DETERMINANTS OF PRETERM BIRTH AMONG MOTHERS WHO GAVE BIRTH AT A REFERRAL HOSPITAL, NORTHWEST ETHIOPIA: UNMATCHED CASE- CONTROL STUDY.

Preterm birth is described as babies that are born alive before the end of 37 weeks of pregnancy. About 15 million babies are born preterm annually and more than a million died due to complications. In developing nations, like Ethiopia, PTB is underreported and underestimated. Objective - to identify determinates of preterm birth among mothers who gave birth at Felege Hiwot Comprehensive Specialized Referral Hospital, Northwest Ethiopia. Unmatched case-control study design was conducted. The total sample size was 558 mothers 140 cases and 418 controls. Simple random sampling was used to select the study population. Data were coded and entered into Epidata, version 3.2, and was analyzed by using STATA version 14. Mothers with a history of antepartum hemorrhage (AOR 3.53, 95% CI 1.31-9.47), premature rupture of membrane (AOR 8.9, 95% CI 4.51-17.57), pregnancy-induced hypertension (AOR 3.65, 95% CI 1.78-7.51), history of multiple pregnancies (AOR 2.49, 95% CI 0.89-6.95), primigravida (AOR 0.16, 95% CI 0.03-0.97) and Primiparity(AOR 0.054, 95% CI 0.05-0.64) had statistically significant association with experiencing preterm birth. The odds of giving preterm birth were higher among women with antepartum hemorrhage, PIH, PROM, and multiple pregnancies, but lower among primigravida and primiparous mothers.

Prediction of future uniform milk prices in Florida federal milk marketing order 6 from milk futures markets.

This study assessed the accuracy of 3 methods that predict the uniform milk price in Federal Milk Marketing Order 6 (Florida). Predictions were made for 1 to 12 mo into the future. Data were from January 2003 to May 2007. The CURRENT method assumed that future uniform milk prices were equal to the last announced uniform milk price. The F+BASIS and F+UTIL methods were based on the milk futures markets because the futures prices reflect the market's expectation of the class III and class IV cash prices that are announced monthly by USDA. The F+BASIS method added an exponentially weighted moving average of the difference between the class III cash price and the historical uniform milk price (also known as basis) to the class III futures price. The F+UTIL method used the class III and class IV futures prices, the most recently announced butter price, and historical utilizations to predict the skim milk prices, butterfat prices, and utilizations in all 4 classes. Predictions of future utilizations were made with a Holt-Winters smoothing method. Federal Milk Marketing Order 6 had high class I utilization (85 +/- 4.8%). Mean and standard deviation of the class III and class IV cash prices were $13.39 +/- 2.40/cwt (1 cwt = 45.36 kg) and $12.06 +/- 1.80/cwt, respectively. The actual uniform price in Tampa, Florida, was $16.62 +/- 2.16/cwt. The basis was $3.23 +/- 1.23/cwt. The F+BASIS and F+UTIL predictions were generally too low during the period considered because the class III cash prices were greater than the corresponding class III futures prices. For the 1- to 6-mo-ahead predictions, the root of the mean squared prediction errors from the F+BASIS method were $1.12, $1.20, $1.55, $1.91, $2.16, and $2.34/cwt, respectively. The root of the mean squared prediction errors ranged from $2.50 to $2.73/cwt for predictions up to 12 mo ahead. Results from the F+UTIL method were similar. The accuracies of the F+BASIS and F+UTIL methods for all 12 fore-cast horizons were not significantly different. Application of the modified Mariano-Diebold tests showed that no method included all the information contained in the other methods. In conclusion, both F+BASIS and F+UTIL methods tended to more accurately predict the future uniform milk prices than the CURRENT method, but prediction errors could be substantial even a few months into the future. The majority of the prediction error was caused by the inefficiency of the futures markets to predict the class III cash prices.