Limits...
Time series analysis of malaria in Afghanistan: using ARIMA models to predict future trends in incidence

View Article: PubMed Central - PubMed

ABSTRACT

Background: Malaria remains endemic in Afghanistan. National control and prevention strategies would be greatly enhanced through a better ability to forecast future trends in disease incidence. It is, therefore, of interest to develop a predictive tool for malaria patterns based on the current passive and affordable surveillance system in this resource-limited region.

Methods: This study employs data from Ministry of Public Health monthly reports from January 2005 to September 2015. Malaria incidence in Afghanistan was forecasted using autoregressive integrated moving average (ARIMA) models in order to build a predictive tool for malaria surveillance. Environmental and climate data were incorporated to assess whether they improve predictive power of models.

Results: Two models were identified, each appropriate for different time horizons. For near-term forecasts, malaria incidence can be predicted based on the number of cases in the four previous months and 12 months prior (Model 1); for longer-term prediction, malaria incidence can be predicted using the rates 1 and 12 months prior (Model 2). Next, climate and environmental variables were incorporated to assess whether the predictive power of proposed models could be improved. Enhanced vegetation index was found to have increased the predictive accuracy of longer-term forecasts.

Conclusion: Results indicate ARIMA models can be applied to forecast malaria patterns in Afghanistan, complementing current surveillance systems. The models provide a means to better understand malaria dynamics in a resource-limited context with minimal data input, yielding forecasts that can be used for public health planning at the national level.

Electronic supplementary material: The online version of this article (doi:10.1186/s12936-016-1602-1) contains supplementary material, which is available to authorized users.

No MeSH data available.


Related in: MedlinePlus

Seasonal variation of malaria and environmental variables (2005–2014). From top left in clock wise order: a monthly variation of malaria, b monthly variation of Temperature, d monthly variation of rainfall, c monthly variation of vegetation index
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC5120433&req=5

Fig1: Seasonal variation of malaria and environmental variables (2005–2014). From top left in clock wise order: a monthly variation of malaria, b monthly variation of Temperature, d monthly variation of rainfall, c monthly variation of vegetation index

Mentions: The dataset covers 129 months, starting from January 2005 to September 2015. The total number of suspected (including confirmed) malaria cases reported throughout the period was 2,243,452 with a mean of 20,772 clinical cases per month, and standard error of 1097 cases. The number of reported cases per month ranged from 4309 to 47,779, consistent with the seasonal nature of malaria in the country. Indeed, looking at the seasonal distribution of cases over the years (Fig. 1a), malaria cases peak between June–September, around the time when temperature is high and rainfall low (Fig. 1b, d), and lag vegetation variation by few months (Fig. 1c). Geographically, in descending order, eastern (1,351,530), north eastern (366,635), northern (239,230), southern (145,220), central (87,227), and western (53,610) regions report the most cases.Fig. 1


Time series analysis of malaria in Afghanistan: using ARIMA models to predict future trends in incidence
Seasonal variation of malaria and environmental variables (2005–2014). From top left in clock wise order: a monthly variation of malaria, b monthly variation of Temperature, d monthly variation of rainfall, c monthly variation of vegetation index
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC5120433&req=5

Fig1: Seasonal variation of malaria and environmental variables (2005–2014). From top left in clock wise order: a monthly variation of malaria, b monthly variation of Temperature, d monthly variation of rainfall, c monthly variation of vegetation index
Mentions: The dataset covers 129 months, starting from January 2005 to September 2015. The total number of suspected (including confirmed) malaria cases reported throughout the period was 2,243,452 with a mean of 20,772 clinical cases per month, and standard error of 1097 cases. The number of reported cases per month ranged from 4309 to 47,779, consistent with the seasonal nature of malaria in the country. Indeed, looking at the seasonal distribution of cases over the years (Fig. 1a), malaria cases peak between June–September, around the time when temperature is high and rainfall low (Fig. 1b, d), and lag vegetation variation by few months (Fig. 1c). Geographically, in descending order, eastern (1,351,530), north eastern (366,635), northern (239,230), southern (145,220), central (87,227), and western (53,610) regions report the most cases.Fig. 1

View Article: PubMed Central - PubMed

ABSTRACT

Background: Malaria remains endemic in Afghanistan. National control and prevention strategies would be greatly enhanced through a better ability to forecast future trends in disease incidence. It is, therefore, of interest to develop a predictive tool for malaria patterns based on the current passive and affordable surveillance system in this resource-limited region.

Methods: This study employs data from Ministry of Public Health monthly reports from January 2005 to September 2015. Malaria incidence in Afghanistan was forecasted using autoregressive integrated moving average (ARIMA) models in order to build a predictive tool for malaria surveillance. Environmental and climate data were incorporated to assess whether they improve predictive power of models.

Results: Two models were identified, each appropriate for different time horizons. For near-term forecasts, malaria incidence can be predicted based on the number of cases in the four previous months and 12 months prior (Model 1); for longer-term prediction, malaria incidence can be predicted using the rates 1 and 12 months prior (Model 2). Next, climate and environmental variables were incorporated to assess whether the predictive power of proposed models could be improved. Enhanced vegetation index was found to have increased the predictive accuracy of longer-term forecasts.

Conclusion: Results indicate ARIMA models can be applied to forecast malaria patterns in Afghanistan, complementing current surveillance systems. The models provide a means to better understand malaria dynamics in a resource-limited context with minimal data input, yielding forecasts that can be used for public health planning at the national level.

Electronic supplementary material: The online version of this article (doi:10.1186/s12936-016-1602-1) contains supplementary material, which is available to authorized users.

No MeSH data available.


Related in: MedlinePlus