SEARCH · Search PubMed

Results for “Undercount”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6Linked to original sources

PubMed1998

Spinal cord injury incidence in Mississippi: a capture-recapture approach.

BACKGROUND: Many studies have investigated spinal cord injury incidence rates. Few, however, have adjusted for the underascertainment. The current study used the capture-recapture method to estimate the ascertainment-corrected spinal cord injury incidence rate in Mississippi. METHODS: Two sources were used for case ascertainment: Mississippi's spinal cord injury registry and hospital reports. The two-sample capture-recapture method was used to adjust for undercount. RESULTS: Two hundred one spinal cord injuries were reported to or found by the Mississippi State Department of Health in 1993, with a crude incidence rate of 7.8 per 100,000 population per year among hospital admissions and prehospital fatalities. Using the two-sample capture-recapture method, it is estimated that the incidence rate would be 9.3 per 100,000 population per year. CONCLUSION: Capture-recapture estimates suggest that Mississippi's spinal cord injury incidence rate is more than twice the national average.

Epidemiologic Methods↗

PubMed1991

Platelet studies in the pathogenesis of thrombocytopenia in May-Hegglin anomaly.

Thrombocytopenia has a variable incidence in the May-Hegglin anomaly, a rare congenital syndrome characterized by inclusions resembling Dohle bodies within leukocytes, thrombocytopenia, and giant platelets. We studied a 16-year-old girl and three relatives with the syndrome to elucidate the pathogenesis of the thrombocytopenia and the degree of impaired hemostasis. On confirmation of automated platelet counts by manual techniques we found that both aperture-impedence and light-scatter instruments tended to undercount the actual number of platelets by 30-50%. An enzyme immunoassay for platelet-associated IgG showed elevated levels in all patients (mean of 2.3 +/- 0.6 ng/10(6) platelets versus 0.8 +/- 0.6 in normals, p less than 0.0001). Bleeding time and platelet aggregation in response to adenosine diphosphate, arachidonic acid, epinephrine, ristocetin, and collagen were all normal. No clinical bleeding was observed in any of the patients in the face of major surgery or childbirth. The findings suggest the possibility of a peripheral destructive and perhaps an immune pathogenesis of the thrombocytopenia of May-Hegglin anomaly and confirm that clinical bleeding may not accompany the severe thrombocytopenia seen in this disorder.

Adolescent↗

PubMed2006

How much work-related injury and illness is missed by the current national surveillance system?

OBJECTIVE: We sought to estimate the undercount in the existing national surveillance system of occupational injuries and illnesses. METHODS: Adhering to the strict confidentiality rules of the U.S. Bureau of Labor Statistics, we matched the companies and individuals who reported work-related injuries and illnesses to the Bureau in 1999, 2000, and 2001 in Michigan with companies and individuals reported in four other Michigan data bases, workers' compensation, OSHA Annual Survey, OSHA Integrated Management Information System, and the Occupational Disease Report. We performed capture-recapture analysis to estimate the number of cases missed by the combined systems. RESULTS: We calculated that the current national surveillance system did not include 61% and with capture-recapture analysis up to 68% of the work-related injuries and illnesses that occurred annually in Michigan. This was true for injuries alone, 60% and 67%, and illnesses alone 66% and 69%, respectively. CONCLUSIONS: The current national system for work-related injuries and illnesses markedly underestimates the magnitude of these conditions. A more comprehensive system, such as the one developed for traumatic workplace fatalities, that is not solely dependent on employer based data sources is needed to better guide decision-making and evaluation of public health programs to reduce work-related conditions.

Databases, Factual↗

PubMed2003

Estimating private sector professional fees for VA providers.

OBJECTIVES: To describe new methods used to estimate inpatient and outpatient Medicare-based professional fees for Veterans Health Administration (VA) services. METHODS: National VA utilization files provided estimates of inpatient physician services, whereas local provider and utilization files gave counts of outpatient services by physicians, nurse practitioners, physician assistant, clinical psychologists, and clinical social workers. Services from ambulatory surgery, emergency room, and clinics (eg, dermatology and gastroenterology) were coded by study health information management staff (coders). VA-based billing information was edited against Medicare guidelines. Estimates for VA services without comparable Medicare fees were obtained from other commercial sources. RESULTS: Hypothetical professional fees for VA services were 17% more ($109 million vs. $93 million) than the VA budget for physicians over 1 fiscal year at six sites. Total payments of nearly $21 million were generated for VA inpatient care. In fiscal year 1999, there were 30,209 admissions (of which 4549 were psychiatric) to the study sites; 30,518 discharges; 229,783 inpatient days, including 27,235 in critical care units; and 38,348 surgical days of care. DISCUSSION: Differences between the VA and the private sector maybe overstated because VA salaries of nonphysicians were not included in the VA budgets. Conversely, the extent to which VA professional services were undercounted in VA information systems used in this study may understate the difference. Future research may consider additional data collection approaches or information systems enhancements to enumerate more accurately all provider services that are reimbursable in the private sector.

Budgets↗

PubMed2005

Accuracy of Medicaid payer coding in hospital patient discharge data: implications for Medicaid policy evaluation.

BACKGROUND: Ambulatory care-sensitive hospitalization rates derived from hospital discharge data have been used to compare ambulatory care across insurance and delivery system groups. OBJECTIVE: We sought to quantify the impact of coding inaccuracies in hospital discharge data on counts of hospitalizations for ambulatory care-sensitive conditions among Medicaid beneficiaries. METHODS: This was a cross-sectional comparison of administrative databases of all California Medicaid beneficiaries younger than 65 years of age. We compared the number of hospitalizations that were attributed to Medicaid beneficiaries in California's hospital discharge data for 1994 to 1999 with the number derived from a file that linked hospital discharge data with the Medicaid eligibility file. RESULTS: Hospital discharge data undercounted 28.2% of hospitalizations for ambulatory care-sensitive conditions among Medicaid beneficiaries and overcounted 13.4% of such admissions among non-Medicaid beneficiaries. Approximately 5% of hospitalizations for ambulatory care-sensitive conditions captured as Medicaid admissions in routine hospital discharge data were among patients who gained Medicaid coverage as a result of the hospitalization. Patients who acquire Medicaid coverage as a result of a hospitalization are much more likely to be placed into Medicaid fee for service rather than Medicaid managed care which biases comparisons of these 2 delivery models. CONCLUSION: Caution should be used in the interpretation of Medicaid hospitalization rates as calculated from routine hospital discharge data. State agencies that provide hospital discharge data should consider the opportunity to improve the evaluation of Medicaid services by linking hospital discharge data with Medicaid enrollment files.

Adult↗

PubMed2006

Validity of cancer registry medicaid status against enrollment files: implications for population-based studies of cancer outcomes.

BACKGROUND: Poor access to or inadequate health insurance contributes to disparities in cancer incidence and mortality. Cancer registry "payer source" data is collected by many cancer registries in the United States and has been used to compare cancer outcomes across insurance types. OBJECTIVES: We evaluated the validity of cancer registry data on patient Medicaid status against enrollment data from Medi-Cal, California's Medicaid program. METHODS: Data from the statewide California Cancer Registry for persons under age 65 years diagnosed with 1) any cancer in 1998 and 1999 or 2) with invasive cervical cancer between 1996 and 1999 were obtained and linked probabilistically to Medi-Cal enrollment files. We compared registry Medicaid status, determined from payer source information, against linkage results and used crosstabulations to calculate sensitivity, specificity, and positive predictive value. These measures were compared across different hospital and patient characteristics and cancer types. RESULTS: Cancer registry Medicaid status data had poor sensitivity (48%), good specificity (98%), and moderate positive predictive value (77%). Measures of validity did not vary substantially by cancer type, stage, patient age, sex, vital status, race/ethnicity, socioeconomic status, or diagnosing hospital size. Registry data undercounted the number of Medicaid patients by 52% and incorrectly assigned Medicaid as a payer to approximately 2% of patients. CONCLUSIONS: As a result of the poor validity of cancer registry Medicaid status data, caution should be used when interpreting cancer outcomes by insurance type calculated from registry payer source data. Linkage of registry data to Medicaid enrollment files represents a more accurate means of identifying Medicaid insurance status.

Adolescent↗

PubMed2006

Trends in colorectal cancer mortality by ethnicity and socio-economic position in New Zealand, 1981-99: one country, many stories.

BACKGROUND: Ethnicity and socio-economic position are important determinants of colorectal cancer (CRC) mortality. In this paper, we determine trends in colorectal cancer mortality by ethnicity and socio-economic position in New Zealand. METHODS: Cohort studies of the entire New Zealand population for 1981-84, 1986-89, 1991-94 and 1996-99 (linking Census and mortality datasets) allowed direct determination of trends in CRC mortality by income and education. For ethnicity, we used routine unlinked Census and mortality data, but with correction factors applied for undercounting of Mâori and Pacific deaths. RESULTS: ETHNICITY: CRC mortality trends varied markedly. There were small (10-20%) decreases among non-Mâori non-Pacific people, a 50% increase among Mâori, and up to 10-fold increase among Pacific people. By 1996-99, all three ethnic groups had similar CRC mortality. SOCIO-ECONOMIC POSITION: For females, differences in CRC mortality by education and income increased over time e.g. poor females had a 40% higher CRC mortality than rich females in 1996-99, compared with no difference in 1981-84 (p for trend 0.04). In men, increases in inequality were seen by income but not education. CONCLUSION: The observed ethnic trends probably reflect differential trends in exposure to etiological risk factors. Social inequalities in colorectal cancer mortality appear to be increasing.

Adult↗

PubMed2000

How much do persons with Alzheimer's disease cost Medicare?

BACKGROUND: Medicare claims are increasingly being used to identify persons with chronic diseases such as Alzheimer's disease (AD) for the purpose of determining the cost to Medicare of caring for such persons. Past work has been limited by the use of only 1 or 2 years of claims data to identify cases, leading to worries that this might lead to an undercount of prevalent cases and bias cost findings. OBJECTIVES: To analyze the average total cost to the Medicare program in 1994 of persons with a claims-based diagnosis of AD, using a 12-year period of claims history to identify prevalent cases, and to investigate the effect on cost of time since diagnosis. DESIGN: A cross-sectional design with a 12-year retrospective period to identify persons with AD. SETTING: Medical care practices, hospitals, and other providers of services to Medicare beneficiaries in the US in 1994. SUBJECTS: Respondents to the screener (n = 10,858) and community (5429) and institutional (n = 1341) questionnaire of the 1994 National Long Term Care Survey, with and without a claims-based diagnosis of AD. MEASUREMENTS: Average total cost to Medicare in 1994, measured as the actual amount Medicare paid for inpatient, outpatient, home health, skilled nursing facility, hospice, and Part B services, including payments to physicians, and other items such as durable medical equipment. We also measured disability in a variety of ways, including cognition, activity limitations, and residence in a nursing home. RESULTS: The average total cost to Medicare of persons with a claims-based diagnosis of AD was $6021 versus $2310 (P < .001) for persons without a diagnosis. When adjusting for patient characteristics, the ratio of cost between persons with AD and those without was reduced to about 1.6 to 1. Time since diagnosis was an important predictor of average total cost in 1994, with each additional year since diagnosis resulting in a $248 (P = .04) decrease in total cost (about 10% of the total sample mean cost of $2426). There was mixed evidence that persons with a diagnosis of AD incurred less cost than otherwise similarly disabled Medicare beneficiaries. CONCLUSIONS: Time since diagnosis with AD is an important predictor of cost and one that should be explicitly included in any rate-setting formula. Expanding the period used to identify cases resulted in an increase in the unadjusted ratio of cost of a Medicare beneficiary with AD relative to one without primarily because our control group costs are lower compared with those of past work.

Activities of Daily Living↗

PubMed1994

Undersensing during ventricular tachyarrhythmias in a third-generation implantable cardioverter defibrillator: diagnosis using stored electrograms and correction with programming.

Third-generation implantable cardioverter defibrillators with stored electrograms allow diagnosis of various sensing problems that may lead to an inappropriate device response. Undersensing of ventricular tachyarrhythmias is a potentially serious problem, as it may lead to failure to deliver therapy. To determine the incidence of this problem, we reviewed 98 patients with Ventritex Cadence defibrillator systems and found 2 patients in whom defibrillation therapy was delayed or aborted because of undersensing during induced ventricular tachyarrhythmias. In both cases, examination of stored electrograms revealed variation in electrogram amplitude, which presumably resulted in failure of the autogain feature to increase its sensitivity enough to count each complex. During charging, criteria for redetection of sinus rhythm were met because of this undercounting, leading to failure to deliver defibrillation therapy. This problem was detected in both patients 4-6 weeks following device implant during device testing, and both patients had been started on antiarrhythmic drug therapy prior to this testing. Programming the sinus redetection parameter from nominal to slow, increasing the number of beats necessary to confirm resumption of sinus rhythm, corrected the problem in both patients. Device testing in the electrophysiology laboratory, routinely postoperatively and following initiation of antiarrhythmic drug therapy, and the ability to retrieve stored electrograms are useful in detecting such sensing anomalies.

Aged↗

PubMed1997

A population-based study of fatal and nonfatal firearm-related injuries.

OBJECTIVE: To determine population-based firearm-related morbidity and mortality for Allegheny County, PA (population = 1.3 million), for the year 1994. METHODS: Fatalities were identified from a review of death certificates. To identify nonfatal cases, an active surveillance was conducted at all 24 acute care EDs in the county. The ED surveillance used 2 existing sources of case identification from each hospital to minimize undercount. RESULTS: Firearms were the leading cause of injury death to county residents, accounting for 155 deaths. The crude mortality rate from firearms was 11.7/100,000. Black males aged 15-19 years were most at risk for a firearm fatality (293/100,000). There were 514 nonfatal firearm injuries, producing a case fatality rate of 23%. The highest age-specific rate for nonfatal firearm-related injuries treated in the county EDs was observed for black males aged 15-19 years (2,245/100,000), which is 58 times higher than the firearm-related injury rate for the entire county population (38.7/100,000). CONCLUSION: Firearm-related injury and death are a significant public health problem in Allegheny County. Although the crude mortality rate from firearms in the county is lower than the reported national rate, the observed rate for nonfatal injuries in the black youth of this community is the highest firearm injury incidence rate ever reported. Local surveillance of firearm-related injuries, including nonfatal events, is needed to more accurately demonstrate the magnitude of this problem.

Adolescent↗

PubMedDate not supplied

Urocystolith detection: comparison of survey, contrast radiographic and ultrasonographic techniques in an in vitro bladder phantom.

Urocystoliths of 9 mineral types from 437 canine patients submitted to the University of Minnesota Urolith Bank were imaged in a urinary bladder phantom. Imaging techniques simulated were survey radiography, pneumocystography, double contrast cystography (two iodine concentrations) and real-time ultrasonography (3.5 MHz, 5.0 MHz, 7.5 MHz). Imaging techniques were compared for accuracy of urocystolith detection, accuracy of urocystolith enumeration, and tendencies for over or undercounting. Across urocystolith mineral types, the false negative rates (no urocystoliths detected in a given case) for survey radiographs range from 2 to 27%. Pneumocystographic techniques are one-half as likely to yield false negative results as are survey radiographic techniques. Underestimates of urocystolith numbers and false negatives are likely using 80 mg iodine/ml double contrast cystography because calcium-based urocystoliths are isopaque in this contrast medium dilution. The 200 mg iodine/ml double contrast cystographic techniques are unlikely to yield false negative diagnoses even for very small (< or = 1.0 mm) urocystoliths and is comparable to pneumocystography for detection and slightly better for enumeration. The likelihood of an ultrasonographic false negative for urocystoliths decreases with increasing MHz. Under optimal conditions using a 7.5 MHz mechanical sector transducer, the false negative rates were comparable to double contrast cystography, but rates increased notably with lower MHz transducers.

Animals↗

PubMed2000

Children are not goldfish--mark/recapture techniques and their application to injury data.

OBJECTIVES: Mark/recapture (or capture-recapture) is a simple technique commonly applied to estimate the hypothetical total (including undercount) in a register composed of cases from two or more independent and separately incomplete case lists. This paper seeks to illustrate serious drawbacks in the use of the mark/recapture technique when applied to injuries. SETTING AND SUBJECTS: Northumbrian children under 15 years of age who were seriously injured in motor vehicle accidents (MVAs) over a five year period ascertained from two data sources: police reports and hospital inpatient records. METHODS: Individuals (n) appearing in both police (S) and hospital (H) case lists are identified using various matching criteria. The separate and combined influence of age, sex, and casualty class (cyclist, passengers, pedestrians) on the probability of such matching is estimated using multivariate techniques. The hypothetical total incidence of child MVA victims (N) is calculated from N = (S x H)/n. MAIN OUTCOMES: Estimates of the incidences of "serious" injuries in MVAs under various conditions of stratification and matching. The overall procedure is tested for conformity with accepted criteria for valid use of mark/recapture. RESULTS: About one third of the 1009 police and 836 hospital records could be exactly matched. There were significant variations in matching proportions by class of accident (pedestrian v passenger v cyclist). This selective recapture or "heterogeneity" was not affected by sex, but was independently influenced by the age of the child. Further uncertainty was introduced when matching criteria were slightly relaxed. Estimates of the total population of children with serious injuries vary accordingly from 1729 to 2743. A number of plausible reasons why these two data sources might not be unbiased or mutually independent samples of the total target population are proposed as explanations for this heterogeneity. CONCLUSION: This typical example of two sample mark/recapture estimation in an epidemiological setting can be shown to violate virtually all the requirements for valid use of the technique. Very little can be deduced accurately about the scale or characteristics of an unobserved group by the use of mark/recapture applied to two overlapping health event registers.

Accidents, Traffic↗

PubMed2002

Characteristics of pregnant women in motor vehicle crashes.

OBJECTIVES: Motor vehicle crashes are the leading cause of hospitalized trauma during pregnancy. Maternal injury puts the fetus at great risk, yet little is known about the incidence, risks, and characteristics of pregnant women in crashes. SETTING AND METHODS: Police reported crashes were analyzed from the National Automotive Sampling System Crashworthiness Data System. Since 1995, this system recorded pregnancy/trimester status. Pregnant and non-pregnant women 15-39 years of age were compared by age, driver status, seat belt use, and treatment. Belt use and seating position were examined by trimester. RESULTS: There were 427 pregnant occupants identified (weighted n=32 810, 2.6%, SE 12 585, rate 13/1000 person years). The mean age was 24.9 compared with 24.8 years (pregnant v non-pregnant). Cases were distributed by trimester as follows: first 29.8%, second 36.4%, and third 33.8%. Pregnant women were drivers 70% of the time compared with 71% for non-pregnant women. No belt use was 14% compared with 13% (pregnant v non-pregnant). Mean injury severity was lower for pregnant women but they were more likely to transported or hospitalized. Improper belt use decreased after the first trimester and there was little change in driver proportion by trimester. Third trimester hospitalization rates increased. CONCLUSIONS: Pregnant occupants in crashes have similar profiles of restraint use, driver status, and seat position but different treatment indicators compared to non-pregnant occupants. Trimester status has relatively little impact on crash risk, seating position or restraint use. Undercounting of pregnant cases was possible, even so, 1% of all births were reported to be involved in utero in crashes. Little research has focused on developmental outcomes to infants and children previously involved in exposure to these crashes.

Accidents, Traffic↗

PubMed2000

Epidemiology and prevention of injuries among adolescent workers in the United States.

In the United States, most adolescents are employed at some time during their teen years. Provisions of the Fair Labor Standards Act, the nation's child labor law, address some of the risks of adolescent employment; however, > 70 teenage workers are estimated to die from work-related injuries each year, with greater risk among males, whites, and older adolescents. Furthermore, the nonfatal work-related injuries of the > or = 64,000 teens who receive emergency department care each year are believed to significantly undercount actual injuries. Many of these nonfatal injuries are severe, with 15%-45% leading to work restriction or permanent disabilities. Significant methodological issues limit the ability of existing surveillance systems to monitor youth worker injury. Risk factor and intervention research is very limited, but does suggest the role of some worker characteristics, management practices, and training issues. However, these factors need more careful study. This review considers existing information about occupational injuries among adolescents and identifies needs for research and policy attention.

Accidents, Occupational↗

PubMed1996

Pulmonary nodules: effect on detection of spiral CT pitch.

PURPOSE: To compare spiral computed tomography (CT) performed at increased pitch with spiral CT performed at standard pitch in the detection of pulmonary nodules. MATERIALS AND METHODS: Spiral CT scanning of the thorax was performed with a pitch of 1.0 in 109 patients with pulmonary nodules due to metastases. The patients were also randomly assigned to undergo further scanning with a pitch of 1.2 (n = 34), 1.5 (n = 37), 2.0 (n = 38) at the same scanning session. The scan pairs were analysed for number, size, and distribution of nodules. RESULTS: A bias toward undercounting was noted on scans with a pitch of 1.5 and 2.0; however, this was not statistically significant. Correlation coefficients were r = .982, r = .977, and r = .989 for scans of pitch 1.2, 1.5, and 2.0, respectively. Disease in one patient would have been prospectively understaged from findings on a scan of pitch 2.0 because of poor conspicuity of a small solitary nodule. CONCLUSION: Findings from scans with increased pitch generally agree well with those from scans with standard pitch; however, there is a greater risk of understaging of disease in patients with solitary nodules as pitch increases. Pitch should be limited to no greater than 1.5 for initial staging of pulmonary metastatic disease.

Adult↗

PubMed1984

Body potassium by four-pi 40K counting: an anthropometric correction.

Four-pi whole body counting for the 1.46 meV photon of 40K has apparent advantages over single-crystal or two-pi counters in efficiency and in subject geometry independence. However, our studies of obese populations have disclosed a systematic undermeasurement of 40K, suggesting that nonhomogeneous K distribution results in systematic undercounting of 40K. In the current study 42K, emitting a 1.52 meV photon, was used in 109 volunteers ranging from 50 to 181 kg, and multiregression covariance analysis was applied to develop correction formulas based on anthropometrics. These corrections quantitatively account for the unappreciated loss of 40K and 42K photons in annular adipose tissue that surrounds the lean body, in which most K+ is concentrated. The correction ranges from 1 to 28% and is a linear (although different) function of weight in both sexes. Thus corrected, body potassium measurements, taken in conjunction with exchangeable sodium and water measurements, provide estimates for whole body osmolality that match measured serum values. Such a quantitative accounting for previously "lost" cation in 58 subjects provides independent evidence for the appropriateness and accuracy of the correction. With this correction, body potassium was recalculated in the 1,492 adult members of a previously reported group of 3,083 subjects.

Absorption↗

PubMed2005

Estimating the occurrence of amyotrophic lateral sclerosis among Gulf War (1990-1991) veterans using capture-recapture methods.

OBJECTIVE: Using data from a recent report that indicated a 2-fold higher risk of amyotrophic lateral sclerosis (ALS) among veterans of the 1991 Gulf War, we applied capture-recapture methodology to estimate possible under-ascertainment of ALS cases among deployed and non-deployed military personnel who were on active duty during that war. STUDY DESIGN AND SETTING: One of the most serious concerns facing field epidemiological investigations is that of case ascertainment bias, particularly when it is differential among the study groups. Capture-recapture methods, however, have promise as an approach to assessing the impact of case ascertainment bias in such studies. To overcome potential limitations of any one approach, three different estimation methods were used: log-linear models, sample coverage, and ecological models, to obtain a comprehensive view of under-ascertainment bias in these populations. RESULTS: All three approaches indicated differential undercount of ALS cases with modest under-ascertainment likely to have occurred among non-deployed military personnel, but little under-ascertainment among the deployed. After correcting the rates for under-ascertainment, the age-adjusted risk of ALS remained elevated among military personnel who had been deployed to S.W. Asia during the 1991 Gulf War, confirming the earlier report. CONCLUSIONS: Capture-recapture methods are a useful approach to assessing the magnitude of case ascertainment bias in epidemiological studies from which ascertainment-adjusted estimates of rates and relative risks can be calculated.

Adult↗

PubMed1997

Geographic distribution of stroke mortality among immigrants to the United States.

BACKGROUND AND PURPOSE: This study examines the geographic distribution of stroke mortality among immigrants and natives of the United States. METHODS: National Center for Health Statistics and Bureau of the Census data were used to determine the geographic distribution of age-adjusted, race-, and race/sex-specific stroke mortality rates among immigrants and natives of the United States for 1979 to 1981. RESULTS: For whites and blacks and for each of the respective race/sex groups, immigrants had markedly and highly statistically significantly lower age-adjusted stroke mortality rates than either the entire US-born resident population or the US-born interregional migrant population. The spatial pattern of immigrant rates did not parallel the patterns for US-born populations. Immigrant rates were highest in the West and lowest in the Midwest for whites and highest in the Midwest and lowest in the Northeast for blacks, whereas for both US-born whites and blacks, resident and native rates were highest in the South and lowest in the Midwest. With few exceptions, region-specific immigrant rates for whites and blacks were significantly lower than rates for either US-born regional residents, US-born migrants to the regions, or US-born natives of the regions. In contrast, white immigrants to the West had significantly higher rates than US-born groups in that region. CONCLUSIONS: Selection factors strongly influence stroke mortality rates among immigrants to the United States. The aberrantly high rates among white immigrants to the West may in part reflect a bias due to large census undercounts of this population.

Adult↗