More prevalent severe obesity may explain black/white difference in stage at diagnosis of breast cancer

Black women are typically diagnosed with breast cancer at a later stage than white women, putting them at greater risk of dying from the disease. Higher rates of severe obesity among black women compared with white women may be a major factor in this disparity, concludes a study supported in part by the Agency for Health Care Policy and Research (HS06910). It shows that severely obese women are more than three times as likely to be diagnosed with cancer at later, less treatable stages of the disease.

In this study, black women were twice as likely as white women to be moderately overweight and more than six times as likely to be severely obese. Also, black women were twice as apt to be diagnosed when the tumor was larger or had spread to nearby lymph nodes. Overall, severe obesity (40 percent over average weight) accounted for about one-third of the racial difference in stage of breast cancer diagnosis, even after accounting for other factors such as the patient's age, socioeconomic status, history of breast cancer screening, and certain lifestyle and reproductive factors. Being slightly or moderately overweight had no significant effect on stage of diagnosis. This suggests there may be a threshold effect, with a cutoff of severe obesity affecting the stage at which breast cancer is diagnosed, explains Beth A. Jones, Ph.D., M.P.H., the study's principal investigator.

Dr. Jones and her colleagues at Yale University's School of Medicine analyzed body mass index, socioeconomic status, and a number of other factors, including medical care variables, as possible predictors of stage and diagnosis of breast cancer among 145 black women and 177 white women diagnosed with new cases of breast cancer in Connecticut in the late 1980s. Excess body weight has been associated with increased levels of the hormone, estrogen, which could speed the development of breast cancer. The effect of severe obesity on stage at diagnosis in this study was limited to women whose tumors were estrogen receptor-positive.

It has been thought that black women are diagnosed with breast cancer at later stages due to factors such as lower socioeconomic status and less access to health care and screening mammography. But accounting for these factors did not diminish the role of severe obesity, according to the researchers.

They note that women who are severely overweight may require special mammography procedures, such as larger films and additional views, for accurate screening, and these additional measures may not always be done. The researchers conclude, however, that the negative impact of severe obesity on stage at diagnosis is more likely to be mediated by endocrinologic (hormonal) processes than by screening processes. The good news is that losing weight, although difficult, can increase the odds that a woman who contracts breast cancer will be diagnosed at an earlier, more treatable stage.

More details are in "Severe obesity as an explanatory factor for the black/white difference in stage of diagnosis of breast cancer," by Dr. Jones, Stanislav V. Kasl, Ph.D., Mary G. McCrea Curnen, M.D., Dr.P.H., and others, in the September 1997 American Journal of Epidemiology 146(5), pp. 394-404.

Return to Contents

Outcomes/Effectiveness Research

Using less expensive antibiotics to treat children's ear infections could save money without compromising care

Treating common ear infections in children with antibiotics such as amoxicillin instead of more costly choices could save millions of dollars a year without changing recovery rates, according to researchers supported by the Agency for Health Care Policy and Research (HS07816). Their study, which looked at children covered under Colorado's Medicaid program, was funded as a component of AHCPR's Pharmaceutical Outcomes Research Program.

Middle ear infection—or otitis media—is the most frequent reason for giving antibiotics to children in the United States. Doctors can select from a large number of antibiotics, but no single antibiotic has been found to be superior to another for treating this condition. However, costs vary widely, from $2.94 to $62.80, for example, for a 10-day course of antibiotics for a 19- to 24-month-old child. According to the investigators, if in 1992, only half the prescriptions in the study for cefaclor, cefixime, and amoxicillin/clavulanate had been written instead for amoxicillin alone, Colorado's Medicaid program would have saved $399,412.

The researchers found that more expensive antibiotics, such as amoxicillin/clavulanate or cephalosporins, accounted for only 30 percent of the prescriptions written for the studied population but up to 77 percent of the more than $2 million spent for medications. In contrast, amoxicillin and other less expensive antibiotics, which worked just as well, accounted for 67 percent of the prescriptions but only 21 percent of the costs.

There were no financial incentives or disincentives to influence physicians to choose one antibiotic over another. In general, physicians in hospital outpatient departments and community health centers tended to prescribe less expensive antibiotics, and office-based physicians generally prescribed more expensive antibiotics.

The study, conducted by Stephen Berman, M.D., and his colleagues at the University of Colorado Health Sciences Center, looked at 12,381 children 13 years of age and younger who were enrolled in Colorado's fee for-service Medicaid program and were treated for a new episode of acute otitis media in 1991 and 1992.

For more information, see "Otitis media-related antibiotic prescribing patterns, outcomes, and expenditures in a pediatric Medicaid population," by Dr. Berman, Patricia J. Byrns, M.D., Jessica Bondy, M.H.A., and others, in the October 1997 issue of Pediatrics 100(4), pp. 585-592.

Low birthweight PORT publishes latest findings

Many maternal behavioral and biological factors contribute to preterm birth and low birthweight, which in turn can increase infants likelihood of disease and death. The Patient Outcomes Research Team on Prevention of Low Birthweight in Minority and High-Risk Women is supported by the Agency for Health Care Policy and Research (PORT contract 290-92-0055) to find ways to prevent and manage low birthweight.

In a recent interview, PORT leader Robert L. Goldenberg, M.D., of the University Alabama at Birmingham, discussed the relationship of intrauterine infection and preterm birth. PORT researchers have also published findings from six other studies. The interview and the studies are summarized here.

Goldenberg, R.L. (1997, June). "When infection causes preterm labor." OBG Management, pp. 55-68.

In this interview, Dr. Goldenberg notes that 20 percent of all preterm births are related to intrauterine infection. He discusses the relationship between bacterial vaginosis and preterm labor, the use of cytokines (proteins released during an immune response) and other substances as biochemical markers for risk of preterm delivery, and the efficacy of employing antibiotics to reduce the incidence of preterm delivery in high-risk women. Dr. Goldenberg points out that about 80 percent of preterm births prior to 30 weeks' gestation are related to intrauterine infection. As gestational age advances, the percentage of preterm births related to infection declines.

The most common pathogens involved in intraamniotic infections that lead to preterm labor are Ureaplasma urealyticum, followed by Gardnerella vaginalis, Mycoplasma hominis, and peptostreptococcus. However, it has not been shown that universal screening for these organisms would help reduce the rates of preterm labor and delivery. Dr. Goldenberg believes that their presence in the vagina most likely is not a risk factor for preterm delivery; it is only when these pathogens ascend to the uterus in large quantities that they can trigger preterm labor. He suggests that bacterial vaginosis screening makes sense in women who have had a previous pregnancy marked by preterm labor.

Peralta-Carcelen, M., Fargason Jr., C.A., Coston, D., and Dolan, J.G. (1997, July). "Preferences of pregnant women and physicians for 2 strategies for prevention of early-onset group B streptococcal sepsis in neonates," Archives of Pediatric and Adolescent Medicine 151, pp. 712-718.

Early-onset Group B streptococcal (GBS) disease is the most common cause of neonatal sepsis in the United States. About 15 percent to 40 percent of mothers carry GBS, but less than 2 in 1,000 newborns will contract the infection from their mothers. The researchers presented 86 pregnant women at a university obstetric clinic with a computerized, interactive decisionmaking model based on two strategies to minimize the risk of neonatal GBS sepsis: one suggested by the American Academy of Pediatrics (AAP) and one recommended by the American College of Obstetricians and Gynecologists (ACOG). They also presented a random sample of 40 pediatricians and 40 obstetricians with the same decisionmaking model. All were asked to rank the five criteria on which they based their decision: a wish to avoid diagnostic tests in healthy infants, the risk of infection in an infant, knowledge of maternal GBS status, cost, and the risk of anaphylactic shock in mothers treated with antibiotics.

The AAP policy recommends screening all mothers for GBS at 28 week's gestation. Mothers who screen positive and have any risk factor are treated with intrapartum antibiotics. ACOG recommends treating mothers with intrapartum antibiotics only if they have risk factors such as prolonged rupture of membranes, fever, or preterm labor (which can indicate vaginal infection).

The researchers found that 81 percent of women and 65 percent of pediatricians, but only 15 percent of obstetricians, studied preferred the AAP strategy. Not surprisingly, 85 percent of obstetricians preferred the ACOG policy. Maternal and physician priorities underlying these decisions varied. Mothers, pediatricians, and obstetricians all ranked risk of infection in an infant as the most important criterion in their decisions. But women ranked knowing their GBS status as more important than did the physicians. Also, 88 percent of women, 73 percent of pediatricians, and 43 percent of obstetricians thought physicians should use the computer-based decisionmaking model employed in this study to involve patients in choosing a GBS prevention strategy.

Gardner, M.O., Goldenberg, R.L., Cliver, S.P., and others (1997). "Maternal serum concentrations of human placental lactogen, estradiol, and pregnancy specific B1-glycoprotein and fetal growth retardation." Acta Obstetricia et Gynecologica Scandinavica Suppl. 165 (76), pp. 56-58.

Fetal growth restriction (FGR) is associated with increased rates of infant disease and death. Women who are 18 week's pregnant and have high levels of human placental lactogen (hPL), estradiol, and pregnancy-specific B1-glycoprotein (SP1) have less risk of FGR than pregnant women with lower levels of these hormones. Unfortunately, assays for these hormones lack the specificity and sensitivity necessary to be clinically useful as screening tests for FGR, even for women at increased risk for this problem, concludes the Low Birthweight PORT.

The researchers analyzed stored blood samples obtained from 200 women with risk factors for FGR to correlate blood concentrations of hPL, estradiol, and SP1 with FGR and found no significant differences in the prevalence of FGR among women with the lowest levels of these hormones. However, only 4.5 percent of infants whose mothers had levels of all three hormones in the highest quartile were diagnosed with FGR compared with 33 percent of the remaining infants. Women with the highest quartile of estradiol levels at 18 week's gestation had half the risk of FGR of women in the lower three quartiles (16 percent vs. 34 percent).

The prevalence of FGR in women with hPL levels in the highest quartile was one-third that of women with levels in the lower three quartiles (12 percent vs. 35 percent). The prevalence of FGR among women with levels of SP1 in the highest quartile was less than half that of women with levels in the lower three quartiles (14 percent vs. 35 percent). These findings indicate that high levels of these hormones are related to a lower risk of FGR but that low levels do not predict FGR.

Tamura, T., Goldenberg, R.L., Johnston, K.E., and others (1997). "Serum concentrations of zinc, folate, vitamins A and E, and proteins, and their relationships to pregnancy outcome." Acta Obstetricia et Gynecologica Scandinavica Suppl 165(76), pp. 63-70.

A pregnant womans blood concentrations of zinc, vitamins A and E, and some proteins—all of which are markers of nutritional status—do not significantly correlate with fetal growth retardation or birthweight. On the other hand, high levels of the protein alpha-2-macroglobulin, which is believed to play a vital role in immunological and inflammatory reactions, are found in pregnant women who deliver growth-retarded infants, according to this study. The researchers measured serum levels of zinc, folate, vitamins A and E, and proteins: alpha-2-macroglobulin, retinol-binding protein (RBP), prealbumin, and albumin, from the serum samples of 289 high-risk pregnant women selected from the 1,545 women who participated in a study in Birmingham, Alabama, between 1986 and 1988. They correlated these laboratory values with birthweight and fetal-growth retardation, Apgar score of infants, and maternal infections around the time of delivery.

Concentrations of the protein alpha-2-macroglobulin were higher in white than in black women, in smokers than in nonsmokers, and in women with a higher body mass index (BMI) than those with a lower BMI. Higher serum levels of this protein were also associated with a significant decrease in infant birthweight. In addition, pregnant women with higher serum folate concentrations at either 18 or 30 week's gestation had fewer maternal infections, fewer growth-retarded infants, and more infants with higher birthweight and Apgar scores than women with lower levels of serum folate. Like folate, which prevents anemia during pregnancy, zinc is important for normal fetal growth and development during pregnancy. But unlike some other studies, this one did not show that a deficiency of maternal zinc results in fetal growth retardation.

Goldenberg, R.L., Cliver, S.P., Neggers, Y., and others (1997). "The relationship between maternal characteristics and fetal and neonatal anthropometric measurements in women delivering at term: A summary." Acta Obstetricia et Gynecologica Scandinavica Suppl 165(76), pp. 8-13.

Fetal ultrasound exams can be used to pinpoint when in pregnancy various maternal characteristics first have an effect on fetal measurements. For instance, black race, cigarette smoking, drug use, having a previous low birthweight infant, maternal hypertension, and being short or thin or failing to gain weight during pregnancy result in an infant weighing 3 to 10 oz (100 to 300 g) less at birth, according to this study. Cigarette smoking appears to have a relatively late effect on most fetal size measurements, but it does not affect head size. Maternal height has its greatest effect on femur length measurements and ultimately on crown-heel length, while the effect of maternal thinness and poor weight gain have their most early and predominant effects on fetal abdominal circumference measurements.

As early as 18 weeks, black fetuses have longer femur lengths than white fetuses. However, eventually black fetal femur lengths equal those of white fetuses. Head and abdominal circumferences of black fetuses are smaller than those of white fetuses beginning relatively early in pregnancy and remain so until term. This accounts for much of the birthweight difference between term black and white infants. Female fetuses are consistently smaller than male fetuses in nearly all measurements and weigh less at birth, despite having greater skinfold thickness (fat content). These findings are based on fetal ultrasound measurements at 18, 24, 30, and 36 week's gestation and a comparison of neonatal body measurements obtained at birth in relation to various maternal characteristics for 1,205 single mother-infant pairs. The data are from a prospective study of low-income, multiparous women with risk factors for fetal growth restriction who delivered at the University of Alabama at Birmingham between December 1985 and October 1988.

Hickey, C.A., Cliver, S.P., Goldenberg, R.L., and others (1997, June). "Low prenatal weight gain among low-income women: What are the risk factors?" BIRTH 24(2), pp. 102-108.

A mother's weight gain during pregnancy is a critical determinant of fetal growth, yet one in four nonobese women has low prenatal weight gain (22 pounds or less). For black women, having a mistimed or unwanted pregnancy or caring for more than one preschool child at home doubles the odds for low pregnancy weight gain. For white women, working more than 40 hours per week when employed increases nine-fold their odds for low pregnancy weight gain, finds this study by PORT researchers.

Not having one's own car for errands also increased a black woman's odds of having low pregnancy weight gain. Not having a car may be a marker for material deprivation, and caring for more than one preschool child could certainly result in increased energy expenditure during pregnancy. But both of these associations may, in turn, relate to the underlying issues of unintended pregnancies and pregnancy spacing, suggest the researchers.

Low-income women with risk factors for fetal growth restriction who participated in a prospective followup study between December 1985 and November 1988 were interviewed during their first prenatal visit to obtain sociodemographic information, reproductive history, reported pre-pregnancy weight, and to determine whether the pregnancies were wanted and/or planned. The women also completed a questionnaire at 24 to 26 weeks' gestation to assess household characteristics, income, transportation, employment, and other characteristics. Of the 1,518 women participating in this study, a total of 806 nonobese women and their infants (536 black, 270 white) were included in this analysis. Overall, about 25 percent of women gained 22 pounds or less (27 percent black, 22 percent white).

Goldenberg, R.L., Hickey, C.A., Cliver, S.P., and others (1997). "Abbreviated scale for the assessment of psychosocial status in pregnancy: Development and evaluation." Acta Obstetricia et Gynecologica Scandinavica Suppl 165(76), pp. 19-29.

A new short scale developed by the Low Birthweight PORT can be administered during an office visit to assess a pregnant woman's psychosocial status, which in turn can predict fetal growth retardation and birthweight. Currently, assessment of affective states (emotion and mood), cognitive factors (beliefs about self and the world), and stress requires the use of several separate scales or inventories, which are too time-consuming to administer in outpatient clinics. This study used factor analysis of data from five existing psychosocial scales administered to a low-income, high-risk population to develop a shorter scale to assess a woman's psychosocial status during pregnancy.

The abbreviated scale included 28 items grouped into 6 categories: negative affect, positive affect, positive self-esteem, low mastery/self-esteem, worry, and stress. The abbreviated scale was highly correlated with the 59-item pool of items from the 5 original scales. The distribution of fetal growth retardation and preterm delivery was similar for the abbreviated and combined scales. Within the low-income, high-risk group of women studied, the abbreviated scale predicted fetal growth restriction and birthweight but not gestational age at birth or preterm delivery.

Stroke prevention studies examine outcomes and appropriateness of carotid endarterectomy

Stroke is the third leading cause of mortality and leading cause of disability in the United States. Randomized controlled trials (RCTs) have shown carotid endarterectomy (surgical removal of plaque from the carotid artery) to be more effective than medical therapy for reducing stroke risk in symptomatic persons with 70 percent or more of the carotid artery blocked. However, these trials are performed on selected patients in carefully controlled situations. A recent study supported by the Agency for Health Care Policy and Research (HS08422) shows that this evidence only applies to patients and hospitals that are comparable to those of the trial participants.

The researchers found that the risk of death is substantially higher when a Medicare patient with more coexisting illnesses than a typical RCT participant undergoes carotid endarterectomy at a hospital with a mortality rate which exceeds that of a typical trial hospital. A second AHCPR-supported study conducted by the Stroke Prevention Patient Outcomes Research Team (PORT contract 290-91-0028) shows that between 34 percent and 82 percent of these procedures are performed for appropriate conditions, depending on how much surgical risk is considered acceptable.

Stukenborg, G.J. (1997, July). "Comparison of carotid endarterectomy outcomes from randomized controlled trials and Medicare administrative databases." Archives of Neurology 54, pp. 826-832.

The efficacy of carotid endarterectomy is uncertain for patients whose characteristics and circumstances are different from those enforced within RCTs, explains George J. Stukenborg, Ph.D., of the University of Virginia School of Medicine. His study shows that nearly half of Medicare patients (46 percent) underwent the procedure in 1989 at a hospital with a perioperative mortality rate which exceeded trial standards, and more than one-fourth of Medicare patients (29 percent) had coexisting conditions which would have excluded them from trial participation.

Medicare patients treated at hospitals with less than ideal mortality rates for the procedure had a substantially higher risk of death within 2 years. Medicare patients with coexisting illnesses that exceeded trial participation criteria (cancer, major organ illnesses, dementia, cardioembolic disorders, uncontrolled hypertension, unstable angina pectoris, or liver disease) had nearly twice the odds of death.

These findings are based on analysis of Medicare claims data for 41,493 patients aged 65 years and older who received carotid endarterectomy during 1989. At that time, many of the hospitals performed a low annual volume of the procedure. Only 151 (or 6 percent) of the 2,531 hospitals in which study participants received their surgery performed over 50 procedures in 1989; 864 hospitals (about 34 percent) performed between one and five carotid endarterectomy procedures in 1989. A total of 23 percent of Medicare patients who underwent the procedure in 1989 died or were readmitted to the hospital within 2 years of surgery. According to the author, administrative databases can be used to calculate outcomes for populations at large to complement results of RCTs and thereby give a more realistic picture of the actual effectiveness of certain medical procedures.

Matchar, D.B., Oddone, E.Z., McCrory, D.C., and others (1997, August). "Influence of projected complication rates on estimated appropriate use rates for carotid endarterectomy." Health Services Research 32(3), pp. 325-342.

Appropriate use rates for carotid endarterectomy may vary widely depending on what is considered acceptable surgical risk, according to a recent study by the Stroke PORT investigators and colleagues. When a strict 3 percent or lower rate of surgical complications (i.e., in-hospital complications or death) is considered acceptable, only about one-third (34 percent) of carotid endarterectomies are considered appropriate. However, the percentage increases three times (to 82 percent) when acceptable surgical risk is relaxed to 7 percent, a rate more comparable to clinical trials.

The project convened a geographically balanced expert panel of nine physicians chosen because of their clinical expertise and influence in their respective areas of practice. Before meeting, the panelists rated 1,222 possible reasons for carotid endarterectomy, assigning each a rating of 1 to 9 with 1 indicating that the carotid endarterectomy was clearly inappropriate, and 9 indicating that the surgery was appropriate. At a face-to-face meeting in Santa Monica, CA in 1991, the panel reduced the 1,222 possible indications to 611 and assigned an appropriateness rating to each that would vary with surgical risk. The panel also categorized surgical risk qualitatively into four categories. These were based on the predicted probability of a combined in-hospital stroke and death: low (less than 3 percent), elevated (3 to 5 percent), high (5 to 7 percent), and very high (more than 7 percent).

Investigators applied the ratings to the clinical indications for carotid endarterectomies performed on 1,160 patients in 12 American Medical Centers Consortium (AMCC) institutions. Charts of patients were abstracted, and clinical indications and perioperative complications were recorded. The indications for surgery were transient ischemic attack (mini stroke) for 51 percent of patients, stroke for 21 percent, and asymptomatic carotid artery stenosis for 24 percent. Most patients (82 percent) receiving carotid endarterectomy had high-grade lesions.

Total mortality was low, ranging from 0 to 4 percent. Major complications (death, stroke, intracranial hemorrhage, or heart attack) varied across medical centers from 2 percent to 11 percent. However, when the thresholds for operative risk were placed at the values first defined by the expert panel, only 33.5 percent of the procedures could be considered appropriate. The percentage of appropriate procedures increased to 82 percent with relatively minor changes to the risk categories: low (less than 7 percent), elevated (7 to 9 percent), high (9 to 11 percent), and very high (more than 11 percent).

This study highlights the difficulty in judging the appropriateness of procedures and the need to quantify surgical risk, especially for procedures like carotid endarterectomy, in which small differences in complication rates can be critical.

People over 45 with pneumonia report fewer symptoms than younger people but may be at higher risk of death or complications

Older people with pneumonia report fewer symptoms than those 44 years of age and younger, according to a study by the Pneumonia Patient Outcomes Research Team (PORT) supported by the Agency for Health Care Policy and Research (HS06468). However, unimpressive symptoms should not be misconstrued as indicating that older patients are less ill. In fact, regardless of the symptoms they report, older people are at higher risk of death and complications from pneumonia than younger people.

Older patients may treat some symptoms as less bothersome and thus be less likely to mention them. On the other hand, their lower reporting of some symptoms, such as fever and chills, may reflect actual physiological changes that accompany aging, explains PORT leader, Wishwa N. Kapoor, M.D., M.P.H., of the University of Pittsburgh. The research team correlated age with 5 respiratory symptoms and 13 nonrespiratory symptoms in 1,812 adults with clinical or x-ray evidence of community-acquired pneumonia at three university hospitals, one community hospital, and one staff-model health maintenance organization.

Overall, advanced age had the greatest effect on total symptom score. On average, patients aged 45 through 64 years, 65 through 74 years, and 75 years or older had 1.4, 2.9, and 3.3 fewer total symptoms than patients aged 18 through 44 years. Even after adjusting for increased severity of illness at the time of diagnosis and other patient factors, the mean number of symptoms from the youngest to oldest age group was 10.3, 8.9, 7.4, and 7.0, respectively.

For instance, older patients had nearly half the prevalence of pleuritic chest pain across the four age groups. The oldest patients had less than half the prevalence of three nonrespiratory symptoms (headache, myalgia or muscle pain, and inability to eat) of the youngest patients. Older patients were least apt to report symptoms related to fever (chills and sweats) and symptoms of pain (chest pain, headache, and myalgia). In contrast, there were small differences across age groups for respiratory symptoms (cough and shortness of breath).

See "Influence of age on symptoms at presentation in patients with community-acquired pneumonia," by Joshua P. Metlay, M.D., Ph.D., Richard Schulz, Ph.D., Yi-Hwei Li, Ph.D., and others, in the July 1997 Archives of Internal Medicine 157, pp. 1453-1459.

Visual functioning and patient preferences should guide decisions about cataract surgery

About 1.3 million cataract extractions are performed on Medicare beneficiaries in the United States each year. Often, measures of patients' visual acuity indicate only mild impairment when in reality cataracts are causing significant problems in daily life. Thus, physicians should rely more on patients visual functioning and personal needs and preferences than on measures of visual acuity when assessing the need for cataract surgery, according to the Cataract Patient Outcomes Research Team (PORT), led by Earl P. Steinberg, M.D., M.P.P., of The Johns Hopkins University School of Medicine, and supported by the Agency for Health Care Policy and Research (HS06280).

The researchers interviewed 47 patients (older than 50 years) randomly selected from Baltimore ophthalmologists' offices, who were scheduled to have routine cataract extraction within 3 months for an age-related cataract. The patients were asked to rate the effects of visual impairment on specific aspects of their daily lives, including work (job or home), leisure time activities, walking, driving, interacting with people, concentrating and remembering things, feelings of frustration, and feelings of depression. They were then asked to assign a preference value to their current vision on a scale ranging from 0 for complete blindness to 10 which would be perfect vision.

Patients' preference values for their preoperative vision were more closely related to problems in specific aspects of daily life—especially feelings of depression and problems in social interaction—than to visual acuity in the operative eye, better eye, or worse eye, or a weighted average of visual acuity in both eyes. More than 60 percent of patients reported frustration with visual impairment, and more than 25 percent reported that their current vision caused problems with driving, leisure-time activities, walking, and working.

Details are in "Preference values for visual states in patients planning to undergo cataract surgery," by Eric B. Bass, M.D., M.P.H., Stacey Wills, M.P.H., Ingrid U. Scott, M.D., and others in the July 1997 Medical Decision Making 17, pp. 324-330.

---

Return to Contents
Proceed to Next Section