Racial differences in osteoarthritis pain and function: potential explanatory factors.

OBJECTIVE: This study examined factors underlying racial differences in pain and function among patients with hip and/or knee osteoarthritis (OA). METHODS: Participants were n=491 African Americans and Caucasians enrolled in a clinical trial of telephone-based OA self-management. Arthritis Impact Measurement Scales-2 (AIMS2) pain and function subscales were obtained at baseline. Potential explanatory variables included arthritis self-efficacy, AIMS2 affect subscale, problem- and emotion-focused pain coping, demographic characteristics, body mass index, self-reported health, joint(s) with OA, symptom duration, pain medication use, current exercise, and AIMS2 pain subscale (in models of function). Variables associated with both race and pain or function, and which reduced the association of race with pain or function by >or=10%, were included in final multivariable models. RESULTS: In simple linear regression models, African Americans had worse scores than Caucasians on AIMS2 pain (B=0.65, P=0.001) and function (B=0.59, P<0.001) subscales. In multivariable models race was no longer associated with pain (B=0.03, P=0.874) or function (B=0.07, P=0.509), indicating these associations were accounted for by other covariates. Variables associated with worse AIMS2 pain and function were: worse AIMS2 affect scores, greater emotion-focused coping, lower arthritis self-efficacy, and fair or poor self-reported health. AIMS2 pain scores were also significantly associated with AIMS2 function. CONCLUSION: Factors explaining racial differences in pain and function were largely psychological, including arthritis self-efficacy, affect, and use of emotion-focused coping. Self-management and psychological interventions can influence these factors, and greater dissemination among African Americans may be a key step toward reducing racial disparities in pain and function.

Daily pain variations among patients with hand, hip, and knee osteoarthritis.

OBJECTIVE: This study examined within-day osteoarthritis (OA)-related pain patterns and associated patient characteristics. METHODS: Participants with physician diagnoses and self-reported symptoms of hand (N=40), hip (N=32), and knee (N=85) OA recorded pain using a handheld computer on one weekday and one weekend day, with ratings beginning immediately after waking, then approximately every 2h following. Pain was rated on a sliding visual analog scale with hidden coding of 1-100. Multivariable linear mixed models examined associations of patient demographic characteristics, enrollment site (Durham VA Medical Center vs Duke University Medical Center), joint site, body mass index, and pain medication use with within-day pain range (maximum minus minimum pain rating) and area under the curve (AUC) of pain ratings, which incorporates the magnitude of all pain measurements. RESULTS: Pain patterns differed substantially across individuals. The sample means of the average, maximum, and minimum weekday pain scores were 35.3, 54.4, and 17.9, respectively. The mean pain range was 36.4, and the mean pain AUC was 564.3 (possible range: 16-1600). Pain scores were similar on weekends. In multivariable mixed models, both knee and hip OA were associated with a greater within-day pain range than hand OA. Only VA enrollment site was associated with a significantly greater pain AUC. CONCLUSION: There is substantial within-day range in OA-related pain. Both pain range and overall within-day magnitude vary according to patient characteristics. Patients' records of within-day pain patterns could be used in clinical encounters to tailor recommendations for the timing of medication use and behavioral strategies.

The National Registry of Veterans with amyotrophic lateral sclerosis.

BACKGROUND: The Department of Veterans Affairs (VA) Cooperative Studies Program has established a National Registry of Veterans with Amyotrophic Lateral Sclerosis (ALS). This article describes the objectives, methods, and sample involved in the registry. METHODS: United States military veterans with ALS were identified through national VA electronic medical record databases and nationwide publicity efforts for an enrollment period of 4 1/2 years. Diagnoses were confirmed by medical record reviews. Registrants were asked to participate in a DNA bank. Follow-up telephone interviews are conducted every 6 months to track participants' health status. RESULTS: As of September 30, 2007, 2,400 veterans had consented to participate in the registry, 2,068 were included after medical record review, 995 were still living and actively participating, and 1,573 consented to participate in the DNA bank. 979 participants had been enrolled in the registry for at least 1 year, 497 for at least 2 years, and 205 for at least 3 years. Fourteen studies have been approved to use registry data for epidemiological, observational, and interventional protocols. CONCLUSION: This registry has proven to be a successful model for identifying large numbers of patients with a relatively rare disease and enrolling them into multiple studies, including genetic protocols.

Occurrence of amyotrophic lateral sclerosis among Gulf War veterans.

BACKGROUND: In response to Gulf War veterans' concerns of high rates of ALS, this investigation sought to determine if Gulf War veterans have an elevated rate of ALS. METHODS: A nationwide epidemiologic case ascertainment study design was used to ascertain all occurrences of ALS for the 10-year period since August 1990 among active duty military and mobilized Reserves, including National Guard, who served during the Gulf War (August 2, 1990, through July 31, 1991). The diagnosis of ALS was confirmed by medical record review. Risk was assessed by the age-adjusted, average, annual 10-year cumulative incidence rate. RESULTS: Among approximately 2.5 million eligible military personnel, 107 confirmed cases of ALS were identified for an overall occurrence of 0.43 per 100,000 persons per year. A significant elevated risk of ALS occurred among all deployed personnel (RR = 1.92; 95% CL = 1.29, 2.84), deployed active duty military (RR = 2.15, 95% CL = 1.38, 3.36), deployed Air Force (RR = 2.68, 95% CL = 1.24, 5.78), and deployed Army (RR = 2.04; 95% CL = 1.10, 3.77) personnel. Elevated, but nonsignificant, risks were observed for deployed Reserves and National Guard (RR = 2.50; 95% CL = 0.88, 7.07), deployed Navy (RR = 1.48, 95% CL = 0.62, 3.57), and deployed Marine Corps (RR = 1.13; 95% CL = 0.27, 4.79) personnel. Overall, the attributable risk associated with deployment was 18% (95% CL = 4.9%, 29.4%). CONCLUSIONS: Military personnel who were deployed to the Gulf Region during the Gulf War period experienced a greater post-war risk of ALS than those who were not deployed to the Gulf.

Detection and localization of a single binary trait locus in experimental populations.

The advancements made in molecular technology coupled with statistical methodology have led to the successful detection and location of genomic regions (quantitative trait loci; QTL) associated with quantitative traits. Binary traits (e.g. susceptibility/resistance), while not quantitative in nature, are equally important for the purpose of detecting and locating significant associations with genomic regions. Existing interval regression methods used in binary trait analysis are adapted from quantitative trait analysis and the tests for regression coefficients are tests of effect, not detection. Additionally, estimates of recombination that fail to take into account varying penetrance perform poorly when penetrance is incomplete. In this work a complete probability model for binary trait data is developed allowing for unbiased estimation of both penetrance and recombination between a genetic marker locus and a binary trait locus for backcross and F2 experimental designs. The regression model is reparameterized allowing for tests of detection. Extensive simulations were conducted to assess the performance of estimation and testing in the proposed parameterization. The proposed parameterization was compared with interval regression via simulation. The results indicate that our parameterization shows equivalent estimation capabilities, requires less computational effort and works well with only a single marker.