Pragmatic clinical trials embedded in healthcare systems: generalizable lessons from the NIH Collaboratory

BACKGROUND

The clinical research enterprise is not producing the evidence decision makers arguably need in a timely and cost effective manner; research currently involves the use of labor-intensive parallel systems that are separate from clinical care. The emergence of pragmatic clinical trials (PCTs) poses a possible solution: these large-scale trials are embedded within routine clinical care and often involve cluster randomization of hospitals, clinics, primary care providers, etc. Interventions can be implemented by health system personnel through usual communication channels and quality improvement infrastructure, and data collected as part of routine clinical care. However, experience with these trials is nascent and best practices regarding design operational, analytic, and reporting methodologies are undeveloped.

METHODS

To strengthen the national capacity to implement cost-effective, large-scale PCTs, the Common Fund of the National Institutes of Health created the Health Care Systems Research Collaboratory (Collaboratory) to support the design, execution, and dissemination of a series of demonstration projects using a pragmatic research design.

RESULTS

In this article, we will describe the Collaboratory, highlight some of the challenges encountered and solutions developed thus far, and discuss remaining barriers and opportunities for large-scale evidence generation using PCTs.

CONCLUSION

A planning phase is critical, and even with careful planning, new challenges arise during execution; comparisons between arms can be complicated by unanticipated changes. Early and ongoing engagement with both health care system leaders and front-line clinicians is critical for success. There is also marked uncertainty when applying existing ethical and regulatory frameworks to PCTS, and using existing electronic health records for data capture adds complexity.

Major Physical Health Conditions and Risk of Suicide

INTRODUCTION

Most individuals make healthcare visits before suicide, but many do not have a diagnosed mental health condition. This study seeks to investigate suicide risk among patients with a range of physical health conditions in a U.S. general population sample and whether risk persists after adjustment for mental health and substance use diagnoses.

METHODS

This study included 2,674 individuals who died by suicide between 2000 and 2013 along with 267,400 controls matched on year and location in a case-control study conducted in 2016 across eight Mental Health Research Network healthcare systems. A total of 19 physical health conditions were identified using diagnostic codes within the healthcare systems' Virtual Data Warehouse, including electronic health record and insurance claims data, during the year before index date.

RESULTS

Seventeen physical health conditions were associated with increased suicide risk after adjustment for age and sex (p<0.001); nine associations persisted after additional adjustment for mental health and substance use diagnoses. Three conditions had a more than twofold increased suicide risk: traumatic brain injury (AOR=8.80, p<0.001); sleep disorders; and HIV/AIDS. Multimorbidity was present in 38% of cases versus 15.5% of controls, and represented nearly a twofold increased risk for suicide.

CONCLUSIONS

Although several individual conditions, for example, traumatic brain injury, were associated with high risk of suicide, nearly all physical health conditions increased suicide risk, even after adjustment for potential confounders. In addition, having multiple physical health conditions increased suicide risk substantially. These data support suicide prevention based on the overall burden of physical health.

Between-visit changes in suicidal ideation and risk of subsequent suicide attempt

BACKGROUND

While clinicians are expected to routinely assess and address suicide risk, existing data provide little guidance regarding the significance of visit-to-visit changes in suicidal ideation.

METHODS

Electronic health records from four large healthcare systems identified patients completing the Patient Health Questionnaire or PHQ9 at outpatient visits. For patients completing two questionnaires within 90 days, health system records and state vital records were used to identify nonfatal and fatal suicide attempts. Analyses examined how changes in PHQ9 item 9 responses between visits predicted suicide attempt or suicide death over 90 days following the second visit.

RESULTS

Analyses included 430,701 pairs of item 9 responses for 118,696 patients. Among patients reporting thoughts of death or self-harm "nearly every day" at the first visit, risk of suicide attempt after the second visit ranged from approximately 2.0% among those reporting continued thoughts "nearly every day" down to 0.5% among those reporting a decrease to "not at all." Among those reporting thoughts of death or self-harm "not at all" at the first visit, risk of suicide attempt following the second visit ranged from approximately 0.2% among those continuing to report such thoughts "not at all" up to 1.2% among those reporting an increase to "nearly every day".

CONCLUSIONS

Resolution of suicidal ideation between visits does imply a clinically important reduction in short-term risk, but prior suicidal ideation still implies significant residual risk. Onset of suicidal ideation between visits does not imply any special elevation compared to ongoing suicidal ideation. Risk is actually highest for patients repeatedly reporting thoughts of death or self-harm.

First Presentation With Psychotic Symptoms in a Population-Based Sample

OBJECTIVE

Increasing evidence supports the effectiveness of comprehensive early intervention at first onset of psychotic symptoms. Implementation of early intervention programs will require population-based data on overall incidence of psychotic symptoms and on care settings of first presentation.

METHODS

In five large health care systems, electronic health records data were used to identify all first occurrences of psychosis diagnoses among persons ages 15-59 between January 1, 2007, and December 31, 2013 (N=37,843). For a random sample of these putative cases (N=1,337), review of full-text medical records confirmed clinician documentation of psychotic symptoms and excluded those with documented prior diagnosis of or treatment for psychosis. Initial incidence rates (based on putative cases) and confirmation rates (from record reviews) were used to estimate true incidence according to age and setting of initial presentation.

RESULTS

Annual incidence estimates based on putative cases were 126 per 100,000 among those ages 15-29 and 107 per 100,000 among those ages 30-59. Rates of chart review confirmation ranged from 84% among those ages 15-29 diagnosed in emergency department or inpatient mental health settings to 19% among those ages 30-59 diagnosed in general medical outpatient settings. Estimated true incidence rates were 86 per 100,000 per year among those ages 15-29 and 46 per 100,000 among those ages 30-59.

CONCLUSIONS

When all care settings were included, incidence of first-onset psychotic symptoms was higher than previous estimates based on surveys or inpatient data. Early intervention programs must accommodate frequent presentation after age 30 and presentation in outpatient settings, including primary care.

Assessing cow-calf welfare. Part 1: Benchmarking beef cow health and behavior, handling; and management, facilities, and producer perspectives

Assessment programs are one way beef producers communicate information about animal welfare to retailers and the public. Programs that monitor cattle through the production cycle (e.g., the Global Animal Partnership) or at individual stages (e.g., slaughter; the North American Meat Institute) exist, but to date, there is no assessment program addressing welfare specifically in the cow-calf sector. The objectives of this study were to measure cow-calf health and handling welfare outcomes and gather management, facility, and producer perspective information to 1) describe current practices and 2) inform assessment design. A welfare assessment, designed using features of similar beef and dairy programs, was conducted on 30 California ranches that varied in size (mean 1,051 cows [SD 1,849], range 28 to 10,000 cows) and location within the state. Cattle health and behavior and stockperson handling were measured during a routine procedure (e.g., pregnancy checks) on breeding females ( = 3,065). Management and producer perspectives were evaluated through an interview, and facility features were recorded at the chute and water access points. Cattle health problems were rare and seen only on specific ranches (e.g., prevalence of lame cattle: mean 1.3% [SD 1.5], range 0 to 7.1%). Cattle behavior and stockperson handling varied between ranches (e.g., cattle balking: mean 22.0% [SD 21.9], range 1.6 to 78.3%; electric prod use: mean 23.5 [SD 21.5], range 0 to 73.0%). Although some management and facility characteristics were shared by most (e.g., all ranches castrated bull calves; 86% used alleyways with an anti-back gate), other aspects varied (e.g., weaning age: mean 8.2 mo [SD 1.4], range 6 to 11 mo; 43% used shade cover over chute). Most producers shared similar perspectives toward their herd health management plan, but their responses varied when asked to evaluate an animal's pain experience. In terms of assessment design, there were challenges with feasibility (e.g., scheduling a ranch visit on a day cattle were processed was difficult), validity (e.g., cattle may back up calmly to adjust posture or quickly in response to an aggressive handler; without this context, the welfare implications of this behavior are unclear), and comparability (e.g., an explicit animal observation period needed to be defined to make comparisons across ranches). Future assessment programs should consider these qualities when selecting measures to evaluate welfare.

A pragmatic randomized clinical trial of behavioral activation for depressed pregnant women

OBJECTIVE

Depression among pregnant women is a prevalent public health problem associated with poor maternal and offspring development. Behavioral activation (BA) is a scalable intervention aligned with pregnant women's preference for nonpharmacological depression care. This is the first test of the effectiveness of BA for depression among pregnant women, which aimed to evaluate the effectiveness of BA as compared with treatment as usual (TAU).

METHOD

Pregnant women (mean age = 28.75 years; SD = 5.67) with depression symptoms were randomly assigned to BA (n = 86) or TAU (n = 77). Exclusion criteria included known bipolar or psychotic disorder or immediate self-harm risk. Follow-up assessment occurred 5 and 10 weeks postrandomization and 3 months postpartum using self-report measures of primary and secondary outcomes and putative targets.

RESULTS

Compared with TAU, BA was associated with significantly lower depressive symptoms (d = 0.34, p = .04) and higher remission (56.3% vs. 30.3%, p = .003). BA also demonstrated significant advantage on anxiety and perceived stress. Participants attended most BA sessions and reported high satisfaction. Participants in BA reported significantly higher levels of activation (d = 0.69, p < .0002) and environmental reward (d = 0.54, p < .003) than those who received TAU, and early change in both of these putative targets significantly mediated subsequent depression outcomes.

CONCLUSIONS

BA is effective for pregnant women, offering significant depression, anxiety, and stress benefits, with mediation analyses supporting the importance of putative targets of activation and environmental reward. (PsycINFO Database Record

Analysis of depression trajectory patterns using collaborative learning

BACKGROUND

Depression is a common, complex, and dynamic mental disorder. Mitigating depression has become a national health priority as it affects 1 out of 10 American adults and is the most common mental illness seen in primary care. The emerging use of electronic health record (EHR) provides an unprecedented information infrastructure to understand depression trajectories.

OBJECTIVE

We aim to effectively analyze patterns in the collected depression trajectories of a treatment population and compare several methods to predict individual trajectories for monitoring treatment outcomes.

METHODS

Our data includes longitudinal Patient Health Questionnaire (PHQ)-9 scores over 4 years for assessing depression severity from the Mental Health Research Network. We analyzed > 3,000 patients with at least six PHQ-9 observations who have ongoing treatment. We used smoothing splines to model individual depression trajectories. We then used K-means clustering and collaborative modeling (CM) to identify subgroup patterns. We further predicted the individuals' PHQ-9 scores based on depression trajectories learnt from individual growth model (IGM), mixed effect model (MEM), CM, and similarity-based CM (SCM), and compared their predictive performances.

RESULTS

We found five broad trajectory patterns in the ongoing treatment population: stable high, stable low, fluctuating moderate, an increasing and a decreasing group. For prediction, the root mean square error (rMSE) in the testing set for IGM, MEM, CM, and SCM are 12.53, 5.91, 5.18, and 3.21.

LIMITATIONS

Our EHR data provide limited information on patients' demographic, socioeconomic, and other clinical factors that may be relevant to improve model performances.

CONCLUSION

We established a trajectory-based framework for depression assessment and prognosis that is adaptable to model population heterogeneity using EHR data. Collaborative modeling outperformed other established methods.

Population-based outreach versus care as usual to prevent suicide attempt: study protocol for a randomized controlled trial

BACKGROUND

Suicide remains the 10th-ranked most frequent cause of death in the United States, accounting for over 40,000 deaths per year. Nonfatal suicide attempts lead to over 200,000 hospitalizations and 600,000 emergency department visits annually. Recent evidence indicates that responses to the commonly used Patient Health Questionnaire (PHQ9) can identify outpatients who are at risk of suicide attempt and suicide death and that specific psychotherapy or Care Management programs can prevent suicide attempts in high-risk patients. Motivated by these developments, the NIMH-funded Mental Health Research Network has undertaken a multisite trial of two outreach programs to prevent suicide attempts among outpatients identified by routinely administered PHQ9 questionnaires.

METHODS/DESIGN

Outpatients who are at risk of suicide attempt are automatically identified using data from electronic health records (EHRs). Following a modified Zelen design, all those identified are assigned to continued usual care (i.e., no contact) or to be offered one of two population-based outreach programs. A Care Management intervention includes systematic outreach to assess suicide risk, EHR-based tools to implement risk-based care pathways, and care management to facilitate recommended follow-up. A Skills Training intervention includes interactive online training in Dialectical Behavior Therapy skills, supported by reminder and reinforcement messages from a skills coach. Each intervention supplements, rather than replaces, usual care; participants may receive any other services normally available. Interventions are delivered primarily by secure messaging through EHR patient portals. Suicide attempts and deaths following randomization are identified using state vital statistics data and health system EHR and insurance claim data. Primary evaluation will compare risk of suicide attempt or death over 18 months according to the initial assignment, regardless of intervention participation. Recruitment is underway in three health systems (Group Health Cooperative, HealthPartners, and Kaiser Permanente Colorado). Over 2500 participants have been randomized as of 1 March 2016, with enrollment averaging approximately 100 per week.

DISCUSSION

Assessing the effectiveness of population-based suicide prevention requires adherence to the principles of pragmatic trials: population-based enrollment, accepting variable treatment participation, assessing outcomes using health record data, and analyses based on intent-to-treat.

TRIAL REGISTRATION

ClinicalTrials.gov registration # NCT02326883 , registered on 23 December 2014.

Coping With Suicidal Thoughts: A Survey of Personal Experience

OBJECTIVE

The objective of the study was to describe use of services and self-care strategies by people experiencing suicidal thoughts.

METHODS

Constituents of the Depression and Bipolar Support Alliance (N=611) completed an anonymous online survey regarding experience of suicidal ideation and use of a range of clinical services, community supports, and self-care strategies.

RESULTS

Mental health providers were the most frequently used and the most favorably rated source of support. Peer supports were less frequently used but also favorably rated. Emergency rooms and telephone crisis clinics were used less frequently and were rated less favorably. The most frequently used self-care strategies included engaging in distracting activities or social activities, using positive affirmations, exercising, and engaging in personal spiritual practices.

CONCLUSIONS

Peer support may be an underutilized resource for coping with suicidal thoughts. Unfavorable ratings for emergency rooms and crisis clinics may indicate a need to develop more collaborative models of emergency care. Frequent use of spiritual practices suggests greater attention to spirituality may be a useful strategy in suicide prevention.

Facilitating Action for Suicide Prevention by Learning Health Care Systems

The Mental Health Research Network (MHRN), funded by the National Institute of Mental Health to serve as a national laboratory to improve mental health care, includes researchers embedded in 13 health systems in 15 states. This column describes practice changes and effectiveness and exploratory research undertaken by MHRN partners when they found a sustained elevated risk of suicide attempts among patients who reported suicidal ideation on the nine-item Patient Health Questionnaire. Challenges described include finding common ground between what health care systems and funding agencies find compelling, choosing study designs that balance research and clinical tensions, and implementing studies in ways that minimize disruption to health systems. The authors conclude that the greatest benefit to working collaboratively with care system partners is the opportunity to improve care and to simultaneously measure the impact of change.

Racial-Ethnic Differences in Psychiatric Diagnoses and Treatment Across 11 Health Care Systems in the Mental Health Research Network

OBJECTIVE

The objective of this study was to characterize racial-ethnic variation in diagnoses and treatment of mental disorders in large not-for-profit health care systems.

METHODS

Participating systems were 11 private, not-for-profit health care organizations constituting the Mental Health Research Network, with a combined 7,523,956 patients age 18 or older who received care during 2011. Rates of diagnoses, prescription of psychotropic medications, and total formal psychotherapy sessions received were obtained from insurance claims and electronic medical record databases across all health care settings.

RESULTS

Of the 7.5 million patients in the study, 1.2 million (15.6%) received a psychiatric diagnosis in 2011. This varied significantly by race-ethnicity, with Native American/Alaskan Native patients having the highest rates of any diagnosis (20.6%) and Asians having the lowest rates (7.5%). Among patients with a psychiatric diagnosis, 73% (N=850,585) received a psychotropic medication. Non-Hispanic white patients were significantly more likely (77.8%) than other racial-ethnic groups (odds ratio [OR] range .48-.81) to receive medication. In contrast, only 34% of patients with a psychiatric diagnosis (N=548,837) received formal psychotherapy. Racial-ethnic differences were most pronounced for depression and schizophrenia; compared with whites, non-Hispanic blacks were more likely to receive formal psychotherapy for their depression (OR=1.20) or for their schizophrenia (OR=2.64).

CONCLUSIONS

There were significant racial-ethnic differences in diagnosis and treatment of psychiatric conditions across 11 U.S. health care systems. Further study is needed to understand underlying causes of these observed differences and whether processes and outcomes of care are equitable across these diverse patient populations.

Risk of suicide attempt and suicide death following completion of the Patient Health Questionnaire depression module in community practice

OBJECTIVE

To examine the association between thoughts of death or self-harm reported on item 9 of the Patient Health Questionnaire (PHQ) depression module and the risk of suicide attempt or suicide death over the following 2 years.

METHOD

In 4 health care systems participating in the Mental Health Research Network, electronic records identified 509,945 adult outpatients completing 1,228,308 PHQ depression questionnaires during visits to primary care, specialty mental health, and other outpatient providers between January 1, 2007 and December 31, 2012. 9,203 nonfatal suicide attempts were identified using health system records of inpatient or outpatient encounters for self-inflicted injury. 484 suicide deaths were identified using cause-of-death codes from state mortality data.

RESULTS

Cumulative hazard of suicide attempt during 2 years ranged from approximately 0.5% among those reporting thoughts of death or self-harm "not at all" to 3.5% among those reporting such thoughts "nearly every day." Cumulative hazard of suicide death during 2 years ranged from approximately 0.04% among those responding "not at all" to 0.19% among those responding "nearly every day." The excess hazard associated with thoughts of death or self-harm declined with time, but remained 2- to 5-fold higher for at least 18 months. Nevertheless, 39% of suicide attempts and 36% of suicide deaths within 30 days of completing a PHQ occurred among those responding "not at all" to item 9.

CONCLUSIONS

In community practice, response to PHQ item 9 is a strong predictor of suicide attempt and suicide death over the following 2 years. For patients reporting thoughts of death or self-harm, suicide prevention efforts must address this enduring vulnerability.

Organized Self-Management Support Services for Chronic Depressive Symptoms: A Randomized Controlled Trial

OBJECTIVE

This study aimed to determine whether a self-management support service was more effective than treatment as usual in reducing depressive symptoms and major depressive episodes and increasing personal recovery among individuals with chronic or recurrent depressive symptoms.

METHODS

The study was a randomized controlled trial of a self-management support service consisting of depression self-management training, recovery coaching, and care coordination. The 18-month intervention included regular telephone or in-person contacts with a care manager and a structured group program co-led by a professional therapist and a trained peer specialist. Intervention (N=150) and control (N=152) participants ages ≥ 18 with chronic or recurrent depressive symptoms were recruited from five clinics in Seattle, Washington. Outcome measures included the Hopkins Symptom Checklist depression scale, the Recovery Assessment Scale, the Patient-Rated Global Improvement scale, and the percentage of participants with a major depressive episode. Interviewers were masked to treatment condition.

RESULTS

Repeated-measures estimates of the long-term effect of the intervention versus usual care (average of the six-, 12-, and 18-month outcomes adjusted for age, gender, and site) indicated that intervention participants had less severe symptoms (p=.002) and higher recovery scores (p=.03), were less likely to be depressed (odds ratio [OR]=.52, p=.001), and were more likely to be much improved (OR=1.96, p=.001).

CONCLUSIONS

These findings support providing regular outreach care management and a self-care group offering a combined behavioral and recovery-oriented approach for people with chronic or recurrent depressive symptoms.

Adherence to common cardiovascular medications in patients with schizophrenia vs. patients without psychiatric illness

OBJECTIVE

The purpose of the study was to examine whether individuals with diagnoses of schizophrenia were differentially adherent to their statin or angiotensin-converting enzyme inhibitor/angiotensin receptor blocker (ACEI/ARB) medications compared to individuals without psychiatric illness.

METHOD

Using electronic medical record data across 13 Mental Health Research Network sites, individuals with diagnoses of schizophrenia or schizoaffective disorder receiving two or more medication dispensings of a statin or an ACEI/ARB in 2011 (N=710) were identified and matched on age, sex and Medicare status to controls with no documented mental illness and two or more medication dispensings of a statin in 2011 (N=710). Medication adherence, and sociodemographic and clinical characteristics of the study population were assessed.

RESULTS

Multivariable models indicated that having a schizophrenia diagnosis was associated with increased odds of statin medication adherence; the odds ratio suggested a small effect. After adjustment for medication regimen, schizophrenia no longer showed an association with statin adherence. Having a schizophrenia diagnosis was not associated with ACEI/ARB medication adherence.

CONCLUSIONS

Compared to patients without any psychiatric illness, individuals with schizophrenia were marginally more likely to be adherent to their statin medications. Given that patterns of adherence to cardioprotective medications may be different from patterns of adherence to antipsychotic medications, improving adherence to the former may require unique intervention strategies.

Using the Patient Health Questionnaire (PHQ-9) and the Edinburgh Postnatal Depression Scale (EPDS) to assess suicidal ideation among pregnant women in Lima, Peru

We sought to examine the concordance of two suicidal ideation items from the Patient Health Questionnaire-9 (PHQ-9) and the Edinburgh Postnatal Depression Scale (EPDS), to evaluate the prevalence of suicidal ideation among pregnant women, and to assess the co-occurrence of suicidal ideation with antepartum depressive symptoms. A cross-sectional study was conducted among 1,517 pregnant women attending prenatal care clinics in Lima, Peru. Item 9 of the PHQ-9 assesses suicidal ideation over the last 14 days while item 10 of the EPDS assesses suicidal ideation in the past 7 days. The two suicidal ideation items have a high concordance rate (84.2 %) but a moderate agreement (the Cohen's kappa = 0.42). Based on the PHQ-9 and the EPDS, 15.8 and 8.8 % of participants screened positive for suicidal ideation, respectively. Assessed by the PHQ-9, 51 % of participants with suicidal ideation had probable depression. In prenatal care clinics, screening for suicidal ideation is needed for women with and without depressive symptoms. Future studies are needed to identify additional predictors of antepartum suicidality, determine the appropriate duration of reporting period for suicidal ideation screening, and assess the percentage of individuals with positive responses to the two suicidal ideation items at high risk of planning and attempting suicide.

Antidepressants are not overprescribed for mild depression

OBJECTIVE

To evaluate overprescribing of antidepressant medication for minimal or mild depression.

METHOD

Electronic records data from 4 large health care systems identified outpatients aged 18 years or older starting a new episode of antidepressant treatment in 2011 with an ICD-9 diagnosis of depressive disorder (296.2, 296.3, 311, or 300.4). Patient Health Questionnaire-9 (PHQ-9) depression severity scores at time of treatment initiation were used to examine the distribution of baseline severity and the association between baseline severity and patients' demographic and clinical characteristics.

RESULTS

Of 19,751 adults beginning treatment in 2011, baseline PHQ-9 scores were available for 7,051. In those with a baseline score, 85% reported moderate or severe symptoms (PHQ-9 score of 10 or more), 12% reported mild symptoms (PHQ-9 score of 5 to 9), and 3% reported minimal symptoms (PHQ-9 score of less than 5). The proportion reporting minimal or mild symptoms when starting treatment increased with age, ranging from 11% in those under age 65 years to 26% in those aged 65 and older. The proportion with minimal or mild symptoms was also moderately higher among patients living in wealthier neighborhoods and those treated by psychiatrists. Nevertheless, across all subgroups defined by sex, race/ethnicity, prescriber specialty, and treatment history, the proportions with minimal or mild symptoms did not exceed 18%. Secondary analyses, including weighting and subgroup analyses, found no evidence that estimates of baseline severity were biased by missing PHQ-9 scores.

CONCLUSIONS

In these health systems, prescribing of antidepressant medication for minimal or mild depression is much less common than suggested by previous reports. Given that this practice may sometimes be clinically appropriate, our findings indicate that overprescribing of antidepressants for mild depression is not a significant public health concern.

Problems With Using Patient Retention in the Evaluation of Mental Health Providers: Differences in Type of Dropout

OBJECTIVE

Retention, or its opposite, dropout, is a common metric of psychotherapy quality, but using it to assess provider performance can be problematic. Differences among providers in numbers of general dropouts, "good" dropouts (patients report positive treatment experiences and outcome), and "bad" dropouts (patients report negative treatment experiences and outcome) were evaluated.

METHODS

Patient records were paired with satisfaction surveys (N=3,054). Binomial mixed-effects models were used to examine differences among providers by dropout type.

RESULTS

Thirty-four percent of treatment episodes resulted in dropout. Of these, 14% were bad dropouts and 27% were good dropouts. Providers accounted for approximately 17% of the variance in general dropout and 10% of the variance in both bad dropout and good dropout. The ranking of providers fluctuated by type of dropout.

CONCLUSIONS

Provider assessments based on patient retention should offer a way to isolate dropout type, given that nonspecific metrics may lead to biased estimates of performance.

Psychiatric comorbidity and 30-day readmissions after hospitalization for heart failure, AMI, and pneumonia

OBJECTIVE

In 2012, the Centers for Medicare and Medicaid Services implemented a policy that penalizes hospitals for "excessive" all-cause hospital readmissions within 30 days after discharge from an index hospitalization for heart failure (HF), acute myocardial infarction (AMI), and pneumonia. The aim of this study was to investigate the influence of psychiatric comorbidities on 30-day all-cause readmissions following hospitalizations for HF, AMI, and pneumonia.

METHODS

Data from 2009-2011 were derived from the HMO Research Network Virtual Data Warehouse of 11 health systems affiliated with the Mental Health Research Network. All index inpatient hospitalizations for HF, AMI, and pneumonia were captured (N=160,169). Psychiatric diagnoses for the year prior to admission were measured. All-cause readmissions within 30 days of discharge were the outcome variable.

RESULTS

Approximately 18% of all individuals with index inpatient hospitalizations for HF, AMI, and pneumonia were readmitted within 30 days. The rate of readmission was 5% greater for individuals with a psychiatric comorbidity compared with those without a psychiatric comorbidity (21.7% and 16.5%, respectively, p<.001). Depression, anxiety, and dementia were associated with more readmissions of persons with index hospitalizations for each general medical condition and for all the conditions combined (p<.05). Substance use and bipolar disorders were linked with higher readmissions for those with initial hospitalizations for HF and pneumonia (p<.05). Readmission rates declined overall from 2009 to 2011.

CONCLUSIONS

Individuals with HF, AMI, and pneumonia experience high rates of readmission, but psychiatric comorbidities appear to increase that risk. Future interventions to reduce readmission should consider adding mental health components.