A COSINOR Model Approach to Variation in Seasonality of Number of Patients Enrollment at the Adult ART Clinic.

International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VI, Issue IX, September 2021|ISSN 2454-6194

A COSINOR Model Approach to Variation in Seasonality of Number of Patients Enrollment at the Adult ART Clinic.

Oyenuga, Iyabode Favour and Akinyemi, Joshua Odunayo
¹Department of Mathematics and Statistics, Faculty of Science, The Polytechnic, Ibadan, Nigeria.
²Department of Epidemiology and Medical Statistics, Faculty of Public Health, College of Medicine,
University of Ibadan, Nigeria.

Abstract: Objective: This study aims to explore broadly the seasonal variation of the number of patients’ enrollment at the adult antiretroviral treatment (ART) clinic, University College Hospital, Ibadan, Nigeria. It also described the variations in seasonality distinctly for the segment 2006-2008, 2009-2011, 2012-2014, 2015-2017 and 2006-2017 in order to ascertain whether there have been changes in time of maxima over the years
Methods: Analysis of data extracted from electronic records of adult patients enrolled for HIV/AIDS care and treatment at the ART clinic, University College Hospital, Ibadan which covers a period from January 2006 to December 2017 spanning twelve years. Variations on seasonality was investigated using COSINOR model approach.
Results:There was a steady decrease in mean number of patients’ enrollment from 212 in 2006-2008 to 63 in 2015-2017 which accounted for about 7% extent of seasonal variation above and below the annual mean with the estimated peak occurring in the month of October. The extent of the seasonal variations above and below the mean level was about 7%, 11%, 16% and 10% in 2006-2008, 2009-2011, 2012-2014 and 2015-2017 respectively.
Conclusion: The entire model fitted for the COSINOR model had a good fit except for the year 2006-2008 which is not statistically significant. The seasonal variation in the number of patients enrollment rate showed the peak relatively constant for the last quarter of the first nine years but later shifted to May , that is, the second quarter of the year for the last three years. Changing in seasonal pattern may be as a result of changes in funding cycles in Nigeria by the Federal government because more financial resources are being allocated for the procurement of the medicines.

Keywords: COSINOR, ART, enrollment, seasonal variation.

I.INTRODUCTION

The cosinor models introduced by Halberg et al. (1967) and Nelson et. al (1979) and was used to model seasonality. It captures a seasonal pattern using sinusoid and suitable for relatively seasonal patterns that are symmetric and stationary. It is used in diverse applications to fit one or more cosine curves of a given period to rhythmic data [Elkum and Myles (2006); Klisch et al. (2006); Nikhil et al (2007). The percentage of the total variance in the time series data explained by the cosine-curve-model approximation of the

Read More