Oct (2010) 28th

A Field Study on the Use of Clay Ceramic Water Filters and Influences on the General Health of Eweje Village, Odeda, Ogun state, Nigeria, Africa
Anand PLAPPALLY1, 3,*, Karen MALATESTA2, Katie C. FRIEDMAN1, 4, Enoch DARE5, Taiwo OGUNYALE6, Ismaiel YAKUB1, 2, Alfred SOBOYEJO3 , Megan LEFTWICH2, Andrew USORO1,7, Ron RIVERA8 and Winston SOBOYEJO1,2
1

Princeton Institute of Science and Technology of Materials (PRISM), 70 Prospect Avenue, Princeton, NJ, 08544; 2 Department of Mechanical and Aerospace Engineering, Princeton University, Princeton NJ, 08544; 3 FABE Department, The Ohio State University, 590 Woody Hayes Drive, Columbus, OH 43210; 4 Department of Chemical Engineering, Princeton University, Princeton, NJ 0854; 5 University of Agriculture, Abeokuta, Nigeria, Africa; 6 Eweje Village Physician, Nigerian Ministry of Health, Federal Government of Nigeria, Africa; 7 Department of Chemistry, Princeton University, Princeton, NJ 08544; 8 Potters for Peace, Ceramic Water Filter Program, Managua, Nicaragua; Received: Accepted: Published:

Abstract Field study and surveys were conducted to evaluate the interdisciplinary parameters influencing health of people using ceramic filters for water purification. A total of 52 families were distributed filters at Eweje Village, Odeda local government area, Ogun State, Nigeria. Surveys contained questions related to hygiene, health, water source and treatment, socio-economic and educational status of people and their use of clay ceramic water filters. Several parameters were studied which include time of use of water filter, maintainability, education, societal economics, and social the status of the people using the filters. There was interdependence between these parameters. Health of the Eweje Village community was greatly influenced by the number of people using the filter, the length of filter usage, education, maintainability, access to medical facilities, and economic status in decreasing order of influence. A novel multi parameter multivariate regression approach clearly enumerates the hierarchy of the effects of the influencing variables on the health of Eweje community. Apart from population and time of filter utility, access to medical services influenced health in this rural community using ceramic water filters.

Key words: Rural, Health, Water, Filters, Education, Regression, Africa

Corresponding Author: A.Plappally, e-mail:[email protected], Phone:001 614506 4332, Fax:001 614 2929448,

INTRODUCTION Eight hundred eighty-four million people are still without potable water with only 5 years to go to meet the Millennium Development Goal of the WHO-UN (WHO 2010). 34% of the deprived people live in Sub-Saharan Africa. In the last 18 years there has been 10% increase in total population of those who have access to potable water (WHO 2010). As per predictions by the World Bank in 2003, by 2015, 5-10% of the population of Middle East and North Africa, Latin America and the Caribbean will still be without reliable potable water. Similarly, approximately 15% of South Asia and 25% of Sub Saharan Africa will not have access to potable water resources (Hillie et al. 2009). Several water filtration technologies have been started by educational initiatives and non-governmental organization in the near future to resolve potable water scarcity (Sobsey et al. 2008). Chlorination with safe storage, chemical coagulants such as WaterMaker (Control Chemical , Alexandria, VA), PuR (Proctor and Gamble, Cincinnati, OH), sodium hypochlorite (SFH/Nigeria) , nut/seed organic medicinal materials, sunlight exposure techniques such as SODIS, SOLAIR, UV

Oct (2010) 28th

A.K Plappally et al.

radiation techniques, filtration techniques such as nano-membrane filtration, reverse osmosis technique, Pureit (HLL Ltd., Unilever Inc., India), organic additive based ceramic filters, Kanchan MIT arsenic filter and bio-sand filters are some of the most studied and surveyed techniques used around the globe for water purification (CDC 2008; Clasen et al. 2006; Clasen et al. 2007; Hillie et al. 2009; Duke et al. 2006; Ngai et al. 2006; Sobsey et al. 2007; Sobsey et al. 2008). Education initiatives have been taken in Nigeria near Sub-Saharan Africa under the research auspices of Potters for Peace, Princeton University and Ohio State University. UNICEF joined with a local non-governmental organization in Myanmar formed the Community Development Association to initiate water purification technology to the household level (Naing 2007). More than 3,000 ceramic water filters have been distributed in the Phyu village and schools in Myanmar. More than 80% of the households near the delta and coastal areas use these filters regularly and customer satisfaction is about 90% (Naing 2007). With this ceramic filter any particle or organisms that are larger than 1 micron are trapped in the filter. Millions of these porous clay ceramic filters are in use in several countries in African, Asian, and South American continents (Plappally et al. 2009). Studies on performance of clay ceramic filters in Bolivia conducted under the nongovernmental organization Food for the Hungry International showed a decrease in the cases of diarrhea by around 45% (Clansen et al. 2006). In the studies conducted by Sobsey et al in 2008, ceramic filters and biosand filters were found to best fit the sustainability criteria in the field with consumers (Sobsey et al. 2008). The main objectives of the socioeconomic, educational and behavioral study were, a) To assess the health impact due to use of these ceramic water filters. b) To evaluate which of the economic and social parameters in Eweje influenced the health of the people using ceramic filters for more than 6 months. c) Study the interaction between the social and economic variables arising in this study which were predictors for health status of people using ceramic filters in Eweje. METHODS Manufacturing Process Ceramic water filters were manufactured with locally available clay and sawdust. These filters were low cost but considered as efficient and sustainable technology to treat drinking water in developing countries (Sobsey et al, 2008). The porous clay ceramic filters were manufactured from moistened suspensions containing clay sawdust (C-S) in 4555, 65-35, 50-50, 55-45 ratio by volume. Due to the plasticity of the moistened clay-sawdust blend, it could mold under stress to any shape as required. The filters were cast in the shape of a frustum (Donachy 2004). Sintering these filter molds to around 900oC introduces numerous pores into the mold serving its filtration capabilities (Lee et al. 2001; Franz, 2005; Dies, 2003; Oyanedel-Craver et al. 2008). These frustum shaped filters had length of axis dimensions of 26cm, lower base diameters of 20cm, and upper base diameters of 23cm respectively. The filter wall and base had a thickness of 0.5cm and 1cm respectively (Plappally et al. 2009). The flow characteristics of the 50-50 filter were far better than that from the other C-S ratios mentioned above (Lee et al. 2001; Plappally et al. 2009). Microbial Removal Efficiency test for the filters Before distribution of the filters for the field testing and survey, microbial filtration experiments were performed on series of filters of 45-55, 50-50 55-45, and 65-35 volume ratios. To determine the filtration efficiency, 1020 ml cultures of the nonpathogenic Escherichia coli K-12 strain W3110 were grown in Millers LB agar at 37o C for 1824 hrs with vigorous aeration either by shaking at approximately 200 220 rpm or by stirring (VWR digital stirrer/hotplate). This testing was carried out at Mechanical and Aerospace department at Princeton University, NJ, USA. Four milliliters of this stationary phase culture were mixed into 4 L of sterile purified water, producing a pre-filtrate suspension containing 106 to 107 cells/ ml. The entire 4 L of pre-filtrate was poured rapidly into a water-saturated filter, and 34 L of the filtrate was collected in a 5 gallon plastic pail lined with sterile plastic. The numbers of viable cells in the pre-filtrate and filtrate suspensions were determined by appropriate dilution into sterilized purified water and plating onto Millers LB agar.

Oct (2010) 28th

o

A.K Plappally et al.

The colonies were counted after overnight incubation at 37 C and used to calculate viable cells/ml. If the viable count of the filtrate was low, cells present in larger filtrate samples (10 - 100 ml) were collected using sterile filtration assemblies (Millipore). The filter was then removed from the filtration assembly, placed directly onto Millers LB agar and incubated overnight at 37oC. To decontaminate filters between experiments, filters were either rinsed thoroughly with purified water and dried in full sunlight for 58 hrs or rinsed with 95% ethanol followed by drying at room temperature. The efficiency of E. coli filtration was calculated as shown below. First, the final E. coli concentration was calculated from:

E .coli = 1

Nf N pf

where Nf is the number of viable cells per milli-Liter of filtrate and Npf is the number of viable cells per milli-liter in the prefiltrate. The log reduction value is defined as (Lantagne 2001), . Setting For testing these filters in real life conditions in a developing country these filters were manufactured and tested in Eweje, Odeda Local Government Area, Ward 1, Abeokuta, Ogun State, Nigeria, West Africa (OSG OdedaLGA 2010).

Figure 1: The map of Ogun State in Nigeria showing Odeda which is located in the north central region of Ogun State ( Courtesy: Ogun Land Information Systems, Ogun State Bureau of Lands and Survey).

Fig.1 illustrates that Odeda local government shares boundaries with South Abeokuta in the south, North Abeokuta in the west, Obafemi Owode local governments in the east, and Oyo state in the north respectively. The climate is tropical with heavy rainfall from April to July and in September and October. Average temperature is about 32oC but humidity is high at 95% (OSG Odeda LGA, 2010). It is very important to know the climate before surveying health since climatic change is a very pertinent factor influencing the health of the population. Each and every family in Eweje was asked for their consent of participation in this survey. The main sources of water are ponds and rivulets running near Eweje village. Manufactured ceramic filters were distributed to 53 families for free in February 2009. These families were surveyed on the effectiveness of the ceramic filters in purifying water. The survey was initiated in February 2009 when the filters were

Oct (2010) 28th

A.K Plappally et al.

distributed. Survey was carried out every 3 months while the filters were utilized by the people. The survey was quantitative and qualitative in nature with an objective to reveal the effects of the ceramic filter based water treatment on the social and economic status as well as the health of the people surveyed. The survey questions are attached in the Appendix I. It was important to note that several people (23 people) moved out of the Eweje locality to cities due to job and employment opportunities in between April and June 2009. All of them took the filter along with them. There was only one person who resisted participating in this survey and ceramic filter trail study conducted in Eweje village locality who did not believe in any new technology. So the results of 30 families have been documented in this study. Personnel Collaborators from The Ohio State University, Princeton University, University of Agriculture, Abeokuta, Ogun State, Nigeria and graduate students were involved in data collection, filter use dissemination, and health survey. The Ministry of Health, Nigeria played an important role in the health survey by allocating a resident physician, Dr.Taiwo Ogunyale, as a part of this project. Important inputs were provided by the resident physician while preparing the survey sheet for the people residing in Eweje Village. The visits of personnel were devoted primarily to educate the people, interview the population of filter users based on questions on the survey sheet. Each of the visits was carried out by a team comprising of two students acting as the investigator and coordinator. For each visit the team filled out duplicate copies of the questionnaire in their native Yuroba tongue as well as in English. The resident physician reviewed the survey on a monthly basis and included his personal investigation of community health to the sections D, E, F, G, H and I of the survey. The Questionnaires It is very important to have a reasonably significant population for such a survey. A reasonable sample set of 53 families were surveyed on the effectiveness of the potable water filter. The survey was discussed and explained individually to the user of the filter. The consent of the user was obtained. The questionnaire contained questions pertaining to economic status, level of education, family demographics, individual hygiene and health concerns before and after filter use, availability of medicine or medical consultation facility, potable water source, filter use and cleaning frequency, and number of filter units in use per family (See Appendix I). This also became the major parameters being studied in this technical chapter to influence health. Health issues may be due to multiple factors so it is assumed to have non linear characteristics. The survey looked at multiple parameters. Health of the people was predicted with the help of 8 parameters which were expected to closely influence the filter usage. The number of filters for each family X1, filter cleaning frequency X2, children below 5 years of age X3 , members in a family X4, and time of filter usage X8, are five of the quantitative predictor variables used in this study. Qualitative predictor variables were wealth statistics X5, educational qualification X6 and availability of medical consultation X7. Qualitative variables were expressed in percentages assuming and evaluating their knowledge level, material assets and their approach in dealing with health issues respectively (Ogunyale 2009; Soboyejo 2006). The criteria for judging these qualitative variables were developed and agreed by the collaborators from Princeton University, Ohio State University, University of Agriculture, Abeokuta, Nigeria and the Ministry of Health, Nigerian Federal Government. RESULTS Microbial Removal Efficiency results for the filters tested at Princeton University The E. coli removals by filters are presented in Table 1 below. Note that the ranges are from duplicate experiments that were performed on the same specimens. All of the filters exhibited very high E. coli removal rates. In comparison, the filtration efficiency of a typical 50-50 colloidal silver coated filter produced by Potters for Peace was 99.99%; the 50-50 filter and 65-35 filters without colloidal silver coating manufactured at Princeton University had very close filtration efficiencies of 99.99 0.00 and 99.97 0.03% respectively (Lantagne 2001, Bielefeldt et al. 2009). This is enumerated in Table 1 below.

Oct (2010) 28th

Filter 45:55 50:50 55:45 65:35 Test 1 99.97 99.99 99.52 99.99 Test 2 99.85 99.93 99.84 99.99 Average Efficiency + Range (%) 99.91 + 0.06 99.97 + 0.03 99.68 + 0.16 99.99 + 0.00 Log Reduction Value-Test 1 3.49 8.16 2.314 5

A.K Plappally et al.

Log Reduction Value-Test 2 2.84 3.15 2.79 6.9

Table 1: Bacterial filtration results obtained for filters with different clay to saw dust ratios.

It was earlier mentioned that 50-50 filter was having a comparatively better flow characteristic as compared to other filters. And from Table 1 it is found that in test 1, 50-50 filters had competitive microbial filtration efficiency and was better than EPA standard LRV value of 6 for bacterial removal (Clansen et al. 2009). Oyanedel-Craver et al, 2008 and Bielefeldt et al, 2009 reported bacterial removal greater than >2.5 log and > 3log of E. coli pulse-spiked onto 50-50 ceramic ware respectively which is very much comparable with the results in Table 1. Similarly Lantagne in 2001 reported 98.2, 97.0, and 82% removal of total coliform, fecal coliform and fecal streptococcus, respectively, by using filters from Potters for Peace, Nicaragua. It should be noted that there is no silver coating on the filters. Hence 50-50 filters are estimated to provide better service in the field also. Survey results Population is a major parameter without which the survey of the effectiveness test of the filters cannot be carried out. It is found that 30 people responded to the survey regularly for more than 6 months. The other 23 people out of the 53 person sample population under survey were able to provide their inputs only once at the initiation of the survey. Hence their comments have been neglected in the studies. The health status of the 30 families were recorded in percentage and plotted in Fig. 2. The data from survey section D-I was considered and personal individual physical examination was performed by Dr. Taiwo Ogunyale for assessing health for each family and reporting in terms of percentage health response (HR%).

Oct (2010) 28th

A.K Plappally et al.

Figure 2: The variation of health (HR %) as a function of the size of family (P) 6 months after initiation of water filter use in Eweje Village, near Lagos, Nigeria.

From the Fig. 2, it is clear that that with increase in the number of people in a family P, there was better health. It was estimated by World Health Organization in 2002, that 9/10 deaths were children and 54.2 million disability adjusted life years were lost due to unhygienic condition and scarcity of potable water (Parikh et al. 1999; WHO 2002).

Figure 3: The variation of health (HR in %) as a function of number of children below 5 years of age (CP = X3).

In Fig. 3, the X axis represents the distribution of the number of children below 5 years of age. The survey data (CP = X3) is appended in the Appendix II. It was found that children below 5 year of age (CP) helped in making the health (HR) a very random variable. It is important to note that presence of a child within the family made a large difference in the health status. It should be noted that children and mothers were more susceptible to health problems compared to others in the family residing in rural areas in developing nations (Montgomery et al. 2007). It is also true that women and children spent much time fetching potable water preventing them from attending schools (Jalan et al. 2003 and Hillie et al. 2009).

Oct (2010) 28th

A.K Plappally et al.

There is a significant influence of wealth on a family decision making either to consult a doctor or to adopt some type of purified drinking water for them, such as bottled mineral water. It is observed that even when family income was high there was no predictable improvement in health. This is very clear from Fig 4, which supports this statement.

Figure 4: The variation of health response (HR) as a function of family income (I = X5) expressed in percentage

The families with higher economic status were found to either have sound or worse health characteristics. This randomness is due to other influencing parameters such as presence of children below 5 years of age or low maintainability of filter, M. The other factor which may mainly contribute to affect health is education. It has been noted that education of females in the family proved to be a positive influence to heath of the family (Jalan et al. 2003). It was found from the survey that with increase in educational qualifications the awareness of the people to clean the filter was higher. This awareness as plotted in Fig. 5, leads to better condition of health arising from eradication of water borne diseases owing to use of the new filters.

Figure 5: The variation of health (HR) as a function of Family Education (E).

Oct (2010) 28th

A.K Plappally et al.

It is found that health of families with people having 20-60% level of education contributed to low health percentage. People with more than 60% level of education (above tenth grade) did have better health. It is found that loosing school hours was deleterious to health (UNICEF 2007). Education is the only edge to restore a sense of normalcy when illiteracy influences disastrous health. For that reason the WASH programs in schools in countries like Thailand, Mongolia and Niger which imparted awareness to hand washing with soap and other sanitation methodology to its students (UNICEF 2008; UNICEF 2009). Lack of potable water and sanitation facilities in schools prevented female attendance and impacted their learning environment (Hillie et al. 2009; UNICEF 2009). Awareness is considered a way to make people realize the importance of better hygiene practices like hand washing and should be a part of the curriculum in schools throughout developing countries (MNN 2010). We can find that an awareness of the importance of pure drinking water and the exposure to mass media has significantly affected people in developing countries. As a result, the number of people who died of cholera has decreased from 8,500 in 2007 to 42 in 2009 in Africa (MNN 2010). Hence the awareness to hygiene will really improve usage and maintainability, M, of the filters. This is confirmed by Fig. 6 illustrating that an improvement in educational status can improve the awareness of better maintenance.

Figure 6: A trend line plot of Maintainability M and Education E.

From Fig. 7, it is seen that with increase in maintainability, M (filter cleaning frequency X2), health of the people was seen to improve proportionally.

Oct (2010) 28th

A.K Plappally et al.

Figure 7: The variation of health response (HR) as a function of water filter maintenance M.

The survey data tabulated in Fig. 2 to Fig 5 and Fig 7 has been plotted using the SPSS statistical software version 16 (SPSS 16.0 for Windows IBM 2007). Theoretical Development A non linear behavior of health response (HR =Y) is seen from the above results from Fig 2 to Fig 6. A lognormal stochastic multi-parameter model has been proposed in this research to model the health response, Y, of the people at Eweje Village, Lagos State, Nigeria. The parameters X1, X2, .. Xk are manifestations of societal, population, technological, and economic infrastructure influencing the health changes in Eweje due to filter use. So, health response, Y, can be expressed mathematically as follows, for i = 1, 2, 3..k (1) The above expression of Xibi is known as the transfer function. This function mathematically expresses the step-bystep effects of the variables mentioned above with time on the health of the community in Eweje (Soboyejo et al. 2000). This nonlinear behavior is simulated with a stochastic step function from step 0 to step 1 for one of the predictor variables as shown below in Eq. 2. for i = 1, 2, 3 (2) where Y0= a initial value of the stochastic process model, and which is a model constant in this problem. The Eq. 2 is extended by increasing the number of parameters and is written in a general form,

Yn b = X nn Yn1
n= k b b n Y = Yn = Yn 1 X n n = aX 1b1 X 2 2 .......X n = a X ibi

(3)

i =1 For n = 1, 2,..k. (4) Since the predictor random variables have different dimensions, Eq. 4 can be mathematically reformulated as (5)

(6) where Xio is any reference constant with the same units as Xi.

Oct (2010) 28th

Eq. 4, can be expressed as
k

A.K Plappally et al.

yi = ln Yi = ln a + bi ln X i
i =1 (7) From Table 2, it is found that X4, number of members in the family, affects positively the health, Y, due to high frequency in cleaning the filters in highly populated families. This is very clearly enumerated in Fig 8.

0.6

0.5

0.4

ln(X4/Y)

0.3

0.2

0.1

0.0 0.0 0.5 1.0 1.5 2.0 2.5

LnX4

Figure 8: Scatter Plot of the new transformed variable X4/Y as a function of X4.

The new response variable is derived using the principle of minimal realization using a quotient variable (Sussman 1977). The new quotient variable G is fraction of the most influencing predictor variable to the actual response variable (Tabuada and Pappas 2005). Hence a predictor variable transformation is performed (Sussman 1977). Then the Eq.7 can be written as (8) Here the predictor variable transformation is applied such that it does not affect the actual behavior of the response variable Y. To test the validity of the assumed transformation X4/Y over multiplicative transformation of Y as well as raw data Y, a Kolmogorov Smirnov goodness of fit test was performed at a 99% level of confidence and is shown in Table 2 below.

Kolmogorov Smirnov Test at =0.01 Y Ln Y Ln G Dn p value

Table 2: Dn and

Critical Value n 30 30 0.29 0.01

0.173 0.03

0.127 0.15

0.2 0.009

for Kolmogorov Smirnov Test for the different models for the Health Response Y (Ang et al. 1975).

Here if the Dn value is less than the critical at =0.01, the proposed distribution for the variable is accepted (Ang et al. 1975). All the variable predictor parameters for health response (HR=Y) were highly correlated to each other and has been enumerated in Table 3 below. Filter cleaning frequency X2 is highly correlated with number of members per family X4. It is clear from the correlation Table 3, that educational awareness X5 is found to increase with increase in wealth X6 and vice versa. This is because better education provided people with high paying jobs, or people with enough money were able to avail good education.

10

Oct (2010) 28th

1 2 3 4 5 6 7 8 0.23 0.04 0.149 -0.032 -0.015 0.01 0.065 0.406 0.508 0.421 0.16 0.03 0.41 0.734 0.339 0.123 0.263 0.208 0.482 0.568 0.02 0.313 0.55 0.064 0.288
0.441 0.181 0.035

A.K Plappally et al.

2 3 4 5 6 7

Table 3: Correlation coefficients for each pair of the random parameters influencing health namely, the number of filters for each family X1, filter cleaning frequency X2, children below 5 years of age X3, members in a family X4, wealth statistics X5, educational qualification X6, and availability of medical consultation X7 and time of filter usage X8.

The correlated predictor matrix column elements in Table 3, with correlation coefficient the independent variables Vi and are scaled as (Haldhar et al. 2000; Krishnamoorthy 2006).

are linearly transformed to

(9)

The left hand side of Eq. 9, are standard normal variates of the predictor variables lnXi, j for ij and i=1, 2, 3 having zero mean and unit standard deviation. Here and are the parameters of lognormal distribution. , which constitutes all the principal components in the transformation matrix . The Vi, j for i=1, 2, 3 in Eq. 9, are the independent normal predictor variables with i, j for i=1, 2, 3 the corresponding variances. An explanation of the procedure in Eq 9 is elaborated in the Appendix III. Influences of the predictor variables on health status of people at Eweje The multiplicative nature of Y is preserved in transformation G and the multivariate approach developed above is applied to fulfill the assumptions of independent variables for regression (Plappally 2010). The multivariate approach helps in identifying the actual independent influences of the predictor variables on the newly transformed response variable G which now consolidates the health parameter Y. So the regression model coefficients are shown in Table 4 below.

Model
G 0.29 0.0842 -0.018 -0.022 -0.027 -0.024 -0.014 -0.030 -0.103
.

S
99 0.0184
Table 4 Multivariate model constants for Eq. 8

Table 4 below has been derived with the help of Minitab 15 statistical software (Minitab 15 2007). From Table 4, it is clear that model constants which are corresponding coefficients of the predictor variables in the Eq. 13 vary as,

This would mean that apart from population X4 which is a major influencing parameter, the other highly influencing parameters can be written as below in their increasing order of influence. X1 < X6 < X2 < X3 < X5 < X7 < X8 It is found that with time for which filter is used X8; there has been a steady decline in health issues in the Eweje locality. This confirms a long term effect of better potable water from the filters on health (Brown et al. 2007). This also predicts the

11

Oct (2010) 28th

A.K Plappally et al.

usage of the filters with satisfaction. Secondly, people who had close access to medical facilities X7 prevented themselves from diseases. This means that people did see their local medical officer for his/her advice on health issues. 83 out of total 124 people who participated in the survey sought medical services. It is very important to note that even though Odeda people are less educated academically they have good knowledge of traditional medicine (Olatokun et al. 2009). Out of 83 people, who used medical services 39 were either traditional healers or people with knowledge of medical herbs or traditional medicine. Within these 39 people there were 26 who traded in medicinal herbs also. A total of 21 women had knowledge of traditional medicine. This confirmed that mostly women have these knowledge bases who happen to be the primary water collectors (Olatokun et al. 2009; McCarton 2009). Financial status X5 as well as money did have a prominent effect in determining the health trends in Eweje but was less influential than availability of medical facilities and duration for which the filters were used. It is very important to find X5 as the third major influence confirming the finding by WHO that 1.8 million water related diseases were basically from low income countries. This actually provides an insight on basic psychology of the people swaying away from buying expensive new technology to avail potable water. Education and awareness played a pertinent role but not to expectation (Jalan et al. 2003). This would support a requirement for vigorous awareness programs educating people about sanitation, water and hygiene and their correlations (UNICEF 2010). Conclusion Multivariate stochastic regression formulation was used for identifying the individual socio-economic parameters affecting health of the society at Eweje Village, Ogun State, Nigeria. Summarizing the major outcomes of the health response effects due to clay ceramic water filter use as explained below. 1. Apart from population, the duration of filter usage has been a major parameter for influencing general health at Eweje Village. This confirms a considerable reduction in water borne diseases at Eweje after the introduction of ceramic water filters. 2. Accessibility of medical facilities or knowledge of traditional medicine within the people at Eweje had a great influence on the health outcome due to filter usage. The knowledge of traditional medicine should be accounted under the parameter education. 3. Financial status played a major role in influencing health at Eweje. 4. Random parameters influenced health in an order as shown below, Number of filters for each family, X1 < educational qualification, X6 < filter cleaning frequency, X2 < children below 5 years of age, X3 < wealth statistics, X5 < availability of medical access, X7 < time of filter usage X8, < members in a family X4. It is suggested and advised that Nigeria as well as other African nations should improve their education policy and cater education to each and every citizen irrespective of race, caste, tribe, religion and financial status of people. Another major suggestion is that traditional medicinal knowledge of people should not be ignored and provisions should be provided to improve transfer of traditional medicine knowledge to newer generations. This would also help in development and breakthrough in the fields of bio materials, agriculture and medicine. Acknowledgements This work is a part of the study conducted under the research partnership of Princeton University and the Ohio State University. This research partnership is funded through the Division of Materials Research, National Science Foundation (DMR 0231418), USA and The Grand Challenges Program at Princeton University. The authors also thank the Food, Agricultural and Biological Engineering Department, the Ohio State University, Columbus, OH for their support throughout the work. The authors also thank Eweje village community for participating in this suvey and providing excellent contribution for generation of this knowledge. We also thank Late Ron Rivera and also pay our tributes without whose guidance it would be impossible to cater cheap potable clay ceramic filtration device in Nigeria as well as throughout the developing world cutting across cultures and international borders. Authors also thank the Ministry of Health, Nigeria for their collaboration. REFERENCES
Ang, AH-S. and Tang WH. 1975. Probability Concepts in Engineering Planning and Design. Vol. I-Basic Principles. John Wiley and Sons, New York. Bielefeldt, AR, Kowalski K. and Summers RS. 2009. Bacterial treatment effectiveness of point-of-use ceramic water filters, Water Research 43(14):3559-3565

12

Oct (2010) 28th

A.K Plappally et al.

Brown J, Sobsey M, Proum S. 2007. WSP Field Notes, Improving Household Drinking Water Quality: Use of Ceramic Water Filters in Cambodia, WSP, UNICEF, August, 2009, pp.1- 44. CDC, 2008. Household Water Treatment Options in Developing Countries: Household Chlorination. (Retrieved from http://www.cdc.gov on Jan 13th 2010). Clasen T, Brown J and Collin SM. 2006. Preventing diarrhea with household ceramic filter: Assessment of a pilot project in Bolivia. International Journal of Environmental Health Research, June 2006, 16(3): 231-239. Clasen T, Nadakatti S and Menon S. 2006. Microbiological performance of a water treatment unit designed for household use in developing countries, Tropical Medicine and International Health, 11(9): 13991405. Clasen T, Schmidt W, Rabie R, Roberts I & Cairncross S. 2007. Interventions to improve water quality for preventing diarrhoea: systematic review and metaanalysis. BMJ, 335: 7597. Clasen T, Naranjo J, Frauchiger D, and Gerba C. 2009. Laboratory Assessment of a Gravity-Fed Ultra filtration Water Treatment Device Designed for Household Use in Low-Income Settings. American Journal of Tropical Medical Hygiene. 80(5): 819-823. Dies WR. 2003. Development of a ceramic water filter for Nepal. Master Thesis. MIT Department of Civil and Environmental Engineering. Cambridge, Ma: June 2003. Donachy B. 2004. Manual/Guide for Health Trainers and Others Involved in the Monitoring of the Colloidal Silver Ceramic Water Filter. Potters for Peace. Duke WF, Nordin RN, Baker D and Mazumdar A. 2006. The Use and Performance of Biosand Filters in the Artibonite Valley of Haiti: A Field Study of 107 Hoseholds. Rural and Remote Health. 6:570. Franz A. 2005. A performance study on Ceramic Candle filters in Kenya including tests for Coliphage removal, Masters Thesis. MIT Department of Civil and Environmental Engineering. Cambridge, Ma: June 2005. EF. 2009. Fundamentals, Rwanda ceramic water filter factory project, News from the Environ Foundation, Fall, pp.1. Haldhar A, Mahadevan S. 2000. Probability, Reliability and Statistical Methods in Engineering Design. NY, John Wiley and Sons. Hillie T, Munasinge M, Hlope M, Deraniyagala Y. 2009. Nanotechnology, Water and Development, for the commissioned as Part of the Meridian Institutes Global Dialogue on Nanotechnology and the Poor: Opportunities and Risks, pp. 1-44. Jalan J and Ravallion M. 2003. Does piped water reduce diarrhea for children in rural India? Journal of Econometrics, 112(1): 153-173 Jalan J, Somanathan E and Chaudhuri S. 2003. Awareness and the demand for environmental quality: Drinking water in urban India, Indian Statistical Institute, Planning Unit, New Delhi Discussion Papers 03-05, Indian Statistical Institute, New Delhi, India. Kiureghian AD and Liu P-L. 1986. Structural Reliability Under Incomplete Probability Information, Journal of Engineering Mechanics, 112(1): 85-104. Krishnamoorthy K. 2006. Handbook of Statistical Distributions with Applications, Boca Raton Chapman and Hall. Lantagne DS , Quick R and Mintz ED. 2006. Household Water Treatment And Safe: Storage Options in Developing Countries: A review of current implementation practices, From: Water Stories: Expanding Opportunities in Small-Scale Water and Sanitation Projects, Woodrow Wilson International Center for Scholars. Lantagne D, Klarman M, Mayer A., Preston A, Napotnik J, Jellison K. 2010. Effect of Production Variables on Microbiological Removal in Locallyproduced Ceramic Filters for Household Water Treatment, International Journal of Environmental Health Research. 1369-1619. Lee C. 2001. Investigation into the properties of Filtron, (http://www.edc-cu.org/pdf/scotland%20study.pdf) Project at University of Strathclyde, UK. McCarton L. 2009. Household Water Management in Sierra Leone, Dublin Institute of Technology, retrieved from http://www.rcsi.ie/hwts09/Liam Mccarton HOUSEHOLD WATER MANAGEMENT IN SIERRA LEONE.ppt, December 2009. Minitab 15. 2007. Minitab statistical software, release v.15 [Computer software]. State College, PA. MNN 2010. Maxims New Network: Guinea: Cholera, Water purification and child health (UNICEF) , March 19th 2010 (Retrieved on 28th March 2010). Montgomery MA. and Elimelech M. 2007. Water and Sanitation in developing countries: including health in the equation, Environmental science and Technology, 41(1):17-24. Naing W. 2007. Ceramic water filters improve water quality for rural communities in Myanmar, retrieved from <http://www.unicef.org/wash/myanmar_40738.html> Yangon, Myanmar, 28 August 2007. Ngai T, Murcott S, Shrestha RR, Dangol B, Maharjan M. 2006. Development and Dissemination of KanchanTM Arsenic Filter in Rural Nepal. Water, Science & Technology. 6(3). Ogunyale T. 2009. Personnel communication with the collaborators from the Ohio state University, Princeton University and University of Agriculture, Abeokuta. Olatokun WM and Ayanbode OF. 2009. Use of indigenous knowledge by women in a Nigerian rural community, Indian Journal of Traditional Knowledge, 8(2):287-295. OSG, OdedaLGA 2010. Information retrieved from website of Odeda Local Government, Nigeria, Area (http://www.ogunstate.gov.ng/eGovernment/index.php?option=com_content&view=article&id=122&Itemid=114 as retrieved on Dec, 31 2009) Oyanedel-Craver AV, amd Smith JA. 2008. Sustainable Colloidal Silver impregnated Ceramic Filter for Point of Use Water Treatment, Environmental Science and Technology, 42(3): 927-933 Parikh J, Smith KR, Laxmi V.1999. Indoor air pollution: A reflection on gender bias. Economic and Political Weekly.34 (9): 53944. Plappally AK, Yakub I, Brown LC, Soboyejo WO, Soboyejo ABO. 2009. Theoretical and Experimental Investigation of Water Flow through Porous Ceramic Clay Composite Water Filter. Fluid Dynamics and Material Processing. 5 (4):373-398. Plappally AK. 2010. Theoretical and Empirical Modeling of Flow, Strength, Leaching and Micro-Structural Characteristics of V Shaped Porous Ceramic Water Filters. PhD Dissertation, The Ohio State University, Columbus, OH. Soboyejo ABO. 2006. Probabilistic Methods in Engineering and Bio systems engineering, retrieved online from <hcgl.eng.ohio-state.edu/~fabe735/> accessed and retrieved in July 2009. Soboyejo ABO and Nestor KE. 2000. A New Statistical Biomechanics Approach to Modeling of Bone Strength in Broiler Chickens and Turkeys. Part I Theoretical Development. Part II Validation of Theoretical Models. Biological Engineering Technical Paper No.1: In Transaction of American Society of Agricultural Engineers. 43 (6): 1997-2006. Sobsey MD, Stauber CE, Casanova LM, Brown JM and Elliott MA. 2008. Point of Use Household Drinking Water Filtration: A Practical, Effective solution for Providing Sustained Access to Safe Drinking Water in the Developing World, Environmental Science and Technology, 42: 4261-4267. SPSS 16.0 for Windows, IBM, 2007. Chicago: SPSS Inc. Sussman H. 1997. Existence and Uniqueness of Minimal Realizations of Nonlinear Systems. Math System Theory, 10:263-284. Tabuada P and Pappas GJ. 2005. Quotients of fully non linear control systems, SIAM Journal on Computing, 43(5):1844-1866.

13

Oct (2010) 28th

A.K Plappally et al.

UNICEF 2007. UNICEF: World Water Day, commentary, UNICEF Television, retrieved on May 07, 2007. UNICEF, 2008, Water and sanitation challenges in Niger. July 17th 2008, UNICEF Television program. UNICEF 2009. WASH Awareness program UNICEF 2009, Afghanistan Government and UNICEF- A Plea for Water, Sanitation and Hygiene. December 19, 2009. UNICEF 2009. Children benefit from improved hygiene following the tsunami in Thailand. Retrieved from http://www.unicef.org/infobycountry/Thailand_52273.html. WHO 2008. The top ten causes of death, Retrieved from www. Who.int January 2010. WHO 2010. Progress on Sanitation and Drinking-water: 2010 Update. WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland, pp. 1-60.

APPENDIX I Survey Questionnaire Interviewer __________________ Family Code _______________________ Date __________________ Consent: We are researchers from University of Agriculture, Abeokuta, Nigeria and Princeton University New Jersey, USA who are working in collaboration with The Ohio State University, Ohio, USA. We would like to ask you about your water consumption sources, your concerns of water quality and effect of primary water source and purification source on the health and wealth of your family. We have consulted with Physician Dr. Taiwo Ogunyale from the Federal Government of Nigeria, serving in your locality, regarding the questions pertaining to your general health. However, we are not medically trained personnel but Dr Taiwo Ogunyale is accompanying us for this survey. Participation in this survey is voluntary. However, we hope that you will participate in this survey since your views are important. All your responses are confidential. We will not share your name with anyone and your answers will not be used for any other purpose other than research. At the conclusion of the interview we will provide you with a free ceramic water filter which is good for one year. You are requested to use this filter for your potable water needs. We would like to perform followup interview throughout this time to chart the effects of the potable water on your health. At this time do you have any question on this research work and survey? Finally do you provide a verbal consent of your involvement in this study? Thank you UAA, Nigeria, PU, USA, The OSU, USA Research team and Dr Taiwo Ogunyale, Government of Nigeria

Section A Dwelling Identification (Preferred interviewee: Head male or head female) Interviewee _______________________ Address: Street/ Road Block/Lot House/ Floor Group Type: Control _____________ Experiment ___________________ Section B

14

Oct (2010) 28th

Identifying Household Members No. Name (Last, First) Write down the names of individuals that eat. Sleep or consume water together within this Household Member role Partner Other of Focal mother

A.K Plappally et al.

Focal Member

Focal Child

Primary Water Collector

Sex Male Female

1 2

Continued No. Age 0.001 for less than 1 month old 0- for dont know Date of Birth Year/ Month Age 0-5yr First name Focal Children Only

6-11yr

+12 yr

1 2 3 Continued No.

Relationship Reference to Head of household Head Partner/Spouse Son/Daughter Step Son/daughter Father/Mother Father/Mother in law Brother/ Sister Brother/Sister in law Grandparent Uncle/ aunt Nephew / Niece Cousin Other Relative Domestic employee

Education Highest % 0 % :No education 10% : Grade 1 15%: Grade 2 20%: Grade 3 25%: Grade 4 30%: Grade 5 35%: Grade 6 40% Grade 7 45%: Grade 8 50%: Grade 9 55%: Grade10 60%: Grade 11 65%: Grade 12 70%: Diploma 80%: Bachelors 90%: Honors 95%: Masters 100%: Masters &Higher

Currently in school

Name Focal Children in School

1 2 3

15

Oct (2010) 28th

Section C: Wealth Index Preferred interviewee: Head of the family Asset Items Electricity Boat/motor Stove(coal/wood/paraffin) Television Sewing Machine Animal drawn Cart Watch Radio Iron Civic Essentials Toilet Type None/Bush/Field Flush/ Pour Flush toilet Pit Latrine Yes No

A.K Plappally et al.

Garbage Disposal Hire the service Cooking Fuels Yes Electricity LPG Natural gas Biogas Kerosene Coal/Lignite Wood/charcoal Straws/shrubs Animal Dung Other. Section D : Water Source Preferred interviewee: Primary Water collector or Head Female or Focal Mother Water Source Yes Piped Water Tube well/Borehole Dug Well Water from Spring Rainwater Tanker Truck No No Municipal service Throw in street/ River/Creek Bury/Burn other

1.

16

Oct (2010) 28th

River/Lake/pond Sachet Water Bottled Water Others 2. Do you know anything to make the water safer for consumption Yes 3. No but let it stand and settle No

A.K Plappally et al.

Dont know

What do you usually do to make the water safer to drink? (Ask indvidually) Yes Boil Add bleach/cholrine Strain through cloth Use Water Filter Solar disinfection Alum Multiple Treatment (Why & When & How) Other Dont Know No

4. 5.

How long does it take to fetch potable water? ------------------------------------What is the main source of hand washing water in your household? -----------------------------------Section E Physical Health Preferred interviewee: All in Household other than Children less than 5 years of age How do you describe your present health Excellent 1 Very Good 2 Good 3 Fair 4 Poor 5 Dont Know 0 In last month how many day did you have difficulty to perform your daily routine because of health reasons? None Every Day Number of days 0 1

1.

2.

3.

In the last month, how many days did you stay in bed because you were ill? None Every Day Number of days 0 1

4. 5.

Compared to a year ago how is your health?

17

Oct (2010) 28th

No change Much Better A little better A little worse Worse 6. 0 1 2 3 4

A.K Plappally et al.

In the last month, how much physical pain did you experience due to ill health? None Little Moderate Severe Very Severe In the last month did you have cough? Yes No Dont Know If yes

7.

Dry cough

Cough & Phlegm

Cough & blood

Other

8.

In the last month did you have Diarrhea? Yes No Dont Know If yes

Bloody

Diarrhea & Mucus

Liquid & Pale Diarrhea

Other

9.

In the last two weeks did you have fever? Yes No Dont Know

1 2 0

10. In the last month did you experience? Health conditions Head ache Fever Runny nose Nausea/Vomit Worms in Stool Back pain Joint pain Stomach ache

Yes

No

NA

1.

Section H Treatment Was anyone consulted about the treatment for the illness?

18

Oct (2010) 28th

Yes No Do not know Who was consulted about the illness ?(Note down all the applied) No one was consulted Doctor/Nurse Traditional Healer Chemist/Pharmacist Do not know others 1 2 0 0 1 2 3 4 5

A.K Plappally et al.

2.

3.

If no consultation or medical help was sought why not? ___________ No facility in Eweje village 0 Facility very far 1 Distrust of facility staff 2 No medication in Facility 3 No money to pay 4 Symptoms were not severe 5 Know what to do in this case 6 No time 7 Facility closed 8 Other 9 Was any type of treatment prescribed? If so, by whom? --------If not treatment was sought why not? --------Was any type of treatment undertaken that was not prescribed by a doctor/healer (eg. Home cure) Yes No Do not know How did the sick individual respond? No difference Reduced work time/Half day Stopped working/Missed School Section I Questions for Household Member in School What grade are you currently completing? 1 2 0

19

Oct (2010) 28th

0 % :No education 10% : Grade 1 15%: Grade 2 20%: Grade 3 25%: Grade 4 30%: Grade 5 35%: Grade 6 40% Grade 7 45%: Grade 8 50%: Grade 9 55%: Grade10 60%: Grade 11 65%: Grade 12 70%: Diploma 80%: Bachelors 90%: Honors 95%: Masters 100%: Masters &Higher

A.K Plappally et al.

1. 2. 3. 4. 5.

1. 2. 3. 4. 5.

In the last two weeks of school, how many days did you attend school? _______ In the last two weeks did you miss school due to illness? _______ Do you feel you could have avoided this illness if you had good access of potable water? _______ Do you know what your school does to make your water safer to drink? _______ Where is the water source located? _______ Section J (Preferred interviewee: Head female unless otherwise stated) Who within the family contributes to household monetarily or laboriously? _________ Who is the main individual that determines how money is spent in this family for common household expenses? _________ Who is the main individual that determines how money is spent in this family for medical treatment? _________ Did you work for money in the last two weeks? ____________ If so what is your occupation? _________________ How many days in the last two weeks did you work? ____________ Did you miss work in the last two weeks? ______________ If so, how many days? ________________ What do you think caused your illness? ____________ Do you feel you would be able to avoid illness if you had access to cleaner water? ______________ At work, what is your main source of drinking water? ___________ Do you know what does your work place do to make your water safer to drink? ____________ Where is the water source located? ______________ Section K (Preferred Interviewee: Primary Water Collector or Focal Mother) How many ceramic water filters do you have? ___________ How often do you use and refill the filters? ______________ What do you use the filtered water for? ___________ What is the frequency of cleaning the filters? ______________

20

Oct (2010) 28th

Appendix II Data Collected from 30 Families at Eweje Village, near Abeokuta, Ogun State, Nigeria, Africa

A.K Plappally et al.

Y-Health Response in %

X1Number of Filters for each family

X2- Filter Cleaning Frequency

X3 Children Below 5 Years of Age 3 0.4 0.2 2 2 0.4 0.2 2 2 0.3 0.3 0.5 0.2 0.3 0.2 1 0.2 3 0.2 2 3 1 0.3 0.1 1 2 0.2 0.2 1 1

X4 Members in a Family

X5- Wealth Status in %

X6 Education Status in %

X7Availability of Medical consultation

X8- Time for Filter Usage

80 70 50 45 65 96 47 97 78 80 95 75 65 55 45 45 65 90 55 55 65 85 85 65 60 60 60 55 65 45

1 1 1 1 1 1 1 1 2 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

80 90 90 95 80 55 26 95 78 85 98 55 86 40 30 60 70 90 85 80 90 20 80 60 80 80 40 30 40 50

4 1 3 5 5 5 1 11 6 5 5 4 2 3 2 3 3 8 3 8 9 5 4 1 4 6 1 1 3 3

80 26 82 85 90 55 40 90 55 98 85 90 75 85 90 80 55 90 90 75 90 45 65 50 75 80 55 55 55 80

30 25 60 42 87 90 25 96 62 75 65 90 60 80 75 60 80 90 80 85 80 55 72 30 60 55 60 60 60 50

40 85 60 28 88 90 15 80 40 65 90 95 85 55 87 45 85 55 50 55 55 60 65 55 45 80 55 80 80 60

3 3 3 3 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 0.5 3 3

21

Oct (2010) 28th

Appendix III

A.K Plappally et al.

From Eq. 9, deviation of predictor variable measurement lnXi,j from the sample mean of each of the predictor variables is written as (Plappally 2010)

where, A is auxiliary n x 1 matrix and n is the number of total measurements for each of the k predictor variables. Here . Then D is written as

The covariance of the predictor variables can be calculated by using the fact that Euclidean Inner product can be written as matrix multiplication (For example, where A and B are column vectors) as shown below

It is known that correlation values can be derived from covariance values of the predictor variables as depicted below, (A) This can also be written as

Here that is a square matrix, k x k, and if is a non-zero vector in the null space of dimensional space of the predictor variable data set, then

and exist in the n (B) (C)

Here is the null space of the characteristic equation term and left hand also represents two orthogonal vectors in the inner product space. It should be noted that contains all vectors perpendicular to the column spaces containing all the predictor variables (Plappally 2010). There will be one eigen space for each distinct eigenvalue for i=1,2..n and j =1,2k when n>k and so there will be k eigen spaces for . Trace of the Eigen value matrix also defines the total variance of the predictor variables. is an orthogonal square matrix of size 3x3 on the right hand side of Eq. 13, with normalized Eigen vectors participating predictor variables. The normalized Eigen vectors can be represented as for the three

Then , which constitutes all the principal components in the transformation matrix . The Vi, j for i=1, 2, 3 in Eq. 13, are the independent normal predictor variables with i, j for i=1, 2, 3 the corresponding variances (Plappally 2010).

22