Predicting Blood Donations in a Tertiary Care Center Using Time Series Forecasting.

The current algorithm to support platelets stock management assumes that there are always sufficient whole blood donations (WBD) to produce the required amount of pooled platelets. Unfortunately, blood donation rate is uncertain so there is the need to backup pooled platelets productions with single-donor (apheresis) collections to compensate periods of low WBD. The aim of this work was to predict the daily number of WBD to a tertiary care center to preemptively account for a decrease of platelets production. We have collected 62,248 blood donations during 3 years, the daily count of which was used to feed (standalone and ensemble versions of) six prediction models, which were evaluated using the Mean Absolute Error (MAE). Forecast models have shown better performances with a MAE of about 8.6 donations, 34% better than using means or medians alone. Trend lines of donations are better modeled by autoregressive integrated moving average (ARIMA) using a frequency of 365 days, the trade-off being the need for at least two years of data.

[Comparing the Application of Hema-Obs RSS to 250 Pregnancies from Obstetrics/Hematology Consultation in Centro Hospitalar São João, Portugal with the Application of Galit Sarig RSS to 90 Pregnancies from Rambam Health Care Campus, Israel]. / Comparação da Aplicação da EER Hema-Obs a 250 Gestações da Consulta de Hematologia-Obstetrícia do Centro Hospitalar São João, Portugal Versus EER Galit Sarig a 90 Gestações no Rambam Health Care Campus, Israel.

INTRODUCTION: Pregnant women with thromboembolic diseases, previous thrombotic episodes or thrombophilia family history were supervised in a multidisciplinary Obstetrics/ Hematology consultation in Centro Hospitalar São João EPE, Porto, Portugal. For the evaluation and medication of these women, a risk stratification scale was used. PURPOSES: The aim of this study was to validate a Risk Stratification Scale and thromboprophylaxis protocol by means of comparing it with a similar scale, developed and published by Sarig. MATERIAL AND METHODS: We have compared: The distribution, by risk groups, obtained through the application of the two scales on pregnant women followed at Centro Hospitalar São João, Porto, Portugal, consultation; the sensibility and specificity for each one of the scales (DeLong scale, applied to Receiver Operating Characteristic) curves; the outcomes in pregnancies followed in Hospital São João, Porto, PortugalResults: According to our Hema-Obs risk stratification scale, 29% were allocated to low-risk, 47% to high-risk and 24% to very-high-risk groups. According to Galit Sarig risk stratification scale, 24% were considered low-risk, 53% moderate, 16% high-risk and 7% as very high-risk group. In our study we observed 9% of spontaneous abortions, in comparison with 18% in the Galit Sarig cohort. From the application of Receiver Operating Characteristic curve to both risk stratification scales, the results of the calculated areas were 58,8% to our Hema-Obs risk stratification scale and 38,7% to Galit Sarig risk stratification scale, with a Delong test significancie of p = 0.0006. CONCLUSIONS: We concluded that Hema-Obs risk stratification scale is an effective support for clinical monitoring of therapeutic strategies.

Introdução: Na consulta de Hematologia-Obstetrícia do Centro Hospitalar São João EPE, as grávidas com trombofilias, episódios trombóticos anteriores e/ou história familiar de trombofilia são avaliadas e medicadas, recorrendo a uma escala de estratificação de risco (EER Hema-Obs).Objectivos: Pretende-se, com este trabalho, avaliar a eficácia desta escala de estratificação de risco, comparando-a com uma escala semelhante, desenvolvida e publicada por Sarig (escala de estratificação de risco Sarig).Material e Métodos: Procedeu-se às seguintes comparações: distribuição, por grupos de risco, obtida pela aplicação das duas escalas, em simultâneo, a 250 grávidas seguidas, na consulta do Centro Hospitalar São João EPE; sensibilidade e especificidade para cada uma das escalas (teste DeLong aplicado às curvas Receiver Operating Characteristic); desfechos nas gestações seguidas no Centro Hospitalar São João EPE e pelo grupo de Sarig no Rambam Health Care Campus (Israel).Resultados: A estratificação de risco nas 250 grávidas foi: a) com a escala de estratificação de risco Hema-Obs - Risco Materno (29%), Alto Risco Materno-Fetal (47%), Muito Alto Risco Materno-Fetal (24%); b) com a escala de estratificação de risco Sarig - Baixo (24%), Intermédio (53%), Alto (16%), Muito Alto (7%). Aplicando as curvas Receiver Operating Characteristic a ambas as escalas, resultam áreas calculadas de 58,8% para a escala de estratificação de risco Hema-Obs e de 38,7% para a escala de estratificação de risco de Sarig, correspondendo a uma diferença estatisticamente significativa (p = 0,0006 pelo teste de DeLong). Nas gestações acompanhadas no Centro Hospitalar São João EPE verificaram-se 91% de gestações bem-sucedidas e 9% de abortamentos; nas gestações acompanhadas por Sarig verificaram-se 82% de gestações bem-sucedidas e 18% de abortamentos.Conclusões: Conclui-se que a escala de estratificação de risco Hema-Obs constitui um suporte eficaz para estratégias terapêuticas de acompanhamento clínico.