I guess that depends on what you mean ‘work’. Does developmentally sensitive care lead to more responsive, humane, and less disturbing care, improving the environment for our preterm infants, I think the answer is clearly yes.
If you mean does NIDCAP improve commonly used objective outcome measures in preterm infants, the answer has to be ‘not proven’. Arne Ohlsson (who is supposed to be retired but you wouldn’t guess it) and Sue Jacobs have just published a systematic review. They found no good evidence of benefit for short term medical complications of prematurity, nor for long term neuro-developmental outcomes. The babies may go home a few days earlier (about 0.5 weeks), But apart from that there is little evidence of benefit.
A few provisos, NIDCAP is a basketful of interventions, some of which may be beneficial and others not so much. There are other models of developmentally sensitive and appropriate care which may be more effective, or less, and also need to be well tested. Don’t throw away all your developmental care interventions, lets try testing them (like cycled lighting for example) and find things that work best.
There is indeed some evidence that cycled lighting might be preferable to persistent darkness, even for the very preterm infant who ‘should’ still be in a low light environment. (See here and here and here). It may also be that some noise, especially the human voice, is better than being very quiet, some very preliminary data from Terrie Inder, presented at PAS suggests that maybe being too quiet isn’t good.
I think the time has come for objective evaluation of the different parts of ‘developmental care’ to ascertain those that are essential, and those that may be useless of harmful.
One thing that has always intrigued me is that Dr H Als, who invented the form of developmentally sensitive care that we call NIDCAP, based on a theory of neurodevelopment that she made up called the ‘synactive theory’, always finds in her studies that all of the outcomes are better with NIDCAP, even in tiny underpowered studies. The latest included only 13 infants in the NIDCAP group, (with 17 controls) and supposedly found that brain structure and brain function were better on all of the measures that were performed. Now I don’t know about you, but the phrase ‘to good to be true’ springs to mind.
According to the manuscript this was a study registered as NCT00914108; but that study, which was only registered in 2009, recruited patients between 1996 and 2002 according to the entry on clinicaltrials.gov.
In this current manuscript it is stated that the infants were recruited from 2005 to 2008. So it cannot be 00914108. That is just sloppy.
This appears to be the study registered as NCT00166660, a single center study of 30 preterm SGA infants, which would match the recruitment dates. But I am not sure, there are a few things that appear different, babies were supposed to be 28 to 33 weeks gestation at birth according to the registration record, but infants in this study were eligible if they were at least 26 4/7 according to the article (which is a weird cutoff), with a birth weight below the 5th %le.
From the publication it is impossible to figure out what was the primary outcome variable, from the registration record of 00166660 it appears that there were 3:
- Assessment of Preterm Infants’ Behavior (APIB) [ Time Frame: 2 weeks corrected age (2w CA)
- Quantified Electroencephalography (qEEG) [ Time Frame: 2 weeks Corrected Age (2w CA)
- Magnetic Resonance Imaging(MRI) [ Time Frame: 2 weeks Corrected Age (2w CA)
Clearly there was no power at all for 3 primary outcome variables, the 2nd and 3rd being so ill-defined that you could do a hundred analyses and only report those that you thought were significant. Indeed there were apparently 40 coherence factors analyzed for the EEG, 5 of which are reported as being ‘significant’.
The statistics section of the study calculates the power of the study using the 30 enrolled infants, which is different to doing an a prior power analysis and calculating how many you need.
Dr Ohlsson and Jacobs review has received a large number of letters to the editor, most of which state something like, ‘we know it works, so we don’t believe this review’, often stated very forcefully. Many of the letters point out the statistically significant results that have been found, such as a Bayley 2 at 9 months of age, which is marginally better in NIDCAP babies to controls, but ignore the fact that this review addresses outcomes that are similar to those examined in the majority of neonatal intervention studies. If NIDCAP cannot be shown to improve these outcomes is it worth the investment?
The authors of the review submit a long response which addresses all of the substantive issues, if you are interested I suggest you read their response rather than sending me a comment!
Being sensitive to the developmental needs and the sensorial sensitivity of preterm infants makes great sense, and many nurses have helped to make the NICU a more humane, less overstimulating and less intrusive place. My own daughter benefited from such sensitivities; but that is not the same as claiming that NIDCAP improves brain structure!