Layout 1


ISDS Annual Conference Proceedings 2012. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and
reproduction in any medium, provided the original work is properly cited.

ISDS 2012 Conference Abstracts

Applying Zero-inflated Mixed Model to School
Absenteeism Surveillance in Rural China
Xiaoxiao Song1, Tao Tao1, Qi Zhao1, Fuqiang Yang2, Palm Lars3, Diwan Vinod4, Hui Yuan2
and Biao* Xu1

1School of Public Health, Fudan University, Shanghai, China; 2Jiangxi Provincial Center for Disease Prevention and Control,
nanchang, China; 3Future Position X, Gavle, Sweden; 4ICHAR, Karolinska Instituet, Stockholm, Sweden

Objective
To describe and explore the spatial and temporal variability via

ZIMM for absenteeism surveillance in primary school for early de-
tection of infectious disease outbreak in rural China.

Introduction
Absenteeism has great advantages in promoting the early detec-

tion of epidemics1. Since August 2011, an integrated syndromic sur-
veillance project (ISSC) has been implemented in China2.
Distribution of the absenteeism generally are asymmetry, zero infla-
tion, truncation and non-independence3. For handling these encum-
brances, we should apply the Zero-inflated Mixed Model (ZIMM).

Methods
Data for this study was obtained from the web-based data of ISSC

in 62 primary schools in two counties of Jiangxi province, China
from April 1th, 2012 to June 30st, 2012. The ZIMM was used to ex-
plore: 1)the temporal and spatial variability regarding occurrence and
intensity of absenteeism simultaneously, and 2) the heterogeneity
among the reporting primary schools by introducing random effects
into the intercepts. The analyse was processed in the SAS procedure
NLMIXED4.

Results
The total 4914 absenteeism events were reported in the 62 primary

schools in the study period. The rate of zero report was 49.88% (Fig.1).
According to ZIMM, there are fixed and random effect parameters in
this model (Table 1). Firstly, for the fixed parameters, the spatial vari-
able (county) was not significantly different both the occurrence and in-
tensity model, while for the temporal variable (month), the probability
of absenteeism occurrence was significantly different over three months
(!=-0.165, p =0.026), suggesting a decreasing of school absenteeism
from April to June. Meanwhile, a statistical significant difference in the
intensity of absenteeism was also found over the three months (!=-
0.073, p=0.007). Secondly, the random effect of intensity model was
statistically significance (p=0.008), which strongly indicated a hetero-
geneity in intensity of absenteeism among the surveillance schools.
Whereas the random effect of occurrence model by logistic regression
showed a non-statistical difference (p=0.774) among the schools sug-
gesting the homogeneity in the occurrence of absenteeism.

Conclusions
School absenteeism data has greater uncertain than many other

sources and easier fluctuate by some factors such as holiday, season,
family status and geographic distribution. Thus, the spatial and tem-
poral dynamics should be taken into account in controlling fluctuate
of absenteeism. Moreover, school absenteeism data are correlated
within each school due to repeated measures. Applying the ZIMM,
the occurrences and intensity of absenteeism could be evaluated to re-
duce the bias and improve the prediction precision. The ZIMM is an
appropriate tool for health authorities in decision making for public
health events.

Table 1 Fixed parameters and variance components estimates for the
absenteeism using ZIMM

Fig. 1 Absenteeism from Apr. 1st to Jun. 30th 2012

Keywords
surveillance; absenteeism; zero-inflated mixed model; occurrence;
intensity

References

1. Lenert L, Kirsh D, Johnson J, Aryel RM. Absenteeism. In: Wagner
MM, Moore AW, Aryel RM, editors. Handbook of Biosurveillance
Burlington: Academic Press, 2006:361-68.

2. Yan W-r, Nie S-f, Xu B, Dong H-j, Palm L, Diwan VK. Establishing a
web-based integrated surveillance system for early detection of in-
fectious disease epidemic in rural China: a field experimental study.
Bmc Medical Informatics and Decision Making 2012;12.

3. Calama R, Mutke S, Tome J, Gordo J, Montero G, Tome M. Modelling
spatial and temporal variability in a zero-inflated variable: The case
of stone pine (Pinus pinea L.) cone production. Ecological Modelling
2011;222(3):606-18.

4. Tooze JA, Grunwald GK, Jones RH. Analysis of repeated measures
data with clumping at zero. Statistical Methods in Medical Research
2002;11(4):341-55.

*Biao Xu
E-mail: bxu@shmu.edu.cn

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 5(1):e92, 2013