Four Footed Pharmacists: Indications of Self-Medicating Livestock in Karamoja, Uganda.Following observations of goats’ possible self-medication browsing the anti-parasitic plant, Albizia anthelmintica,
an ethnobotanical survey was undertaken to examine whether livestock
engage in other self-medicating behaviors, and if people also use the
same medications. Information was gathered over a five-month period from
147 Karamojong pastoralists and healers using a checklist of questions.
There were 124 observations for 50 proposed self-medicating behaviors,
primarily eating plants, to treat a total of 35 disease conditions. Of
the plant species, 72% were also prepared by informants to treat human
or veterinary diseases. Species importance was estimated by four
factors: >3 user citations, informant consensus factor >0.4,
fidelity level >40% and presence in the local pharmacopoeia. Eight
species fulfilled all of these factors, and 12 had at least three. These
results provide support for the hypothesis that animals graze specific
plants when ill and suggest that people have developed some of their
knowledge through animal observation.
Evidence for animals’ self-medication has accumulated over the past two decades (Engel 2002; Hart 2005; Huffman 2003; Lozano 1998). Research has concentrated on Africa’s great apes. Janzen (1978)
was the first to suggest that ingested secondary plant compounds
actually help animals to combat parasites. Research has since identified
chimpanzees’ self-treatment for internal parasitism through
leaf-swallowing of Aspilia spp. (Asteraceae) (Wrangham and Nishida 1983) and bitter pith chewing of Vernonia amygdalina Delille (Asteraceae) (Huffman and Seifu 1989) as well as other species (Lozano 1998).Other
studies approach proving self-medication by identifying and isolating
biologically active compounds responsible for specific pharmacological
effects, for example V. amygdalina (Huffman et al. 1993) Albizia grandibracteata Taub. (Fabaceae) and Trichilia rubescens Oliv. (Meliaceae) (Krief et al. 2005)
from which antiparasitic and antibacterial compounds have been isolated
following observations of chimpanzees. Watt and Beyer-Brandwijk (1962) earlier documented that indigenous East and South African people use some of these plants as medications.
The study of self-medication in animals is known as ‘zoopharmacognosy’ (Rodriguez and Wrangham 1993),
defined as the study of secondary plant components or other
non-nutritive substances used by animals for self-medication (Huffman 1997). Huffman (2007)
recently broadened the definition to include behavior and non-plant
substances used to suppress disease or to enhance animal health. Use of
soils and their properties has been well-documented in non-human
primates and elephants (Engel 2002) as have other behaviors that do not include ingestion of soils or plants (Clark and Mason 1985; Lozano 1998).
Ethnoveterinary
knowledge (EVK) refers to people’s knowledge, skills, practices, and
beliefs about the care of their animals (McCorkle 1986).
EVK has documented many local remedies, or ethnoveterinary medicines
(EVM), that animal keepers use for livestock health. These EVMs are
often taken to mean the use of medicinal herbs, but also include wild
plant and mineral products, and processes such as bone setting and
vaccination. However, EVK always has a human intervention component
(McCorkle et al. 1996).
There
has been little reference to livestock or other domestic animals in the
field of zoopharmacognosy. Moreover, most research has been by
behaviorists’ observations of wild or zoo animals, predominantly
primates. On the other hand, many farm animals lack access to
self-medication because they are confined and given a specifically
developed diet that has little bearing on what they would get in the
wild (Engel 2002). Research has shown in vivo antiparasitic effects of tanniferous plants that small ruminants may browse and graze (Niezen et al. 1998; Paolini et al. 2004).
However, definitive work on sheep self-medicating, when challenged with
illness-producing foods, was the first demonstration of multiple
malaise-medicine associations supporting zoopharmacognosy (Villalba et
al. 2006).
This paper investigates self-medication by domestic animals, as
observed by Karamojong pastoralists of north-eastern Uganda, where the
only fodder animals get is what they forage.
In 1998,
an ethnoveterinary survey catalogued over 70 different medicines
pastoralists in central Karamoja use to treat and prevent livestock
diseases (Gradé and Shean unpublished). Some of these veterinary
remedies are also used to treat human afflictions. Following this
survey, validation field trials of selected EVMs were undertaken to
evaluate their efficacy in the field (Gradé 2001).
Early in one such study, the first author observed an abrupt drop in
individual goats’ internal parasite loads (measured by daily fecal eggs
per gram and fecal egg count reduction), whereas this was not observed
in other species studied (unpublished data). Shepherds suggested that
this was related to selective browsing of a bitter plant, Albizia anthelmintica Brongn. (Fabaceae). It was explained that goats suffering from internal parasites tend to graze A. anthelmintica, followed by expulsion of gross worms in the feces and improvement in their symptoms (Gradé et al. 2007). This observation led us to search for other self-medicating behaviors by similar livestock in the project area.
The
primary objective of the present study was to investigate
self-medicating behaviors of Karamojong livestock, to examine the
hypothesis that animals seek out and graze specific medicinal plants or
employ other behaviors when sick. The survey addressed three central
questions: Do animals (domestic and wild) perform self-medicating
behaviors? If so, what plants or other materials do they use? And
finally, do people locally use these same items to treat disease in
livestock and/or themselves (i.e., are they included in the local
pharmacopoeia)? A secondary objective was to select those
plants/materials that were most important in order to validate their use
potential for promotion within the subregion, e.g. establishing them in
home gardens, follow-up field trials and possible drug development.
Methods
Study Site
The
region of Karamoja is located in northeastern Uganda. This study took
place in southwestern Karamoja between 2° to 2°03′ N and 34°15′ to
34°40′ E (Fig. 1).
Mean annual rainfall ranges from 500 to 750 mm, with 380 to 500 mm in
the plains and higher levels in the surrounding mountain ranges. Daily
temperatures average 30–35° C. The region has a semi-arid to arid
environment and is prone to cyclical droughts. Pratt and Gwynne (1977)
grouped Karamoja in Eco-Climate Zone VI of East Africa, in which
evaporation greatly exceeds precipitation (Karamoja Data Centre
unpublished data).
Fig. 1
Study area (credit: John O. Grade)
Thomas (1943) described the vegetation of Karamoja as consisting of Acacia-Combretum-Terminalia species associations, with an herb layer of Hyparrhenia, Setaria, Themeda, Chrysopogon and Sporobolus species. In contrast with neighboring districts, the short vegetation shows evidence of heavy grazing.
Karamoja consists of five geopolitical districts. Total population is around 935,000 (UBOS 2002) and contains five distinct Nilotic peoples in the plains and two small Kuliak groups found along the mountains (Gulliver 1952).
All of these groups live in conflict with one another and neighboring
tribes. One of the Eastern Nilotic ethnic groups in Uganda, the
Karimojong, has three major clans, from which the county names are
derived. Two of these clans, Bokora and Pian, are the focus of this
paper. The Bokora county population is 95,000, while Pian has 38,000
people (UBOS 2002).
They share a transhumant agro-pastoral lifestyle and speak the same
language, with slight differences. However, due to armed bilateral
cattle rustling, there are strong cultural taboos against sharing
livestock information between clans (Mirzeler and Young 2000).
These
pastoralists rely almost entirely on livestock for survival and
cultural events. Karamojong are semi-nomadic and have minimal health
care infrastructure for humans or livestock, with only one doctor per
57,133 people (Netherlands Development Organisation unpublished) and one
veterinarian for 90,000 livestock (Moroto District Veterinary Office
unpublished; Uganda Wildlife Authority unpublished). Culturally, people
rarely disclose their true number of animals, so even this low ratio may
be overestimated. Thus, the Karamojong rely almost exclusively on local
knowledge and medicines for treating diseases in both people and
animals (Gradé et al. 2007).
Daily
patterns and problem solving revolve around their livestock: when to
graze, when to water, when and how to protect livestock from wild
animals or raiders, how often to milk or bleed for human consumption,
which animals to select for sacrifice, and so on. Thus, they are highly
motivated to observe and well-manage their livestock.
Methods
We conducted our study, interviewing 12 groups (Table 1),
from April to August 2005. Twenty four traditional healers
participated, not only in discussions, but also in collections. These
healers are representatives of the Bokora and Pian- registered
traditional healer associations. An additional 123 pastoralists from the
community joined the discussions; of these 147 total respondents, 55%
were from Bokora and 45% were from Pian. One survey was performed at a
cattle market in Pian, while 11 were held in home villages of
traditional healers.
Table 1
Breakdown of community member numbers surveyed
Groups: (home, market) a
Registered Traditional Healers: (men, women)
Non-healers: (men, women)
Pian
8: (7, 1)
21: (16, 5)
60: (31, 29)
Bokora
3: (3, 0)
3: (2, 1)
63: (31, 32)
total
12
24
123
anumber of groups and their interview location
Information was gathered by a checklist of questions performed during field excursions and group discussions (Martin 1995).
Participants were asked to list behaviors in livestock that were
performed indicative of self-medication or symptomatic relief of disease
conditions; and also whether these practices, behaviors and remedies
were used by people in the treatment of disease in animals and/or
people. Data were cross-referenced by the first author’s observations to
determine if the observed self-medications were part of the
pharmacopoeia of Karamojong-prepared medications. They were also asked
from where they believed their indigenous knowledge had originated.
Informants
were asked to narrate observations with the following processes: 1)
animals having an obvious ailment (one the pastoralists could visually
diagnose); 2) these animals displaying behavior that is absent or rare
when healthy, such as grazing an unpalatable plant part; 3) following
this behavior, they observe an improvement in the animals’ symptoms; and
subsequently 4) the animal ceasing the said behavior as symptoms
improve. Provided their observation included step two (i.e. animals
displayed a behavior that is absent or rare when healthy) and most of
the other steps, the observation was included in the results. We are
intrigued by the possibility of a fifth step, in which after observation
of steps one through four, pastoralists experiment with using the
relevant self-medicating material as a prepared medicine. For points one
through four, we relied upon the pastoralists’ observations, memories,
and willingness to share information. Pastoralists are the best resource
for self-medicating livestock observations due to their constant close
proximity to the animals, high motivation to observe and monitor
disease, and years of accumulated experience on animal care, diagnosis,
and treatment. It is important to note the potential barriers and layers
of translation between the author and the animals in this study. There
was a clear barrier between the animals and pastoralists, who discerned
disease conditions – including subjective conditions such as headache
and nausea – from healthy animal behavior. There were also barriers
separating the first author and pastoralists, due to differences in
language, and in cultural views of disease etiology, medication, and
livestock behavior. Therefore indices of informant consensus factor
(ICF) and fidelity level (FL) were used to guide interpretation of our
observations and efforts were made to draw responses from the emic
perspective.
Indigenous disease terminology was
verified by the primary author, an experienced veterinarian. Diseases,
processes, and disorders were grouped according to the usage category
standard for human symptoms and ailments developed by Cook (1995).
In alphabetical order, the usage categories are as follows: digestive
system disorders, ill-defined, infection/infestations, injuries,
metabolic disorders, nutritional disorders, pregnancy/birth/puerperium
disorders, prophylaxis, respiratory disorders and skin disorders.
Self-medicating remedies were grouped according to botanical family
(Angiosperm Phylogeny Group 2003), if plant-based; as ‘mineral’ if inorganic; or as ‘water’ if water was a component.
Plants mentioned by participants were photographed and collected in situ.
Collected plants were sent to Makerere University Herbarium in Kampala
for botanical identification according to Flora of Tropical East Africa.
Karamojong names for some plants are connected to more than one
scientific name; we refer to these data as ethnospecies (Reyes-Garcia et
al. 2006).
All plant species and material are listed with small caps in
vernacular. Vouchers are kept at Makerere herbarium and a community
herbarium located at the study’s partner NGO, KACHEP, in Nabilatuk,
Karamoja.
We conducted our survey as part of KACHEP’s
ongoing community development project focusing on animal health care,
local capacity building and EVK awareness. Key informants were healers
associations’ formations registered at the national level with whom the
first author has a long standing (8 years) relationship, facilitating
their whose associations were, development of intellectual property
rights awareness, and their goals and vision; therefore prior consent
was obtained.
Data Analysis
Interview
responses were compiled into narrative form, coded and entered into an
Excel spreadsheet. Analysis included calculation of ICF (Equation 1) and FL (Equation 2). ICF was adapted from Trotter and Logan’s informant agreement ratio (1986)
to measure informant groups’ consensus and thereby infer consensus
among different animals observed. We specifically looked into three
areas; individual self-medicating remedies and categories of these
remedies, and usage categories of disease (e.g. a specific plant
species, and a particular plant family and category of disease as
described by Cook). In this method, the relative importance of each use
is calculated directly from the degree of consensus in informants’
responses (Phillips 1996).
The importance of different plants or uses is assessed by the
proportion of informant groups who independently corroborate different
animals’ self-medicating behaviors. The ICF has a maximum of 1.0
(maximum consensus) and a minimum of 0 (no consensus). ICF for an
individual self-medicating remedy, for example, is calculated, as
follows: number of observations of the remedy (obs) subtracted by the
number of ailments (use category of disease, UC), divided by the number
of observations of the particular remedy minus one:
ICF=obs−UCobs−1
(1)
We
used ICF to evaluate the hypothesis that livestock self-medicate. A
self-medicating behavior (SM) with a positive ICF indicates that
livestock in different areas displayed a consistent behavior; i.e. that
local people observed animals with a given disease tended to perform a
particular SM. Among usage categories, a positive ICF indicates
specificity of the SM for a given disease condition; i.e., animals
performing a given SM tended to have the same disease condition, rather
than haphazardly chewing an unpalatable plant part for any illness. ICF
for disease categories took into account a summation of observations
within the category. Each SM was counted only once, even if it appeared
in more than one disease within the usage category.
Fidelity level (FL, Equation 2), adapted from Phillips (1996),
was used to determine the most important self-medicating species in
each specific usage category. Fidelity level is derived from
FL(%)=NpN×100
(2)
where Np is the number of observations of a particular self-medicating behavior for one usage category, and N is the total observations for the behavior for any use.
We
are confident that our findings are significant, particularly given
that Bokora and Pian forbid sharing livestock-related information
between clans. Elevated ICF and FL are therefore strong indicators of
multiple, independent and congruent observations.
We
estimated plant/material importance as a function of four criteria. The
species must have at least three uses in order to be considered, The
other three criteria are: ICF >0.4 (Moerman 2007), FL >40% and presence of the treatment in the local pharmacopoeia.
Results
There
were 124 use citations (observations) for 50 proposed self-medicating
behaviors used to treat a total of 35 disease conditions, as observed by
12 informant groups. The domesticated animals mentioned were goats,
cattle, sheep, donkeys, camels, poultry and dogs, in decreasing order of
self-medicating behavior observations. Cats and pigs were not
mentioned. There were single self-medicating behavior mentions for three
wild animal species; buffalo, kudu and guinea fowl. As stated in the
methods section, to be considered a self-medicating activity, the
informants observed animals displaying a behavior that is absent or rare
when healthy, as part of the following process: 1) animals having an
obvious ailment (with symptoms pastoralists could visually diagnose); 2)
these animals displaying behavior that is absent or rare when healthy,
such as grazing an unpalatable plant part; 3) subsequent improvement in
the animals’ symptoms; and 4) subsequent cease in self-medicating
behavior. All the 50 proposed self-medicating behaviors fulfilled
criterion two, almost all of the 50 also fulfilled steps one, three and
four with step four being the least, but still fetching > 70% of 124
observations.
Most Observed Self-Medicines
Table 2
shows the 50 observed self-medicating remedies and behaviors in
alphabetical order, divided in three parts, i.e. specific oral remedies,
non-specific oral remedies and non-oral self-medicating behaviors. The
most commonly mentioned self-medicating behavior was grazing Ekosimabu (Loudetia superba De Not), which was reported nine times out of the total of 124 reports. Three Karamojong remedies had eight observations: Akawoo (Cymbopogon giganteus Chiov. or Bothriochloa insculpta (A. Rich.) A. Camus), Edoot soils, and Loletio (Eragrostis pilosa P. Beauv.). Two species had six observations: Ecucukwa (Aloe tweediae Christian or A. dawei Berger) and Ekapangiteng (Albizia anthelmintica). Note that both Akawoo and Ecucukwa are ethnospecies with more than one scientific name.
Table 2
List
of self-medication remedies and behaviors, (a) for specific oral
treatments’ (b) for non-specific oral (c) for behavior, not ingestion of
remedy
material (part used)
Scientific name
Family
Obs
UC
ICF
PM
FL
(a)
Abalongit
Mineral - soda ash
‘mineral’
2
2
0
yes
50 R, D
Abir (wp)
Sphaeranthus ukambensis O Huffm.
Asteraceae
1
1
0
yes
100 D
Abwach (b, l)
Warburgia ugandensis Sprague
Canellaceae
3
3
0
yes
33
Akawoo(l)
Cymbopogon giganteus Chiov.
Poaceae
8
3
0.71
no
50 I
Bothriochloa insculpta(A. Rich.) A. Camus
Poaceae
Akouma
Anthill or termite mound
‘mineral’
1
1
0
yes
100 N
Amojoi (l, fl)
Agave sisalana Perrine ex Engelm.
Agavaceae
1
1
0
no
100 P
Apuna (l)
Bulbostylis pusilla (A. Rich) C. B. Cl. Subsp congolensis (De Wild) R. Haines
Cyperaceae
5
4
0.25
no
40 N
Cassava (l)
Manihot esculenta Crantz
Euphorbiaceae
1
1
0
yes
100 D
Ecorgorum (l)
Not collected
NI
1
1
0
yes
100 I
Ecucukwa(l)
Aloe tweediaeChristian
Aloaceae
6
4
0.4
yes
50 X
Aloe dawei Berger
Aloaceae
Edapal (l, fr)
Opuntia cochenillifera (L) Mill
Cactaceae
3
3
0
yes
33
Edipidipi(l, b)
Erythrococca bongensisPAX
Euphorbiaceae
3
2
0.5
yes
67 I
Edoot
Clay or anthill like soils
‘mineral’
8
4
0.57
yes
63 N
Egigith (wp-r)
Cissus quadrangularis L.
Vitaceae
2
2
0
yes
50 I, ID
Ekadolia(l, fr)
Capparis tomentosaLam.
Capparidaceae
3
2
0.5
yes
67 X
Ekamuria (l, b)
Carissa edulis (Forsk.) Vahl.
Apocynaceae
1
1
0
yes
100 I
Ekapangiteng(l, b)
Albizia anthelminticaBrongn.
Fabaceae
6
2
0.8
yes
83 I
Ekapeliman (l)
Acacia nilotica (L.) Del.
Fabaceae
2
1
1.0
yes
100 X
Ekara (l)
not collected
NI
1
1
0
yes
100 X
Ekere(l, s)
Harrisonia abyssinicaOliv.
Simaroubaceae
5
3
0.5
yes
60 I
Ekorete (l, fr)
Balanites aegyptiaca (L.) Del.
Balanitaceae
1
1
0
yes
100 X
Ekosimabu(wp)
Loudetia superbaDe Not.
Poaceae
9
5
0.5
no
43 I
Eligoi (wp-r)
Euphorbia tirucalli L.
Euphorbiaceae
2
2
0
yes
50 N, X
Cissus spp.
Vitaceae
Elira (l, b)
Melia azedarach Linn.
Meliaceae
1
1
0
yes
100 X
Eminit (l, b)
Acacia gerradii Beuth.
Fabaceae
1
1
0
yes
100 X
Epederu (l, fr)
Tamarindus indica L.
Fabaceae
1
1
0
yes
100 X
Epeeru(wp-r)
Cassia nigricansVahl.
Fabaceae
5
1
1.0
yes
100 I
Epetet (l, b, fr)
Acacia aerfote (Forssk.) Schweinf.
Fabaceae
2
2
0
yes
50 I, J
Erogorowete (l, b)
Capparis spp.
Capparidaceae
2
2
0
yes
50 I, X
Eteteleit (l, b)
Acalypha fruticosa Forssk.
Euphorbiaceae
2
2
0
yes
100 X
Etirir (l, b)
Acacia spirocarpa Hochst. ex A. Rich.
Fabaceae
1
1
0
yes
100 X
Ewonokori (l, b)
Capparis spp.
Capparidaceae
2
1
1.0
yes
100 I
Eyelel (l, fr)
Acacia drepanolobrium Sjöstedt
Fabaceae
1
1
0
yes
100 X
Jokpolon (l)
Hyparrhenia spp.
Poaceae
1
1
0
no
100 I
Khat (l)
Celastrus edulis (Forssk.) Vahl
Celastraceae
1
1
0
yes
100 ID
Loletio(l)
Eragrostis pilosa(L.) P. Beauv.
Poaceae
7
4
0.5
no
57 N
Longarwe (l, b)
Cissus spp.
Vitaceae
1
1
0
yes
100 I
Synadenium grantii Hook f.
Euphorbiaceae
Losgirai (l)
Amaranthus spinosus L.
Amaranthaceae
1
1
0
yes
100 I
Mumwa (wp-r)
Sorghum vulgare Pers.
Poaceae
1
1
0
yes
100 X
water ofAri Ekosimabu(l)
Ari = bend in the river, lined withL. superba
Poaceae and ‘water’
3
2
0.5
no
67 I
(b)
Bad water
non-plant
‘water’
1
1
0
no
100 ID
†Bitter, smelly plants (l, b)
not specific
NI
3
1
1.0
yes
100 X
Burnt grass and its ashes (wp)
not specific
Poaceae
3
3
0
yes
33
Good grass and water (l)
not specific
Poaceae and ‘water’
1
1
0
no
100 X
Ninya (l)
not specific
Poaceae
1
1
0
no
100 D
Sour grass near clay soil (l)
not collected
Poaceae
3
3
0
no
33
Water
non-plant
‘water’
1
1
0
yes
100 I
(c)
Fence rubbing
not specific
‘mineral’
3
3
0
no
100 I
Rain standing
non-plant
‘water’
1
1
0
no
100 I
Rocky soil
non-plant
‘mineral’
1
1
0
no
100 I
TOTALS
124
10
72%
part used
= plant part used; l = leaves, b = stem bark, fr = fruit, fl = flower, r
= root bark, wp-r = whole plant, wp-r = wp without roots. If the remedy
is not a plant or is not ingested – this is left blank
Obs = # of observations or use citations of this self-medicating remedy by informant groups
PM = medicine is also prepared for human and/or veterinary use
FL = Fidelity Level% (Equation 2).
With the highest UC noted, except if FL 33% - thus three UC, where D:
digestive, I: infection/infestations, ID: ill-defined, J: injury, N:
nutritional, P: pregnancy/birth/peurperium, R: respiratory, S; skin
diseases, X: prophylaxis
NI = not identified
† = group of 3–8 plants
Those self-medicating behaviors that have high importance factors are bold.
Forty-seven of the 50 self-medicating behaviors, or 94%, were taken orally (part a and b in Table 2). Only three behaviors had no oral component (part c, Table 2).
Forty-two, or 84%, of the self-medicating behaviors documented here
involved plants. Five had a water component, and six had a mineral
component. Three behaviors spanned two of the above categories.
Table 3
lists families associated with livestock self-medication, in decreasing
frequency of observations. The Poaceae family was represented by 10
species, whereas Fabaceae had eight and Euphorbiaceae had five species
that livestock had been observed to graze or browse when ill. A general
category of ‘mineral’ was designated for geophagia (soil eating) and
behaviors such as walking on rocks or rubbing against fences. This
category had five remedies.
Table 3
Families associated with karamojong livestock zoopharmacognosy
Family
# species mentioned
Obs
UC
ICF
most common UC observed
Poaceae
10
36
6
0.86
Infection
Fabaceae
8
19
4
0.83
Infection
‘Mineral’1
5
18
6
0.71
Nutrition
Euphorbiaceae
5
8
4
0.57
Infection
Capparidaceae
3
7
2
0.83
Infection
Vitaceae
3
5
4
0.25
Infection
Aloaceae
2
6
4
0.4
Infection
Other (13 families)
1
1
1
0
UC = # of different usage categories ‘treated’ by this family
Obs = # of observations or events of particular family by informant/group
1 = Mineral group was included as a common category, not as a botanical family
The
Poaceae family also had the most observations, with 36 citations in six
different usage categories. This included seven citations for Ekicuyan,
a common but somewhat vague nutritional ailment described below. The
Fabaceae family had 19 use reports for four usage categories, six of
these reports involved internal parasites (part of infection/infestation
usage category). The mineral group had 18 observations for six usage
categories, including five for Ekicuyan.
Most Commonly Observed Diseases, Processes, and Disorders
Table 4 lists the 35 different disease conditions grouped according to usage category (Cook 1995),
in decreasing frequency. In order to protect Karamojong intellectual
property rights, the specific medicinal applications of each plant are
not given at this instance, although the most common remedy for each
category is highlighted in table 4. Furthermore, the discussion will
raise one example, Albizia anthelmintica
for internal parasitism, as the plant and its use are widely
distributed public knowledge and the NGO and CBO partners felt
comfortable with this decision. Internal parasitism was the most common
condition addressed by self-medicating behaviors. This condition was
observed 13 times, with eight different self-medicating behaviors.
Twelve observations each for disease prevention and health were noted,
while Ekicuyan had 11 observations with eight different remedies. Ekicuyan, a culturally bound syndrome, can refer to human or veterinary disease and is described by pastoralists as heartburn or Akicwe, anthropomorphically described as ‘salt-craving’ or ‘meat-deficiency.’
Table 4
Usage
categories in decreasing order of observations (obs), listed with
individual diseases, processes and disorders (dpd) in decreasing order,
number of unique self-medicating behaviors (sm) for each usage category
and informant consensus factor (icf)
Usage category
individual diseases, processes and disorders*
Obs
SM
ICF
most common remedy
I
Infections/ infestations
Internal parasitism (13)
Ebaibai - FMD
Lopid - anaplasmosis
48
22
0.55
A. anthelmintica and C. nigricans
Tick infestation (6)
Eyaliyal∼ tetany
Malaria fever
Heartwater (5)
Lice
Ekore - chest pain
Lokit - ECF (4)
Loleo - rinderpest
Loukoi - CBPP
Emitina - goat mange (3)
Lokecuman - black quarter
Scabies
X
Prophylaxis
Disease prevention (12)
Health (12)
Non specific-health (7)
31
25
0.59
A. nilotica and C. tomentosa
N
Nutritional Disorders
Ekicuyan∼‘heartburn’ (11)
Thin (low BCS) (5)
Anorexia (not eating)
18
8
0.2
E. pilosa
P
Pregnancy/Birth/ Puerperium
Milk production (8)
8
8
0
D
Digestive System Disorders
Ekitubon - bloat
Constipation
Nausea
6
5
0.2
L. superba
Diarrhea
ID
Ill-Defined Symptoms
Ill-defined
Lethargy
Weak calves
4
4
0
R
Respiratory System Disorder
Coughing
Difficulty breathing
3
3
0
S
Skin/Subcutaneous Cellular Tissue Disorders
Poor coat
3
3
0
J
Injuries
Snake bite
Wound
2
2
0
M
Muscular-Skeletal System Disorders
Headache
1
1
0
TOTALS
124
50
0.6
* = # observations (Obs) listed with individual DPD when ≥3
SM = # of self-medicating remedies used for usage category
The most frequently used remedy for each category is listed, in bold if it had high importance.
Usage Groups
Table 4
lists the usage categories in decreasing order of user citations.
Infections/infestations had the most individual diseases, processes or
disorders and the most user citations with 49, or 39.5% of total.
Following this were the prophylaxis group with 31, 18 in nutritional
disorders, and eight for milk production, which was the sole process
mentioned in the pregnancy/birth/puerperium disorders category.
Self-medication Indices
As
stated in the methods section, we estimated plant/material importance
with following criteria: three use citations, ICF >0.4, FL >40%
and presence of the treatment in the local pharmacopoeia.
ICF (Equation 1) was used to measure the level of consensus in individual self-medicating behaviors (Table 2), families (Table 3) and usage categories (Table 4,
ICF = obs-SM/obs -1). Fourteen of the 50 identified self-medicating
behaviors, or 30%, had an ICF >0.4. Four remedies had the highest
possible ICF of 1.0. When considering remedy use at the individual
ailment level, vs. the more general, grouped ailment usage level used in
this study, Albizia anthelmintica
has the highest ICF, 0.8. All but one of its observations was for
internal parasitism (infections/infestations category); the other
observation was under the prophylaxis category. The mineral group had an
ICF of 0.71. Only two of the 10 usage categories (Table 4)
had ICF >0.4: prophylaxis (ICF = 0.59) and infection/infestation
(ICF = 0.55), whereas nutritional disorders and digestive system
disorders both had low positive ICF of 0.2.
Table 2
shows the fidelity levels (last column) for each self-medicating remedy
according to usage category. The infection category had the highest
fidelity level. Not including those behaviors with only one user
citation (which yields 100% FL by default); six remedies had perfect
fidelity levels. There were 14 self-medicating behaviors with more than
three use citations and a FL >40%.
Thirty-six of
the 50 remedies, or 72%, were also used as a prepared medication by the
informants to treat their animals and/or their families (Table 2).
Thirty-six, or 76.6%, of the 47 orally applied self-medicating remedies
are also prepared medications. If the prophylaxis category is removed
(in other words, counting only observations in apparently sick or weak
animals), the overlap is 65.8% or 25 of 38; if we only include oral
treatments, the overlap increases to 71% (25 of 35).
Table 2
sets the most important remedies in boldface. Four importance factors
were considered, eight self-medicating remedies fulfilled all of these
factors, and 12 had at least three factors.
Indigenous Knowledge Origin
When
asked to free-list the sources of their indigenous knowledge (IK),
informants identified three main sources: 1) Creator-God; 2) people,
both dead and living; and 3) animals. All 12 participant groups reported
receiving IK from the first two categories through dreams, visions,
oral traditions, personal observation, and study of tradition. Only two
informant groups noted obtaining IK from animals by observing the
animals’ behavior; furthermore, it was only one or two individuals that
admitted this – much to the laughter of others.
Discussion
Importance Factors
Although
there are no universally agreed upon criteria for distinguishing
self-medication (SM) from routine eating or other non-self-medicating
activities, several tools have been proposed, such as evaluating plant
parts bioactivities of chimpanzee diets (Krief et al. 2006), Huffman’s steps (Huffman and Seifu 1989) and comparisons of what chimps eat as to what local people use medicinally (Huffman et al. 1996; Krief et al. 2005). The importance factors used in this paper drew on Krief et al.’s (2005)
work comparing chimpanzees’ diets with the pharmacopoeia of people
throughout the world. We also relied on the field of ethnobotany, which
has developed methods for studying consensus between cultures (Phillips 1996).
Any individual observation of a SM may be important and relevant to the
argument that animals self-medicate. However, a SM that is reported
more often (in this study, we set three use citations as an a priori
minimum), we believe is more likely to represent a true SM. The
self-medication hypothesis is further supported in this study by
similarity in SM and Karamojong pharmacopoeia, >0.4 ICF, and >40%
FL. Therefore the self-medication hypothesis is especially supported by
those boldfaced remedies in Table 2, as they fulfill the above criteria.
Our
results show that most self-medicating behaviors were orally, and that
most were plant-based. Both of these findings are consistent with the
existing zoopharmacognosy literature (Clayton and Wolfe 1993; Engel 2002; Huffman 2003; Krief, Hladik, and Haxaire 2005; Lozano 1998; Nègre et al. 2006).
Fur rubbing has also been observed in black lemurs and capuchins with
millipedes, in several bird species with ants, other non-human primates
with plant resins and leaves (Birkinshaw 1999). However, in our study, the informants did not specify which tree species the livestock rubbed upon.
Small
sample size is a possible limitation of this study, as it may not have
been adequate to distinguish some valid self-medicating activities from
background noise; therefore we set three use citations as a minimum for
importance. Indeed, the sample size of this study is notably smaller
than in other research involving ICF, which may involve thousands of
user citations (Heinrich 2000; Heinrich et al. 1998).
On the other hand, zoopharmacognosy research has generally been
anecdotal, following one animal or a small group of animals with much
fewer use citations, in the range of one to 30 (Huffman and Seifu 1989;
Krief et al. 2005; Wrangham and Nishida 1983).
However, due to the innovative style of this study, recording every
observation supposed to be a self-medicating activity is important in
that it would support the self-medication hypothesis.
One
might assume that the more abundant a particular plant, the more
commonly animals graze it. However, pastoralists did not remark on what
their livestock typically ate, but rather what they specifically ate
while visibly displaying an illness. Goats were the exception, as they
routinely browse on bitter plants. Informants thought many of these
remedies ‘strengthened their animals and kept them from falling ill’
which we etically translated to mean disease prevention, and overall
health promotion. This is shown in the prophylaxis usage category which
had 26 of the 50 different self-medicating remedies. This indicates that
pastoralists associate more than half of their livestock’s
self-medicating behaviors as routine ways to enhance their health
without shepherds’ involvement. Prophylaxis is a difficult concept. Many
of the observations appearing in this category, are those when the
individual informant could not ascertain the animal’s illness or why the
animal they observed was acting ‘queerly’ or eating something that
would be out-of-the-ordinary. So they in turn presumed that the animals’
behavior was to strengthen it or prevent it from falling ill. However,
one might suspect that observers are anthropomorphizing: if they
consider a plant to be medicinal-only or unpalatable as food, they may
assume that their animals perceive it the same way, and thus attribute a
medicinal purpose to use by a healthy animal that may well not exist.
Or, they might follow the line of reasoning, because they collect and
prepare the same plant as medicine – the animal might be doing the same
thing. However, Karamojong pastoralists do understand that diseases
spread, from animal to animal, through vectors (animate and inanimate).
They will separate and quarantine those animals that are ill, or keep
their healthy animals from mingling with others they perceive as ill.
They will lead their animals to certain grazing areas or natural
salt-licks when the suspect illness or force-feed minerals or medicines
to keep a new dam healthy or to ‘strengthen’ her (Gradé personal
observations).
Parasitic diseases, overall, are the
most common ailment which livestock self-medicate as recognized by
Karamojong pastoralists. Thirty-seven, or 29.8%, of the 124
self-medicating observations were for internal or external parasites and
the diseases they transmit. This is consistent with existing
zoopharmacognosy literature: the most studied plants, Vernonia amygdalina and Aspilia spp., act against internal parasites (Lozano 1998) and parasitism is the most common ailment that has been noted in zoopharmacognosy (Engel 2002; Huffman 2003), additionally Clark and Mason (1985)
reported on external parasites, i.e. nesting behavior in birds. The
most commonly mentioned dewormer for livestock and humans in our study
was Albizia anthelmintica, which
was also the remedy with the highest importance factors. Even using the
more stringent usage level, ICF is 0.8 (i.e. internal parasitism vs.
general usage category infection/infestation) and its FL (83%) was the
highest for a remedy with >6 observations. FL was used to quantify
importance of species for a given purpose (usage category).
The
idea that pastoralists experimented with the relevant remedy as a
prepared medicine is supported by the observation some respondents
claimed part of their IK came from animal observation, and is further
addressed with 72% self-med to prepared med overlap. The overlap was
higher than the 36% that Krief et al. (2005)
found in comparing ethnobotanical uses to typical chimpanzee diets in
Kibale, Uganda. Our higher levels could be that our study focused on
self-medicating activities whereas their chimpanzee study examined the
entire diet. Additionally, our study looked only at the local
pharmacopoeia whereas theirs used a world-wide ethnobotanical database.
In this instance, our higher level self-to-prepared med overlap suggests
that some of Karamoja’s ethnomedicine origins were derived from animal
observations.
Future objective and quantifiable research in this area should include:
a)
Objective diagnosis and documentation of animals’ disease condition with laboratory analysis.
b)
Documentation of the frequency of behavior that is absent or rare when healthy, such as grazing an unpalatable plant part.
c)
Documentation of resolution or improvement in disease condition, through standardized observation and/or laboratory means.
d)
Documentation of cessation of the unusual behavior after improvement in symptoms.
e)
Finally,
reproduction of the beneficial effect in other diseased animals
following administration of the relevant self-medicating behavior.
Trained behaviorists would more objectively evaluate steps b, c, and d. Villalba et al. (2006)
essentially performed the last step in their sheep study by proving
that animals choose appropriate plants to ameliorate their illness as a
learned behavior.
Objective monitoring that are
scientifically quantifiable and verifiable (steps a, c, and e), make
internal parasitism attractive for further research. Much like Aspilia spp. and Vernonia amygdalina have drawn primatologists’ attention (Lozano 1998), Albizia anthelmintica
is a particularly good candidate for further exploration of livestock
pharmacognosy. In addition to its high importance factors in this study,
A. anthelmintica has shown anthelmintic activity in several studies (Gathuma et al. 2004; Githiori et al. 2003; Gradé et al. 2007; Koko et al. 2000).
To
the knowledge of the authors, this is the first time that pastoralists’
wealth of knowledge has been utilized as part of self-medicating
behavior research, or that ICF and FL have been used to objectively
evaluate the zoopharmacognosy hypothesis. Our results indicate that
animals display multiple behaviors consistent with self-medication.
There is also reason to suggest that pastoralists have developed some of
their ethnopharmacological knowledge from careful animal observation.
Finally, high importance species merit further investigations;
identification/isolation of biologically active compounds (in particular
those that are in the infection/infestations category), promotion at
the local level, and perhaps drug development and primary health care.
Acknowledgements
Many
thanks to the registered healers of Bokora and Pian and to the staff of
KACHEP for sharing your wisdom and welcoming us into your hearts. The
traditional healers and pastoralists of Karamoja are the owners of the
information presented in this paper and any benefits that may arise from
the use of this information will belong to them. Gratitude to the
editing assistance of friends and reviewers. For identification of plant
specimens, we thank Olivia Wanyana Maganyi and Protase Rwaburindore of
Makerere University for their expertise. We are also grateful to the
Uganda National Council for Science and Technology for permission to
work in Karamoja. Funding has been through Christian Veterinary Mission,
Seattle, USA.
