Fecal and Urinary Excretion Profiles

Generation rate, composition, and physical and chemical nature of both feces and urine were recorded as of Table 1. Each recorded datum was the mean of the data from the reported study. Some published papers reported two or more independent studies so these papers contributed more than one value to the data set. The mean and median of each variable were both calculated as measures of central tendency and data were checked for normality by calculating a coefficient of skewness (Young, 1962 ):


Table 1

	Feces unit	Urine unit
Variable	of measure	of measure
Generation	g/cap/day	L/cap/day
Frequency of defecation	motions/24 hr	urinations/24 hr
Water content	% total mass	% total mass
Organic composition	% total mass	% dry mass
Components of solids	% total mass	% total mass
Inorganic composition	% dry mass	% dry mass
Daily excretion of elements	g/cap/day	g/cap/day, mg/L
Chemical nature
pH	pH	pH
COD and BOD	mg/g wet mass	mg/L
Physical form
Bristol stool form	Linear scale (1–7)
Diarrhea prevalence	% of population

σ = Standard deviationn = Valid number of casesBox and whisker plots were created using Statistica 11 software (Statsoft Inc., Tulsa, OK, USA, 2011). Outliers of each data set were defined using a standard default outlier coefficient value (Burns et al., 2005 (link)).

No outliers were removed from the data set but were identified in the graphical output. Full statistical calculations were only conducted on variables that had at least seven values but a median value is given for data when there were less than seven values.
A summary of studies used in the statistical analysis are outlined in Table 2, including the location and number of studies. A large proportion (80%) of the data set was from studies conducted in Europe and North America. A distinction was therefore made between low and high income countries by the measure of development; using the Human Development Index (HDI), a composite index measuring average achievement in three basic dimensions of human development; life expectancy, education, and income (UNDP, 2011 ).
Table 2

Country	n	HDI*	References
Africa	2	3/4a	Cranston and Burkitt (1975 (link)), Burkitt et al. (1980 (link))
Australia	2	1	Birkett et al. (1996 (link)), Hovey et al. (2003 (link))
Burma	1	4	Myo-Kin et al. (1994 (link))
Canada	3	1	Burkitt et al. (1980 (link)), Vuksan et al. (1999 (link))
China	3	2	Jie et al. (2000 (link)), Chen et al. (2008 (link)), Bai and Wang (2010 )
Denmark	2	1	Maclennan and Jensen (1977 (link)), Jensen et al. (1982 (link))
Developing countries	2	3/4a	Feachem et al. (1978 )
Europe and North America	1	1/2b	Feachem et al. (1978 )
European	1	1b	Mykkänen et al. (1998 (link))
Finland	4	1	Reddy et al. (1975 (link)), Reddy et al. (1978 (link)), Jensen et al. (1982 (link)), Mykkänen et al. (1998 (link))
Germany	1	1	Erhardt et al. (1997 (link))
Guatemala	1	3	Calloway and Kretsch (1978 (link))
Holland	4	1	Stasse-Wolthuis et al. (1980 (link)), Van Faassen et al. (1993 (link)), Gaillard (2002 ), Wierdsma et al. (2011 (link))
India	1	3	Shetty and Kurpad (1986 (link))
Iran	1	2	Adibi et al. (2007 (link))
Japan	7	1	Glober et al. (1977 ), Polprasert and Valencia (1981 ), Tarida et al. (1984 (link)), Saitoh et al. (1999 (link)), Danjo et al. (2008 (link)), Shinohara et al. (2010 (link)), Hotta and Funamizu (2009 (link))
Kenya	1	4	Cranston and Burkitt (1975 (link))
New Zealand	1	1	Pomare et al. (1981 )
North America	1	1b	Vuksan et al. (2008 (link))
Peru	1	2	Crofts (1975 (link))
Singapore	1	1	Chen et al. (2000 (link))
South Africa	2	3	Burkitt et al. (1972 ), Walker (1975 (link))
Spain	1	1	Roig et al. (1993 (link))
Sweden	4	1	Reddy et al. (1978 (link)), Vinneras (2002 ), Vinnerås et al. (2006 )
Thailand	2	2	Danivat et al. (1988 ), Schouw et al. (2002 (link))
Tonga	1	2	Pomare et al. (1981 )
UK	26	1	Olmsted et al. (1934 ), Connell et al. (1965 (link)), Southgate and Durnin (1970 (link)), Burkitt et al. (1972 ), Goy et al. (1976 (link)), Wyman et al. (1978 (link)), Prynne and Southgate (1979 (link)), Stephen and Cummings (1980 (link)), Eastwood et al. (1984 (link)), Eastwood et al. (1986 (link)), Davies et al. (1986 (link)), Cummings et al. (1987 (link)), Sandler and Drossman (1987 (link)), Cummings et al. (1992 (link)), Murphy et al. (1993 (link)), Cummings et al. (1996 (link)), Lewis and Heaton (1997 (link)), Chen et al. (1998 (link)), Reddy et al. (1998 (link)), Rivero-Marcotegui et al. (1998 (link)), Aichbichler et al. (1998 (link)), Almeida et al. (1999 ), Magee et al. (2000 (link)), Chaplin et al. (2000 (link)), Woodmansey et al. (2004 (link)), Silvester et al. (2011 (link))
USA	18	1	Canfield et al. (1963 ), Watts et al. (1963 (link)), Diem and Lentner (1970 ), Goldsmith and Burkitt (1975 (link)), Cummings et al. (1978 ), Glober et al. (1977 ), Goldberg et al. (1977 (link)), Beyer and Flynn (1978 (link)), Reddy et al. (1978 (link)), Calloway and Kretsch (1978 (link)), Kien et al. (1981 (link)), Polprasert and Valencia (1981 ), Tucker et al. (1981 (link)), Schubert et al. (1984 (link)), Parker and Gallagher (1988 ), Zuckerman, et al. (1995 (link)), Aichbichler et al. (1998 (link)), McRorie et al. (2000)

*Human Development Index Classifications (UNDP, 2011 ): 1. Very high, 2. High, 3. Medium, 4. Low.^aClassification not available, presumed to be ranking 3 or 4.^bClassification not available, presumed to be ranking 1 or 2.

Preliminary data analysis indicated that fiber intake was a major cause of variation in fecal generation and composition. There were a sufficient number of studies that had examined the effects of fiber intake on fecal output to enable further analysis to be undertaken on these data. The total dietary fiber intake was related to the generation of feces in linear and nonlinear regression analyses.