Biography:

In the past Julien I.E. Hoffman has collaborated on articles with Peter Oishi and Lucy S. Brevetti. One of their most recent publications is Clinical studyThe natural history of ventricular septal defects in infancy☆. Which was published in journal The American Journal of Cardiology.

More information about Julien I.E. Hoffman research including statistics on their citations can be found on their Copernicus Academic profile page.

Julien I.E. Hoffman's Articles: (35)

Clinical studyThe natural history of ventricular septal defects in infancy☆

AbstractSixty-two infants with ventricular septal defects were first catheterized under 1 year of age and followed up for one to five years; 40 were recatheterized.Thirty-six infants were born in the Bronx Municipal Hospital Center or Lincoln Hospital (local group). These patients were unselected, in that the decision to catheterize them was based only on the clinical diagnosis of a ventricular septal defect and not on an assessment of its size or the presence of symptoms. As far as we know, these infants include most if not all of the children born with ventricular septal defects in these hospitals during the period of study. With this assumption, the incidence of ventricular septal defects per 1,000 live births was 0.94 for full term infants, 4.51 for premature infants, and 1.35 for the whole group.Twenty-six other infants were referred from other hospitals or doctors (referred group), and, compared to the local group, they had larger defects, were more often in congestive heart failure and were more often catheterized after 6 months of age.In the whole series, 31 of 62 had congestive heart failure. This began before 6 months of age in all and occurred much earlier in premature than in full term infants.In the local group, spontaneous functional closure of the ventricular septal defect took place in 13 of 36 (36%); this was proved by recatheterization in 10 and autopsy in 1. Marked decrease in size of the defect took place in another 10 (28%). In the referred group (26 cases) there were 2 (8%) spontaneous closures and 7 (27%) reductions in defect size. For the whole series, 52 per cent of the patients had defects which were known to have closed or become smaller; 32 per cent were doing well clinically (most were not recatheterized so that some of these defects might have become smaller); and only 16 per cent of the infants were seriously affected by their lesions. Complete closure occurred between 7 and 12 months of age, most often when the defect was small, but it could occur with large defects. Decrease in defect size, however, occurred as frequently with large as with small defects, and was found in many children with congestive heart failure and pulmonary arterial hypertension.The 10 infants with large ventricular septal defects who did not do well included 4 who had pulmonary arterial banding for severe, uncontrolled congestive heart failure (2 died); 1 premature baby who died at 6 weeks of age with severe congestive heart failure; and 5 with a very high pulmonary vascular resistance. One of these 5 had a high pulmonary vascular resistance when 1 month old, and this did not change significantly at 5 and 13 months of age. The other 4, however, had low pulmonary vascular resistance when first catheterized at 3, 6, 6 and 11 months of age, respectively, and significant rises of resistance to pathologic levels when recatheterized at 9, 15, 26 and 40 months of age, respectively. This suggests that a rise in pulmonary vascular resistance in infancy is not rare when the ventricular septal defect is big and that, even in these children, the pulmonary vascular resistance does not usually persist at the high level present at birth but first falls post natally to normal levels before rising.These 5 infants all had large left to right shunts, high left atrial pressures and marked pulmonary arterial hypertension. In 3 out of the 4 whose pulmonary vascular resistances rose after the first catheterization, there were no distinctive clinical indications of this rise which was detected only at recatheterization. The other child had increasing right ventricular hypertrophy on the electrocardiogram.From these findings we suggest that, since in infancy spontaneous closure or reduction in size of the ventricular septal defect is so common, most infants should be treated conservatively in the hope that spontaneous improvement will occur. If the defect is very big and the child is in severe congestive heart failure which cannot be well controlled, then surgical closure of the defect or banding of the pulmonary artery should be done. If, however, the defect is big but congestive heart failure can be controlled medically, then the infant should be followed up with the hope that spontaneous improvement will occur. Whether clinical improvement occurs or not, the infant should be recatheterized six to nine months after the initial catheterization to determine if a rise in pulmonary vascular resistance has occurred; if it has then surgery should be advised.

Pediatric cardiologyLeft to right atrial shunts in infants☆

AbstractAt cardiac catheterization, 24 infants less than 1 year old had a left to right atrial shunt without other major cardiac lesions. Only 6 infants had cardiac symptoms, and there was no relation between symptoms and the size of the shunt. Six infants (including 2 who were less than 1 week old and were asymptomatic) had a pulmonary to systemic flow ratio (Qp:Qs) greater than 3. Clinical features were often atypical. Fifteen had harsh grade 3–46 systolic murmurs; only 10 had wide fixed splitting of the second heart sound. Splitting of the second heart sound was unrelated to the Qp:Qs. Two infants with a large shunt had no cardiomegaly. Electrocardiograms were not typical of atrial septal defect.The shunt disappeared in 9 of 12 infants who underwent recatheterization and was not clinically detectable in 3 others whose shunt was large when under 2 weeks of age. Many shunts probably occurred through an incompetent foramen ovale, sometimes secondary to left-sided lesions. However, some children had a classic atrial septal defect that closed spontaneously. Since spontaneous closure of atrial septal defects occurs and there are many atypical features of these lesions in infants and younger children, it is possible that the incidence of this lesion is underestimated in infancy and early childhood.

Clinical studyThe natural history of ventricular septal defects in infancy☆

AbstractSixty-two infants with ventricular septal defects were first catheterized under 1 year of age and followed up for one to five years; 40 were recatheterized.Thirty-six infants were born in the Bronx Municipal Hospital Center or Lincoln Hospital (local group). These patients were unselected, in that the decision to catheterize them was based only on the clinical diagnosis of a ventricular septal defect and not on an assessment of its size or the presence of symptoms. As far as we know, these infants include most if not all of the children born with ventricular septal defects in these hospitals during the period of study. With this assumption, the incidence of ventricular septal defects per 1,000 live births was 0.94 for full term infants, 4.51 for premature infants, and 1.35 for the whole group.Twenty-six other infants were referred from other hospitals or doctors (referred group), and, compared to the local group, they had larger defects, were more often in congestive heart failure and were more often catheterized after 6 months of age.In the whole series, 31 of 62 had congestive heart failure. This began before 6 months of age in all and occurred much earlier in premature than in full term infants.In the local group, spontaneous functional closure of the ventricular septal defect took place in 13 of 36 (36%); this was proved by recatheterization in 10 and autopsy in 1. Marked decrease in size of the defect took place in another 10 (28%). In the referred group (26 cases) there were 2 (8%) spontaneous closures and 7 (27%) reductions in defect size. For the whole series, 52 per cent of the patients had defects which were known to have closed or become smaller; 32 per cent were doing well clinically (most were not recatheterized so that some of these defects might have become smaller); and only 16 per cent of the infants were seriously affected by their lesions. Complete closure occurred between 7 and 12 months of age, most often when the defect was small, but it could occur with large defects. Decrease in defect size, however, occurred as frequently with large as with small defects, and was found in many children with congestive heart failure and pulmonary arterial hypertension.The 10 infants with large ventricular septal defects who did not do well included 4 who had pulmonary arterial banding for severe, uncontrolled congestive heart failure (2 died); 1 premature baby who died at 6 weeks of age with severe congestive heart failure; and 5 with a very high pulmonary vascular resistance. One of these 5 had a high pulmonary vascular resistance when 1 month old, and this did not change significantly at 5 and 13 months of age. The other 4, however, had low pulmonary vascular resistance when first catheterized at 3, 6, 6 and 11 months of age, respectively, and significant rises of resistance to pathologic levels when recatheterized at 9, 15, 26 and 40 months of age, respectively. This suggests that a rise in pulmonary vascular resistance in infancy is not rare when the ventricular septal defect is big and that, even in these children, the pulmonary vascular resistance does not usually persist at the high level present at birth but first falls post natally to normal levels before rising.These 5 infants all had large left to right shunts, high left atrial pressures and marked pulmonary arterial hypertension. In 3 out of the 4 whose pulmonary vascular resistances rose after the first catheterization, there were no distinctive clinical indications of this rise which was detected only at recatheterization. The other child had increasing right ventricular hypertrophy on the electrocardiogram.From these findings we suggest that, since in infancy spontaneous closure or reduction in size of the ventricular septal defect is so common, most infants should be treated conservatively in the hope that spontaneous improvement will occur. If the defect is very big and the child is in severe congestive heart failure which cannot be well controlled, then surgical closure of the defect or banding of the pulmonary artery should be done. If, however, the defect is big but congestive heart failure can be controlled medically, then the infant should be followed up with the hope that spontaneous improvement will occur. Whether clinical improvement occurs or not, the infant should be recatheterized six to nine months after the initial catheterization to determine if a rise in pulmonary vascular resistance has occurred; if it has then surgery should be advised.

Pediatric cardiologyLeft to right atrial shunts in infants☆

AbstractAt cardiac catheterization, 24 infants less than 1 year old had a left to right atrial shunt without other major cardiac lesions. Only 6 infants had cardiac symptoms, and there was no relation between symptoms and the size of the shunt. Six infants (including 2 who were less than 1 week old and were asymptomatic) had a pulmonary to systemic flow ratio (Qp:Qs) greater than 3. Clinical features were often atypical. Fifteen had harsh grade 3–46 systolic murmurs; only 10 had wide fixed splitting of the second heart sound. Splitting of the second heart sound was unrelated to the Qp:Qs. Two infants with a large shunt had no cardiomegaly. Electrocardiograms were not typical of atrial septal defect.The shunt disappeared in 9 of 12 infants who underwent recatheterization and was not clinically detectable in 3 others whose shunt was large when under 2 weeks of age. Many shunts probably occurred through an incompetent foramen ovale, sometimes secondary to left-sided lesions. However, some children had a classic atrial septal defect that closed spontaneously. Since spontaneous closure of atrial septal defects occurs and there are many atypical features of these lesions in infants and younger children, it is possible that the incidence of this lesion is underestimated in infancy and early childhood.

Clinical studyThe natural history of ventricular septal defects in infancy☆

AbstractSixty-two infants with ventricular septal defects were first catheterized under 1 year of age and followed up for one to five years; 40 were recatheterized.Thirty-six infants were born in the Bronx Municipal Hospital Center or Lincoln Hospital (local group). These patients were unselected, in that the decision to catheterize them was based only on the clinical diagnosis of a ventricular septal defect and not on an assessment of its size or the presence of symptoms. As far as we know, these infants include most if not all of the children born with ventricular septal defects in these hospitals during the period of study. With this assumption, the incidence of ventricular septal defects per 1,000 live births was 0.94 for full term infants, 4.51 for premature infants, and 1.35 for the whole group.Twenty-six other infants were referred from other hospitals or doctors (referred group), and, compared to the local group, they had larger defects, were more often in congestive heart failure and were more often catheterized after 6 months of age.In the whole series, 31 of 62 had congestive heart failure. This began before 6 months of age in all and occurred much earlier in premature than in full term infants.In the local group, spontaneous functional closure of the ventricular septal defect took place in 13 of 36 (36%); this was proved by recatheterization in 10 and autopsy in 1. Marked decrease in size of the defect took place in another 10 (28%). In the referred group (26 cases) there were 2 (8%) spontaneous closures and 7 (27%) reductions in defect size. For the whole series, 52 per cent of the patients had defects which were known to have closed or become smaller; 32 per cent were doing well clinically (most were not recatheterized so that some of these defects might have become smaller); and only 16 per cent of the infants were seriously affected by their lesions. Complete closure occurred between 7 and 12 months of age, most often when the defect was small, but it could occur with large defects. Decrease in defect size, however, occurred as frequently with large as with small defects, and was found in many children with congestive heart failure and pulmonary arterial hypertension.The 10 infants with large ventricular septal defects who did not do well included 4 who had pulmonary arterial banding for severe, uncontrolled congestive heart failure (2 died); 1 premature baby who died at 6 weeks of age with severe congestive heart failure; and 5 with a very high pulmonary vascular resistance. One of these 5 had a high pulmonary vascular resistance when 1 month old, and this did not change significantly at 5 and 13 months of age. The other 4, however, had low pulmonary vascular resistance when first catheterized at 3, 6, 6 and 11 months of age, respectively, and significant rises of resistance to pathologic levels when recatheterized at 9, 15, 26 and 40 months of age, respectively. This suggests that a rise in pulmonary vascular resistance in infancy is not rare when the ventricular septal defect is big and that, even in these children, the pulmonary vascular resistance does not usually persist at the high level present at birth but first falls post natally to normal levels before rising.These 5 infants all had large left to right shunts, high left atrial pressures and marked pulmonary arterial hypertension. In 3 out of the 4 whose pulmonary vascular resistances rose after the first catheterization, there were no distinctive clinical indications of this rise which was detected only at recatheterization. The other child had increasing right ventricular hypertrophy on the electrocardiogram.From these findings we suggest that, since in infancy spontaneous closure or reduction in size of the ventricular septal defect is so common, most infants should be treated conservatively in the hope that spontaneous improvement will occur. If the defect is very big and the child is in severe congestive heart failure which cannot be well controlled, then surgical closure of the defect or banding of the pulmonary artery should be done. If, however, the defect is big but congestive heart failure can be controlled medically, then the infant should be followed up with the hope that spontaneous improvement will occur. Whether clinical improvement occurs or not, the infant should be recatheterized six to nine months after the initial catheterization to determine if a rise in pulmonary vascular resistance has occurred; if it has then surgery should be advised.

Pediatric cardiologyLeft to right atrial shunts in infants☆

AbstractAt cardiac catheterization, 24 infants less than 1 year old had a left to right atrial shunt without other major cardiac lesions. Only 6 infants had cardiac symptoms, and there was no relation between symptoms and the size of the shunt. Six infants (including 2 who were less than 1 week old and were asymptomatic) had a pulmonary to systemic flow ratio (Qp:Qs) greater than 3. Clinical features were often atypical. Fifteen had harsh grade 3–46 systolic murmurs; only 10 had wide fixed splitting of the second heart sound. Splitting of the second heart sound was unrelated to the Qp:Qs. Two infants with a large shunt had no cardiomegaly. Electrocardiograms were not typical of atrial septal defect.The shunt disappeared in 9 of 12 infants who underwent recatheterization and was not clinically detectable in 3 others whose shunt was large when under 2 weeks of age. Many shunts probably occurred through an incompetent foramen ovale, sometimes secondary to left-sided lesions. However, some children had a classic atrial septal defect that closed spontaneously. Since spontaneous closure of atrial septal defects occurs and there are many atypical features of these lesions in infants and younger children, it is possible that the incidence of this lesion is underestimated in infancy and early childhood.

Clinical studyThe natural history of ventricular septal defects in infancy☆

AbstractSixty-two infants with ventricular septal defects were first catheterized under 1 year of age and followed up for one to five years; 40 were recatheterized.Thirty-six infants were born in the Bronx Municipal Hospital Center or Lincoln Hospital (local group). These patients were unselected, in that the decision to catheterize them was based only on the clinical diagnosis of a ventricular septal defect and not on an assessment of its size or the presence of symptoms. As far as we know, these infants include most if not all of the children born with ventricular septal defects in these hospitals during the period of study. With this assumption, the incidence of ventricular septal defects per 1,000 live births was 0.94 for full term infants, 4.51 for premature infants, and 1.35 for the whole group.Twenty-six other infants were referred from other hospitals or doctors (referred group), and, compared to the local group, they had larger defects, were more often in congestive heart failure and were more often catheterized after 6 months of age.In the whole series, 31 of 62 had congestive heart failure. This began before 6 months of age in all and occurred much earlier in premature than in full term infants.In the local group, spontaneous functional closure of the ventricular septal defect took place in 13 of 36 (36%); this was proved by recatheterization in 10 and autopsy in 1. Marked decrease in size of the defect took place in another 10 (28%). In the referred group (26 cases) there were 2 (8%) spontaneous closures and 7 (27%) reductions in defect size. For the whole series, 52 per cent of the patients had defects which were known to have closed or become smaller; 32 per cent were doing well clinically (most were not recatheterized so that some of these defects might have become smaller); and only 16 per cent of the infants were seriously affected by their lesions. Complete closure occurred between 7 and 12 months of age, most often when the defect was small, but it could occur with large defects. Decrease in defect size, however, occurred as frequently with large as with small defects, and was found in many children with congestive heart failure and pulmonary arterial hypertension.The 10 infants with large ventricular septal defects who did not do well included 4 who had pulmonary arterial banding for severe, uncontrolled congestive heart failure (2 died); 1 premature baby who died at 6 weeks of age with severe congestive heart failure; and 5 with a very high pulmonary vascular resistance. One of these 5 had a high pulmonary vascular resistance when 1 month old, and this did not change significantly at 5 and 13 months of age. The other 4, however, had low pulmonary vascular resistance when first catheterized at 3, 6, 6 and 11 months of age, respectively, and significant rises of resistance to pathologic levels when recatheterized at 9, 15, 26 and 40 months of age, respectively. This suggests that a rise in pulmonary vascular resistance in infancy is not rare when the ventricular septal defect is big and that, even in these children, the pulmonary vascular resistance does not usually persist at the high level present at birth but first falls post natally to normal levels before rising.These 5 infants all had large left to right shunts, high left atrial pressures and marked pulmonary arterial hypertension. In 3 out of the 4 whose pulmonary vascular resistances rose after the first catheterization, there were no distinctive clinical indications of this rise which was detected only at recatheterization. The other child had increasing right ventricular hypertrophy on the electrocardiogram.From these findings we suggest that, since in infancy spontaneous closure or reduction in size of the ventricular septal defect is so common, most infants should be treated conservatively in the hope that spontaneous improvement will occur. If the defect is very big and the child is in severe congestive heart failure which cannot be well controlled, then surgical closure of the defect or banding of the pulmonary artery should be done. If, however, the defect is big but congestive heart failure can be controlled medically, then the infant should be followed up with the hope that spontaneous improvement will occur. Whether clinical improvement occurs or not, the infant should be recatheterized six to nine months after the initial catheterization to determine if a rise in pulmonary vascular resistance has occurred; if it has then surgery should be advised.

Pediatric cardiologyLeft to right atrial shunts in infants☆

AbstractAt cardiac catheterization, 24 infants less than 1 year old had a left to right atrial shunt without other major cardiac lesions. Only 6 infants had cardiac symptoms, and there was no relation between symptoms and the size of the shunt. Six infants (including 2 who were less than 1 week old and were asymptomatic) had a pulmonary to systemic flow ratio (Qp:Qs) greater than 3. Clinical features were often atypical. Fifteen had harsh grade 3–46 systolic murmurs; only 10 had wide fixed splitting of the second heart sound. Splitting of the second heart sound was unrelated to the Qp:Qs. Two infants with a large shunt had no cardiomegaly. Electrocardiograms were not typical of atrial septal defect.The shunt disappeared in 9 of 12 infants who underwent recatheterization and was not clinically detectable in 3 others whose shunt was large when under 2 weeks of age. Many shunts probably occurred through an incompetent foramen ovale, sometimes secondary to left-sided lesions. However, some children had a classic atrial septal defect that closed spontaneously. Since spontaneous closure of atrial septal defects occurs and there are many atypical features of these lesions in infants and younger children, it is possible that the incidence of this lesion is underestimated in infancy and early childhood.

Clinical studyThe natural history of ventricular septal defects in infancy☆

AbstractSixty-two infants with ventricular septal defects were first catheterized under 1 year of age and followed up for one to five years; 40 were recatheterized.Thirty-six infants were born in the Bronx Municipal Hospital Center or Lincoln Hospital (local group). These patients were unselected, in that the decision to catheterize them was based only on the clinical diagnosis of a ventricular septal defect and not on an assessment of its size or the presence of symptoms. As far as we know, these infants include most if not all of the children born with ventricular septal defects in these hospitals during the period of study. With this assumption, the incidence of ventricular septal defects per 1,000 live births was 0.94 for full term infants, 4.51 for premature infants, and 1.35 for the whole group.Twenty-six other infants were referred from other hospitals or doctors (referred group), and, compared to the local group, they had larger defects, were more often in congestive heart failure and were more often catheterized after 6 months of age.In the whole series, 31 of 62 had congestive heart failure. This began before 6 months of age in all and occurred much earlier in premature than in full term infants.In the local group, spontaneous functional closure of the ventricular septal defect took place in 13 of 36 (36%); this was proved by recatheterization in 10 and autopsy in 1. Marked decrease in size of the defect took place in another 10 (28%). In the referred group (26 cases) there were 2 (8%) spontaneous closures and 7 (27%) reductions in defect size. For the whole series, 52 per cent of the patients had defects which were known to have closed or become smaller; 32 per cent were doing well clinically (most were not recatheterized so that some of these defects might have become smaller); and only 16 per cent of the infants were seriously affected by their lesions. Complete closure occurred between 7 and 12 months of age, most often when the defect was small, but it could occur with large defects. Decrease in defect size, however, occurred as frequently with large as with small defects, and was found in many children with congestive heart failure and pulmonary arterial hypertension.The 10 infants with large ventricular septal defects who did not do well included 4 who had pulmonary arterial banding for severe, uncontrolled congestive heart failure (2 died); 1 premature baby who died at 6 weeks of age with severe congestive heart failure; and 5 with a very high pulmonary vascular resistance. One of these 5 had a high pulmonary vascular resistance when 1 month old, and this did not change significantly at 5 and 13 months of age. The other 4, however, had low pulmonary vascular resistance when first catheterized at 3, 6, 6 and 11 months of age, respectively, and significant rises of resistance to pathologic levels when recatheterized at 9, 15, 26 and 40 months of age, respectively. This suggests that a rise in pulmonary vascular resistance in infancy is not rare when the ventricular septal defect is big and that, even in these children, the pulmonary vascular resistance does not usually persist at the high level present at birth but first falls post natally to normal levels before rising.These 5 infants all had large left to right shunts, high left atrial pressures and marked pulmonary arterial hypertension. In 3 out of the 4 whose pulmonary vascular resistances rose after the first catheterization, there were no distinctive clinical indications of this rise which was detected only at recatheterization. The other child had increasing right ventricular hypertrophy on the electrocardiogram.From these findings we suggest that, since in infancy spontaneous closure or reduction in size of the ventricular septal defect is so common, most infants should be treated conservatively in the hope that spontaneous improvement will occur. If the defect is very big and the child is in severe congestive heart failure which cannot be well controlled, then surgical closure of the defect or banding of the pulmonary artery should be done. If, however, the defect is big but congestive heart failure can be controlled medically, then the infant should be followed up with the hope that spontaneous improvement will occur. Whether clinical improvement occurs or not, the infant should be recatheterized six to nine months after the initial catheterization to determine if a rise in pulmonary vascular resistance has occurred; if it has then surgery should be advised.

Pediatric cardiologyLeft to right atrial shunts in infants☆

AbstractAt cardiac catheterization, 24 infants less than 1 year old had a left to right atrial shunt without other major cardiac lesions. Only 6 infants had cardiac symptoms, and there was no relation between symptoms and the size of the shunt. Six infants (including 2 who were less than 1 week old and were asymptomatic) had a pulmonary to systemic flow ratio (Qp:Qs) greater than 3. Clinical features were often atypical. Fifteen had harsh grade 3–46 systolic murmurs; only 10 had wide fixed splitting of the second heart sound. Splitting of the second heart sound was unrelated to the Qp:Qs. Two infants with a large shunt had no cardiomegaly. Electrocardiograms were not typical of atrial septal defect.The shunt disappeared in 9 of 12 infants who underwent recatheterization and was not clinically detectable in 3 others whose shunt was large when under 2 weeks of age. Many shunts probably occurred through an incompetent foramen ovale, sometimes secondary to left-sided lesions. However, some children had a classic atrial septal defect that closed spontaneously. Since spontaneous closure of atrial septal defects occurs and there are many atypical features of these lesions in infants and younger children, it is possible that the incidence of this lesion is underestimated in infancy and early childhood.

Chapter 3 - Some Practical Aspects

AbstractThis chapter begins with a survey of statistical programs. It then discusses variables, measurement scales, how to display data sets with comments on accuracy and precision, and ends with a brief discussion of statistical notation and the concept of weighted values.

Chapter 5 - Basic Probability

AbstractThis chapter describes the bases of probability theory, including the addition and multiplication rules, and leads by way of conditional probability to the simple presentation of Bayes' theorem. Some basic problems are worked out, including the famous Monty Hall and birthday problems. Understanding these issues is central to medical diagnoses.

Chapter 8 - Other Continuous Distributions

AbstractAlthough the normal distribution is the most frequently used continuous distribution in statistics, it is not the only one. Other important distributions are the uniform, exponential, logarithmic, Weibull, chi-square, and F distributions. The first four of these have use in specific circumstances. The last two are used very frequently; their bases are presented here, but each receives more detailed chapters later in the book.

Chapter 10 - Hypothesis Testing: The Null Hypothesis, Significance, and Type I Error

AbstractThe null hypothesis is that there is no difference between two (or more) groups apart from random variation, and is central to statistical hypothesis testing. If the differences between the groups are more than can be accounted for by random variation, then the judgment may be made that the groups are different and unlikely to have been drawn from the same population. This judgment is never absolute, and is given a probability (based on the calculations from the data). If the probability that the groups came from the same population is very low, the difference is termed “significant,” but there is much variation in how the term is interpreted. If we decide that the two groups are different, but in fact they really did come from the same population, then we have made a Type I error. The calculated probability value gives the probability of that error.

Chapter 13 - Hypergeometric Distribution

AbstractThis chapter gives the basis for Fisher's test (Fisher–Irwin test), which is an important variant of the chi-square test and is often used in its place.

Chapter 19 - Negative Binomial Distribution

AbstractThis distribution is used to predict the number of failures before a success occurs in a sequence of Bernoulli trials. In addition, it is used to model a distribution that shows excessive variance because some categories show an excess and some a deficiency of counts. For example, filariasis is a rare disease that might be expected to have a distribution with the number of affected patients per village following a Poisson distribution. In fact, there are an excess number of villages with no disease or with much disease, and relatively few villages have intermediate numbers. This type of distribution is referred to as clumped, contagious, or overdispersed, and is often well fitted by a negative binomial distribution. The inference drawn is that some villages are in areas without insect vectors and others where the vectors are very plentiful.

Chapter 23 - t-Test Variants: Crossover Tests, Equivalence Tests

AbstractThe t-test can be modified to allow the same subjects to be used to test two different treatments (one of which may be a placebo) while considering any effects that the order of administration of the treatments might have—the crossover test. In addition the t-test can be modified to demonstrate that a new treatment is equivalent to or not inferior to an existing treatment.

Chapter 25 - Analysis of Variance I. One-Way

AbstractThis method is the basis for almost all statistical hypothesis testing, and is described in detail. It allows us to compare the means of >2 groups to determine if they come from the same or different populations, and to set confidence limits on the differences. New concepts such as a priori and a posteriori inferences are introduced, as well as a detailed discussion of dealing with the multiple comparison problem. Heterogeneity of variance is analyzed and recommendations made for handling it.

Chapter 33 - Logistic Regression

AbstractWe often need to determine probabilities for a yes or no outcome in medical diagnosis, and require special methods. These are basically simple but computationally tedious, and usually require professional programs for their implementation.

Chapter 34 - Poisson Regression

AbstractThis is a complicated subject that deals with predicting counts that cannot be handled by any other form of regression. Its methodology is described, as well as alternative methods of analysis that will need statistical consultation.

Back | Next (Page 1 of 2)
Advertisement
Join Copernicus Academic and get access to over 12 million papers authored by 7+ million academics.
Join for free!

Contact us