline to whether any such studies had ever actually been done.  (The technical term is "extinction", if you want to do the same thing.)  There are no long-term studies that I could find, but I did find two studies that looked at the psychological status of children shortly after sleep training.  Both of these seem to have passed unnoticed by the very people who are supposedly most fascinated by the psychological status of children following sleep training.  Call me cynical, but am I wrong in thinking that this might possibly have something to do with the fact that both studies actually showed children to be, if anything, somewhat more secure following CIO?

So, I replied to Hathor's claim with a quick summary of the above.  Since the list of references she gave in reply was fairly typical of the kind of stuff that gets presented as evidence in these debates, I'll go through them.

One reference to a speech by James McKenna in which he cited primate studies into short-term mother-infant separation.  Now, I can't comment directly on how these studies might or might not relate to CIO, because direct references weren't given in Hathor's quote or anywhere else on the 'Net that I could find.  However, a Medline search on "mother-infant separation" shows that, while lengthy separations do indeed appear to be harmful to infants, infants separated from their mothers for brief periods of time only were actually less fazed by separation when older than primates who hadn't undergone such separations.

One newspaper article about Margot Sunderland's new book, The Science Of Parenting.  I haven't read the whole book, as yet, but I've read the section on sleep training.  No references to studies on CIO.

Two articles about the infamous Commons and Miller paper.  I call it infamous because it gets mentioned in tones of reverence all the time in CIO debates.  It is, according to popular legend about it, a study by two Harvard psychiatrists that showed CIO to be harmful.  The only part of that that's correct is that the authors do indeed work at Harvard. 

The Commons and Miller paper wasn't a study and wasn't about CIO.  (And the authors are psychologists, not psychiatrists.)  It was a discussion of the many ways in which child-rearing practices differ in two different societies (the USA and the Gusii tribe of Kenya) and what kind of long-term effects this might have on children reared in the two societies.  It's a fascinating paper, but it isn't a study.

One reference to a study stating that all of 186 hunter-gatherer societies looked at in one study practiced co-sleeping.  Which tells us, um, precisely zero about the effects of CIO.

One webpage on the general evils of leaving babies to cry, devoid of any actual references.

And one article about a study showing that infant rats who received plenty of affection from their mothers were more secure than infant rats who received little maternal attention.  Which, as I discussed above, adds to the already sizeable body of evidence that giving your child little attention overall is A Bad Thing, but tells us nothing about the effects of a specific short-term intervention such as CIO.

My dissent on the issue of whether this constituted adequate evidence of the evils of CIO caused, as you can imagine, some debate.  Since there are now quite a number of questions for me in the second comment thread still awaiting a reply, I decided to move the discussion over here and answer them in this post.

What exactly are you looking for for something to be a study?

Well, not wanting to sound tautologous or anything, but a study involves studying something.  When someone says that CIO is harmful but doesn‘t actually provide any evidence to back this up, that’s an opinion.  When someone speculates on whether CIO may be harmful, that’s a theory.  When someone makes an attempt to assess the state of children following CIO, that’s a study.  (Whether or not it’s a good study is, of course, a whole separate and important question.)

Or to have compelling information for you to see that CIO is not a good thing for babies?

I’m not trying to claim it’s a “good thing” (although I believe that, for some babies, it’s a better thing than the alternative).  I’m objecting to the claim that research has proved it to be a harmful thing.  But, to answer your question: if well-conducted studies into the psychological state of children following sleep training showed them to be psychologically worse off after CIO, then that would be compelling evidence.

If I may be so bold as to ask, what exactly are you doing on a site that is pro co-sleeping trying to defend CIO?

Objecting to misinformation.  I don’t object to people being anti-CIO; I do object to people claiming the evidence states something that it doesn’t.

Or at least trying to say that there needs to be studies to prove that co-sleeping is benificial (sic)?

I haven’t said that.

I guess it all comes down to doing what works best for your family, taking into consideration that babies/children are people too, and that they have needs that they can not meet themselves do to their age.

Doing what works best for your family is exactly my philosophy, as well.  However, my experience is that when that statement is followed by that sort of qualifier in this sort of debate, what it actually means is that you don’t believe CIO is ever going to be what works best for anyone's family.  And, having read a lot of different stories from different people with different experiences, I can’t agree with that.

There are may ways to help a child learn to sleep that do not involve them having to cry for extended periods of time.

And I’d like to see them much more widely known (by which I do not just mean the blanket “Co-sleeping will solve all your problems!  What more could you possibly need to know?” recommendation that seems to be all that some attachment parenting advocates have to offer).  I’d also, however, like to see it more widely recognised that – like everything else in parenting – they aren’t universal solutions that work for all children and all families.

But I think we need to remember that there are a lot of parents out there who might well have tried alternative solutions to sleep problems with their children if they’d known about them, but who didn’t know about them and thus tried some form of CIO.  Now, leaving these families thinking “Damn, if only I’d known about that at the time!  Could have saved us an unpleasant few evenings” is one thing; leaving them thinking “Oh, no!  There’s scientific evidence that the way I handled things was actually damaging for my child!“ is another.  If we’re going to do that to parents, we ought to be damn sure we have our facts straight first.  If there isn’t any actual evidence that CIO is harmful then we shouldn‘t be claiming that there is, no matter how vehement our personal opinions on the subject.

 

Touche on the Harvard study, I haven’t seen the actual paper the article was based on.

Well, if you want to, you can read it here.  Right where I said it would be, in fact.

But a comparative multi-disciplinary investigation of different societies is not necessarily less valid than lab-controlled experiments. It’s what anthropologists do.

It's a valid research method for some things, although I don't think it would be a good way of studying CIO – there are so many differences between different societies that it wouldn't be possible to single out one specific brief episode during childhood and pinpoint the effects of that.  However, the objection I was making is not that their paper is an anthropological study, but that it isn't a study at all.  It's a discussion of previous research into the topic, and it doesn't contain any actual information on how the different methods of child-rearing affect children.  It simply theorises on how the differences might affect children, and suggests this as a topic for further research.

 

These [the children in the first CIO study] are 6-24 month old children they studied. How would you guess they rated the security and anxiety of these children?

They used a modified version of a scale called the Flint Infant Security Scale, filled in by the parents.  The second study I cited used the same scale, and also visual analogue scales to measure the parents' impressions of how depressed and how anxious/insecure their children seemed.

I personally can’t see how being left alone to sleep can make anyone more secure.

I've found that dealing successfully with a situation I originally thought to be beyond me usually leaves me feeling more secure.  Knowing that I can deal with it leaves me with more confidence in my own abilities.

It's also worth remembering that children who have difficulty getting to sleep and wake frequently in the night are often sleep-deprived themselves.  If adults find it easier to cope with life's stresses when well-rested, why shouldn't the same be true of children?

To me this abstract is pretty unconvincing.

That's fine.  I'm not out to bang a CIO-is-wonderful drum here – that isn't the way I feel at all.  What I'm trying to point out is that the existing evidence doesn't show it to be harmful.

I don’t believe in CIO.  Sarah, you obviously do to some extent

What I believe in is finding solutions that work for individual families, individual children.  I believe that sometimes, that solution is going to be CIO.  And I believe that though another method could potentially have worked just as well or better in most (not all) cases where CIO is used, that doesn't mean that using CIO in those cases was actually harmful.

 

Anyway, people also used to widely believe in ’spare the rod spoil the child’ and were full of evidence of how spanking led to better children.

And stories like that don't tell you that we should be extremely careful about not claiming that the evidence supports a particular way of doing things purely because that's what it suits us to believe?

 

I just don’t see how a three or six month old baby for example can know the difference between just having been left in his safe nursery and having been abandoned completely.

Well, when his mother turns up again, I think he's going to figure out that it was the former.

 

And how do you really know that a three month old really isn’t hungry, or that something isn’t really bothering him?

In fact, I don't know any experts who advocate using sleep training for a baby as young as three months.  But, assuming that you didn't feel that to be the crucial point of your question: By knowing your child and by using common sense.  For example, if you've just nursed your child and he isn't taking any more milk then it's a fair bet that hunger isn't the problem.

And besides that, why are only physical needs valid when speaking about babies? Certainly judging by the numbers of relationship gurus out there, all the books, all the Dr. Phils and beyond, we in North America believe that we have emotional needs that deserve to be met.

Certainly.  But that doesn't mean that someone has to be available to meet them every minute throughout the day and night.  I don't expect my partner to drop absolutely everything he's doing to talk to me whenever the fancy takes me, even if it's 4 a.m. and he's in a sound sleep.  I know that he has other things to do that are important; and I know that that doesn't detract from his love for me or his ability to be supportive and available to me overall.

 

Why is it less valid for a baby to be lonely than it is for an adult to be lonely?

It isn't.  But, similarly, why should it be so much more valid?  If a friend staying with you was regularly expecting you to come and keep her company regardless of what hour of the day and night it was or what else you might need to do, how long would it be before you started saying no some of the time?

 

I mean no offense by this, but I don’t really need you to answer these questions. I know what the answers are for me.

 

Which is good.  The point at which I start having a problem with these sorts of discussions is when people start deciding that they know what the answers are for everybody else.

I think ultimately all there is to this topic is to follow your heart, as Julinda and Serendipity said above.

And if your heart leads you to the conclusion that CIO is the right answer for your baby?

 

I hope these articles make people think about this issue a little bit more, to reconsider, to tune into their heart and see what is right for *them*.

I'd love it if there were more articles that did that, but I don't think either Rosa Brooks' or Hathor's had that aim.  What Hathor, like Brooks, really wants other people to do is to tune into her heart and do things the way she thinks is right.  That's the problem I have with this issue, as with so much else in parenting; so many people think they've got the one right way that's going to work for all children, just as though children weren't individuals as much as the rest of us.

But that's not why I wrote the reply I did to Hathor's post.  I replied to it because I believe that she was not correct in claiming that the existing scientific evidence proves CIO to be harmful.  And I hope I'd have had the guts to say so even if I was passionately anti-CIO on a personal level.  Judging the evidence on the basis of what we want it to show is a temptation that's impossible to avoid altogether – but we should be willing to be as honest as we can be about what it actually shows.