Advanced

Insulinantikroppars påverkan på risken för hypoglykemi hos patienter med diabetes typ 2 behandlade med inhalerat eller subkutant insulin

Wahlgren, Linnea LU (2010) STAM01 20101
Department of Statistics
Abstract
A higher levels of insulin antibodies has been seen with diabetes patients who inhale insulin instead of injecting it subcutaneously, which is the most common way to take insulin. Whether or not this increased level of insulin antibodies affect the patients condition is not fully investigated in all aspects. This essay aims to, in different ways describe and analyse if a higher level of insulin antibodies is related to a higher risk of hypoglycemia (hypo, unhealthy low levels of blood sugar). Several different approaches are presented and different statistical methods applied to data from a clinical trial with 618 patients with diabetes type 2. We consider the hypo’s severity, proportion of patients with hypo, number of hypos per patient,... (More)
A higher levels of insulin antibodies has been seen with diabetes patients who inhale insulin instead of injecting it subcutaneously, which is the most common way to take insulin. Whether or not this increased level of insulin antibodies affect the patients condition is not fully investigated in all aspects. This essay aims to, in different ways describe and analyse if a higher level of insulin antibodies is related to a higher risk of hypoglycemia (hypo, unhealthy low levels of blood sugar). Several different approaches are presented and different statistical methods applied to data from a clinical trial with 618 patients with diabetes type 2. We consider the hypo’s severity, proportion of patients with hypo, number of hypos per patient, time to first hypo (with Kaplan Meier plots and Cox proportional hazards model) and different extensions of Cox proportional hazards model. The results are discussed and compared between the two different treatment methods (inhaling or injecting insulin) and between different levels of antibodies.
It is shown that the patients inhaling insulin run a lower risk of getting
hypos, and an increased level of antibodies is not shown to increase the risk of a hypo. (Less)
Please use this url to cite or link to this publication:
@misc{1746910,
  abstract     = {A higher levels of insulin antibodies has been seen with diabetes patients who inhale insulin instead of injecting it subcutaneously, which is the most common way to take insulin. Whether or not this increased level of insulin antibodies affect the patients condition is not fully investigated in all aspects. This essay aims to, in different ways describe and analyse if a higher level of insulin antibodies is related to a higher risk of hypoglycemia (hypo, unhealthy low levels of blood sugar). Several different approaches are presented and different statistical methods applied to data from a clinical trial with 618 patients with diabetes type 2. We consider the hypo’s severity, proportion of patients with hypo, number of hypos per patient, time to first hypo (with Kaplan Meier plots and Cox proportional hazards model) and different extensions of Cox proportional hazards model. The results are discussed and compared between the two different treatment methods (inhaling or injecting insulin) and between different levels of antibodies.
It is shown that the patients inhaling insulin run a lower risk of getting
hypos, and an increased level of antibodies is not shown to increase the risk of a hypo.},
  author       = {Wahlgren, Linnea},
  keyword      = {diabetes,Statistics,Poisson regression,Cox Proportional Hazards model,Negative binomial regression,Extended Cox models,Andersen and Gill,PWP,Kaplan Meier,WLW,Logistic regression,Survival analysis,Pearsons Chi-square test,Mantel-Haenszels Chi-square test,antibodies,insulin,inhaled insulin},
  language     = {swe},
  note         = {Student Paper},
  title        = {Insulinantikroppars påverkan på risken för hypoglykemi hos patienter med diabetes typ 2 behandlade med inhalerat eller subkutant insulin},
  year         = {2010},
}