Mean absolute percentage error (MAPE), the measure most often used for evaluating subnational demographic estimates, is not always valid. We describe guidelines for determining when MAPE is valid. Applying them to case study data, we find that MAPE understates accuracy because it is unduly influenced by outliers. To overcome this problem, we calculate a transformed MAPE (MAPET) using a modified Box-Cox method. Because MAPE-T is not in the same scale as the untransformed absolute percentage errors, we provide a procedure for calculating MAPE-R, a measure in the same scale as the original observations. We argue that MAPE-R is a more appropriate summary measure of average absolute percentage error when the guidelines indicate that MAPE is not valid.

The influence of age, gender, and aspirin ingestion on plasma levels of tissue plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) activities was studied in a reference population of 35 men and 35 women between the ages of 20 and 65 years. The t-PA values (mean +/- SD) in the women before and after 5 minutes of venous occlusion were 3.8 +/- 1.4 and 7.8 +/- 4.4 micrograms/L, respectively; in men these values were 3.3 +/- 1.2 and 8.8 +/- 8.9 micrograms/L. Men had higher mean PAI levels than did women (5.0 vs. 2.5 kU/L). T-PA showed an inverse relationship to PAI in both sexes. There was a negative correlation of t-PA levels with age, whereas PAI levels were positively correlated. The ingestion of a single dose of aspirin (650 mg) did not alter PAI or t-PA activities. This study indicates that factors such as age and sex may need to be considered when reference populations are developed for clinical studies of fibrinolysis.

No abstract

Twenty healthy young men were exercised on a treadmill according to the protocol of Balke and Ware. Mean duration of exercise was 24.9 +/- 5.7 min and mean maximum heart rate was 195 +/- 9. Fibrinolytic activity was markedly accelerated with euglobulin lysis times decreasing to 36% of control values and fibrinogen-fibrin degradation products increasing 109% after exercise. Assays for fibrin monomer were negative in all samples. In vivo fibrinogen A alpha-chin degradation was assessed by sodium dodecyl sulfate polyacrylamide gel electrophoresis of reduced samples of fibrin monomer isolated by clotting plasma samples in the presence of 0.1 M epsilon-aminocaproic acid and 1% disodium ethylenediaminetetraacetate. The A alpha-chain, the fibrinogen chain most susceptible to plasmin degeneration, showed no evidence of increased degeneration after exercise. Gel scans showed no decrease in the ratio of total alpha-chain to beta- and gamma-chains after exercise. The ratio of intact alpha 1 chain (alpha 1, 67,000 mol wt) to total alpha-chain was 0.66 +/- 0.13 before exercise, 0.64 +/- 0.14 immediately after exercise, and 0.65 +/- 0.13 1 h after exercise. The rate and extent of crosslinking of the alpha-chain of fibrin formed by clotting plasma samples was unaltered by exericse. These data suggest that physiologically significant fibrinogenolysis does not occur with strenuous exercise, even when fibrinolytic activity is markedly accelerated.

Plasma levels of a fast-acting plasminogen activator inhibitor (PAI), which neutralizes both tissue plasminogen activator (t-PA) and urokinase, are markedly increased in endotoxin-treated rabbits. The ability of this inhibitor to prevent the fibrinolysis that occurs after a thrombogenic stimulus was investigated in a rabbit model. Normal and endotoxin-treated male New Zealand rabbits were infused with ancrod, an enzyme that causes noncrosslinked fibrin formation in vivo. Ancrod stimulated t-PA activity by 90% in normal rabbits and caused hypofibrinogenemia but did not increase PAI levels or induce fibrin deposition in target organs. Rabbits injected with endotoxin (10 micrograms/kg) showed an increase in PAI from less than 1 to 32 U/mL 4 hours later. When ancrod was infused at this time, 90% of the rabbits developed renal fibrin thrombi. Fibrin deposition was recorded in 40% of the rabbits that received a lower dose of endotoxin (1.0 microgram/kg) and had a PAI level of 14 U/ml at the time of ancrod infusion 4 hours later. Fibrin deposition did not occur in the endotoxin-treated rabbits that received normal saline. These data suggest that high levels of PAI inhibit fibrinolysis in vivo, thereby promoting fibrin clot deposition following a thrombogenic stimulus.