Content-Type: text/html

COGNITIVE-AFFECTIVE AND BEHAVIOR-AFFECTIVE
DIMENSIONS: A COMPARATIVE ANALYSIS OF PARTICIPATORY & DIFFUSION APPROACHES
IN THE DESTIGMATIZATION OF LEPROSY
A Case Study in Gwalior, India
 
 
 
By
 
Pradeep K. Krishnatray
Srinivas R. Melkote
 
 
Please direct all correspondence regarding this paper to: Dr. S. Melkote,
Graduate College, Bowling Green State University, Bowling Green, OH 43403.
(Telephone: 419/372-9324);
Fax: 419/372-8569; E-mail: <a href="/cgi-bin/wa?LOGON=A3%3Dind9710B%26L%3DAEJMC%26E%3D7BIT%26P%3D942329%26B%3D--%26T%3DTEXT%252FPLAIN%3B%2520charset%3DUS-ASCII" >[log in to unmask]</a>
 
 
Paper submitted to the Theory and Methodology Division of the AEJMC
 
 
Pradeep K. Krishnatray ( Ph.D., Bowling Green State University, 1996) is a
professor of communication at Mudra Institute for Communications, Ahmedabad,
India.  Srinivas R. Melkote (Ph.D, University of Iowa, 1984) is a professor of
telecommunication in the School of Communication Studies at Bowling Green State
University, Ohio, USA.
 COGNITIVE-AFFECTIVE AND BEHAVIOR-AFFECTIVE DIMENSIONS:
A COMPARATIVE ANALYSIS OF PARTICIPATORY & DIFFUSION
APPROACHES IN THE DESTIGMATIZATION OF LEPROSY
A Case Study in Gwalior, India
 
 
 
ABSTRACT
This was an experimental study designed to determine the relative effectiveness
of diffusion and participatory strategies in (health campaigns) and the effect
of caste on the dependent variables of knowledge, perception of risk, and
behavioral involvement that were conceptualized as contributing to leprosy
destigmatization in Madhya Pradesh state, India.  Multivariate analysis of
covariance (MANCOVA) procedure found significant difference between the
communication treatments on the dependent variables. The discriminant analysis
procedure was used to locate the source of difference. This procedure identified
two significant discriminant functions: cognitive-affective and
behavior-affective dimensions. The participatory treatment showed higher
knowledge and lower perception of risk on the cognitive-affective dimension, and
higher behavioral involvement on the behavior-affective dimension, but the
diffusion treatment showed only lower self-perception of risk on the
behavior-affective dimension. The study concluded that participatory strategies
promoting dialogue, interaction and incorporating people's knowledge and action
component result in increased knowledge, lower perception of risk, higher
behavioral involvement, and hence, destigmatization . COGNITIVE-AFFECTIVE AND
BEHAVIOR-AFFECTIVE DIMENSIONS:
A COMPARATIVE ANALYSIS OF PARTICIPATORY & DIFFUSION
APPROACHES IN THE DESTIGMATIZATION OF LEPROSY
A Case Study in Gwalior, India
 
        Stigma may be defined as negative attitudes and prejudice toward a person that
results in avoidance of social interaction. According to Goffman (1963) stigma
does not reside in the attribute but is ascribed by "normal" others to those
whose characteristics are seen as not fulfilling normative expectations. The
"undesired differentness" may sometimes be so strong as to preclude recognition
of any other qualities possessed by the person.
        The "undesired differentness" a person is perceived as possessing typically
relates to attributes of physical disfigurement, blemishes of individual
character and/or personality, and social categorization such as race, national
origin, and religious affiliation (Goffman, 1963). The first, the focus of this
study, includes physical handicaps and somatic conditions. Disfigurement-
related stigmas affecting social interaction were ordered along six dimensions
(Jones, et al., 1984). The first dimension of concealability refers to the
hidden or obvious condition of stigma. Visible conditions attract high stigma
and impede social interaction whereas conditions low on visibility permit the
bearer of stigma to interact with greater ease. The second dimension on which
stigma differs is its level of disruptiveness (or obtrusiveness). Disruptiveness
can be thought of as characteristics that make social interaction uncertain,
unpredictable or awkward for the participants. The third dimension of aesthetic
quality of stigma refers to how physically unattractive the stigma may be.
Physical disfigurements that appear repellent, ugly or upsetting evoke more
stigma. The fourth dimension is condition's origin. The extent of interaction
depends on how responsible a person is held to be for the stigmatizing
condition. The fifth dimension is the course of stigma over time. Interpersonal
contact is less with those whose condition is perceived as irrecoverable. The
sixth and final dimension on which to locate any stigma is perceived peril, that
is, the extent to which others feel physically, psychologically or morally
threatened by the stigmatizing condition. Behavioral involvement is avoided with
those whose condition is perceived as risky for physical or social well being
(Goffman, 1963; Jones et al., 1984).
        The association between stigma, disease and social interaction was examined by
Skinsnes (1964), a leprologist, who constructed a hypothetical disease scenario
expressing the ultimate in physical disfigurement and negative social
consequences. Such a disease would be manifestly visible, progressively
deforming, chronic, and of unusually long duration; it would have an insidious
onset, long incubation period, high endemicity, and appear to be incurable.
Persons suffering from it would be neglected, avoided, segregated, and
ostracized.
        Leprosy's disease characteristics fit Skinsnes' description. It is described as
a chronic disease caused by a bacterium, Mycobacterium leprae, which is normally
contracted through the respiratory tract. The M. leprae have a long generation
time. The clinical evolution of the disease is slow, often extending into many
years (Browne, 1985). One of the two types, the multibacillary (MB), is
considered contagious. Although the 'oldest disease known to man,' leprosy had
no effective therapeutic treatment until recently. The disease develops
insidiously for years, infiltrating skin and nerve cells before causing
permanent damage to limbs and eyes. The disfigurement often associated with the
disease leads to stigmatization of the patient. Hirmani (1992) stated that
leprosy "is looked down upon not because it kills the patient or is highly
contagious but because it disfigures the body, thereby attracting social stigma"
(p. 181).
        In India, stigmatization of leprosy patient existed at the legal level until
recently. The Indian Christian Marriage Act (1872), the Muslim Marriage Act
(1939), and the Hindu Marriage Act (1956) granted divorce on grounds of leprosy.
The electoral laws disqualified patients from contesting elections. The Motor
Vehicles Act (1939) prohibited them from obtaining a driver's license though
only 25 per cent suffer from sensory loss of the limbs. Provisions in the legal
manuals of public transport companies imposed travel restrictions. Insurance
companies charged higher premium. And, the various local accommodation acts
authorized owners to evict  leprosy patients.  They continue to be quarantined
at home and, sometimes, evicted from village.  Incidents of leprosy patients'
dead bodies being buried rather than cremated, as is the Hindu custom, have also
been recorded (Patankar, 1992).
        Stigma has been examined in different settings using various methodologies.
Goffman (1959) dealt mainly with casual interaction between relatively unknown
people ("unfocused gatherings"). Gussow and Tracy (1968) examined more permanent
social settings where the stigmatized were in continuous interaction with the
'normal' others. The present study used camp settings, employed by two
communication strategies (i.e. diffusion and participatory approaches), to
determine behavioral interaction between leprosy patients and community members.
 
LEPROSY COMMUNICATION
        Two communication approaches have been employed in the Indian state of Madhya
Pradesh to destigmatize and eradicate leprosy. The first approach, employed by
the federal and state government, is based on the Survey-Education-Treatment
(SET) model. SET can be construed as a specific adaptation of the diffusion of
innovations approach that informs all health and agricultural programs
implemented by the government. The model clearly establishes association between
stigma and lack of knowledge of disease.
        In order to create knowledge and facilitate interaction, the SET model aims at:
     l. Developing knowledge on the nature of leprosy, its amenability to
     cure, recognition of early signs of the disease and prevention of
     deformities;
     2. Creating a reasoned and rational attitude towards patients, so
     that they are not ostracized from society, displaced from their homes,
     jobs, marriages, etc.; and
     3. Promoting social integration of the leprosy patients. (emphasis
     added; Director General of Health Services [DGHS], 1992; 1993).
On the other hand, the participatory communication model, as implemented by the
Danish Leprosy (DANLEP) unit-trained government health workers of two districts
in Madhya Pradesh, de-emphasizes the role of knowledge in the destigmatization
process and underlines the need to promote interaction between the community and
the patients (Patankar,1992).  This model incorporates a social-action based
participatory strategy wherein community members, patients, doctors and health
workers are brought together in mixed-group residential camps to share
experiences, discuss, build capabilities, and remove attendant stigma.
        The literature indicates a relationship between stigma and knowledge, attitude,
and social interaction. Valencia (1986) hypothesized that increased knowledge of
their disease would lead leprosy patients to adopt more emotionally neutral
attitudes toward leprosy. Investigating the compliance behavior of leprosy
patients, Hertroijs (1974) found that educated people were more sensitive to
stigma and, therefore, likely to discontinue treatment. Siller et al. (1967)
conducted extensive interviews with 65 people and found that their interaction
with people with physical disability was influenced by fears of ostracization
and self-affliction. In their study of leprosy patients in an Ethiopian
hospital, Giel and Luijk (1970) found that 47 per cent of their sample was
divorced because of the disease. Forty three percent said that people in their
village avoided them.
        In India, Kushwah et al. (1981) conducted a longitudinal study at a leprosy
clinic in Gwalior. Findings of the study corroborated Hertroijs assumption that
attenders at clinics tend to belong to low income, low education, and less
prestigious occupational categories. Kumar (1983) interviewed leprosy patients
and members of their community in Madras city in the state of Tamil Nadu in
India. The study found that most patients (82%) and nonpatients (80%) possess
only "fair" awareness of the early signs of leprosy and that large majorities of
both patients (75%) and non-patients (80%) seriously misunderstood the way in
which leprosy is contracted. Pal and Girdhar (1985) who conducted a study at the
Central Jalma Institution for Leprosy in Agra, India, found that the majority of
patients were uneducated (53%), from rural areas (63%), and only eight percent
knew the actual cause of their disease. Mutatkar and Tare (1988) surveyed 1200
respondents and 400 patients in four villages of the two Indian states of Orissa
and Andhra Pradesh (A.P.). They concluded that although "people have scientific
knowledge of leprosy it has very low impact on their having positive attitude
towards leprosy patients" (p. 4). Caste-based discrimination showed that
low-caste patients were discriminated against by greater number of family
members, relatives, and friends than high-caste patients in the four villages of
Orissa and  A. P. In a knowledge, perception, and behavior study conducted among
281 households in six villages in a south Indian state, Rao (1992) found that 89
percent respondents considered deformity as sign of leprosy. Skin patches were
not recognized as a sign of leprosy. Over 30 percent attributed the disease to
God's wish, fate, and sins committed by patients in previous life. Only three
percent attributed the disease to germ or an "organism."
 
HYPOTHESES
Communication Treatment Factor
        The government of the Indian state of Madhya Pradesh acknowledged the "great
success"1_ achieved by the participatory practices in eradicating leprosy in two
districts of the state. Patankar (1992) evaluated the success of the
participatory strategies. He found that the annual case detection rate of
leprosy in the Rajnandgaon district increased from 1.1 per 1,000 in 1987-1988
(the first year of introduction of multidrug therapy and participatory practices
in the county) to 1.3 in 1991. During the same period, the prevalence rate
dropped from 4.0 to 1.8. The deformity rate among new cases declined from 6.2 to
5.6 per 1,000. This data reflects two trends: Increase in the number of new
cases of leprosy (case detection) and regularity of treatment (case holding).
        Supportive action of the community led to the establishment of a "troika of
relationship" between the community, the patient, and the health worker.
Patankar (1992) concluded people's participation came about because:
     While retaining emphasis on case detection, the main concern was to
     address itself to social stigma. The programme addressed itself to
     people's concern such as ulcer cure and care and disability prevention.
     This resulted in people's positive response and openness. Leprosy workers
     experienced their movements in an altogether different world--situations
     as never before experienced in any health programme, let alone leprosy.
     (p. 2)
The federal health department, on the other hand, highlighted the success of its
diffusion strategy. Evaluation of its education activities after the
introduction of the multi-drug therapy (MDT) showed that over 70 % of the 425
respondents in a community were aware of the leprosy workers visiting the
villages. Over 50 % knew that leprosy was not hereditary. Sixty per cent
believed that leprosy was least contagious. More than 70 % knew that leprosy was
curable. Sixty seven per cent knew that a leprosy patient could be safely kept
at home. A related study of 466 patients showed that an overwhelming number
(over 90 %) lived with their families. Half of them were employed. The
proportion of self-reporting cases was found to be 48.2 % indicating fear
removal and reduced stigma attached to the disease. The 1992 status report of
the government concluded "It was obvious that the programme had percolated to
the people. The general attitude of the community towards the disease was
positive.... The general belief that the leprosy patients were shunned by one
and all was not borne out by facts" (DGHS, 1992, p. 48).
        The evaluation of participatory and diffusion strategies presented here
indicates that both  strategies identify stigma as an impediment in the
eradication of leprosy. It seems reasonable that as long as patients are
discriminated against, they would prefer hiding the disease. Both the strategies
assume that destigmatization would enable suspected cases to voluntarily report
for diagnosis and treatment. However, differences between the two strategies are
seen in the importance assigned to knowledge, attitude, and behavior. The
diffusion strategy underscores the importance of dispelling 'myths and
misconceptions,' and the need to create awareness and a 'reasoned and rational
attitude' to stop exclusion of the patients. One of its education objectives is
"disseminating correct information about the disease and, in doing so, helping
to remove the social stigma it carries by means of systematic health education
for the community" (DGHS, 1993, p. 2). The assumption clearly seems to be that
people lack proper understanding of the concept, cause, and cure of leprosy, and
diffusion of information will produce desirable attitudinal and behavioral
changes. While the diffusion model adopts a cognitive, knowledge-based route,
the participatory model seeks to follow a behavior-based approach. According to
Patankar (1992), "knowledge itself does not remove fear; it may even accentuate
it. The confidence about one's capability to prevent the progression of the
disease, and experience of cure, would alone ward off fear from the mind" (p.
6). Since both diffusion and participatory strategies share the common objective
of eradication of leprosy but differ in their method of destigmatization, the
question arises: Which of the two strategies currently being employed is better?
The first hypothesis of this study, therefore, stated:
      H 1     There will be a difference between communication treatments
     (participatory, diffusion, and control) on the dependent variables of
     knowledge, perception of risk, self-perception of risk, and behavioral
     involvement when pre-treatment differences are controlled.
Caste Factor
        Information campaigns including health education campaigns have been shown to
differentially impact social groups (Tichenor et al., 1970). Studies in India
have found better-off sections of the society benefiting more than those who
were socially or economically disadvantaged (Shingi and Mody, 1976). Caste has
been found to be an important determinant of access to education and health.
Corrie (1995) indicated that low caste female literacy rate in Madhya Pradesh
state is only 7%. Naik (1982) pointed out that of all the children who enter
schools, nearly half drop out by grade V, only about 15% reach grade XII, and
less than one per cent reach college. Most of the children who cannot continue
beyond grade V belong to low castes.
        Since most low castes live in villages, their access to modern health care
facilities is limited. Of the 450,000 doctors in India, only 21,000 work in the
countryside where about 70 % of the population lives. Only 27 % of the country's
hospitals are in rural areas (Kaul, 1993). In the Madhya Pradesh state, 74 % of
sub-health centers, 55 % of primary health centers, and 73% of community health
centers function without a building. There are no doctors in 375 of the 1841
primary health centers (Government of Madhya Pradesh, 1995).
        Since access to and availability of health and educational facilities is a
function of caste, it is assumed that leprosy-related education will have a
differential impact on low and high (who tend to study longer and have better
means of access) castes. According to Jayaraman (1981) "a convenient way of
describing such differences is to distinguish two categories: the superordinate
and subordinate" (p. 8). In this study, the categorization of caste into low and
high is based on Jayaraman's classification. (Low caste refers to the Scheduled
Castes or "those castes or parts of or groups within castes" as are declared to
be Scheduled Castes by the President of India by public notification under
Article 34 (1) of the Constitution of India. All other castes are grouped
together as the high caste.) The second hypothesis, therefore, stated that:
     H2 There will be a difference between the two caste levels on the
     dependent variables of knowledge, perception of risk, self-perception of
     risk, and behavioral involvement when pre-treatment differences are
     controlled.
METHOD
Location
        The data for the 3 X 2 factorial design experimental study were collected by
organizing health education camps in three nonrandomly selected villages of
Gwalior district in Madhya Pradesh state.   While the selection of the villages
was non-random, it was based on a set of demographic, geographic, educational,
and other relevant criteria that made them representative of other villages in
Gwalior district of Madhya Pradesh.  These selection factors included:
population size, number of households, presence of high and low castes in the
villages, distance from the highway, educational and health care facilities,
socio-cultural conditions, geographical contiguity, access to the mass media,
registered leprosy cases, and non-exposure to sustained and systematic leprosy
eradication campaigns.  The villages were randomly assigned to the difffusion,
participatory and control treatments.
Procedure
        A total of 269 adult male and female subjects belonging to low and high castes
were randomly chosen. The 3X2 completely crossed communication treatment by
caste factorial design featured three levels of communication treatment
(diffusion, participation, and control) and two levels of caste (low and high).
Table 1 shows the distribution of subjects.
 
(TABLE 1 ABOUT HERE)
        Ninety one subjects were imparted leprosy education over three days in the
diffusion treatment. The diffusion treatment was administered by two former
directors of the Gandhi Memorial Leprosy Foundation.   Eighy nine subjects
attended the participation treatment over a period of three days. The
participatory treatment was administered by two experienced health workers from
the Rajnandgaon district.   The strategy here was to generate a dialogue between
the participants, the patients and the health workers on issues of concept,
cause and cure of leprosy.  The purpose of the interactions was not necessarily
for achieving closure on issues but to create a space for contest and
convergence of different points of view and experience.  The Hydro-Oleo-Therapy
(HOT)  was undertaken three times a day.  This is an inexpensive water and oil
massage of ulcers and wounds of the patients by the health worker in the
presence of other participants.  The act of the health worker scrubbing and
applying medication on ulcers was supposed to raise questions on infectivity and
patient-avoidance behaviors.  These were followed by live case demonstrations
where patients displayed their wounds and skin patches and narrated their trials
and tribulations not only in dealing with leprosy but also its social
consequences. About 30 subjects attended the diffusion and participatory
villages respectively each day. Thus, at the end of the third day about 90
subjects had received the experimental treatment in diffusion and participatory
villages.   In the control group, 89 subjects were interviewed during a two-day
skin diseases camp. The skin disease camp was conducted by the local leprosy
staff. About 45 subjects attended the camps each day in the control village.
        The questionnaire was developed in consultation with two medical doctors
working at the Central Jalma Leprosy Institute, Agra. It was pretested in seven
villages of Gwalior district. The subjects at the three camps were pre-tested
and post-tested each day. The questionnaires for the pretest and posttests were
identical except for two additional items in the posttest. A team of six trained
interviewers administered the questionnaires. Six leprosy patients were present
at the three camps.
 
Measures
        The independent variables were a) the communication treatment with three
levels: diffusion, participatory, and no treatment, and b) two caste levels (low
and high).  The dependent variables included in the analysis were summative
indices of knowledge of cause, knowledge of spread, perception of risk, and
behavioral involvement. The posttest also included a single self-perception of
risk item.
Knowledge
        Knowledge of cause was measured with items such as: leprosy is caused by sins
committed in past life/god's wrath/heredity/fate/wrong food/ and germ. The
six-item knowledge of cause scale was based on misconceptions widely shared
among rural people (DGHS, 1993; Patankar, 1992).  Knowledge of spread scale
consisted of three questions that probed knowledge of the route of transmission:
Do all patients spread leprosy? Does touching the wounds on hand and feet of
patients spread leprosy? Does leprosy spread by coughing and sneezing? The two
knowledge scales used no (0) and yes ( l ) responses to questions. The knowledge
scales were constructed separately for the pretest and posttest data using raw
scores. (alpha reliability scores for knowledge of cause were .59 for pretest
scale and .60 for posttest scale; alpha for pretest knowledge of spread scale
was .70 and posttest scale was .73)
Perception of Risk
        The perception of risk scale measured the probability of contracting leprosy in
the event of establishing varying degrees of physical contact with the patient.
Subjects reported their risk for contracting leprosy on a 10-item scale
developed for this study. Perceived risk related to the estimation of
contracting leprosy for an individual when she/he sits next to, talks, sleeps,
drinks,  eats with persons with leprosy or when the patient lives in the same
village. The posttest included an additional item that assessed respondent's own
estimation of contracting leprosy after having spent a day with patients in the
health camp. Perceived risk items were measured on a five point scale with (1)
denoting 100%, (2) denoting 75%, (3) denoting 50%, (4) denoting 25% and (5)
denoting zero per cent  (alpha reliability score for pretest scale was .81 and
for the posttest scale it was .79).
 Behavioral Involvement
        Behavioral involvement was assessed with items about actual behaviors such as
talking with, sitting next to, touching the hands and feet of, and drinking
water served by persons with leprosy. The posttest included an additional item
concerning inspection of skin patches of other people in the camp. Behavioral
involvement items were measured on a scale ranging from none at all (0), once
(1), few times (2) and many times (3). (alpha reliability score for pretest
scale was .86 and for the posttest scale it was .89).
 
RESULTS
        The analysis is based on responses of 262 cases. Since this study used multiple
dependent variables that were significantly correlated with each other at
pretest (p = .0001) and posttest (p = .0001), the data were analyzed using the
MANOVA procedure.
Pretest MANOVA
        Significant interaction effect between communication treatment and caste was
found on the dependent variables of knowledge of cause, knowledge of spread,
perception of risk, and behavioral involvement (Pillai's test = .08; F = 2.62;
p< .01; hypoth. DF =8.0; Error DF = 508).  Further analysis of main effects,
therefore, was not done.
        To achieve pre-treatment equivalence, the pretest dependent variables were used
as covariates to test the hypotheses.  The dependent variables for the posttest
MANCOVA were knowledge of cause, knowledge of spread, perception of risk,
self-perception of risk, and behavioral involvement.
Posttest MANCOVA
Hypothesis 1
        The hypothesis of significant communication treatment differences was
supported. (Pillai's test = .60; F = 21.17; Hypoth DF = 10.0; Error DF = 492; p
=.0001).  Therefore, the combined means on the scales of knowledge of cause,
knowledge of spread, perception of risk, self-perception of risk, and behavioral
involvement were significantly different for the diffusion, participatory, and
control treatments when pretest MANOVA differences were controlled.
 
 
Hypothesis 2
        The hypothesis of significant differences between low and high castes on the
specified dependent variables of knowledge, perception of risk and behavior
involvement  was not supported. (Pillai's test = ..02; F = 1.12; Hypoth DF =
5.0; Error DF = 245.0; p =.35).
        Also, no significant interaction effect between the two independent variables
was found at the posttest.
 
Discriminant Analysis
        Given the correlation between the multiple dependent variables, step-down
F-tests were not conducted to locate the source of difference for the
communication treatments..  A multivariate post-hoc test, the direct
discriminant function analysis, was performed using the five discriminant
(dependent) variables to locate and explain differences between the three
communication treatment groups. Communication treatment groups were diffusion,
participatory, and control villages. The discriminant variables were knowledge
of cause, knowledge of spread, perception of risk, self-perception of risk, and
behavioral involvement.
 
(TABLE 2 ABOUT HERE)
        Two significant discriminant functions were obtained as shown in Tables 2 & 3.
The first significant discriminant function (labeled cognitive-affective
dimension) explained 73% of the between-group variability. After removal of the
first function, there was still strong association between groups and dependent
variables (p <.0001). The second discriminant function (labeled
behavior-affective dimension) accounted for 27% of between-group variability.
Cognitive-Affective Dimension (Discriminant Function 1)
        The loading matrix of correlations between discriminant variables and the first
discriminant function, as seen in Table 2, suggests that the best variables for
distinguishing between the diffusion, participatory and control groups on the
first function are knowledge of cause and perception of risk (loadings less than
.50 are not interpreted). This discriminant function could be termed the
cognitive-affective dimension of destigmatization.
 
        When the group centroids were plotted on the significant discriminant function,
data in Table 2 indicate that the participatory group had higher knowledge of
cause and lower perception of risk than diffusion group and control groups [i.e.
positive group means on this function would reflect high knowledge and low
(i.e., positive) perception of risk]2.
(TABLE 3 ABOUT HERE)
 
Behavior-Affective Dimension (Discriminant Function 2)
        The loading matrix of correlations between discriminant variables and the
second discriminant function, as seen in Table 3, suggests that the best
variables for distinguishing between the diffusion, participatory and control
groups on the second function are behavior involvement and self-perception of
risk (loadings less than .50 are not interpreted). This discriminant function
could be termed the behavior-affective dimension of destigmatization.
        When the group centroids were plotted on the significant discriminant function,
data in Table 3 indicate that the diffusion group has lower self-perception of
risk but participatory and control groups had higher behavioral involvement
[i.e.negative group means on this function reflect lower self-perception of risk
and positive means show higher behavioral involvement]3.
        In conclusion, this study found that on the first dimension extracted by the
discriminant analysis, the participatory strategy adopted for destigmatization
of leprosy was more effective than the diffusion strategy. That is, the
participatory group showed lower perception of risk and higher knowledge of
cause. At the same time, and independent of the first, on the second
multivariate dimension, the diffusion group showed lower self-perception of
risk, but the participatory group showed higher behavioral involvement.
        This finding has important implications for communication programs and
campaigns that problematize stigma as a critical factor in health contexts. A
natural corollary of the finding is that there are two alternate routes to
destigmatization: the cognitive-affective and the behavior-affective. In the
cognitive-affective route, corresponding to the first dimension, the predominant
elements of persuasion are knowledge and perception. The high correlation of
knowledge underscores the significance of information (see Table 2). The
contribution of  behavior to destigmatization (loading of .44), though
important, is relatively small. In the behavior-affective dimension, or the
second dimension, the importance of behavioral involvement and self-perception
of risk is significantly large and occurs despite the absence of knowledge. It
can, therefore, be hypothesized that the participatory treatment (vis-a-vis the
diffusion treatment) has within its strategy elements that reduce stigma by
engaging subjects at a cognitive and affective plane (i.e. participatory group
had higher knowledge of cause and lower perception of risk on the first
dimension). The results also suggested that the participatory treatment relative
to the diffusion treatment has a strong behavioral component (i.e. participatory
group had higher behavioral involvement on the second dimension) that brings
about destigmatization on its own (i.e., devoid of knowledge).
DISCUSSION
        The significance of the findings of the study are explained by locating the
diffusion and participatory strategies in the information-education and
communication-action perspectives respectively.
Information-Education Perspective
        It is argued here that the limited efficacy of the diffusion model in
generating greater behavioral involvement can be better appreciated by situating
it in the information-education perspective. This perspective is predominantly
informed by clinical construction of disease. It derives its persuasive thrust
from the message-based theories that postulate that dissemination of medical
information will bring about desirable change (Devine and Hirt, 1992). Anchored
in the presumption that people either do not know, or not know enough, or know
incorrectly, the diffusion model strategizes information with the explicit
purpose of 'exposing,' 'creating awareness,' and 'imparting knowledge.' Emphasis
is placed on media selection. Information campaigns using multiple channels are
appropriately designed to catalyze the diffusion of clinical knowledge. Backer
et al. (1992) state: "Frequently, mass media campaigns concerning health issues
involve communicating the findings from biomedical research almost as soon as
they are generated" (p. 6).
        The diffusion strategy of leprosy eradication in India subscribes to many of
the ideas embedded in the information-education perspective. The emphasis on
knowledge is clearly seen in the objectives of the program. However, it is
noteworthy that diffusion strategy that emphasizes dissemination of clinical
information showed minimal correlation with knowledge on the first discriminant
function (see Table 2). Several studies have indicated a positive relation
between increased knowledge and behavioral involvement (Montgomery, 1987;
Valente et al., 1996). Therefore, it is likely that failure to produce high
knowledge may have contributed to the lack of behavioral involvement on the part
of the subjects exposed to the diffusion treatment. If this interpretation were
true, it suggests that diffusion campaigns based on information-education
perspective must pay special attention to the knowledge component of the
strategy. This task is particularly difficult with the changing nature of
knowledge itself. Leprosy was once regarded as hereditary among medical experts.
Later, it was widely believed to be transmitted by contact between patients and
non-patients. More recent research posits leprosy as spreading through the
respiratory route. The best summation is, perhaps, offered by the Chief of the
Leprosy Division of WHO: "The exact mechanism of transmission of leprosy is not
known" (Noordeen, 1985, p. 25).
        The changing nature of knowledge about leprosy is one part of the issue. The
other relates to the discrepancy between experts' representation and people's
construction of leprosy. Table 4 presents the differences in perception about
leprosy between the medical/ scientific system and the lay referral system of
the Indian village communities.
(TABLE 4 ABOUT HERE)
 
The participatory group by contextualizing information/knowledge in the personal
and social experiences of the community may have been successful in generating
higher knowledge scores vis-a-vis the diffusion group.
Communication-Action Perspective
        Why did the participatory treatment group show higher behavioral involvement on
the second multivariate dimension of destigmatization when self-perception of
risk for contracting leprosy was relatively high?  Unlike the medical-scientific
orientation of the diffusion strategy, the participatory model as employed in
the Indian state of Madhya Pradesh directly addressed people's fear of
contracting leprosy by its dialogic and action components. The two components
take as their premise the position that because the community stigmatizes the
patient it is the community that first needs to be released from fear of ulcers
and deformity before the patient can be influenced to seek treatment (Patankar,
1992). The strategy's dialogic component draws sharp distinctions between
experts' scientific understanding and people's cultural interpretation of the
disease. Thus, it recognizes different constructions of leprosy and creates
space for contest and convergence between them. The immediate outcome of such a
strategy is dialogue: collective and serious engagement of the group in issues
of concept, cause, and cure of leprosy. The health workers abet the
participatory process by refraining from directing attention to themselves:
questions are deflected to the group, answers are avoided, clarifications are
withheld, and conclusions are discouraged; instead questions are raised to keep
the dialogue going. Their main arguments tend to be based not on logic and
reason, but analogy. The purpose of dialogue is clearly three-fold: to remove
the health worker from occupying a position of centrality (or authority), to
energize the group to collectively analyze multiple conceptualizations of
leprosy by a process of cognitive agitation or, as Patankar states, "education
by confusion management" (personal discussion), and to create redundancies in
the communication situation so that an open and free environment built by this
process reinforces the action component.
        The action component of the participatory strategy mainly rests on
HydroOleoTherapy (HOT)--an inexpensive water-and-oil massage of ulcers and
wounds patients undertake thrice a day. The manifest purpose of HOT is to let
the group see for itself that ulcers, perceived to be source of infection,
gradually heal. But the latent function of the therapy is more powerful. The act
of the health worker scrubbing and applying medication on ulcers raises doubts
and questions of infectivity and patient-avoidance behavior.
        These doubts and questions are explained by an additional component of the
action strategy: live case demonstration. Unlike the diffusion approach that
uses video film and slides, the participatory method strategizes its main
arguments and attention around patients. Patients display their wounds and skin
patches and narrate their trial and tribulation not only in dealing with the
disease but also its social consequences. The sharing of life stories and
personal experiences often creates strong emotional bonds between the community
and the patients.
        The effectiveness of the participatory strategy in generating higher behavior
involvement is additionally explained within the framework of the contact
hypothesis (Allport, 1954). The hypothesis suggests that health behavioral
changes occur due to experiential learning. In contrast to modeling which is
essentially learning by imitation, contact can be hypothesized as learning by
direct personal interaction with stigmatized categories of people (Miller and
Brewer, 1984). Contact is not merely a structural property of communication as
having the health worker, community members, and the patients together in the
camp. Rather, it is the open, equal, and transactional quality of interaction
between the three actors that is the defining characteristic of contact. In the
case of the diffusion treatment, the monologic nature of communication in which
the health workers delivered what they believed to be the correct knowledge of
leprosy precluded the real possibility of dynamic three-way learning
interaction. An obvious implication for health communication campaigns is
deemphasizing content and contextualizing information in the personal and social
experiences of the group.
 
IMPLICATIONS FOR ACTION
        This study points to the possibility of reducing stigma using communication
campaigns.  It further suggests that strategies that address local cultural
ideologies, engage patients, community and health worker in triadic dialogue,
and incorporate action component have greater potential of reducing stigma.
        Destigmatization efforts can simultaneously occur at multiple levels and
proceed along different directions.  An important public health policy direction
would entail merger of the vertical national leprosy eradication program (NLEP)
with the general primary health care system.  Over the years, the Indian program
for leprosy control has grown into "the largest of its kind in the world" (DGHS,
1992, foreword). But, as Spicker (1984) argues, exclusive and specialized
programs targeting specific groups only perpetuate stigma.  Some scholars have,
therefore, called for integration of the eradication program into the larger
public health framework.  Antia (1988) recommends that "leprosy medicine and
leprosy surgery must be integrated into the general medicine and general
surgery.....so that special centres which propagate the stigma of the disease
are gradually phased out" (p.55).
        A second related public policy measure would focus on resource re-allocation:
gradually scaling down assistance to voluntary efforts that permanently
segregate patients and increasing support for initiatives such as:
community-based preventive, promotive, and rehabilitative efforts to educate
members about early signs, early detection, and early treatment of cases;
community participation in delivery and referral systems; training of members of
the newly-formed village health committees and the village health worker; and
public efforts that offer hitherto neglected services like reconstructive
surgery, and rehabilitation of patients by training them in income-generating
professions.  After all, an economically independent and self-reliant person can
be the best spokesperson for desegregation.
NOTES
1. In a letter written to members of the research team, the Deputy Secretary
(Public Health and Family Planning) to the government of Madhya Pradesh said "A
project on eradication of leprosy being sponsored by the Danish government is in
operation in the districts of Rajnandgaon and Durg of Madhya Pradesh. It has
been a great success in the districts of Rajnandgaon and Durg whereas it has not
been so in some other districts where the project is also in operation."
2. Higher scores on perception of risk item actually denote lower perception of
risk because higher numerical codes were used to denote lower perception of risk
(see perception of risk scale in the "Measures" subsection)
3.  As the participatory group means were located on the positive end of the
dimension, this group had higher behavioral involvement scores vis-a-vis the
diffusion group.  Also,  as the diffusion group mean was located on the negative
end of the dimension, the diffusion group had higher self perception of risk
score.  However, in the coding scheme adopted for measuring self perception of
risk, higher scores denoted lower self perception of risk (see self perception
of risk in the "Measures" subsection).  Hence, diffusion group had lower self
perception of risk
 REFERENCES
 
Allport, G. W. (1954). The nature of prejudice. Cambridge, MA: Addison- Wesley.
Antia, N. H. (1988). Role of coercive surgery in leprosy rehabilitation. In
        Department of Health
     and Family Welfare (Ed.), Leprosy eradication: An experiment in
     community action       Bhopal: Government of Madhya Pradesh.
Backer, T. E., Rogers, E. M., & Sopory, P. (1992). Designing health
communication campaigns:        What works? Newbury Park, CA: Sage.
Browne, S.G. (1985).  The history of leprosy.  In R.C. Hastings (Ed.), Leprosy,
(pp. 1-13).     London: Churchill Livingstone.
Corrie, B. P. (1995). A human development index for the dalit child in India.
Social Indicators       Research, 34, 395-409.
Devine, P. G., & Hirt, E. R. (1992). Message strategies for information
campaigns: A social     psychological analysis. In C. T. Salmon (Ed.), Information
        campaigns: Balancing social     values and social change (pp. 229-258).         Newbury
Park, CA: Sage.
Directorate General of Health Services. (1992). National leprosy eradication
programme: Status       report. New Delhi: Ministry of Health and Family Welfare.
Directorate General of Health Services. (1993). National leprosy eradication
programme in
     India: Guidelines for multidrug treatment in endemic districts. New
     Delhi: Ministry of Health and Family Welfare.
Giel, R., & Luijk, N. (1970). Leprosy in Ethiopian society. International
Journal of Leprosy, 38,         181-197.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, New
York:   Doubleday Anchor Books.
Goffman, E. (1963). Stigma: Notes on the management of spoiled identity.
Englewood Cliffs, NJ:   Prentice-Hall.
Government of Madhya Pradesh. (1995). The Madhya Pradesh human development
report.         Bhopal: Project Office, Directorate of Institutional Finance.
Gussow, Z., & Tracy, G. S. (1968). Status, ideology, and adaptation to
stigmatized illness: A  study of leprosy. Human Organization, 27(4), 316-325.
Hertroijs, R. (1974). A study of some factors affecting the attendance of the
patients in a leprosy   control scheme. International Journal of Leprosy, 42,
419-427.
Hirmani, A. B. (1992). Health education for early detection of "leprosy". In
Leprosy: A      reference guide. New Delhi.
Jayaraman, R. (1981). Caste and class: Dynamics of inequality in Indian society.
        Delhi:  Hindustan.
Jones, E. E., Farina, A., Hastorf, A. H., Markus, H., Miller, D. T., & Scott, R.
A.      (1984). Social  stigma: The psychology of the marked relationships. New
York: W. H. Freeman.
Kaul, R. (1993).  Caste, class and education: politics of the capitation fee
phenomenon in   Karnataka.  New Delhi: Sage.
Kumar, A. (1983). Community awareness about leprosy and participation in
national leprosy        programme. Leprosy in India, 55, 701-711.
Kushwah, S. S., Govila, A. K., Upadhyay, S., & Kushwah, J. (1981). A study of
social stigma   among leprosy patients attending leprosy clinic in Gwalior.
        Leprosy in India, 53, 221-      225.
Miller, N., & Brewer, M. B. (1984). Groups in contact: The psychology of
desegregation. New      York: Academic Press.
Montgomery, S. B. (1987). Behavioral change in a cohort of homosexual men at
        risk for AIDS.  Unpublished doctoral dissertation, University of Michigan, Ann
Arbor.
Mutatkar, R. K., & Tare, S. P. (1988). Evaluation of impact of health education
in SET pattern.         Centre for Social Science Research on Leprosy. Wardha, India:
Gandhi Memorial Leprosy         Foundation.
Naik, J. P. (1982). The education commission and after. New Delhi: Allied
Publishers.
Noordeen, S. K. (1985). Elimination of leprosy as a public health problem.
        Leprosy Review, 63      (1), 1-4.
Pal, S., & Giridhar, B. K. (1985). A study of knowledge of disease among leprosy
        patients and    attitudes of community towards them. Indian Journal of Leprosy,
57, 620-623.
Patankar, P. (1992). Innovative approaches to NLEP in Madhya Pradesh:
Case-Studies-
     Rajnandgaon and Durg; January 1987 to July 1992. Report submitted to
     the Department of Health, Government of Madhya Pradesh, India.
Rao, V. (1992). Leprosy in rural India. Delhi: Manak Publications.
Shinghi, P., & Modi, B. (1976). The communication effects gap: A field
experiment on television        and agricultural ignorance in India. Communication
Research, 3, 171-190.
Siller, J., Chipman, A., Ferguson, L. T., & Vann, D. H. (1967). Attitudes of the
nondisabled     toward the physically disabled. New York: New York University
School of Education.
Skinsnes, O. K. (1964). Leprosy rationale. Military Medicine, 130 (9), 927-929.
Spicker, P.  (1984). Stigma and social welfare. New York: St. Martin's Press.
Tichenor, P. J., Donohue, G. A., & Olien, C. N. (1970). Mass media flow and
differential growth     in knowledge. Public Opinion Quarterly, 34, 159-170.
Valencia, L. B. (1986). Psycho-social and environmental factors in triad model
for     the     management of control program: Focus on leprosy. The Southeast  Asian
Journal of      Tropical Medicine and Public Health, 17, 442-451.
Valente, T. W., Saba, W. P., Merritt, A. P., Perez, A., & Beltran, L. R. (1996).
The Bolivia     national reproductive health promotion campaign. Paper presented at
the annual      convention of the International Communication Association, Chicago.
  TABLE 1
Communication treatment and caste-wise distribution of subjects
 
Treatment
__________________________________________________________________
Caste           Diffusion       Participatory   Control Total
__________________________________________________________________
Low             40              38              44              122
High            51              51              45              147
Total           91              89              89              269
__________________________________________________________________
 TABLE 2
Cognitive-Affective Discriminant Function
_________________________________________________________________________
Significant Discriminant Function Extracted:
Knowledge of Cause & Perception of Risk (Cognitive-Affective)
(Wilks' Lambda = .43; p< .0001; eigen value= .8; canonical correlation = .67:
variance explained = 73.12%)
 
Pooled within-groups correlation between discriminant variables and discriminant
function
                                        Cognitive Affective
                                        Discriminant Function
Perception of risk                      0.767
Knowledge of cause                      0.680
Self-perception of risk                 0.446
Behavioral involvement          0.438
Knowledge of spread                     0.380
 
Group Centroids (means) on the significant discriminant function:
Participatory Group             1.14
Diffusion Group         -.14
Control Group                   -1.04
 
______________________________________________________________________________
 
 TABLE 3
Behavior-Affective Discriminant Function
_________________________________________________________________________
Significant Discriminant Function Extracted:
Behavioral Involvement (+) Vs. Self-Perception of Risk (-)
(Wilks' Lambda = .77; p< .0001; eigen value= .3; canonical correlation = .48;
variance explained = 26.88%)
 
Pooled within-groups correlation between discriminant variables and discriminant
function
                                        Behavior-Affective
                                        Discriminant Function
Perception of risk                      -0.335
Knowledge of cause                      -0.113
Self-perception of risk                 -0.597
Behavioral involvement          +0.761
Knowledge of spread                     -0.147
 
Group Centroids (means) on the significant discriminant function:
Participatory Group             0.31
Diffusion Group         -.76
Control Group                   0.46
 
______________________________________________________________________________
 TABLE 4
. Differences in perception of leprosy between medical system
and lay referral system
__________________________________________________________________
 Variable                       Medical                 Lay referral
                                system                          system
__________________________________________________________________
Concept of                      Disease                 Curse/sin/disease
Leprosy
 
Cause                           Bacteria                        Mysterious
 
Contagiousness/         Paucibacillary: No              Highly contagious
spread                          Multibacillary: Yes
 
Early signs                     Anesthetic skin         Diseases associated
                                patches; swollen                with pain. As skin
                                nerves                          patches do not cause
                                                                any suffering, early
                                                                signs disregarded
 
Ulcers and                      Are a consequence of            Are the cause of
deformity                       neglect; avoidable              transmission of                                                         by early
detection.              disease. Not                                                                                            avoidable.
 
Concept of                      Both kinds of leprosy           Is not curable.
cure                            curable.
 
Method of                       Medication,                     Segregation.
cure                            Hospitalization,
                                Rehabilitation.
 
Definition of                   Cessation of bacterial          Regain lost parts
cure                            activity ('bacterial kill').    of body.
__________________________________________________________________