Central nervous system involvement in multiple myeloma.
Explore the source record for details and available documents.
Adult↗
SEARCH · PubMed Health
Results for “Selection Bias”
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Biased selection under an experimental enrollment and marketing Medicare HMO broker.
All studies conducted to date suggest that nearly all Medicare HMOs have experienced favorable risk selection in their Medicare HMO enrollments. While there is little definitive empiric knowledge about the extent to which Medicare HMOs can and do encourage favorable selection through marketing and enrollment activities, it has been speculated that centralizing all such functions through an independent broker could reduce enrollment selection bias. In 1985, the Health Care Financing Administration initiated a three-year demonstration of a HMO broker model in Portland, Oregon, known as Health Choice, Incorporated (HCI). This study reports empiric findings that provide no evidence to support claims of the efficacy of an enrollment brokerage function in reducing Medicare HMO enrollment selection bias.
Contract Services↗
Consumer information and biased selection in the demand for coverage supplementing Medicare.
This study examines how the relationship between health insurance knowledge and the health status of health insurance consumers influences their decisions to purchase insurance coverage. Data from the federal Medicare health insurance program for the elderly in the United States are used. The basic Medicare program provides a limited amount of coverage for health care services obtained from any provider in the private fee-for-service (FFS) market. Beneficiaries of this program may choose to supplement the basic coverage which they receive by two mechanisms: either they may purchase private insurance designed to fill some of the gaps left by the federal program ('Medigap' policies), thereby remaining in the FFS market and preserving their choice of provider, or they may enroll in health maintenance organizations (HMOs), thereby leaving the FFS market and agreeing to use only those providers affiliated with the HMO, and in return receiving broader coverage at little additional out-of-pocket cost. The study was made possible by a unique data set which combines measures of beneficiary knowledge of Medicare coverage with measures of perceived health status, socio-economic characteristics, and insurance coverage choices for a sample of Medicare beneficiaries who participated in an educational workshop about their insurance coverage options. These data were used to estimate a multinomial logistic model of the determinants of insurance choices, where the options included the two listed above and a basic Medicare option. The study explicitly recognizes the interaction between insurance information and health status in health plan choice. These results show that knowledge of coverage does have a differential impact on the decision to purchase health insurance depending on health status. With a high level of knowledge, sicker beneficiaries are less likely to have basic Medicare alone, compared with HMOs or Medigap policies, while healthier beneficiaries are less likely to be enrolled in HMOs, compared with Medigap policies. This finding has important implications for the use of health status measures to adjust capitated payment formulas when knowledgable consumers have the option to enroll in HMOs or remain in the FFS environment. In the absence of health status adjusters for the HMO capitation payments, high levels of coverage knowledge may exacerbate inherent selection bias among these coverage options by healthier and sicker consumers of health insurance.
Aged↗
Biased selection under the senior health plan prior use capitation formula.
A widely acknowledged shortcoming of the current AAPCC capitation formula for Medicare HMOs is its inability to adjust capitation levels for differences in health status among enrolled groups. Prior use models have been proposed as one alternative to the current AAPCC risk classes. From 1984 through 1987, Senior Health Plan (SHP) participated in a HCFA-sponsored demonstration project in which capitation payments were determined by a prior use formula incorporating information on inpatient hospital days and Part B deductibles. This paper contains findings about selection bias in SHP enrollment from analyses of preenrollment reimbursements and postenrollment mortality rates of SHP enrollees compared with Medicare FFS beneficiaries and enrollees of other Medicare HMOs in the Minneapolis-St. Paul area.
Capitation Fee↗
Symptoms and selection bias: the influence of selection towards specialist care on the relationship between symptoms and diagnoses.
Observations with respect to the relationship between symptoms and diseases can seriously be biased by selection phenomena. This selection may occur from the general population, via consultation behavior, diagnostic and therapeutic activities of the general practitioner, and by referral. Relationships may be suggested and reproduced even if they do not exist in unselected populations, as a product of diagnostic routines. Correction for selection bias can only be achieved by choosing proper comparison groups. While this can be done in a general practice setting, this is almost impossible after referral, as is demonstrated in this paper. Surprisingly, the most unbiased estimation of the relationship between symptoms and diseases after referral can be made from patient groups that are referred for reason unrelated to the disease under study. Definitive answers for the general practitioner can only be provided by prospective studies from a primary care setting. In the meantime, however, biased relationships can be maintained by teaching knowledge derived from a specialist experience.
Anemia↗
Reporting and selection bias in case-control studies of congenital malformations.
Retrospective studies of congenital malformations frequently rely on exposures reported by study subjects. Differential error in exposure reporting by cases and controls, which has alternatively been referred to as "recall bias" and "reporting bias," may result in a biased effect measure. Some authors have attempted to avoid reporting bias by comparing exposures between two malformed groups, rather than between cases and nonmalformed controls. This approach, however, may introduce its own bias, which we call selection bias. Both reporting bias and selection bias are shown to be algebraically equivalent to bias arising from exposure misclassification. The magnitudes of these biases are compared for a range of plausible parametric values. The case-control design is sensitive to both differential reporting and selection bias, and the choice of study design involves balancing these two sources of bias.
Case-Control Studies↗
The role of health practices, health status, and prior health care claims in HMO selection bias.
To examine selection bias in terms of demographics, self-reported health practices, professionally evaluated health status, and prior health care utilization by employees and dependents, we examined health insurance choice between traditional insurance and a health maintenance organization (HMO) by central office employees of a large southwestern utility company. The HMO attracted a relatively younger and female population compared with the traditional plan. We found no difference in health practices and health status measures between the two groups. Consistent with recent studies, the HMO group experienced lower health claims cost the year prior to enrollment compared with persons who remained in the traditional plan. This difference was largely due to dependents' utilization-a factor that should be examined in future studies and considered by those structuring insurance premiums for HMOs and traditional plans.
Adult↗
Interpreting the term selection bias in medical research.
The issue of selection bias is often raised in the critical appraisal of medical studies, but it is often poorly defined and misunderstood. This paper, describes three common patterns of use of the term selection bias and their effects on study results. The three ways in which selection bias is used are related to 1) selection of representative subjects, 2) selection of subjects to exposures, and 3) selection of subjects at outcome. Avoidance of bias in the first of these issues, selection of representative subjects, enhances the ability to generalize a study's results. The other two uses of selection bias relate to the internal validity of studies. The selection of subjects to exposures without randomization in observational studies can distort results because of confounding variables. The selection of study subjects at outcome in case-control and cross-sectional studies can distort study findings if selection into the study is distorted according to exposure status. Readers of medical studies should understand the different implications of these uses to improve their critical evaluation of studies. Writers and discussants should be aware of these differences and provide clarifying details when they use the term.
Case-Control Studies↗
Evaluation of selection bias in a cross-sectional survey.
Selection bias is inherent in all occupational cohorts. Selection bias at entry has long been known and is commonly referred to as a "healthy worker effect." Less well appreciated is selection during the life of a cohort resulting from life-style factors (e.g., cigarette smoking); aging with accompanying chronic diseases, economic and demographic circumstances; and diseases that might result from exposures suffered by the cohort being studied, that influence whether individuals remain in a trade. These factors weigh differently at different times. Thus, at any point in time, "surviving" members of a cohort reflect an amalgam of selection factors. When such groups are studied in cross-sectional surveys there can be uncertainty whether clinical, radiological and physiological findings are necessarily representative for the trade or occupation as a whole. We analyzed the results of a large clinical field survey of long-term asbestos insulation workers to investigate whether the non-participants differed substantially from those who were examined. Five thousand three hundred and fifty-five (5,355) men, of an initial cohort of 17,800 established January 1, 1967, had reached 30 or more years from onset of their work by July 1, 1981. All were invited to come for examination. Two thousand and seventy-seven (2,077) came, and 3,278 did not. We questioned a sample of 1,393 non-responders to see why they failed to appear. The answers did not give evidence of significant health-related selection influence. Sickness only infrequently kept them away. We then followed both groups--those examined and those not examined--to the end of 1987 for their mortality experience. There was no great difference. The non-responders had somewhat fewer deaths overall and proportionately fewer of asbestos-associated cancers, such as mesothelioma and lung cancer. The results indicated that, in this cohort, there did not seem to be health-related selection bias that determined whether or not cohort members responded to invitations for examinations.
Asbestos↗
Case-control studies in clinical research: mechanism and prevention of selection bias.
The mechanism by which selection bias occurs in case-control studies is explained to an audience of clinicians using a simple conceptual framework and a graphical presentation. A case-control study consists in comparing the frequency of exposure in a group of subjects having the studied disease (the cases) relative to another group free of that disease (the controls). Cases and controls can be thought of as arising from a hypothetical cohort study. Thus, enrolled cases are a fraction F1 of the exposed who developed the disease plus a fraction F3 of the unexposed who developed the disease during a given period. Similarly, enrolled controls are a fraction F2 of the exposed who did not develop disease plus a fraction F4 of the unexposed who did not develop the disease. A selection process is inherent to the design of case-control studies but it leads to selection bias only when the ratio of F1 x F4/F2 x F3 is not equal to unity. Examples demonstrate the implication of sampling fractions for designing and interpreting case-control studies performed in clinical settings.
Case-Control Studies↗
Subject selection biases in clinical trials: data from a multicenter schizophrenia treatment study.
To evaluate subject selection biases in clinical trials, demographic characteristics (gender, race, and age) of subjects at different phases of evaluation for a multicenter maintenance trial in schizophrenia were examined. Six thousand twelve diagnostically appropriate subjects were screened for the study; of these, 1,320 met eligibility criteria and 528 (9% of the screened sample) entered the study. Women, blacks, and older subjects were more likely not to meet eligibility criteria; women and older subjects were more likely and blacks were less likely to refuse study participation. Overall, compared with the screened population, the sample of subjects who entered the study contained proportionately fewer women (33 vs. 43%), more blacks (48.5 vs. 41%), and fewer older subjects (mean age of the entered sample was 29.4 +/- 7.4 vs. 34.8 +/- 11.3 years for the screened population). Having identified these selection factors, a second goal was to assess the potential clinical relevance of selection biases of these magnitudes on clinical trials using models of hypothetical studies with different degrees of selection bias. These showed that selection biases would rarely change overall study outcomes to a clinically relevant degree. However, in our models, selection biases did limit the ability to make inferences about results for select small subgroups of the study population. Investigators should consider collecting data on the recruitment process to allow estimation of the effects of selection biases on the generalizability of their findings.
Acute Disease↗
Who says yes? Identifying selection biases in a psychosocial intervention study of multiple sclerosis.
The purpose of this work is to examine whether patients' sociodemographic and medical characteristics are associated with participation in a randomized controlled trial of two psychosocial interventions. After elimination of basic ineligibles from a registry of multiple sclerosis patients, 325 patients were sent a letter inviting participation in the randomized trial. Among those invited, 29% expressed an interest in participating, 19% scheduled an intake interview and 13% were actually randomized. chi 2 and logistic regression analyses were used to determine what factors were associated with successive stages of participation or non-participation in the study. Differential referral and participation rates could be traced in part to physician factors. With respect to patient factors, people were more likely to participate in the successive stages of recruitment if they had a higher median family income, lived a moderate distance from the hospital, and were disabled from working. Multivariate analyses of patient factors revealed that having a higher income and being disabled from work were the strongest predictors of participation, after adjusting for the effects of course of disease, disability status, and other sociodemographic predictors. Implications of these findings are discussed in terms of potential sources of bias in the selection of eligible patients, as well as the role the investigator may play in highlighting or underplaying this selection bias. Suggestions are made to minimize the bias-generating barriers.
Adult↗
Controlling for selection bias in the evaluation of Alcoholics Anonymous as aftercare treatment.
OBJECTIVE: The purpose of this research was to control for self-selection bias in the evaluation of Alcoholics Anonymous (AA) as aftercare treatment. Observational studies of alcoholism aftercare treatment are subject to selection bias whenever the self-selection process results in important differences in unobserved casemix dimensions between treatment groups. METHOD: The sample included 118 male veterans discharged from inpatient alcohol treatment, 85% of whom were followed-up at 3 months. Drinking outcomes were measured by self-reported abstinence in the third month after discharge. The aftercare treatment effect of AA was estimated using standard logistic regression analysis and instrumental variables analysis. Instruments included the subject's ability to drive oneself to AA meetings and the presence/absence of an AA meeting in the subject's town of residence. A Hausman exogeniety test was used to determine whether the standard logistic regression results were subject to self-selection bias. RESULTS: Estimates from the standard logistic regression yielded a positive (OR = 3.7) and significant (p = .018) treatment effect for AA attendance. However, the instrumental variables analysis yielded a smaller (OR* = 1.7) and insignificant treatment effect estimate (p = .782). The Hausman exogeniety test indicated that the treatment effect estimate from the standard logistic regression was subject to significant self-selection bias (chi2 = 83.9, 1 df, p <.01). CONCLUSIONS: The AA aftercare treatment effect observed in this sample was due to differences in unobserved casemix factors between the treatment groups. Results suggest that previous AA aftercare research may have also been subject to self-selection bias. Researchers of substance abuse outcomes should consider analyzing nonexperimental data using instrumental variables methodologies.
Adult↗
Affirmative actions: can the discriminant accuracy of a test be determined in the face of selection bias?
Clinical estimates of test efficacy can be distorted by the differential referral of positive and negative test responders for outcome verification. Accordingly, a series of computer simulations was performed to quantify the effects of various degrees of this selection bias on the observed true-positive rate, false-positive rate, and discriminant accuracy of a hypothetical test. The error in observed true- and false-positive rates was positive with respect to diagnosis, and negative with respect to prognosis. The magnitude of error was highly correlated with the magnitude of bias associated with the test response (primary selection bias), but not with the magnitude of bias associated with additional independent factors (secondary selection bias). Mathematical correction for preferential referral based on the test response using a previously published algorithm completely removed the correlation with primary selection bias for both diagnosis and prognosis. Although a significant correlation with secondary selection bias persisted at intermediate base rates, its magnitude was small. Discriminant accuracy was assessed in terms of area under a receiver operating characteristic (ROC) curve. Biased values of true- and false-positive rates were distributed along the curve defined by the actual true- and false-positive rates of the test for both diagnosis and prognosis. As a result, the areas under ROC curves calculated from biased true- and false-positive rates were within 2% of the areas calculated from the actual rates. Only when the primary and secondary observations were independent with respect to one outcome and dependent with respect to the other outcome did a systematic error appear in ROC area.(ABSTRACT TRUNCATED AT 250 WORDS)
Computer Simulation↗
The effects of sample selection bias on racial differences in child abuse reporting.
OBJECTIVE: The aim was to examine whether design features of Wave 1, 1980 National Incidence Study (NIS) data resulted in sample selection bias when certain victims of maltreatment were excluded. METHOD: Logistic regression models for the probability of child abuse reports to the child protective services (CPS) were estimated using maximum likelihood methods for Black (n = 511) and White (n = 2499) child abuse cases. The models were estimated with and without correction for selection bias using a two-step procedure proposed by Heckman. RESULTS: Substantial differences were found in the characteristics of Black and White victims by source of report and by type of maltreatment. Also found were sizeable differences within each racial group between sampled agencies and nonsampled agencies. Sample selection bias affected the estimation of both White and Black child abuse reporting rates. In the Black sample, however, the effect of sample selection bias was to reduce the statistical significance of the impacts of reporting agency and physical and sexual abuse on report rates. In the White sample, most significant factors in the basic model remained statistically significant with correction for selection bias. CONCLUSIONS: Selection bias was found to be caused by the exclusion of family, friends, and neighbors in the NIS sample design. Such exclusion has the effect of altering the interpretation of the determinants of child abuse reporting among Blacks, but not among Whites. Thus, conclusions about racial differences in child maltreatment must be reached cautiously, given the NIS study design.
Child↗
Measuring second-order selection bias in a work site health program.
Magnitude and direction of second-order self-selection bias were assessed in a sample of 93,807 IBM employees who participated in the company's Voluntary Health Assessment (VHA) Program, by comparing repeat participants with one-time participants and by simulating selection into the repeat sample. One-time and repeat VHA participants differed systematically but not uniformly in several health characteristics. Repeat participants improved significantly in risk-relevant behaviors and health risk measures. Simulation of selection bias by excluding healthier or less healthy participants from the repeat VHA sample showed findings of gain to be robust. In studies of gain, second-order selection bias cannot automatically be assumed to inflate gain nor to be of sufficient magnitude to affect conclusions about program effects. Simulation is a useful tool for gauging direction and magnitude of selection bias.
Adult↗
Sources of selection bias in evaluating social programs: an interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation method.
This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.
Bias↗
The lack of selection bias in a snowball sampled case-control study on drug abuse.
BACKGROUND: Friend controls in matched case-control studies can be a potential source of bias based on the assumption that friends are more likely to share exposure factors. This study evaluates the role of selection bias in a case-control study that used the snowball sampling method based on friendship for the selection of cases and controls. METHODS: The cases selected fro the study were drug abusers located in the community. Exposure was defined by the presence of at least one psychiatric diagnosis. Psychiatric and drug abuse/dependence diagnoses were made according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) criteria. Cases and controls were matched on sex, age and friendship. The measurement of selection bias was made through the comparison of the proportion of exposed controls selected by exposed cases (p1) with the proportion of exposed controls selected by unexposed cases (p2). If p1 = p2 then, selection bias should not occur. RESULTS: The observed distribution of the 185 matched pairs having at least one psychiatric disorder showed a p1 value of 0.52 and a p2 value of 0.51, indicating no selection bias in this study. CONCLUSIONS: Our findings support the idea that the use of friend controls can produce a valid basis for a case-control study.
Adolescent↗