The Houston Vision Assessment Test (HVAT): an assessment of validity. The Cataract Outcome Study Group

PURPOSE

Cataract surgery is one of the most successful procedures in medicine, and outcome is typically measured by a single factor - improvement in visual acuity. Health-related functional outcome testing, which quantifies the patient's self-reported perception of impairment, can be integrated with objective clinical findings. Based on the patient's self-assessed lifestyle impairment, the physician and patient together can make an informed decision on the treatment that is most likely to benefit the patient.

METHOD

A functional outcome test (the Houston Vision Assessment Test - HVAT, copyrighted 1990, 1992) was evaluated in a cataract population of 149 patients from seven study centers. Test results were correlated with objective ophthalmic endpoints. The HVAT divides an estimated total impairment into subcomponents of Visual Impairment (correctable by cataract surgery) and non-visual Physical Impairments (co-morbidities not affected by cataract surgery).

RESULTS

In this prospective study, the average Visual Impairment score improved by 19 points (65%) following cataract extraction (the mean HVAT Visual Impairment score was 29 points before surgery and 10 points at 5 months post-surgery, a change of 65%). Physical Impairment remained unchanged by surgery. Visual acuity was a poor predictor of Visual Impairment.

CONCLUSIONS

The HVAT has 11 simple questions. It may be self-administered and is available on the Internet: http://www.DHAC.com. The physician may err if his decision in favor of cataract surgery is based only on visual acuity. The HVAT has the potential to guide the decision-making process between patient and physician.

Outcomes research and community health information systems

The developments in outcomes research and community health information systems are pursuing essentially parallel courses in this country, and yet integrating the lessons from each other could yield more informed health policy decisions. This article provides the linkage between these two developing fields, discusses indicators of community health, describes several approximations to an outcomes-based community health information system in this country, and specifies a model (POPULIS) of such a system from Canada. Although POPULIS can serve as a prototype for health policy decision making, the outcomes-based community health information system, which can be constructed in this country, is constrained by the pluralistic system of health care we have and the lack of any centralized planning about health.

A framework for assessing the effectiveness, efficiency, and equity of behavioral healthcare

OBJECTIVE

To evaluate the effectiveness, efficiency, and equity of behavioral healthcare and to guide an assessment of the current state of the art of behavioral health-oriented health services research.

STUDY DESIGN

The framework is grounded in previous conceptual work by the authors in defining a prevention- and outcomes-oriented continuum of healthcare and in identifying and integrating the concepts and methods of health services research and policy analysis for assessing healthcare system performance.

PATIENTS AND METHODS

The defining assumptions are that (1) the denominator for behavioral healthcare services must encompass a look at the population, not just the patients, who manifest behavioral health risks; and (2) the delivery system to address these needs must extend beyond acute, treatment-oriented services to include both primary prevention and aftercare services for chronic relapsing conditions.

RESULTS

Current policy and practice in behavioral healthcare reveal the absence of a comprehensive, coordinated continuum of care; substantial variation in policy and financial incentives to encourage such development; and poorly defined or articulated outcome goals and objectives. The current state of the art of research in this area reflects considerable imprecision in conceptualizing and measuring the effectiveness, efficiency, and equity criteria. Further, these 3 criteria have not been examined together in evaluating system performance.

CONCLUSIONS

The first era of behavioral healthcare focused on cost savings in managed care alternatives; the second is focusing on quality and outcomes; a third must consider the issues of equity and access to behavioral healthcare, especially for the most seriously ill and vulnerable, in an increasingly managed care-dominated public and private policy environment.

A comparative cost analysis of newborn screening for classic congenital adrenal hyperplasia in Texas

OBJECTIVE

Texas mandates a two-test newborn screening program for congenital adrenal hyperplasia (CAH): one test at birth and a second test at approximately one to two weeks after birth. The authors compared the dollar cost of detecting infants with CAH clinically and through the screening program.

METHODS

The authors estimated the costs of screening newborns in 1994 for CAH, including resources used by the Texas Department of Health and the broader cost to society.

RESULTS

Fifteen infants with classic CAH were diagnosed in Texas in 1994 among 325,521 infants born (1:21,701 cumulative incidence). Seven infants were detected clinically and the others were detected through screening, six on the first screen and two on the second screen. The first screen identified all previously undetected infants with severe salt-wasting CAH. The cumulative cost to diagnose the seven infants detected clinically was $79,187. The incremental costs for the screening program were $115,169 per additional infant diagnosed through the first screen and $242,865 per additional infant diagnosed through the second screen.

CONCLUSIONS

If the goal is early diagnosis of infants with the severe salt-wasting form of CAH, a single screen is effective. If the goal is to detect infants with the simple virilizing form of the disorder who may benefit from early treatment, the second screen is necessary, but it is not as cost-effective as the first screen.

What is outcomes research and what can it tell us?

In 1989, Congress enacted P.L. 101-239, establishing the Agency for Health Care Policy and Research (AHCPR). Since then, practically every specialty society has developed practice guidelines for its practitioners, and every health care organization has embarked on some outcomes research. Outcomes research has become a fashion, meaning all things to all people, and runs the risk of becoming meaningless. This article attempts to reduce the confusion by clarifying what outcomes research is not delineating its several levels, along with the methods, tools, and examples appropriate to each level. The contributions of outcomes research to health care delivery innovation to date have been modest. Ultimately, we need community health information systems, which have not only the structural and process variables but also include the outcome results, to guide decision making with regard to the health of entire communities and the appropriate investment of resources to improve health.

Physiological levels of testosterone kill salmonid leukocytes in vitro

Adult spring chinook salmon (Oncorhynchus tshawytscha) elaborate high plasma concentrations of testosterone during sexual maturation, and these levels of testosterone have been shown to reduce the salmonid immune response in vitro. Our search for the mechanism of testosterone's immunosuppressive action has led to the characterization of an androgen receptor in salmonid leukocytes. In the present study we examined the specific effects that testosterone had on salmonid leukocytes. Direct counts of viable leukocytes after incubation with and without physiological levels of testosterone demonstrate a significant loss of leukocytes in cultures exposed to testosterone. At least 5 days of contact with testosterone was required to produce significant immunosuppression and addition of a "conditioned media" (supernatant from proliferating lymphocytes not exposed to testosterone) did not reverse the immunosuppressive effects of testosterone. These data lead us to conclude that testosterone may exert its immunosuppressive effects by direct action on salmonid leukocytes, through the androgen receptor described, and that this action leads to the death of a significant number of these leukocytes.

Managed care and community-oriented care: conflict or complement?

Motivated by the need for fundamental change, reform of the health care delivery system is continuing despite the recent failure of national initiatives. One aspect of this reform is the restructuring of managed care systems to include low-income, at-risk populations in their health delivery program: It is a move that threatens current "safety-net" providers, which already serve these populations with programs that combine public health and traditional primary care. This paper explores this potential conflict by providing a brief history and comparison of the main features of the community-oriented primary care (COPC) and health maintenance organization (HMO) models. The authors provide a frame-work that contrasts the structure, process, and outcome characteristics of these two models, delineating key similarities and differences. The frame-work is used in profiling a service delivery system model that integrates the two systems and in discussing issues related to operationalizing the proposed integration.

Characterization of an androgen receptor in salmonid lymphocytes: possible link to androgen-induced immunosuppression

Cytosol of rainbow trout (Oncorhynchus mykiss) leukocytes demonstrated specific and saturable binding of [3H]testosterone (Kd = 0.99 +/- 0.17 nM and Bmax = 30.4 +/- 4.9 fmol/mg protein; based on a total of 6 determinations from three different cytosolic pools). Specific binding of [3H]testosterone was high in leukocytes and other tissues with known androgen binding affinity (plasma, skin, and liver) and low in other tissues (heart, muscle, and red blood cells). Specific binding of [3H]testosterone was displaced by testosterone and dihydrotestosterone. Androstenedione displaced 50% of specifically bound [3H]testosterone between 10- and 100-fold excess, while 17-alpha-methltestosterone, 11-ketotestosterone, and progesterone displaced 50% of specifically bound [3H]testosterone between 100- and 500-fold excess. Cortisol, 17 beta-estradiol, 17 alpha,20 beta-dihydroxyprogesterone, the synthetic androgen mibolerone, and the synthetic estrogen ethenylestradiol did not displace [3H]testosterone binding, even at 500-fold excess. Treatment of cytosol with proteolytic enzyme significantly reduced the specific binding of [3H]testosterone. HPLC analysis determined that [3H]testosterone was not metabolized during assay incubation with cytosol. These data strongly suggest that androgen receptors exists in salmonid leukocytes and support the hypothesis that these receptors play a role in androgen induced immunosuppression.

Avoidable hospitalizations and socio-economic status in Galveston County, Texas

Population-based hospitalization rates for preventable conditions are derived for a low-income population in Galveston County, Texas using discharge data from four area hospitals and demographic data from the U.S. Census. Comparisons are made with similar rates for the general populations of two eastern states (Maryland and Massachusetts) and New York City. Results support the hypothesis that low-income persons lack access to primary care, leading to higher rates of hospitalization for preventable conditions. Alternative explanations for differences found are examined.

A method of developing and weighting explicit process of care criteria for quality assessment

The use of explicit criteria to evaluate how well processes of care conform to accepted standards is a key method of quality assessment. Synthesizing four decades of literature, we devised an inexpensive, 6-step method of developing reliable, content-valid, explicit process criteria. This paper describes the method using a set of congestive heart failure criteria. In step 1 of the Criteria Development Method, criteria are derived from state-of-the-art clinical literature. In step 2, criteria are refined by expert panels. In this study, panelists refined the items by mail in a three-round Delphi process. In step 3, decisions about unit-or differential item weighting are made; we derived differential item weights from the panelists' third-round ratings. Step 4 consists of flagging items which may yield little information, i.e., consensus items of low import, and nonconsensus items. Numeric flags were computed using third-round median ratings and their interquartile ranges. Selection of a scoring method to summarize scores and communicate results is done in step 5. In step 6, chart reviewers are trained, inter-rater reliability is measured, and items with poor reliability are culled. This straightforward developmental method can be used to devise explicit process criteria for use in ambulatory or hospital settings and to evaluate care delivered by different types of providers. The method yields reliable criteria representing accepted standards of current clinical practice. This high content validity is a sine qua non for convergent and predictive validity, both of which must be demonstrated in empirical studies in which the criteria are compared against external yardsticks.

Multiple molecular forms of gonadotropin-releasing hormone in the brain of an elasmobranch: evidence for IR-lamprey GnRH

These studies investigated brains of skate, Raja erinacea (order Rajiformes, class Chondrichthyes), for gonadotropin-releasing hormone (GnRH) peptides by chromatograph and immunoreactivity with region-specific antisera raised against mammalian GnRH and lamprey GnRH. The region-specific antibody to lamprey GnRH-I was produced following conjugation to bovine serum albumin using the bis-diazotized benzidine method. This antibody was characterized by assaying a range of increasing dilutions of the known vertebrate GnRHs, as well as analogs to lamprey GnRH-I. Two analogs, lamprey [Phe2]GnRH-I and lamprey [Leu7]GnRH-I, were synthesized by solid phase peptide synthesis using a benzhydrylamine resin as the supporting medium and purified by chromatography. This antibody demonstrated less than 0.01% cross-reactivity with all GnRH peptides tested, suggesting a highly specific antibody with a region of amino acids 2-8 that appears essential for binding. In the skate brain, five immunoreactive (IR) GnRH forms were detected, four of which eluted in the same positions as synthetic mammal and chicken GnRH-I (which coelute): lamprey GnRH-I, salmon and chicken GnRH-II, and one that was an unidentified form. A minor peak coeluted with lamprey GnRH-III. The major form in the skate brain is considered to have eluted with synthetic mammalian GnRH. These studies confirm an earlier report of an IR-mammalian GnRH peptide and provide new evidence for IR-lamprey GnRH in the brain of an elasmobranch.

Testosterone alters the immune response of chinook salmon, Oncorhynchus tshawytscha

We examined the effects of the high concentrations of sex steroid hormones detected in the plasma of sexually maturing chinook salmon (Oncorhynchus tshawytscha) on the functioning of the immune system. Leukocytes from the anterior kidney of juvenile spring chinook salmon were incubated in the presence of steroid, and their ability to form specific antibody-producing cells (plaque-forming cells) was used as a measure of immunocompetence. Testosterone and cortisol, but not 17 beta-estradiol or aldosterone, were found to significantly reduce the plaque-forming response in vitro. Testosterone and cortisol administered together had a significantly greater effect than did either administered alone. The plaque-forming response was significantly affected by season. The magnitude of the steroid-induced immunosuppression was reduced during winter and increased again in spring.

Occupational physical activity and other risk factors for preterm birth among US Army primigravidas

We examined the relation of occupational physical activity to the risk of preterm birth among US Army active-duty primigravidas between 1981 and 1984 using 604 cases (preterm deliveries; less than or equal to 37 weeks gestation) and 6,070 controls (term and post-term deliveries). Women employed in the highest physical activity levels had increased odds of preterm delivery ranging from 1.69 to 1.75. The relation was not changed by adjustment for the effects of age, marital status, socioeconomic status, or education. Missing data suggest cautious interpretation.

The validity of single-item, self-assessment questions as measures of adult physical activity

Individual energy expenditure (kcal/kg/day) was calculated from a detailed set of questions from the Health Promotion and Disease Prevention Supplement of the 1985 National Health Interview Survey. Responses to three single-item, self-assessment questions were compared to the energy expenditure variable to test criterion validity. Spearman's correlation coefficient revealed moderate correlations between energy expenditure and corresponding levels of self-assessed leisure-time physical activity for each single-item question (r = 0.14 to 0.41). For purposes of measuring prevalence of physical activity, the energy expenditure variable was used to categorize individuals into activity levels. The single-item questions were found to have Spearman's correlations with the categorical measures ranging from 0.11 to 0.37 for leisure-time activities. Generally higher correlations were found for males and younger age groups (18-34 years). The relationships were interpreted as being weak relative to an expected correlation of 0.75 for criterion validation. However, the single-item questions show promise for obtaining proxy estimates of the degree of leisure-time physical activity in a population.

An analysis of ambulatory care quality assessment research

This article offers a framework that distinguishes quality as an attribute of the processes of medical care from effectiveness as the results of that care. The review of process-outcome studies in the light of this framework reveals not only a confusion of concepts but a variety of methodological and measurement problems. Among the measurement problems is the lack of assessment of the reliability and validity of the process and outcome measures in most quality studies in ambulatory medical care. This article proposes a minimum set of elements to be considered in developing a valid quality assessment method.

A reassessment of the additive scoring of health practices

Over the past 20 years, investigators have been refining the connection between behavioral practices, popularly known as health habits, and health status. Repeated study has demonstrated that the number of healthful practices, regardless of which ones are adopted, provides a reliable predictor of mortality. Few studies, however, have questioned the validity of summing such diverse practices as smoking and physical activity together to form a single practice score. The purpose of this study was to raise some questions about this widely adopted scoring procedure and to reasses the problems connected with its use. Data are drawn from the Texas Behavioral Risk Factor Survey of 1982. The approach contrasts practice profiles formed from all possible combination of practices, representing full information about them, and the scores produced by collapsing practices onto a single dimension. Special attention is given to the meaningfulness of the information lost in the scoring process and to the implications this may have for the health practice-to-health status relationship.

Problems in estimating the prevalence of physical activity from national surveys

Health policy in the United States has paid scant notice to physical activity until recently. This current policy focus on physical activity has revealed not only that there is less than adequate data about it but also that the single survey questions used for the purpose of measuring its prevalence should be interpreted carefully. A case in point is the example presented in this article, which gives estimates of physical activity prevalence levels for women of child-bearing ages from several National Center for Health Statistics (NCHS) surveys. The amount of physical activity reported, and thus the proportion of women at risk for various diseases due to limited activity, depends on how the question is asked and the type of responses offered as options. Various questions used in three different NCHS surveys produced prevalence estimates of limited physical activity levels ranging from 3.9 to 39.1%. These findings have important implications for survey assessments of physical activity at all policy levels.

Ischemic heart disease: footprints through the data

This study addresses the question of whether the rise in ischemic heart disease mortality has been just "a paper epidemic" as asserted in an earlier issue of this journal. Age-standardized death rates, proportions expected to die, mean ages at death, and cause specific contributions to changes in overall life expectancies were calculated for acute and chronic ischemic heart disease and for males and females for the years 1931 to 1980 using published vital statistics data. These multiple analyses reveal: a true epidemic of acute ischemic heart disease has occurred, affecting males exclusively or to a greater degree than females and it is now on the decline, fairly stable and more nearly comparable mortality for both males and females for chronic ischemic heart disease, and continuing problems of classification obscure the true levels of mortality for both the acute and chronic entities.

The independent contributions of socioeconomic status and health practices to health status

The objective of this study was to determine whether the much-repeated finding of a relationship between socioeconomic status and health status is explained by individuals' health practices. The investigation was carried out using data tapes from the 1977 Health Interview Survey in which a one-third subsample of adults was asked a series of questions related to the seven nonmedical health practices identified in the Alameda County Study. The group selected for analysis comprised 15,892 white, responding adults. With age controlled statistically, perceived health status was found to be associated with socioeconomic status, whether the indicator was educational level, family income, or occupation, and to number of positive health practices. When number of health practices, in addition to age and other socioeconomic indicators was controlled for, the association was still positive and significant. The finding of an independent contribution by socioeconomic status to health status emphasizes that individual health habits are not the only influence on health status.

Agreement of self-reported and physiologically estimated fitness status in a symptom-free population

A population of 204 symptom-free clerical and white-collar employees who volunteered for a corporate-based health promotion program was studied to determine agreement between self-reported and physiologically determined fitness status. Physiologic fitness was estimated using exercise treadmill performance. There were statistically significant differences between self-reported and treadmill-estimated fitness status. The study population rated their fitness substantially higher than that estimated by treadmill performance. Self-reported fitness was found to be poorly correlated with physiologic fitness, indicating independent distributions. These relationships persisted across gender. The results question the use of self-reported estimates of physical health in health services research.