Pain is not good for patients of any age, but the consequences of pain for neonatal patients are perhaps more important than for older patients. Pain has serious short and long term consequences in the newborn. In the short term, it provokes an acute increase of intracranial pressure and blood pressure, oxygen desaturation, and a release of free radicals [3]. Subsequent metabolic and endocrine responses lead to hyperglycemia, and cardiovascular and respiratory instability. Long term effects of untreated pain include altered responses to pain in older life, extending into adolescence [4]. And yet, despite these well-known consequences, analgesia is grossly underused for newborn infants [5–7]. As an example, even though caregivers recognize that endotracheal intubation is painful, as many as 87 % of units do not premedicate before an endotracheal intubation in non-emergency conditions. Analgesia is rarely given to newborns for lumbar punctures and even sometimes chest tubes are inserted without local or systemic analgesia [5, 7]. This would be inconceivable in older patients, despite there being much less evidence of long term adverse consequences, perhaps because they can physically protest during painful procedures. Although there are concerns regarding the overutilization of opioids, in neonatology underutilization of analgesia is more of a concern than overutilization.

For more minor procedures, oral administration of a sucrose solution is known to be effective, but even this inexpensive simple intervention is underutilized [8]. Rigorous studies are performed to find out how to better alleviate or eliminate this pain. Unfortunately, during many of these clinical trials, babies in the control groups receive no analgesia [9, 10]. Placebo groups in such studies raise serious ethical concerns, but have unfortunately been included in 89 % of neonatal analgesia studies [11]. Some of these studies examine intramuscular injections or heelstick punctures, which might be considered to be relatively minor, but even very invasive procedures such as circumcision have undergone clinical trials where no analgesia was provided to the control-babies [12, 13]. The declaration of Helsinki [14] requires that in clinical trials any new treatment or drug should be compared to standard of care, if any exists, and not to a placebo. Ironically, prolonged pain and suffering are often used as reasons for withholding active intensive care in preterm infants. Intolerable, persistent and untreatable pain is an ethically appropriate reason for limiting interventions. Failure to prevent and treat pain is not.

Noises in an incubator are far higher than those allowed for other patients in the hospital. The incubator fans produce noises of 45 dB or even higher than this level in some cases [2, 15] while in hospital the background noise should not exceed 35 dB [16]. It is hard to understand why the maximal noise threshold tolerated in nurseries is higher than 35 dB.

Variations of practices exist everywhere in medicine. Usually, these variations are due to random factors, economic incentives, availability of resources or a different combination of therapies. Variations of outcomes for preterm infants are frequently due explicit policies or philosophies. Interestingly, different industrialized countries, which all have access to the same information from the medical literature, come to very different conclusions. Survival for babies born at 24 weeks is 81 % in Japan, 60 % in Canada, 33 % in the United Kingdom, and is as low as zero in some centers in the Netherlands [17]. Those who are more aggressive obviously think that intervention is ethically appropriate. Those who defend less aggressive treatment think that over-treatment is more problematic than under-treatment. Both argue that they are doing what is in the best interest of neonates and their families. Both invoke local or national policies, based on local or national outcome data, to justify their approach. In the end, values become facts, which reinforce the values and policies. Policy statements for preterm infants often state survival and handicap as justification for optional intervention. The fact that such outcome statistics would not be used to justify non-intervention approaches at later ages suggests that some other powerful factors are at work [18, 19].

In empiric research, physicians and students were more frequently willing to withhold resuscitation and intensive care from sick neonates than from older children or adults with similar or even much worse prognoses [20–25]. Many healthcare providers thought that resuscitation was in the best interests of a preterm infant born at 24 weeks gestation. Interestingly, even more thought that resuscitation was in the best interests of a 2 month-old or a multiply disabled 7 year-old who were described as having similar or worse outcomes [20–25]. More than 50 % of those who thought resuscitation to be in the best interest of the 24-weaker also said that they would accept a parental decision to provide comfort care [23]. On the other hand, a minority would do so for the 2 month old and the disabled 7 year-old. These findings were found in multiple countries with different healthcare and cultures [20–25]. Such responses suggest that decisions for preterm newborns are made using different values than those for older children. Many clinical guidelines propose withholding interventions for neonates as a result of poor survival at mortality rates that have never been used to propose withholding resuscitation later in life [26, 27].

These observations extend to resuscitation decisions in real life. For example, survival to hospital discharge of adults after an out of hospital cardiac arrest is about 7 % with a high risk of subsequent disability [28]. This outcome is much worse than survival/disability outcomes of babies born at 23 weeks of gestation. Yet, the former routinely have resuscitation instituted, whereas among the latter in many countries, resuscitation is actively discouraged, and in others it is considered optional. For example, among 12,390 adults with cardiac arrest recently studied, 159 were lucky enough to have a witnessed cardiac arrest in public close to an automatic defibrillating device and had it applied by a bystander, which increased their survival to about 34 % [28]. This survival rate is considered sufficiently positive to drive the widespread purchase, installation and maintenance of automated external defibrillators throughout North America. In stark contrast, many clinical guidelines propose avoiding the resuscitation of 23 week infants, who have a similar survival rate and better long term outcomes if active care is instituted [19].

For older children, near certain death or profound disability seem necessary before withholding or withdrawing LSIs are considered by HCP [29]. For the newborn, end of life decisions are generally taken considering not only babies’ best interest, but also the parents’ interests [30]. In a questionnaire study [31], medical respondents were more likely to wish to resuscitate a “precious IVF” baby of an older mother than a baby of an young single mother. Are physicians who care for older incompetent patients also influenced by older maternal age for life and death decisions of their patients?

NICUs cost-effectiveness has been highly scrutinized and seems to be held to higher standard than ICU for older patients. Some policy statements regarding counseling of women at risk of delivering preterm have used the just distribution of resources as a reason for being cautious in resuscitating fragile preterm infants [26]. While it is true that NICU care for neonates is expensive, so is intensive care for older patients. Surprisingly, in spite of these high costs, every study of the cost-effectiveness of NICUs shows them to be far more cost effective than many widely accepted treatments [32]. Twenty-nine percent of adult ICU bed days are used by patients who die, compared to 8 % in the NICU [33]. NICUs are cost effective because most of the money is spent on babies who survive. Even when the long term costs of survivors who are disabled are included, the advantage of neonatal units still holds. A standard measure of cost-effectiveness that combines survival and quality of life is dollars per quality-adjusted life-year (QALY). Most NICU survivors live a long time without serious impairments and economically productive lives. Thus, the high initial costs are amortized over a lifetime and lead to relatively low figures on dollars/QALY. The quality adjusted costs for life of an infant born at 24 weeks has been calculated at around 6000$ US/QALY [34]. Most critical interventions for adults cost more than 70,000$/QALY.