Search

The analysis of infant mortality data provides an opportunity for developing preventive strategies to improve this indicator of a population's health. All infant deaths in North Carolina during a 5-year period (1980 through 1984) were analyzed using the International Classification of Diseases, 9th revision (ICD-9), and a system for linked birth and death records that allows the analysis of birth certificate information on deaths. Causes of death were aggregated based on common etiology such as prematurity or obstetric-related conditions rather than the more traditional organ system taxonomy of the ICD-9 codes. Analyses were carried out separately for very low birth weight (less than or equal to 1500 g), moderately low birth weight (1501 through 2500 g), and normal birth weight (greater than 2500 g) babies. Maternal characteristics identified from the birth certificate were also compared with the different causes of death. Prematurity-related conditions accounted for 37.5% of all deaths, ranking far above the 17.4% for congenital anomalies and 12.9% for sudden infant death syndrome. For normal birth weight babies, sudden infant death syndrome ranked first, followed by congenital anomalies and nonperinatal infections. For the moderately low birth weight babies, congenital anomalies ranked first, with sudden infant death syndrome second and prematurity-related conditions third. For the very low birth weight babies, prematurity-related conditions accounted for nearly 70% of the deaths, with obstetric conditions and congenital anomalies ranking second and third, respectively. Maternal risk factors identified an overrepresentation of nonwhite, unmarried, and young teenage mothers and mothers with less than adequate prenatal care.(ABSTRACT TRUNCATED AT 250 WORDS)

The impact of a rural regional perinatal care (RPC) program was assessed by a quasi-experimental, controlled, population-based design. Outcome measures included changes in five-year average fetal and neonatal mortality rates as well as short-term obstetric and newborn morbidity. Declines in fetal and neonatal as well as birthweight specific mortality rates were observed for both pilot and control regions, for both races, and especially for 1501-2500 g infants. However, comparisons of preprogram (1966-74) and postprogram (1975-80) average yearly changes showed no statistically significant differences between regions. While the incidence of prenatal morbidity was the same for both regions, intrapartum and newborn morbidity significantly favored the pilot region. These results were difficult to interpret. Program relevant implications of the findings in relation to rural RPC in North Carolina are discussed. Specific benefits appeared to be associated with the development of two high-risk maternity clinics and a Level II center capability in the pilot region. The importance of community support and public/private sector cooperation in relation to RPC is noted.

Interrupted time series designs are frequently employed to evaluate program impact. Analysis strategies to determine if shifts have occurred are not well known. The case where statistical fluctuations (errors) may be assumed independent is considered, and a segmented regression methodology presented. The method discussed ia applied to the assessment of changes in local and state perinatal postneonatal mortality to identify historical trends and will be used to evaluate the impact of the North Carolina Regionalized Perinatal Care Program when seven years of post-program mortality data become available. The perinatal program region is contrasted with a control region to provide a basis for interpretation of differences noted. Relevant segmented regression models provided good fits to the data and highlighted mortality trends over the last 30 years. Considerable racial differences in these trends were identified, particularly for postneonatal mortality. Segmented regression is considered relevant for the analysis of interrupted time series designs in other applications when errors can be taken to be independent. Thus, the methodology may be regarded as a general statistical tool for evaluation purposes.

The effectiveness of a rural regionalized perinatal care (RPC) program was evaluated by a controlled, population-based design. The RPC program, begun in a carefully selected study region in July 1975, evolved into a system of care which included the following major components: identification of high-risk pregnancies and high-risk newborn infants; obstetric and newborn consultation and referral services between Level I, II, and III centers; professional education for physicians, nurses, and other health professionals; and nutrition and social work consultation. Substantial resources were made available from 1975 to 1980 to implement this "total package" of RPC. A matched, control region was identified which, except for the RPC program, was as comparable as possible to the study region. It was hypothesized that the RPC intervention would have the following effects when the study region was compared with the control region: reduction in fetal and neonatal mortality, no increase in postneonatal mortality, and reduction in obstetric and newborn morbidity. These findings were reported previously. This paper presents results of hypothesized reductions in adverse developmental, neurologic, maternal-infant attachment, and selected physical health outcomes. A sample of 447 infants was assessed by an extensive battery of instruments at 1 year, adjusted for gestational age. The most notable findings were mothers' reports of receptive language development and observations of maternal attachment behaviors that significantly favored the study region. No significant differences between study and control regions were observed for Bayley Mental and Motor Scores, abnormal neurologic signs, and the physical health measures.(ABSTRACT TRUNCATED AT 250 WORDS)

During the past two decades, intervention strategies designed to improve the health status of mothers and infants have been widely disseminated, yet relatively few have been evaluated for effectiveness. Moreover, most reported investigations have involved straightforward comparisons of aggregate data, employing various degrees of methodological control. In this study, vital statistics data were used to assess the effects of the North Carolina Maternity and Infant Care (MIC) Project on use of prenatal care and low birth weight. A weighted least squares procedure was used to control for selected maternal characteristics and identify significant interactions. Analyses of total population data indicated only minor MIC effects. However, more careful scrutiny of subpopulation data suggests that MIC impacts differed across categories of maternal risk status, with the greatest influence observed among mothers and infants at greatest risk. These findings raise several questions regarding the genesis of differential effects and suggest areas of special concern in conducting, interpreting, and using evaluations of programs for mothers and infants.

A survey, conducted in the summer of 1974, to determine the extent of regionalized perinatal care in the United States revealed 28 states to have programs in operation. Evaluation was a common concern of these programs. A model for the evaluation of Regionalized Perinatal Care Programs is presented here with the North Carolina program used as an example. Evaluation was mandated as an integral component of the program in North Carolina, and this model has been developed in response to that mandate.

A state's declining neonatal mortality rate was studied over the decade 1968 to 1977 to determine how much of the decrease resulted from improved risk characteristics in the newborn population. A shift was found over time, mostly among white births, toward higher birth weights and longer gestations. Overall, there was a 19% reduction in the low-birth-weight rate. Considerable improvement in birth weight-specific mortality was also found, especially in babies weighing 1,000 to 2,500 g during the last 5 years of the study. A neonatal risk matrix model was applied to the changes over time to separate the improvement due to shifts in population risk characteristics from that due to better care of the newborn. Thirty-four percent of the decade's decrease in NMR was attributable to changes in the birth weight and gestational age makeup of the newborn population. This "better babies" component was more prominent early in the decade and among the white population.