 To determine whether publicly reporting hospital scores on antibiotic timing in pneumonia (percentage of patients with pneumonia receiving antibiotics within 4 hours) has led to unintended adverse consequences for patients.

Retrospective analyses of 13,042 emergency department (ED) visits by adult patients with respiratory symptoms in the National Hospital Ambulatory Medical Care Survey, 2001-2005.

Rates of pneumonia diagnosis, antibiotic use, and waiting times to see a physician were compared before and after public reporting, using a nationally representative hospital sample. These outcomes also were compared between hospitals with different antibiotic timing scores.

 There were no differences in rates of pneumonia diagnosis (10% vs 11% of all ED visits, P = .72) or antibiotic administration (34% vs 35%, P = .21) before and after antibiotic timing score reporting. Mean waiting times to be seen by a physician increased similarly for patients with and without respiratory symptoms (11-minute vs 6-minute increase, respectively; P = .29). After adjustment for confounders, hospitals with higher 2005 antibiotic timing scores had shorter mean waiting times for all patients, but there were no significant score-related trends for rates of pneumonia diagnosis or antibiotic use.

 Despite concerns, public reporting of hospital antibiotic timing scores has not led to increased pneumonia diagnosis, antibiotic use, or a change in patient prioritization.  (Am J Manag Care. 2009;15(2):137--144)

Take-Away Points There has been concern that publicly reporting hospital scores on antibiotic timing in pneumonia (percentage of patients with pneumonia who received antibiotics within 4 hours) has led to unintended adverse consequences for patients.

A national sample of hospitals revealed little evidence that this public reporting has led to widespread overdiagnosis of pneumonia or inappropriate antibiotic administration.

 Explainable variation in hospitals’ antibiotic timing scores is primarily attributable to differences in patients’ waiting times to see a physician, rather than differences in rates of pneumonia diagnosis or antibiotic administration. 

Future monitoring of the effects of public reporting programs may provide valuable guidance to policy makers, especially in areas of controversy.

To encourage improvement in hospitals’ quality of care, the Hospital Quality Alliance (HQA) began an initiative to collect and publicly report hospital-level performance on 10 quality measures in 2004.

 More than 98% of US acute care hospitals supply performance data to the HQA,

 but concerns have been raised about potential unintended consequences of public reporting.

Hospitals’ responses to the HQA measure “Initial Antibiotic Received within 4 Hours of Hospital Arrival” have been of particular concern. Hospitals feeling pressure to improve antibiotic timing performance could potentially “play for the test” by encouraging the premature (and potentially inaccurate) diagnosis of pneumonia, giving antibiotics indiscriminately to patients with respiratory symptoms, or inappropriately prioritizing patients likely to have pneumonia ahead of others whose medical conditions may be more urgent.

Prior studies from single institutions and self-selected hospitals participating in a pay-for-performance pilot program suggest that incentives tied to pneumonia antibiotic timing scores have led to increased rates of inaccurate pneumonia diagnosis and inappropriate antibiotic administration in emergency departments (EDs).

 However, whether public reporting on antibiotic timing has had similar effects on a national scale is unknown.

Because antibiotic timing scores are thought to reflect care delivered in EDs,

 we used a national database of ED visits and compared the care for patients with respiratory symptoms before and after the start of public reporting. We assessed whether these patients were more likely to be diagnosed with pneumonia, to be prescribed antibiotics, and to have shorter waiting times to see a physician (compared with patients who did not have respiratory symptoms, reflecting patient prioritization). To test the hypothesis that hospitals with higher scores were “playing for the test,” we also assessed differences on these 3 measures between hospitals scoring higher and lower on the pneumonia antibiotic timing measure.

We used patient visit data from the Emergency Department module of the nationally representative National Hospital Ambulatory Medical Care Survey (NHAMCS), which is administered annually by the National Center for Health Statistics (NCHS).

 A rotating panel of nonfederal, general, and short-stay hospitals participate in the NHAMCS, and visits from each hospital in the panel are sampled approximately every 15 months.

Trained hospital staff record patient and clinical data on standardized forms for each visit. Patient data include demographic information, expected source of payment, and nursing home residence. Clinical data include up to 3 reasons for visit, up to 3 physician diagnoses (in the ED), up to 8 medications administered during the ED visit (but not the timing of medication administration), waiting time to see a physician (2003-2005 only), triage vital signs, and orientation to person, place, and time.

The NHAMCS Emergency Department module collected 182,332 patient visit records between 2001 and 2005. Among contacted hospitals over the study period, 90% to 95% participated. Visits are weighted to allow extrapolation of survey results to national estimates.

 For our analysis the NCHS created anonymous hospital identifiers that allowed longitudinal tracking of each participating hospital. The NCHS institutional review board approved all NHAMCS protocols, and the confidentiality of the data is protected by law.

Hospital Antibiotic Timing Scores and Hospital Characteristics 

We obtained publicly available hospital-level HQA performance data on the timing of initial antibiotics delivered to patients admitted with pneumonia for 2004 and 2005. Hospital scores were calculated as the percentage of adult patients discharged from the hospital with a diagnosis of pneumonia who received their first dose of antibiotics within 4 hours of hospital arrival. Detailed specifications for this measure are available elsewhere.

We also linked hospitals’ pneumonia antibiotic timing scores to other hospital characteristics obtained from the 2005 database of the American Hospital Association: number of beds, geographic region, urban location, ownership (for-profit, not-for-profit, and government), status of membership in the Council of Teaching Hospitals, and percentage of patients covered by Medicare and Medicaid. Of the 507 unique hospitals participating in the NHAMCS Emergency Department module during 2001-2005, NCHS staff matched 503 (99%) to their HQA pneumonia antibiotic timing scores and other corresponding hospital characteristics.

Consistent with prior literature, we included for analysis only hospitals reporting stable antibiotic timing scores, defined as scores calculated using at least 25 patient discharges during the year 2005.

 Of the 503 NHAMCS sample hospitals, 118 (23%) were excluded based on this criterion.

Our study population included ED visits during 2001-2005 by patients age 18 years and older whose primary reason for visit was “symptoms referable to the respiratory system” or “diseases of the respiratory system,” excluding conditions limited to the upper respiratory tract (eg, nasal congestion). Among included visits, the most common specific reasons for visit were cough (50%), shortness of breath (24%), and “labored or difficult breathing” (11%). In supplementary analyses, inclusion of visits for upper respiratory conditions did not substantively alter our results. Visits were included regardless of patients’ dispositions at the end of each ED visit (eg, admitted to hospital, transferred, discharged to home). Supplementary analyses limited to visits resulting in hospital admission did not substantively alter our results.

Outcome Variables: Processes of Emergency Department Care

We had 3 major outcome variables: ED diagnosis, antibiotic use, and waiting time to see a physician. We used 

International Classification of Diseases, Ninth Revision, Clinical Modification

 (ICD-9-CM) codes to classify ED diagnoses as pneumonia (with the same codes used for antibiotic timing score reporting),

 bronchitis, congestive heart failure, or other. It was possible for a single visit to carry more than 1 of these diagnoses, and we counted visits receiving diagnoses in more than 1 category toward the total in each applicable diagnostic category.

Antibiotic use was identified using the NCHS drug classification system.

 As in previous studies, we classified antibiotic use in visits for asthma and congestive heart failure as inappropriate when pneumonia was not also an ED diagnosis.

 In supplementary analyses we also included antibiotic use in bronchitis as inappropriate without substantive changes to the results.

We performed 2 main comparisons. First, we analyzed nationwide longitudinal trends and differences in the outcome variables (ED diagnosis, antibiotic use, and waiting time to see a physician) before and after the start of public score reporting among ED visits for respiratory symptoms. We designated January 1, 2004, as the first day of the public reporting period because this was the first day of care that could contribute to publicly available antibiotic timing scores. Because the antibiotic timing measure was first published in October 2003,

 we designated October 1, 2003, as the first day of the reporting period in supplementary analyses with substantively similar results.

Second, we conducted a cross-sectional analysis of relationships between 2005 antibiotic timing scores and the outcome variables, restricting our analysis to ED visits in the public reporting period (2004-2005). We hypothesized that if hospitals were “playing for the test” in order to raise their scores, patients with respiratory symptoms visiting hospitals with the highest 2005 scores would experience the highest rates of pneumonia diagnosis, the highest rates of antibiotic use, and (relative to patients without respiratory symptoms) the shortest waiting times.

We assessed relationships between categorical variables using the 

 test. Due to the nonnormal distribution of waiting time to see a physician, waiting times were modeled using generalized log-linear regression. 

For the longitudinal analysis we constructed multivariable logistic regression models predicting each categorical outcome variable (ED diagnosis and antibiotic use) as a function of time period (prereporting vs during reporting) and the above-listed hospital-level characteristics from the American Hospital Association database. To adjust for case severity and patient characteristics known to be associated with ED processes of care,

 the models included patient age, sex, orientation to person/place/time (dichotomous), presence of fever (triage temperature >100.4°F), residence in a nursing home  dichotomous), race and ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other), anticipated source of payment (private, Medicare, Medicaid, or other), and season of the visit (winter, spring, summer, or fall). We constructed multivariable log-linear regression models for waiting time, including the same potential confounders. For the cross-sectional analysis we created similar models using hospitals’ 2005 antibiotic timing scores (continuous variable) rather than time period as the major predictor.

In order to assess the contribution of ED care processes to variation in antibiotic timing scores, we constructed supplementary linear regression models predicting the 2005 antibiotic timing score of the hospital associated with each ED visit as a function of pneumonia diagnosis, antibiotic use, and waiting time. We assessed the independent, mutually adjusted contribution of each explanatory variable to overall variation in scores by comparing the partial R2 values for each predictor. All analyses were performed using SUDAAN version 9.0.0 (Research Triangle Institute, Research Triangle Park, NC) to account for data clustering and the complex NHAMCS sampling design.

There were 385 hospitals participating in the NHAMCS from 2001 to 2005 that also reported pneumonia antibiotic timing scores based on at least 25 observations in 2005. There were no significant differences between the NHAMCS hospital sample and the overall population of US hospitals reporting at least 25 antibiotic timing score observations in 2005 (

Based on the NHAMCS sample, there were an estimated 40 million (95% confidence interval, 39 to 42 million) ED visits to hospitals by adults with respiratory symptoms between 2001 and 2005. These visits for respiratory symptoms represented 11% of all ED visits, and this percentage was constant across all data years. There were no significant differences between the prereporting (2001-2003) and reporting (2004-2005) periods in patient sociodemographic factors (age, sex, race, and ethnicity), nursing home residence, or expected payment source (data not shown). However, patients in the prereporting period were less likely to have a fever (8% prereporting vs 11% during reporting, 

 = .003) and to be oriented to person, place, and time (77% prereporting vs 85% during reporting, 

There was no significant difference in the rate of pneumonia diagnosis among ED visits for respiratory symptoms between the prereporting and during reporting periods (

). Antibiotic use during ED visits increased among visits resulting in an ED diagnosis of pneumonia (70% prereporting vs 78% during reporting, 

 = .01), but there were no significant changes in antibiotic use among patients with other ED diagnoses. Mean waiting times to be seen by a physician increased similarly for patients with and without respiratory symptoms (11-minute vs 6-minute increase, respectively; 

 = .29). Adjustment for potential confounders including fever and orientation did not significantly alter

In our second analysis we examined cross-sectional relationships between hospitals’ 2005 pneumonia antibiotic timing scores and the outcome variables during public reporting. Pneumonia diagnosis rates did not significantly increase with antibiotic timing score (

). (Adjusted diagnosis rates are given in the 

 available at www.ajmc.com.) Patients visiting higher-scoring hospitals were more likely to receive an antibiotic (31% in the lowest score quintile vs 44% in the highest, 

 = .004 for trend) and had shorter mean waiting times regardless of respiratory symptoms (66 vs 38 minutes with respiratory symptoms, 

 <.001 for trend; 69 vs 38 minutes without respiratory symptoms, 

 = .45 for trend in difference between symptom categories). After adjustment for confounders, only the relationship between shorter waiting times and higher-scoring hospitals remained a statistically significant trend.

Differences between hospitals in rates of pneumonia diagnosis, antibiotic prescribing, and waiting time explained only 4% of overall variation in 2005 antibiotic timing scores, suggesting a weak relationship between the measured ED processes of care and hospital antibiotic timing scores (data not shown). Of this explainable score variation, waiting times accounted for 79%, and antibiotic usage accounted for 3%.

Despite fears that publicly reporting hospitals’ pneumonia antibiotic timing scores would lead to increased pneumonia diagnosis, indiscriminate antibiotic use, and inappropriate prioritization of patients with respiratory symptoms,

 we found little evidence of these unintended consequences in a nationally representative cohort of ED visits. Rates of pneumonia diagnosis and overall antibiotic use did not exhibit significant changes over time. Waiting times to see a physician increased similarly for patients with and without respiratory symptoms over 2003-2005, arguing against higher prioritization of patients likely to have pneumonia.

Moreover, cross-sectional analyses of ED visits during the public reporting period revealed that after adjustment for confounders, only waiting times differed significantly between hospitals with higher and lower antibiotic timing scores. Successful efforts to shorten waiting times for all patients would be better described as quality improvements than as adverse consequences.

Although hospitals in the highest antibiotic timing score quintile had the highest rates of antibiotic administration for inappropriate diagnoses, the lack of a statistical trend between antibiotic use and timing scores suggests that excessive antibiotic administration does not significantly contribute to the timing score ranking for most hospitals. Also, analysis of overall score variation revealed that only a very small percentage was attributable to differences in antibiotic administration rates. These findings are consistent with the overall stability of antibiotic administration rates before and after the start of public reporting.

However, if there are persistent concerns that hospitals seeking to achieve the highest antibiotic timing scores will, in the future, have increased rates of inappropriate antibiotic administration, then strategies focused on the top-scoring hospitals may be considered. For example, publicly reporting a score “band” rather than an exact score for hospitals scoring above a certain threshold could attenuate the incentive to achieve scores in the range generating these concerns.

The implications of our longitudinal analysis differ from those suggested by some earlier reports. However, key differences in cohort design between our analysis (which included all patients presenting to EDs with respiratory complaints) and earlier single-institution studies (which included only those patients admitted with a pneumonia diagnosis) may not allow direct comparison of findings.

 A study among self-selected Premier client hospitals in the Hospital Quality Incentive Demonstration (HQID) revealed higher rates of antibiotic use for inappropriate diagnoses (heart failure, asthma, and chronic obstructive pulmonary disease) at hospitals with higher antibiotic timing scores.

 However, HQID hospitals faced financial incentives directly tied to antibiotic timing performance, and while all hospitals in our analysis publicly reported their antibiotic timing scores, no data were available to identify which hospitals in our analysis had antibiotic timing—based financial incentives. It is possible that compared with public  reporting, financial incentives could have different effects on patient treatment patterns. Prior studies agreeing with our findings demonstrated associations between ED overcrowding (an established contributor to longer waiting times) and lower antibiotic timing scores.

Our study has limitations. First, the NHAMCS does not assess the accuracy of ED diagnoses, so we cannot directly conclude that diagnostic accuracy was unaffected by public reporting. However, because rates of pneumonia diagnosis did not change, any increased inaccuracy would have had to split equally between errors of commission and mission. Second, we lacked complete clinical information (eg, presence or absence of infiltrate on chest X-ray, comorbid illnesses) to perform fuller case-mix adjustment. Third, our measure of patient prioritization (waiting times to see a physician) did not extend to other important processes of ED care, such as immediacy of imaging.

 Fourth, our analysis included ED visits by patients whose primary reason for visit was a symptom referable to the lower respiratory tract. It is possible that public reporting of antibiotic timing scores would have a different impact on patients with symptoms that were less specific for pneumonia (eg, delirium, fever, chest pain). Finally, absence of proof is not proof of absence. Failure to detect significant unintended consequences of public antibiotic timing score reporting could be due to insufficient statistical power. However, observed rates of pneumonia diagnosis and antibiotic use were remarkably stable over time, with waiting times for all patients (regardless of respiratory symptoms) accounting for the majority of explainable between-hospital variation in antibiotic timing scores.

External incentive programs designed to encourage healthcare quality improvement are becoming increasingly common, and concerns about unintended consequences of these programs have surfaced in a variety of clinical settings.

 Many of these concerns focus on providers “playing for the test” or “gaming the system.”

 In response to concerns that the pneumonia antibiotic timing measure had led to adverse unintended consequences, the Joint Commission and the National Quality Forum changed the cutoff for timely initial antibiotic in pneumonia from 4 hours to 6 hours after hospital arrival and excluded cases of “diagnostic uncertainty” from score calculation.

 The Infectious Diseases Society of America eliminated the time cutoff altogether, recommending only that initial antibiotics be received in the ED.

 To the extent that the outcomes we examined reflect possible adverse unintended consequences, our results do not support these changes.

In summary, we found that during the first 2 years of public reporting of hospital pneumonia antibiotic timing scores, concerns about potential widespread unintended consequences were not substantiated by the national experience. Patterns of ED pneumonia diagnosis and antibiotic use among patients with respiratory symptoms have remained stable over time. The EDs of hospitals with higher antibiotic timing scores distinguish themselves by having shorter waiting times to see a physician, suggesting that these scores communicate information of real importance to patients and payers. However, providers’ concerns about the potential adverse consequences of public reporting and pay-for-performance programs deserve attention.

 Monitoring systems that target these concerns and prospectively measure the patient-level effects of quality improvement initiatives may provide valuable guidance (and reassurance) to policy makers.

Acknowledgments  We thank Robert Krasowski, MA, MS, Susan Schappert, MA, and Peter Meyer, MA, MPH, at the National Center for Health Statistics for invaluable assistance in database linkage, management, and storage. We thank Stuart Lipsitz, ScD, Division of General Medicine, Brigham and Women’s Hospital, for statistical consultation. We thank Ashish Jha, MD, MPH, of the Division of General Medicine, Brigham and Women’s Hospital, and Dale Bratzler, DO, MPH, of the Oklahoma Foundation for Medical Quality for helpful comments on earlier drafts of this manuscript.  Author Affiliations: From the Division of General Medicine (MWF, JAL), Brigham and Womens’ Hospital, Boston, MA; the Department of Medicine (JAL) Harvard University, Boston, MA; the Division of General Internal Medicine (AM), University of Pittsburgh; and RAND Health (AM), Pittsburgh, PA.  Funding Source: The study was supported by the Brigham and Women’s Hospital Primary Care Teaching and Education Fund. Dr Friedberg was supported by a National Research Service Award from the Health Resources and Services Administration (5 T32 HP1100120). Dr Mehrotra was supported by a Career Development Award (KL2 RR-024154-03) from the National Center for Research Resources (NCRR), a component of the National Institutes of Health. Dr Linder was supported by a Career Development Award (K08 HS014563) from the Agency for Healthcare Research and Quality. No funding source had a role in the study design, analysis, or manuscript preparation.   Author Disclosure: Dr Linder reports having received grant support from Roche Pharmaceuticals and Pfizer, Inc. The other authors (MWF, AM) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. Dr Friedberg had full access to all of the de-identified study data, which are stored in secure facilities at the National Center for Health Statistics. Dr Friedberg performed and takes responsibility for the accuracy of the data analysis. Partial results from this work were presented at the 2007 AcademyHealth Annual Research Meeting, June 4, 2007, Orlando, FL.   Authorship Information: Concept and design (MWF, AM, JAL); acquisition of data (MWF, JAL); analysis and interpretation of data (MWF, AM, JAL); drafting of the manuscript (MWF, JAL); critical revision of the manuscript for important intellectual content (AM, JAL); statistical analysis (MWF, JAL); obtaining funding (MWF, JAL); and supervision (JAL).  Address correspondence to: Jeffrey A. Linder, MD, MPH, Division of General Medicine, Brigham and Women’s Hospital, 1620 Tremont St, BC-3-2X, Boston, MA 02120. E-mail: jlinder@partners.org.

1. Williams SC, Schmaltz SP, Morton DJ, et al. 

Quality of care in U.S. hospitals as reflected by standardized measures, 2002-2004. 

 Care in U.S. hospitals—the Hospital Quality Alliance program. 

 National study on the quality of emergency department care in the treatment of acute myocardial infarction and pneumonia. 

Articles

The American Journal of Managed Care

Nationwide data on hospital emergency department visits reveal little evidence of unintended adverse consequences associated with publicly reporting hospitals’ antibiotic timing in pneumonia.

Reporting Hospitals" Antibiotic Timing in Pneumonia: Adverse Consequences for Patients?

 To examine the level of care in a large provider organization with respect to the Diabetes Physician Recognition Program (DPRP) standards of care and describe treatment patterns, diabetesrelated complications, and achievement of clinical goals among patients stratified by glycemic control levels.

 Observational, retrospective, chart review study assessing care among patients with diabetes.

 Diabetic patients aged ≥5 years who were prescribed insulin or oral hypoglycemics/ antihyperglycemics for at least 12 months by 1 of 14 DPRP-participating physicians from the provider organization were eligible. A sample of patient medical charts was collected. Descriptive statistics were generated to assess demographic and clinical variables, with subanalyses for patients in the 3 glycosylated hemoglobin (A1C) cohorts. Data were used to describe the demographics, disease prevalence, comorbidities, clinical outcomes, and treatment patterns of the study population. Results were assessed according to national treatment guidelines.

 Almost all DPRP recognition measures were met and/or exceeded. More than 90% of patients received appropriate assessments. The majority achieved the A1C, blood pressure, and low-density lipoprotein control levels recommended by national treatment guidelines. Patients with multiple comorbidities had worse levels of control, with only 14.3% of patients achieving all 3 treatment goals. Nearly 30% of patients had diabetes-related complications, most commonly kidney disease.

 Differences in control and treatment patterns exist in patients with varying levels of glycemic control. Opportunities exist to improve diabetes care through goal attainment. Further research is needed to determine whether specific measures of care correlate with levels of glycemic control.

A retrospective chart review collected clinical data on patients with diabetes to assess the level of care according to the Diabetes Physician Recognition Program (DPRP) recommendations.

Almost all DPRP recognition measures were met and/or exceeded.

The majority of patients achieved the glycosylated hemoglobin, blood pressure, and lowdensity lipoprotein control levels recommended by national treatment guidelines, although fewer than 15% were at goal with all 3 measures.

Further research may determine whether these results are reproducible and whether specific measures of care correlate with levels of glycemic control.

An estimated 20.8 million Americans—7% of the population&mdash;have diabetes.

 This proportion is expected to increase 198% to encompass 11.5% of the population by 2050.

According to 2002 data, total direct and indirect costs of managing diabetes were $132 billion—twice the cost of care of an otherwise-matched nondiabetic patient population.

Because the burden of diabetes is largely derived from long-term microvascular and macrovascular complications, the American Diabetes Association (ADA) standards of care reflect a multifaceted approach directed at smoking, glycosylated hemoglobin (A1C), blood pressure (BP), and lipid treatment targets to impact vascular consequences.

 Early, aggressive care targeted at A1C control can delay or abate the advent of complications.

 Nationally recognized statistics provide insight into the impact of controlling glucose, BP, and lipids on complications

: for every 1% decrease in A1C, microvascular complications are reduced by 40%; effective BP control can impact macrovascular disease by 33% to 50% and microvascular disease by 33%; and cholesterol management can decrease cardiovascular complications by 20% to 50%.

To provide physicians with resources that support consistent delivery of multifaceted quality care, the National Committee for Quality Assurance and the ADA created the Diabetes Physician Recognition Program (DPRP) in 1997.

 Using established measures of care, participating providers collect data on diabetic patients. Providers who meet or exceed national performance thresholds receive public recognition for providing care according to evidence-based measures. Providers recognized by DPRP may gain health plan support in the form of data collection assistance, pay-for-performance incentives, offset application fees, and entry into high-performance networks.

In a large Midwest provider organization, we examined care related to DPRP standards through a retrospective chart review of patients with diabetes, looking at treatment patterns, diabetes-related complications, and clinical goal achievement among patients stratified into groups of good (A1C <7%), fair (A1C 7%-9%), and poor (A1C >9%) glycemic control. Based on DPRP methodology, participants aged ≥5 years with a diabetes diagnosis and/or notation of prescribed insulin or oral hypoglycemics/antihyperglycemics for >12 months were included. Fourteen DPRP physicians from the provider organization volunteered to participate. Eligible patients were sequentially identified by using DPRP sampling criteria. An estimated sample size of 411 charts was required to detect a difference in A1C control. Oversampling was built in to account for unavailable/missing charts. Each physician identified a minimum of 34 patients for review, resulting in 501 charts collected and analyzed.

The study protocol was approved by the New England Institutional Review Board. Medical records from 2005-2006 were reviewed to collect demographics and clinical measures, including diabetes type, comorbidities, risk factors, target organ damage, measures of diabetes care, laboratory data, and prescribing patterns. Statistics were generated to assess demographic and clinical variables, with subanalyses

conducted for the 3 cohorts. Only patients with A1C values in their charts were included in this subanalysis. Results were assessed according to treatment guidelines established by ADA standards,

The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7),

 the Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel [ATP] III),

 and Healthcare Effectiveness Data and Information Set and DPRP standards.

Most patients (490 of 501) had type 2 diabetes. Mean age was 62 years (SD ± 12.7 years), with 40.9% of patients aged >65 years. More than half were male (55.5%), and 88.0% were overweight or obese (mean body mass index of 33.0 kg/m2). Almost all had >1 comorbidity, most commonly hypertension (98.4%), hyperlipidemia (73.3%), and cerebrovascular disease (40.3%). Most (71.3%) had diabetes, hypertension, and hyperlipidemia. The average number of physician visits was 4.4 per year (range, 1-20 visits).

Overall Achievement of Diabetes Care Measures

More than 90% of the patients received appropriate assessments, except urinalysis to assess nephropathy (74.3%) and an eye exam (47.1%). Rates for these assessments were below the minimum DPRP criteria of 80% and 60%, respectively. Additionally, 375 (74.9%) patients received weight loss counseling. All 3 values (A1C, low-density lipoprotein [LDL], and BP) were recorded for 434 (86.6%) patients. The majority achieved control levels recommended by national treatment guidelines. Mean A1C was 6.9% (SD ± 1.3%), consistent with ADA’s goal of <7%; 62.4% of patients achieved this level. Mean LDL was 94.1 mg/dL (SD ± 31.8 mg/dL), consistent with ATP III’s goal of <100 mg/dL; 62.1% of patients achieved this level. Although mean BP in this group (129.7/76.4 mm Hg) was consistent with JNC 7 recommendations for diabetics (<130/80 mm Hg), only 37.2% achieved this level. Not surprisingly, multiple comorbidities worsened levels of control. Only 62 of 434 patients (14.3%) with A1C, BP, and LDL values recorded achieved all 3 goals. Despite these levels, 29.3% of patients (n = 147) had documentation of a diabetesrelated complication, most commonly some form of kidney disease (78 had microalbuminuria, 2 had macroalbuminuria, and 28 had a nephropathy diagnosis), followed by retinopathy (33 patients). As shown in the 

, DPRP performance was consistent with the overall trend toward greater measure attainment and improved patient care over time.

 Among DPRP physicians, the rate of A1C control (<7%) improved from 37% to 46% from 2000 to 2003 and was reported at 62% in this provider organization. Similar improvement was seen for LDL and BP control.

Although most patients received antihyperglycemic medication therapy, 12.8% with diabetes alone, 11.2% with comorbid hypertension, and 7.1% with comorbid hyperlipidemia received no medication. Most received monotherapy or therapy with only 1 therapeutic class of antihyperglycemics (40.1%), followed by dual therapy (30.3%). Patients with diabetes and comorbid hypertension were more likely to receive dual antihypertensive therapy (30.4%) or triple therapy (30.2%) than monotherapy (28.2%), whereas most with comorbid hyperlipidemia received antihyperlipidemic monotherapy (76.6%).

Achievement of Measures and Treatment Patterns by Subgroups To assess differences in care, the data were further analyzed by subgroups based on the mean A1C level achieved: in subgroup 1, A1C was <7% (n = 303); in subgroup 2, A1C was 7% to 9% (n = 145); in subgroup 3, A1C was >9% (n = 38). The subgroups were similar with respect to age, race, body mass index, and comorbidities. The mean number of physician visits was lowest in subgroup 1 (4.3 visits), followed by subgroup 2 (4.4 visits), and subgroup 3 (5.4 visits). Examination of DPRP measures for diabetes care demonstrated that most patients in all 3 subgroups received appropriate disease assessment for A1C, BP, LDL, and foot exams. However, no subgroup achieved the minimum DPRP criteria for eye exams, smoking, or urinalysis (

A general trend of worsening hypertension and hyperlipidemia control was observed for patients with poorer A1C control. In subgroup 3 (mean A1C 10.2%), mean systolic BP was slightly higher than the JNC 7 goal for patients with diabetes (132.1/77.4 mm Hg), and mean LDL exceeded the ATP III goal (119.2 mg/dL). Across all 3 subgroups, only about one third of patients achieved BP <130/80 mm Hg; more in subgroups 1 and 2 achieved BP <140/90 mm Hg than in subgroup 3. Similarly, the proportion achieving LDL <100 mg/dL decreased from 63.0% in subgroup 1 to 35.3% in subgroup 3; whereas more patients in all subgroups achieved LDL <130 mg/dL, relatively fewer achieved this goal in subgroup 3.

Our results demonstrate that the level of diabetes care in the large provider organization met commonly accepted measures of care. Physicians in this group performed better than the historic DPRP performance from 1997 to 2003 (as shown in the Figure),

 although the small sample size of reporting physicians in this study should be noted. The overall trend seen in DPRP indicates such programs may improve quality of care. Their value is providing not only resources to consistently measure care, but also a positive method to promote quality of care. Providers gain value from public recognition of their quality of care and receive preferred provider status or financial incentives in some quality initiatives. Diabetes Physician Recognition Program specifically focuses on goal attainment for LDL, A1C, and BP because these have been linked to improved clinical outcomes in diabetic patients.

 Diabetes Physician Recognition Program performance measures are based on national guidelines, establishing a link between nationally recognized recommendations and actual clinical practice. Diabetes Physician Recognition Program’s impact on improving measures of care is well documented; participating DPRP providers show 50% to 100% improvements in screening rates for nephropathy and lipids as well as improved BP and LDL control.

Although control rates and process measures were consistent with national treatment goals and DPRP measurement criteria, they highlight opportunities to improve diabetes care. For example, fewer than 15% of patients with comorbid diabetes, hypertension, and hyperlipidemia were at goal with all 3 measures. Additionally, stratification by level of A1C control demonstrated that the level of BP and LDL control worsened in patients with poor A1C control (>9%). The proportion of patients with LDL <100 mg/dL, in particular, dropped from as high as 65% to 35% of patients with poor A1C control. Although not as steep a decline, BP <140/90 mm Hg worsened as A1C control worsened. Interestingly, despite the fact that patients in the poor A1C control subgroup had worse mean A1C, systolic and diastolic BP, and mean LDL, more patients in this subgroup received the recommended testing (A1C, BP, LDL, eye exams, and diabetic foot exams). These patients were not substantially different from the good and moderate A1C subgroups in terms of level of comorbid hypertension or hyperlipidemia, but did have a higher incidence of risk factors or target organ damage (family history, heart failure, premature cardiovascular disease, and smoking). As expected, more patients in the poor A1C subgroup received dual or triple antihyperglycemic medication therapy; unexpectedly, however, more patients in this group received no antihypertensive or antihyperlipidemic medication therapy. These findings may indicate a subset of patients inherently difficult to treat and a potential role for more aggressive medication therapy in these patients with suboptimal A1C control (>9%) may benefit from additional treatment to lower cardiovascular risk.

Limitations to this analysis should be noted:

Data collected from medical charts may be inherently limited by lack of documentation of tests, medications, or lab values in the chart.

Medications identified in the charts were those that were prescribed; whether patients actually received the medication and level of adherence are unknown.

Because data were obtained from 1 provider group in 1 state, results may not be representative of other DPRP-reporting physicians or patients under their care.

Because there was no control group, a causal relationship between the DPRP and improvements in diabetes care cannot be proven.

Participating physicians were identified on a volunteer basis and may not be reflective of all physicians in the provider organization.

Due to stratification by A1C level, the sample size of subgroup 3 (A1C >9%) was relatively small; thus, caution should be taken when extrapolating results from this subgroup.

Still, this study identifies opportunities for improving diabetes care, including aggressive pharmacotherapy to control comorbidities that may increase cardiovascular risk. Using more than 1 class of antihyperglycemic medication to achieve A1C <7% in accordance with ADA treatment recommendations,

 using new medications with novel mechanisms of action, or targeting different physiologic causes of diabetes also may help reduce the number of patients not at A1C goal. Weight management also should be addressed. Excessive weight can hamper drug treatment efforts by increasing insulin resistance and glucose intolerance.16 Future initiatives should incorporate weight reduction in an effort to mitigate the negative metabolic consequences of excess weight and further improve outcomes.

As guidelines and recommendations are updated to accommodate changes in patient population and treatment modalities, it is important to measure care against these guidelines to validate quality, compare benchmarks with other organizations, and receive recognition by healthcare stakeholders such as employers. Attaining standardized measures is important not only to provide quality care but also to mitigate costs of care and provide financial resources to those responsible for attaining goals. These results reveal differences in control levels and treatment patterns in patients with varying levels of glycemic control and opportunities to improve diabetes care through goal attainment. Further research is needed to determine whether these results are reproducible in other physician groups and to identify whether specific measures of care correlate with levels of glycemic control.

Acknowledgment: The authors would like to thank Orsolya Lunacsek for her work with the data analysis. Author Affiliations: From The Medical Group of Ohio (BW, BH), Columbus; Xcenda (EC, JHJ), Palm Harbor, FL; and Novartis Pharmaceuticals Corporation (CAP, PAM), East Hanover, NJ. Dr Wall is now an independent consultant in Columbus, OH. Funding Source: Novartis Pharmaceuticals Corporation provided funding for this study. Author Disclosure: Drs Chiao and Jackson are employees of Xcenda, a consulting company that received payment for their involvement and preparation of this manuscript. Dr Plauschinat is an employee of Novartis Pharmaceuticals Corporation and reports owning stock in that company. Drs Wall, Miner, and Humphrey report no relationship or financial interest that would pose a conflict of interest with the subject of this article. Authorship Information: Concept and design (BW, CAP, PAM, JHJ); acquisition of data (BW, JHJ); analysis and interpretation of data (EC, JHJ); drafting of the manuscript (EC, CAP, JHJ, BH); critical revision of the manuscript for important intellectual content (BW, EC, CAP, JHJ, BH); provision of study materials or patients (BW, JHJ); obtaining funding (JHJ); administrative, technical, or logistic support (BW, PAM, BH); and supervision (BW, BH). Address correspondence to: Benjamin Humphrey, MD, The Medical Group of Ohio, 445 Hutchinson Ave, Ste 950, Columbus, OH 43235-5677. E-mail: bhumphrey@themgo.com.

1. Centers for Disease Control and Prevention. 

National diabetes fact sheet: general information and national estimates on diabetes in the United States, 2005.

 Atlanta, GA: Centers for Disease Control and Prevention, US Dept of Health and Human Services; 2005.

2. Narayan KM, Boyle JP, Geiss LS, Saaddine JB, Thompson TJ.

 Impact of recent increases in incidence on future diabetes burden: US. 2005-2050. 

Census product update: US population projected at 420 million in 2050. January 28, 2005.

 http://www.census.gov/mp/www/cpu/fact_of_the_day/004127.html. Accessed November 13, 2007.

 Economic costs of diabetes in the U.S. in 2002. 

Standards of medical care in diabetes— 2007.

6. Diabetes Control and Complications Trial Research Group.

 The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. 

7. UK Prospective Diabetes Study (UKPDS) Group.

 Intensive bloodglucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). [Published correction appears in 

8. UK Prospective Diabetes Study (UKPDS) Group.

 Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). [Published correction appears in Lancet. 1998;352(9139):1558]. 

9. Nathan DM, Cleary PA, Backlunc JY, et al; Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group.

 Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. 

10. National Institute of Diabetes and Digestive and Kidney Diseases.

National Diabetes Statistics fact sheet: general information and national estimates on diabetes in the United States, 2005.

 Bethesda, MD: National Institutes of Health, US Dept of Health and Human Services; 2005.

11. National Committee for Quality Assurance

http://www.ncqa.org/tabid/139/Default.aspx. Accessed November 21, 2008.

12. Chobanian AV, Bakris GL, Black HR, et al; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee.

 The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. [Published correction appears in 

13. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults

. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). 

Guidelines and performance measures for diabetes. 

15. Gaede P, Vedel P, Larsen N, Jensen G, Parving H, Pedersen O. 

Multifactorial intervention and cardiovascular disease in patients with Type 2 diabetes.

Obesity fact sheets. http://www. obesity.org/information/factsheets.asp. Accessed November 21, 2008.

Retrospective chart review to assess the impact of the Diabetes Physician Recognition Program showed that most patients achieved control levels recommended by national treatment guidelines.


Provider Organization Performance Assessment Utilizing Diabetes Physician Recognition Program

 To examine the effects of antihypertensive step therapy on prescription drug utilization and spending, and other medical care utilization and spending.

 Employers who had implemented step therapy were compared with employers who had not implemented step therapy. Data were drawn from the 2003 through 2006 MarketScan Research Databases. The study sample included employees and dependents who used antihypertensives (11,851 patients whose employer implemented a step-therapy protocol and 30,882 patients in the comparison group without step therapy). Multivariate generalized estimating equation models were used to estimate the immediate and timevarying effects of step therapy on medical and prescription drug spending and utilization, while controlling for important covariates and adjusting for clustering by patient.

 Results showed an initial 7.9% reduction in antihypertensive medication days supplied and an initial 3.1% reduction in medication costs among antihypertensive users in the step-therapy plans. However, these percentages grew in each subsequent quarter. Antihypertensive users in step-therapy programs also experienced an increase in inpatient admissions and emergency room visits. After an initial decline in spending, the step-therapy group incurred $99 more per user in quarterly expenditures than the comparison group.

 The intended effect of step therapy is to substitute cheaper and equivalently effective medications for more expensive medications. As this study demonstrates, step therapy may create barriers to receiving any medication, resulting in higher medical utilization and costs. Further research is needed to understand why these unintended consequences occur and how they might be avoided.

This study examined the effects of step therapy on prescription drug, inpatient, and emergency room utilization and spending among users of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers. Although step therapy led to medication cost savings, it also had unintended consequences:

The number of days supplied for antihypertensives declined. 

The number of people discontinuing therapy increased immediately after step therapy was implemented. 

Inpatient and emergency room admissions and costs were higher in step-therapy programs.

Managed care organizations and insurance plans are increasingly adopting step therapy in an effort to contain costs by steering patients away from more costly pharmacotherapies. Step therapy requires a member to try the first-line medication(s) within a drug class, often a generic alternative, prior to receiving coverage for a second-line agent, usually a branded product.

 Currently, most pharmacy benefit managers implement step therapy using “smart edit” logic and grandfathering those members who had obtained a prescription for the target (second-line) drug in the recent past. At the point of service, the smart edit reviews the member’s claims history (both electronically and in real time) for evidence of prior use of the first-line agent(s). If a claim is found, the system covers the second-line agent; otherwise, the claim is rejected. After claim rejection, members have the opportunity to have their prescriber change the prescription to the first-line agent or to submit a request for coverage of the secondline agent through a prior authorization.

There is a small but growing literature on step-therapy programs. In 2007 Yokoyama and colleagues evaluated hypertension-related pharmacy use and costs for 3 managed care plans that implemented an angiotensin receptor blocker (ARB) step-therapy intervention compared with 1 health plan with no ARB step-therapy intervention.

 The step-therapy intervention used a smart edit of patient pharmacy claim history in the preceding 3-month period. The ARB claim was rejected if there was no prior use of these drugs, in which case the pharmacist or patient had to contact the prescriber to obtain an alternative to the ARB or a prior authorization. The researchers found that within 12 months of follow-up, a step-therapy intervention for ARBs was associated with an 18% ratio of ARB users to the total number of patients using angiotensin-converting enzyme (ACE) inhibitors or ARBs compared with a 31% ratio in a health plan without the ARB step-therapy intervention. Of the patients who attempted to obtain an ARB and were rejected in the step-therapy group, 44.6% of patients went through the prior authorization process and received an ARB as initial therapy, 48.8% received other antihypertensive therapy, and 6.6% did not receive any antihypertensive therapy. Antihypertensive drug costs were about 13% lower for the ACE/ARB patients in the intervention group.

Motheral and colleagues examined the effect of prescription step-therapy programs in terms of plan-sponsor savings and member experience at the point of service.

 Specifically, they studied a 20,000-member plan that implemented 3 step-therapy programs in September 2002. The step-therapy programs covered proton pump inhibitors, selective serotonin reuptake inhibitors, and nonsteroidal anti-inflammatory drugs. Pharmacy claims from September 1, 2001, through June 30, 2003, were examined to compare changes in pharmacy per-member per-month (PMPM) net cost for the intervention group with changes in pharmacy PMPM net cost for a random sample of members from commercial plans without the step-therapy programs. A mailed, self-administered survey was sent to members with a step edit from September 1, 2002, through December 31, 2002. They found that the employer experienced a decrease of $0.83 PMPM in net cost after implementing step therapy, while the comparison group had an increase of $0.10 PMPM for these therapy classes.

Of the approximately 657 members who experienced a step-therapy edit and were mailed a survey, 33% returned the survey. Member-reported outcomes indicated that approximately 30% of patients received a generic after the step-therapy edit, 23% were granted a medical exception for the brand, and another 16% paid out-of-pocket for the brand medication. Nearly 17% received no medication, and another 10% received a sample or a nonprescription alternative. Compared with those who received first-line therapy, those who paid out-of-pocket for the brand medication and those who did not receive any medication were less likely to be satisfied with their pharmacy benefit.

Dunn and colleagues evaluated the impact on utilization and costs of a generic step-therapy program for antidepressant drugs in a health maintenance organization.

 In the program, brand-name antidepressants were covered only after a trial with a generic antidepressant, excluding tricyclic antidepressants. The authors found that the generic antidepressant dispensing rate increased by 20 percentage points (32.5%-52.5%) in the intervention group but only 7.4 points in the comparison group (24.9%-32.3%). Antidepressant costs per day of therapy in the intervention group decreased 11.7% versus a 2.7% decrease in the comparison group. The combination of change in drug costs and utilization resulted in a 13% decrease in antidepressant drug costs, from $4.16 PMPM in 2004 to $3.62 PMPM in 2005, versus a 7.6% decrease (from $3.57 to $3.30 PMPM) in the comparison group.

While prior research examined the effects of step therapy on prescription drug utilization and spending, this study focused on the effects of step therapy on all types of medical care utilization and spending. Specifically, the goal of this study was to examine the impact of a step-therapy program for ACE/ARBs on prescription drug and medical utilization and spending. Our estimates of the effects of step therapy are based on 2 ACE/ARB step-therapy programs implemented at different points in time, not a single program implementation. This strengthens the study design, providing an estimate of the common effects of step-therapy programs on utilization and spending.

The data source is the MarketScan Research Database from Thomson Reuters, representing the healthcare experience of enrollees in employer-sponsored health plans in the United States. Although the MarketScan Database in its entirety represents the healthcare experiences of more than 60 employers, we selected employers who had either (1) recently implemented step therapy (as a treatment group) or (2) could be verified as not having a step-therapy program in place (as a comparison group). Two firms had recently implemented steptherapy programs and were selected as the treatment group. As an additional requirement, the step-therapy program had to be implemented equally across all benefit plans available to an enrollee. To reduce the likelihood of selection bias, step therapy could not be selected or deselected by the individual enrollee. Two employers that did not have a step-therapy program in effect during the same time served as a control group. 

The step-therapy programs identified were implemented at different times, from 2003 through 2005, depending on the particular employer and prescription drug class. Exact dates of implementation are not disclosed to protect the identities of the employers.

This analysis focuses on step-therapy programs for antihypertensive medications because these are among the most commonly used medications. Two employers in the study implemented step therapy for antihypertensives, in particular ACE inhibitors and ARBs. The step-therapy program required certain (first-level or preferred) ACE inhibitors or ARBs to be used for a period of time (eg, 130 days) before using another (second-level) ACE inhibitor or ARB. Both step-therapy programs included the same antihypertensive medications for steps 1 and 2:

Step 1: benazepril, captopril, enalapril, fosinopril, lisinopril, moexipril, Quinaretic.

Step 2: Atacand, Avalide, Avapro, Benicar, Cozaar, Diovan, Hyzaar, Micardis, Teveten.

The sample was limited to employees and dependents of the selected employers (step-therapy and comparison group) who were continuously enrolled from 2003 through the third quarter of 2006, under 65 years of age, and not eligible for Medicare. An analytic file was created using a panel data (repeated measures) framework with 1 observation each quarter for each individual. Each enrollee in the analytic file had 15 observations (ie, quarters of data). The total sample comprised 269,561 enrollees. Of this group, 66,308 were enrolled in step-therapy plans and 203,253 were enrolled in comparison plans. We also focused on the subset of enrollees in each plan who were antihypertensive users (N = 42,733). Of these, 11,851 were in step-therapy plans and 30,882 were in comparison plans.

The dependent variables were prescription drug and medical costs and utilization. Six quarterly utilization measures were examined for antihypertensive users: the number of antihypertensive (ACE/ARB) days supplied (within the time period covered by the quarter), the antihypertensive (ACE/ARB) discontinuation rate (as defined by a gap of at least 60 days with no evidence of use of any drug in the class), the number of prescriptions filled, the number of emergency room visits, and the number of inpatient admissions. Five spending measures for antihypertensive users were examined: outpatient prescription drug, emergency room, inpatient medical (admissions), outpatient medical (nondrug), and total (medical and drug). Expenditures were defined as the total amount reimbursed to providers from all sources of payment including the health plan, the patient, and any third party (coordination of benefits amount). One utilization measure was examined for all enrollees: the antihypertensive (ACE/ARB) days supplied.

Statistical evaluations were performed using UNIX SAS version 9.01 (SAS Institute Inc, Cary, NC) and Stata 9.0 for Windows (StataCorp LP, College Station, TX). Chi-square and Student 

 tests were used to compare demographic and clinical characteristics as well as outcome variables between the step-therapy and comparison groups after step therapy had been implemented.

), where i is person and t is time in quarters.

The explanatory variables were defined as follows: Y: medical care utilization or expenditure measure. 

: a 0/1 dummy variable that equals 1 after the employer instituted step therapy and 0 otherwise. 

: a numeric counter measuring the number of quarters since step therapy began (ie, 1, 2, 3...). 

: a time trend variable to capture the linear utilization and spending trends common to both the step-therapy and comparison groups, with the first quarter of 2003 coded as 1 and the third quarter of 2006 coded as 15. Quadratic trends were captured with the Quarter

: demographic characteristics including age, sex, median income in the patient’s ZIP code of residence, and percentage of college graduates in the patient’s ZIP code of residence (from the US Census). 

: clinical characteristics including number of 

 (ICD-9-CM) codes in the prior year and Deyo-Charlson Comorbidity Index over the prior year.

Multivariate generalized estimating equation models were used to estimate the effects of step therapy on spending and utilization while controlling for important covariates and adjusting for clustering by patient.6 Utilization variables, representing counts of each type of service, were estimated with a negative binomial distribution and a log link. Discontinuation was estimated by using a binomial distribution and a logit link. Expenditures were estimated by using a gamma distribution with a log link.

The effects of step therapy are captured in the coefficients of the Step and 

 variables. The Step variable captures the immediate, and static, effects of step therapy, and the 

Step therapy with antihypertensives may lead to medication cost savings but may have unintended effects on other medical care utilization and spending.

February 2009

Reporting Hospitals" Antibiotic Timing in Pneumonia: Adverse Consequences for Patients?

Provider Organization Performance Assessment Utilizing Diabetes Physician Recognition Program

The Effects of Antihypertensive Step-Therapy Protocols on Pharmaceutical and Medical Utilization and Expenditures

Impact of 2 Employer-Sponsored Population Health Management Programs on Medical Care Cost and Utilization

Cost-Effectiveness of Long-Term Oxygen Therapy for Chronic Obstructive Pulmonary Disease

"No-Shows": Who Fails to Follow Up With Initial Behavioral Health Treatment?

A Blueprint for Pharmacy Benefit Managers to Increase Value