AUR antibiotic use rate BPD bronchopulmonary dysplasia CDC Centers for Disease Control and Prevention DOT days of therapy ELBW extremely low-birthweight EOS early-onset sepsis NRN Neonatal Research Network VLBW very low-birthweight e516 NeoReviews clear benefits for infants with suspected or confirmed infection, but appropriate narrow-range coverage and treatment durations remain an issue. (3) However, there is increasing evidence pointing to the risks of adverse outcomes of antibiotic exposure among infants without cultureconfirmed infection. Measures to improve both prevention of infection and accuracy in diagnosing infection should result in decreased antibiotic use. To measure and ultimately optimize antibiotic use, there is a need to consistently characterize antibiotic use in NICUs and nurseries, as well as across centers. A thorough understanding of current antibiotic use and reliable tools for measuring use are first and critical steps to appropriately applying the principles of antimicrobial stewardship to the neonatal population. ANTIBIOTIC USE AMONG PRETERM INFANTS Preterm infants frequently receive empirical treatment with antibiotics soon after birth because of the risk of early-onset sepsis (EOS). Preterm labor and/or premature rupture of membranes result from maternal infection in about a third of cases, (4)(5) and 1 in 90 very low-birthweight infants (VLBW; See end of article for authors’ affiliations Correspondence to: Dr Hoehn, Department of Anesthesiology and Critical Care Medicine, 9th floor, Main Building, Children’s Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, PA 19104, USA; hoehn@email.chop. edu Accepted 27 September 2004.. Arch Dis Child Fetal Neonatal Ed 2005;90:F267–F269. doi: 10.1136/adc.2004.065078 Background: Although parents of neonates with congenital heart disease are often asked permission for their neonates to participate in research studies, little is known about the factors parents consider when making these decisions. Objective: To determine the reasons for parents’ decisions about participation in research studies. Methods: Qualitative analysis of the unsolicited comments of 34 parents regarding reasons for agreeing or declining to participate in research studies. Parents’ comments were offered spontaneously during interviews about clinical care decisions for neonates with congenital heart disease. Results: Parents cited five types of reason for or against permitting their newborn to participate in research studies: societal benefit (n = 18), individual benefit for their infant (n = 16), risk of study participation (n = 10), perception that participation posed no harm (n = 9), and anti-experimentation views (n = 4). Conclusion: Addressing parental decision making in the light of these reasons could enhance the parental permission process for parents of critically ill neonates. What factors do parents consider when they are asked to enrol their critically ill neonates in research studies? Although these complex decisions are common in many neonatal intensive care units, little is known about how parents conceive these decisions.1 Retrospective studies of parental permission in neonatal randomised controlled trials have found potential deficiencies in consent in 70% of parents interviewed.2 Parental duress associated with having a sick neonate may preclude informed permission.3 Evidence addressing this issue is mixed. One study found that 18 months after participation in neonatal research, 12% of the parents did not remember that they had been asked to consent to a study.4 In contrast, when parents whose newborns had participated in a clinical trial and then had died were asked permission to have a post mortem examination performed, they reported not feeling pressured when making this decision.5 A study of parents’ perspectives, using both retrospectively and prospectively collected data, found that 34–43% of parents had chosen to participate in a study because they believed that their neonate would get better care in the study.6 Personal benefit was also given as a reason to choose participation by 64% of parents in another large retrospective study; other reasons parents cited included societal benefit (48.6%) and no risk to their child (39.2%).2 We report here our analysis of a collection of unsolicited comments about neonatal research participation that were provided by parents in the context of a qualitative study on the effect of either prenatal or postnatal diagnosis of congenital heart malformation on parental permission for neonatal cardiac surgery.7 Given the paucity of prospectively collected data on parental decision making with regard to neonatal research, these unprompted reasons offered by the parents provide a fortuitous opportunity to examine factors that parents consider when contemplating whether to have their newborn child participate in research. METHODS After institutional review board approval, parents of neonates having cardiothoracic surgery were approached from 1 November 2001 through 1 May 2002 and invited to participate in a study evaluating the impact of prenatal diagnosis on parental permission for neonatal cardiac surgery.7 Semistructured qualitative interviews were performed between postoperative day 1 and 7 (mean 3). During these interviews, 34 of 49 parents mentioned, unsolicited, their reasons for or against participation in neonatal research. All of these parents were asked, as a follow up question, if they felt any pressure to participate in the studies. This report represents a qualitative analysis of these unsolicited, and originally unanticipated, parental comments about research participation. The specific research studies presented to the parents (and their decisions about participation) were obtained from the principle investigator for each study and are presented in table 1. The studies offered to the parents for their neonate’s participation depended on diagnosis and type of planned surgery. The number of studies in which the parents were invited to participate ranged from one to four and varied over the course of data collection.8 9 All the interviews were audio-