From population reference to national standard: new and improved birthweight charts.

BACKGROUND: Antenatal detection of intrauterine growth restriction remains a major obstetrical challenge, with the majority of cases not detected before birth. In these infants with undetected intrauterine growth restriction, the diagnosis must be made after birth. Clinicians use birthweight charts to identify infants as small-for-gestational-age if their birthweights are below a predefined threshold for gestational age. The choice of birthweight chart strongly affects the classification of small-for-gestational-age infants and has an impact on both research findings and clinical practice. Despite extensive literature on pathological risk factors associated with small-for-gestational-age, controversy exists regarding the exclusion of affected infants from a reference population. OBJECTIVE: This study aims to identify pathological risk factors for abnormal fetal growth, to quantify their effects, and to use these findings to calculate prescriptive birthweight charts for the Dutch population. MATERIALS AND METHODS: We performed a retrospective cross-sectional study, using routinely collected data of 2,712,301 infants born in The Netherlands between 2000 and 2014. Risk factors for abnormal fetal growth were identified and categorized in 7 groups: multiple gestation, hypertensive disorders, diabetes, other pre-existing maternal medical conditions, maternal substance (ab)use, medical conditions related to the pregnancy, and congenital malformations. The effects of these risk factors on mean birthweight were assessed using linear regression. Prescriptive birthweight charts were derived from live-born singleton infants, born to ostensibly healthy mothers after uncomplicated pregnancies and spontaneous onset of labor. The Box-Cox-t distribution was used to model birthweight and to calculate sex-specific percentiles. The new charts were compared to various existing birthweight and fetal-weight charts. RESULTS: We excluded 111,621 infants because of missing data on birthweight, gestational age or sex, stillbirth, or a gestational age not between 23 and 42 weeks. Of the 2,599,640 potentially eligible infants, 969,552 (37.3%) had 1 or more risk factors for abnormal fetal growth and were subsequently excluded. Large absolute differences were observed between the mean birthweights of infants with and without these risk factors, with different patterns for term and preterm infants. The final low-risk population consisted of 1,629,776 live-born singleton infants (50.9% male), from which sex-specific percentiles were calculated. Median and 10th percentiles closely approximated fetal-weight charts but consistently exceeded existing birthweight charts. CONCLUSION: Excluding risk factors that cause lower birthweights results in prescriptive birthweight charts that are more akin to fetal-weight charts, enabling proper discrimination between normal and abnormal birthweight. This proof of concept can be applied to other populations.

Choosing a hospital for surgery: the importance of information on quality of care.

OBJECTIVE: To assess whether patients use information on quality of care when choosing a hospital for surgery compared with more general hospital information. METHODS: In this cross-sectional study in 3 Dutch hospitals, questionnaires were sent to 2122 patients who underwent 1 of 6 elective surgical procedures in 2005-2006 (aorta reconstruction [for treatment of aneurysm], cholecystectomy, colon resection, inguinal hernia repair, esophageal resection, thyroid surgery). Patients were asked which information they had used to choose this hospital and which information they intended to use if they would need similar surgical treatment in the future. RESULTS: In total, 1329 questionnaires were available for analysis (response rate 62.6%). Most patients indicated having used the hospital's good reputation (69.1%) and friendly hospital atmosphere (63.3%) to choose a hospital. For future choices, most patients intended to use the fact that they were already treated in that hospital (79.3%) and the hospital's good reputation (74.1%). Regarding quality-of-care information, patients preferred a summary measure (% patients with ''textbook outcome'') over separate more detailed measures (52.1% v. 38.0%, χ2 = 291, P < 0.01). For future choices, patients intend to use more information items than in 2005-2006, both in absolute terms (9 v. 4 items, t = 38.3, P < 0.01) as relative to the total number of available items (41.3% [40.1%-42.5%] v. 29.2% [28.1%-30.2%]). CONCLUSION: Patients intended to use more information for future choices than they used for past choices. For future choices, most patients prefer a summary measure on quality of care over more detailed measures but seem to value that they were already treated in that hospital or a hospital's good reputation even more.