Sachin J. Shah

Health consequences of the US Deferred Action for Childhood Arrivals (DACA) immigration programme: a quasi-experimental study

BACKGROUND The effects of changes in immigration policy on health outcomes among undocumented immigrants are not well known. We aimed to examine the physical and mental health effects of the Deferred Action for Childhood Arrivals (DACA) programme, a 2012 US immigration policy that provided renewable work permits and freedom from deportation for a large number of undocumented immigrants. METHODS We did a retrospective, quasi-experimental study using nationally representative, repeated cross-sectional data from the US National Health Interview Survey (NHIS) for the period January, 2008, to December, 2015. We included non-citizen, Hispanic adults aged 19-50 years in our analyses. We used a difference-in-differences strategy to compare changes in health outcomes among individuals who met key DACA eligibility criteria (based on age at immigration and at the time of policy implementation) before and after programme implementation versus changes in outcomes for individuals who did not meet these criteria. We additionally restricted the sample to individuals who had lived in the USA for at least 5 years and had completed high school or its equivalent, in order to hold fixed two other DACA eligibility criteria. Our primary outcomes were self-reported overall health (measured on a 5 point Likert scale) and psychological distress (Kessler 6 [K6] scale), the latter was administered to a random subset of NHIS respondents. FINDINGS Our final sample contained 14 973 respondents for the self-reported health outcome and 5035 respondents for the K6 outcome. Of these individuals, 3972 in the self-reported health analysis and 1138 in the K6 analysis met the DACA eligibility criteria. Compared with people ineligible for DACA, the introduction of DACA was associated with no significant change among DACA-eligible individuals in terms of self-reported overall health (b=0·056, 95% CI -0·024 to 0·14, p=0·17) or the likelihood of reporting poor or fair health (adjusted odds ratio [aOR] 0·98, 95% CI 0·66-1·44, p=0·91). However, DACA-eligible individuals experienced a reduction in K6 score compared with DACA-ineligible individuals (adjusted incident risk ratio 0·78, 95% CI 0·56-0·95, p=0·020) and were less likely to meet screening criteria for moderate or worse psychological distress (aOR 0·62, 95% CI 0·41-0·93, p=0·022). INTERPRETATION Economic opportunities and protection from deportation for undocumented immigrants, as offered by DACA, could confer large mental health benefits to such individuals. Health consequences should be considered by researchers and policy makers in evaluations of the broader welfare effects of immigration policy. FUNDING None.

Financial Stress and Outcomes after Acute Myocardial Infarction

Background Little is known about the association between financial stress and health care outcomes. Our objective was to examine the association between self-reported financial stress during initial hospitalization and long-term outcomes after acute myocardial infarction (AMI). Materials and Methods We used Prospective Registry Evaluating Myocardial Infarction: Event and Recovery (PREMIER) data, an observational, multicenter US study of AMI patients discharged between January 2003 and June 2004. Primary outcomes were disease-specific and generic health status outcomes at 1 year (symptoms, function, and quality of life (QoL)), assessed by the Seattle Angina Questionnaire [SAQ] and Short Form [SF]-12. Secondary outcomes included 1-year rehospitalization and 4-year mortality. Hierarchical regression models accounted for patient socio-demographic, clinical, and quality of care characteristics, and access and barriers to care. Results Among 2344 AMI patients, 1241 (52.9%) reported no financial stress, 735 (31.4%) reported low financial stress, and 368 (15.7%) reported high financial stress. When comparing individuals reporting low financial stress to no financial stress, there were no significant differences in post-AMI outcomes. In contrast, individuals reporting high financial stress were more likely to have worse physical health (SF-12 PCS mean difference −3.24, 95% Confidence Interval [CI]: −4.82, −1.66), mental health (SF-12 MCS mean difference: −2.44, 95% CI: −3.83, −1.05), disease-specific QoL (SAQ QoL mean difference: −6.99, 95% CI: −9.59, −4.40), and be experiencing angina (SAQ Angina Relative Risk = 1.66, 95%CI: 1.19, 2.32) at 1 year post-AMI. While 1-year readmission rates were increased (Hazard Ratio = 1.50; 95%CI: 1.20, 1.86), 4-year mortality was no different. Conclusions High financial stress is common and an important risk factor for worse long-term outcomes post-AMI, independent of access and barriers to care.

Payments for Acute Myocardial Infarction Episodes-of-Care Initiated at Hospitals With and Without Interventional Capabilities

Background—It is unknown whether hospitals with percutaneous coronary intervention (PCI) capability provide costlier care than hospitals without PCI capability for patients with acute myocardial infarction. The growing number of PCI hospitals and higher rate of PCI use may result in higher costs for episodes-of-care initiated at PCI hospitals. However, higher rates of transfers and postacute care procedures may result in higher costs for episodes-of-care initiated at non-PCI hospitals. Methods and Results—We identified all 2008 acute myocardial infarction admissions among Medicare fee-for-service beneficiaries by principal discharge diagnosis and classified hospitals as PCI- or non-PCI-capable on the basis of hospitals’ 2007 PCI performance. We added all payments from admission through 30 days postadmission, including payments to hospitals other than the admitting hospital. We calculated and compared risk-standardized payment for PCI and non-PCI hospitals using 2-level hierarchical generalized linear models, adjusting for patient demographics and clinical characteristics. PCI hospitals had a higher mean 30-day risk-standardized payment than non-PCI hospitals (PCI,

Regional associations between medicare advantage penetration and administrative claims-based measures of hospital outcomes.

20 340; non-PCI,

Effect of Variation in Published Stroke Rates on the Net Clinical Benefit of Anticoagulation for Atrial Fibrillation

19 713; P<0.001). Patients presenting to PCI hospitals had higher PCI rates (39.2% versus 13.2%; P<0.001) and higher coronary artery bypass graft rates (9.5% versus 4.4%; P<0.001) during index admissions, lower transfer rates (2.2% versus 25.4%; P<0.001), and lower revascularization rates within 30 days (0.15% versus 0.27%; P<0.0001) than those presenting to non-PCI hospitals. Conclusions—Despite higher PCI and coronary artery bypass graft rates for Medicare patients initially presenting to PCI hospitals, PCI hospitals were only

Appointment "no-shows" are an independent predictor of subsequent quality of care and resource utilization outcomes.

627 costlier than non-PCI hospitals for the treatment of patients with acute myocardial infarction in 2008.

Targeted Reminder Phone Calls to Patients at High Risk of No-Show for Primary Care Appointment: A Randomized Trial.

Background:Risk-standardized measures of hospital outcomes reported by the Centers for Medicare and Medicaid Services include Medicare fee-for-service (FFS) patients and exclude Medicare Advantage (MA) patients due to data availability. MA penetration varies greatly nationwide and seems to be associated with increased FFS population risk. Whether variation in MA penetration affects the performance on the Centers for Medicare and Medicaid Service measures is unknown. Objective:To determine whether the MA penetration rate is associated with outcomes measures based on FFS patients. Research Design:In this retrospective study, 2008 MA penetration was estimated at the Hospital Referral Region (HRR) level. Risk-standardized mortality rates and risk-standardized readmission rates for heart failure, acute myocardial infarction, and pneumonia from 2006 to 2008 were estimated among HRRs, along with several markers of FFS population risk. Weighted linear regression was used to test the association between each of these variables and MA penetration among HRRs. Results:Among 304 HRRs, MA penetration varied greatly (median, 17.0%; range, 2.1%–56.6%). Although MA penetration was significantly (P<0.05) associated with 5 of the 6 markers of FFS population risk, MA penetration was insignificantly (P≥0.05) associated with 5 of 6 hospital outcome measures. Conclusion:Risk-standardized mortality rates and risk-standardized readmission rates for heart failure, acute myocardial infarction, and pneumonia do not seem to differ systematically with MA penetration, lending support to the widespread use of these measures even in areas of high MA penetration.

ST-elevation myocardial infarction patients can be enrolled in randomized trials before emergent coronary intervention without sacrificing door-to-balloon time.

Oral anticoagulation (OAC) dramatically reduces risk for ischemic stroke in patients with atrial fibrillation (AF) but at the cost of increasing risk for major hemorrhage, including intracranial hemorrhage (ICH), which is frequently fatal (1, 2). As such, the decision to recommend OAC to a patient with AF should be based on the expected net clinical benefit of OACthat is, the difference between the reduction in ischemic stroke risk and increase in bleeding risk weighted by the severity of each of these outcomes (3, 4). To help make the decision, several professional societies have published guidelines recommending that physicians assess risk for ischemic stroke using the CHA2DS2-VASc (congestive heart failure, hypertension, age, diabetes, stroke, and vascular disease) score (5). The most recent guidelines from the European Society of Cardiology (ESC) and joint guidelines from the American Heart Association, American College of Cardiology, and Heart Rhythm Society (AHA/ACC/HRS) recommend using OAC with a CHA2DS2-VASc score of 2 or greater (excluding sex as a risk factor in the ESC guidelines) (6, 7). Both guidelines use the rationale that above the specified threshold score, the expected benefits of OAC outweigh the expected harms. These guidelines implicitly assume that the CHA2DS2-VASc score accurately represents an individual patients risk for ischemic stroke without anticoagulation, that the risk for ischemic stroke for a given score is consistent across populations, and that OAC will reduce relative ischemic stroke risk for a patient with AF by roughly two thirds across the range of underlying stroke risk. Recent work by Quinn and colleagues (8) raised doubts about the second assumption that the risk for stroke for a given CHA2DS2-VASc score is consistent across populations. Their systematic review found that stroke rates for patients with a given score who are not receiving anticoagulant therapy varied substantially from one study population to the next. For example, for a CHA2DS2-VASc score of 2, stroke rates ranged from 0.48% to 7.84% per year. The degree to which this variation results in a meaningfully different assessment of the net clinical benefit of anticoagulation has not been quantified. We measured the effect of variation in published rates of ischemic stroke on the expected net clinical benefit of anticoagulation in patients with incident AF who were not receiving anticoagulant therapy. To this end, we compared estimated gains with OAC in quality-adjusted life-years (QALYs) by incorporating stroke rates for untreated patients from 4 prominent AF cohorts into a Markov state-transition decision analytic model applied to a large U.S. community-based cohort with incident AF. Methods Study Overview and Design We applied the Markov state-transition model to 33434 patients with incident AF in the ATRIA-CVRN (AnTicoagulation and Risk Factors In Atrial FibrillationCardiovascular Research Network) cohort (9). The decision analytic model estimated the net clinical benefit of OAC in QALYs using a well-established Markov state-transition model (4, 10, 11). We made 4 sequential measurements of the net clinical benefit of OAC in this cohort; all model parameters were held constant except the ischemic stroke rate for a corresponding CHA2DS2-VASc score. In sequential measurements, we used ischemic stroke rates corresponding to CHA2DS2-VASc scores from ATRIA (9), the Swedish AF cohort study (12), and the Danish National Patient Registry (13) and the imputed rates from the SPORTIF (Stroke Prevention using ORal Thrombin Inhibitor in atrial Fibrillation) study quoted in the AHA/ACC/HRS guidelines (6, 14). For each patient, we produced 4 estimates of the net clinical benefit of OAC such that differences in net clinical benefit were the result of variation in published stroke rates. Decision Analytic Model We used a previously published Markov state-transition model with 28 states that compares the following 3 strategies: no antithrombotic therapy, aspirin, and OAC with warfarin (4, 10, 11). We did not use results for the aspirin strategy in this analysis because guidelines now discourage use of aspirin to prevent stroke in AF (7). For the base case, we modeled use of warfarin as the anticoagulant. We also estimated the effect of nonvitamin K antagonist oral anticoagulants (NOACs) instead of warfarin by reducing the risk for ICH by 52% in the anticoagulant group of the modelreduction in ICH is the major clinical advantage of NOACs over warfarin (15). We used a proprietary computer program (Decision Maker) to build the model and analyze the results. During each monthly cycle, patients face a chance of stroke and hemorrhage, either of which may lead to death, neurologic sequelae, or symptom resolution. The simulation runs for the entire life expectancy of the patient, adjusted for age and sex. We summarize base-case values for model parameters in Table 1; we describe the decision tree and health states in Supplement Figures 1 and 2. Table 1. Markov Model Inputs, Including Probabilities, Rates, and Quality of Life* Supplement. Supplementary Material We calculated a CHA2DS2-VASc score for each patient and, in consecutive runs, assigned the score to the corresponding annualized ischemic stroke rate reported in the ATRIA, SPORTIF, Swedish, and Danish cohorts (Table 1). We based predicted annual rates of major extracranial bleeding for each patient on a modified HAS-BLED (hypertension, abnormal liver or renal function, stroke history, bleeding predisposition, labile international normalized ratio, elderly, drug or alcohol use) score (16). Because this is a study of incident AF, labile international normalized ratio does not apply. We included the patient-specific predicted risk for ICH among untreated patients using a multivariable regression model developed in a Swedish registry population of 90490 untreated patients with AF (11, 12). We modeled differential outcomes of hemorrhagic events on the basis of whether the events occurred while patients were receiving anticoagulant therapy. Because ATRIA-CVRN did not collect data on alcohol use, we did a sensitivity analysis examining the effect of missing alcohol data (Supplement Table 3) and found that it would not meaningfully change the results. Population Used in Decision Analytic Model We applied the decision analytic model to the baseline characteristics of patients in the ATRIA-CVRN cohort. The cohort comprises 37492 patients with incident AF in the integrated health care delivery systems of Kaiser Permanente Northern and Southern California; we have previously published details of the cohort assembly (31). In brief, patients were included in the cohort if they had a new diagnosis of AF between 1 January 2006 and 30 June 2009 (defined by International Classification of Diseases, Ninth Revision, codes) and electrocardiographic data with no prior known diagnosis of AF. The ATRIA-CVRN cohort did not exclude patients with mitral stenosis or a history of valve replacement; such patients account for 1.5% of the cohort. Clinical characteristics of patients in the cohort were determined by searching inpatient, outpatient, laboratory, and pharmacy databases for the relevant International Classification of Diseases, Ninth Revision, codes; medications; or laboratory values in the year before each patients diagnosis of AF; specific codes are available on request (9, 32). The study captured variables required to calculate the CHA2DS2-VASc stroke risk score and modified HAS-BLED risk score. The study did not collect data on alcohol use. Because the decision analytic model we used was not designed to assess risk for bleeding with concurrent use of OAC and nonaspirin antiplatelet medication (such as clopidogrel), we excluded 4058 patients who had a prescription for nonaspirin antiplatelet medication at the time of AF diagnosis. Study-Specific Ischemic Stroke Rates As noted above, reported ischemic stroke rates without anticoagulation for a given CHA2DS2-VASc score vary considerably between studies. We chose rates from 4 studies. We included ATRIA because it is one of the largest U.S. studies of a community-based cohort of patients with AF. We included SPORTIF because the current AHA/ACC/HRS and ESC guidelines both report imputed ischemic stroke rates from this study. Finally, we included the Swedish and Danish studies because the ESC guidelines cite both as evidence to use the CHA2DS2-VASc score to assess stroke risk and to offer anticoagulation to patients with a score of 2 or more. We note that the cohorts vary in the outcomes included in their reported event rates. The ATRIA and SPORTIF studies reported ischemic stroke and peripheral embolism; the Swedish study reported just ischemic stroke; and the Danish study reported ischemic stroke, peripheral embolism, and pulmonary embolism (pulmonary embolism accounted for 7.7% of the studys outcome events). In prior reports, investigators have measured net clinical benefit using stroke equivalents in a nondecision analytic approach (3, 3335). To relate this analysis to previous analyses of stroke equivalents, we did a parallel analysis of net clinical benefit using stroke equivalents (Supplement Tables 1 and 2). Statistical Analysis We determined differences in benefit by comparing QALYs when patients used and did not use OAC. We determined the 95% CIs of population benefit by using 1000 bootstrapped samples and present them as a percentage of the point estimate. Because the distribution was skewed, we compared the 4 estimates of benefit based on different stroke rates using the signed-rank test. We used the Bonferroni method to adjust for multiple comparisons. We did analyses using SAS, version 9.4 (SAS Institute). Role of the Funding Source This study received no external funding. Results Table 2 shows baseline characteristics of the 33434 patients in the ATRIA-CVRN cohort: 45% were aged 75 years or older at entry, and 45% were women.