Scott A. Lorch

Building a Stronger Instrument in an Observational Study of Perinatal Care for Premature Infants

An instrument is a random nudge toward acceptance of a treatment that affects outcomes only to the extent that it affects acceptance of the treatment. Nonetheless, in settings in which treatment assignment is mostly deliberate and not random, there may exist some essentially random nudges to accept treatment, so that use of an instrument might extract bits of random treatment assignment from a setting that is otherwise quite biased in its treatment assignments. An instrument is weak if the random nudges barely influence treatment assignment or strong if the nudges are often decisive in influencing treatment assignment. Although ideally an ostensibly random instrument is perfectly random and not biased, it is not possible to be certain of this; thus a typical concern is that even the instrument might be biased to some degree. It is known from theoretical arguments that weak instruments are invariably sensitive to extremely small biases; for this reason, strong instruments are preferred. The strength of an instrument is often taken as a given. It is not. In an evaluation of effects of perinatal care on the mortality of premature infants, we show that it is possible to build a stronger instrument, we show how to do it, and we show that success in this task is critically important. We also develop methods of permutation inference for effect ratios, a key component in an instrumental variable analysis.

The Regionalization of Pediatric Health Care

Regionalization of health care is a method of providing high-quality, cost-efficient health care to the largest number of patients. Within pediatric medicine, regionalization has been undertaken in 2 areas: neonatal intensive care and pediatric trauma care. The supporting literature for the regionalization of these areas demonstrates the range of studies within this field: studies of neonatal intensive care primarily compare different levels of hospitals, whereas studies of pediatric trauma care primarily compare the impact of institutionalizing a trauma system in a single geographic region. However, neither specialty has been completely regionalized, possibly because of methodologic deficiencies in the evidence base. Research with improved study designs, controlling for differences in illness severity between different hospitals; a systems approach to regionalization studies; and measurement of parental preferences will improve the understanding of the advantages and disadvantages of regionalizing pediatric medicine and will ultimately optimize the outcomes of children.

Routine Cerebrospinal Fluid Enterovirus Polymerase Chain Reaction Testing Reduces Hospitalization and Antibiotic Use for Infants 90 Days of Age or Younger

OBJECTIVE. The goal was to evaluate the impact of cerebrospinal fluid enterovirus polymerase chain reaction testing on the length of hospitalization and the duration of antibiotic use for infants ≤90 days of age with suspected aseptic meningitis. METHODS. This retrospective cohort study was conducted at an urban, tertiary-care childrens hospital. Data were collected for 478 patients ≤90 days of age for whom cerebrospinal fluid enterovirus polymerase chain reaction testing was performed during the enteroviral seasons of 2000 to 2006. The length of hospitalization and the duration of antibiotic use were assessed. RESULTS. Cerebrospinal fluid enterovirus polymerase chain reaction test results were positive for 154 patients (34.8%). The mean length of stay was 3.65 days. The median polymerase chain reaction turnaround time was 23 hours. In multivariate analysis, having a positive cerebrospinal fluid enterovirus polymerase chain reaction result was associated with a 1.54-day decrease in the length of stay and a 33.7% shorter duration of antibiotic use. When patients were stratified according to the presence or absence of pleocytosis, both groups demonstrated significant reductions in the length of stay with positive cerebrospinal fluid enterovirus polymerase chain reaction results (1.32 and 1.38 days, respectively). Furthermore, increasing the polymerase chain reaction turnaround time by 24 hours increased the length of stay by 13.6% for patients with positive cerebrospinal fluid enterovirus polymerase chain reaction results. CONCLUSIONS. Having positive cerebrospinal fluid enterovirus polymerase chain reaction results decreases the length of hospitalization and the duration of antibiotic use for young infants. These results support the routine use of this test during periods of peak enterovirus prevalence.

Association of late-preterm birth with asthma in young children: practice-based study

OBJECTIVE: To evaluate the association of late-preterm birth with asthma severity among young children. METHODS: A retrospective cohort study was performed with electronic health record data from 31 practices affiliated with an academic medical center. Participants included children born in 2007 at 34 to 42 weeks of gestation and monitored from birth to 18 months. We used multivariate logistic or Poisson models to assess the impact of late-preterm (34–36 weeks) and low-normal (37–38 weeks) compared with term (39–42 weeks) gestation on diagnoses of asthma and persistent asthma, inhaled corticosteroid use, and numbers of acute respiratory visits. RESULTS: Our population included 7925 infants (7% late-preterm and 21% low-normal gestation). Overall, 8.3% had been diagnosed with asthma by 18 months. Compared with term gestation, late-preterm gestation was associated with significant increases in persistent asthma diagnoses (adjusted odds ratio [aOR]: 1.68), inhaled corticosteroid use (aOR: 1.66), and numbers of acute respiratory visits (incidence rate ratio: 1.44). Low-normal gestation was associated with increases in asthma diagnoses (aOR: 1.34) and inhaled corticosteroid use (aOR: 1.39). CONCLUSION: Birth at late-preterm and low-normal gestational ages might be an important risk factor for the development of asthma and for increased health service use in early childhood.

Metoclopramide for the Treatment of Gastroesophageal Reflux Disease in Infants: A Systematic Review

OBJECTIVES. Metoclopramide is a commonly used drug to treat gastroesophageal reflux disease in infants. Given its widespread use and growing concern about toxicity in this population, we conducted a systematic review of metoclopramide for the treatment of gastroesophageal reflux disease in infants. METHODS. We performed a systematic search of PubMed and bibliographies of relevant review articles. We included cohort, case-control, and intervention studies of the efficacy, effectiveness, or toxicity of metoclopramide therapy for gastroesophageal reflux disease in infants. We excluded case reports, case series, review articles, and abstracts. RESULTS. Twelve articles met our inclusion criteria. Of these, 11 were prospective trials, and 5 were randomized, blinded clinical trials. Study size ranged from 6 to 77 patients. Eight studies showed patient improvement with metoclopramide in at least 1 measured outcome; 1 study showed worsening symptoms with metoclopramide. Of the 5 randomized, blinded trials, 2 showed no effect of metoclopramide on any outcome, and 2 showed a significant placebo effect. Four studies commented on adverse effects of therapy, with irritability being the most frequently reported potential adverse effect of therapy. Other reported adverse effects included dystonic reactions, drowsiness, oculogyric crisis, emesis, and apnea. Among studies, there was marked heterogeneity in the patient populations, dosing, and outcomes studied. Therefore, a meta-analysis was not performed. We both agreed on a US Preventive Service Task Force rating of “poor” for the level of evidence, leading to an “inconclusive” recommendation for the safety and efficacy of metoclopramide in infants. CONCLUSIONS. The current literature is insufficient to either support or oppose the use of metoclopramide for gastroesophageal reflux disease in infants. In the future, large blinded randomized clinical trials are needed to determine the efficacy and toxicity of metoclopramide in this population.

The Hospital Compare Mortality Model and the Volume–Outcome Relationship

OBJECTIVE We ask whether Medicares Hospital Compare random effects model correctly assesses acute myocardial infarction (AMI) hospital mortality rates when there is a volume-outcome relationship. DATA SOURCES/STUDY SETTING Medicare claims on 208,157 AMI patients admitted in 3,629 acute care hospitals throughout the United States. STUDY DESIGN We compared average-adjusted mortality using logistic regression with average adjusted mortality based on the Hospital Compare random effects model. We then fit random effects models with the same patient variables as in Medicares Hospital Compare mortality model but also included terms for hospital Medicare AMI volume and another model that additionally included other hospital characteristics. PRINCIPAL FINDINGS Hospital Compares average adjusted mortality significantly underestimates average observed death rates in small volume hospitals. Placing hospital volume in the Hospital Compare model significantly improved predictions. CONCLUSIONS The Hospital Compare random effects model underestimates the typically poorer performance of low-volume hospitals. Placing hospital volume in the Hospital Compare model, and possibly other important hospital characteristics, appears indicated when using a random effects model to predict outcomes. Care must be taken to insure the proper method of reporting such models, especially if hospital characteristics are included in the random effects model.

One-Year Respiratory Outcomes of Preterm Infants Enrolled in the Nitric Oxide (to Prevent) Chronic Lung Disease Trial

OBJECTIVE To identify whether inhaled nitric oxide treatment decreased indicators of long-term pulmonary morbidities after discharge from the neonatal intensive care unit. STUDY DESIGN The Nitric Oxide (to Prevent) Chronic Lung Disease trial enrolled preterm infants (<1250 g) between 7 to 21 days of age who were ventilated and at high risk for bronchopulmonary dysplasia. Follow-up occurred at 12 +/- 3 months of age adjusted for prematurity; long-term pulmonary morbidity and other outcomes were reported by parents during structured blinded interviews. RESULTS A total of 456 infants (85%) were seen at 1 year. Compared with control infants, infants randomized to inhaled nitric oxide received significantly less bronchodilators (odds ratio [OR] 0.53 [95% confidence interval 0.36-0.78]), inhaled steroids (OR 0.50 [0.32-0.77]), systemic steroids (OR 0.56 [0.32-0.97]), diuretics (OR 0.54 [0.34-0.85]), and supplemental oxygen (OR 0.65 [0.44-0.95]) after discharge from the neonatal intensive care unit. There were no significant differences between parental report of rehospitalizations (OR 0.83 [0.57-1.21]) or wheezing or whistling in the chest (OR 0.70 [0.48-1.03]). CONCLUSIONS Infants treated with inhaled nitric oxide received fewer outpatient respiratory medications than the control group. However, any decision to institute routine use of this dosing regimen should also take into account the results of the 24-month neurodevelopmental assessment.

Variation in treatment of neonatal abstinence syndrome in US children's hospitals, 2004-2011.

Objective:Neonatal abstinence syndrome (NAS) is a drug withdrawal syndrome experienced by opioid-exposed infants. There is no standard treatment for NAS and surveys suggest wide variation in pharmacotherapy for NAS. Our objective was to determine whether different pharmacotherapies for NAS are associated with differences in outcomes and to determine whether pharmacotherapy and outcome vary by hospital.Study Design:We used the Pediatric Health Information System Database from 2004 to 2011 to identify a cohort of infants with NAS requiring pharmacotherapy. Mixed effects hierarchical negative binomial models evaluated the association between pharmacotherapy and hospital with length of stay (LOS), length of treatment (LOT) and hospital charges, after adjusting for socioeconomic variables and comorbid clinical conditions.Result:Our cohort included 1424 infants with NAS from 14 children’s hospitals. Among hospitals in our sample, six used morphine, six used methadone and two used phenobarbital as primary initial treatment for NAS. In multivariate analysis, when compared with NAS patients initially treated with morphine, infants treated with methadone had shorter LOT (incidence rate ratio (IRR)=0.55; P<0.0001) and LOS (IRR=0.60; P<0.0001). Phenobarbital as a second-line agent was associated with increased LOT (IRR=2.09; P<0.0001), LOS (IRR=1.78; P<0.0001) and higher hospital charges (IRR=1.84; P<0.0001). After controlling for case-mix, hospitals varied in LOT, LOS and hospital charges.Conclusion:We found variation in hospital in treatment for NAS among major US childrens hospitals. In analyses controlling for possible confounders, methadone as initial treatment was associated with reduced LOT and hospital stay.

Evaluating Risk-Adjusted Cesarean Delivery Rate as a Measure of Obstetric Quality

OBJECTIVE: To validate the risk-adjusted cesarean delivery rate as a measure of obstetric quality through its association with maternal and neonatal outcomes for all pregnancies (model 1) and in primiparous patients with singleton pregnancies (model 2). METHODS: We constructed a population-based cohort of 845,651 patients from 401 hospitals representing all deliveries in California and Pennsylvania between 2004 and 2005. We used linked birth certificate and hospital admission records for mother and neonate to estimate the correlation between risk-adjusted cesarean delivery and a composite of adverse maternal outcomes, adverse neonatal outcomes, and four obstetric patient safety indicators from the Agency for Healthcare Research and Quality (AHRQ). RESULTS: In both models, risk-adjusted cesarean delivery rates were negatively correlated with both the maternal and neonatal composite outcomes and the AHRQ patient safety indicators for birth trauma, injury with instrumented vaginal delivery, and cesarean delivery. Approximately 60% of the 107 hospitals with lower-than-expected risk-adjusted cesarean delivery rates had higher-than-expected rates of at least one of the six adverse outcomes, compared with 19.6% of the hospitals with higher-than-expected risk-adjusted cesarean delivery rates and 36.1% of the hospitals with expected rates (P<.001). CONCLUSION: Lower-than-expected risk-adjusted cesarean delivery rates in all patients or when restricted to a more homogeneous group of primiparous patients with term singleton pregnancies are associated with higher-than-expected adverse maternal or neonatal outcomes. Higher-than-expected risk-adjusted cesarean delivery rates do not result in improved outcomes. LEVEL OF EVIDENCE: II

Delayed Acyclovir Therapy and Death Among Neonates With Herpes Simplex Virus Infection

OBJECTIVE: To determine the association of delayed acyclovir therapy with death among neonates with herpes simplex virus (HSV) infection. METHODS: A multicenter, retrospective, cohort study was conducted between January 1, 2003, and December 31, 2009, with 1086 neonates (age: ≤28 days) with HSV infection from 41 tertiary care childrens hospitals. Early acyclovir therapy was defined as initiation of intravenous acyclovir treatment within 1 day after hospital admission, and delayed acyclovir therapy was defined as initiation of treatment >1 and ≤7 days after hospital admission. Multivariate logistic regression models determined the association between delayed acyclovir therapy and death, with the use of propensity scores for each neonates likelihood of receiving delayed acyclovir treatment to control for differences in illness severity between groups. RESULTS: The median age was 10 days. Delayed acyclovir therapy was administered to 262 neonates (24.1%). In most cases (86.2%) of delayed receipt, acyclovir administration occurred on the second or third day of hospitalization. The overall mortality rate was 7.3% (95% confidence interval: 5.8%–9.0%); 9.5% of those who received delayed acyclovir treatment and 6.6% of those who received early acyclovir treatment died. In a multivariate analysis, delayed acyclovir therapy was associated with significantly greater odds of death (adjusted odds ratio: 2.63 [95% confidence interval: 1.36–5.08]) compared with early acyclovir therapy. CONCLUSIONS: In this multicenter observational study of neonates with HSV infection, delayed initiation of acyclovir therapy was associated with in-hospital death. Our data support the use of empiric acyclovir therapy for neonates undergoing testing for HSV infection.