 To assess quality of depression care during the high-risk period following a psychiatric hospitalization.

 Retrospective administrative data analysis.

 Using Veterans Health Administration (VHA) administrative data, we assessed mental health follow-up within 7 and 30 days of psychiatric hospitalizations for major depression from 2004 to 2008. Adequate antidepressant medication coverage and number of psychotherapy visits were assessed within 90 days of discharge. Multivariable logistic regression was used to identify patient demographic and clinical characteristics associated with each quality indicator.

 Of the 45,587 patients discharged from a psychiatric inpatient stay with a diagnosis of major depressive disorder, 39.4% and 75.8% received an outpatient visit within 7 and 30 days of discharge, respectively; 58.7% of patients received adequate antidepressant coverage (72 of 90 days) and 12.9% received adequate psychotherapy encounters (8 visits). Receipt of outpatient mental health visits and of adequate psychotherapy were less likely among patients who were male, aged <35 or >65 years, had >3 major general medical comorbidities, lived >30 miles from a VHA clinic, or whose hospital length of stay was <7 days. Patients with comorbid substance use disorders were less likely to receive adequate antidepressant treatment.

 To optimize evidence-based depression care after a psychiatric hospitalization, health systems might increase receipt of psychotherapy by considering potential barriers related to age, medical condition, and distance. Patients with comorbid substance use disorders or their providers may need additional services to support antidepressant treatment.

	To enhance treatment of depression during the high-risk period following a psychiatric hospitalization, health systems should:

		Ensure timely access to outpatient mental health care and consider additional transition support, particularly for patients without mental health encounters prior to their psychiatric admission.

		Increase receipt of an adequate course of psychotherapy. Patients aged either <35 or >65 years, those with >3 serious medical comorbidities, and those residing >30 miles from a clinic were particularly less likely to receive psychotherapy.

		Increase antidepressant medication coverage for depressed patients with comorbid substance use disorders.

	Immediately following discharge from a psychiatric hospitalization, patients are at high risk for functional impairment, rehospitalization, and suicide.

 Suicide risk is high among depressed patients, and it is nearly 5 times higher during the 12 weeks after a psychiatric hospitalization compared with other time periods.

 To reduce risks of adverse outcomes, it is important for health systems to optimize the quality of depression care immediately following psychiatric hospitalization.

	Quality of post-discharge depression care encompasses several domains, including timeliness of follow-up with outpatient mental health providers and the delivery of an adequate course of antidepressant treatment or psychotherapy. To date, research studies of factors associated with postdischarge follow-up have not focused on care for patients with depression. Among general populations of psychiatric inpatients, nearly half of patients that schedule a post-hospitalization follow-up appointment do not show up for the first appointment, and slightly less than half have no follow-up within 30 days of discharge.

 Prior studies of depression-treatment indicators have focused on new and recurrent episodes of depression in predominantly outpatient settings, rather than during the high-risk, post-hospital period.

 One study of patients diagnosed with depression measured adherence to antidepressant medication regimens following a psychiatric hospitalization and found nearly half had inadequate adherence; however, the study excluded patients with no follow-up appointments and did not examine antidepressant treatment in the context of other quality indicators.

	In this study, we examined administrative data from the Veterans Health Administration (VHA) to determine the patient characteristics associated with the quality of post-hospital depression care across various quality indicators. The VHA provides comprehensive inpatient and outpatient specialty mental health care to eligible veterans and has an electronic medical record system to support communication between inpatient and outpatient services and across facilities. By describing factors affecting post-hospital care in a large integrated health system, study findings may inform both VHA and non-VHA efforts to improve post-hospital transition care.

	Patient data were obtained from the VHA’s National Registry for Depression, which includes comprehensive inpatient and outpatient records, including diagnoses, pharmacy records, and demographic information on all patients receiving treatment for depression within the VHA.

 Patients were included if they received a diagnosis of major depressive disorder (MDD;

 International Classification of Diseases, 9th Edition, Clinical Modification [ICD-9-CM]

 codes 296.3 or 296.2) during an inpatient psychiatric hospitalization between fiscal years 2004 and 2008. We excluded patients who had diagnoses of bipolar disorder, schizoaffective disorder, or schizophrenia, because these patients may require substantially different post-discharge care. The study was conducted following an institutional review board approval of database analyses through the Veterans Affairs Ann Arbor Healthcare System.

	For each patient, we identified the first inpatient psychiatric stay that included a diagnosis of MDD during the study period and used the discharge date from that hospitalization as the index discharge date. Adequacy of outpatient follow-up was determined using 2 indicators: whether patients received an outpatient visit with a mental health provider within 7 days and within 30 days of the index discharge date. These have been included as measures in the National Committee for Quality Assurance’s 

Healthcare Effectiveness Data and Information Set 

(HEDIS), which tracks performance on health-related quality measures among commercial, Medicare, and Medicaid health plans within the United States.

 Analysis of these measures was limited to patients who did not have another inpatient admission within 7 or 30 days, respectively.

	Adequacy of antidepressant treatment was determined by whether patients received at least a 72-day supply of an antidepressant during the initial 90 days following the index discharge date and refilled their medication at least once after their initial discharge prescription. A similar measure was shown to predict decreased odds of psychiatric hospitalization among VHA outpatients initiating treatment for depression.

 Coverage for 72 out of 90 days corresponds to a medication possession ratio of 0.80 (80%), a commonly used cut off for adequacy of medication receipt.

 The requirement of at least 1 refill was chosen so that a large initial prescription would not automatically satisfy the measure. Adequate psychotherapy was defined as the receipt of 8 individual or group psychotherapy visits (Current Procedural Technology [CPT] codes: 90804-90815, 90845, 90847, 90849, 90853, 90857) during the 90-day period following the index discharge date. These CPT codes include insight-oriented, behavior-modifying, supportive, family, and multi-family psychotherapy and psychoanalysis. A cut-off of 8 visits was used for adequate treatment based on manualized clinical trials of psychotherapy for depression and prior retrospective studies of psychotherapy adequacy.

 Analysis of receipt of adequate antidepressant treatment and adequate psychotherapy in the 90 days following psychiatric hospitalization was limited to patients who had <30 inpatient days during the follow-up period.

	Demographic characteristics included age (categorized as <35, 35-49, 50-64, 65-79, 80 years), gender, race (white, black, other, unknown), Hispanic ethnicity, and marital status (married or unmarried). The distance patients lived from their nearest VHA facility was approximated using a straightline distance from the population centroid of the patient’s ZIP code of residence (categorized as <30 miles, 30 to 60 miles, or >60 miles).

	Comorbid mental health conditions diagnosed during the index hospitalization were categorized as substance use disorder, post-traumatic stress disorder (PTSD), other anxiety disorders, or personality disorder. General medical comorbidity was measured using the Charlson Comorbidity Index

 and scores were categorized as 0 (least morbidity), 1-2, or 

	Length of stay during the index hospitalization was categorized as <7 days, 7 to 14 days, 15 to 30 days, or >30 days. We also created indicators of whether patients had mental health visits, received antidepressant treatment, or received psychotherapy in the 90 days before the index hospitalization.

	Continuous variables (eg, age, distance, inpatient length of stay) were categorized to facilitate interpretation of resulting odds ratios (ORs) and because prior studies have demonstrated non-linear relationships between patient characteristics, such as age and depression care.

	We calculated the frequencies of each quality indicator by each of the covariates. In bivariate analyses for each covariate, we used X

 tests to determine whether there were significant differences in receipt of the quality indicators.

	Multivariable logistic regressions were then used to predict receipt of each of the quality indicators adjusting for all covariates simultaneously. We also calculated Spearman rank correlation coefficients to assess the relationship between all possible pairs of covariates. Alpha was set at 0.05 for all comparisons. All analyses were conducted in SAS version 9.2 (SAS Institute Inc, Cary, North Carolina).

	The demographic and clinical characteristics of the 45,587 patients discharged from an inpatient psychiatry stay with a diagnosis of MDD between 2004 and 2008 are shown in 

. Patients had a mean age of 51.6 years; 89.2% were male; 72.6% were white, and 21.2% were black; and 5.5% were Hispanic. Comorbidity with other mental health conditions was common, with 46.5% having a comorbid substance use disorder, 33.7% having comorbid PTSD, and 11.7% having a comorbid anxiety disorder other than PTSD. Only 25.8% of patients had none of these other diagnoses during their hospital stay.

	Among all patients sampled, 39.4% received a mental health follow-up visit within 7 days of hospital discharge and 75.8% received a mental health follow-up visit within 30 days. Adequate antidepressant coverage was provided to 58.7% of patients; 12.9% of patients received adequate psychotherapy in the 90 days postdischarge. X

 tests demonstrated a statistically significant relationship between each of the covariate measures and at least 1 of the outcomes. We retained all covariates for the multivariate models and subsequently report only those results.

. In adjusted analyses, patients were less likely to receive a follow-up mental health appointment within 7 days if they were aged 

65 years (OR 0.77, 95% confidence interval [CI] 0.70-0.85) for age 65-79 vs <35 years); male (OR 0.91, 95% CI 0.85-0.97); married (OR 0.93, 95% CI 0.89-0.97); residing 

30 miles from a VHA clinic (OR 0.71, 95% CI 0.66-0.76 for 30-60 miles vs <30 miles); had 

1 major medical comorbidities (OR 0.78, 95% CI 0.70-0.86 for Charlson 3 or more vs 0); or received antidepressant treatment prior to hospitalization (OR 0.94, 95% CI 0.89- 0.98; 

	Patients were more likely to follow up within 7 days if they were aged 35 to 49 years versus <35 (OR 1.14, 95% CI 1.06-1.22 for age 35-49 vs <35); had a comorbid substance use disorder (OR 1.39, 95% CI 1.34-1.45); had comorbid PTSD (1.11, 95% CI 1.06-1.16); had a hospital length of stay for 

7 days (OR 1.26, 95% CI 1.18-1.34 for 15-30 days vs <7 days); had any outpatient mental health encounter before being hospitalized (OR 1.34, 95% CI 1.27-1.42); or had a psychotherapy encounter prior to being hospitalized (OR 1.57, 95% CI 1.49-1.66).

 Similar trends were seen for receiving a mental health follow-up within 30 days except that black, unknown, and other race were associated with decreased odds of follow-up (OR 0.89, 95% CI 0.84-0.94 for black vs white patients); patients who were aged 50 to 64 years (OR 1.48, 95% CI 1.37-1.60 vs age <35); married (OR 1.19, 95% CI 1.13-1.25); had a comorbid anxiety disorder other than PTSD; (OR 1.12, 95% CI 1.04-1.21) or had prior antidepressant use (OR 1.27, 95% CI 1.20-1.34) had increased odds of follow-up. Having a comorbid substance use disorder was no longer associated with follow-up.

Adequate Antidepressant Treatment Duration.

 Receipt of adequate antidepressant use was more likely among patients who were aged >35 years (OR 1.47, 95% CI 1.36-1.58 for age 35-49 vs <35) but <80 years (OR 1.12, 95% CI 0.97-1.28 for age 80 vs <35); were married (OR 1.33, 95% CI 1.27-1.39); had comorbid PTSD (OR 1.07, 95% CI 1.02-1.12) or other anxiety disorder (OR 1.25, 95% CI 1.17-1.34); lived 30 to 60 miles from a VHA clinic (OR 1.15, 95% CI 1.07-1.23 vs <30 miles); had a hospital length of stay >7 days (OR 1.34, 95% CI 1.28-1.41 for 7 to 14 days vs <7 days); or had antidepressant use (OR 3.08, 95% CI 2.93-3.23) or psychotherapy (OR 1.09, 95% CI 1.03-1.15) prior to hospitalization. Patients were less likely to receive adequate antidepressant treatment if they were male (OR 0.78, 95% CI 0.73-0.84), black (OR 0.56, 95% CI 0.53-0.59 vs white) or other or unknown race, Hispanic (OR 0.82, 95% CI 0.75-0.90); had a comorbid substance use disorder (OR 0.66, 95% CI 0.63-0.69) or personality disorder (OR 0.86, 95% CI 0.81-0.92); or had 

3 Charlson comorbidities (OR 0.85, 95% CI 0.77-0.93 vs Charlson of 0).

 Receipt of adequate psychotherapy was more likely among patients aged 35 to 64 years compared with younger or older patients (OR 1.37, 95% CI 1.23-1.53 for 35-49 vs <35). Blacks were not significantly different from whites in receipt of psychotherapy, and Hispanic patients were less likely to receive adequate psychotherapy (OR 0.79, 95% CI 0.69-0.90 vs non-Hispanic). Living 

30 miles from a VHA clinic (OR 0.56, 95% CI 0.50-0.64 for 30-60 miles compared with <30 miles); having 

3 Charlson comorbidities (OR 0.62, 95% CI 0.53-0.74 vs Charlson of 0); and having received an antidepressant prior to hospitalization (OR 0.89, 95% CI 0.83-0.95) were associated with decreased odds of receiving adequate psychotherapy. Having a comorbid substance use disorder (OR 1.47, 95% CI 1.38-1.56); PTSD (OR 1.39, 95% CI 1.31-1.48); length of hospital stay of 

7 days (OR 1.34, 95% CI 1.25-1.43 for 7-14 days vs <7 days); or psychotherapy prior to the hospitalization (OR 2.51, 95% CI 2.31-2.73) were all associated with increased odds of adequate psychotherapy after discharge.

 The largest correlations among the covariates were between Charlson comorbidity index and age (

 <.001), and being married and having a substance use disorder (

	Effective transition from inpatient to outpatient care is intended to sustain the process of recovery and to prevent relapse, readmission, or worse outcomes. The post-hospital transition may be particularly important for patients with MDD because it is the highest risk period for suicide. Per the HEDIS quality measure, we found that fewer than half of patients hospitalized for major depression had outpatient mental health follow-up within 7 days of discharge (39.4%), similar to the rate found in the general US population among Medicare (38.1%) and Medicaid (42.6%) beneficiaries in 2008.

 Follow-up within 30 days of discharge was substantially more common (75.8%). Unfortunately, follow-up within these time frames has not yet been shown to improve the outcomes of depressed patients. Follow-up within 30 days of discharge was not associated with reduced risk of suicide among psychiatric inpatients in a prior VHA study, but 7-day follow-up or the effect on longer-term outcomes of follow-up among patients with depression was not included in this prior study and should be a focus for subsequent work.

	Many patients received adequate psychopharmacologic treatment following a hospitalization for depression (58.7%) while relatively few patients received adequate psychotherapy post-discharge (12.9%), despite the fact that post-hospital needs likely remain high. The rate of adequate psychotherapy post-discharge is higher than that observed among VHA patients, who are primarily outpatients, newly diagnosed with depression (<5%), but it is lower than that observed among US community residents with depression or anxiety disorders (16%-23%).

 Such differences in receipt of care may reflect a general treatment bias toward pharmacologic treatment among providers or patients within the VHA, and, potentially, additional biases favoring pharmacologic treatment for patients with severe versus milder forms of depression. However, because 39% of patients received at least 1 psychotherapy visit within 90 days prior to their hospitalization, patients may be referred for psychotherapy after a hospitalization yet may have difficulty adhering to an adequate course of treatment in the post-hospital period. Although little information is available regarding the role of psychotherapy for depression following a psychiatric hospitalization, combining psychotherapy with pharmacologic treatment of depression is cost-effective and is particularly more effective than either modality alone for patients with severe or chronic depression.

 Extensive training in evidence-based psychotherapies has been made available to VHA providers; whether this improves access to psychotherapy for recently hospitalized patients is yet to be determined.

Articles

WEB EXCLUSIVE

Different patient characteristics predict adequate antidepressant treatment after hospitalization, received by 58.7% of patients, versus adequate psychotherapy, received by 12.9% of patients.

Depression Care Following Psychiatric Hospitalization in the Veterans Health Administration

To study risk factors for and likelihood of unplanned hospital admission and readmission in persons aged 

We used data from clinical records of the primary care centers, pharmacy database, and hospital discharge records for persons aged 

65 years registered in primary care centers of referral hospitals in the Baix Llobregat healthcare area. Unplanned hospital admission was defined as any unscheduled hospitalization in 2008; unplanned readmission was defined as any unscheduled admission occurring within 6 months after discharge date of index admission. Logistic regression analysis was used to identify predictors of unplanned hospital admission and readmission.

The population included 28,430 individuals. Among them, 2103 (7%) experienced an unplanned admission and 365 (1.3%) an unplanned readmission. The readmission rate for the admitted population was 18.7%. The strongest predictive factor of unplanned admission was >2 admissions in the previous 2 years (odds ratio [OR] 24.9, 95% confidence interval [CI] 16.0-38.7 for 2007; OR 15.6, 95% CI 8.6-28.0 for 2006). Factors associated with unplanned readmission were aged 

2 admissions during the previous year (OR 20.4, 95% CI 14.1-29.5). The area under the receiver operating characteristics curve was 0.78 for unplanned admission and 0.85 for unplanned readmission in the development sample and 0.76 and 0.81, respectively, in the validation sample.

 Aged persons and those who used more hospital services in previous years had a higher probability of hospital admission and readmission.

	Predictive models of people at high risk for hospital admission and readmission were developed using primary care, pharmacy, and hospital care databases.

		Aged people with respiratory and/or cardiac comorbidities and with greater use of hospital care services in previous years had a higher probability of hospital admission and readmission.

		 Application of this predictive model to identify people at risk may have a considerable impact on the care of patients with high comorbidity, making for a more proactive healthcare system and at the same time reducing the cost of avoidable hospitalizations.

	Persons of advanced age have a greater number of chronic diseases and more comorbidity than the remaining population. This population has a considerable economic impact on the healthcare system, particularly when there is a lack of coordination between primary care and hospital services. Patients with chronic diseases and high comorbidity may not always receive optimum care because sometimes they have not been identified as persons at risk or with special needs. Some studies have estimated that the population with greater needs represents around 10% of the total, but consumes about 70% of available healthcare resources.

 Unplanned hospital admissions and readmissions are included among the indicators of this higher healthcare expenditure. Short periods of time between admission and readmission (eg, 30 days) are used as indicators of quality of hospital care. Longer periods are useful as indicators of the population at risk and to asses the impact on the health system.

	Managed care models have been developed to promote integration and coordination of the healthcare system, and to provide patients with different interventions according to their level of need.

These models have been implemented by various organizations.

One of these models stratifies the population’s healthcare needs into 4 levels and provides patients with different interventions according to their level of need, while the other model focuses on the high-risk population and offers specific resources, particularly nursing case management, to achieve a more proactive approach in the care of high-risk patients. The aim of this strategy is to improve the patient’s clinical outcome and quality of life, while reducing the use of more costly services such as unplanned hospital admission. These models have also been applied in the National Health Service in England. Managed care models are based on identification of patients at risk of emergency hospitalization or high consumption of resources, or of requiring highcost care in the future.

	A systematic review published in 1991 identified 44 studies related to predictive models of hospital admission or readmission.

 Forty-six studies published between 1991 and 2010 that developed predictive models to identify persons at risk were retrieved; most of them cohort studies (n = 30). Eight standardized instruments

 designed to recognize the population at higher risk and to develop interventions to reduce unplanned admissions have been published.

	The healthcare system in Catalonia is a public system, financed by taxes, with universal healthcare coverage and free entry for all citizens. It is oriented toward primary care, which is the first contact and level of access users have with the healthcare system (cure, care, prevention, and health promotion), and it is mainly provided in primary care centers.

The hospitalization rate in Catalonia was 306.3/1000 for the population over age 64 years in 2009.

 A comparable figure of admission rate in acute, short-stay hospitals was 350.8/1000 for the United States in 2006.

	With the currently available information about the Spanish healthcare system, it would be appropriate to incorporate tools to help identify subpopulations at risk of high consumption of resources.

	The aims of this study were to analyze the risk factors for and the likelihood of unplanned hospital admission and readmission in the population over age 64 years who were registered with a primary care center in 1 of the geographical healthcare areas of Catalonia. This tool could potentially provide useful information to establish interventions focused on preventing future unplanned admissions and readmissions.

	This is a retrospective cohort study including all persons more than 64 years of age who were registered in the primary care area of Baix Llobregat Litoral within the health region of Costa de Ponent in Catalonia (Spain). The health services provided in this area include 7 primary care teams and 2 public referral hospitals, Viladecans Hospital, a first-level center, and Bellvitge Hospital, a third-level center for more complex cases. The primary care centers and hospitals provide almost all healthcare services in this area.

	The data were obtained from the e-CAP, the computerized information system used in primary care; the pharmacy database, which records dispensation of drugs; and the hospital discharge data from Viladecans and Bellvitge. Data used to build the predictive models came from the period encompassing 2006, 2007, and 2008. Data from 2006 and 2007 were used as background information on healthcare service use in both the primary care and hospital settings. To validate the model, data from the period including 2007, 2008, and 2009, were used, with data from 2007 and 2008 being the background information.

	The dependent variables were unplanned hospital admission between January 1, 2008, and December 31, 2008, and unplanned readmission within 6 months after discharge day of the index admission.

	The exclusion criteria for index admissions were elective or planned admission, readmissions during the same study period, admission of patients who resided outside the area, and hospital discharges without recording the admission or discharge date or the type of admission. In readmissions, the exclusion criteria were persons dead at discharge of index admission and individuals transferred from one hospital to another hospital in the same readmission.

	The sociodemographic factors analyzed were age (categorized into 65-69, 70-74, 75-79, and 

80 years), sex, place of residence, morbidity, and diagnoses recorded in the e-CAP, based on the 

International Classification of Diseases, 10th Revision, Clinical Modification

 into large groups, from which the more specific, common diagnoses according to the literature were chosen. These included insulin-dependent diabetes (E10), non—insulindependent diabetes (E11), heart failure (I50), ischemic heart disease (I20-I50), bronchitis (J40, J41, and J42), emphysema (J43), chronic obstructive pulmonary disease (COPD) (J44), asthma (J45), bronchiectasis (J47), Alzheimer’s disease (G30), and femoral neck fracture (S72), plus the total number of concurrent diagnoses (

5). In addition, the following factors were analyzed: consumption of primary care resources in the previous 2 years, consultations by telephone contact (0 vs 

4), and use of hospital resources in the previous 2 years, including previous unplanned admissions (

9). Categorization of these variables was carried out in keeping with data in the literature and according to their skewed distribution in the population studied.

	An analysis of the internal validity, quality, and consistency of the information obtained was carried out following a previously designed protocol.

	This study was performed in accordance with national and international guidelines (ethics code, Declaration of Helsinki) and with legal regulations on the confidentiality of personal data (Law 15/1999 of December 13 on the protection of personal data).

	Logistic regression was used to identify factors predictive of unplanned hospital admission in a 12-month period and readmission within 6 months. The model building process consisted of 3 steps: selecting the variables, building the model, and validating the model. Unplanned hospital admission and readmission models were adjusted for sex and age. Model fitting was realized with an iterative process, selecting in each step the candidate variable to add to the model, using a statistical significance level of <.01. To choose the best model, the likelihood ratio test was used to compare the models in each iteration.

	Validation of the model was performed by testing calibration, discrimination, and colinearity, and analyzing the residuals and influence values.

SPS version 18 (SPS Inc, Chicago, Illinois) and STATA version 11 (StataCorp, College Station, Texas) for Windows were used for the statistical analyses.

	A total of 28,430 persons over 64 years of age (57.6% women) were registered in a primary care center in the study area. Among them, 30.3% were 80 years or older, 35.7% were registered in Castelldefels, and 35.9% were registered in Viladecans, Spain.

	The sociodemographic characteristics of the entire study population, patients who had any hospital admission (planned or unplanned), and patients who had only an unplanned hospital admission in 2008, are presented in 

. Among the total population, 7.4% (n = 2103) had an unplanned admission; 51.7% of this group were women and 48% were 80 years or older (

). The most prevalent diagnoses related to unplanned admissions compared with the total population were COPD, heart failure, ischemic heart disease, non—insulin-dependent diabetes, and having 5 or more concurrent diagnoses (

). Compared with the total population, hospitalized patients in this group consumed an increased amount of healthcare resources, and this increase was higher in 2007 than in 2006.

	Among the total population, 1.3% (n = 365) had an unplanned readmission within 180 days from the index admission (53.7% men, 53.7% 80 years or older), and 4.1% had any readmission (unplanned or planned; 

). In the admitted population the rate of readmission was 18.7% (365/1952) at 6 months and 13.6% (265/1952) at 30 days. Compared with the total population, readmitted patients had a higher prevalence of almost all the variables analyzed. Emergency admission in the 2 years prior to readmission was also higher in the readmitted population (

	The predictive factors for unplanned hospital admission were sex (male), age (the older the patient, the higher the risk), COPD, heart failure, 5 or more concurrent diagnoses, dispensation of 4 or more medications, and the use of hospital resources in the 2 years prior to admission (emergency admissions, planned admissions, and cumulative days of hospital stay). Associated factors for unplanned readmission were sex (male), age (the older the patient, the higher the risk), insulin- dependent diabetes, ischemic heart disease, heart failure, the number of dispensed drugs, and emergency admissions and days of hospital stay accumulated in the previous year. The most powerful predictive factor for both admission and readmission was unplanned admissions in the previous year (odds ratio [OR] 24.9, 95% confidence interval [CI] 16.0-38.7 and OR 20.4, 95% CI 14.1-29.5, respectively; 

	The area under the receiver operating characteristic (ROC) curve of hospital admission was 0.78 in the study sample and 0.76 in the validation sample (

). The variance influence factors were 1.03 to 1.24 and the Hosmer-Lemeshow 

 value was .50. For readmission, ROC curves were 0.85 in the study sample and 0.81 in the validation sample (Figure). The variance influence factors were 1.00 to 1.14, and the Hosmer-Lemeshow 

	In this study, predictive models for unplanned hospital admission and readmission were developed in a health area of Catalonia using retrospective data from the primary care, pharmacy, and hospital databases. Hospitalization in previous years was the most powerful risk factor for unplanned hospitalization and rehospitalization. These predictive models enable identification of persons at higher risk and establishment of proactive interventions with greater intensity in the highrisk categories, so that more effective allocation of resources to meet population needs can be achieved.

	The fact that the Spanish healthcare system is oriented toward primary care, with easy and universal access to care, could be associated with low rates of hospitalizations, both the total rates and specifically the rates for ambulatory care—sensitive conditions.

 Nevertheless, factors related to th use of primary care did not achieve statistical significance in our predictive models. As almost all the elderly population had consulted regularly with primary care in previous years (mean number of visits to a primary care center per year in the study sample was 10.8 in 2008), these variables lost their ability to discriminate between those who were admitted and those who were not admitted to the hospital. Moreover, in spite of the high mean number of visits to a primary care center in the Spanish elderly population compared with the number of visits in other countries, the present study did not include assessment of the “quality” or effectiveness of those visits.

	No differences were found in the probability of admission or readmission related to the place of residence, despite the existing variability in clinical practice in Spain.

 This may be because the area analyzed was relatively small, including only 2 hospitals and 7 primary care teams.

	The rates of unplanned readmission for the admitted population in our study were 18.7% at 6 months and 13.6% at 30 days, both lower than the 20.0% generally found in the United States at 30 days.

	Previous studies in Spain have analyzed admission and/or readmission for specific diseases or specialties. In Catalonia (Baix Empordà) the relationship between population morbidity and healthcare costs was examined.

 In Valencia, relationships between comorbidity and hospital stay, mortality, and readmission were analyzed using the Charlson index.

The results from the current study cannot be compared with the results of these studies because we examined the risk of unplanned hospitalization using various databases, not using only hospital data. In addition, the study from Baix Empordà analyzed primary care and hospital cost together as a dependent variable.

Predictive models for hospital admissions and readmissions in persons aged >65 years have been developed through a combination of comorbidity and previous healthcare use.

Predictive Model for Emergency Hospital Admission and 6-Month Readmission

 To develop a cost-offset model from a US payer perspective comparing glaucomatous progression and costs among primary open-angle glaucoma (POAG) patients using bimatoprost, latanoprost, or travoprost.

 A Markov cohort model was used to estimate glaucomatous progression for POAG patients over 7 years. The model assumed bimatoprost-treated patients had lower resulting intraocular pressure (IOP) (by 1 mm Hg) for all presenting IOP categories than latanoprost- or travoprost-treated patients. Patients with lower IOP were assumed to have lower probability of progression. Those that progressed were assumed to do so at a rate of —0.6 dB per year. Direct costs associated with mean deviation score categories were applied to each treatment cohort to calculate the expected 7-year costs of treating patients with each prostaglandin analogue (PGA). Literature was used to support assumptions. A budget impact analysis was conducted where all travoprost patients switched to generic latanoprost and where all bimatoprost patients switched to generic latanoprost. The base case market share was 22% bimatoprost, 23% travoprost, and 55% latanoprost.

 Model results demonstrate that for a managed care plan with 9500 PGA-treated glaucoma patients, exclusive bimatoprost use would prevent progression in 136 additional individuals compared with exclusive travoprost or latanoprost treatment. Model results demonstrate that greater IOP reduction from bimatoprost is associated with increased cost savings compared with latanoprost or travoprost treatments.

 Model results demonstrate that greater IOP reduction from bimatoprost could reduce managed care spending.

	Greater intraocular pressure (IOP) reduction with bimatoprost is associated with cost savings compared with treatment with latanoprost or travoprost.

		In a health plan with 9500 prostaglandin analogue—treated glaucoma patients, exclusive bimatoprost use would prevent progression in an estimated 136 incremental patients over 7 years compared with exclusive travoprost or latanoprost treatment.

		Exclusive treatment with bimatoprost would generate cost savings from reduced resource utilization estimated at $4009 per avoided progression of early glaucoma and $4543 per avoided progression of advanced glaucoma over 7 years.

		Cost benefits of lower IOP provide incentive for health system—level policies to promote bimatoprost use.

	More than 2 million people in the United States 40 years and older have glaucoma.

 Primary open-angle glaucoma (POAG) is the most common form of the disease. While there are numerous risk factors associated with glaucoma progression,

 intraocular pressure (IOP) is one of the only known modifiable risk factors that can be controlled to prevent progressive optic neuropathy.

	The incidence of glaucoma is expected to rise with the continued growth of the elderly population and is a leading cause of preventable blindness in the United States.

 Glaucoma costs the healthcare system an estimated $2.9 billion annually in direct medical expenditures.

 An analysis using a Markov model to replicate health events over the remaining lifetime of someone newly diagnosed with glaucoma estimated that the cost of care for people with POAG was higher than that of people without the disease by $1688 over their expected lifetime, or approximately $137 per year.

 Lee et al estimated the average direct cost of treatment for glaucoma to range from $523 per patient per year to $2125 per patient per year depending upon the stage of disease (early vs severe).

 Given the burden of glaucoma on the healthcare system, there is a financial incentive to delay or prevent glaucomatous progression and associated visual field loss.

	Multiple medical therapies that are intended to lower IOP are available, the most common of which are prostaglandin analogues (PGAs).

 and are not associated with the significant systemic side effects associated with topical beta blockers, such as bronchospasm, shortness of breath, depression, fatigue, impotence, hair loss, heart failure, and bradycardia.

	Latanoprost, travoprost, and bimatoprost are once-daily PGAs that lower IOP effectively and continuously, allowing for flat diurnal curves with few systemic side effects.

 Numerous randomized clinical trials have demonstrated that patients achieve greater IOP lowering with bimatoprost than either latanoprost or travoprost.

 Furthermore, a recent meta-analysis of multiple studies concluded that bimatoprost has greater efficacy than latanoprost or travoprost.

 reported on the latanoprost to bimatoprost switching patterns of an HMO. The study found that the HMO achieved a high switch rate, with few patients switching back. Moreover, there was a statistically significant reduction in mean IOP for patients who switched from latanoprost to bimatoprost. Similar studies that examined high switch rates from latanoprost to travoprost did not demonstrate a statistically significant reduction in mean IOP.

 The impact of this superior efficacy of bimatoprost on direct medical costs has not been evaluated to date.

	The primary objective of this study was to develop a cost-offset model from a payer perspective to compare glaucomatous progression and costs among POAG patients treated with bimatoprost, latanoprost, or travoprost in the United States. The analysis aimed to support the hypothesis that a reduction in IOP leads to a decreased risk of glaucoma progression, which in turn leads to cost savings from less healthcare resource utilization; specifically, due to bimatoprost’s greater efficacy in IOP reduction compared with travoprost and latanoprost, its use should lead to cost savings for payers. Fewer advanced glaucoma patients would result in lower expenses as increased resources needed for advanced disease stages are avoided.

) was constructed (1) to evaluate the difference in glaucomatous progression rates with bimatoprost, travoprost, or latanoprost; and (2) to provide a cost-offset analysis for these PGAs. This analysis was driven by the link between the treatments’ clinical efficacy, as represented by IOP after treatment (called the resultant IOP) and resource utilization. The model accounted for various healthcare resources (detailed below), the use of which increased with the resultant IOP. As the model assumed that IOP increase is associated with increased glaucomatous progression risk (detailed below), this analysis demonstrated the cost savings that arise from decreasing resource utilization due to the prevention of disease progression. While a cost-offset analysis provides useful insight into whether an intervention is worth financing from a budgetary standpoint, it is based on efficacy and resource utilization assumptions that may differ from real world rates and practices.

	The model calculated results over a 7-year period in a hypothetical health plan of 1,000,000 members. Given the slow-progressing nature of glaucoma, a 7-year time horizon was utilized, as in previous studies.

 Progression was defined as the loss of visual field as it is measured by the mean deviation (MD) score. POAG prevalence was assumed to be 1.9% (ie, 19,000 plan members had glaucoma).

 the PGA treatment rate was estimated at 50% (N = 9500).

	A subsequent budget impact analysis assumed bimatoprost as a first-line treatment with PGA market share as follows: 22% bimatoprost, 23% travoprost, and 55% latanoprost.

 This budget impact analysis compared the total cost to the payer, including the cost of the PGAs, medical costs (such as physician visits), and other glaucoma medication costs between different market share scenarios. In general, these budget impact analyses are useful for estimating systemwide (eg, medical and pharmacy) budget impact; they are often used by managed care organizations

 to justify therapy reimbursement. While these analyses are limited in that they do not account for the full impact of clinical and patient outcomes and therefore do not establish the overall value of the product, they are not intended to do so; budget impact analyses are intended to present a financial perspective when considering a particular therapeutic decision.

	Two scenarios were explored: In scenario A, all bimatoprost patients were switched to generic latanoprost, resulting in a market share of 0% bimatoprost, 23% travoprost, and 77% latanoprost. In scenario B, all travoprost patients were switched to generic latanoprost; market shares were 22% bimatoprost, 0% travoprost, and 78% latanoprost.

	This model accounted for 3 different payer situations: all Medicare Part D patients, all commercial patients, and a mix of Medicare Part D and commercial patients. Medicare Part D patients included the assumed number of individuals covered by the plan who were 65 years or older. Commercial patients included the assumed number of individuals on the plan who were younger than 65 years.

	Intraocular pressure ranges 19-21, 22, 23, 24, 25, and 26 (all mm Hg) and the percentage of patients allocated to them were based on a population-based study (the Beaver Dam Eye Study).

 The resultant IOP with bimatoprost treatment was calculated by reducing the presenting IOP by 27%.

 As the model assumes a 1 mm Hg reduction advantage for bimatoprost over latanoprost or travoprost,

 the resultant IOPs for the latter PGAs were calculated by adding 1 mm Hg to the resultant IOP scores calculated for bimatoprost. This resultant IOP value was the basis of the patients’ probability of progression (discussed in the Progression Rate section).

	As supported by literature, a baseline mean deviation (MD) score of —4 decibels

 (dB) for early glaucoma patients or —10 dB

 for advanced glaucoma patients was assigned to the “worse seeing eye” of each patient cohort.

 Baseline MD score for the “better seeing eye” was estimated based upon a correlation with the worse seeing eye using the following equation:

	BetterEYEMD = (WorseEYEMD —1.83) / 1.307.

	The annual probability of progression was estimated using the resultant IOP following PGA treatment. It was estimated from literature that a 1 mm Hg decrease in IOP would reduce the probability of glaucoma progression by 10%

 was the benchmark. Based on the Early Manifest Glaucoma Trial data, the probability that a patient with an IOP of 21 mm Hg would progress was 10.2%.

 Therefore, the calculation for progression probability for IOP values other than 21 mm Hg was estimated using the following equation:

	After PGA treatment, the probability of progression remained constant throughout the model. Based on data from the Early Manifest Glaucoma Trial,

 patients who progressed did so at a mean rate of —0.6 dB/year (Table 1).

	The model also considered the impact of IOP fluctuation on probability of progression. The percentage of patients with IOP fluctuation greater than 3 mm Hg was 7.8% for latanoprost and 2.5% for bimatoprost.

 The percentage of patients on travoprost with high fluctuation was assumed to be similar to or greater than the percentage of those on latanoprost

; therefore, the percentage of travoprost patients with IOP fluctuation greater than 3 mm Hg defaulted to 7.8%. Inclusion of IOP fluctuation in the model was supported by a literature review that demonstrated that individuals who had an IOP fluctuation greater than 2 standard deviations would be 3 times more likely to experience glaucomatous progression.

 The risk ratio (progression due to IOP fluctuation) was calculated as follows: 30% (probability of progression among patients with IOP fluctuation of >2 mm Hg)

 / 9.7% (probability of progression among patients with IOP fluctuation of 

	The annual per-patient PGA cost was calculated based on drug cost (average wholesale price [AWP] or Medicare), dispensing fees ($2 for each brand name drug and $3 for generic drugs), copays,

 rebates, and pharmacy discounts (which were 15% of the monthly cost of branded products and 17% of the monthly cost of generic products; Table 1). Rebates and copays were subtracted from the price to get the net cost to the managed care payer. As there were no data for the generic price of latanoprost, it was assumed that the cost of generic latanoprost would be 80% of the branded price. A lower copay for the generic ($10 instead of $25)

 was also assumed. The number of office visits and visual field tests were predicted from the MD score and IOP value.

 The annual number of concomitant glaucoma medications (non-PGA) was predicted from the MD score (Y[number of glaucoma medications] = 0.70 — 0.0455 [MD score]), as derived from literature on the number of medications by stage.

 Surgery was applied as a per-eye cost and the number of surgeries per year was estimated using literature on the annual rate of surgery by stage.

 For nonprogressors, the annual rate of surgery was assumed to be the literature-based rate for the stage of disease corresponding to the MD score in the model. For progressors, this annual rate was assumed to be higher based on literature showing that those who had disease stage progression were more likely to receive surgery (rate ratio of 1.92).

 The model conservatively assumed that patients who underwent surgery discontinued their medication and were therefore dropped out of the cohort. Direct medical and non-PGA pharmacy costs, which were based on resource utilization estimates, were applied to each treatment cohort to calculate the expected 7-year costs (

). Unit costs for office visits, visual field tests, additional non-PGA glaucoma medications, and surgeries were based on resource utilization and 2008 Current Procedural Terminology codes (Table 2).

 A discount rate of 3% for future costs was applied.

 Results were presented as total annual costs and annual per member per month (PMPM) costs.

	Two 1-way sensitivity analyses were conducted for the estimated reduction in risk of progression per 1 mm Hg change in IOP. The base case assumed that there is a 10% reduction in the risk of progression per 1 mm Hg decrease in IOP. The sensitivity analyses assumed that this reduction in risk is 19%. This parameter change was tested for both less severe and more severe glaucoma patients.

	Four 1-way sensitivity analyses were conducted for the price of generic latanoprost. Since the analysis was originally conducted, generic latanoprost has entered the market. Therefore, the sensitivity analyses assessed how the results would change should the price of generic latanoprost be as low as $16.44

 These results were calculated for patients with less severe and more severe disease.

	This study is a health-related outcomes analysis based on available peer-reviewed literature and therefore was not subject to approval by an institutional review board.

Long-Term Progression Rates: Base Case for Early and Advanced Glaucoma Patients

Treatment of glaucoma with bimatoprost is associated with cost savings compared with treatment with latanoprost or travoprost because of greater intraocular pressure reduction.

Cost-Offset Analysis: Bimatoprost Versus Other Prostaglandin Analogues in Open-Angle Glaucoma

 While the 2011 implementation of “meaningful use” legislation for certified electronic health records (EHRs) promises to change quality reporting by overcoming data capture issues affecting quality measurement, the magnitude of this effect is unclear. We compared the measured quality of laboratory monitoring of Healthcare Effectiveness Data and Information Set (HEDIS) medications based on specifications that (1) include and exclude patients hospitalized in the measurement year and (2) use physician test orders and patient test completion.

 Among patients 18 years and older in a large multispecialty group practice utilizing a fully implemented EHR between January 1, 2008, and July 31, 2008, we measured the prevalence of ordering and completion of laboratory tests monitoring HEDIS medications (cardiovascular drugs [angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, digoxin, and diuretics] and anticonvulsants [carbamazepine, phenobarbital, phenytoin, and valproic acid]).

 Measures excluding hospitalized patients were not statistically significantly different from measures including hospitalized patients, except for digoxin, but this difference was not clinically significant. The prevalence of appropriate monitoring based on test orders typically captured in the EHR was statistically significantly higher than the prevalence based on claims-based test completions for cardiovascular drugs.

 HEDIS quality metrics based on data typically collected from claims undermeasured quality of medication monitoring compared to EHR data. The HEDIS optional specification excluding hospitalized patients from the monitoring measure does not have a significant impact on reported quality. Integration of EHR data into quality measurement may significantly change some organizations’ reported quality of care.

	Integration of electronic health record (EHR) data into quality measurement may significantly change some organizations’ reported quality of care.

		For example, measuring laboratory monitoring quality from claims data (test completion) underestimates care quality compared with measurements based on EHR data (physician test ordering).

		If quality-of-care measurements improve concurrent with “meaningful use” implementation, it may be difficult to discern whether this change reflects true quality improvements or measurement changes.

Shifting from claims to integrated electronic health records to calculate quality metrics will improve reported quality attributable to data capture changes, not true quality improvements.

September 2011

Depression Care Following Psychiatric Hospitalization in the Veterans Health Administration

Predictive Model for Emergency Hospital Admission and 6-Month Readmission

Cost-Offset Analysis: Bimatoprost Versus Other Prostaglandin Analogues in Open-Angle Glaucoma

Quality Measurement of Medication Monitoring in the Meaningful Use Era

Disparities in Antidepressant Adherence in Primary Care: Report From Israel

Potential Role of Comanagement in Rescue of Surgical Patients

Electronic Health Records, Clinical Decision Support, and Blood Pressure Control

Compliance, Persistence, and Switching Patterns for ACE Inhibitors and ARBs

Mental Illness and Warfarin Use in Atrial Fibrillation

Inappropriate Antibiotic Prescribing in Managed Care Subjects With Influenza