Home Print this page Email this page
Users Online: 323
Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2018  |  Volume : 7  |  Issue : 1  |  Page : 37-40

Seasonality of leptospirosis and its association with rainfall and humidity in Ratnagiri, Maharashtra

1 Medical Officer, District Tuberculosis Centre, Vazirabad, Nanded, India
2 Medical Officer, ADHS Leprosy, Ratnagiri, India
3 District Health Officer, Ratnagiri, Maharashtra, India
4 District Tuberculosis Officer, Ratnagiri, Maharashtra, India

Date of Web Publication1-Mar-2018

Correspondence Address:
Dr. Shivshakti D Pawar
TB Centre, Vazirabad, Nanded - 431 601, Maharashtra
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/ijhas.IJHAS_35_16

Rights and Permissions

CONTEXT: Leptospirosis is the disease which has worldwide occurrence, but the incidence is more in tropical countries. Disease outbreaks depend on the climatic factors which allow better survival of bacteria.
AIMS: The aim of the study was (1) to study the seasonal pattern of leptospirosis cases and with rainfall and relative humidity and (2) to forecast the leptospirosis cases' occurrence based on the model.
SUBJECTS AND METHODS: Retrospective time series analysis was carried out of leptospirosis cases registered in Ratnagiri, Maharashtra, during January 2011 to December 2015.
STATISTICAL ANALYSIS: Patterns of occurrence of monthly cases and patterns of monthly rainfall and relative humidity were studied using cross-correlation function, autocorrelation function, partial autocorrelation function, and simple seasonal model (selected by expert modeler) were applied using SPSS version 20.
RESULTS: A significant seasonal pattern is noted in the leptospirosis cases' occurrence. Cross-correlation function shows a significant highest correlation 1 month lag between the occurrence of heavy rainfall and outbreak of cases. SPSS expert modeler used to forecast the cases and could predict the 80% variability.
CONCLUSIONS: Seasonal pattern of cases of leptospirosis was observed along with the correlation of rainfall. This forecasted model could be used by health administrators effectively to arrange the adequate resources on time to manage the outbreaks.

Keywords: Cross-correlation functions, humidity, rainfall, seasonality, simple seasonal model

How to cite this article:
Pawar SD, Kore M, Athalye A, Thombre P S. Seasonality of leptospirosis and its association with rainfall and humidity in Ratnagiri, Maharashtra. Int J Health Allied Sci 2018;7:37-40

How to cite this URL:
Pawar SD, Kore M, Athalye A, Thombre P S. Seasonality of leptospirosis and its association with rainfall and humidity in Ratnagiri, Maharashtra. Int J Health Allied Sci [serial online] 2018 [cited 2023 Feb 4];7:37-40. Available from: https://www.ijhas.in/text.asp?2018/7/1/37/226260

  Introduction Top

Leptospirosis is one of the most widespread diseases worldwide. World Health Organization reports that there occur >500,000 cases per year in the world.[1] Although its incidence ranges from 0.1 to 1 case per 100,000, in temperate countries, >10 cases per 100,000 may occur and during outbreaks in humid environment, it may reach >100 cases per 100,000 population.[2] The incidence is significantly higher in tropical countries than temperate regions, mainly because in humid and warm environment leptospira species have longer survival.[2]

Leptospirosis is caused by spirochete of genus Leptospira interrogans. These are aerobic, motile, and slow-growing bacilli having optimal growth at temperature 30°C and are able to survive in soil and water bodies for a longer period.[3] Infection in humans is either by direct contact with the urine of an infected animal or indirectly through the contaminated environment.[3] Rodents are the most efficient epidemiological reservoirs for pathogenic leptospires.[2] Agricultural workers are most infected during cultivation periods. Flooding carries the urine of the infected animals at distant places and determines the size of an epidemic.[3]

Ratnagiri, a district in the southwestern part, in the Konkan region of Maharashtra state in India, situated on Arabian sea coast at 16.98 N, 73.3 E with an average elevation of 11 m. Population of Ratnagiri according to Census 2011 is 76,229. The climate is tropical, and temperatures are highest from March to May, with rainy and humid season between mid-June and November, and the annual rainfall is around 3188 mm.[4]

Leptospirosis is an important example of reemerging infectious disease that affects humans. Its infection spreads through the contaminated water, so it is indirectly environment-mediated disease. Risk of outbreaks and its drivers depend on interactions between humans, animals, and environmental conditions. We aimed to study, therefore, the 5-year transmission dynamic patterns of the leptospirosis and how it is affected by environmental changes. Simultaneously, we projected the future occurrence of leptospirosis cases based on a model using past 5-year data.

  Subjects and Methods Top

The monthly number of cases of leptospirosis from January 2011 to December 2015 was availed with due permission from District Health Office, Ratnagiri. All cases were defined by at least one positive biological test (patients with microscopic agglutination test titer >1/400, or with positive polymerase chain reaction result, or with a positive blood culture). Monthly meteorological data of the same period were obtained from Tutiempo Network website.[5]

Statistical analysis

Statistical analysis was conducted with SPSS version 20 (IBM, SPSS Tnc., Chicago, Illinois, USA) using expert modeler for forecasting and time series analysis (TSA). TSA was used to identify the temporal patterns in series of cases of leptospirosis with the relationship between rainfall and relative humidity from January 2011 to December 2015. TSA is the method consisting of series of observations which are correlated over time by befitting best model. Stationarity of each parameter is tested using augmented Dicky–Fuller test. Autocorrelation and partial autocorrelation function series also applied for checking trend and seasonality. Using expert modeler in SPSS, we tested the correctness of model using Ljung-Box (modified Box-Pierce) statistics. We checked cross-correlations of a number of cases of leptospirosis with rainfall and relative humidity. Using total leptospirosis cases as a dependent variable and rainfall and average relative humidity as independent variables, we forecasted the number of leptospirosis cases up to December 2017 in a simple seasonal model.

  Results Top

[Table 1] summarizes that a number of leptospirosis cases were 123 in 2011 and decreased number of cases in next studied years till 2015; in 2015, only 36 cases occurred. When studied month-wise occurrence of cases in each year, number of cases were started increasing in each year from July which corresponds to rainy season and decreased to minimum in December in each year.
Table 1: Monthly leptospirosis incidence from January 2011

Click here to view

[Figure 1] shows the relation between a number of leptospirosis cases in each month with the rainfall and average relative humidity in each month. The peak of leptospirosis cases corresponds with the peak of rainfall and humidity in each year. It shows a cyclical seasonal pattern as a peak of cases follows a similar peak in rainfall and humidity. Seasonal pattern of cases shown by the autocorrelation function depicting a significant peak at lag of 12 which is confirmed by partial autocorrelation function (not shown here due to constraint of space). Peak in cases followed a significant peak in the rainfall in previous 1 or 2 months. [Figure 3] and [Figure 4] show the graph of cross-correlations between total patients in month with rainfall in the month and relative humidity. Significant positive cross-correlations were detected between monthly leptospirosis cases and monthly rainfall and relative humidity, which are lagged by 0 to 1 month and shows the strongest significant correlation (r = 0.78 between rainfall and number of cases and r = 0.70 between relative humidity and number of patients) [Figure 2].
Figure 1: Relation between number of leptospirosis cases and rainfall and relative humidity

Click here to view
Figure 2: Cross-correlation function between number of leptospirosis cases and rainfall

Click here to view
Figure 3: Cross-correlations between number of leptospirosis cases and relative humidity

Click here to view
Figure 4: Forecasting the number of leptospirosis cases with using simple seasonal model

Click here to view

Time series sequence chart showed seasonal periodic fluctuations. Hence, expert modeler of SPSS version 20 suggested a simple seasonal model as the best-fitted model for these time series data. The Ljung-Box (modifies Box-Pierce) test indicated that the model was correctly specified for the data, P > 0.05 [Table 2]. The expert modeler detected no outliers in the data. Here, stationary R-squared value was used for testing goodness of fit. This TSA provides an estimate of proportion of total variation in the time series that is explained by the model and mainly by the R-squared when there is a seasonal pattern. Maximum values of R-square indicate a better fit of the model. A value here of 0.80 meant that the model could explain 80% of the observed variations in the time series.
Table 2: Model statistics for leptospirosis

Click here to view

[Table 3] summarizes that, from the month of June to December, the seasonal adjusted factor (SAF %) of leptospirosis was >zero, that is, in these months, the leptospirosis cases were more above the typical months. Among these months, August with SAF = 445.6 had the highest SAF, that is, in this month, the leptospirosis cases were >445.6% higher compared with the typical months. SAF for rainfall and relative humidity was highest in month of July, 386% and 122%, respectively (not shown in figure). These findings corroborate with the findings of cross-correlations.
Table 3: Seasonal adjustment factor (%) for leptospirosis

Click here to view

[Figure 4] shows that observed and the predicted model values (simple seasonal model) are matched reasonably well, and there was a consistency in seasonality and trend. This same model was used to forecast the incidence of leptospirosis cases for future from January 2016 to December 2017. The forecasted incidence also showed a seasonal pattern of the disease and peak of cases occurred during August to September.

  Discussion Top

The monthly number of reported cases of leptospirosis from January 2011 to December 2015 showed a consistent seasonality pattern. The peak of cases was in August in each year approximately. There is a significant time lag of rainfall (heavy rainfall) and relative humidity in Ratnagiri recorded 1 month previously and a number of cases in recorded in the month considered, and cross-correlation is strongly positive. Our finding is similar to a study conducted by Lhomme et al. and similar study conducted by Desvars et al.[6],[7]

During the rainy season, there is a formation of water bodies which helps leptospires to survive for a longer time for approximately 1–2 months; increased moisture and humidity also help to grow pathogenic leptospires, which ultimately leads to increased number of cases due to prolonged contact of at-risk people with water bodies.[2]

Our model, based on the two meteorological parameters, estimates 80% of variation of the monthly number of leptospirosis cases. Nevertheless, the peak incidence observed in July–August in 2011 (>50 cases) was not predicted by our model (predicted cases = 30 cases), but model calculated a good and exact prediction of a number of cases observed in August 2013 and 2014.

Other parameters also affect the incidence of leptospirosis cases in the environment such as the concentration of oxygen and iron in water or soil and water pH.[2] Variation in the human activities also influences the probability of occurrence of more number of cases. We did not include tropical cyclone data as well as climatic depressions which may influence the number of cases.

However, this simple seasonal model could be most useful for predicting leptospirosis cases to be well prepared to timely and adequately provide health-care services in need.

  Conclusions Top

There is a seasonal peak of leptospirosis cases, depicting the rise of cases along with increased rainfall and relative humidity. Based on previous 5-year data, future possible peaks of leptospirosis calculated and can be effectively used in mitigating epidemics by intervening as a public health response.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

World Health Organization. Wkly Epidemiol Rec No. 74. Geneva: World Health Organization; 1999. p. 242-6.  Back to cited text no. 1
Park K. Park's Textbook of Preventive and Social Medicine. 19th ed. Jabalpur: M/s Banarasidas Bhanot; 2007. p. 243.  Back to cited text no. 2
Faine S, Adler B, Bolin C, Perolat P. Leptospira and Leptospirosis. 2nd ed. Melbourne, Australia: Medisci Press; 1999. p. 296.  Back to cited text no. 3
National Informatics Centre. Ratnagiri City and District Information. Available from: http://www.ratnagiri.nic.in/gazetter/gom/home.html. [Last accessed on 2016 Feb 29].  Back to cited text no. 4
Tutiempo Network, S.L. Climate RATNAGIRI Climate data from 1973-2016. Available from: http://www.en.tutiempo.net/climate/ws-431100.html. [Last accessed on 2016 Feb 20].  Back to cited text no. 5
Lhomme V, Grolier-Bois L, Jouannelle J, Elisabeth L. Lepto in Martinique from 1987 to 1992: Results of an epidemiological, Clinical and biological study. Med Mal Infect 1996;26:42-6.  Back to cited text no. 6
Desvars A, Jego S, Chiroleu F, Bourhy P, Cardinale E, Michault A, et al. Seasonality of human leptospirosis in Reunion Island (Indian Ocean) and its association with meteorological data. PLoS One 2011;6:e20377.  Back to cited text no. 7


  [Figure 1], [Figure 2], [Figure 3], [Figure 4]

  [Table 1], [Table 2], [Table 3]

This article has been cited by
1 Impact of rainfall on the transmission of leptospirosis in Si Sa Ket, Thailand
S Wichapeng,S Chadsuthi,C Modchang
Journal of Physics: Conference Series. 2021; 1719(1): 012024
[Pubmed] | [DOI]
2 Leptospira in river and soil in a highly endemic area of Ecuador
Erin Miller,Veronica Barragan,Jorge Chiriboga,Chad Weddell,Ligia Luna,Dulce J. Jiménez,John Aleman,Joseph R. Mihaljevic,Sonora Olivas,Jane Marks,Ricardo Izurieta,Nathan Nieto,Paul Keim,Gabriel Trueba,J. Gregory Caporaso,Talima Pearson
BMC Microbiology. 2021; 21(1)
[Pubmed] | [DOI]
3 Prediction mapping of human leptospirosis using ANN, GWR, SVM and GLM approaches
Ali Mohammadinia,Bahram Saeidian,Biswajeet Pradhan,Zeinab Ghaemi
BMC Infectious Diseases. 2019; 19(1)
[Pubmed] | [DOI]
4 A Cross-Sectional Assessment of Knowledge, Attitude, and Practice Toward Leptospirosis among Rural and Urban Population of a South Indian District
Sivakumar Rathinam,Rajesh Vedhanayaki,Kandasamy Balagiri
Ocular Immunology and Inflammation. 2019; : 1
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Subjects and Methods
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded281    
    Comments [Add]    
    Cited by others 4    

Recommend this journal