Stigma and Discrimination in Health-Care Provision to

Drug Users: The Role of Values, Affect, and

Deservingness Judgments

NATALIE SKINNER N. T. FEATHER1

National Centre for Education and School of Psychology
Training on Addiction Flinders University
and School of Psychology Adelaide, Australia
Flinders University
Adelaide, Australia

TOBY FREEMAN ANN ROCHE

National Centre for Education and National Centre for Education and
Training on Addiction and Training on Addiction
School of Psychology Flinders University
Flinders University Adelaide, Australia
Adelaide, Australia
This study examined the role of values, affect, and deservingness judgments in
health professionals’ views of patients with stigmatized conditions (e.g., drug de-
pendence). Participants were 277 nurses who responded to a survey containing 2
scenarios of a nurse providing high- or low-quality care to a patient with a con-
dition related to prolonged use of alcohol or heroin. Affective responses to the
patient were more positive for nurses with higher self-transcendence values, and
more negative for nurses with higher conservation values. Deservingness judgments
were predicted by positive and negative affect toward the patient, but not by at-
tributions of responsibility for drug use. Deservingness judgments emerged as
strong predictors of nurses’ satisfaction with the provision of high- or low-quality
care. The findings imply that the deservingness judgments made by nurses reflected
strong entitlement norms concerning the provision of proper care for patients that
were independent of patients’ perceived responsibility for their condition.

There is evidence that some health professionals have negative views of

individuals with stigmatized conditions (e.g., drug dependence, Hepatitis C,
HIV/AIDS) and are reluctant to provide high-quality care to these pa-
tients (Abed & Neira-Munoz, 1990; Abouyanni et al., 2000; Farmer &

1
Correspondence should be addressed to N. T. Feather, School of Psychology, Flinders
University, GPO Box 2100, Adelaide, Australia, 5001. E-mail: [email protected]

163

Journal of Applied Social Psychology, 2007, 37, 1, pp. 163–186.

r 2007 Copyright the Authors
Journal compilation r 2007 Blackwell Publishing, Inc.
164 SKINNER ET AL.

Greenwood, 2001; Karam-Hage, Nerenberg, & Brower, 2001; Melby, Bo-

ore, & Murray, 1992; National Centre in HIV Social Research, 2001; Roche
& Richard, 1991). The present study focuses on health professionals’ re-
sponses to the stigma of problematic alcohol use and drug dependence.
Surveys of health professionals indicate that a significant proportion hold
negative or stereotypical views of individuals with drug dependence that are
likely to compromise the provision of high-quality care. For example, Abed
and Neira-Munoz (1990) found that a small but significant proportion (10%)
of general practitioners agreed with the statement ‘‘Drug addicts deserve
whatever misfortune befalls them.’’ A number of studies with nurses have
found that negative and punitive attitudes toward drug users are relatively
common (Aalto, Pekuri, & Seppa, 2001; Melby et al., 1992; Norman, 2001).
Not only are these attitudes contrary to our expectations concerning profes-
sional ethics in the health sector, but the perception that some health pro-
fessionals are judgmental, unsympathetic, or hostile may discourage
individuals with drug-related problems from accessing health-care services
(McLaughlin, McKenna, & Leslie, 2000; Telfer & Clulow, 1990). Therefore, a
research priority is to identify the antecedents of some health professionals’
negative and discriminatory views of patients with stigmatized conditions,
such as problematic alcohol use and drug dependence.
Providing high-quality medical care is, in essence, a helping behavior.
The large research literature on the antecedents of helping behaviors sug-
gests that attributions of responsibility for a stigmatized condition play a
central role in people’s willingness to provide assistance and support (Menec
& Perry, 1998; Schmidt & Weiner, 1988; Schwarzer & Weiner, 1991; Weiner,
1988, 1995; Weiner, Perry, & Magnusson, 1988). However, health profes-
sionals routinely provide high-quality care and treatment to individuals who
hold a significant degree of personal responsibility for their health condi-
tions (e.g., heart disease, obesity). Therefore, responsibility attributions
alone may not be sufficient to account for some health professionals’ re-
luctance to provide high-quality care to individuals with stigmatized con-
ditions. Indeed, in the context of a hospital, they may play a relatively minor
role (if any), given that the normative context prescribes that patients are
entitled to proper care.
An alternative perspective is to focus on the social-justice implications of
providing care to individuals with stigmatized conditions (e.g., drug de-
pendence). Provision of medical care to individuals with drug-related prob-
lems represents a dilemma of social justice (equitable access to high-quality
care) and distributive justice (high-quality care is a scarce resource). There-
fore, the extent to which a poor standard of care is perceived to be just and
deserved is likely to be a key cognition influencing health professionals’
acceptance of this outcome. The current study investigates the role of
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 165

deservingness judgments in health professionals’ satisfaction with both high

and low standards of care for patients with drug dependence.

Deservingness Judgments

Judgments of deservingness relate to the justice or fairness of an outcome:

A just and deserved outcome is likely to be viewed with satisfaction and
approval, whereas an unjust and undeserved outcome will be met with dis-
approval and displeasure (Feather, 1999b). There is now a large body of
research showing that deservingness judgments influence affective reactions
(e.g., resentment, sympathy, schadenfreude) to a range of outcomes, such as
the arrest and prosecution of environmental protestors (Feather, 2002a), the
resolution of industrial conflicts (Feather, 2002b), the imposition of manda-
tory sentences by a court (Feather & Souter, 2002), and high or low achieve-
ment in university exams (Feather & Nairn, 2005; Feather & Sherman, 2002).
There is also evidence that deservingness judgments are likely to influence
health professionals’ reactions to clients with a stigmatized health condition
(e.g., psychiatric illness). Feather and Johnstone (2001) presented nurses
with a hypothetical scenario in which a psychiatric patient with schizo-
phrenic or a personality-disorder patient became aggressive and attacked
hospital property. The patient was either assisted in a positive way (i.e.,
patient’s hand was bandaged and patient was listened to by the nurses) or
responded to in a negative way (i.e., patient was rebuked and sent away).
Feather and Johnstone found that stronger perceptions of the patient’s re-
sponsibility for the aggressive incident were associated with more anger and
less sympathy toward the patient. These affective reactions, in turn, pre-
dicted deservingness judgments: Perceived deservingness of the negative
treatment increased with higher anger and lower sympathy, whereas per-
ceived deservingness of help and assistance increased with higher sympathy
and lower anger. Stronger perceptions that help or punishment was deserved
were associated with higher positive affect in response to each outcome. It is
important to note, however, that consistent with health-care entitlement
norms, the help provided by the nurse was perceived as more deserved and
resulted in higher positive affect when compared to the situation in which
the patient was rebuked.

The Present Study

Building on Feather and Johnstone’s (2001) study, the current study

investigates nurses’ responses to two hypothetical scenarios in which either
high or low quality of care was provided to individuals who either used
166 SKINNER ET AL.

Responsibility
for
drug use

Affect toward the Deservingness of Satisfaction with

drug user high- /low-quality high- /low-quality
care care

Values

Figure 1. Proposed theoretical model.

heroin or consumed alcohol to excess. Heroin and alcohol were chosen for
this study because both drugs are known to be associated with significant
personal and social harms, and significant stigma is attached to heroin ad-
diction and heavy alcohol consumption. Therefore, we expect that issues
regarding access to and deservingness of medical care will be particularly
salient in this context.
The proposed theoretical model is presented in Figure 1 and will be dis-
cussed in further detail. As Figure 1 indicates, we expect that the patient’s
perceived deservingness of high- or low-quality care will be the most proximal
predictor of nurses’ satisfaction with each outcome. There are three possible
antecedents of deservingness judgments that are considered: affective reac-
tions to the drug user, attributions of responsibility for drug use, and values.

Affective Responses to the Drug User

Findings from previous research are consistent with the assumption from
deservingness theory that deservingness judgments about another person’s
positive or negative outcome are moderated by how much that person is
positively or negatively evaluated (e.g., liked or disliked) by the person making
the judgments (e.g., Feather, 1999a, 1999b; Feather & Nairn, 2005; Feather &
Sherman, 2002). Consistent with these findings, we expect that deservingness
of high-quality care (a positive outcome) will increase with stronger positive
affect and weaker negative affect toward the drug user. The opposite pattern is
expected for deservingness of low-quality care (a negative outcome).

Responsibility Attributions

In Feather and Johnstone’s (2001) study, deservingness judgments relat-

ing to the nurse’s positive or negative response to the patients’ aggressive
episode were related to how much the patient was deemed to be responsible
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 167

for the aggressive episode. Note that this study involved patients who were
expected to differ in their perceived responsibility for the aggressive incident,
and, indeed, the results showed that the patient with schizophrenia was
deemed to be less responsible when compared with the patient with per-
sonality disorder. In contrast to Feather and Johnstone’s study, the current
study involves patients who are similar in the sense that they are drug users
(alcohol, heroin), and responsibility judgments were targeted to their drug
use, rather than to a specific behavioral incident within the hospital.
It might be argued that the more these patients were perceived to be
responsible for their drug use, the more they would be perceived to deserve
low- rather than high-quality care, and the more angry and less sympathetic
the nurses would feel toward them. Consistent with this argument, Skitka
and Tetlock (1992, 1993) demonstrated that attributions of responsibility for
health care (e.g., AIDS patients, organ-donor recipients) resulted in less
generous allocation of (hypothetical) public resources. Research has also
shown that feelings of anger and lowered sympathy toward another person
are more evident when that person is perceived to be more responsible for a
negative event or outcome than when perceived responsibility is lower or
absent (e.g., Menec & Perry, 1998; Weiner, 1995; Weiner et al., 1988).
Complicating this argument as it relates to deservingness, however, is the
fact that people’s deservingness judgments are sometimes likely to reflect
beliefs about entitlement, especially in contexts in which entitlement norms
are salient. People sometimes confuse deservingness with entitlement in their
everyday language. Feather (1999b, 2002a, 2003) distinguished between de-
servingness and entitlement on conceptual grounds. He proposed that
deservingness refers to judgments about positive or negative outcomes that
are contingent on a person’s positive or negative behavior (e.g., a student
deserving a high grade on an exam because he or she studied hard). In
contrast, entitlement refers more to the external normative context of rules,
laws, or other quasilegal prescriptions (e.g., an employee becomes entitled to
a promotion on the basis of an affirmative-action policy). Keeping these
fairness principles apart has important implications. For example, in some
situations, deservingness and entitlement may be compatible; while in other
situations they may come into conflict (Feather, 2002a, 2003; Feather &
Johnstone, 2001; Feinberg, 1970).
We can add further to this analysis. While judgments of deservingness for
a contingent outcome presuppose personal causation for the action that
produced the outcome and some degree of perceived responsibility, there is
no reason to expect that a person’s perceived entitlement to an outcome
should depend in part on that person’s perceived responsibility. A judgment
that a patient who uses drugs is personally responsible for his or her
condition should not mean that he or she is not entitled to proper care, given
168 SKINNER ET AL.

the norms under which the hospital and its employees operate and the caring
roles that these employees are expected to perform. Note that norms of
entitlement had an important influence on nurses’ responses in Feather and
Johnstone’s (2001) study, and we expect them to do so in the current study
as well.
The implication of the preceding argument is that if the nurses responded
to questions about deservingness as if these questions concerned entitlement,
then their responses may be unrelated to the patient’s perceived responsi-
bility for drug abuse. Deservingness, then, would be equated with entitle-
ment, and all patients would be perceived to be entitled to good, quality
care, regardless of their perceived responsibility for their drug-related ail-
ments. Beliefs about entitlement would dictate the responses to be made,
especially when the context is one in which there are strong entitlement
norms about proper treatment directed toward the welfare of patients and
where, in contrast to Feather and Johnstone’s (2001) study, there is an
absence of a specific behavior (e.g., an aggressive episode) that would engage
deservingness judgments. Entitlement norms would dictate that nurses
should express sympathy (or positive affect), rather than anger (or negative
affect), toward patients who are under their care, regardless of patients’
perceived responsibility. Hence, the paths in Figure 1 between responsibility
and deservingness, and between responsibility and affect are presented with
considerable qualification, and, indeed, they may be absent.

Values

The focus so far has been on relations between responsibility, affect, and
deservingness. In the current study, we also are interested in the role of
values as possible antecedents of responsibility attributions and affective
responses to the drug user. Values reflect general beliefs that people hold
about desirable or undesirable ways of behaving (e.g., being honest, being
loving) and about general goals or end states of existence (e.g., equality,
freedom; Feather, 1999b).
Values are assumed to be central components of the self-concept and to
be important influences on a person’s attitudes and behaviors. They are used
as criteria to guide the way people construe situations in terms of what
objects and events, behaviors and goals, or actions and outcomes are per-
ceived as desirable or undesirable, or as possessing positive or negative
valence (Feather, 1975, 1994, 1995; Rokeach, 1973), influencing the way
actions and their outcomes are evaluated (Feather, 1995). For example,
people with strong right-wing authoritarian values are more likely to view
crimes as serious (depending also on the context and the perpetrator of the
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 169

offense; Altemeyer, 1988; Christie, 1993; Feather, 1996, 1998). In the current
study, we explore the role of the higher order conservation and self-
transcendence value dimensions described by Schwartz (1992, 1994). These
values were chosen because they were most likely to relate to the issues of
drug use and provision of medical care that are examined in the current
study.
Conservation values encompass values that emphasize the importance of
conformity, tradition, and security values for self, as opposed to values that
emphasize the importance of openness to change (e.g., stimulation, self-
direction values). We expect that an emphasis on conservation values will be
associated with more punitive judgments, and will be reflected in perceptions
that drug users are largely responsible for their drug use and also in lower
positive affect and stronger negative affect toward the drug user. Evidence
from studies of right-wing authoritarians is consistent with this assumption.
Right-wing authoritarians tend to have stronger conservation values, to be
more punitive in their judgments, and to be more negative in their social
attitudes in situations in which these values are threatened (Feather, 1996,
1998, 1999b).
Self-transcendence values encompass universalism values that empha-
size the importance of general social goals (e.g., equality, social justice for
self), and benevolence values that emphasize the importance of prosocial
values that are more interpersonal in nature (e.g., being forgiving, being
helpful). They are opposed to values that emphasize self-enhancement
(e.g., power, achievement values). We expect that an emphasis on self-
transcendence values will be associated with more compassionate and
prosocial judgments, and will be reflected in perceptions that drug users are
less responsible for their drug use and also in higher positive affect and
lower negative affect toward the drug user. Evidence from values research
is consistent with this prediction. For example, people with stronger uni-
versalism values tend to perceive a crime committed by a member of the
public as less serious, depending on the context and nature of the crime
(Feather, 1996, 1998).

Summary

This is the first study to examine the role of deservingness judgments in

health professionals’ provision of care to drug users. The study also builds
on previous research by investigating the effects of nurses’ values on their
reactions to drug users. Finally, the study provides an opportunity for de-
termining how far Feather and Johnstone’s (2001) findings with psychiatric
patients can be generalized to a new area (i.e., drug users).
170 SKINNER ET AL.

Method
Participants

In a single mailing, questionnaires were sent to a random sample of 800

nurses registered with the Nurses Registration Board in New South Wales,
Australia in December 2002. Participants were mailed a hard copy or com-
pleted an online version of the questionnaire posted on the university
Web site. A total of 277 surveys were returned (34.6% response rate). Of
these surveys, 141 concerned alcohol use and 136 concerned heroin use. For
the total sample (N 5 277), participants’ ages ranged from 22 to 75 years
(M 5 48.1, SD 5 9.3), and number of years working as a nurse ranged from
1 to 53 years (M 5 24.5, SD 5 10.7).

Procedure

A 2 (Quality of Care: high vs. low)  2 (Drug Type: alcohol vs. heroin)
mixed within–between subjects design was used. The study was introduced
as an exploration of how nurses view individuals who use licit or illicit drugs.
The questionnaires were completed under anonymous conditions.
Participants were provided with a hypothetical scenario of a drug user
presenting to an Emergency Department (ED) nurse with a serious medical
condition associated with one type of drug use. Participants were randomly
assigned to a scenario addressing problematic alcohol use or injection of
heroin. The scenarios contained identical wording, except for the specific
type of drug and associated medical condition (heroin and a skin abscess;
alcohol and a stomach ulcer). The medical conditions were chosen carefully
to ensure that they represented common and realistic conditions associated
with each drug type, and were of equivalent seriousness with regard to
treatment priorities in an ED setting. The heroin scenario was as follows:
On a relatively quiet Sunday night, MJ presents at an Emer-
gency Department with a large and painful abscess on their
arm. On close questioning, MJ reveals to the triage nurse that
the abscess has developed from regular injection of heroin over
the past 5 years.
The alcohol scenario was as follows:
On a relatively quiet Sunday night, MJ presents at an Emer-
gency Department with a strong stomach pain. On close ques-
tioning, MJ reveals to the triage nurse that they have previously
been diagnosed with a large stomach ulcer that has developed
from regular excessive drinking sessions over the past 5 years.
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 171

Participants then responded to items addressing perceptions of the patient’s

responsibility for the drug use, and positive and negative affect toward the
patient.
Participants then were presented with two scenarios describing the nurse
as providing high- or low-quality care. Quality of care was operationalized
in terms of the degree to which the nurse provided help and assistance to the
patient, and the timeliness of treatment for the medical condition. The high-
quality care scenario described the nurse providing information on treat-
ment and support services and arranging for treatment of the drug user’s
medical condition as soon as possible. The low-quality care scenario de-
scribed the nurse providing a reprimand and placing the drug user as ‘‘non-
urgent’’ on the priority list for treatment. Each scenario was followed by
scales addressing perceived deservingness of the quality of care provided,
and positive and negative affective responses to each outcome. The scenarios
were presented in counterbalanced order. Order of scenario presentation did
not significantly affect responses on these scales.

Measures Concerning Perceptions of the Drug User

After reading the first part of each scenario, participants were presented
with five questions concerning (a) perceived responsibility for drug use,
measured by the extent to which the patient was responsible for the drug use
(1 5 not at all responsible to 5 5 very responsible); and (b) positive (sym-
pathy, concern) and negative (anger, disappointment) affective responses to
the patient (e.g., 1 5 not at all angry to 5 5 very angry).
The affect items were chosen on the basis of previous research on af-
fective responses to stigmatized others (Feather & Johnstone, 2001; Menec
& Perry, 1998; Schmidt & Weiner, 1988). A principal-components factor
analysis with varimax rotation was conducted on the four scales. There were
two factors that emerged with eigenvalues greater than 1. Anger and dis-
appointment items loaded on a single factor with factor loadings of .82 and
.87, respectively. Scores on the two items were averaged to produce a mea-
sure of negative affect toward the drug user. Sympathy and concern items
loaded on a second factor with factor loadings of .83 and .85, respectively.
Scores on these two items were averaged to produce a measure of positive
affect toward the drug user.

Measures Concerning the Quality of Care Provided to the Drug User

Following each scenario describing high- or low-quality care, partici-

pants were presented four questions concerning (a) the extent to which the
172 SKINNER ET AL.

patient deserved the high- or low-quality care (1 5 not at all deserving to

5 5 very deserving); and (b) satisfaction with the quality of care provided
(pleased, satisfied, disappointed, annoyed; e.g., 1 5 not at all pleased to
5 5 very pleased).
The selection of affect items was based on previous research on affective
responses to outcomes experienced by others (Feather, 1996; Feather &
Deverson, 2000; Feather & Johnstone, 2001). Separate principal-compo-
nents factor analyses with varimax rotation were conducted on the four
scales for the high- and low-quality care scenarios. For each scenario, one
factor emerged with an eigenvalue greater than 1. Factor loadings for the
disappointment, annoyance, pleased, and satisfied items were - .78, - .79,
.78, and .83, respectively, for the high-quality care scenario; and - .80, - .77,
.82, and .88 for the low-quality care scenario. Scores on the disappointment
and annoyance items were reverse-coded, and scores on the four affect items
then were averaged to obtain an overall measure of satisfaction with the
quality of care provided.

Internal Reliabilities

Internal reliabilities, as measured by Cronbach’s alpha, were acceptable

for items related to negative (a 5 .60) and positive (a 5 .60) affective re-
sponses to the drug user. They were good for satisfaction with high-quality
care (a 5 .77) and with low-quality care (a 5 .82).

Values Measure

Values were measured using the Schwartz Value Survey (SVS; Schwartz,
1992). Participants indicated the extent to which 57 values are a ‘‘guiding
principle in your life’’ using the standard SVS 9-point response scale ranging
from - 1 (opposed to my values) to 0 (not important) to 7 (of supreme im-
portance). The scale taps 10 different value types that reflect four higher
order value dimensions. The higher order value dimensions of conservation
and self-transcendence were considered in the current model.2
Self-transcendence value types were tapped by two scales: universalism
(i.e., understanding, appreciation, tolerance, protection for the welfare of all
people and for nature; a 5 .83) and benevolence (i.e., preservation and en-
hancement of the welfare of people with whom one is in frequent personal
contact; a 5 .77; Schwartz, 1994). Conservation value types were tapped by

2
Examination of correlation matrices indicates that the values that were excluded from the
current analyses did not demonstrate large or consistent relationships with the dependent
variables.
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 173

three scales: tradition (i.e., respect, commitment, acceptance of the customs

and ideas that traditional culture or religion provide; a 5 .60), conformity
(i.e., restraint of actions, inclination, and impulses likely to upset or harm
others and violate social expectations and norms; a 5 .69), and security (i.e.,
safety, harmony, and stability of society, of relationships, and of self;
a 5 .60; Schwartz, 1994).
Schwartz (1992) recommended controlling for differences in participants’
use of the SVS by computing each participant’s mean importance rating for
the 57 values and controlling for its effect using partial correlation. To
control for individual differences in scale use while maintaining an accept-
able degree of model parsimony (i.e., to avoid including a large number of
additional variables in the model), we centered each participant’s impor-
tance rating for each value by subtracting the mean importance rating for
the 57 values for each participant. Participants’ final score on each of the 10
value types was the average of the set of centered items making up each
value type. Scores on the higher order value dimensions were calculated by
averaging the final score on the set of values that made up each higher order
value dimension.

Results

Mean Scores for Scenarios

Table 1 presents mean scores on items related to perceptions of the

patient in the alcohol and heroin scenarios. Overall, participants perceived
drug users to hold high levels of responsibility for their alcohol or drug use,
and reported low levels of negative affect (anger, disappointment) and rela-
tively high levels of positive affect toward these individuals (sympathy,
concern). Scores on these measures did not differ significantly between the
heroin and alcohol scenarios.
As Table 2 shows, overall participants perceived high-quality care to be
highly deserved and low-quality care to be undeserved. They reported high
satisfaction in response to high-quality care and low satisfaction in response
to low-quality care. A 2 (Quality of Care: high vs. low)  2 (Drug Type:
heroin vs. alcohol) mixed within–between subjects MANOVA indicates that
for both drug types, participants perceived high-quality care to be more
deserved, and they reported greater satisfaction with high- compared to low-
quality care; differences that, in accordance with our previous discussion,
reflect the influence of entitlement norms for patients within a hospital
setting. Participants’ scores on the centered higher order value dimensions
indicate that self-transcendence value types (M 5 0.82, SD 5 0.45) were
174 SKINNER ET AL.

Table 1

Means for Items Related to Perceptions of the Patient According to Drug

Type

Alcohol (n 5 141) Heroin (n 5 136)

Variable M SD M SD t
Responsibility 4.08 0.94 4.03 0.98 0.41
Positive affect 3.88 0.82 3.84 0.78 0.47
Negative affect 2.43 1.04 2.57 1.08 - 1.08
Note. df 5 275; None of the t tests were significant (p>.05).

rated as slightly more important, as compared to conservation value types

(M 5 - 0.02, SD 5 0.51).

Correlations Between Variables

Correlations between variables are presented in Table 3. With respect to

the predictions of the theoretical model (Figure 1), the central features of the
correlation matrix are as follows:
1. Perceived responsibility for heroin use increased with higher im-
portance assigned to conservation values.
2. Positive affect toward the patient (sympathy, concern) increased
with higher importance assigned to self-transcendence values.
3. Negative affect (anger, disappointment) toward the patient in-
creased with higher importance assigned to conservation values
and decreased with higher importance assigned to self-transcen-
dence values.
4. Perceived deservingness of high-quality care increased with stronger
positive affect (sympathy, concern) and decreased with stronger
negative affect (anger, disappointment) toward the patient.
5. Perceived deservingness of low-quality care increased with stronger
negative affect (anger, disappointment) and decreased with stronger
positive affect (sympathy, concern) toward the patient.
6. Satisfaction with high/low quality of care increased with the per-
ceived deservingness of each outcome.
Table 2

Means for Deservingness of High Versus Low Quality of Care and Satisfaction With Outcome According to Drug Type

Alcohol Heroin Univariate Fs

Drug Quality of Interaction

Variable M SD M SD M SD M SD (A) care (B) (A  B)
Deservingness of high- / 4.63 0.68 1.31 0.67 4.70 0.58 1.29 0.57 0.50 2547.40 0.68
low-quality care
Satisfaction with high- / 4.69 0.54 1.61 0.84 4.69 0.51 1.67 0.71 0.81 2153.20 0.14
low-quality care
p o .001.
STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION
175
Table 3

Correlations Between Variables

176 SKINNER ET AL.

Variable 1 2 3 4 5 6 7 8 9
1. Self-transcendence F - .19 - .16 .27 - .18 .20 - .32 .10 - .28
2. Conservation - .13 F .11 - .03 .23 - .16 .12 .01 .07
3. Responsibility for drug use - .01 .23 F .01 .14 - .07 .06 .02 .04
4. Sympathy/concern .24 - .02 - .04 F - .11 .37 - .38 .31 - .42
5. Anger/disappointment - .21 .32 .08 - .12 F - .28 .37 - .16 .28
6. Deservingness of high-quality care .22 - .02 .01 .31 - .23 F - .64 .62  - .50
7. Deservingness of low-quality care - .28 .07 - .13 - .28 .32 - .43 F - .40 .62
8. Satisfaction with high-quality care .21 .12 .15 .35 - .05 .55 - .37 F - .49
9. Satisfaction with low-quality care - .20 .01 - .02 - .47 .06 - .37 .59 - .38 F
Note. Alcohol correlations are presented above the diagonal, while heroin correlations are presented below the diagonal.
p o .05. p o .01. p o .001.
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 177

Regression Analyses

The proposed model was tested using path analysis. The analysis used
multiple regression with pairwise deletion of missing cases to generate
standardized regression coefficients. Path coefficients were obtained by
regressing each variable on those that directly impinged on it. Figure 2
summarizes the results of the regression analysis in a path diagram (non-
significant paths for both drugs are not shown, but will be discussed in
further detail).

Affective Responses to the Drug User and Perceived Deservingness of High-

and Low-Quality Care

As Figure 2 indicates, the current findings were generally supportive of

the proposed model. The conservation value dimension (security, tradition,
conformity) was associated with stronger feelings of anger and disappoint-
ment toward the patient. The self-transcendence value dimension (benevo-
lence, universalism) was associated with higher levels of sympathy and
concern for the drug user. Contrary to expectations, the conservation value
dimension was not associated with reduced positive affect toward the drug

Key:
Responsibility Alcohol
for Heroin
drug use
.09
.23* 2
R = .01
R2 = .05

Negative affect Deservingness of .62*** Satisfaction with

.22* -.25** the high-quality .55*** the high-quality
Conservation .30*** (angry/disappointed) -.19* care care
R2 = .06
R2 = .14 R2 = .20 R2 = .38
.34*** R2 = .12 R2 = .30
.29**

-.09
-.18* .34***
.29**

Deservingness of .62*** Satisfaction with

.26** Positive affect -.34*** the low-quality .59*** the low-quality
Self-transcendence (sympathy/concern) -.24** care care
.24**
R22 = .07 R22 = .26 R22 = .38
R = .06 R = 16 R = 35

Figure 2. Theoretical model of affective responses to the drug user and deservingness and
satisfaction (standardized coefficients). Path coefficients for negative affect on deservingness of
low-quality care and for positive affect on deservingness of high-quality care were identical
when rounded to two decimal places. p o .05. p o .01. p o .001.
178 SKINNER ET AL.

user, nor was the self-transcendence value dimension associated with re-
duced negative affect (with the exception of a weak effect in the heroin
scenario). In addition, conservation and self-transcendence value dimen-
sions did not predict perceived responsibility for drug use in the alcohol
scenario, and only the conservation value dimension predicted perceived
responsibility in the heroin scenario.
As expected, affective responses to the patient predicted the perceived
deservingness of high- and low-quality care. Low-quality care was perceived
to be more deserved when negative affect was stronger and positive affect
toward the patient was weaker. The opposite pattern was evident for per-
ceived deservingness of high-quality care. However, perceived responsibility
for drug use did not predict positive or negative affect toward the drug user
or deservingness judgments. As expected, perceived deservingness of each
outcome (i.e., high- or low-quality care) strongly predicted satisfaction with
each respective outcome.

Tests for Nonmediated Effects

The presence of nonmediated effects that were not predicted by the

model were tested by using Baron and Kenny’s (1986) regression method-
ology. Nonmediated effects are present if a variable in the model is asso-
ciated with a variable appearing later in the model, even after controlling for
intervening variables. Specifically, we examined the presence of direct effects
of (a) values on deservingness of high- or low-quality care; and (b) positive
affect, negative affect, and responsibility on satisfaction with the nurse’s
response.
Only two nonmediated effects were found. After controlling for per-
ceived deservingness, positive affect toward the patient had a direct impact
on satisfaction with low-quality care (alcohol, b 5 - .25, p o .01; heroin,
b 5 - .34, p o .001). For the heroin scenario, positive affect toward the drug
user had a direct impact on satisfaction with high-quality care (b 5 .21, p o
.01). These findings indicate that higher levels of positive affect toward the
patient (sympathy, concern) predicted lower satisfaction with low-quality
care (alcohol, heroin) and higher satisfaction with high-quality care (heroin).
Therefore, these results are not unexpected.

Comparison of Model Across Drug Type

It is possible that the effects of different variables in the model might

differ according to whether the drug involved is alcohol or heroin. The
equivalence of path coefficients across drug types was tested using Kline’s
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 179

(1998) structural equation modeling multiple-group comparison procedure.

The fit indexes were low when the path coefficients from the two models
were constrained to be invariant, w2(62, N 5 277) 5 185.82, p o .00 (normed
fit index 5 .658, non-normed fit index 5 .695), and none of the Lagrange
multiplier modification indexes were significant. This indicates that none of
the paths in the model were significantly different across drug type.

Discussion

This study supports the utility of a social-justice perspective in under-

standing the dynamics of health-care practices regarding stigmatized con-
ditions (e.g., heroin dependence, alcohol-related problems). Specifically,
judgments of patients’ deservingness of low (or high) quality of care influ-
enced health professionals’ satisfaction with these standards of care. The
results also support the influence of affective responses to individuals on
judgments that high- or low-quality care is deserved.
Positive affect (sympathy, concern) predicted deservingness of high-
quality care, while negative affect (anger, disappointment) predicted de-
servingness of low-quality care. In addition, general value orientations in-
fluenced affective reactions toward the drug user. In particular, conservation
values predicted negative affect toward both the heroin user and the alcohol
user.

Deservingness or Entitlement?

Our results suggest that the nurses’ judgments of deservingness reflected

the strong influence of entitlement norms within the hospital setting. We say
that for two reasons. First, the ratings of deservingness for high- and low-
quality care were at the high and low ends of the rating scale, respectively
(Table 1). So were the composite ratings of satisfaction with high- and low-
quality care. These large differences in ratings are consistent with an inter-
pretation that assumes that the nurses equated deservingness with entitle-
ment in this context; responding in terms of a belief, inculcated in their
training and emphasized within the hospital, that all patients are entitled to
high-quality care, regardless of their perceived responsibility for their med-
ical conditions.
Second, the fact that perceived responsibility did not predict patients’
deservingness or nurses’ affective reactions to either patient is consistent
with an interpretation that entitlement norms dominated their judgments.
As argued previously, perceived responsibility would be expected to
influence judged deservingness when there is a specific and identifiable
180 SKINNER ET AL.

behavioral outcome that can be related to personal causation, as occurred

with the aggressive episode in Feather and Johnstone’s (2001) study. There
was no such specific, salient behavioral outcome in the present study. The
nurses simply responded to patients who were admitted to hospital with a
drug-related complaint, and patients’ perceived responsibility for their con-
ditions was not an issue. Instead, patients were perceived as entitled to
sympathy and concern, rather than to responses expressing anger and dis-
appointment. Patients were also perceived as entitled to high-quality, rather
than low-quality care (see means in Tables 1 and 2).
This interpretation raises interesting issues about the distinction between
deservingness and entitlement, and about how respondents interpret ques-
tions about deservingness. The present results show that in contexts (e.g., a
hospital setting) in which entitlement norms are highly salient, deservingness
judgments may primarily reflect beliefs about entitlement. Similarly, a per-
son left a legacy in a will generally would be perceived as more entitled to the
legacy if he or she were close kin (i.e., a son or daughter), as compared with
a friend who rendered help (Feather, 2003).
In other contexts, however, the focus may be on personal causation and
intentionality, and deservingness judgments may be largely unaffected by
issues of entitlement. Deserving promotion within an organization, for ex-
ample, may be perceived as depending more on an employee’s contributions
and hard work for the organization, rather than on his or her entitlement to
promotion based on formal or informal rules.
In some cases, deservingness judgments may reflect perceived responsi-
bility for an action and its outcome as well as the effects of entitlement
norms, as was the case in Feather and Johnstone’s (2001) study. In that
study, a specific behavioral episode occurred involving a patient with either
schizophrenia or personality disorder who differed in the perceived respon-
sibility for the aggressive incident. These possible interactions between de-
servingness and entitlement have been previously discussed (Feather, 2002a;
Feather & Johnstone, 2001) and investigated in recent studies (Boeckmann
& Feather, 2005; Feather, 2003). They raise important issues for future
research.
It might be argued that the failure of perceived responsibility to predict
either affective reactions to each patient or deservingness of high- or low-
quality care could be a result of inadequate measurement of perceived re-
sponsibility. However, the measure we used has been employed successfully
in previous research addressing complex social problems (e.g., Feather,
1996, 1998, 1999a, 1999b, 2002b, 2003; Feather & Deverson, 2000; Feather
& Johnstone, 2001; Feather & Nairn, 2005; Feather & Oberdan, 2000; Fea-
ther & Sherman, 2002; Feather & Souter, 2002). Therefore, our interpre-
tation that deservingness judgments in the current study strongly reflect
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 181

entitlement norms remains as the most likely explanation of the failure of

perceived responsibility to predict either affect or deservingness.

Affective Reactions to the Drug User

Consistent with previous research (Feather & Johnstone, 2001), negative

affective reactions to the drug user were associated with more negative de-
servingness judgments (i.e., high-quality care less deserved, low-quality care
more deserved); whereas, positive affective reactions were associated with
more compassionate deservingness judgments (i.e., high-quality care more
deserved, low-quality care less deserved). These affective responses reflect
the effects of entitlement norms. They also may partly reflect private feelings
that the nurses held: feelings that could affect the nurses’ own beliefs about
each patient’s deservingness or entitlement to care, but only in a minor way,
given the fact that these beliefs were affected strongly by entitlement norms
within the hospital setting.

Values

The current findings support the influence of values on affective reac-

tions to others in distress. They are consistent with results from other re-
search that has related reported affect to measures of right-wing
authoritarianism and value types (e.g., Altemeyer, 1981, 1988; Feather,
1996, 1998). The importance assigned to conservation values (conformity,
security, tradition) was associated with higher negative affect toward the
drug user; whereas, the importance assigned to self-transcendence values
was associated with stronger reported positive affect toward the drug user.
Note, however, that in the path analysis, the conservation value dimen-
sion did not predict lower positive affect toward drug users, and the self-
transcendence value dimension did not predict lower negative affect. Con-
sideration of the context represented by the hypothetical scenario may pro-
vide further insight. Feather (1999b) proposed that values are activated by
situational cues, and that the likelihood of activation also depends on the
importance placed on a particular value relative to other values within a
person’s value system.
In the current study, individuals who placed high importance on con-
servation value types would be expected to be more sensitive to those aspects
of the scenario that threatened these core beliefs (e.g., deviance and non-
conformity associated with drug addiction). Therefore, in this context, the
activation of conservation value types would be related to a perceived threat
to important values and with negative affect (anger, disappointment)
182 SKINNER ET AL.

associated with this threat. In contrast, individuals who placed a high value
on self-transcendence value types would be expected to be more sensitive to
aspects of the scenario that related to the provision of care and assistance.
Therefore, in this context, the activation of self-transcendence value types
would be related to the potential for fulfilling important values and to
positive affect (sympathy, concern) associated with this potential fulfilment.

Methodological Limitations

Much of the research on individuals’ affective reactions to others in

distress has been conducted with university students. The present study
makes an important contribution to this research literature by examining the
responses of health professionals to a realistic scenario of health-care pro-
vision. However, the limitations of hypothetical scenarios are acknowledged
(Parkinson & Manstead, 1993). To what extent nurses’ responses to a
hypothetical scenario reflect the quality of care they are likely to provide is a
difficult question to study directly. This question does not lend itself easily
to experimental or observational studies because both types of design raise
significant ethical issues concerning the participation of vulnerable individ-
uals (i.e., alcohol or drug users). However, the consistent relationship of
deservingness judgments to affective responses to high- or low-quality care
suggests that these deservingness judgmentsFwhich we propose largely re-
flect entitlement normsFare key variables that are likely to impact the
quality of care that health professionals provide to individuals with stig-
matized conditions.
Injection of heroin and problematic alcohol use are stigmatized drug-
taking behaviors that often evoke strong emotional responses as a result of
their association with criminal or antisocial acts (e.g., drunk driving, do-
mestic violence, theft) and their historical link with moral decline (e.g., lack
of will power, hedonism). The current study supports the utility of a social-
justice perspective for enhancing the understanding of health-care delivery
for stigmatized conditions, such as drug dependence. It also indicates the
importance of addressing both affective responses to stigmatized groups
(e.g., anger, sympathy, concern) and deservingness and entitlement judg-
ments in education and training programs for health professionals.

References

Aalto, M., Pekuri, P., & Seppa, K. (2001). Primary health care nurses’ and
physicians’ attitudes, knowledge, and beliefs regarding brief intervention
for heavy drinkers. Addiction, 96, 305-311.
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 183

Abed, R. T., & Neira-Munoz, E. (1990). A survey of general practitioners’

opinion and attitude to drug addicts and addiction. British Journal of
Addiction, 85, 131-136.
Abouyanni, G., Stevens, L. J., Harris, M. F., Wickes, W. A., Ramakrishna,
S. S., & Ta, E., et al. (2000). GP attitudes to managing drug- and
alcohol-dependent patients: A reluctant role. Drug and Alcohol Review,
19, 165-170.
Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg, Manitoba,
Canada: University of Manitoba Press.
Altemeyer, B. (1988). Enemies of freedom: Understanding right-wing au-
thoritarianism. San Francisco: Jossey-Bass.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
distinction in social psychological research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychology,
51, 1173-1182.
Boeckmann, R. J., & Feather, N. T. (2005). Gender, discrimination beliefs,
group-based guilt, and responses to affirmative action for Australian
women. Manuscript submitted for publication.
Christie, R. (1993). Some experimental approaches to authoritar-
ianism: II. Authoritarianism and punitiveness. In W. F. Stone,
G. Lederer, & R. Christie (Eds.), Strength and weakness: The
authoritarian personality today (pp. 99-118). New York: Springer-
Verlag.
Farmer, R., & Greenwood, N. (2001). General practitioners’ management of
problem drinkers: Attitudes, knowledge, and practice. Drugs: Education,
Prevention, and Policy, 8, 119-129.
Feather, N. T. (1975). Values in education and society. New York: Free
Press.
Feather, N. T. (1994). Human values and their relation to justice. Journal of
Social Issues, 50, 129-151.
Feather, N. T. (1995). Values, valences, and choice: The influences of values
on the perceived attractiveness and choice of alternatives. Journal of
Personality and Social Psychology, 68, 1135-1151.
Feather, N. T. (1996). Reactions to penalties for an offense in relation to
authoritarianism, values, perceived responsibility, perceived seriousness,
and deservingness. Journal of Personality and Social Psychology, 71,
571-587.
Feather, N. T. (1998). Reactions to penalties for offenses committed by
the police and public citizens: Testing a social-cognitive process model
of retributive justice. Journal of Personality and Social Psychology, 75,
528-544.
184 SKINNER ET AL.

Feather, N. T. (1999a). Judgments of deservingness: Studies in the psychol-

ogy of justice and achievement. Personality and Social Psychology
Review, 3, 86-107.
Feather, N. T. (1999b). Values, achievement, and justice: Studies in the psy-
chology of deservingness. New York: Kluwer Academic/Plenum.
Feather, N. T. (2002a). Reactions to supporters and opponents of uranium
mining in relation to status, attitude similarity, and right-wing authori-
tarianism. Journal of Applied Social Psychology, 32, 1464-1487.
Feather, N. T. (2002b). Values and value dilemmas in relation to judgments
concerning outcomes of an industrial conflict. Personality and Social
Psychology Bulletin, 28, 446-459.
Feather, N. T. (2003). Distinguishing between deservingness and entitle-
ment: Earned outcomes versus lawful outcomes. European Journal of
Social Psychology, 33, 367-385.
Feather, N. T., & Deverson, N. H. (2000). Reactions to a motor-vehicle
accident in relation to mitigating circumstances and the gender and moral
worth of the driver. Journal of Applied Social Psychology, 30, 77-95.
Feather, N. T., & Johnstone, C. (2001). Social norms, entitlement, and
deservingness: Differential reactions to aggressive behavior of schizo-
phrenic and personality disorder patients. Personality and Social Psy-
chology Bulletin, 27, 755-767.
Feather, N. T., & Nairn, K. (2005). Resentment, envy, schadenfreude, and
sympathy: Effects of own and others’ deserved or undeserved status.
Australian Journal of Psychology, 57, 87-102.
Feather, N. T., & Oberdan, D. (2000). Reactions to penalties for an offence
in relation to ethnic identity, responsibility, and authoritarianism. Aus-
tralian Journal of Psychology, 52, 9-16.
Feather, N. T., & Sherman, R. (2002). Envy, resentment, schadenfreude,
and sympathy: Reactions to deserved and undeserved achievement and
subsequent failure. Personality and Social Psychology Bulletin, 28,
953-961.
Feather, N. T., & Souter, J. (2002). Reactions to mandatory sentences in
relation to the ethnic identity and criminal history of the offender. Law
and Human Behavior, 26, 417-438.
Feinberg, J. (1970). Doing and deserving. Princeton, NJ: Princeton Univer-
sity Press.
Karam-Hage, M., Nerenberg, L., & Brower, K. J. (2001). Modifying resi-
dents’ attitudes about substance abuse treatment and training. American
Journal on Addictions, 10, 40-47.
Kline, R. B. (1998). Principles and practice of structural equation modeling.
New York: Guilford.
 STIGMA AND DISCRIMINATION IN HEALTH-CARE PROVISION 185

McLaughlin, D. F., McKenna, H., & Leslie, J. C. (2000). The perceptions

and aspirations illicit drug users hold toward health care staff and the
care they receive. Journal of Psychiatric and Mental Health Nursing, 7,
435-441.
Melby, V., Boore, J. R. P., & Murray, M. (1992). Acquired immunodefi-
ciency syndrome: Knowledge and attitudes of nurses in Northern Ire-
land. Journal of Advanced Nursing, 17, 1068-1077.
Menec, V. H., & Perry, R. P. (1998). Reactions to stigmas among Canadian
students: Testing an attribution–affect–help judgment model. Journal of
Social Psychology, 138, 443-453.
National Centre in HIV Social Research. (2001). C-change. Report of the
enquiry into Hepatitis C related discrimination. Sydney, N.S.W. Anti-
Discrimination Board of New South Wales.
Norman, R. (2001). Have you got an attitude problem? Caring for illicit
drug-using patients. Contemporary Nurse, 10, 83-91.
Parkinson, B., & Manstead, A. S. (1993). Making sense of emotion in stories
and social life. Cognition and Emotion, 7, 295-323.
Roche, A. M., & Richard, G. P. (1991). Doctors’ willingness to intervene in
patients’ drug and alcohol problems. Social Science and Medicine, 33,
1053-1061.
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Schmidt, G., & Weiner, B. (1988). An attribution–affect–action theory
of behavior: Replications of judgments of help-giving. Personality and
Social Psychology Bulletin, 14, 610-621.
Schwartz, S. H. (1992). Universals in the content and structure of values:
Theoretical advances and empirical tests in 20 countries. Advances in
Experimental Social Psychology, 25, 1-65.
Schwartz, S. H. (1994). Are there universal aspects in the structure and
contents of human values? Journal of Social Issues, 50, 19-45.
Schwarzer, R., & Weiner, B. (1991). Stigma controllability and coping as
predictors of emotions and social support. Journal of Social and Personal
Relationships, 8, 133-140.
Skitka, L. J., & Tetlock, P. E. (1992). Allocating scarce resources: A con-
tingency model of distributive justice. Journal of Experimental Social
Psychology, 28, 491-522.
Skitka, L. J., & Tetlock, P. E. (1993). Providing public assistance: Cognitive
and motivational processes underlying liberal and conservative
policy preferences. Journal of Personality and Social Psychology, 65,
1205-1223.
Telfer, M. B., & Clulow, L. (1990). Heroin misusers: What they think of
their general practitioners. British Journal of Addiction, 85, 137-140.
186 SKINNER ET AL.

Weiner, B. (1988). An attributional analysis of changing reactions to persons

with AIDS. In R. A. Berk (Ed.), The social impact of AIDS in the US
(pp. 123–132). Cambridge, MA: ABT Press.
Weiner, B. (1995). Judgments of responsibility. New York: Guilford.
Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional analysis
of reactions to stigmas. Journal of Personality and Social Psychology, 55,
738-748.