A SAS macro for a clustered permutation test

The clustered permutation test is a nonparametric method of significance testing for correlated data. It is often used in cluster randomized trials where groups of patients rather than individuals are randomized to either a treatment or control intervention. We describe a flexible and efficient SAS macro that implements the 2-sample clustered permutation test. We discuss the theory and applications behind this test as well as details of the SAS code.

Dual energy x-ray absorptiometry of the distal femur may be more reliable than the proximal tibia in spinal cord injury

OBJECTIVE

To evaluate the precision of dual energy x-ray absorptiometry scanning at 2 skeletal sites at the knee (proximal femur and distal tibia) in people with SCI.

DESIGN

Cross-sectional.

SETTING

Veterans Affairs Medical Center.

PARTICIPANTS

Subjects (N=20) with chronic SCI.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Precision as determined by root mean square coefficient of variation (RMS-CV) and root mean square standard deviation (RMS-SD).

RESULTS

At the distal femur the root RMS-CV was 3.01% and the RMS-SD was 0.025g/cm2. At the proximal tibia the RMS-CV was 5.91% and the RMS-SD was 0.030g/cm2.

CONCLUSIONS

Precision at the distal femur is greater than at the proximal tibia and we recommend it as the preferred site for the longitudinal assessment of bone mineral density at the knee in chronic SCI.

Predictors of cardiopulmonary hospitalization in chronic spinal cord injury

OBJECTIVE

To investigate longitudinal risk factors of hospitalization for circulatory and pulmonary diseases among veterans with chronic spinal cord injury (SCI). Circulatory and respiratory system illnesses are leading causes of death in patients with chronic SCI, yet risk factors for related hospitalizations have not been characterized.

DESIGN

Prospective cohort study.

SETTING

Veterans Affairs (VA) Boston Healthcare System, Boston, Massachusetts. PARTICIPANTS/DATA SOURCES: Veterans (N=309) greater than or equal to 1 year post-SCI from the VA Boston Chronic SCI cohort who completed a health questionnaire and underwent spirometry at study entry. Baseline data were linked to 1996 through 2003 hospitalization records from the VA National Patient Care Database.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Cardiopulmonary hospital admissions, the predictors of which were assessed by multivariate Cox regression.

RESULTS

Of 1478 admissions observed, 143 were a result of cardiopulmonary (77 circulatory and 66 respiratory) illnesses. Independent predictors were greater age (3% increase/y), hypertension, and the lowest body mass index quintile (<22.4 kg/m2). A greater percentage-predicted forced expiratory volume in 1 second was associated with reduced risk. SCI level and completeness of injury were not statistically significant after adjusting for these risk factors.

CONCLUSIONS

Cardiopulmonary hospitalization risk in persons with chronic SCI is related to greater age and medical factors that, if recognized, may result in strategies for reducing future hospitalizations.

Poisson regression for modeling count and frequency outcomes in trauma research

The authors describe how the Poisson regression method for analyzing count or frequency outcome variables can be applied in trauma studies. The outcome of interest in trauma research may represent a count of the number of incidents of behavior occurring in a given time interval, such as acts of physical aggression or substance abuse. Traditional linear regression approaches assume a normally distributed outcome variable with equal variances over the range of predictor variables, and may not be optimal for modeling count outcomes. An application of Poisson regression is presented using data from a study of intimate partner aggression among male patients in an alcohol treatment program and their female partners. Results of Poisson regression and linear regression models are compared.

Determinants of lung volumes in chronic spinal cord injury

OBJECTIVE

To characterize determinants of lung volumes in chronic spinal cord injury (SCI).

DESIGN

Cross-sectional.

SETTING

VA Boston Healthcare System.

PARTICIPANTS

White men (N=330) with chronic SCI.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Questionnaire responses and measurements of lung volumes.

RESULTS

Adjusted for SCI severity and stature, greater body mass index (BMI) was associated (all P<.05) with lower total lung capacity (TLC) (-38.7 mL x kg(-1) x m(-2)), functional residual capacity (FRC) (-73.9 mL x kg(-1) x m(-2)), residual volume (RV) (-40.4 mL x kg(-1) x m(-2)), and expiratory reserve volume (ERV) (-32.2 mL x kg(-1) x m(-2)). The effect of BMI on RV was most pronounced in quadriplegia (-72 mL x kg(-1) x m(-2)). Lifetime smoking was associated with a greater FRC (5.3 mL/pack-year) and RV (3.1 mL/pack-years). The effects of lifetime smoking were also greatest in quadriplegia (11 mL/pack-year for FRC; 7.8 mL/pack-year for RV). Time since injury, independent of age, was associated with a decrease in TLC, FRC, ERV, and RV (P<.05). Age was not a predictor of TLC once time since injury was considered.

CONCLUSIONS

Determinants of FRC, TLC, ERV, and RV in chronic SCI include factors related and unrelated to SCI. The mechanisms remain to be determined but likely involve the elastic properties and muscle function of the respiratory system and perhaps the effects of systemic inflammation related to adiposity. Addressing modifiable factors such as obesity, muscle stiffness, and smoking may improve respiratory morbidity and mortality in SCI by improving pulmonary function.

Association between mobility mode and C-reactive protein levels in men with chronic spinal cord injury

OBJECTIVE

To assess clinical determinants of systemic inflammation in persons with chronic spinal cord injury (SCI).

DESIGN

Cross-sectional survey.

SETTING

Veterans Affairs medical center.

PARTICIPANTS

As part of an epidemiologic study assessing SCI-related health conditions, 63 men with chronic SCI provided a blood sample and information regarding locomotive mode and personal habits.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURE

Plasma high-sensitivity C-reactive protein (CRP).

RESULTS

The mean +/- standard deviation age was 56+/-14y, and participants were assessed 21+/-13y after injury. Adjusting for heart disease, hypertension, and body mass index (BMI), the mean CRP in 12 motorized wheelchair users (5.11mg/L) was not significantly greater than 23 participants who used a manual wheelchair (2.19mg/L) (P=.085) but was significantly greater than the 17 who walked with an assistive device (1.41mg/L) (P=.005) and the 12 who walked independently (1.63mg/L) (P=.027). CRP was significantly greater in participants with obesity but was not related to age, smoking, or SCI level and severity. CRP was elevated in participants reporting a urinary tract infection (UTI) or pressure ulcer within a year, but adjustment for this did not account for the elevated CRP in motorized wheelchair users.

CONCLUSIONS

These results suggest that CRP in chronic SCI is independently related to locomotive mode, BMI, and a history of pressure ulcers and UTI. It is suggested that future studies in SCI investigate whether modifying these factors influence systemic inflammation and cardiovascular health.

Association of elevated alanine aminotransferase with BMI and diabetes in older veteran outpatients

AIMS

We determined age-stratified associations of elevated alanine aminotransferase (ALAT), as a likely indicator of non-alcoholic steatonecrosis (NASH), with body mass index (BMI) and diabetes among older male, veteran outpatients.

METHODS

Administrative files at three, urban Veteran Administration Healthcare Systems were surveyed. 47.916 male outpatients, mean age 69.2 years, without ICD-9CM documented liver diseases or alcohol- or drug-abuse, had the highest of one or more ALAT results analyzed. Logistic regression determined age-stratified predictors of elevated ALAT among groups <50 ("younger-aged"), 50-70 ("middle-aged") and >70 ("older-aged"). BMI, diabetes and "statin" use were predictors.

RESULTS

1272 (2.7%) patients had their highest ALAT greater than the upper limit of normal (mean=98.4U/L). 4.9% of all diabetics and 15.2, 6.1 and 3.4% of the "younger-aged", middle-aged" and "older-aged" diabetics, respectively, had elevated ALATs. BMI values from 25 to <30 and 30 to 45, compared with those from >20 to <25, predicted elevated ALAT in the two younger groups. In contrast, among "older-aged" patients, diabetes independently predicted elevated ALAT, whereas BMI did not.

CONCLUSIONS

The cross-sectional findings support previous epidemiologic data associating ALAT elevations with obesity and diabetes, and provide new evidence that in elderly male patients, ALAT elevations, absent other known liver diseases, may selectively associate with diabetes.

Longitudinal change in FEV1 and FVC in chronic spinal cord injury

RATIONALE

Although respiratory dysfunction is common in chronic spinal cord injury (SCI), determinants of longitudinal change in FEV(1) and FVC have not been assessed.

OBJECTIVES

Determine factors that influence longitudinal lung function decline in SCI.

METHODS

A total of 174 male participants (mean age of 49 and 17 yr after injury) completed a respiratory questionnaire and underwent spirometry over an average follow-up of 7.5 years (range, 4-14 yr).

MEASUREMENTS AND MAIN RESULTS

In multivariate models, longitudinal decline in FEV(1) was significantly related to continued smoking, persistent wheeze, an increase in body mass index, and respiratory muscle strength. Aging was associated with an accelerated decline in FEV(1) (for ages <40, 40-60, >60 yr: -27, -37, and -71 ml/yr, respectively). Similar effects were observed for FVC.

CONCLUSIONS

Longitudinal change in FEV(1) and FVC was not directly related to level and severity of SCI, but was attributable to potentially modifiable factors in addition to age. These results suggest that weight control, smoking cessation, trials directed at the recognition and treatment of wheeze, and efforts to improve respiratory muscle strength may slow lung function decline after SCI.

Matrix Gla protein is associated with risk factors for atherosclerosis but not with coronary artery calcification

OBJECTIVE

Atherosclerotic coronary artery calcification (CAC) is associated with increased coronary heart disease (CHD) risk. Matrix Gla protein (MGP) is an inhibitor of calcification in vivo. However, little is known regarding the distribution of circulating MGP and its associations with CHD risk factors or with CAC in humans.

METHODS AND RESULTS

Serum MGP concentrations were determined in 2 independent populations of men and women free of clinically apparent cardiovascular disease: study A, n=316, mean age 58 years, and study B, n=452, mean age 68 years. CAC was determined by computed tomography. Mean MGP concentrations were 98.4 and 198 ng/mL in men, and 97.4 and 201 ng/mL in women, in study A and study B, respectively. In both cohorts, MGP levels were higher with increasing age. In age-adjusted analyses, there was an association of circulating MGP with increasing Framingham CHD risk score (in study A, P=0.003 in men and P=0.016 in women, respectively; in study B, a nonsignificant increase in men and P=0.05 in women, respectively). Significant associations of circulating MGP with high-density lipoprotein and other individual CHD risk factors were also noted in both cohorts. There were no consistent associations between MGP and CAC after adjustment for CHD risk score in the 2 cohorts.

CONCLUSIONS

MGP is associated with individual CHD risk factors and the Framingham CHD risk score in men and women free of clinically apparent CHD. The relation of MGP with CAC deserves further study in larger populations.

Health care utilization and receipt of cholesterol testing by veterans with and those without mental illness

OBJECTIVE

We examined the relationship between mental illness, health care utilization and rates of cholesterol testing.

METHODS

We conducted a retrospective cohort study using Veterans Affairs (VA) administrative data on 64,490 United States veterans who used VA New England Health Care System outpatient services between January 1998 and June 2001. A total of 10,100 veterans (15.7%) had a mental illness treated with medication. We examined the interaction between mental illness and outpatient service utilization with respect to the likelihood of receiving a cholesterol test, adjusting for major demographic and clinical covariates.

RESULTS

Among veterans using VA outpatient services infrequently, those with mental illness were less likely than non-mentally ill control subjects to receive a cholesterol test during the study period (first quartile adjusted OR=0.45, 95% CI=0.37-0.54; second quartile adjusted OR=0.50, 95% CI=0.45-0.57). Mentally ill subjects with more frequent utilization of VA services were as likely as (third quartile adjusted OR=1.01, 95% CI=0.91-1.13) or more likely than (fourth quartile adjusted OR=2.73, 95% CI=2.46-3.03) non-mentally ill subjects to receive cholesterol testing.

CONCLUSIONS

Mental illness was associated with a lower likelihood of cholesterol testing in subjects who used fewer VA outpatient services. The observed disparity attenuated at higher levels of service utilization.

Association of dietary and biochemical measures of vitamin K with quantitative ultrasound of the heel in men and women

INTRODUCTION

Low vitamin K nutritional status is associated with increased fracture risk but is inconsistently related to bone mineral density (BMD), suggesting that vitamin K may affect components of bone strength not measured by BMD, such as microarchitecture. Quantitative ultrasound (QUS) may assess trabecular orientation, providing information on the mechanical properties of bone and may serve as a potential alternative to BMD for gaining insight to the relation between vitamin K and bone strength. We therefore examined the association of vitamin K nutritional status measured in several different ways with QUS in men and women who participated in the Framingham Osteoporosis Study.

METHODS

From 1996 to 2001, broadband ultrasound attenuation (BUA) and speed of sound (SOS) of the calcaneus (heel) were measured in 583 men and 768 women (mean age 59 years). Vitamin K nutritional status was assessed between 1995 and 1998 by three separate measures: plasma phylloquinone concentration, serum percent undercarboxylated osteocalcin (%ucOC) and dietary vitamin K intake. Multiple linear regression analyses were used to calculate regression coefficients in order to evaluate the associations between both measures of QUS and each measure of vitamin K nutritional status. Regression analyses were conducted separately for subgroups of participants defined by gender, menopause status and current use of estrogen replacement medication.

RESULTS

Among the men, plasma phylloquinone concentration was positively associated with both BUA (P<0.01) and SOS (P=0.02) of the heel. Neither serum %ucOC nor dietary vitamin K intake, however, was associated with QUS measures. Among women, none of the three measures of vitamin K nutritional status were associated with either BUA or SOS, regardless of menopause status or use of estrogen. Although QUS is associated with vitamin K nutritional status in men, the observed relation was not consistent among subgroups of participants.

CONCLUSION

These findings suggest that QUS may not be the best method for elucidating the role of vitamin K on the skeleton.

Characterizations of long-term anxiolytic benzodiazepine prescriptions in veteran patients

To characterize long-term prescriptions for commonly prescribed anxiolytic benzodiazepines to veteran patients and to identify factors associated with high daily doses, we analyzed the linked pharmacy and administrative databases from New England Veterans Healthcare System (VISN 1). We analyzed treatment episodes of 3 months or longer with the 4 most commonly prescribed agents: alprazolam, clonazepam, diazepam, and lorazepam. Descriptive statistics and univariate and multivariate analyses described the sample and tested associations of pharmacological and clinical variables for patients prescribed the top 5% of average daily doses ("high-dose" prescriptions). From 16,630 full or partial treatment episodes for all 4 agents analyzed within a 42-month window, average daily doses were predominantly moderate, age-sensitive, and stable; refill lag intervals were short. Patients on "high-dose" prescriptions for the 4 agents combined, compared with "middle quartile" dose prescriptions, in adjusted analyses, were younger, more likely to have posttraumatic stress disorder (odds ratio [OR], 2.6; 95% confidence interval [CI], 2.17-3.13), substance abuse (OR, 1.50; 95% CI, 1.25-1.80), and anxiety (OR, 1.33; 95% CI, 1.11-1.60) and were more likely to be receiving concurrent oxycodone/acetaminophen (OR, 2.05; 95% CI, 1.64-2.56), anxiolytic benzodiazepine (OR, 1.51; 95% CI, 1.12-2.03), antidepressant (OR, 2.15; 95% CI, 1.80-2.58), and neuroleptic (OR, 2.03; 95% CI, 1.69-2.44) prescriptions. These results indicate that veteran patients prescribed anxiolytic benzodiazepines typically receive modest, nonincreasing doses over long-term treatment episodes. However, those on the highest average daily doses, typically more than recommended guidelines, are more likely to have clinical diagnoses and concurrent prescriptions for psychoactive medications indicative of more complex and, perhaps, problematic management.

Three complementary definitions of polypharmacy: methods, application and comparison of findings in a large prescription database

PURPOSE

To develop and compare three operational definitions of polypharmacy using a large prescription database.

METHODS

We defined Cumulative polypharmacy as all prescriptions filled during a 178 day window--which captured 95% of eventual refills as calculated from Kaplan-Meier and cumulative incidence curves. Continuous polypharmacy was all prescriptions filled in two such windows 6 months apart. Simultaneous polypharmacy was the number of prescriptions active on a particular day, as determined by fill dates and amount of medication given. We applied these definitions to the outpatient prescription files of New England veterans and compared the resulting estimates of polypharmacy using descriptive statistics.

RESULTS

118,013 patients received at least one prescription between January 1998 and July 1999. Cumulative polypharmacy averaged 3.54 (SD = 4.95) medications and continuous polypharmacy averaged 1.96 (SD = 3.23). Examination of simultaneous polypharmacy over 40 2-week intervals revealed an average of 2.63 (CI 2.61-2.65), a minimum of 1.09 (CI 1.08-1.10) and maximum of 4.94 (CI 4.92-4.96). One arbitrarily selected observation point had an average of 3.87 (SD = 3.17).

CONCLUSIONS

Our definitions of cumulative and continuous polypharmacy serve to set upper and lower bounds for the estimate of polypharmacy. Our method for simultaneous polypharmacy gives numbers that diverge in some respects, but it is better at showing transient changes in medications. The methods are complementary and allow exploration of various aspects of medication use, such as cumulative medication exposure over time, the influence of chronic medical problems, and the causes of rapid changes in medications.

Characterizations of long-term oxycodone/acetaminophen prescriptions in veteran patients

BACKGROUND

Long-term management of chronic pain with opioids may be stable over time or may be complicated by problematic dose increases, drug dependencies, and toxic effects. To determine clinical contexts in which stability or problems may occur, we examined the pharmacologic and clinical correlates of long-term prescriptions of oxycodone/acetaminophen, a commonly prescribed short-acting opioid formulation.

METHODS

We analyzed linked, archival outpatient pharmacy and clinical databases from the New England Veterans Integrated Service Network between January 1, 1998, and June 30, 2001. Durations, doses, and dose changes of oxycodone/acetaminophen prescriptions and concurrent use of long-acting opioids, benzodiazepines, tricyclic antidepressants, and anticonvulsants were determined.

RESULTS

In aggregate, 2195 patients (31% with cancer diagnoses per the International Classification of Diseases, Ninth Revision, Clinical Modification) received oxycodone/acetaminophen for more than 9 months at a mean prescribed daily dose of 3.9 tablets per day (range, 0.5-13.0 tablets per day) with minimal changes in daily prescribed mean dose over time. Patients with cancer were more likely than other patients to receive concurrent long-acting opioids. For patients without cancer, a higher mean daily dose was associated with duration, older age, human immunodeficiency virus (HIV) and/or AIDS, and with prescribed benzodiazepines and long-acting opioids; concurrent benzodiazepine prescriptions were associated with anticonvulsant prescriptions and with psychogenic pain and alcohol abuse and/or dependence diagnoses.

CONCLUSIONS

In veteran patients who received long-term oxycodone/acetaminophen prescriptions, mean daily doses were typically modest and stable, likely reflecting a selection of patients with successful, long-term management. Among patients without cancer, however, associations of higher oxycodone/acetaminophen doses with benzodiazepine prescriptions, psychogenic pain, alcohol abuse, and HIV/AIDS may portend opioid prescription management problems.

Patterns of Dispensed Disulfiram and Naltrexone for Alcoholism Treatment in a Veteran Patient Population

BACKGROUND

Short-term treatment trials indicate that two Food and Drug Administration-approved agents, disulfiram and naltrexone, may each curtail alcohol consumption, but two large 1-year Veterans Administration cooperative studies showed no long-term benefits for these agents over placebo. To assess whether these agents are being prescribed for extended periods, as an indicator of long-term use in nonexperimental settings, we compared dispensing patterns in a veteran patient population.

METHODS

The New England Veterans Integrated Service Network outpatient pharmacy files between January 1, 1998, and June 30, 2001, were analyzed; only patients with prescriptions on or after March 1, 1998, were included. Measurements for each patient included data on new and refilled prescriptions of disulfiram, naltrexone, and control medications. Prescription survival curves with right censoring were constructed. Distinct treatment episodes were defined by having six or more months between the end date of a prior prescription and the start date of a new prescription.

RESULTS

From eight New England Veterans Integrated Service Network centers, 754 patients were dispensed disulfiram, and 971 were dispensed naltrexone, encompassing 873 and 1075 treatment episodes, respectively. Treatment episode durations were virtually identical for both drugs: more than 35% of episodes were 1 month or shorter, more than 50% were 2 months or shorter, and 75% were 5 months or shorter. Concurrently prescribed neuroleptic or statin medications predicted longer disulfiram and naltrexone treatment episodes. However, for patients newly prescribed common neuroleptic, antidepressant, or statin agents, the risks for discontinuing disulfiram or naltrexone were 1.4 to 2.3 times greater than for discontinuing these other agents.

CONCLUSIONS

In clinical settings, veteran patients were likely to be dispensed either disulfiram or naltrexone for only several months or less. The contexts and reasons for these predominantly short-term treatment episodes or the benefits derived were not known and merit further study.

Risk factors for central serous chorioretinopathy: a case-control study

PURPOSE

To identify systemic factors associated with the development of central serous chorioretinopathy (CSCR).

DESIGN

Retrospective, case-control study.

PARTICIPANTS AND CONTROLS

312 cases and 312 controls.

RESULTS

By use of a multivariate analysis, the previously described risk factors, systemic steroid use (odds ratio [OR], 37.1; 95% confidence interval [CI], 6.2-221.8), and pregnancy (OR, 7.1; 95% CI, 1.0-50.7), were strongly associated with CSCR. Additional risk factors identified by this study include antibiotic use (OR, 6.2; 95% CI, 1.0-37.9), alcohol use (OR, 4.9; 95% CI, 1.5-16.3), untreated hypertension (OR, 3.3; 95% CI, 1.3-8.5), and allergic respiratory disease (OR, 2.5; 95% CI, 1.2-5.1).

CONCLUSIONS

A wide variety of systemic factors are associated with CSCR. Prospective studies are warranted to evaluate the nature and significance of these associations further.

Vitamin K intake and bone mineral density in women and men

BACKGROUND

Low dietary vitamin K intake has been associated with an increased risk of hip fracture in men and women. Few data exist on the association between dietary vitamin K intake and bone mineral density (BMD).

OBJECTIVE

We studied cross-sectional associations between self-reported dietary vitamin K intake and BMD of the hip and spine in men and women aged 29-86 y.

DESIGN

BMD was measured at the hip and spine in 1112 men and 1479 women (macro x +/- SD age: 59 +/- 9 y) who participated in the Framingham Heart Study (1996-2000). Dietary and supplemental intakes of vitamin K were assessed with the use of a food-frequency questionnaire. Additional covariates included age, body mass index, smoking status, alcohol use, physical activity score, and menopause status and current estrogen use among the women.

RESULTS

Women in the lowest quartile of vitamin K intake (macro x: 70.2 microg/d) had significantly (P < or = 0.005) lower mean (+/- SEM) BMD at the femoral neck (0.854 +/- 0.006 g/cm(2)) and spine (1.140 +/- 0.010 g/cm(2)) than did those in the highest quartile of vitamin K intake (macro x: 309 microg/d): 0.888 +/- 0.006 and 1.190 +/- 0.010 g/cm(2), respectively. These associations remained after potential confounders were controlled for and after stratification by age or supplement use. No significant association was found between dietary vitamin K intake and BMD in men.

CONCLUSIONS

Low dietary vitamin K intake was associated with low BMD in women, consistent with previous reports that low dietary vitamin K intake is associated with an increased risk of hip fracture. In contrast, there was no association between dietary vitamin K intake and BMD in men.

The analysis of survival data with a non-susceptible fraction and dual censoring mechanisms

It is known that the ages of onset of many diseases are determined by both a genetic predisposition to disease as well as environmental risk factors that are capable of either triggering or hastening the onset of disease. Difficulties in modelling onset ages arise when a large fraction fail to inherit the disease-causing gene, and multiple reasons for censoring result in unobserved onset ages. We present a parametric Bayesian model that includes subjects with missing age information, non-susceptible subjects and allows for regression on risk factor information. The model is fit using Markov chain Monte Carlo simulation from the posterior distribution, and allows the simultaneous estimation of the proportion of the population at risk of disease, the mean onset age of disease, survival after disease onset, and the association of risk factors with susceptibility, onset age and survival after onset. An example employing Huntington's disease data is presented.

Insulin-like growth factor binding proteins 4 and 5 and bone mineral density in elderly men and women

Insulin-like growth factor-1 (IGF-I) plays a central role in the maintenance of bone mass. To test whether two major IGF-I binding proteins, IGFBP-4 and IGFBP-5, are related to bone mineral density (BMD), we studied a sample of the Framingham Offspring Cohort participants (99 men and 101 women, ages 60-87). Serum levels of IGF-I, IGFBP-4, and IGFBP-5 were measured by previously validated radioimmunoassays (CVs approximately 10%). BMDs of the proximal femur and lumbar spine were measured using a Lunar DPX-L densitometer. In males, but not females, IGF-I and IGFBP-5 were inversely associated with age (r = 0.34 and r = -0.28, respectively; P <0.01), while IGFBP-4 levels were positively associated with age (P <0.01). Multivariate means for BMD (adjusted for age, body mass index, height, smoking, and in women, estrogen use) were computed across quartiles of IGFBP-4 and IGFBP-5 and IGFBP-4/IGFBP-5 ratio. In women, but not men, IGFBP-5 was positively associated with femoral neck BMD (P = 0.03), however, after statistical adjustment for IGF-I, this association was no longer significant. No other associations were observed for BMD at any other site. Further study is necessary for elucidation of the gender differences in the possible influence of IGF system components on bone mass.

Modeling the effects of genetic factors on late-onset diseases in cohort studies

Many late-onset diseases are caused by what appears to be a combination of a genetic predisposition to disease and environmental factors. The use of existing cohort studies provides an opportunity to infer genetic predisposition to disease on a representative sample of a study population, now that many such studies are gathering genetic information on the participants. One feature to using existing cohorts is that subjects may be censored due to death prior to genetic sampling, thereby adding a layer of complexity to the analysis. We develop a statistical framework to infer parameters of a latent variables model for disease onset. The latent variables model describes the role of genetic and modifiable risk factors on the onset ages of multiple diseases, and accounts for right-censoring of disease onset ages. The framework also allows for missing genetic information by inferring a subject's unknown genotype through appropriately incorporated covariate information. The model is applied to data gathered in the Framingham Heart Study for measuring the effect of different Apo-E genotypes on the occurrence of various cardiovascular disease events.

Genome screen for quantitative trait loci contributing to normal variation in bone mineral density: the Framingham Study

A genome-wide scan was performed in a randomly ascertained set of 330 extended families from the population-based Framingham Study to identify chromosomal regions possibly linked to bone mineral density (BMD). A set of 401 microsatellite markers was typed at a 10-centimorgan (cM) average density throughout the genome. BMD was measured at the femoral neck, trochanter, Ward's area, and lumbar spine in 1557 participants of both Framingham cohorts. BMDs were adjusted for age, body mass index (BMI), height, alcohol, caffeine, calcium and vitamin D intakes, smoking, physical activity, and estrogen use in women within each sex and cohort. Strong heritabilities (values between 0.543 and 0.633) were found for the adjusted BMD at all sites. Two-point and multipoint quantitative linkage analyses were performed for each BMD site using the maximum likelihood variance components method. By two-point screening, loci of suggestive linkage were identified on chromosomes 6 and 21, with the maximum log10 of the odds ratio (LOD) scores of 2.34 for the trochanter at D21S1446 and 2.93 for the femoral neck at D6S2427. Lumbar spine BMD had maxima at D6S2427 (LOD = 1.88) and at D12S395 (LOD = 2.08). Multipoint linkage analysis revealed suggestive linkage of trochanteric BMD at a broad (approximately 20 cM) interval on chromosome 21q, with the peak linkage close to D21S1446 (LOD = 3.14). LOD scores were 2.13 at 8q24 with Ward's BMD and 1.92 at 14q21.3 with lumbar spine BMD. This largest genome screen to date for genes underlying normal variation in BMD, adjusted for a large number of covariates, will help to identify new positional candidate genes, otherwise unrecognized.

Bone marrow aspiration, biopsy, and culture in the evaluation of HIV-infected patients for invasive mycobacteria and histoplasma infections

Bone marrow (BM) aspiration and biopsy are used commonly in clinical practice to diagnose invasive tissue infections caused by Mycobacterium avium intracellulare (MAC), Mycobacterium tuberculosis (TB), and Histoplasma capsulatum (HC) in patients with human immunodeficiency virus-1 (HIV) infection. However, the value of these invasive procedures relative to other diagnostic approaches has not been clearly defined. To determine the value of BM culture and BM histology in the diagnosis of opportunistic MAC/TB and HC infections in immunosuppressed patients with HIV, we retrospectively reviewed the records of 56 adult patients with HIV who underwent a single BM aspiration, biopsy, and culture because of unexplained fever and/or other clinical features suggestive of MAC/TB or HC infection. Thirty-two patients (57%) were ultimately diagnosed with MAC/TB or HC infection by positive cultures of BM, blood, sputum, or bronchoalveolar lavage fluid or by the histologic detection of organisms in biopsies of BM or other tissues. The diagnostic sensitivity of BM cultures was equal to that of blood cultures (20/32, or 63%). Granuloma and/or histologically apparent organisms were seen in BM biopsy specimens in 11 of 32 individuals (34%) ultimately diagnosed with MAC/TB or HC infections. Among these 11 cases, both granuloma and acid-fast staining organisms were found in the BM biopsy specimens of 2 individuals for whom both BM and blood cultures were negative. Certain clinical symptoms and signs at the time of BM examination were found by logistic regression analysis to be significantly associated with a subsequent diagnosis of MAC/TB or HC infections; these included high fever, long duration of febrile days prior to BM examination, and elevated direct bilirubin. In conclusion, while the diagnostic sensitivity of BM cultures was found to be no greater than that of blood cultures in detecting MAC/TB or HC infections in immunosuppressed HIV+ patients, histopathologic examination of BM specimens resulted in the relatively rapid identification of nearly one third of infected patients who underwent BM examination, and also identified infections in some patients who were culture negative. These findings support the continued use of BM aspiration, biopsy, and culture for the diagnosis of opportunistic MAC/TB or HC infections in immunosuppressed HIV+ patients, particularly when selected clinical features are present.

Dietary vitamin K intakes are associated with hip fracture but not with bone mineral density in elderly men and women

BACKGROUND

Vitamin K has been associated with bone mineral density (BMD) and risk of hip fracture. The apolipoprotein (apo) E4 allele (APOE*E4) has been associated with bone fracture through a putative effect on vitamin K transport in blood.

OBJECTIVE

The objective was to determine the associations between vitamin K intake, apo E genotype, BMD, and hip fracture in a population-based cohort of elderly men and women.

DESIGN

Dietary vitamin K intake was assessed with a food-frequency questionnaire in 335 men and 553 women (average age: 75.2 y) participating in the Framingham Heart Study in 1988-1989. Incidence of hip fractures was recorded from 1988 to 1995. BMD at the hip, spine, and arm was assessed on 2 separate occasions (1988-1989 and 1992-1993). Comparisons between apo E genotype and BMD were made relative to E4 allele status (at least 1 epsilon4 allele compared with no epsilon4 allele).

RESULTS

Individuals in the highest quartile of vitamin K intake (median: 254 microg/d) had a significantly lower fully adjusted relative risk (0.35; 95% CI: 0. 13, 0.94) of hip fracture than did those in the lowest quartile of intake (median: 56 microg/d). There were no associations between vitamin K intake and BMD in either men or women. No association was found between the E4 allele and BMD, and there were no significant interactions between the E4 allele and phylloquinone intake and BMD or hip fracture.

CONCLUSIONS

Low vitamin K intakes were associated with an increased incidence of hip fractures in this cohort of elderly men and women. Neither low vitamin K intake nor E4 allele status was associated with low BMD.

Characteristics of nurse-midwife patients and visits, 1991

OBJECTIVES

This study describes the patient populations served by and visits made to certified nurse-midwives (CNMs) in the United States.

METHODS

Prospective data on 16,729 visits were collected from 369 CNMs randomly selected from a 1991 population survey. Population estimates were derived from a multistage survey design with probability sampling.

RESULTS

We estimated that approximately 5.4 million visits were made to nearly 3000 CNMs nationwide in 1991. Most visits involved maternity care, although fully 20% were for care outside the maternity cycle. Patients considered vulnerable to poor access or outcomes made 7 of every 10 visits.

CONCLUSIONS

Nurse-midwives substantially contribute to the health care of women nationwide, especially for vulnerable populations.