Jeanine E. Ransom

Mortality, Disability, and Nursing Home Use for Persons with and without Hip Fracture: A Population‐Based Study

OBJECTIVES: To compare persons with and without hip fracture for subsequent mortality and change in disability and nursing home (NH) use.

Incidence and prevalence of CIDP and the association of diabetes mellitus

Background: The reported prevalence of chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) varies greatly, from 1.9 to 7.7 per 100,000. CIDP is reported to occur more commonly in patients with diabetes mellitus (DM) but has not been rigorously tested. Objectives: To determine the incidence (1982–2001) and prevalence (on January 1, 2000) of CIDP in Olmsted County, Minnesota, and whether DM is more frequent in CIDP. Methods: CIDP was diagnosed by clinical criteria followed by review of electrophysiology. Cases were coded as definite, probable, or possible. DM was ascertained by clinical diagnosis or current American Diabetes Association glycemia criteria. Results: One thousand five hundred eighty-one medical records were reviewed, and 23 patients (10 women and 13 men) were identified as having CIDP (19 definite and 4 probable). The median age was 58 years (range 4–83 years), with a median disease duration at diagnosis of 10 months (range 2–64 months). The incidence of CIDP was 1.6/100,000/year. The prevalence was 8.9/100,000 persons on January 1, 2000. Only 1 of the 23 CIDP patients (4%) also had DM, whereas 14 of 115 age- and sex-matched controls (12%) had DM. Conclusions: 1) The incidence (1.6/100,000/year) and prevalence (8.9/100,000) of chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) are similar to or higher than previous estimates. 2) The incidence of CIDP is similar to that of acute inflammatory demyelinating polyradiculoneuropathy within the same population. 3) Diabetes mellitus (DM) is unlikely to be a major risk covariate for CIDP, but we cannot exclude a small effect. 4) The perceived association of DM with CIDP may be due to misclassification of other forms of diabetic neuropathies and excessive emphasis on electrophysiologic criteria.

Incidence of traumatic brain injury across the full disease spectrum: A population-based medical record review study

Background: Extremely few objective estimates of traumatic brain injury incidence include all ages, both sexes, all injury mechanisms, and the full spectrum from very mild to fatal events. Methods: We used unique Rochester Epidemiology Project medical records-linkage resources, including highly sensitive and specific diagnostic coding, to identify all Olmsted County, MN, residents with diagnoses suggestive of traumatic brain injury regardless of age, setting, insurance, or injury mechanism. Provider-linked medical records for a 16% random sample were reviewed for confirmation as definite, probable, possible (symptomatic), or no traumatic brain injury. We estimated incidence per 100,000 person-years for 1987–2000 and compared these record-review rates with rates obtained using Centers for Disease Control and Prevention (CDC) data-systems approach. For the latter, we identified all Olmsted County residents with any CDC-specified diagnosis codes recorded on hospital/emergency department administrative claims or death certificates during 1987–2000. Results: Of sampled individuals, 1257 met record-review criteria for incident traumatic brain injury; 56% were ages 16–64 years, 56% were male, and 53% were symptomatic. Mechanism, sex, and diagnostic certainty differed by age. The incidence rate per 100,000 person-years was 558 (95% confidence interval = 528–590) versus 341 (331–350) using the CDC data-system approach. The CDC approach captured only 40% of record-review cases. Seventy-four percent of missing cases presented to the hospital/emergency department; none had CDC-specified codes assigned on hospital/emergency department administrative claims or death certificates; and 66% were symptomatic. Conclusions: Capture of symptomatic traumatic brain injuries requires a wider range of diagnosis codes, plus sampling strategies to avoid high rates of false-positive events.

Comorbid conditions associated with Parkinson's disease: A population-based study

The burden of comorbidity in Parkinsons disease (PD) remains unclear. All Olmsted County, Minnesota, residents with incident PD in 1976–1995 (n = 197) plus one age‐ and sex‐matched non‐PD referent subject per case were followed for all clinical diagnoses from 5 years before through 15 years after index (i.e., year of PD onset for each case and same year for the referent subject). Both members of a case–referent pair were censored at death or emigration of either member to ensure equivalent follow‐up. Cases and referent subjects were compared for summary comorbidity (Charlson index) and for the likelihood of having one or more diagnoses within each International Classification of Diseases chapter/subchapter. Before index, the groups were similar for all comparisons. After index, cases had a higher likelihood of diagnoses within the chapters “Mental Disorders” and “Diseases of the Genitourinary System,” and within the subchapters “Organic Psychotic Conditions,” “Other Psychoses,” “Neurotic/Personality/Other Nonpsychotic Disorders,” “Hereditary/Degenerative Diseases of Central Nervous System,” “Symptoms,” “Other Diseases of Digestive System,” “Other Diseases of Urinary System,” “Diseases of Veins/Lymphatics/Other Circulatory System Diseases,” “Fractures of Lower Limb,” “Other Diseases of Skin/Subcutaneous Tissue,” “Osteopathies/Chrondropathies/Acquired Musculoskeletal Deformities,” and “Pneumonia and Influenza.” The excess morbidity and mortality observed for persons with PD are consistent with recognized PD sequelae.

Temporal Association of Changes in Fasting Blood Glucose and Body Mass Index With Diagnosis of Pancreatic Cancer

OBJECTIVES:Although the association between diabetes mellitus (DM) and pancreatic cancer is well described, temporal patterns of changes in fasting blood glucose (FBG) and body mass index (BMI) before pancreatic cancer diagnosis are not known.METHODS:We reviewed the medical records of pancreatic cancer cases seen at the Mayo Clinic from 15 January 1981 through to 9 July 2004 and selected those residing within 120 miles of Rochester, MN and who were seen at the Mayo Clinic within 30 days of the date of cancer diagnosis (index date). We identified approximately two matched controls per case residing locally and seen at Mayo in the year of their case index date. For the 736 cases and 1,875 controls with at least one outpatient FBG measurement, we abstracted all FBG values, and corresponding heights and weights up to 60 months before the index date and grouped them into 12-month intervals preceding the index. We compared FBG and BMI in each interval between cases and controls.RESULTS:Mean FBG values were similar between cases, compared with controls, in the intervals from months −60 to −48 (102 mg per 100 ml vs. 100 mg per 100 ml, P=0.34) and from months −48 to −36 (106 mg per 100 ml vs. 102 mg per 100 ml, P=0.09), but progressively increased in the intervals from months −36 to −24 (105 mg per 100 ml vs. 100 mg per 100 ml, P=0.01), from months −24 to −12 (114 mg per 100 ml vs. 102 mg per 100 ml, P=0.001), and from months −12 to +1 (123 mg per 100 ml vs. 102 mg per 100 ml, P<0.0001). Though mean BMI values were generally similar in cases and controls up to 12 months before index, they were significantly lower in cases vs. controls in the interval from months −12 to +1 (P<0.001).CONCLUSIONS:Pancreatic cancer is characterized by progressive hyperglycemia beginning up to 24 months before cancer diagnosis in the setting of decreasing BMI. Pancreatic cancer can potentially be diagnosed early if biomarkers are identified that can distinguish pancreatic cancer–induced DM from type II DM.

Recent trends in the prevalence of coronary disease: a population-based autopsy study of nonnatural deaths.

BACKGROUND Despite increases in obesity and diabetes mellitus, mortality caused by coronary disease continues to decline. Recent trends in coronary disease prevalence are unknown. METHODS There were 3237 deaths among Olmsted County, Minnesota, residents aged 16 through 64 years during the 1981-2004 period. Of the 515 due to accident, suicide, homicide, or a manner that could not be determined, 425 individuals (82%) had coronary anatomy graded. Pathology reports were reviewed for the grade of coronary disease (range, 0-5) assigned each of 4 arteries: left anterior descending (LAD), left circumflex (LCx), right coronary artery (RCA), and left main artery (LMA). High-grade disease was defined as more than a 75% reduction in cross-sectional luminal area (grade >or=4) in any of LAD, LCx, or RCA or more than 50% reduction (grade >or=3) in LMA. Evidence of any disease was defined as a grade higher than 0 in any artery. Calendar-year trends were analyzed as linear and nonlinear functions. RESULTS Over the full period (1981-2004), 8.2% of the 425 individuals had high-grade disease, and 83% had evidence of any disease. Age- and sex-adjusted regression analyses revealed temporal declines over the full period (1981-2004) for high-grade disease, any disease, and grade of coronary disease. Declines in the grade of coronary disease ended after 1995 (P <or= .01 for every artery) and possibly reversed after 2000 (P = .06 for LCx). CONCLUSIONS Declines in coronary disease prevalence overall (during 1981-2004) reinforce arguments that any increased prevalence resulting from improved survival among persons with disease was offset by reductions in disease incidence. Study findings suggest that declines in coronary disease prevalence have ended. The question of whether recent trends are attributable to increasing obesity and diabetes mellitus awaits further investigation.

Use of Nursing Home After Stroke and Dependence on Stroke Severity A Population-Based Analysis

BACKGROUND AND PURPOSE There are few population-based data available regarding nursing home use after stroke. This study clarifies the use of a nursing home after stroke, as well as its dependence on stroke severity, in a defined population. METHODS All first stroke events among residents of Rochester, Minn, during 1987-1989 were ascertained, subtyped, and assigned Rankin disability scores (RS) before the event, at maximal deficit, and at specified intervals after stroke. Persons were followed from the date of stroke event to death, emigration from Rochester, or December 31, 1994, in complete community-based medical records and Minnesota Case Mix Review Program data tapes to determine nursing home residency before stroke and at 90 days and 1 year after stroke, proportion of survival days in a nursing home, and cumulative risk of admission to a nursing home. RESULTS There were 251 cases of first cerebral infarction, 24 intracerebral hemorrhages, and 15 subarachnoid hemorrhages among residents of Rochester during 1987-1989. The maximal deficit RS was 1 or 2 for 62 (25%), RS 3 for 72 (29%), and RS 4 or 5 for 117 (47%) of the cerebral infarct patients. Among patients surviving to 90 days or 1 year after cerebral infarction, 25% were in nursing home at 90 days and 22% at 1 year, respectively. Within these maximal deficit RS categories, the percentages of follow-up time spent in a nursing home during the first post-cerebral infarction year are as follows: RS 1 to 2, 4%; RS 3, 10%; and RS 4 to 5, 54%. Multivariate logistic regression revealed that increasing age and RS 4 to 5 at maximal deficit were independent predictors (P<0.0001) of nursing home residency at 90 days and 1 year after stroke, whereas stroke type was not an independent predictor. At 1 year after cerebral infarction, the Kaplan-Meier estimates of proportion of people with at least 1 nursing home admission were 11% for RS 1 to 2, 22% for RS 3, and 68% for RS 4 to 5. CONCLUSIONS This study provides unique population-based data regarding the short- and long-term use of a nursing home after stroke and its dependence on stroke severity. More than 50% of people with a severe cerebral infarction are in a nursing home 90 days and 1 year after the stroke, and by 1 year, nearly 70% will have required some nursing home stay. Age and stroke severity are independent predictors of nursing home residency after stroke.

Impact of data fragmentation across healthcare centers on the accuracy of a high-throughput clinical phenotyping algorithm for specifying subjects with type 2 diabetes mellitus

OBJECTIVE To evaluate data fragmentation across healthcare centers with regard to the accuracy of a high-throughput clinical phenotyping (HTCP) algorithm developed to differentiate (1) patients with type 2 diabetes mellitus (T2DM) and (2) patients with no diabetes. MATERIALS AND METHODS This population-based study identified all Olmsted County, Minnesota residents in 2007. We used provider-linked electronic medical record data from the two healthcare centers that provide >95% of all care to County residents (ie, Olmsted Medical Center and Mayo Clinic in Rochester, Minnesota, USA). Subjects were limited to residents with one or more encounter January 1, 2006 through December 31, 2007 at both healthcare centers. DM-relevant data on diagnoses, laboratory results, and medication from both centers were obtained during this period. The algorithm was first executed using data from both centers (ie, the gold standard) and then from Mayo Clinic alone. Positive predictive values and false-negative rates were calculated, and the McNemar test was used to compare categorization when data from the Mayo Clinic alone were used with the gold standard. Age and sex were compared between true-positive and false-negative subjects with T2DM. Statistical significance was accepted as p<0.05. RESULTS With data from both medical centers, 765 subjects with T2DM (4256 non-DM subjects) were identified. When single-center data were used, 252 T2DM subjects (1573 non-DM subjects) were missed; an additional false-positive 27 T2DM subjects (215 non-DM subjects) were identified. The positive predictive values and false-negative rates were 95.0% (513/540) and 32.9% (252/765), respectively, for T2DM subjects and 92.6% (2683/2898) and 37.0% (1573/4256), respectively, for non-DM subjects. Age and sex distribution differed between true-positive (mean age 62.1; 45% female) and false-negative (mean age 65.0; 56.0% female) T2DM subjects. CONCLUSION The findings show that application of an HTCP algorithm using data from a single medical center contributes to misclassification. These findings should be considered carefully by researchers when developing and executing HTCP algorithms.

Trends in the epidemiology of osteomyelitis: a population-based study, 1969 to 2009.

BACKGROUND The epidemiology of osteomyelitis in the United States is largely unknown. The purpose of this study was to determine long-term secular trends in the incidence of osteomyelitis in a population-based setting. METHODS The study population comprised 760 incident cases of osteomyelitis first diagnosed between January 1, 1969, and December 31, 2009, among residents of Olmsted County, Minnesota. The complete medical records for each potential subject were reviewed to confirm the osteomyelitis diagnosis and to extract details on anatomical sites, infecting organisms, etiological risk factors, and outcomes. RESULTS The overall age and sex-adjusted annual incidence of osteomyelitis was 21.8 cases per 100,000 person-years. The annual incidence was higher for men than for women and increased with age (p < 0.001). Rates increased with the calendar year (p < 0.001) from 11.4 cases per 100,000 person-years in the period from 1969 to 1979 to 24.4 per 100,000 person-years in the period from 2000 to 2009. The incidence remained relatively stable among children and young adults but almost tripled among individuals older than sixty years; this was partly driven by a significant increase in diabetes-related osteomyelitis from 2.3 cases per 100,000 person-years in the period from 1969 to 1979 to 7.6 cases per 100,000 person-years in the period from 2000 to 2009 (p < 0.001). Forty-four percent of cases involved Staphylococcus aureus infections. CONCLUSIONS The reasons for the increase in osteomyelitis between 1969 and 2009 are unclear but could comprise a variety of factors, including changes in diagnosing patterns or increases in the prevalence of risk factors (e.g., diabetes) in this population.

The Epidemiology and Health Care Burden of Tennis Elbow A Population-Based Study

Background: Lateral elbow tendinosis (epicondylitis) is a common condition both in primary care and specialty clinics. Purpose: To evaluate the natural history (ie, incidence, recurrence, and progression to surgery) of lateral elbow tendinosis in a large population. Study Design: Descriptive epidemiology study. Methods: The study population comprised a population-based incidence cohort of patients with new-onset lateral elbow tendinosis between January 1, 2000, and December 31, 2012. The medical records of a 10% random sample (n = 576) were reviewed to ascertain information on patient and disease characteristics, treatment modalities, recurrence, and progression to surgery. Age- and sex-specific incidence rates were calculated and adjusted to the 2010 US population. Results: The age- and sex-adjusted annual incidence of lateral elbow tendinosis decreased significantly over time from 4.5 per 1000 people in 2000 to 2.4 per 1000 in 2012 (P < .001). The recurrence rate within 2 years was 8.5% and remained constant over time. The proportion of surgically treated cases within 2 years of diagnosis tripled over time, from 1.1% during the 2000-2002 time period to 3.2% after 2009 (P < .00001). About 1 in 10 patients with persistent symptoms at 6 months required surgery. Conclusion: The decrease in incidence of lateral elbow tendinosis may represent changes in diagnosis patterns or a true decrease in disease incidence. Natural history data can be used to help guide patients and providers in determining the most appropriate course at a given time in the disease process. The study data suggest that patients without resolution after 6 months of onset may have a prolonged disease course and may need surgical intervention.