Page 58 - Contributed Paper Session (CPS) - Volume 8
P. 58

CPS2184 M Lutfor Rahman



                                 Transmission network analysis of infectious
                                              disease in Portugal
                                                         1
                                                                      2
                                        M Lutfor Rahman , Carla Nunes
                         1  Institute of Statistical Research & Training, University of Dhaka, Bangladesh
                           2  National School of Public Health, NOVA University of Lisbon, Portugal

                  Abstract
                  Proper identification of the key factors responsible for an infectious disease
                  transmission  would  contribute  to  control  rapid  spread  of  the  disease.  This
                  study aims to model transmission network of an infectious disease focused on
                  a  case-study  of  tuberculosis  network  in  Portugal.  Characteristics  of  index-
                  patients (confirmed TB diseased), contacts and exposition were considered.
                  There were 495 participants in the study, considering 69 were index patients
                  (confirmed TB diseased) and 426 identified contacts.  Identified contacts were
                  screened at the TB centre and other relevant information was collected at the
                  time of screening. Some different methods were applied in a sequential order:
                  simple measures of association, binary logistic regression, hierarchical logistic
                  regression  and  network  logistic  regression.  These  different  models  have
                  different  assumptions,  namely  independency  between  observations  and
                  index-contact  relationship  of  the  network,  and  the  results  were  compared.
                  Based on binary logistic regression only the factors symptomatic period of
                  index  patients,  sputum,  age  of  contacts,  and  exposure  duration  were
                  responsible for contacts being infected. The hierarchical logistic  regression
                  selected  the  factors age,  diabetes  mellitus,  sleeping  together  as  important
                  variables for TB infection. The network regression reveals that the older age of
                  the contacts and household interaction are the risk factors of TB infection. The
                  network logistic regression was better to model the TB transmission using a
                  lower number of variables, being a promising tool in the analysis of infectious
                  disease with lower methodological limitations.

                  Keywords
                  Tuberculosis; transmission; network logistic regression; risk factors

                  1.  Introduction
                      Infectious disease, also known as communicable or transmissible disease,
                  is caused by infectious agents including viruses, bacteria, fungi, parasites, and
                  other  macro  parasites.  Emerging  infectious  diseases  continue  to  expose
                  national  and  global  unpreparedness  for  prevention  and  control  of  disease
                  outbreaks.  Tuberculosis  is  an  infectious  disease  usually  caused  by  the
                  Mycobacterium, thus the disease known as Mycobacterium tuberculosis or M.
                  Tuberculosis.  According  to  World  Health  Organization  (WHO)  tuberculosis

                                                                      47 | I S I   W S C   2 0 1 9
   53   54   55   56   57   58   59   60   61   62   63