(Circulation. 1995;91:2753-2761.)
© 1995 American Heart Association, Inc.
Articles |
Sex Differences in Patient Acceptance of Cardiac Transplant Candidacy
Presented in part at the 1993 Clinical Research Meeting of the American Federation for Clinical Research, Washington, DC, May 1993, and published in abstract form in Clin Res. 1993;41:200A.
From the Division of Circulatory Physiology, Department of Medicine, Columbia University College of Physicians and Surgeons (K.D.A., D.M.M.), and the General Clinical Research Center (J.E.G.), the Division of General Medicine, Department of Medicine (J.S.S.), and the Leonard Davis Institute of Health Economics (J.S.S.), University of Pennsylvania, Philadelphia.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background The overwhelming majority of cardiac transplant recipients are men. This can be partially explained by the earlier age at which heart failure develops in men. However, an underrepresentation of women also may reflect physician referral or selection biases or differences in patients' access to or acceptance of heart transplantation.
Methods and Results We investigated whether sex bias occurred in the transplant candidate selection process at a single cardiac transplant center. We prospectively evaluated 386 individuals <70 years of age (295 men, 91 women) referred for management of moderate to severe heart failure and/or cardiac transplant evaluation. Age, race, sex, heart failure type, New York Heart Association class, left ventricular ejection fraction, peak exercise oxygen consumption, disease duration, resting hemodynamic measurements, comorbidity index score, health insurance coverage, and estimated household income were recorded. For patients not accepted for transplantation, the reason for rejection was also obtained. Univariable and multivariable (logistic regression) analyses were performed comparing men and women and patients accepted and those not accepted for cardiac transplantation. Female sex was independently associated with rejection for cardiac transplantation (odds ratio, 2.57; P=.01). However, the reason for rejection was more likely to be patient self-refusal for women than for men (29% versus 9%), and female sex was independently associated with patient self-refusal (odds ratio, 4.68; P=.003). When patients who refused transplant were reclassified as accepted for transplant, female sex was no longer associated with nonacceptance. However, lower patient income was associated with nonacceptance for transplant.
Conclusions We found no evidence of sex bias in the selection of cardiac transplant recipients at our center. These findings suggest that the underrepresentation of women among cardiac transplant recipients may result, in part, from a sex difference in treatment preference, with a decreased willingness of women to undergo transplantation. The reasons for the difference in acceptance rates between men and women need to be elucidated.
Key Words: transplantation • patient acceptance of health care • sex factors
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The sex of the patient may influence physicians' decisions about the use of diagnostic and therapeutic procedures. Several studies have suggested that physicians undertake a less aggressive approach to the diagnoses and treatment of coronary artery disease in women, with less frequent use of invasive cardiac procedures1 2 3 and later referral for bypass surgery.4
In 1993, 2298 cardiac transplantations were performed in the United States, with women constituting only 21.5% of recipients.5 Survival after cardiac transplantation now approaches 85% at 1 year6 and is nearly equivalent in women and men.6 7 8 9 10 Quality of life after cardiac transplantation is good,11 12 13 14 15 16 with no reported difference in quality of life between male and female recipients. Possible explanations for the underrepresentation of women among patients receiving cardiac transplants include the later age at which women develop heart failure17 18 and the lower heart failure mortality rate among women with the syndrome.19 20 Other potential causes of an underrepresentation of women include referral bias, selection bias, or less access to or acceptance of cardiac transplantation among women. Since the safety and efficacy of cardiac transplantation does not differ by sex, the lower use of the procedure in women warrants exploration. No prior studies have examined sex bias in the selection of cardiac transplant candidates. Accordingly, we studied a consecutive series of patients with moderate to severe congestive heart failure presenting to a single cardiac transplantation center to investigate possible sex-based differences in the selection of cardiac transplant candidates.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Study Population
Four hundred eight consecutive patients with moderate to severe heart failure (left ventricular ejection fraction [LVEF]
40%)
were
referred to the Heart Failure Clinic at the Hospital of the University
of Pennsylvania from July 1986 through December 1991. Data for this
study were collected in the course of an ongoing investigation of
prognostic indicators in patients with heart failure. Twenty-two
patients >70 years old were excluded, leaving 386 eligible patients
(295 men, 91 women) included in the present study.
Data Collection
All patients underwent a standard evaluation
consisting of a
clinical history, physical examination, chest radiograph, 12-lead ECG,
serum chemistries, and determination of LVEF (by radionuclide
ventriculography, cardiac catheterization, or two-dimensional
echocardiography). Substance abuse was defined as a pattern of heavy
ethanol ingestion during the prior 6 months or illicit drug use.
Determination of New York Heart Association (NYHA) functional
classification was made by two cardiologists.21 Maximal
treadmill exercise testing with measurement of peak oxygen consumption
(peak 
O2) was performed in
278
ambulatory patients according to the modified Naughton protocol.
Measurements of mixed expired oxygen, mixed expired carbon dioxide, and
expired volume were determined at rest and every 30 seconds throughout
exercise with a metabolic cart (Sensor Medics). Sixty-nine patients
were bedridden with NYHA class IV failure and unable to exercise; they
were assigned a O2 of 3.5
mL · kg-1 · min-1 (1 metabolic
equivalent). Peak O2 was
not measured
or assigned in 39 patients who had severe angina or noncardiac
limitations to exercise (eg, orthopedic or arthritic problems). Cardiac
output was determined as the average of three measurements assessed by
the thermodilution technique.
A comorbidity index score was determined for each patient according to a modification of a method previously described by Charlson et al.22 In the original methodology, the presence or absence of 18 disease conditions is assessed for each patient. Each condition is assigned a predetermined severity weight; the comorbidity index score equals the sum of the weights for each condition present. A history of congestive heart failure and myocardial infarction constituted 2 of the original 18 disease categories; our modification removed these 2 categories so that only noncardiac diseases remained in the comorbidity index, leaving a potential range of scores from 0 to 35.
Each patient's type of insurance coverage (ie, commercial insurance, health maintenance organization, Blue Cross/Blue Shield, Medicare, Medicaid, or uninsured) at the time of the initial evaluation was obtained retrospectively through computerized billing records. Four male patients who received their care at the Philadelphia Veterans Affairs Medical Center and 1 male patient who was insured by workers' compensation were classified as Medicaid-insured for some analyses. Initially uninsured patients who were successfully assisted in obtaining insurance in the first few weeks after referral to our clinic (ie, before decisions about acceptance or rejection as transplant candidates) were classified as insured. Twelve patients (4 nonwhite women, 4 nonwhite men, 4 white men) were uninsured. Since lack of insurance was an absolute barrier to transplantation at our center, these patients were excluded from the analysis of reasons patients were not accepted for cardiac transplantation and from all multivariable analyses (inclusion of uninsured patients in these analyses would have resulted in confounded estimates of the effects of other variables of interest). Household income was estimated from tables of the median household income by zip code from the 1990 census. Estimated incomes were grouped into quartiles from lowest (1st to 24th percentile) to highest (74th to 99th percentile). Decisions regarding transplant candidacy were made after completion of a full evaluation of patients who were on an optimal medical regimen.
Statistical Analysis
Sample statistics and distributions of
all continuous variables
were examined, and logarithmic and square-root transformations were
performed to satisfy parametric test assumptions of normality or equal
variance when appropriate. Differences between men and women, white and
nonwhite individuals, and patients accepted and those rejected as
transplant candidates were tested by independent sample t
tests (for equal or unequal variances) or Mann-Whitney-Wilcoxon
rank-sum tests for continuous variables and by Pearson's
2, continuity adjusted 2,
or Fisher's exact test for categorical variables as appropriate.
Results for continuous variables are expressed as mean±SD. A
two-tailed value of P.05 was considered significant with
no adjustments made for multiple testing. Multivariable analyses of
predictors of acceptance of transplant candidacy and of self-refusal as
the cause of rejection (among patients not accepted as transplant
candidates) were performed by logistic regression using a stepwise
backward elimination technique: All variables that were marginally
significant (P.20) with univariable analyses and any
statistical interactions of these variables suggested by preliminary
analyses were initially entered, with the least significant variables
(P>.05) sequentially eliminated. Results of the
multivariable analyses are expressed as adjusted odds ratios (ORs) with
their associated 95% confidence intervals (CIs). All analyses were
performed with SAS version 6.07 (SAS Institute).
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Clinical Characteristics
Characteristics of the entire patient population and separately for men and women are shown in Table 1
. Patients ranged
in age from 18 to 70 years, with a mean±SD age of 52±11 years.
Mean
NYHA functional class was 2.9±0.9, with 11% class I, 16% class II,
44% class III, and 30% class IV. The duration of heart failure before
study enrollment averaged 25±37 months. Mean LVEF was 20±8%,
mean
pulmonary capillary wedge pressure 22±10 mm Hg, mean cardiac index
2.1±0.6
L · min-1 · m-2, and
mean peak O2
12.4±6.6 mL
O2 · kg-1 · min-1.
The
average comorbidity index score was 0.7±1.1.
|
Whites constituted 71% of the study sample, with 26% black, 2% Hispanic, 1% Asian, and 0.5% from other racial/ethnic groups. The cause of heart failure was ischemic in 52% and nonischemic in 48% of cases. Twenty-one percent of patients had a history of substance abuse. The distribution of health insurance types for the sample was 39% Blue Cross/Blue Shield, 22% commercial, 18% Medicare, 13% Medicaid, 4% health maintenance organization, 1% Veterans Affairs, 0.3% workers' compensation, and 3% none.
There were no significant differences
between men and women by age,
NYHA class, prior duration of heart failure, LVEF, cardiac index, or
comorbidity index score. The pulmonary capillary wedge pressure was
higher in men than in women, as was the peak
O2. Men were more likely to
be white,
to have an ischemic cardiomyopathy, and to have a substance abuse
history. Health insurance coverage differed between men and women
(P=.02), with men particularly less likely to have Medicaid
(9% of men versus 24% of women). Estimated household income also
differed between men and women (P=.01), with more women in
the lowest income quartile and more men in the second lowest and the
highest income quartiles.
Accepted Versus Rejected Patients
One hundred eighteen
patients were accepted and 268 patients were
rejected as transplant candidates. The reasons patients were rejected
were prospectively identified and recorded. Among patients not
accepted, 38% were considered too well for transplant candidacy. We
have shown previously that patients with a peak
O2 
14 mL
O2 · kg-1 · min-1
have an
expected 1-year survival >90%.23 Since this is better
than the expected 1-year survival after transplantation, a peak
O2 14
mL · kg-1 · min-1 was a
strong,
albeit not absolute, contraindication to transplant candidacy.
Additional reasons for rejection included significant noncardiac
comorbidity, 21%; age 65 years, 11%; psychosocial reasons (eg,
ongoing substance abuse, serious psychiatric illness requiring
treatment, medical noncompliance), 10%; and pulmonary vascular
resistance >6 Wood's units despite acute vasodilator therapy, 2%. Of
those not accepted, 8% had "other" reasons: recommendation of
alternative interventions (eg, valve replacement, coronary artery
bypass graft surgery, coronary angioplasty, or internal
cardioverter/defibrillator implantation) or a further period of
observation (eg, for certain patients with acute myocarditis,
peripartum cardiomyopathy, or new-onset heart failure of unexplained
cause). Five patients who died before their evaluations were completed
were also classified as not accepted for other reasons. Fourteen
percent of patients refused to be placed on the transplant waiting list
despite a recommendation of transplantation as their best therapeutic
option. These patients, who had no medical contraindication to cardiac
transplantation, were classified as not accepted because of patient
refusal ("refusers").
Univariable comparisons between
patients accepted and those rejected as
transplant candidates are shown in Table 2. There were
no significant differences in age or prior duration of congestive heart
failure between patients accepted and those not accepted for cardiac
transplantation. NYHA functional class, mean pulmonary capillary wedge
pressure, and comorbidity index score were lower, and LVEF, cardiac
index, and peak O2 were
higher in
those rejected as transplant candidates. Ischemic pathogenesis was less
common, while nonwhite race and a history of substance abuse were more
frequent in patients not accepted. Women constituted a larger
proportion of the patients rejected for transplant.
|
The distribution of insurance status also differed between patients accepted and those rejected for cardiac transplantation (P=.003). Only patients with health insurance were accepted (4.5% of patients not accepted for transplantation were uninsured). Accepted patients were less likely to be insured by Medicaid (7% versus 17%) and more frequently had Blue Cross/Blue Shield, commercial, or health maintenance organization insurance coverage (78% versus 60%).
Estimated household income differed markedly between accepted and rejected patients (P<.001). Rejected patients were much more commonly in the lowest income quartile (33% versus 8%) and far less commonly in either of the top two income quartiles (42% versus 65%). These differences persisted even after exclusion of the 12 uninsured patients (P<.001).
Reasons for Rejection: Men Versus Women
Reasons that patients
were rejected differed significantly between
men and women (2=27.27, P<.001)
(Table 3). Women were much less likely to have been
considered too well (19% of women versus 42% of men, OR 0.32,
P=.001). Women were far more likely to have refused
transplant. Among the 268 patients rejected, 29% of women versus 9%
of men were refusers (OR 4.08, P<.001). All but 2 of the
women who refused transplant would have been placed on the waiting list
as status 2 (ie, not requiring an intravenous inotrope or mechanical
support). These women rejected a therapeutic option recommended to
improve their chances of survival over the next few months to years.
The other reasons for rejection were distributed similarly between men
and women. Of note, reasons for rejection did not differ significantly
between whites and nonwhites (2=9.96,
P=.13).
|
Patient Refusal Versus All Other Reasons for Nonacceptance
Table 4 shows the characteristics of patients who
refused transplant candidacy contrasted with those of patients who were
rejected as transplant candidates for all other reasons. Age, NYHA
class, prior duration of congestive heart failure, LVEF, pulmonary
capillary wedge pressure, cardiac index, and the prevalence of an
ischemic cause and substance abuse were similar between the groups.
Peak exercise oxygen consumption and comorbidity index score were
significantly lower in refusers. Women constituted a larger proportion
of patients who refused transplant candidacy than of patients who were
rejected as candidates for all other reasons (54% versus 22%, OR
4.08, P<.001). There was only a trend toward differences in
insurance status (P=.10), with patients who refused
transplant candidacy more likely to be insured through Medicaid,
Veterans Affairs, or workers' compensation. Estimated household income
did not differ between the groups (P=.50).
|
Multivariable Analyses: Predictors of Rejection and of Patient
Refusal
Multivariable logistic regression analysis was performed to
identify variables that were independently associated with
rejection of transplant candidacy (Table 5). The
interactions of race by sex and race by substance abuse were evaluated
(the interaction of sex by substance abuse was eliminated because of
redundancy, since all but 2 patients with a history of substance abuse
were men). Female sex (adjusted OR 2.57, P=.01) was
independently associated with rejection. A history of substance abuse,
NYHA class I or II symptomatology (versus NYHA class III or IV), the
lowest quartile of estimated household income (versus any other income
quartile), higher comorbidity index score, greater peak
O2, and greater LVEF were
also
independently associated with nonacceptance. Race was not a significant
predictor in this model (OR for nonwhites versus whites 1.74,
P=.14).
|
Since patients who refused transplant would
have been accepted as
transplant candidates, another multivariable logistic regression
analysis was performed after refusers were reclassified as accepted
transplant candidates (Table 6). Sex was no longer
independently associated with nonacceptance after reclassification of
refusers (adjusted OR for women 1.15, P=.67). Race and
nonacceptance remained unassociated after reclassification (adjusted OR
for nonwhites 0.96, P=.96).
|
Data from the 256 insured
patients not accepted as transplant
candidates were then reexamined to identify variables independently
associated with patient refusal of transplant candidacy. Women were
independently more likely to refuse transplant candidacy than were men
(adjusted OR 4.68, 95% CI 2.01 to 10.89, P=.003), as were
patients with lower comorbidity index scores (adjusted OR 0.50 for each
additional point, 95% CI 0.30 to 0.84, P=.01). An
association between peak O2
and
patient refusal of transplant candidacy (adjusted OR 0.94 for each
additional mL
O2 · kg-1 · min-1,
95% CI ratio 0.88 to 1.00, P=.06) was of borderline
significance. Race (adjusted OR 1.59, P=.29) was not a
predictor of patient refusal, nor were estimated household income and
type of insurance.
Since men were more frequently judged too well for transplant, they had less opportunity to refuse transplantation than women did (since patients were classified as refusers only if they were clinically appropriate transplant candidates). The analysis to identify variables independently associated with patient refusal of transplant candidacy was therefore repeated after data from individuals considered too well for transplantation were excluded. Women remained significantly more likely to refuse transplant candidacy (adjusted OR 3.75, 95% CI 1.46 to 9.68).
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Decisions about transplant candidacy are made at individual transplantation centers by a committee of physicians and other health professionals, guided by their collective clinical judgment. As the national cardiac transplant waiting list grows, the chance that an individual candidate will receive a donor heart decreases. Transplantation physicians attempt to optimize survival for the entire heart failure population at risk by accepting only those patients with the greatest projected survival benefit from the procedure. Potential candidates, referring physicians, and the general public must be confident that these decisions are made without bias.
Some cause for concern exists, because studies have identified a lower rate of renal transplantation in women (versus men) with end-stage renal disease requiring dialysis, despite adjustment for age, race, comorbidity, health insurance, and income.24 25 Several investigations have suggested that physicians undertake a less aggressive approach to the diagnosis and treatment of coronary artery disease in women, with less frequent use of invasive cardiac procedures1 2 3 and later referral for bypass surgery.4 26 While confirming sex differences in the use of these procedures, recent investigations that more thoroughly adjusted for important covariates that may influence physician behavior (eg, age, angina severity, extent of coronary disease) have found the adjusted use rates of coronary angiography and revascularization procedures to be more comparable.27 28 In one study, the referral of patients for coronary bypass surgery appeared to reflect more judicious use of the procedure among women.27 Many studies have found coronary artery bypass graft surgery to be less effective in women than in men,4 26 29 30 31 32 but these differences may result from a greater frequency of urgent or emergent surgery,4 26 29 32 smaller coronary arteries,26 32 and less frequent use of an internal mammary artery graft in women.4 32
The use of a procedure may appropriately differ by sex if, after adjustment for all potentially confounding variables, its utility also differs by sex.33 However, when the effectiveness of a procedure is similar in men and women (as in cardiac transplantation), reasons for differential use should be examined. Therefore, we studied patients presenting to a single cardiac transplantation center over a 5-year period to determine whether there was evidence of sex bias in the selection of cardiac transplant candidates.
Effect of Sex on Transplant Candidacy
The major finding of
this study is that the
underrepresentation of female transplant candidates at our
center resulted from a sex difference in patient treatment preference
rather than from physician selection bias. To the best of our
knowledge, this is the first study to identify a sex difference in
treatment preference in the areas of cardiology and transplantation.
In univariable analyses, we found that men were more likely than women to be accepted for cardiac transplantation. Other factors associated with acceptance as a transplant candidate included increased NYHA functional class, pulmonary capillary wedge pressure, and lower LVEF, cardiac index, peak exercise oxygen consumption, and comorbidity index score. An ischemic origin and the absence of a substance abuse history were also associated with acceptance as a transplant candidate. This clinical profile of patients accepted and rejected for transplantation at our center is consistent with the recently published experience of a large transplantation center34 and with the currently recommended indications for transplantation.35 White race, insurance coverage providing higher levels of reimbursement, and an estimated household income in any of the top three quartiles were also associated with a greater likelihood of transplant candidacy. In a multivariable analysis adjusting for all of these factors, female sex remained associated with rejection.
The reasons patients were rejected for transplantation were analyzed to explore possible explanations for the sex difference in acceptance rates.The reasons for rejection did differ significantly between men and women, with women far more likely to have refused transplant candidacy.
Since our aim was to determine whether or not selection bias by physicians was present, we sought to eliminate the effect of patient treatment preferences. Therefore, patients who refused transplant candidacy were reclassified as accepted, and the multivariable analysis of factors associated with rejection as a transplant candidate was repeated. In this analysis, sex was no longer associated with rejection as a transplant candidate. These results suggest that sex-related patient treatment preferences had confounded the association between sex and transplant candidacy. This is the first study to suggest that the use of a cardiac procedure is influenced by patient preference.
To confirm the relation between sex
and treatment preference,
independent predictors of patient refusal of transplant candidacy were
sought in a multivariable analysis. As expected, female sex
remained strongly associated with patient refusal. Higher comorbidity
index score and greater peak
O2 were
also associated with patient refusal, as they had been in the
univariable analyses. These latter findings are consistent with current
transplantation guidelines.35
Our findings suggest that
the preference of some women to forego
transplant candidacy, rather than physician bias, caused women to be
underrepresented on our transplant waiting list. This is in
contrast to the results of a study of 268 patients on chronic dialysis,
in which no difference was found in the proportion of men and women who
wished to receive a kidney transplant (67% versus
65%).36 A similar lack of treatment preference by sex was
observed among 157 dialysis patients 70 years old in their wish to
receive a renal transplant.24 37 The availability of
an
alternative lifesaving intervention (ie, dialysis) to those patients
who refused renal transplantation may account for the discrepancy
between these results and our own.
We can only speculate as to why women were more likely to choose against transplant candidacy. Societally reinforced sex roles and the social support structures they engender have a strong influence on patients' perceptions of illness and health-seeking behaviors.38 39 Men and women may have different perceptions of illness severity and of subsequent limitations. Women may be more accepting of a reduced functional capacity, particularly if their premorbid lifestyle was less active. Religious and sociocultural differences may also be important in patient's acceptance of transplant. Women tend to be more religious than men40 and may be more accepting of fate.
Cardiac transplantation involves a high level of uncertainty. Women and men may differ in their acceptance of risk and uncertainty and in how they implicitly value time in the present versus time in the future. Transplant imposes an enormous financial burden on patients and their families that women may be less able41 or willing to incur.
Communications between patients and physicians may greatly influence treatment choices. The sex of the physician can affect treatment and patient-physician interactions.42 Reviewing a large insurance claims database, Lurie et al43 found that women were more likely to undergo screening with Pap smears and mammograms if they saw female rather than male physicians. Although few cardiologists are women, at our center, most patients were screened by a female cardiologist. Nevertheless, women's style of communication and the resulting perceptions may differ from those of men.
Women may have unique concerns about posttransplantation life and distortions of body image. The possibility of experiencing common side effects of standard immunosuppressive therapy that distort body image (eg, Cushingoid habitus, hirsutism, and gingival hyperplasia) may have been of particular concern to women.
Effect of Race on Transplant Candidacy
Nonwhite race and
rejection as a transplant candidate were
strongly associated by univariable analysis. However, compared with
white patients, nonwhite patients were more likely to be women (32%
versus 20%, P=.023), to be substance abusers (32% versus
11%, P<.001), and to have estimated household incomes in
the lowest quartile (59% versus 7%, P<.001). Since these
three factors were potent independent predictors of transplant
candidacy, race was not independently associated with transplant
candidacy in the multivariable analyses after these factors were
considered.
Reasons for nonacceptance did not differ significantly between whites and nonwhites, and race was not an independent predictor of patient refusal as the reason for rejection of transplant candidacy. However, after refusers were reclassified as accepted for transplant candidacy, there was a marked change in the adjusted OR for race and in its significance (from OR 1.74, P=.14 to OR 0.96, P=.96). This suggests that treatment preference may have differed by race; the inability to detect a difference between whites and nonwhites in treatment preference may have resulted from inadequate statistical power. In a sample of dialysis patients from the District of Columbia, Callender has shown that blacks less frequently wished to receive a kidney transplant than did whites (20% versus 10%) and that this difference decreased after patients were educated about transplantation.44
Effect of Income on Transplant Candidacy
The finding that
patients in the lowest income quartile were less
likely to be accepted for transplant listing persisted despite
exclusion of uninsured patients, adjustment for substance abuse history
and insurance type, and reclassification of refusers as accepted. A
similar finding has been reported in the renal transplant literature.
In a random national sample of 4118 patients who began dialysis in 1986
or 1987, the likelihood of receiving a first renal transplant increased
by 16% for every $10 000 increase in estimated median household
income by zip code.25 Ozminkowski et al45
used 1986-1987 discharge abstract data from the Hospital Cost and
Utilization Project database (a national sample of more than 500
nonfederal short-term general hospitals) to identify heart and liver
transplant recipients and other patients with heart failure and liver
disease who were hospitalized at the same institutions but did not
receive transplants. After adjustment for differences in the expected
years of survival after transplantation, age, sex, and race, patients
who had the greatest expected ability to pay for the transplantation
(ie, those with private insurance and estimated household incomes above
the national median) were more likely to receive transplants than were
patients with lesser financial resources.
There may be cogent reasons why economically disadvantaged patients were less likely to be accepted as transplant candidates despite the presence of insurance and absence of substance abuse. A history of poor compliance with medical therapy is a contraindication to transplantation. In renal transplant recipients, poor compliance with medical therapy may be more common among those less economically advantaged.46 Although insured, these patients often suffer from inadequate pharmaceutical coverage and other financial hardships that make good compliance difficult or impossible. Nevertheless, we suspect that subtle barriers of education, language, culture, and economics conspire to decrease the frequency with which poorer patients are placed on transplant waiting lists.
Limitations
Several important limitations of this study will
need to be
addressed in future studies. This study was performed at a single East
Coast urban referral center and reflects this center's particular
referral patterns and the treatment decisions made by a small number of
physicians on its staff. These data do not exclude the possibility of
sex bias in the selection of cardiac transplant candidates at other
centers, nor do they explicitly address the potentially greater problem
of sex bias in the referral of heart failure patients to transplant
centers for evaluation. In the previously noted study of Ozminkowski et
al,45 women with heart failure were less likely to receive
a heart transplant than were men, despite adjustment for age, race, and
expected ability to pay. Our data provide limited support for a
referral bias by sex in that men were more likely to be "too
well" for transplant, as evidenced by a higher peak
O2. However, men are
expected to have
a higher peak O2 than are
women
because their muscle mass is proportionally
greater.47
These data were collected in the course of an ongoing study of prognostic indicators in patients with advanced heart failure; they were not collected with the evaluation of selection bias in mind. Therefore, many potentially important covariates were not available for analysis. Estimated household income by zip code is only a rough surrogate for actual household income. More complete financial information (eg, actual income, number of dependents) would have allowed us to better define the relation between finances, transplant candidacy, and treatment preference.
Information about educational status was also lacking. Physicians may identify more with better-educated patients and may communicate more effectively with these individuals. More-educated patients may be better informed about transplantation as a heart failure treatment and may be better able to weigh the relative risks of further medical treatment alone versus those of cardiac transplantation. Whether any of these factors were important in patients' decisions to accept or reject transplant candidacy and to what extent, if any, their impact differs by sex should be prospectively assessed.
Clinical Implications
Patients presenting to other
transplantation centers should be
studied to confirm or refute our findings. If studies at other centers
are confirmatory, it will be important to determine why women are more
likely to refuse transplant candidacy, so that appropriately targeted
interventional strategies can be pursued (eg, increased education about
cardiac transplantation, alternative forms of communication). Although
cardiac transplantation is a dramatic, high-visibility procedure, the
public health impact of an inappropriately low use of the procedure in
women would be quite small, given the relatively few procedures
performed nationwide.5 One would suspect, however, that if
there are sex differences in patients' treatment preferences for heart
transplantation, sex differences may also be present for other
health care treatment choices that affect a far greater number of
individuals. The ramifications of such a finding could be profound,
particularly if sex differences in patients' treatment preferences
were present for treatments that are equally effective in men and
women. In exploring this question, effectiveness must be taken to
encompass the full range of health outcomes (ie, including
quality-of-life measures), which may be valued differently by men and
women.
Many studies have identified variations in the use of services by sex,1 2 3 4 24 25 26 27 28 45 racial,24 25 45 48 49 50 51 52 53 54 and socioeconomic groups25 45 48 54 and have raised the question of physician bias in the application of these services. If we are to redress inequities that variation may represent, investigators must first determine why such differences exist. Administrative and procedurally oriented databases provide only limited information in this regard. Patient-level data, in the form of surveys and interviews, may provide greater insight into why variations in the use of services exist, whether these variations are rational, and if not, how they can best be remedied.
|
Acknowledgments |
|---|
This work was supported by an American Heart Association Delaware Affiliate Fellowship Award, National Research Service Award 1-F32-HL-08697, and Clinical Investigator Development Award K08-HL-02829 from the National Heart, Lung, and Blood Institute, National Institutes of Health (K.D.A.) and by US Public Health Service Research Grant M01-RR00040 from the National Institutes of Health to the General Clinical Research Center at the Hospital of the University of Pennsylvania.
|
Footnotes |
|---|
Reprint requests to Keith D. Aaronson, MD, MS, Division of Circulatory Physiology, Columbia–Presbyterian Medical Center, Milstein Hospital Building, 5th Floor, Room 435, 177 Ft Washington Ave, New York, NY 10032. E-mail kda2@columbia.edu.
Received October 17, 1994; accepted December 14, 1994.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Tobin J, Wassertheil-Smoller S, Wexler J, Steingart R, Budner N, Lense L, Wachspress J. Sex bias in considering coronary bypass surgery. Ann Intern Med. 1987;107:19-25.
2. Steingart R, Packer M, Hamm P, Coglianese M, Gersh B, Geltman E, Sollano J, Katz S, Moyé L, Basta L. Sex differences in the management of coronary artery disease. N Engl J Med. 1991;325:226-230. [Abstract]
3. Ayanian J, Epstein A. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med. 1991;325:221-225. [Abstract]
4. Khan S, Nessim S, Gray R, Czer L, Chaux A, Matloff J. Increased mortality of women in coronary artery bypass surgery: evidence for a referral bias. Ann Intern Med. 1990;112:561-567.
5. Benenson E, ed. UNOS Update Vol 10(6). Richmond, Va: United Network for Organ Sharing; 1994.
6. Kobashigawa J, Kirklin J, Naftel D, Bourge R, Ventura H, Mohanty P, Cintron G, Bhat G. Pretransplantation risk factors for acute rejection after heart transplantation: a multiinstitutional study. The Transplant Cardiologists Research Database Group. J Heart Lung Transplant. 1993;12:355-366. [Medline] [Order article via Infotrieve]
7. Fragomeni S, Kaye M. The registry of the International Society for Heart Transplantation: fifth official report. J Heart Transplant. 1988;7:249-253. [Medline] [Order article via Infotrieve]
8. Crandall B, Renlund D, O'Connell J. Increased cardiac allograft rejection in female heart transplant recipients. J Heart Transplant. 1988;7:419-423. [Medline] [Order article via Infotrieve]
9. Esmore D, Spratt P, Keogh A, Jones B, Chang V. Cardiac transplantation in females. J Heart Transplant. 1989;8:109. Abstract.
10. Fabbri A, Bryan A, Sharples L, Dunning J, Caine N, Schofield P, Wallwork J, Lange S. Influence of recipient and donor gender on outcome after heart transplantation. J Heart Lung Transplant. 1992;11:701-707. [Medline] [Order article via Infotrieve]
11. Caine N, Sharples L, English T, Wallwork J. Prospective study comparing quality of life before and after heart transplantation. Transplant Proc. 1990;22:1437-1439. [Medline] [Order article via Infotrieve]
12. Bunzel B, Grundbock A, Laczkovics A, Holzinger C, Teufelsbauer H. Quality of life after orthotopic heart transplantation. J Heart Lung Transplant. 1991;10:455-459. [Medline] [Order article via Infotrieve]
13. Angermann C, Bullinger M, Spes C, Zellner M, Kemkes B, Theisen K. Quality of life in long-term survivors of orthotopic heart transplantation. Z Kardiol. 1992;81:411-417. [Medline] [Order article via Infotrieve]
14. Jones B, Taylor F, Downs K, Spratt P. Longitudinal study of quality of life and psychological adjustment after cardiac transplantation. Med J Aust. 1992;157:24-26. [Medline] [Order article via Infotrieve]
15. Riether A, Smith S, Lewison B, Cotsonis G, Epstein C. Quality-of-life changes and psychiatric and neurocognitive outcome after heart and liver transplantation. Transplantation. 1992;54:444-450. [Medline] [Order article via Infotrieve]
16. Rosenblum D, Rosen M, Pine Z, Rosen S, Borg-Stein J. Health status and quality of life following cardiac transplantation. Arch Phys Med Rehabil. 1993;74:490-493. [Medline] [Order article via Infotrieve]
17. Smith W. Epidemiology of congestive heart failure. Am J Cardiol. 1985;55:3A-8A. [Medline] [Order article via Infotrieve]
18. Kannel W, Belanger A. Epidemiology of heart failure. Am Heart J. 1991;121:951-957. [Medline] [Order article via Infotrieve]
19. Shocken D, Arrieta M, Leaverton P, Ross E. Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol. 1992;20:301-306. [Abstract]
20.
Ho K, Anderson K, Kannel W, Grossman W, Levy D.
Survival after the onset of congestive heart failure in
Framingham Heart Study subjects.
Circulation. 1993;88:107-115.
21. The Criteria Committee of the New York Heart Association. Diseases of the Heart and Blood Vessels: Nomenclature and Criteria for Diagnosis. 6th ed. Boston, Mass: Little, Brown & Co; 1964.
22. Charlson M, Pompei P, Ales K, MacKenzie C. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383. [Medline] [Order article via Infotrieve]
23.
Mancini D, Eisen H, Kussmaul W, Mull R, Edmunds L,
Wilson J. Value of peak exercise oxygen consumption for optimal
timing of cardiac transplantation in ambulatory patients with heart
failure. Circulation. 1991;83:778-786.
24.
Kjellstrand C. Age, sex, and race inequality in
renal transplantation. Arch Intern Med. 1988;148:1305-1309.
25.
Gaylin D, Held P, Port F, Hunsicker L, Wolfe R, Kahan
B, Jones C, Agodoa L. The impact of comorbid and
sociodemographic factors on access to renal transplantation.
JAMA. 1993;269:603-608.
26. Fisher L, Kennedy J, Davis K, Maynard C, Fritz J, Kaiser G, Myers W. Association of sex, physical size, and operative mortality after coronary artery bypass in the Coronary Artery Surgery Study (CASS). J Thorac Cardiovasc Surg. 1982;84:334-341. [Abstract]
27. Bickell N, Pieper K, Lee K, Mark D, Glower D, Pryor D, Califf R. Referral patterns for coronary artery disease treatment: gender bias or good clinical judgement? Ann Intern Med. 1992;116:791-797.
28. Krumholz H, Douglas P, Lauer M, Pasternak R. Selection of patients for coronary angiography and coronary revascularization early after myocardial infarction: is there evidence for a gender bias? Ann Intern Med. 1992;116:785-790.
29. Loop F, Golding L, MacMillan J, Cosgrove D, Lytle B, Sheldon W. Coronary artery surgery in women compared with men: analysis of risks and long-term results. J Am Coll Cardiol. 1983;1:383-390. [Abstract]
30. Hall R, Elayda M, Gray A, Mathur V, Garcia E, de Castro C, Massumi A, Cooley D. Coronary artery bypass: long term follow-up of 22,284 consecutive patients. Circulation. 1983;68(suppl II):II-20–II-26.
31.
Hannan E, Kilburn H Jr, O'Donnel J, Lukacik G, Shields
E. Adult open heart surgery in New York State: an analysis
of risk factors and hospital mortality rates. JAMA. 1990;264:2768-2774.
32.
O'Connor G, Morton J, Diehl M, Ollmstead E, Coffin L,
Levy D, Maloney C, Plume S, Nugent W, Malenka D, Hernandez F, Clough R,
Birkmeyer J, Marrin C, Leavitt B, for the Northern New England
Cardiovascular Disease Study Group. Differences between men and
women in hospital mortality associated with coronary artery bypass
graft surgery. Circulation. 1993;88:2104-2110.
33. Laskey W. Gender differences in the management of coronary artery disease: bias or good clinical judgement? Ann Intern Med. 1992;116:869-871.
34. Kubo S, Ormaza S, Francis G, Holmer S, Olivari M, Bolman M, Shumway S. Trends in patient selection for heart transplantation. J Am Coll Cardiol. 1993;21:975-981. [Abstract]
35. Mudge G, Goldstein S, Addonizio L, Caplan A, Mancini D, Levine T, Ritsch M, Stevenson L. 24th Bethesda Conference: Cardiac Transplantation. Task Force 3: Recipient Guideline/Prioritization. J Am Coll Cardiol. 1993;22:21-31. [Medline] [Order article via Infotrieve]
36. Kjellstrand C, Ericsson F, Traneus A, Noree L, Lins L. The wish for renal transplantation. ASAIO Trans. 1989;35:619-621. [Medline] [Order article via Infotrieve]
37. Westlie L, Umen A, Nestrud S, Kjellstrand C. Mortality, morbidity and life satisfaction in the very old dialysis patient. Trans Am Soc Artif Organs. 1984;30:21-30.
38. Rodin J, Ickovics J. Women's health: review and research agenda as we approach the 21st century. Am Psychol. 1990;45:1018-1034. [Medline] [Order article via Infotrieve]
39. Puentes-Markides C. Women and access to health care. Soc Sci Med. 1992;35:619-626.
40. deVaus D. Gender differences in religion: a test of the structural location theory. Am Sociol Rev. 1987;52:472-481.
41. Wilson B. Women and poverty: a demographic overview. Women Health. 1988;13:21-40.
42. Travis C. Women and Health Psychology: Biomedical Issues. Hillsdale, NJ: Erlbaum; 1988.
43.
Lurie N, Slater J, McGovern P, Ekstrum J, Quam L,
Margolis K. Preventive care for women: does the sex of the
physician matter? N Engl J Med. 1993;329:478-482.
44. Callender C. Organ donation in blacks: a community approach. Transplant Proc. 1987;19:1551-1554. [Medline] [Order article via Infotrieve]
45. Ozminkowski R, Friedman B, Taylor Z. Access to heart and liver transplantation in the late 1980s. Med Care. 1993;31:1027-1042. [Medline] [Order article via Infotrieve]
46. Rovelli M, Palmeri D, Vossler E, Bartus S, Hull D, Schweizer R. Noncompliance in renal transplant recipients: evaluation by socioeconomic groups. Transplant Proc. 1989;21:3979-3981. [Medline] [Order article via Infotrieve]
47. Wasserman K, Hansen J, Sue D, Whipp B. Principles of Exercise Testing and Interpretation. Philadelphia, Pa: Lea & Febiger; 1987.
48.
Escarce J, Epstein K, Colby D, Schwartz JS.
Racial differences in the elderly's use of medical procedures
and diagnostic tests. Am J Public Health. 1993;83:948-954.
49.
Gonwa T, Morris C, Mai M, Husberg B, Goldstein R,
Klintmalm G. Race and liver transplantation. Arch
Surg. 1991;126:1141-1143.
50. Kasiske B, Neylan J, Riggio R, Danovitch G, Kahana L, Alexander S, White M. The effect of race on access and outcome in transplantation. N Engl J Med. 1991;324:302-307. [Medline] [Order article via Infotrieve]
51. Kjerulff K, Guzinski G, Langenberg P, Stolley P, Adler Moye NE, Kazandjian V. Hysterectomy and race. Obstet Gynecol. 1993;82:757-764. [Medline] [Order article via Infotrieve]
52.
Moore R, Stanton D, Gopalan R, Chaisson R.
Racial differences in the use of drug therapy for HIV disease in
an urban community. N Engl J Med. 1994;330:763-768.
53.
Whittle J, Conigliaro J, Good C, Lofgren R.
Racial differences in the use of invasive cardiovascular
procedures in the Department of Veterans Affairs Medical
System. N Engl J Med. 1993;329:621-627.
54.
Held P, Pauly M, Bovbjerg R. Access to kidney
transplantation: has the United States eliminated income and racial
differences? Arch Intern Med. 1988;148:2594-2600.
This article has been cited by other articles:
 |
|
 |
|
  E. S. Weiss, J. G. Allen, N. D. Patel, S. D. Russell, W. A. Baumgartner, A. S. Shah, and J. V. Conte The Impact of Donor-Recipient Sex Matching on Survival After Orthotopic Heart Transplantation: Analysis of 18 000 Transplants in the Modern Era Circ Heart Fail, September 1, 2009; 2(5): 401 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Csete Gender Issues in Transplantation Anesth. Analg., July 1, 2008; 107(1): 232 - 238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C Oniscu, A. A H Schalkwijk, R. J Johnson, H. Brown, and J. L R Forsythe Equity of access to renal transplant waiting list and renal transplantation in Scotland: cohort study BMJ, November 29, 2003; 327(7426): 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Schaubel, D. E. Stewart, H. I. Morrison, D. L. Zimmerman, J. I. Cameron, J. J. Jeffery, and S. S. A. Fenton Sex Inequality in Kidney Transplantation Rates Arch Intern Med, August 14, 2000; 160(15): 2349 - 2354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Petrie, N. F. Dawson, D. R. Murdoch, A. P. Davie, and J. J. V. McMurray Failure of Women's Hearts Circulation, May 4, 1999; 99(17): 2334 - 2341. [Full Text] [PDF]
|
|
|
|
|
|
H. L. Horton, R. A. Marinchak, S. J. Rials, and P. R. Kowey Gender Differences in Device Therapy for Malignant Ventricular Arrhythmias Arch Intern Med, November 27, 1995; 155(21): 2342 - 2345. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||