Randomized trials have demonstrated the short- to medium-term effectiveness of behavioral infant sleep interventions. However, concerns persist that they may harm children’s emotional development and subsequent mental health. This study aimed to determine long-term harms and/or benefits of an infant behavioral sleep program at age 6 years on (1) child, (2) child-parent, and (3) maternal outcomes.
Three hundred twenty-six children (173 intervention) with parent-reported sleep problems at age 7 months were selected from a population sample of 692 infants recruited from well-child centers. The study was a 5-year follow-up of a population-based cluster-randomized trial. Allocation was concealed and researchers (but not parents) were blinded to group allocation. Behavioral techniques were delivered over 1 to 3 individual nurse consultations at infant age 8 to 10 months, versus usual care. The main outcomes measured were (1) child mental health, sleep, psychosocial functioning, stress regulation; (2) child-parent relationship; and (3) maternal mental health and parenting styles.
Two hundred twenty-five families (69%) participated. There was no evidence of differences between intervention and control families for any outcome, including (1) children’s emotional (P = .8) and conduct behavior scores (P = .6), sleep problems (9% vs 7%, P = .2), sleep habits score (P = .4), parent- (P = .7) and child-reported (P = .8) psychosocial functioning, chronic stress (29% vs 22%, P = .4); (2) child-parent closeness (P = .1) and conflict (P = .4), global relationship (P = .9), disinhibited attachment (P = .3); and (3) parent depression, anxiety, and stress scores (P = .9) or authoritative parenting (63% vs 59%, P = .5).
Behavioral sleep techniques have no marked long-lasting effects (positive or negative). Parents and health professionals can confidently use these techniques to reduce the short- to medium-term burden of infant sleep problems and maternal depression.
Comments
Unethical and unscientific conclusions ignore how infants are harmed
This article shows a lack of ethical monitoring by reviewers and editors. The authors state: "Behavioral sleep techniques have no marked long-lasting effects (positive or negative)." This is an unconscionable and unscientific conclusion since there is no way that the authors studied all possible effects nor did they examine exactly what the control group was doing (so, what are they comparing against?). They apparently ignored the vast amount of mammalian research showing that distressing young offspring leads to long term negative effects on mental and physical health. By allowing these unconscionable conclusions, you are encouraging irresponsible parenting behavior that will do great harm to children.
Conflict of Interest:
None declared
Sleep Training Not Harmful? Methodological Concerns Question Conclusion
The conclusion drawn by Price and colleagues that behavioral sleep techniques did not cause lasting harms (Five-Year Follow-Up of Harms and Benefits of a Behavioral Infant Sleep Intervention: Randomized Trial) can be challenged on several levels.
Treatment Integrity
Reliability of conclusions drawn in longitudinal research rests on treatment integrity of the initial design. In this Infant Sleep Study, initial treatment integrity was irreparably confounded when 40% of mothers in the treatment group did not engage in treatment (behavioral sleep intervention). This invalidates drawing statistically sound conclusions related to treatment effect initially and longitudinally. How can one draw conclusions about the efficacy or safety of an intervention when almost half of the treatment group declined treatment? Furthermore, the authors did not collect or report information on intensity or duration of treatment use or infant distress for those mothers engaged in the advised interventions.
Neither can the authors say for certain that control group mothers did not engage in a behavioral sleep intervention, since they did not collect that information either. Indeed, given the acceptance in the study location of use of behavioral sleep interventions, there is good reason to suspect that the control group might have been contaminated with subjects who used controlled crying.
Compromised Design
The research design is presented as a randomized controlled study. However, those directly involved in the intervention itself, i.e., nurses and the families, were aware of group membership. This introduces an expectancy effect (see Adair, 1984 for review) in which changes in participants' behavior following an intervention can be associated with the expectation of benefit rather than from the intervention itself.
Cortisol Protocol
The author's method of using diurnal cortisol patterns as an indicator of chronic stress caused by the behavioral intervention is highly questionable. The protocol for salivary collection as outlined -- 30 minutes post waking and noontime -- is not supported in the literature as a means of interpreting diurnal patterns of stress. There is no basis in scientific literature for expecting cortisol levels to be elevated five years after a stressor event, i.e., behavioral intervention, unless the child was subject to ongoing stress in the interim.
Need to Reconsider Conclusions
The authors have no valid basis for their assertion that behavioral sleep interventions cause no long-term harm. Thus, this article cannot and should not be presented as evidence of the long-term safety of behavioral sleep interventions that involve leaving infants to cry for varying periods of time. Given the author's own admission of no evidence of long- term benefit to the children in the intervention group - neither in terms of sleep quality nor measures of family mental health -- they and others who promote the use of behavioral sleep interventions should reconsider their assumptions that such treatments are necessary in order to prevent long-term sleep problems or other negative outcomes purported to be caused by untreated night waking. This is a particularly important reconsideration given the plethora of literature supporting synchrony, responsiveness, and breastfeeding--all of which are compromised in behavioral sleep interventions.
Conflict of Interest:
None declared
Re:Unethical and unscientific conclusions ignore how infants are harmed
Even as a lay person, I can see that this study proves nothing. The "control" group, if it even qualifies as control, could have actually gotten frustrated and just left their babies to cry to sleep (CIO)at some point. We don't know. If some or many of them did and if we hold the affects against the mammal studies Dr. Narvaez mentions, the test group who used "controlled crying" methods would actually be BETTER off in terms of unregulated stress (which is the concern of leading neuropsychological thought). So there is really no comparison happening here. In other words, you'd have a similar amount of harm in both groups, as Dr. Narvaez indicates. If these researchers are making the claim that everything that happened crying-wise must have been safe, because "all the kids seemed ok" at 6, that's not good science. I wish Allan Schore, one of the leading researchers of our time on the profound effects of early unregulated distress, would respond to this. One point I'm sure he would make, is that you can't necessarily see the effects of a brain wired for stress until later in life when adversity tests our core resilience.
Conflict of Interest:
None declared
Five Year Follow-up of Harms and Benefits of Behavioral Sleep Intervention: Randomized Trial
The recently reported paper Five Year Follow-up of Harms and Benefits of Behavioral Sleep Intervention: Randomized Trial.1 claims to show that the use of 'controlled comforting' more typically referred to as 'controlled crying' does not cause long-lasting harms or benefits to child, child-parent or maternal outcomes. We are concerned that these claims go beyond what this paper has reliably demonstrated. One of the main concerns about the use of such techniques is their impact on cortisol, and in this study the measure of 'abnormal' cortisol was only available for 46% of the sample, of whom 29% of the intervention group had abnormal cortisol levels compared with 22% in the control group. This difference would probably have been statistically significant had they been able to follow-up more infants. This study also did not utilise a good measure of infant attachment security using only a brief 5-item parent-report of 'disinhibited' attachment, which is typically only identified in children who have experienced abuse or severe neglect.2-3 The study also provides inadequate information about the extent to which these techniques were actually used. One of the main concerns about elevated cortisol levels during the first two years of life is the impact of 'toxic' levels of stress on a number of the developing neural systems, and the 'architecture of regions in the brain that are essential for learning and memory'.4 The current study made no attempt to measure the impact of such techniques on child development or learning. There has been no research to date that has examined the impact of the use of these 'extinction' techniques with infants less than 6 months of age, and the overall research at the current time strongly suggests that young infants should not be left to cry themselves to sleep5 This study does not show that there is no long-term impact of controlled comforting on infants, and more scientific research is needed about the potential benefits and harms before parents can be confidently reassured about the extent to which this technique should be used. In the meantime, there is extensive evidence available concerning the effectiveness of other techniques for promoting sleep and suggesting that parent education/prevention 'may set the standard as the most economical and time -efficient approach to behaviorally-based pediatric sleep problems'.6 1. Price AMH, Wake M, Ukoumunne OC, Hiscock H . Five-year follow-up of harms and benefits of behavioral infant sleep intervention: Randomized Trial. Pediatrics. 2012, 130(4), 643-651. 2. Zeanah CH. "Disturbances of attachment in young children adopted from institutions". Journal of Developmental and Behavioral Pediatrics. 2000, 21 (3): 230-36. 3. Zeanah CH, Scheeringa M, Boris NW, Heller SS, Smyke AT, Trapani J. "Reactive attachment disorder in maltreated toddlers". Child Abuse and Neglect. 2004, 28 (8): 877-88. 4. National Scientific Council on the Developing Child. Excessive Stress Disrupts the Architecture of the Developing Brain: Working Paper 2005, No. 3. p.3. 5. Middlemiss W, Granger DA, Goldberg WA, Nathans L (2012). Asynchrony of mother-infant hypothalamic-pituitary-adrenal axis activity following extinction of infant crying responses induced during the transition to sleep. Early Human Development . 2012, 88: 227-232. 6. Mindell JA, Kunh B, Lewin DS, meltzer LJ, Sadeh A (2006). Behavioral Treatment of bedtime problems and night wakings in infants and young children. Sleep. 29(10): 1263-1276.
Conflict of Interest:
None declared
Long-term effects of sleep training: A flawed methodology
The question of the long-term impacts of sleep training methods is one that is paramount given the preponderance of resources recommending such action. I was quite excited to read Price et al.'s article discussing research looking at this understudied issue. However, after reading the article I admit I'm rather confused as to how the authors can claim that they even addressed the question of interest given the major flaws in the research. Given the space constraints I will only address three of the largest methodological flaws.
Arguably the most serious problem is the lack of control in the control group. Despite randomizing the groups, little to nothing is known about either what the parents in the control group actually did vis-a-vis sleep behaviour or what the nurses discussed with the control group families. Apparently the researchers assumed that these families did not take part in any sleep training with no evidence to support such an assumption (especially when myriad resources suggest sleep training as a remedy for infant sleep problems). As is, the control group is not a true control group with respect to sleep training outcomes.
The second concern pertains to the misuse of the intention-to-treat principle. Although there was an element of self-selecting in the experimental group, that self-selection would not have unfairly biased the outcomes being measured. If indeed there are long-term effects, they have ostensibly been masked by the inclusion of the nearly 43% of the experimental group who refused the experimental protocol (i.e., sleep training). A parallel would be examinations of breastfeeding outcomes. Researchers do not examine these outcomes based on what women intended to do (despite that being important in many ways) but rather what the actual behaviour was. In these instances it is better to collect data on the possible confounds and control for them statistically than to utilize the intention-to-treat principle. Currently, we have outcomes for those who did not sleep train included in the outcomes of those who did which only serves to muddy the waters.
The third concern is that the measures used to assess child outcomes are parent-report. What the authors have presented is an assessment of parental perception of child attachment and behaviour; there is no objective or child-report measure included (with the exception of child health). Parents' perceptions may be colored by their choice (to take part in sleep training or not) and feelings of having intervened rather than the intervention itself and should be supplemented with other measures, especially as the researchers did do a home visit, making this type of assessment possible.
Overall it seems that the authors tried to make their data fit a pressing research question. Unfortunately, what has resulted is a study that has no bearing on the question of interest, and thus more research remains needed. Despite what the authors would like us to believe, we are no closer to knowing the long-term effects of sleep training than we were prior to the publication of this article.
Conflict of Interest:
None declared