Whitney E. Zahnd

Health Literacy Skills in Rural and Urban Populations.

OBJECTIVE To determine whether health literacy is lower in rural populations. METHOD We analyzed health, prose, document, and quantitative literacy from the National Assessment of Adult Literacy study. Metropolitan Statistical Area designated participants as rural or urban. RESULTS Rural populations had lower literacy levels for all literacy types (P<0.001 for each). After adjusting for known confounders, there was no longer a difference in health or prose literacy (P>0.05). However, rural populations had higher document (P=0.04) and quantitative (P=0.01) literacy. CONCLUSION Health literacy is lower in the rural population although this difference is explained by known confounders.

Outcomes in Patients with Rheumatoid Arthritis and Myocardial Infarction

BACKGROUND Patients with rheumatoid arthritis have an increased risk for accelerated atherosclerosis. It is unknown, however, whether rheumatoid arthritis also increases in-hospital mortality after a myocardial infarction or influences the therapy patients receive. METHODS A cross-sectional analysis of 1,112,676 patients with myocardial infarction in the 2003-2005 Nationwide Inpatient Sample was performed. RESULTS Patients with rheumatoid arthritis were 39% more likely to receive medical therapy (odds ratio [OR], 1.39; 95% confidence interval [CI], 1.30-1.49) than interventional therapy. By using logistic regression, we adjusted for confounding variables to determine the effect of rheumatoid arthritis on the selection of therapy and found that rheumatoid arthritis itself was associated with a 38% increased likelihood of undergoing thrombolysis (OR, 1.38; 95% CI, 1.10-1.71) and a 27% increased likelihood of undergoing percutaneous coronary intervention (OR, 1.27; 95% CI, 1.17-1.39). For the primary outcome measure, we determined that patients with rheumatoid arthritis overall had a 24% better in-hospital mortality compared with other patients with a myocardial infarction (OR, 0.76; 95% CI, 0.68-0.86), which was 34% better after adjusting for confounding variables (OR, 0.66; 95% CI, 0.59-0.74). This better in-hospital mortality was seen in patients with rheumatoid arthritis undergoing medical therapy (adjusted OR, 0.67; 95% CI, 0.59-0.75) and percutaneous coronary intervention (adjusted OR, 0.47; 95% CI, 0.32-0.70), but not in patients undergoing thrombolysis or coronary artery bypass grafting. CONCLUSIONS Among patients with myocardial infarction, rheumatoid arthritis was associated with an increased use of thrombolysis and percutaneous coronary intervention. Moreover, patients with rheumatoid arthritis had an in-hospital survival advantage, particularly those undergoing medical therapy and percutaneous coronary intervention.

Rural-urban differences in behaviors to prevent skin cancer: an analysis of the Health Information National Trends Survey.

BACKGROUND There is concern that rural residents may be less likely to engage in behaviors to reduce their risk for skin cancer compared with urban residents. OBJECTIVES First, we sought to determine whether rural residents are less likely to use sunscreen and engage in other skin cancer preventive measures. Second, we sought to determine whether such actions are sufficiently explained by factors known to affect these behaviors or whether such actions are affected by rurality. METHODS We analyzed the 2005 Health Information National Trends Survey, a survey of the noninstitutionalized, adult population performed by the National Cancer Institute. We used logistic regression analysis to adjust for confounding by age, race, income, education, health insurance, smoking, sex, marital status, and region. RESULTS Compared with urban residents, rural residents were 33% less likely (odds ratio = 0.67; 95% confidence interval, 0.57-0.80) to wear sunscreen when exposed to the sun for more than 1 hour. After adjusting for the above confounding variables, however, rural individuals were just as likely as urban individuals to use sunscreen with sun exposure. LIMITATIONS Inability to adjust for unmeasured confounding variables, such as occupational sun exposure, is a limitation. CONCLUSION Rural residents were less likely to use sunscreen. This decreased use of sunscreen, however, was explained by differences in age, race, income, education, and other confounding factors that negatively influence the use of sunscreen.

Joint replacement surgeries among medicare beneficiaries in rural compared with urban areas

OBJECTIVE People in rural areas live farther away from hospitals than do people in urban areas. Thus, there is concern that people living in rural areas may be less willing or able to undergo elective surgical procedures. This study was undertaken to determine whether Medicare beneficiaries in rural areas were less likely to have elective total knee or hip replacement surgeries compared with their urban counterparts. METHODS We performed a cross-sectional study of Medicare beneficiaries, controlling for age, sex, race/ethnicity, and economic status. Beneficiaries were assigned to rural versus urban areas based on their zip code of residence and the 10-point Rural-Urban Commuting Area designation. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. RESULTS Compared with urban beneficiaries, rural beneficiaries were 27% more likely to have total knee or hip replacement surgeries (OR 1.27 [95% CI 1.26-1.28]). After adjusting for age, sex, race/ethnicity, median household income, average house value, mean poverty ratio, and state of residence, rural beneficiaries were still 14% more likely to have total joint replacement surgeries (OR 1.14 [95% CI 1.13-1.16]). Differential use of surgery before and after receiving Medicare eligibility did not explain the findings. While significant sex, racial, and ethnic disparities were present in both rural and urban areas, for the most part these disparities were ameliorated rather than accentuated in rural areas. CONCLUSION Contrary to expectations, our findings indicate that Medicare beneficiaries living in rural areas are more likely to undergo total knee or hip replacement surgeries.

Rural-Urban Differences in Surgical Procedures for Medicare Beneficiaries

OBJECTIVE To determine whether Medicare beneficiaries in rural areas were less likely to undergo a variety of surgical procedures compared with their urban counterparts. DESIGN, SETTING, AND PATIENTS Cross-sectional study of Medicare beneficiaries. MAIN OUTCOME MEASURE Any incidence of the surgical procedures studied. RESULTS Compared with urban Medicare beneficiaries, rural Medicare beneficiaries were more likely to undergo a broad array of surgical procedures: 35% more likely for carotid endarterectomy (odds ratio [OR] = 1.35; 95% confidence interval [CI], 1.33-1.38), 32% for lumbar spine fusion (OR = 1.32; 95% CI, 1.29-1.35), 30% for knee replacement surgery (OR = 1.30; 95% CI, 1.28-1.31), 28% for abdominal aortic aneurysm repair (OR = 1.28; 95% CI, 1.24-1.31), 22% for prostatectomy (OR = 1.22; 95% CI, 1.19-1.24), 19% for hip replacement surgery (OR = 1.19; 95% CI, 1.17-1.21), 18% for aortic valve replacement (OR = 1.18; 95% CI, 1.14-1.21), 16% for open reduction and internal fixation of the femur (OR = 1.16; 95% CI, 1.14-1.18), and 15% for appendectomy (OR = 1.15; 95% CI, 1.11-1.19). To determine whether these differences could be explained by known confounding variables, we then used logistic regression to adjust for age, sex, race/ethnicity, median household income, average house value, mean poverty ratio, and state of residence. Rural beneficiaries were still more likely to undergo all of these surgical procedures. CONCLUSIONS Medicare beneficiaries living in rural areas were more likely to undergo a broad array of surgical procedures compared with those living in urban areas. While allaying some concern about rural access to surgical procedures, the uniformity of these results raises concern that people living in rural areas may have an overall poorer quality of health.

Impact of County Rurality and Urologist Density on Urological Cancer Mortality in Illinois

PURPOSE The urology work force is contracting at a time when service demand is increasing due to demographic changes, especially in rural areas. We investigated the impact of rural status and urologist density on kidney and renal pelvis, bladder and prostate cancer mortality at the county level in Illinois. MATERIALS AND METHODS We stratified the 102 Illinois counties by 2003 RUCCs as urban (36, RUCCs 1 to 3) and rural (66, RUCCs 4 to 9). Area Health Resource Files were used for county demographic data and urologist density. County level age adjusted mortality rates from 1990 to 2010 were derived from National Center for Health Statistics data using SEER*Stat. We examined the associations of urological cancer mortality rates with rural status and urologist density. RESULTS Average urologist density significantly differed between rural and urban counties (1.9 vs 3.4/100,000 population, p < 0.01). The kidney and renal pelvis cancer mortality rate in rural counties was higher than in urban counties while that of prostate cancer was lower (4.9 vs 4.3 and 28.7 vs 32.2/100,000 population, respectively, each p < 0.01). Urologist density correlated with the mortality rate of kidney and renal pelvis cancer (Pearson coefficient -0.33, p < 0.01) but not with the bladder or prostate cancer mortality rate. Multiple regression analysis revealed that rurality and lower urologist density (p = 0.01 and < 0.05) were significantly associated with higher kidney and renal pelvis cancer mortality. CONCLUSIONS Rural residence and low urologist density were associated with increased kidney and renal pelvis cancer mortality on the county level in Illinois. Further expansion and testing of evidence-based telemedicine is warranted because remote technical consultation is now technologically feasible, effective, inexpensive and satisfactory to patients.

Mortality after coronary artery revascularization of patients with rheumatoid arthritis.

OBJECTIVE Patients with rheumatoid arthritis have an increased risk for accelerated atherosclerosis. It is not known, however, whether this disorder is associated with a higher risk of complications after coronary artery revascularization. METHODS We conducted a cross-sectional study of patients in the 2003-2005 Nationwide Inpatient Sample. To determine whether patients with rheumatoid arthritis had higher in-hospital mortality after coronary artery revascularization, we used logistic regression to adjust for age, sex, race/ethnicity, income, rural-urban residency, diabetes, hypertension, hyperlipidemia, Charlson comorbidities (including myocardial infarction, congestive heart failure, and diabetes), elective admission, weekend admission, and primary payer. RESULTS Among patients undergoing coronary artery revascularization, those with rheumatoid arthritis were 49% less likely to die while hospitalized compared with those without rheumatoid arthritis (odds ratio, 0.51; 95% confidence interval, 0.40-0.65) after adjusting for the above confounders. In subgroup analyses that adjusted for the same confounders, patients with rheumatoid arthritis also had a 61% improvement of in-patient mortality when they underwent percutaneous coronary interventions (odds ratio, 0.39; 95% confidence interval, 0.29-0.54) along with a median of 0.32 less days hospitalized (95% confidence interval, 0.28-0.34 days). Similarly, patients with rheumatoid arthritis undergoing coronary artery bypass grafting had a 31% improvement of in-patient mortality (odds ratio, 0.69; 95% confidence interval, 0.48-0.99), with a median of 1.36 less days hospitalized (95% confidence interval, 0.72-1.12 days). CONCLUSION Among patients undergoing coronary artery revascularization, patients with rheumatoid arthritis have an in-hospital survival advantage along with reduced days of hospitalization compared with patients without rheumatoid arthritis.

Manuscript title: Geospatial analysis of Cancer risk and residential proximity to coal mines in Illinois

BACKGROUND Studies have indicated a population-level association between coal mining and cancer incidence and mortality, but few studies specifically examined residential proximity to this exposure using spatial analysis. We utilized a Geographic Information Systems (GIS) approach to perform spatial and statistical analyses to test two coal mining exposure variables and their associations with cancer incidence and mortality in Illinois--the fourth highest coal producing state in the United States. METHODS Data included age-adjusted county-level cancer incidence and mortality for five cancers: all malignant, lung, colorectal, breast (female) and prostate. Coal mining exposure was defined by two variables: coal production group and distance-weighted exposure. Spatial analyses were performed to identify spatial clustering. Correlation and stepwise regression analyses were performed to explore the relationship between cancer incidence and mortality and coal mining exposures. Covariates considered in regression analyses included socioeconomic deprivation, former/current smoking prevalence, race, and rurality. RESULTS Global spatial autocorrelation indicated significant spatial clustering of incidence, mortality and aggregated coal production. Distance-weighted exposure was significantly correlated with coal production group, age-adjusted all cancer incidence and age-adjusted all cancer mortality. Regression analyses indicated an association between recent coal production and colorectal cancer incidence (p=0.009) and mortality (p=0.035) and prostate cancer mortality (p=0.047). Distance weighted exposure was associated with lung cancer incidence (p=0.004) and mortality (p<0.001), and all cancer mortality (p<0.001). CONCLUSION Coal production, incidence and mortality are spatially clustered in Illinois. Exposures to coal mining were associated with elevated risk of multiple cancers, most notably lung and colorectal. The environmental impact of the mining industry is substantial, and exposure of individuals residing near coal mines to known carcinogens is plausible. Future studies are needed to further elucidate the population exposure dynamics of coal mining, and should be explored using individual-level exposures and cancer outcomes.

Effect of Number of Clinics and Panel Size on Patient Continuity for Medical Residents

BACKGROUND Accreditation Council for Graduate Medical Education program requirements for internal medicine residency training include a longitudinal, continuity experience with a panel of patients. OBJECTIVE To determine whether the number of resident clinics, the resident panel size, and the supervising attending physician affect patient continuity. To determine the number of clinics and the panel size necessary to maximize patient continuity. DESIGN We used linear regression modeling to assess the effect of number of attended clinics, the panel size, and the attending physician on patient continuity. PARTICIPANTS Forty medicine residents in an academic medicine clinic. MEASUREMENTS Percent patient continuity by the usual provider of care method. RESULTS Unadjusted linear regression analysis showed that patient continuity increased 2.3% ± 0.7% for each additional clinic per 9 weeks or 0.4% ± 0.1% for each additional clinic per year (P  =  .003). Conversely, patient continuity decreased 0.7% ± 0.4% for every additional 10 patients in the panel (P  =  .04). When simultaneously controlling for number of clinics, panel size, and attending physician, multivariable linear regression analysis showed that patient continuity increased 3.3% ± 0.5% for each additional clinic per 9 weeks or 0.6% ± 0.1% for each additional clinic per year (P < .001). Conversely, patient continuity decreased 2.2% ± 0.4% for every additional 10 patients in the panel (P < .001). Thus, residents who actually attend at least 1 clinic per week with a panel size less than 106 patients can achieve 50% patient continuity. Interestingly, the attending physician accounted for most of the variability in patient continuity (51%). CONCLUSIONS Patient continuity for residents significantly increased with increasing numbers of clinics and decreasing panel size and was significantly influenced by the attending physician.

Evolving disparities in the epidemiology of oral cavity and oropharyngeal cancers

Incidence rates of head and neck cancers (HNC) associated with human papillomavirus (HPVa) infection are increasing while non-HPV-associated (non-HPVa) HNC cancer rates are decreasing. As nearly all sexually active individuals will acquire an HPV infection, it is important to understand epidemiologic trends of HNCs associated with this sexually transmitted disease. We analyzed SEER 9 (1973–2012) and 18 data (2000–2012) for HPVa HNCs (oropharynx area; OP) and non-HPVa (oral cavity area; OC). Incidence rates were examined by gender, race, rurality, geographic location, and time. Joinpoint regression analyses assessed temporal variations. From 1973 to 2012, OC incidence decreased while OP increased, with changes largely driven by males (whose OP rate increased 106.2% vs female decrease of 10.3%). Males consistently had higher rates of both cancer groups across each registry except Alaska, OP rates among blacks changed from significantly above whites to below, and trend analysis indicated significant differences in rates over time by gender, race, and geography. Analysis of SEER 18 found that rates discordantly varied by group and gender across the 18 registries, as did the male/female rate ratio with overall means of 4.7 for OP versus 1.7 for OC (only Alaska and Georgia having overlapping ranges). Our findings indicate that much of the HPVa rate increases were driven by rate increases among males and that there were changing differences in risk between genders, race, and geographic location. The epidemiology of HNCs is complex, with locally relevant factors requiring further research for elucidation of demographic disparities in incidence.