Alan S. Go, MD

To characterize warfarin eligibility and receipt among Veterans Health Administration (VHA) patients with and without mental health conditions (MHCs).

This observational study identified VHA atrial fibrillation (AF) patients with and without MHCs in 2004. We examined unadjusted MHC-related differences in warfarin eligibility and warfarin receipt among warfarin-eligible patients, using logistic regression for any MHC and for specific MHCs (adjusting for sociodemographic and clinical characteristics).

Of 125,670 patients with AF, most (96.8%) were warfarin-eligible based on a CHADS2 stroke risk score. High stroke risk and contraindications to anticoagulation were both more common in patients with MHC. Warfarin-eligible patients with MHC were less likely to receive warfarin than those without MHC (adjusted odds ratio [AOR], 0.90; 95% CI, 0.87-0.94). The association between MHC and warfarin receipt among warfarin-eligible patients varied by specific MHC. Patients with anxiety disorders (AOR, 0.86; 95% CI, 0.80-0.93), psychotic disorders (AOR, 0.77; 95% CI, 0.65-0.90), and alcohol use disorders (AOR 0.62, 95% CI 0.54-0.72) were less likely to receive warfarin than patients without these conditions, whereas patients with depressive disorders and posttraumatic stress disorder were no less likely to receive warfarin than patients without these conditions.

Compared with patients with AF without MHCs, those with MHCs are less likely to be eligible for warfarin receipt and, among those eligible, are less likely to receive such treatment. Although patients with AF with MHC need careful assessment of bleeding risk, this finding suggests potential missed opportunities for more intensive therapy among some individuals with MHCs.

Patients with atrial fibrillation and mental health conditions (MHCs) may not receive anticoagulation treatment, even when they exhibit indications for anticoagulation for stroke prevention. 

Atrial fibrillation patients with MHCs are a clinically complex group; stroke prevention decision making with this group of patients must be particularly deliberative. 

Clinicians should consider MHCs when developing an anticoagulation treatment plan, though presence of MHCs should not be considered an absolute contraindication to anticoagulation.

Atrial fibrillation and atrial flutter (AF, collectively) affect almost 7 million Americans

— including 6% of people aged over 65 years old

—and account for at least 15% of the 700,000 strokes per year in the United States.

 Anticoagulation with warfarin can decrease ischemic stroke risk in AF by more than 60%.

 However, warfarin has a narrow therapeutic window: subtherapeutic and supratherapeutic anticoagulation can cause ischemic and hemorrhagic stroke, respectively, making ongoing laboratory monitoring and dose titration necessary. Consequently, certain clinical scenarios preclude safe warfarin use, such as situations in which the risk of hemorrhage is high or the likelihood is low that the patient will be able to achieve and maintain good anticoagulation control.

 Healthcare providers may include patients with mental health conditions (MHCs) in the latter group due to concerns (founded or not) about effectiveness or safety of anticoagulation (eg, related to comorbidities, concomitant medications, or anticipated nonadherence). The Veterans Health Administration (VHA) has invested considerable resources in maintaining a robust anticoagulation care infrastructure, but does not provide explicit guidance on anticoagulation in the context of mental illness,

despite the high prevalence of MHCs among VHA patients.

 Other national guidelines have provided scant input on this issue,

 and little research characterizes warfarin receipt among AF patients with MHCs.

We therefore performed a descriptive study of national data to determine: 1) whether AF patients with versus without MHCs differ in eligibility for anticoagulation; and 2) whether, among AF patients eligible for warfarin, warfarin receipt differs for patients with versus without MHC. The study was approved by the Institutional Review Board at Stanford University, Stanford, California.

We created a database containing variables characterizing AF care for all VHA patients with prevalent AF, as of the first day of fiscal year 2004. This database was created through a series of processing steps, quality checks, merges, and variable creation from VHA’s centralized inpatient and outpatient treatment and pharmacy databases, linked to Medicare inpatient and outpatient treatment claims data. Medicare pharmacy records were not available.

 describes cohort construction. We used Veterans Affairs (VA) encounter data to identify all veterans using VHA outpatient care during 2004 who had AF as of the first day of 2004. AF was defined as the presence of at least 1 

International Classification of Diseases, Ninth Revision

) AF diagnosis code 427.31 or 427.32 in 2002, and at least 1 confirmatory AF diagnosis in 2003 in VHA or Medicare inpatient or outpatient encounter records (based on records involving a face-to-face visit with a clinician). We excluded patients institutionalized for the majority of 2004, and those who received some of their 2004 VHA care outside of the continental United States (who thus might have incomplete data capture for processes of care). We also excluded patients whose MHC status could not be determined (see description of MHC variable creation below) to create “cohort 1.”

To examine warfarin receipt among patients with and without MHCs, we created “cohort 2,” which was the subset of cohort 1 patients who were apparently eligible for warfarin based on the presence of diagnosed stroke risk factors and absence of diagnosed contraindications to anticoagulation. 

Mental Health Conditions (independent variables)

Starting with the Agency for Health Research and Quality’s Clinical Classifications Software (CCS),

 we conducted an expert panel process to make modifications to the 

 codes uniquely to 5 common specific MHCs (depressive disorders, posttraumatic stress disorder [PTSD], other anxiety disorders, psychotic disorders, and alcohol use disorders) and “other” psychiatric disorders. A patient was considered to be “MHC Yes” if he/she had at least 1 instance of an 

 code falling into an MHC condition category during 2002-2003, plus at least 1 confirmatory 

 code in 2004 (ie, during the period in which warfarin receipt was assessed), in a VHA or Medicare record associated with an outpatient face-to-face visit with a clinician or an inpatient record. A patient was considered to be “MHC No” if he/she had no instance of an MHC 

 code in the entire interval of 2002 to 2004. Patients for whom MHC status was uncertain (ie, those with an MHC diagnosis at baseline [2002-2003] or during the study period [2004], but not both) were excluded, allowing for direct comparisons between the 2 distinct groups of “MHC Yes” and “MHC No.”

Separate dichotomous indicator variables were created for each of the 5 specific MHCs and for “other” MHCs for each patient. A patient could have more than 1 diagnosed MHC.

Eligibility for Anticoagulation (independent variable)

We examined patient characteristics suggesting greater eligibility for anticoagulation (ie, risk factors for stroke), and those suggesting less eligibility for anticoagulation (ie, risk factors for hemorrhage). To characterize stroke risk as of the start of the study period (ie, prior to 2004), we calculated the CHADS2 score, which gives 2 point each for being 75 years or older, having congestive heart failure, hypertension, or diabetes; and 2 points for prior stroke or transient ischemic attack.

 Based on available national guidelines during the study period, CHADS2 was recommended for stroke risk stratification, and warfarin was recommended for patients with AF who had a CHADS2 score of 2 or higher. Warfarin was recommended, but not required, in patients with a CHADS2 score of 1.

 Contraindications to anticoagulation included a history of intracranial hemorrhage, history of other hemorrhage, dementia, cirrhosis, seizure disorder, and end-stage renal disease.

 These conditions, and conditions included in the CHADS2 score, were identified based on 

 codes in 2001 to 2003 VA and Medicare encounter data.

VHA Decision Support System (DSS) pharmacy records contain a record of dispensed medications from all pharmacy orders entered by VHA clinicians. From the DSS data, we identified every outpatient warfarin prescription issued in the VHA in 2004 to patients in our warfarin-eligible analytic cohort (cohort 2). A patient was considered to have received warfarin if he/she received at least 2 VHA outpatient warfarin prescriptions in 2004, with at least 30 days between the start of one prescription and the start of another prescription. Although our focus was on warfarin receipt—rather than warfarin persistency or adherence, which would be better examined in a cohort of new warfarin users&mdash;we required 2 warfarin prescriptions to confirm that the patient was actively prescribed warfarin through the VHA, and that, for example, he/she had not merely been issued a single warfarin prescription as part of an emergency department visit. Prescriptions issued on the last day of an inpatient stay were counted as outpatient prescriptions because they represent discharge medications. Warfarin prescriptions issued by Medicare providers were not available to us.

Patient age, gender, race/ethnicity, and physical comorbidity index were derived from VHA and Medicare patient treatment databases. The Selim physical comorbidity index is a count of common nonpsychiatric medical conditions developed for VHA outpatient case-mix adjustment.

In cohort 1 (AF patients, n = 125,670) and then in cohort 2 (AF patients eligible for warfarin, n = 87,248), we first compared those with versus without an MHC on sociodemographic characteristics, health status, and eligibility for anticoagulation (stroke and hemorrhage risk factors), using χ

 tests for continuous variables. Next, in cohort 2 we descriptively examined the proportion of patients who received warfarin, first by MHC status, and then in the subgroups with the 5 most common MHCs or with “other” MHCs. Finally, we performed 4 logistic regression analyses on cohort 2 to calculate unadjusted and adjusted odds of warfarin receipt (overall and by CHADS2 score) for patients with versus without diagnosed MHCs (Model 1 unadjusted and adjusted), and for patients with versus without specific MHCs (Model 2 unadjusted and adjusted). Because a patient could have more than 1 specific MHC (for example, a patient might have a depressive disorder, PTSD, and alcohol use disorder), Model 2 included a binary indicator variable for each of the 5 most common MHCs, and a binary indicator variable for other MHCs in aggregate, in a single logistic regression model; each estimate represents the conditional effect of each specific MHC, controlling for the presence of the other specific MHCs. Multivariate logistic regression models controlled for age as a continuous variable, gender, and physical comorbidities; overall models additionally controlled for stroke risk (CHADS2). 

After applying cohort selection and exclusion criteria, we identified 125,670 AF patients (cohort 1), and among them, 87,248 (69%) were apparently eligible for warfarin (cohort 2). Among cohort 1 patients, 22,247 (18%) had an MHC. Compared with patients without an MHC, patients with an MHC were younger (18% versus 8%, respectively, were under age 65) and slightly less likely to be male or Caucasian (

). Among cohort 2 patients, 12,190 (14%) had any MHC—the most common specific MHCs were depressive disorders (7% of cohort 2 patients), PTSD (2%), other anxiety disorders (3%), psychotic disorders (1%), and alcohol use disorders (1%).

Eligibility for Anticoagulation Among All AF Patients

In cohort 1, indications for anticoagulation based on CHADS2 stroke risk index were greater in patients with an MHC. Nearly all AF patients (97% in both MHC and no MHC groups) were warfarin-eligible, based on a CHADS2 score of 1 or more (Table 1). A slightly greater proportion of patients with MHC than without MHC had high levels of stroke risk (ie, CHADS2 score ≥2). Regarding specific stroke risk factors in patients with versus without an MHC: respectively, 54% versus 48% had congestive heart failure; 91% versus 90% had hypertension; 41% versus 40% had diabetes; 39% versus 31%, had prior stroke or transient ischemic attack (Table 1).

However, contraindications to anticoagulation (ie, risk factors for hemorrhage) were also more common in patients with MHC than in those without MHC: respectively, 1.4% versus 0.6% had a history of intracranial hemorrhage; 27% versus 20% had a history of other hemorrhage; 17% versus 3% had a history of dementia; 2% versus 1% had a history of cirrhosis; and 0.6% versus 0.5% had a history of end-stage renal disease. (

Receipt of Warfarin Among AF Patients Eligible for Warfarin

Among warfarin-eligible patients with AF, 50.6% of those with MHC and 52.4% of those with no MHC received warfarin (

 additionally shows the rate of receipt of warfarin for patients with each of the 5 most common specific MHCs or with any other MHC. Although we did not specify a minimum prescription days’ supply, 99% of patients who received warfarin had at least 60 days of warfarin coverage in 2004. 

Warfarin-eligible patients with an MHC were less likely to receive warfarin than those without an MHC (Model 1: unadjusted OR, 0.93, 95% CI, 0.90-0.97; adjusted OR, 0.90, 95% CI, 0.87-0.94). In the adjusted model, differences persisted in CHADS2-stratified analyses (

We also analyzed warfarin receipt for patients with specific MHCs, which is important because clinical presentation differs by MHC. Patients with some types of MHC might have particular difficulty understanding or adhering to a complex medication regimen, or their provider's perception might be that they would. The Model 2 estimates presented in Table 2 represent the marginal differences in warfarin receipt associated with any one specific MHC, controlling for the presence of the other specific MHCs.

The association between MHC and warfarin receipt varied substantially by specific MHC type (Table 2, Model 2). After adjustment, patients with anxiety disorders (OR, 0.86; 95% CI, 0.80-0.93), psychotic disorders (OR, 0.77, 95% CI 0.65-0.90), and alcohol use disorders (OR, 0.62; 95% CI, 0.54-0.72) were less likely to have received warfarin than were patients with absence of the specific condition. In CHADS2-stratified analyses, adjusted odds of warfarin receipt was lower for patients with alcohol use disorders at a CHADS2 score of 2-3 (OR, 0.53; 95% CI, 0.43-0.64) than at a CHADS2 score of 1 (OR, 0.70; 95% CI, 0.55-0.90).

Main findings (adjusted ORs for Model 1 and Model 2) remained robust across a series of sensitivity analyses: 1) we included the MHC-indeterminate patients as a third level of exposure; 2) we removed patients who died in 2004 or who developed a new contraindication to warfarin in 2004; and 3) we added clinic site (as a fixed effect) to the models.

Our study provides clinically relevant insights into the connection between mental illness and stroke prevention care in VHA. We found that AF patients with an MHC have an excess burden of both stroke risk factors and contraindications to anticoagulation. Among those apparently eligible for warfarin, patients with an MHC were less likely to receive warfarin, although this association varied by specific MHC type.

Our findings suggest that special clinical decision-making challenges may arise in the management of AF patients with an MHC. On the one hand, patients with an MHC have excess burden of stroke risk factors: therefore, many of them have strong indications for anticoagulation for stroke prevention. On the other hand, AF patients with MHC also have an excess burden of risk factors for hemorrhage, and thus, may have disproportionate rates of contraindications to anticoagulation. In such a setting, assessment of risk versus benefit of warfarin receipt must be individualized, and shared decision-making approaches

Our findings in a national sample of more than 12,000 warfarin-eligible patients with AF with an MHC are consistent with, and expand upon, a prior single-site chart-review analysis, which found lower odds of warfarin receipt in 66 warfarin-eligible AF patients with an MHC.

  The lower frequency of warfarin receipt seen for patients with an MHC in our study is also in line with studies showing that patients with an MHC may be less likely to receive pharmacotherapy for various medical conditions,

 although such MHC-related disparities are not universally seen.

We also found that the odds of warfarin receipt in eligible patients with AF varied by MHC type. This finding is not surprising, given the heterogeneity of the clinical presentation of different MHCs. Patients with alcohol use disorders were less likely to receive warfarin—perhaps related to concerns about drug-drug interactions between warfarin and alcohol, concerns about falls risk, or concerns about adherence

; clinicians’ expectations about whether a patient will adhere to treatment can directly influence their treatment decisions.

 Interestingly, this effect was even stronger at a higher level of stroke risk (CHADS2 score of 2-3) than at a lower level of stroke risk (CHADS2 score of 1). Although our study did not directly examine MHC severity, patients with psychotic disorders also were less likely to receive warfarin, which could reflect clinical impressions that their thought processes were too disorganized to support faithful use of a complicated medication regimen, and the fact that medication nonadherence in patients with psychotic disorders is known to be high.

In contrast, patients with depression and PTSD were no less likely than patients without these conditions to receive warfarin, suggesting that clinicians respond to different mental health conditions differently. Future studies using qualitative methods or examining potential mediators at a more granular level may provide greater insight into drivers of medical decision making, especially now that new alternatives to warfarin (eg, dabigatran, rivaroxaban, apixaban) with decreased intracranial hemorrhage risk and less onerous laboratory monitoring requirements are available.

Our results need to be interpreted in the context of several limitations. First, there is risk of misclassification of MHC status. MHCs often go undetected in a primary care setting; therefore, some patients in our “No MHC” group may have had an undetected MHC. If anything, this would be expected to bias our findings toward the null, given that patients with an undetected MHC probably would be less likely to receive warfarin. Furthermore, our decision to exclude patients with an indeterminate MHC status (MHC diagnoses present at baseline or during the study period, but not both) should have helped to minimize misclassification. There is also the potential for uncaptured processes of care (warfarin receipt). This study was conducted before Medicare Part D (prescription) data were available; thus, any prescriptions issued through Medicare were unavailable. Since VHA patients with an MHC tend to be less likely to receive care outside the VHA, it is possible that rates of warfarin receipt in the “No MHC” group were disproportionately underestimated, again tending to bias our warfarin-receipt findings toward the null.

Additionally, although warfarin remains the mainstay of AF stroke prevention treatment in VHA, AF stroke prevention care has evolved since 2004 when this study was conducted. However, the focus of this study was not on absolute rates of anticoagulation treatment, but rather on MHC-related differences in anticoagulation treatment; such effects would not be expected to have changed substantially over this time period. Furthermore, a recent longitudinal study demonstrated that the rate of warfarin receipt in VA patients with AF remained fairly stable from 2004 (the year of the results presented here) through 2008 (the last year of data examined in that study).

 Finally, this study examined warfarin receipt as a process of care, and thus, we did not examine patient outcomes. Finally, results apply to atrial fibrillation patients who use VHA services, who are primarily men, and cannot necessarily be generalized to women, veterans who do not use VHA, or to nonveterans.

In a large national cohort of patients with AF, those with MHC were more likely than those without MHC to have contraindications to anticoagulation, suggesting that cautious assessment of bleeding risk factors prior to initiation of warfarin may be particularly salient for patients with MHC. Several recent studies suggest that outcomes of anticoagulation therapy in the presence of MHC may be somewhat worse, but that presence of MHC does not consistently preclude safe treatment with warfarin.

 In that context, our finding that warfarin-eligible patients with AF with an MHC were less likely than other AF patients to receive warfarin, raises concerns about potential missed opportunities for more intensive therapy among some patients with an MHC. This supports the recommendation that patients and their healthcare providers weigh the pros and cons of anticoagulation with warfarin versus other management approaches so as to maximize delivery of equitable, high-quality, patient-centered care.

The authors gratefully acknowledge Vu Nguyen and Stephanie Le for their contributions to earlier phases of this project. Support for VA/CMS data was provided by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development, VA Information Resource Center (Project Numbers SDR 02-237 and 98-004).

 VA HSR&D Center for Innovation to Implementation (SKS, MPT, CSP, RHM, PH, VYC, SAF, CTT, SMF), and Health Economics Resource Center (SKS, CSP), and Women’s Health Section, Medical Service (SMF), VA Palo Alto Health Care System, Palo Alto, CA; Stanford University School of Medicine (MPT, CSP, RHM, PH, SMF), Stanford, CA; Department of Veterans Affairs Center for Healthcare Organization and Implementation Research, Edith Nourse Rogers Memorial Veterans Hospital (DB), Bedford, MA; Department of Health Policy and Management, Boston University School of Public Health (DB), Boston, MA; Division of Research, Kaiser Permanente of Northern California (ASG), Oakland, CA; Departments of Epidemiology, Biostatistics, and Medicine, University of California at San Francisco (ASG), San Francisco, CA.

 This work was supported by VA HSR&D IIR 04-248, CDA09-027, and RCS 90-001. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.

 Dr Go has previously received a grant from the FDA. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (SKS, SMF, SAF, MPT, PH, DB, CSP, RHM); acquisition of data (SMF, SAF, CSP); analysis and interpretation of data (SKS, SAF, SMF, MPT, PH, DB, ASG, CSP, RHM); drafting of the manuscript (SKS, DB); critical revision of the manuscript for important intellectual content (SKS, SMF, MPT, PH, DB, VYC, ASG, CSP, RHM); statistical analysis (SKS, SAF, VYC, CSP); obtaining funding (SMF, MPT, CSP); administrative, technical, or logistic support (VYC, RHM, CTT); and supervision (MPT, SMF).

1. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. 

2. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and gender of patients with atrial fibrillation: analysis and implications. 

Articles

Atrial fibrillation patients with mental health conditions are less likely to be eligible for warfarin receipt, and those who are eligible receive warfarin at lower rates.

Anticoagulation in Atrial Fibrillation: Impact of Mental Illness

 To determine whether ethnic differences in the incidence of albuminuria are present in patients with diabetes, and to identify social, behavioral, and provider factors that explain ethnic differences.

 Survey follow-up design with a race-stratified baseline survey (2005-2006) in diabetic patients from a nonprofit, fully integrated healthcare system in Northern California. We followed the 10,596 respondents (30% whites, 20% blacks, 23% Hispanics, 14% Asians, and 13% Filipinos) without evidence of prevalent albuminuria at baseline.

Incident albuminuria was defined by positive dipstick urinalysis (

1) or urine albumin to creatinine level (

30 mg/g), and confirmed with repeat testing at least 3 months later.

The 27,292 person-years of observation yielded 981 incident albuminuria events. Agestandardized rates of albuminuria (per 1000 person-years) ranged from 13.6 (95% confidence interval [CI] 10.5-17.0) in whites to 27.8 (CI 18.2- 38.3) in blacks. In fully adjusted Cox models, the hazard ratio for blacks (1.22, 95% CI 1.09-1.38), Asians (1.35, 95% CI 1.13-1.61), and Filipinos (1.93, 95% CI 1.61-2.32), but not Hispanics, was significantly greater than it was for whites. In some cases, point estimates changed markedly from the base model when fully adjusted for potential confounders. Moreover, adjustment for an array of potentially mediating factors explained only a small proportion of the observed ethnic disparities.

Despite uniform medical care coverage, Filipinos, blacks, and Asians with diabetes developed albuminuria at higher rates than white and Hispanic adults.

	Despite uniform medical care coverage, Filipinos, blacks, and Asians with diabetes developed albuminuria at higher rates than whites.

		After adjustment for confounding, the incidence of albuminuria was 93% greater in Filipinos, 35% greater in Asians, and 22% higher in blacks compared with whites; the difference between Hispanics and whites was not significant.

		Racial differences were not explained by a wide range of potentially mediating factors.

		Albuminuria is a potent predictor of future microvascular and macrovascular complications; thus, these findings have clinical implications for risk stratification and future efforts to reduce ethnic disparities in renal disease.

	Diabetes afflicts 8% of the US population, and its prevalence is expected to double over the next 2 decades.

 Diabetes is at least 2 to 4 times more common among ethnic minorities than it is among non-Hispanic whites.

 Furthermore, minorities have higher mortality rates and microvascular complications of diabetes, such as end-stage renal disease (ESRD).

 Collectively, these statistics have placed diabetes at the center of the president’s Healthy People 2020 initiative to eliminate health disparities.

	Albuminuria is an extremely common consequence of diabetes, with a prevalence of 30% to 50%.

 Microalbuminuria and macroalbuminuria are strongly associated with angiographically determined coronary atherosclerosis, cardiovascular events, kidney failure, and mortality in patients with diabetes, as well as in the general population, independent of conventional cardiovascular risk factors and the estimated glomerular filtration rate (eGFR).

 Although the natural history of diabetic nephropathy is relatively well described, prior studies have rarely examined ethnic variations in the development of albuminuria or identified factors that may explain ethnic differences in the risk of incident albuminuria. The majority of work in this field has focused on ESRD, for which ethnic disparities are established.

 However, ESRD patients represent less than 1% of the diabetic population; therefore, strategies targeting ESRD are limited to a small subset of patients.

 Understanding ethnic differences in rates of albuminuria and the underlying causes for these differences is likely to advance public health objectives for diabetes treatment, because albuminuria is very common and is strongly associated with adverse events, and kidney disease is the complication of diabetes with the greatest effect on minority groups.

	We conducted a prospective, longitudinal cohort study to evaluate ethnic differences in incident albuminuria in a fully insured, ethnically diverse, well-characterized cohort of diabetic patients. Our secondary goal was to identify social, behavioral, and provider-level factors that could explain observed ethnic differences in albuminuria in this population.

	The Diabetes Study of Northern California (DISTANCE) is a National Institute of Diabetes and Digestive and Kidney Diseases— sponsored study of diabetes conducted at Kaiser Permanente Northern California (Kaiser).

 This nonprofit, integrated group practice provides comprehensive healthcare to an ethnically diverse population of more than 3 million people, approximately 30% of the population of Northern California. The DISTANCE cohort comprised an ethnically stratified, random sample of diabetic patients (type 1 and type 2) receiving care from Kaiser. In 2005 to 2006, participants of DISTANCE completed a detailed assessment of demographic, clinical, behavioral, socioeconomic, psychosocial, medical knowledge, and quality of care indicators (62% survey response rate). The complete survey is available at www.distancesurvey.org. The survey was offered in multiple languages including English, Spanish, Cantonese, Mandarin, and Tagalog using certified translations of an English script, and through multiple modalities, including computer-assisted telephone interview, self-administered written format, and the Internet, to maximize accessibility of the survey.

	Survey data were supplemented with clinical information from the Kaiser electronic medical record that comprehensively captures inpatient and outpatient utilization, laboratory testing, clinical measurements (blood pressure, height, and weight), pharmacy data (including prescribed medications and filled prescriptions), processes of care, procedures, and diagnoses. Patient data are linked to the California Death Registry to ascertain death information and the United States Renal Data System for ESRD status.

 Neighborhood socioeconomic status was determined by linkage of participant residential addresses with the year 2000 US Census data at the census-tract level.

	We included all DISTANCE participants who self-identified their ethnicity as white (non-Hispanic), black (non-Hispanic), Hispanic, Asian, or Filipino. All individuals entered the study at the time of survey (index date). We excluded (1) participants belonging to other ethnic groups, those who were multiethnic, or those with missing ethnicity; (2) those without continuous membership at Kaiser (allowing <3 months’ gap); (3) individuals with prevalent albuminuria, defined as microalbuminuria measured on any urine dipstick testing (

1) or urine albumin to creatinine ratio (>

 mg/g) within 2 years prior to the index date; (4) patients with ESRD (defined as receipt of chronic dialysis therapy or kidney transplantation)

; and (5) participants who did not have an additional measurement of albuminuria (urinary albumin excretion or urine dipstick testing) after the index date.

	The primary outcome for the analysis was time from study entry to incident albuminuria. Albuminuria was defined according to National Kidney Foundation guidelines as having urinary albumin excretion levels at or above the threshold for “microalbuminuria” (

30 mg/g on urine albumin to creatinine ratio testing), confirmed on a second consecutive occasion at least 3 months later. The first positive urine test result was used at the date of censor (ie, microalbuminuria incidence date).

	Self-reported ethnicity, our primary predictor, was categorized as white, black, Asian, Hispanic, or Filipino. Filipinos were not included in the Asian category given the unique social, cultural, and linguistic characteristics of this group.

All baseline covariates were derived from the DISTANCE survey or the electronic medical record, using validated algorithms based on laboratory, vital sign, and clinical data closest to the survey date and within the preceding 2 years.

 We included demographics (age, sex); clinical characteristics (blood pressure, low-density lipoprotein, glycosylated hemoglobin [A1C], body mass index, eGFR); comorbid conditions (hypertension, hyperlipidemia); macrovascular disease (heart failure, coronary heart disease, peripheral vascular disease, or cerebrovascular accident); microvascular disease (retinopathy); diabetes characteristics (diabetes duration, receipt of insulin therapy); and current treatment (angiotensin-converting enzyme inhibitor or angiotensin receptor blocker). We calculated eGFR using the Modification of Diet in Renal Disease formula based on age, sex, race, and standardized serum creatinine.

	We also selected variables that could be potential mediators of ethnic differences in the development of albuminuria based on the Institute of Medicine’s definition of disparities as the difference in treatment or access not justified by the differences in health status or preferences of the groups.

 These included socioeconomic factors (education [highest degree attained], income, and neighborhood deprivation index

); language (limited English proficiency and nativity); behaviors (smoking status [current, never, past], adequate exercise [by self-report], self-monitoring of blood glucose levels, and adherence [oral diabetes medication adherence and antihypertensive adherence over the prior year]); and provider factors (poor provider communication,

patient/ provider race and sex concordance). The closed pharmacy system at Kaiser provides near-complete ascertainment of pharmacy utilization and facilitates assessment of medication adherence. We also adjusted relevant clinical factors including use of lipid-lowering and antihypertensive medications (and initiation of new treatments during follow-up), change in systolic blood pressure, change in A1C, and a validated index of medication adherence, “Continuous, multiple- interval measure of medication gaps,” which estimates the percentage of days during follow-up without adequate supply of medications.

	We calculated age-standardized rates of incident albuminuria assuming a Poisson distribution. We estimated the adjusted risk of albuminuria using a Cox proportional hazards model. For all analyses, we incorporated expansion weighting in all models to accommodate the nonproportional sampling fractions used for the race-stratified random sampling design of the DISTANCE survey. Patients were followed from the index date and censored at the time of de-enrollment from Kaiser, death, or the end of follow-up (December 31, 2008). We compared the relative hazard for incident albuminuria associated with ethnicity using white race as the referent group. Multivariable Cox models were adjusted for all covariates described above and listed in 

. We also investigated possible interactions between ethnicity and age, sex, microvascular disease, exposure to angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, blood pressure, body mass index, and eGFR in fully adjusted Cox models. The 

 values for interactions were obtained using cross products between ethnicity with each variable of interest. We performed Markov chain Monte Carlo multiple imputation using 20 imputations to account for missing data.

 We specified serial, nested models to identify potential mediator domains that may explain ethnic differences in the albuminuria incidence by sequentially adding groups of related candidate mediators to the fully adjusted base model.

	Diabetes care at Kaiser is largely consistent with the recommendations of the American Diabetes Association, which advises annual testing for albuminuria. To assess adherence with these guidelines and to determine whether ascertainment of the outcome occurred uniformly across ethnic groups, we compared the frequency of albuminuria testing in the cohort by ethnicity. In a supplemental analysis, we compared the incidence of albuminuria in DISTANCE survey responders with the incidence in survey nonresponders by ethnic group to evaluate the potential for selection bias due to survey nonresponse. Model assumptions of Cox proportional hazards models were checked by comparing plots of log (-log[survival]) versus log of survival time and the Schoenfeld test. All analyses were conducted using SAS version 9.2 (SAS Institute, Inc, Cary, North Carolina). This study was approved by the institutional review board of the Kaiser Foundation Research Institute.

	In total, 20,030 subjects responded to the DISTANCE survey and maintained continuous membership at Kaiser. We excluded 3629 individuals with self-reported multiethnic, other, or unknown ethnicity, 306 persons with ESRD, 4984 participants with prevalent albuminuria, and 515 participants with no albuminuria measurement, leaving a total of 10,596 persons in the analytic cohort. Ethnic minorities were well represented, with a relatively balanced proportion of women within each group (Table 1). The majority of patients were older, with a mean age of approximately 60 years, and 70% to 95% had diagnosed hypertension or hyperlipidemia. A larger proportion of Hispanics, Filipinos, and blacks had A1C levels of >8% compared with whites and Asians. In contrast, Hispanics,

	Asians, and Filipinos had better blood pressure control compared with whites and blacks (marked by a systolic blood pressure of <130 mg Hg). At baseline, approximately 15% to 25% of the cohort had moderate impairment of kidney function (eGFR <60 mL/min per 1.73 m²) without albuminuria; the highest prevalence of reduced GFR was observed in whites. Asians, Filipinos, and whites had higher levels of socioeconomic status than blacks and Hispanics based on annual income levels greater than $65,000, college education, and lower levels of neighborhood deprivation. Limited English proficiency was common in Hispanics, Asians, and Filipinos; the vast majority of Filipinos and a large proportion of Asians and Hispanics were not born in the United States. Poor provider communication was more commonly reported by Filipinos and Hispanics than by whites, blacks, and Asians.

	There were 981 confirmed incident albuminuria events over 3 years of observation (27,292 person-years of observation). In all, 11% of blacks, 10% of Filipinos, 9% of Asians and Hispanics, and 8% of whites developed albuminuria. The mean observation time and the number of albuminuria measurements per individual appeared to be comparable across ethnic groups (

). Seventy percent of the diagnoses were based on 2 positive confirmatory tests (either albumin to creatinine ratio or 24-hour urinary protein); 22% were based on a combination of a dipstick plus either albumin to creatinine ratio or 24-hour urinary protein; and 8% were based on 2 positive dipstick tests (ie, dipstick only). On average, the second (confirmatory) test occurred 9.2 months (SD 5.6 months) after the first positive test. The age-standardized rate of albuminuria (per 1000 person-years) was highest in blacks (27.8, 95% confidence interval [CI] 18.2-38.3), and the CIs did not overlap with those for the reference group (whites:13.6, 95% CI 10.5-17.0). The confidence intervals did overlap between those for whites and the remaining race groups: Filipinos (22.0, 95% CI 14.0-31.1), Asians (15.0, 95% CI 7.7-23.4), and Hispanics (13.8, 95% CI 11.6-16.2). In fully adjusted models, we observed a 93% greater risk of albuminuria in Filipinos, a 35% greater risk in Asians, and a 22% greater risk in blacks compared with whites.

	We also attempted to identify potential mediators of the elevated risk of incident albuminuria associated with ethnicity (

). Hazard ratios were only slightly attenuated for blacks, changed minimally in Hispanics, and strengthened in Asians and Filipinos. In the final saturated model that included all 34 factors, the risk of albuminuria associated with Asian and Filipino ethnicity increased relative to the base model, particularly when clinical factors were introduced to the model. In supplemental analyses, incidence rates of albuminuria were similar in survey nonrespondents in DISTANCE, as well as in the larger Kaiser Permanente Northern California Diabetes Registry, suggesting selection bias was not a major concern.

	To our knowledge, this study is the first to document ethnic differences in the incidence of albuminuria among patients with diabetes. In this large, ethnically diverse, wellcharacterized cohort of diabetic patients receiving relatively uniform healthcare, we found that the fully adjusted risk of albuminuria was 22%, 35%, and 93% greater in blacks, Asians, and Filipinos compared with whites. Only Hispanics were not at significantly elevated risk relative to whites. We also tested a wide spectrum of factors that were hypothesized to explain observed ethnic disparities. Despite extensive adjustment for these candidate mediator variables, we were unable to explain a substantive proportion of the differences we observed between ethnic groups. In fact, model adjustment accentuated differences between the white reference group and Asians and Filipinos. Inflation of effect estimates in response to adjustment is the opposite of what would be expected if the included variables were explanatory. Alternatively, the crude estimates could have underestimated true effect size due to “negative confounding.” This might occur if Asians and Filipinos had a higher-than-expected observed incidence of albuminuria after accounting for observed racial differences in variables that were strongly predictive overall (eg, worse blood pressure, glycemic control, lower GFR, congestive heart failure). It is also possible that ethnic differences in the risk of albuminuria may be attributable to unmeasured, underlying genetic or biologic differences; long-term exposures to poor risk factor control; or other environmental factors not measured in this study.

	It is a common assumption that ethnic differences in diabetic nephropathy have been well characterized. While it is known that ethnic minorities with diabetes experience accelerated rates of ESRD compared with whites, no previous studies have examined ethnic differences in the development of albuminuria. Prior work in this area has been limited to studies that were cross-sectional in design,

 or the provision of dialysis or kidney transplantation,

 or lacked adequate representation of ethnic minorities.

 The results of our longitudinal analyses are consistent with cross-sectional findings from the National Health and Nutrition Examination Survey as well as the Pathways Study, which found a higher prevalence of albuminuria in ethnic minorities compared with whites.

 Our study builds on these observations by extending them to a much larger number of patients, particularly Asians and Hispanics, and strengthening the causal interpretation of the association between ethnicity and albuminuria, by virtue of the prospective cohort design of the current study.

	It is notable that the majority of research aimed at reducing ethnic disparities in diabetic kidney disease has focused on the risk of ESRD. While interventions aimed at curbing ESRD attributable to diabetic nephropathy are needed, we speculate that public health approaches that target albuminuria are more likely to reach a larger number of individuals and thus have a greater impact on ethnic disparities. Albuminuria affects up to 50% of patients with diabetes, while eGFR <60 mL/min per 1.73 m

 is present in 14%, and ESRD occurs in <1%.

 In addition, any degree of albuminuria, determined either by dipstick urinalysis or urine albumin to creatinine ratio, appears to be as strong an indicator as kidneyfunction for identifying patients at risk for both progressive kidney disease and adverse cardiovascular events.

 Based on this information, the National Kidney Foundation—Kidney Disease: Improving Global Outcomes initiative plans to formally include grading of albuminuria levels in the consensus guidelines for staging of chronic kidney disease, with clinical recommendations for each stage, including those aimed at prevention. It will be important for future research to evaluate how implementation of new guidelines impacts ethnic disparities in diabetes-related outcomes.

	The secondary goal of this study was to identify the underlying determinants of the observed ethnic differences in the incidence of albuminuria. All participants had regular ambulatory diabetes care in an integrated healthcare system providing uniform access to medications and services. Despite similar access to care, ethnic differences in the risk of albuminuria were substantial. We tested a large number of factors hypothesized to contribute to diabetes complications in prior studies. In spite of this, we were unable to identify factors that substantially explained observed ethnic differences. While we thoroughly investigated clinical differences between ethnic groups, social factors, and inequalities in the receipt of healthcare, it is possible that relevant factors were not included in this analysis. In addition, long-term exposures to elevated A1C and poorly controlled hypertension, which were not captured in these data, may play an important role. Genetic factors have been widely recognized in renal disease (eg, APOL1 and MYH9), and may play an important role in the residual differences in albuminuria observed in this study.

	Study strengths include the large sample population, balanced representation across multiple ethnic groups, and uniform ascertainment of a wide range of patient factors collected within an integrated healthcare setting. Results should be generalizable to other insured, treated diabetic populations. Although dipstick urinalysis is a semiquantitative method that has less favorable diagnostic properties than urine albumin to creatinine ratio for the assessment of albuminuria, it is widely used in clinical practice, is considerably less expensive, and provides similar information for prognosis and risk stratification.

However, the 2 methods and cut-points (albumin to creatinine ratio of 

1]) may not be fully comparable or used at random. A further strength of this study is the ability to confirm the presence of albuminuria with repeated measurement over a 3-month period, as currently recommended by National Kidney Foundation guidelines.

Transient proteinuria or decreases in GFR are common in clinical practice and may decrease the specificity of such outcomes in cohort studies, which typically ascertain the outcome at a single time point. However, we also relied on an electronic medical record and automated databases for outcomes surveillance in a usual care setting, for which the timing of measurements is determined by the healthcare provider, in contrast to repeated visits by participants at set time points. Annual screening for diabetic nephropathy through urine dipstick or albumin to creatinine testing is currently the standard of care in diabetes within the Kaiser medical system. In addition, we found that the frequency and timing of albuminuria testing were relatively similar between ethnic groups. Another limitation is that the measured covariates may differ biologically between race groups; for example, A1C levels differ between African Americans and whites despite similar serum glucose levels. While we adjusted for the duration of diabetes, this variable was based on time since first diagnosis. Although differences in the duration across ethnic groups were minimal, it is possible that the subjects had undiagnosed diabetes for differing lengths of time.

Despite uniform access to medical care, Filipinos, blacks, and Asians with diabetes developed albuminuria at higher rates than whites.

Ethnic Differences in the Development of Albuminuria: The DISTANCE Study

 To determine whether atrial fibrillation (AF) patients with mental health conditions (MHCs) were less likely than AF patients without MHCs to be prescribed warfarin and, if receiving warfarin, to maintain an International Normalized Ratio (INR) within the therapeutic range.

 Detailed chart review of AF patients using a Veterans Health Administration (VHA) facility in 2003.

 For a random sample of 296 AF patients, records identified clinician-diagnosed MHCs (independent variable) and AF-related care in 2003 (dependent variables), receipt of warfarin, INR values below/above key thresholds, and time spent within the therapeutic range (2.0-3.0) or highly out of range. Differences between the MHC and comparison groups were examined using X

 tests and logistic regression controlling for age and comorbidity.

 Among warfarin-eligible AF patients (n = 246), 48.5% of those with MHCs versus 28.9% of those without MHCs were 

 = .004). Among those receiving warfarin and monitored in VHA, highly supratherapeutic INRs were more common in the MHC group; for example, 27.3% versus 1.6% had any INR >5.0 (

 <.001). Differences persisted after adjusting for age and comorbidity.

 MHC patients with AF were less likely than those without MHC to have adequate management of their AF care. Interventions directed at AF patients with MHC may help to optimize their outcomes.

	Mental illness decreased likelihood of warfarin receipt among patients without recorded contraindications, but clinicians did not list mental illness as a rationale for withholding warfarin.

		Among warfarin-treated patients, those with mental illness were more likely to have some International Normalized Ratio (INR) values substantially above the therapeutic range, but some with mental illness consistently maintained therapeutic INRs.

		Identifying subgroups with mental illness who can/cannot safely take warfarin would inform interventions and guidelines.

The Veterans Health Administration (VHA) National Mental Health Strategic Plan has recently called attention to optimizing treatment of coexisting medical conditions in patients with mental illness. Research is beginning to amass suggesting that patients with mental health conditions (MHCs), representing nearly one third of Americans,

 may receive less intensive care for a range of conditions from diabetes to coronary artery disease to chronic pain.

 However, the influence of mental illness on the management of atrial fibrillation (AF) is not yet well understood. This is important because AF is highly prevalent, especially among older adults, and is associated with increased stroke risk.

 An oral anticoagulant, warfarin, reduces AF-related stroke risk from approximately 5% per year to 1% to 2% per year,

 but also has downsides: it increases major bleeding risk, it typically involves a complex and rigorous dosing schedule, and it requires frequent blood monitoring of coagulation status (International Normalized Ratio [INR]). Given warfarin’s narrow therapeutic window, clinicians may have concerns about whether patients with MHC can safely take warfarin as prescribed and follow through with necessary monitoring tests.

Unfortunately, major AF practice guidelines do not currently address whether mental illness should influence anticoagulant therapy decision making.

 This is concerning, since there are potential adverse consequences of prescribing warfarin for a patient who cannot use it safely: excess major bleeding if overdosed or thromboembolic events if underdosed. Likewise, there are potential adverse consequences of not prescribing warfarin for a patient who could take it safely, such as unnecessarily increased stroke risk.

Therefore, a detailed medical chart review methodology was used at 1 VHA facility to determine:

• Whether VHA AF patients with MHCs were less likely than those without MHCs to receive warfarin, even in the absence of contraindications.

• Whether, among those treated with warfarin and monitored in VHA, those with MHCs were more likely than those without MHCs to have INRs falling outside the standard therapeutic range.

Patients were drawn from a national database of AF/atrial flutter cases developed for another ongoing study, where AF was identified by the presence of

 International Classification of Diseases, Ninth Revision (ICD-9)

 code 427.3x in fiscal year 2001-2002 VHA or Medicare outpatient or inpatient administrative data. Patients with AF/atrial flutter were included regardless of whether or not the arrhythmia was thought to be paroxysmal, and regardless of duration of the diagnosis (except for transient AF; see below). From the 2490 members of this database who received more than 50% of their VHA outpatient care in both 2002 and 2003 at a suburban, West Coast VHA facility, a random sample of 450 was selected for detailed chart review of electronic medical records. After refining the standardized abstraction protocol on 50 cases not included in the study, a trained senior medical student identified information on variables of interest and recorded them on a standardized abstraction form. Ongoing quality control occurred at weekly meetings of the research team to ensure consistency in coding.

 Administrative data supplemented chart data for some variables.

Excluded were 6 patients with inaccurate chart record identifiers, 33 patients who died prior to the end of 2003, 17 patients with transient AF (occurring only during an acute hospitalization, or within 30 days of cardiac surgery or cardiac catheterization), and 98 with no VHA chart evidence of AF based on data from available electrocardiograms, patient problem list, or outpatient provider diagnoses from progress notes. (Among this latter subgroup, 75 were identified as having AF by Medicare

 codes only; likely their AF was managed by a non-VHA provider, and thus the diagnosis did not appear in VHA files.) A total of 296 AF cases (full cohort) were thus included in the study. Of these, 51 patients had evidence of 1 or more potential contraindications to warfarin: history of nonadherence, hemorrhage, falls, dementia, advanced liver disease, or malignancy.

 Among the remaining 246 patients, all of whom were apparently eligible to receive warfarin (warfarin-eligible subcohort), 161 (65.4%) received warfarin within or outside of VHA. Among them, 84 received their INR monitoring in the VHA Anticoagulation Clinic and had reliable INR data available, such that warfarin effect could be assessed; they constituted the warfarin-treated-VHA-monitored subcohort. This study was approved by the Institutional Review Board at Stanford University, Stanford, California.

The primary independent variable was MHC. MHC was defined as (1) 1 or more provider diagnosis of substance use disorder, depressive disorder, anxiety disorder, psychotic disorder, dysfunctional personality disorder, or bipolar disorder (in progress notes or problem lists) during 2001-2003; or (2) receipt of psychopharmacologic agents including antidepressants, antipsychotics, or mood stabilizers (based on the VHA medication log or based on a progress note referring to a non-VHA prescription) during 2001-2003.

Dependent variables included (1) receipt of warfarin in 2003 (yes/no), based upon the chart medication log or a progress note referring to a warfarin prescription by a non-VHA provider; (2) presence of at least 1 subtherapeutic INR in 2003 (<1.5, <2.0) based on laboratory result file data; and (3) presence of at least 1 supratherapeutic INR in 2003 (>3.0, >3.5, >4.0, >4.5, >5.0). A standard procedure was used to calculate time in therapeutic range (ie, the percentage of days in 2003 during which the INR fell within the 2.0 to 3.0 range, linearly interpolating INR values for days on which an actual value was not available).

 Time highly out of therapeutic range was calculated as percentage of days in 2003 during which the INR was less than 1.5 or greater than 4.0.

Other variables included known stroke risk factors for AF patients

: age 75 years or older, congestive heart failure, hypertension, diabetes, and prior nonhemorrhagic stroke or transient ischemic attack, based on 2001-2002 VHA outpatient or inpatient administrative data. Comorbidity was identified from the medical component of the Comorbidity Index (a count of medical conditions common in veterans and used for ambulatory care case-mix adjustment), using VHA administrative data.

 A count of primary care visits in 2003 came from clinic type codes in VHA administrative files. Patients were considered to have received only VHA care if they lacked any forms of outside insurance, based on the medical record, and if there was no mention of receipt of outside medical care or medications, based on primary care and cardiology clinic progress notes.

Among the full cohort, the warfarin-eligible subcohort, and the warfarin-treated-VHA-monitored subcohort, the following characteristics were examined for patients with and without MHC: demographics, stroke risk factors, Comorbidity Index, and utilization.

Among warfarin-eligible patients, the next analyses compared the proportions of those with and without MHC who were receiving warfarin, using a X

 test. In this group, multivariable logistic regression was performed to examine receipt of warfarin as a function of the presence of MHC, first unadjusted, then controlling for age, and then controlling for age and Comorbidity Index.

Next, among warfarin-treated-VHA-monitored patients, proportions were sequentially compared for those with and without MHC who had at least 1 low INR (<2.0, <1.5) and for those with and without MHC who had at least 1 high INR (>3.0, >3.5, >4.0, >4.5, >5.0), using a X

 test. Fisher’s exact test was used to calculate 

 values in instances of small cell size. Then logistic regression was used to examine high or low INR as a function of MHC, first unadjusted, then controlling for age, and then controlling for age and Comorbidity Index.

Among warfarin-treated-VHA-monitored patients, patients with and without MHC were compared with respect to mean time in therapeutic range and to mean time highly out of therapeutic range, using 

Sensitivity analyses tested the stability of findings. Because of a concern that some VHA patients may receive their AF care from non-VHA providers, sensitivity analysis 1 examined the effect of restricting analyses to patients who used VHA as their exclusive source of care (ie, for whom all AF care could be reliably captured). Because of the possibility that some patients with a single recording of an AF diagnosis could have been miscoded as being an AF case, or could have had transient AF (despite efforts to exclude transient AF cases), sensitivity analysis 2 examined the effect of restricting analyses to patients who had at least 2 instances of an AF diagnosis from electrocardiograms or primary care/cardiology clinic notes.

Among the full cohort (n = 296), 29.7% had at least 1 diagnosed MHC: substance abuse 7.8%, depression 18.9%, anxiety 15.2%, psychosis 3.0%, dysfunctional personality 1.0%, and mania 0.7%. Patients with MHC tended to be younger than those without MHC, yet tended to have greater medical comorbidity, including stroke risk factors (

). Among the 17.2% of patients with a listed contraindication to warfarin, the most common contraindications cited were dementia (7.8%), history of falls (2.7%), and history of hemorrhage (3.7%). Of note, no records mentioned MHC as a contraindication to warfarin.

In the warfarin-eligible subcohort, 48.5% of those with MHC versus 28.9% of those without MHC were not treated with warfarin (

); even after adjustment for age and comorbidity, those with MHC were significantly less likely to receive warfarin (

). The frequency of not being treated with warfarin varied by specific MHC, as follows: substance abuse 66.7%, depression 50.0%, anxiety 47.4%, psychosis 25.0%, and personality disorder 50.0%. Note that among the warfarin-eligible subcohort (n = 246), 19.7% of patients with MHC and 15.6% of patients without MHC had a CHADS2 (congestive heart failure, hypertension, age 75 years or older, diabetes mellitus, previous stroke or transient ischemic attack) score below 2, a level at which alternatives to warfarin may be appropriate.

Among the warfarin-treated-VHA-monitored subcohort (n = 84), patients with MHC were more likely than patients without MHC to have high INR values, both before and after adjustment for age. This would put them at increased risk for hemorrhage. There was a trend toward the same effect after additionally adjusting for Comorbidity Index. Differences between patients with versus without MHC were even more pronounced at higher INR thresholds. There was also a nonsignificant trend toward higher rates of subtherapeutic INR (<1.5) among patients with MHC.

Among the warfarin-treated-VHA-monitored subcohort, mean time spent in the 2.0 to 3.0 therapeutic range was lower for the MHC group than for the group without MHC: 56.8 (standard deviation [SD] 16.9) versus 65.9 (SD 18.2) percent of days (

 = .04). Mean time spent highly out of range was higher for the MHC group than for the group without MHC: 9.0 (SD 9.8) versus 3.4 (SD 7.1) percent of days (

In sensitivity analysis 1 (restricting to patients who received all care in VHA), the magnitude and direction of unadjusted differences were similar to those in the main analyses. The same was true in sensitivity analysis 2, restricting to patients with 2 or more AF diagnoses (

Warfarin-eligible AF patients with MHC were substantially less likely than those without MHC to receive warfarin, even after controlling for age and comorbidity. This is of potential concern, since stroke risk factors were greater in AF patients with MHC than in those without MHC. When prescribed warfarin, patients with MHC were substantially more likely than those without MHC to have INRs in the highly supratherapeutic range, with 27% having at least 1 INR value above 5.0. There was also a trend for them to have greater risk of having INRs in the highly subtherapeutic range. As a result, these patients could have increased risk of either thromboembolism (in the setting of low INR) or hemorrhage (in the setting of high INR).

The differences in warfarin receipt observed here are consistent with emerging literature documenting lower-intensity care provided to patients with MHC for a range of medical conditions,

 and could reflect several potential mechanisms. Patient preferences could be important,

but have not been well characterized with respect to MHC. Mental illness can influence physicians’ perceptions of patients,

 and physician expectations of patients—founded or not&mdash;may affect management.

 Alternatively, physicians’ threshold for concluding that falls, dementia, or nonadherence (known to affect prescribing behavior

) should be counted as a warfarin contraindication may be lower for patients with MHCs. Mental illness may also affect communication between physician and patient, or may cause patients to be labeled as “difficult.”

Atrial fibrillation patients with mental illness are less likely to receive warfarin anticoagulation; those who do receive warfarin have excess risk of over-anticoagulation.

Alan S. Go, MD

Articles

Anticoagulation in Atrial Fibrillation: Impact of Mental Illness

Ethnic Differences in the Development of Albuminuria: The DISTANCE Study

Mental Illness and Warfarin Use in Atrial Fibrillation

Glycemic Response to Newly Initiated Diabetes Therapies

Efficacy of Anticoagulation for Stroke Prevention and Risk Stratification in Atrial Fibrillation: Translating Trials into Clinical Practice