Using the case mix of pressure ulcer healing to evaluate nursing home performance.

Pressure ulcer healing is an important quality measure for nursing homes, but the factors that predict healing have not been well studied. Using the Minimum Data Set, the authors identified candidate variables for a logistic regression, risk-adjustment model to predict ulcer healing. The authors then assessed model discrimination and calibration. Finally, the authors compared unadjusted with risk-adjusted performance for the individual facilities within a nursing home chain. Significant predictors of healing included mobility in bed, presence of a stage 2 ulcer (compared with a stage 4 ulcer), absence of paraplegia and quadriplegia, and absence of end-stage illness. The model C statistic was 0.67, and the calibration was acceptable. Judgments about nursing performance varied in 2 cases depending upon whether unadjusted or risk-adjusted performance was used. The model that the authors developed contains credible predictors of healing. Pressure ulcer healing may be one of many indicators used to evaluate nursing home quality.

Racial differences in the prevalence of atrial fibrillation among males.

BACKGROUND: Despite being the most common cardiac arrhythmia, little is known about racial differences in atrial fibrillation (AF) prevalence and whether differences persist after accounting for known risk factors. METHODS: We identified male respondents to the 1999 Large Health Survey of Veteran Enrollees who had an AF diagnosis in the VA administrative database during the preceding two years. RESULTS: Of 664,754 male respondents, 5.3% had AF. By race, age-adjusted prevalence was 5.7% in whites, 3.4% in blacks, 3.0% in Hispanics, 5.4% in native Americans/Alaskans, 3.6% in Asians and 5.2% in Pacific Islanders (p<0.001). Of predisposing conditions, whites were more likely to have valvular heart disease, coronary artery disease and congestive heart failure, blacks had the highest hypertension prevalence; Hispanics had the highest diabetes prevalence. Racial differences remained after adjustment for age, body mass index and these comorbidities. White males were significantly more likely to have AF compared to all races but Pacific Islanders [versus blacks, OR=1.84 (95% CI: 1.71-1.98); versus Hispanics, OR=1.77 (1.60-1.97); vs Native Americans, OR 1.15 [1.04-1.27]; versus Asians, OR=1.41 (1.12-1.77) versus Pacific Islanders, OR=1.16 (0.88-1.53)]. CONCLUSIONS: AF prevalence varies by race. White males have the highest AF burden even after adjustment for known risk factors. Recognition of the high AF prevalence, especially among whites, as well as native Americans and Pacific Islanders, should help guide provider practices for screening among older male patients. Further research is necessary to verify and establish reasons for these racial differences.

The effect of age on hypertension control and management.

BACKGROUND: Despite guidelines recommending similar blood pressure (BP) treatment goals regardless of age, controversy exists regarding treating those > or = 80 years of age. Whether this affects current practice in terms of differences in BP control and number of prescribed antihypertensives by age is unknown. METHODS: This was a cross-sectional study of 59,207 outpatients with hypertension treated at 10 Veterans Health Administration sites. Outcome measures were BP control (< 140/90 mm Hg) and number of antihypertensive medications at the patient's last study visit. Uncontrolled BP was also categorized by whether systolic, diastolic, or both were elevated. RESULTS: Subjects 40 to 49 years and those 50 to 59 years of age had better BP control (adjusted odds ratios 1.35 [95% CI = 1.26 to 1.44] and 1.22 [CI = 1.17 to 1.28] respectively) compared with subjects 60 to 69 years of age; those 70 to 79 years of age and > or = 80 years had worse control (OR = 0.92 for both; respective CIs = 0.88 to 0.96 and 0.86 to 0.99). Antihypertensive medication use increased by successive decade to age 80 years, after which the trend reversed. Adjusted mean number of medications by age were: < 40 years, 2.60; 40 to 49, 2.82; 50 to 59, 2.91; 60 to 69, 3.01; 70 to 79, 3.03; > or = 80 years, 2.90 (P < .05 in pairwise comparisons). The trend of number of medications by age did not vary across hypertension categories, despite systolic hypertension increasing and diastolic hypertension decreasing with age. Subjects < 40 years of age were taking the fewest medications, followed by subjects > or = 80 years and then by those 40 to 49, 50 to 59, 70 to 79, and 60 to 69 years of age. CONCLUSIONS: The oldest hypertension patients, despite worse BP control, are being treated less aggressively with fewer medications than their younger counterparts (those 60 to 79 years of age). Our results suggest that current controversy in treating the oldest hypertensive patients is having an impact on actual practice.

Adequate initial antidepressant treatment among patients with chronic obstructive pulmonary disease in a cohort of depressed veterans.

OBJECTIVE: Depression is common among patients with chronic obstructive pulmonary disease (COPD). Patients with COPD may be more likely to have inadequate treatment with antidepressant medications. We tested the hypothesis that depressed patients with COPD have lower odds of adequate duration of antidepressant therapy in the first 3 months of treatment compared to those without COPD. METHOD: Using administrative and centralized pharmacy data from 14 northeastern Veterans Affairs Medical Centers, we identified 778 veterans with depression (ICD-9-CM codes 296.2x, 296.3x, and 311.xx) who were in the acute phase of antidepressant treatment from June 1, 1999, through August 31, 1999. Within this group, we identified those patients with COPD (23%). An adequate duration of antidepressant treatment was defined as ≥ 80% of days on an antidepressant. We used multivariable logistic regression models to determine the adjusted odds of adequate acute phase antidepressant treatment duration. RESULTS: Those patients with COPD had markedly lower odds of adequate acute phase treatment duration (odds ratio = 0.67, 95% CI = 0.47 to 0.96); this was not observed with other medical diagnoses such as coronary heart disease, diabetes mellitus, or osteoarthritis. CONCLUSIONS: The first few months of treatment appears to be a critical period for depressed patients with COPD who are started on antidepressants. The causes for early antidepressant treatment inadequacy among patients with COPD require further investigation. More intensive efforts may be necessary early in the course of treatment to assure high-quality pharmacologic therapy of depressed patients with COPD.

Purchasing or providing nursing home care: can quality of care data provide guidance.

OBJECTIVES: To examine whether quality of care differed for veterans in Department of Veterans Affairs (VA) nursing homes and those on contract in community nursing homes, and whether the VA was contracting with nursing homes providing better quality of care than other nursing homes. DESIGN: Observational study using administrative databases from 1997 to 1999. SETTING: Ten VA and 650 community nursing homes in New York state. PARTICIPANTS: Four thousand seven hundred sixty-three veteran and 195,438 nonveteran residents of these nursing homes. MEASUREMENTS: Risk-adjusted rates of pressure ulcer development, functional decline, behavioral decline, and mortality. RESULTS: Veterans in VA nursing homes were significantly (P< .05) less likely to develop a pressure ulcer (odds ratio (OR)=0.63) but more likely to experience functional decline (OR=1.6) than veterans in community nursing homes. Residents of community nursing homes with VA contracts were significantly (P< .05) less likely to develop a pressure ulcer (OR=0.91) but more likely to die than residents in noncontract homes. Few nursing homes were consistently among the best or worst performers on all measures; only seven of 650 nursing homes were in the top or bottom decile and 34 in the top or bottom quartile for each measure. CONCLUSION: Large purchasers and providers of nursing home care such as the VA are unlikely to find information on quality of care useful in making decisions on whether they should "make" or "buy" care. Nursing homes performing well on one quality measure may perform poorly on another, and it is difficult to identify nursing homes that are consistently among the best or worst. Other consumers may encounter similar difficulties when using data on nursing home quality.

Hypertension management in patients with diabetes: the need for more aggressive therapy.

OBJECTIVE: Clinical trials have demonstrated the importance of tight blood pressure control among patients with diabetes. However, little is known regarding the management of hypertension in patients with coexisting diabetes. To examine this issue, we addressed 1) whether hypertensive patients with coexisting diabetes are achieving lower levels of blood pressure than patients without diabetes, 2) whether there are differences in the intensity of antihypertensive medication therapy provided to patients with and without diabetes, and 3) whether diabetes management affects decisions to increase antihypertensive medication therapy. RESEARCH DESIGN AND METHODS: We abstracted medical records to collect detailed information on 2 years of care provided for 800 male veterans with hypertension. We compared patients with and without diabetes on intensity of therapy and blood pressure control. Intensity of therapy was described using a previously validated measure that captures the likelihood of an increase in antihypertensive medications. We also determined whether increases in antihypertensive medications were less likely at those visits in which the diabetes medications were being adjusted. RESULTS: Of the 274 hypertensive patients with diabetes, 73% had a blood pressure > or =140/90 mmHg, compared with 66% in the 526 patients without diabetes (P = 0.04). Diabetic patients also received significantly (P = 0.05) less intensive antihypertensive medication therapy than patients without diabetes. Less intensive therapy in diabetic patients could not be explained by clinicians being distracted by the treatment for diabetes. CONCLUSIONS: There is an urgent need to improve hypertension care and blood pressure control in patients with diabetes. Additional information is required to understand why clinicians are not more aggressive in managing blood pressure when patients also have diabetes.

Hemoglobin Level and Hospital Mortality Among ICU Patients With Cardiac Disease Who Received Transfusions.

BACKGROUND: There is a paucity of randomized clinical trial data on the use of red blood cell (RBC) transfusion in critically ill patients, specifically in the setting of cardiac disease. OBJECTIVES: This study examined how hemoglobin (Hgb) level and cardiac disease modify the relationship of RBC transfusion with hospital mortality. The aim was to estimate the Hgb level threshold below which transfusion would be associated with reduced hospital mortality. METHODS: We performed secondary data analyses of Veterans Affairs intensive care unit (ICU) episodes across 5 years. Logistic regression quantified the effect of transfusion on hospital mortality while adjusting for nadir Hgb level, demographic characteristics, admission information, comorbid conditions, and ICU admission diagnoses. RESULTS: Among 258,826 ICU episodes, 12.4% involved transfusions. Hospital death occurred in 11.6%. Without comorbid heart disease, transfusion was associated with decreased adjusted hospital mortality when Hgb was approximately <7.7 g/dl, but transfusion increased mortality above this Hgb level. Corresponding Hgb level thresholds were approximately 8.7 g/dl when comorbid heart disease was present and approximately 10 g/dl when the ICU admission diagnosis was acute myocardial infarction (AMI). Sensitivity analysis using additional adjustment for selected blood tests in a subgroup of 182,792 ICU episodes lowered these thresholds by approximately 1 g/dl. CONCLUSIONS: Transfusion of critically ill patients was associated with reduced hospital mortality when Hgb level was <8 to 9 g/dl in the presence of comorbid heart disease. This Hgb level threshold for transfusion was 9 to 10 g/dl when AMI was the ICU admission diagnosis.

Profiling nursing homes using Bayesian hierarchical modeling.

OBJECTIVES: New methods developed to improve the statistical basis of provider profiling may be particularly applicable to nursing homes. We examine the use of Bayesian hierarchical modeling in profiling nursing homes on their rate of pressure ulcer development. DESIGN: Observational study using Minimum Data Set data from 1997 and 1998. SETTING: A for-profit nursing home chain. PARTICIPANTS: Residents of 108 nursing homes who were without a pressure ulcer on an index assessment. MEASUREMENTS: Nursing homes were compared on their performance on risk-adjusted rates of pressure ulcer development calculated using standard statistical techniques and Bayesian hierarchical modeling. RESULTS: Bayesian estimates of nursing home performance differed considerably from rates calculated using standard statistical techniques. The range of risk-adjusted rates among nursing homes was 0% to 14.3% using standard methods and 1.0% to 4.8% using Bayesian analysis. Fifteen nursing homes were designated as outliers based on their z scores, and two were outliers using Bayesian modeling. Only one nursing home had greater than a 50% probability of having a true rate of ulcer development exceeding 4%. CONCLUSIONS: Bayesian hierarchical modeling can be successfully applied to the problem of profiling nursing homes. Results obtained from Bayesian modeling are different from those obtained using standard statistical techniques. The continued evaluation and application of this new methodology in nursing homes may ensure that consumers and providers have the most accurate information regarding performance.

The relationship of system-level quality improvement with quality of depression care.

OBJECTIVE: To explore the relationship of systemwide continuous quality improvement (CQI) with depression care quality in the Veterans Health Administration (VHA). STUDY DESIGN: Observational study using data from 2 VHA studies. PATIENTS AND METHODS: The Depression Care Quality Study (DCQS) was a retrospective cohort study of depression care quality in the northeastern United States involving 12 678 patients cared for at 14 VHA facilities; it used guideline-based process measures (ie, dosage and duration adequacy). The VHA CQI survey was a cross-sectional survey of systemwide CQI among a representative sample of VHA hospitals; it assessed CQI and organizational culture (OC) at 116 VHA hospitals nationwide and provided data on the 14 study facilities. We used analysis of variance to identify differences in the adequacy of depression care among these facilities. Pearson's correlation was used to identify the relationship of CQI and OC with facility-level depression care adequacy. RESULTS: Mean depression care adequacy differed among the 14 DCQS facilities (P < .0001). Overall dosage adequacy was 90% (range: 87%-92%). Overall duration adequacy was 45% (range: 39%-64%). There was no correlation between CQI and either dosage adequacy (r= .004, P= .98) or duration adequacy (r= -.17, P= .55). Similarly, there was no correlation between OC and either dosage adequacy (r= -.35, P= .22) or duration adequacy (r= -.12, P= .68). CONCLUSION: Although CQI may help bridge the healthcare quality gap, it may not be associated with higher disease-specific quality of care.

Patient factors associated with transfusion practices in Veterans Affairs intensive care units: implications for further research.

PURPOSE: We sought to describe how patient characteristics influence the frequency of red blood cell (RBC) transfusions among critically ill patients after taking into account hemoglobin (Hgb) level. METHODS: This was a retrospective cohort study using secondary analysis of administrative data of Veterans Affairs intensive care unit (ICU) admissions. The outcome of interest was RBC transfusion during the first 30 days of ICU admission. Besides Hgb level, explanatory variables included demographics, admission-related information, comorbid conditions, ICU admission diagnosis, and selected laboratory test results. Logistic regression modeling quantified associations between explanatory variables and transfusion. RESULTS: For 259 281 ICU admissions from 2001 to 2005, the overall incidence of RBC transfusion was 12.5%. Increased age, male gender, admission for acute myocardial infarction (AMI), and comorbid heart disease were independently associated with transfusion. Compared with admission for reference diagnoses, transfusions were more likely for admissions for AMI, unstable angina, and congestive heart failure only at Hgb levels below 11, 9, and 6 g/dL, respectively. CONCLUSIONS: Intensive care unit patients admitted for AMI, unstable angina, and congestive heart-failure had higher likelihood of receiving RBC transfusions below specific Hgb levels varying from 6 to 11 g/dL. Further research is needed to determine how these transfusion practices influence outcomes.

Using administrative data to identify mental illness: what approach is best?

The authors estimated the validity of algorithms for identification of mental health conditions (MHCs) in administrative data for the 133 068 diabetic patients who used Veterans Health Administration (VHA) nationally in 1998 and responded to the 1999 Large Health Survey of Veteran Enrollees. They compared various algorithms for identification of MHCs from International Classification of Diseases, 9th Revision (ICD-9) codes with self-reported depression, posttraumatic stress disorder, or schizophrenia from the survey. Positive predictive value (PPV) and negative predictive value (NPV) for identification of MHC varied by algorithm (0.65-0.86, 0.68-0.77, respectively). PPV was optimized by requiring > or =2 instances of MHC ICD-9 codes or by only accepting codes from mental health visits. NPV was optimized by supplementing VHA data with Medicare data. Findings inform efforts to identify MHC in quality improvement programs that assess health care disparities. When using administrative data in mental health studies, researchers should consider the nature of their research question in choosing algorithms for MHC identification.

Monitoring depression care: in search of an accurate quality indicator.

BACKGROUND: Linking process and outcomes is critical to accurately estimating healthcare quality and quantifying its benefits. OBJECTIVES: The objective of this study was to explore the relationship of guideline-based depression process measures with subsequent overall and psychiatric hospitalizations. RESEARCH DESIGN: This is a retrospective cohort study during which we used administrative and centralized pharmacy records for sample identification, derivation of guideline-based process measures (antidepressant dosage and duration adequacy), and subsequent hospitalization ascertainment. Depression care was measured from June 1, 1999, through August 31, 1999. We used multivariable regression to evaluate the link between depression care and subsequent overall and psychiatric hospitalization, adjusting for patient age, race, sex, socioeconomic status, comorbid illness, and hospitalization in the prior 12 months. SUBJECTS: We studied a total of 12,678 patients from 14 Northeastern VHA hospitals. RESULTS: We identified adequate antidepressant dosage in 90% and adequate duration in 45%. Those with adequate duration of antidepressants were less likely to be hospitalized in the subsequent 12 months than those with inadequate duration (odds ratio [OR],.90; 95% confidence interval [CI], .81-1.00). Those with adequate duration of antidepressants were less likely to have a psychiatric hospitalization in the subsequent 12 months than those with inadequate duration (OR, .82; 95% CI, .69-.96). We did not demonstrate a significant link between dosage adequacy and subsequent overall or psychiatric hospitalization. CONCLUSIONS: Guideline-based depression process measures derived from centralized data sources offer an important method of depression care surveillance. Their accuracy in capturing depression care quality is supported by their link to healthcare utilization. Further work is needed to assess the effect of implementing these quality indicators on depression care.

Measuring the quality of depression care in a large integrated health system.

BACKGROUND: Guideline-based depression process measures provide a powerful way to monitor depression care and target areas needing improvement. OBJECTIVES: To assess the adequacy of depression care in the Veterans Health Administration (VHA) using guideline-based process measures derived from administrative and centralized pharmacy records, and to identify patient and provider characteristics associated with adequate depression care. RESEARCH DESIGN: This is a cohort study of patients from 14 VHA hospitals in the Northeastern United States which relied on existing databases. Subject eligibility criteria: at least one depression diagnosis during 1999, neither schizophrenia nor bipolar disease, and at least one antidepressant prescribed in the VHA during the period of depression care profiling (June 1, 1999 through August 31, 1999). Depression care was evaluated with process measures defined from the 1997 VHA depression guidelines: antidepressant dosage and duration adequacy. We used multivariable regression to identify patient and provider characteristics predicting adequate care. SUBJECTS: There were 12,678 patients eligible for depression care profiling. RESULTS: Adequate dosage was identified in 90%; 45% of patients had adequate duration of antidepressants. Significant patient and provider characteristics predicting inadequate depression care were younger age (<65), black race, and treatment exclusively in primary care. CONCLUSIONS: Under-treatment of depression exists in the VHA, despite considerable mental health access and generous pharmacy benefits. Certain patient populations may be at higher risk for inadequate depression care. More work is needed to align current practice with best-practice guidelines and to identify optimal ways of using available data sources to monitor depression care quality.