Local understanding and practices related to IMCI interventions in eastern Tanzania

This PhD thesis presents findings of the health seeking component of the Tanzania
Essential Health Intervention Project (TEHIP). It was carried out from 1998 to 2001 in
two districts of southern Tanzania where Integrated Management of Childhood Illness
(IMCI) was introduced in 1997. The rationale is that best IMCI services are of little
benefit, if they do not reach community and household levels. Caregivers need to
understand and comply with IMCI core principles, i.e. learn to recognize the correct
danger signs and seek prompt and effective treatment.
The goal of our study was to contribute to increasing “community effectiveness” (Tanner
et al. 1993) of health care in the study districts. Our specific objectives were to generate
local knowledge to better adjust the IMCI interventions to local health seeking behavior
and to improve the ways in which caretakers identify and manage common childhood
illness.
We define health seeking to encompass three dimensions: 1) health concepts including
signs and symptoms recognized by the community; 2) aetiology comprising
interpretations and explanations of illness; and 3) help seeking referring to home
management and all forms of seeking help from experts, whether these are neighbors,
traditional healers or health care staff.
We first investigated the local illness terminology and the relative importance of symptom
recognition and labeling in care-seeking. We found that local illness terms overlap with
biomedical classifications such as “malaria”, but this overlap does not constitute direct
correspondence. Caregivers rarely see a link between malaria and convulsions and
create new links between convulsions and polio, tetanus and epilepsy. We identified
intra-cultural diversity in symptom recognition and severity ranking of the same illness.
Caregivers search for illness labels which are not only a name but contain information
about treatment. In this search they face difficulties due to two reasons: 1) different
illnesses produce similar symptoms, and 2) different persons provide changing and even
contradictory advice and information. We introduce the term “fuzzy concept” and suggest
that fuzziness can be explained by the diverse manifestations of malaria, by intra-cultural
variability and/or by culture change confronting individual persons with multiple
meanings.
In a second step we analyze local aetiologies which we consider equally important for
appropriate care-seeking as prompt recognition of danger signs and symptoms.
Community aetiologies of IMCI related illness encompass a wide spectrum ranging from
natural to supernatural causes. Some caregivers act on these notions, others are not
interested in causes, and the majority remains ambivalent and pragmatic. A closer
analysis of malaria-related aetiology shows that caregivers clearly attribute malaria to
mosquito bites but have fragmented knowledge about the aetiology of homa (fever) and
degedege (convulsions). We suggest that aetiological uncertainty leads to difficulties in
therapy choice and thus to pragmatic ambivalence.
In a third step we assess care-seeking in actual illness episodes. Caregivers make
extensive use of formal health care facilities, not only for homa and malaria but for most
other IMCI related illnesses. Exceptions are the folk illnesses degedege and kimeo
(elongated uvula). The basic distinction found in many parts of Africa also applies to our
study sites: 1) mild and “normal” malaria is first treated at home and if not cured brought
to a formal health care facility; 2) severe forms of the disease presenting convulsions are
rarely considered as malaria but as a distinct illness entity requiring traditional treatment
at home or from a traditional healer. Our most important finding is that many children
who suffer and sometimes even die from convulsions have had not only a history of
homa but have even been diagnosed and treated for malaria in a health facility before
they developed convulsions.
In the fourth and final step we examine the impact of malaria care-seeking patterns on
childhood mortality. Our findings show that nearly 80 percent of malaria-attributable
deaths used modern biomedical care as a first resort, both in the form of antimalarial
pharmaceuticals from shops or formal health care services. If care was sought more
than once in these fatal cases, modern care was included in the first or second resort in
90 percent with convulsions and 99 percent without convulsions.
There clearly is an urgent need for a thorough analysis of what is happening in these
cases. Health providers often formulate the problem of recurrent illness in terms of a
delay in treatment or a lack of adherence to treatment regimes. We argue that victim
blaming does not bring us any further. Our case studies demonstrate that many mothers
make an enormous effort in time, energy and money searching for the best care for their
child. They should be supported rather than blamed by the health system.