Search

OBJECTIVE The objective of this study was to compare results of using web-based and mail (postal) Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) data collection protocols. RESEARCH DESIGN Patients who had been hospitalized in a New England Hospital were surveyed about their hospital experience. Patients who provided email addresses were randomized to 1 of 3 data collection protocols: web-alone, web with postal mail follow-up, and postal mail only. Those who did not provide email addresses were surveyed using postal mail only. Analyses compared response rates, respondent characteristics, and patient-reported experiences. SUBJECTS For an 8-week period, patients were discharged from the study hospital to home. MEASURES Measures included response rates, characteristics of respondents, 6 composite measures of their patient experiences, and 2 ratings of the hospital. RESULTS Response rates were significantly lower for the web-only protocol than the mail or combined protocols, and those who had not provided email addresses had lower response rates. Those over 65 were more likely than others to respond to all protocols, especially for the mail-only protocols. Respondents without email addresses were older, less educated, and reported worse health than those who provided email addresses. After adjusting for respondent differences, those in the combined protocol differed significantly from the mail (postal) only respondents on 2 measures of patient experience; those in the web-only protocol differed on one. Those not providing an email address differed from those who did on one measure. CONCLUSION If web-based protocols are used for HCAHPS surveys, adjustments for a mode of data collection are needed to make results comparable.

Background Recent research and policy initiatives propose addressing the social determinants of health within clinical settings. One such strategy is the expansion of routine data collection on patient Race, Ethnicity, and Language (REAL) within electronic health records (EHRs). Although previous research has examined the general views of providers and patients on REAL data, few studies consider health care workers' perceptions of this data collection directly at the point of care, including how workers understand REAL data in relation to health equity. Objective This qualitative study examines a large integrated delivery system's implementation of REAL data collection, focusing on health care workers' understanding of REAL and its impact on data's integration within EHRs. Results Providers, staff, and administrators expressed apprehension over REAL data collection due to the following: (1) disagreement over data's significance, including the expected purpose of collecting REAL items; (2) perceived barriers to data retrieval, such as the lack of standardization across providers and national tensions over race and immigration; and (3) uncertainty regarding data's use (clinical decision making vs. system research) and dissemination (with whom the data may be shared; eg, public agencies, other providers, and insurers). Conclusion Emerging racial disparities associated with COVID-19 highlight the high stakes of REAL data collection. However, numerous barriers to health equity remain. Health care workers need greater institutional support for REAL data and related EHR initiatives. Despite data collection's central importance to policy objectives of disparity reduction, data mandates alone may be insufficient for achieving health equity.

Supplemental Digital Content is available in the text., Objective: The objective of this study was to evaluate whether incorporating historical clinical information beyond 1 year improves risk adjustment. Data Sources: Administrative data from the Department of Veterans Affairs and Medicare (for veterans concurrently enrolled in Medicare) for fiscal years (FYs) 2011–2015. Study Design: We regressed total annual costs on Medicare hierarchical condition category indicators and risk scores for FY 2015 in both a concurrent and a prospective model using 5-fold cross-validation. Regressions were repeated incorporating clinical information from FY 2011 to 2015. Model fit was appraised using R2 and mean squared predictive error (MSPE). Data Collection: All veterans affairs users (n=3,254,783) with diagnostic information FY 2011–2015. Principal Findings: In a concurrent model, adding additional years of historical clinical information (FY 2011–2014) did not result in substantive gains in fit (R2 from 0.671 to 0.673) or predictive capability (MSPE from 1956 to 1950). In a prospective model, adding additional years of historical clinical information also did not result in substantive gains in fit (R2 from 0.334 to 0.344) or predictive capability (MSPE from 3988 to 3940). Conclusion: Incorporating historical clinical information yielded no material gain in risk adjustment fit.

Supplemental Digital Content is available in the text., Background: Aspirin impacts risk for important outcomes such as cancer, cardiovascular disease, and gastrointestinal bleeding. However, ascertaining exposure to medications available both by prescription and over-the-counter such as aspirin for research and quality improvement purposes is a challenge. Objectives: Develop and validate a strategy for ascertaining aspirin exposure, utilizing a combination of structured and unstructured data. Research Design: This is a retrospective cohort study. Subjects: In total, 1,869,439 Veterans who underwent usual care colonoscopy 1999–2014 within the Department of Veterans Affairs. Measures: Aspirin exposure and dose were obtained from an ascertainment strategy combining query of structured medication records available in electronic health record databases and unstructured data extracted from free-text progress notes. Prevalence of any aspirin exposure and dose-specific exposure were estimated. Positive predictive value and negative predictive value were used to assess strategy performance, using manual chart review as the reference standard. Results: Our combined strategy for ascertaining aspirin exposure using structured and unstructured data reached a positive predictive value and negative predictive value of 99.2% and 97.5% for any exposure, and 92.6% and 98.3% for dose-specific exposure. Estimated prevalence of any aspirin exposure was 36.3% (95% confidence interval: 36.2%–36.4%) and dose-specific exposure was 35.4% (95% confidence interval: 35.3%–35.5%). Conclusions: A readily accessible approach utilizing a combination of structured medication records and query of unstructured data can be used to ascertain aspirin exposure when manual chart review is impractical.

Background General population surveys are increasingly offering broader response options for questions on sexual orientation-for example, not only gay or lesbian, but also "something else" (SE) and "don't know" (DK). However, these additional response options are potentially confusing for those who may not know what the terms mean. Researchers studying sexual orientation-based disparities face difficult methodological trade-offs regarding how best to classify respondents identifying with the SE and DK categories. Objectives Develop respondent-level probabilities of sexual minority orientation without excluding or misclassifying the potentially ambiguous SE and DK responses. Compare 3 increasingly inclusive analytic approaches for estimating health disparities using a single item: (a) omitting SE and DK respondents; (b) classifying SE as sexual minority and omitting DK; and (c) a new approach classifying only SE and DK respondents with >50% predicted probabilities of being sexual minorities as sexual minority. Materials and methods We used the sociodemographic information and follow-up questions for SE and DK respondents in the 2013-2014 National Health Interview Survey to generate predicted probabilities of identifying as a sexual minority adult. Results About 94% of the 144 SE respondents and 20% of the 310 DK respondents were predicted to identify as a sexual minority adult, with higher probabilities for younger, wealthier, non-Hispanic white, and urban-dwelling respondents. Using a more specific definition of sexual minority orientation improved the precision of health and health care disparity estimates. Conclusions Predicted probabilities of sexual minority orientation may be used in this and other surveys to improve representation and categorization of those who identify as a sexual minority adult.

Recent research and policy initiatives propose addressing the social determinants of health within clinical settings. One such strategy is the expansion of routine data collection on patient Race, Ethnicity, and Language (REAL) within electronic health records (EHRs). Although previous research has examined the general views of providers and patients on REAL data, few studies consider health care workers' perceptions of this data collection directly at the point of care, including how workers understand REAL data in relation to health equity.This qualitative study examines a large integrated delivery system's implementation of REAL data collection, focusing on health care workers' understanding of REAL and its impact on data's integration within EHRs.Providers, staff, and administrators expressed apprehension over REAL data collection due to the following: (1) disagreement over data's significance, including the expected purpose of collecting REAL items; (2) perceived barriers to data retrieval, such as the lack of standardization across providers and national tensions over race and immigration; and (3) uncertainty regarding data's use (clinical decision making vs. system research) and dissemination (with whom the data may be shared; eg, public agencies, other providers, and insurers).Emerging racial disparities associated with COVID-19 highlight the high stakes of REAL data collection. However, numerous barriers to health equity remain. Health care workers need greater institutional support for REAL data and related EHR initiatives. Despite data collection's central importance to policy objectives of disparity reduction, data mandates alone may be insufficient for achieving health equity.

Background To enhance the quality of hospice care and to facilitate consumers' choices, the Centers for Medicare and Medicaid Services (CMS) began the Hospice Quality Reporting Program, in which CMS posted the quality measures of participating hospices on its reporting website, Hospice Compare. Little is known about the participation rate and the types of nonparticipating hospices. Objective To examine the factors associated with hospices' nonparticipation in Hospice Compare. Research design We analyzed data from the CMS 2016 Hospice Compare. "Nonparticipants" were those who did not submit any quality measure. With the data of the Provider of Service file, the Healthcare Cost Report Information System, and the Area Health Resources File, multivariate logistic regressions estimated the association between nonparticipants and hospice and market characteristics, including ownership, size, nurse staffing ratio, and market competition intensity. Results Among the 4123 certified hospices subject to penalty from nonparticipation, 259 did not participate in Hospice Compare. California, New Mexico, Texas, and Wyoming had participation rates lower than 80%. Hospices that were for-profit, had no accreditation, had few nurses per patient day, provided no inpatient care, and were located in competitive markets were less likely to participate than other hospices. Conclusions Hospice Compare successfully motivated hospice in participating in the quality report program in most of states. For-profit hospices, hospices with less quality, and hospices located in competitive markets were less likely to participate. Further research is warranted to examine the quality of these nonparticipants, especially in the 4 states with a lower participation rate.

BACKGROUND–: Despite widespread use of comorbidities for population health descriptions and risk adjustment, the ideal method for ascertaining comorbidities is not known. We sought to compare the relative value of several methodologies by which comorbidities may be ascertained. METHODS–: This is an observational study of 1,596 patients admitted to the University of Chicago for community-acquired pneumonia from 1998–2012. We collected data via chart abstraction, administrative data, and patient report, then performed logistic regression analyses, specifying comorbidities as independent variables and in-hospital mortality as the dependent variable. Finally, we compared area under the curve (AUC) statistics to determine the relative ability of each method of comorbidity ascertainment to predict in-hospital mortality. RESULTS–: Chart review (area under curve [AUC] 0.72) and administrative data (Charlson AUC 0.83, Elixhauser AUC 0.84) predicted in-hospital mortality with greater fidelity than patient report (AUC 0.61). However, multivariate logistic regression analyses demonstrated that individual comorbidity derivation via chart review had the strongest relationship with in-hospital mortality. This is consistent with prior literature suggesting that administrative data has inherent, paradoxical biases with important implications for risk adjustment based solely on administrative data. CONCLUSIONS–: Although comorbidities derived through administrative data did produce an AUC greater than chart review, our analyses suggest a coding bias in several comorbidities with a paradoxically protective effect. Therefore, chart review, while labor and resource intensive, may be the ideal method for ascertainment of clinically-relevant comorbidities.

BACKGROUND Systematic coding systems are used to define clinically meaningful outcomes when leveraging administrative claims data for research. How and when these codes are applied within a research study can have implications for the study validity and their specificity can vary significantly depending on treatment received. SUBJECTS Data are from the Surveillance, Epidemiology, and End Results-Medicare linked dataset. STUDY DESIGN We use propensity score methods in a retrospective cohort of prostate cancer patients first examined in a recently published radiation oncology comparative effectiveness study. RESULTS With the narrowly defined outcome definition, the toxicity event outcome rate ratio was 0.88 per 100 person-years (95% confidence interval, 0.71-1.08). With the broadly defined outcome, the rate ratio was comparable, with 0.89 per 100 person-years (95% confidence interval, 0.76-1.04), although individual event rates were doubled. Some evidence of surveillance bias was suggested by a higher rate of endoscopic procedures the first year of follow-up in patients who received proton therapy compared with those receiving intensity-modulated radiation treatment (11.15 vs. 8.90, respectively). CONCLUSIONS This study demonstrates the risk of introducing bias through subjective application of procedure codes. Careful consideration is required when using procedure codes to define outcomes in administrative data.

Author(s): Setodji, Claude M; Burkhart, Q; Hays, Ron D; Quigley, Denise D; Skootsky, Samuel A; Elliott, Marc N | Abstract: BackgroundPatient experience data can be collected by sampling patients periodically (eg, patients with any visits over a 1-year period) or sampling visits continuously (eg, sampling any visit in a monthly interval). Continuous sampling likely yields a sample with more frequent and more recent visits, possibly affecting the comparability of data collected under the 2 approaches.ObjectiveTo explore differences in Consumer Assessment of Healthcare Providers and Systems Clinician and Group survey (CG-CAHPS) scores using periodic and continuous sampling.Research designWe use observational data to estimate case-mix-adjusted differences in patient experience scores under 12-month periodic sampling and simulated continuous sampling.SubjectsA total of 29,254 adult patients responding to the CG-CAHPS survey regarding visits in the past 12 months to any of 480 physicians, 2007-2009.MeasuresOverall doctor rating and 4 CG-CAHPS composite measures of patient experience: doctor communication, access to care, care coordination, and office staff.ResultsCompared with 12-month periodic sampling, simulated continuous sampling yielded patients with more recent visits (by definition), more frequent visits (92% of patients with 2+ visits, compared with 76%), and more positive case-mix-adjusted CAHPS scores (2-3 percentage points higher).ConclusionsPatients with more frequent visits reported markedly higher CG-CAHPS scores, but this causes only small to moderate changes in adjusted physician-level scores between 12-month periodic and continuous sampling schemes. Caution should be exercised in trending or comparing scores collected through different schemes.

Background Patient experience surveys are widely used to capture the patient-reported quality of care and are increasingly being used for formal reporting purposes. There is evidence that certain patient subgroups are less likely to respond to traditional CAHPS surveys. As patient-facing technologies become more common, it is important to examine whether tablet-based patient experience surveys have the potential to promote responses from more diverse populations. Objectives To develop, gain perspectives about, and pilot an English and Spanish low-literacy adaptation of the Consumer Assessment of Healthcare Providers & Systems Clinician & Group Survey (CG-CAHPS) administered on a tablet device at the point of care. Research design Cognitive testing and evaluation of a quality improvement pilot comparing a tablet-based adaptation and traditional paper-based versions of the CG-CAHPS survey. Subjects English-speaking and Spanish-speaking patients receiving primary care in an urban community clinic. Measures To compare the acceptability of low-literacy tablet-based and traditional paper-based patient experience surveys, we examined the concordance of responses between survey modes and preferences for modality, as well as perspectives on usability and reporting care experiences. We examined demographic differences in responses to tablet-based versus mailed surveys from a quality improvement pilot. Results The majority of cognitive interview participants preferred a low-literacy, tablet-based survey over a paper-based survey with traditional wording. In a quality improvement pilot comparing tablet-based administration at the point of care versus mailed surveys, respondents to the tablet-based survey were more likely to be younger and Latino. Conclusions If designed with patient input, tablet-based surveys have the potential to improve the collection of patient experience data among diverse populations.

Introduction The inclusion of reference values for common patient-reported outcomes (PROs) measures in clinical care settings provides a clinically relevant context for an individual patient's PRO scores. PRO reference values are currently not reported in clinical care settings. This is a missed opportunity, as clinicians are familiar with the presence and interpretation of reference values, commonly provided alongside laboratory test results. Incorporating PRO reference values into clinical PRO reporting requires: an understanding of the clinical purpose, the availability of an appropriate reference value, and graphical representation. Methods for pro score interpretation We present reference value terminology adapted for PROs and discuss important differences between using reference values in the PRO score interpretation compared to other types of clinical measures from clinical chemistry. We outline the basic methodological approaches in obtaining a PRO reference sample and calculating reference intervals. Lastly, we provide recommendations on how to present and use PRO reference values in clinical care settings. Discussion There is a strong, long-standing discipline behind reference value development and application in psychology and medicine, allowing for both providers and patients to understand comparisons and identify what is "out of range." PRO reference values can be communicated in a wide range of ways within clinical care settings and are adaptable as required to different patient populations or clinical care situations. However, a notable adoption barrier is the expense and methodological expertise needed to establish and apply PRO reference values effectively in clinical encounters.

Introduction The collection of patient-reported outcomes (PROs) in routine clinical practice provides opportunities to "feed-forward" the patient's perspective to his/her clinical team to inform planning and management. This data can also be aggregated to "feedback" population-level analytics that can inform treatment decision-making, predictive modeling, population-based care, and system-level quality improvement efforts. Methods aiding interpretation and acting on results Three case studies demonstrate a number of system-level features which aid effective PRO interpretation: (1) feed-forward and feedback information flows; (2) score interpretation aids; (3) cascading measurement; (4) registry-enabled learning health care systems; and (5) the maturational development of information systems. Discussion The case studies describe the developmental span of feed-forward PRO programs-from simple to mature applications. The Concord Hospital (CH) Multiple Sclerosis Neurobehavioral Clinic exemplifies a simple application in which PRO data are used before and during clinic visits by patients and clinicians to inform care. The Dartmouth-Hitchcock (D-H) Spine Center exemplifies a mature program which utilizes population-level analytics to provide decision support by predicting outcomes for different treatment options. The Swedish Rheumatology Quality (SRQ) Registry epitomizes an exceptional application which has spread to multiple systems across an entire country. Key points Feed-forward and feedback PRO information systems can better inform, involve, and support clinicians, patients and families, and allow health systems to monitor and improve system performance and population health outcomes. Ideal systems have the capability for multilevel analyses at patient, system, and population levels, and an information technology infrastructure that is linked to associated workflows and a supportive practice culture. As systems mature, they progress beyond the ability to describe and inform towards higher level capabilities including prediction and decision support. Finally, there is additional promise for the integration of patient-reported information that is adjusted (or weighted) by preferences and values to guide shared decision-making and inform individualized precision health care in the future.

Objectives To compare the performance of the health-related quality of life-comorbidity index (HRQoL-CI) with the diagnosis-based Charlson, Elixhauser, and combined comorbidity scores and the prescription-based chronic disease score (CDS) in predicting HRQoL in Agency of Healthcare Research and Quality priority conditions (asthma, breast cancer, diabetes, and heart failure). Methods The Medical Expenditure Panel Survey (2005 and 2007-2011) data was used for this retrospective study. Four disease-specific cohorts were developed that included adult patients (age 18 y and above) with the particular disease condition. The outcome HRQoL [physical component score (PCS) and mental component score (MCS)] was measured using the Short Form Health Survey, Version 2 (SF-12v2). Multiple linear regression analyses were conducted with the PCS and MCS as dependent variables. Comorbidity scores were compared using adjusted R. Results Of 140,046 adult participants, the study cohort included 7436 asthma (5.3%), 1054 breast cancer (0.8%), 13,829 diabetes (9.9%), and 937 heart failure (0.7%) patients. Among individual scores, HRQoL-CI was best at predicting PCS and MCS. Adding prescription-based comorbidity scores to HRQoL-CI in the same model improved prediction of PCS and MCS. HRQoL-CI+CDS performed the best in predicting PCS (adjusted R): asthma (43.7%), breast cancer (31.7%), diabetes (32.7%), and heart failure (20.0%). HRQoL-CI+CDS and Elixhauser+CDS had superior and comparable performance in predicting MCS (adjusted R): asthma (HRQoL-CI+CDS=20.1%; Elixhauser+CDS=19.6%), breast cancer (HRQoL-CI+CDS=12.9%; Elixhauser+CDS=14.1%), diabetes (HRQoL-CI+CDS=17.7%; Elixhauser+CDS=17.7%), and heart failure (HRQoL-CI+CDS=18.1%; Elixhauser+CDS=17.7%). Conclusions HRQoL-CI performed best in predicting HRQoL. Combining prescription-based scores to diagnosis-based scores improved the prediction of HRQoL.

BACKGROUND The Relational Coordination (RC) instrument has been used extensively in the context of health care interprofessional coordination. However, the instrument's applicability to patient experiences of their interactions with professionals is largely untested. OBJECTIVES This study's objectives were to determine: (1) whether the RC instrument could be modified for phone administration to yield internally consistent results when used with a diverse group of patients with complex health needs; and (2) whether the RC factor was invariant across patients of differing education, levels of emotional problems, race, and ethnicity, thereby showing similar interpretation of items across these groups. RESEARCH DESIGN The RC instrument was administered through a phone survey to patients in Texas (n=346) who reported receiving care coordination. Data collection occurred between 2014 and 2016. Cronbach α coefficients and confirmatory factor analysis were used to determine whether the original set of RC items could be used for phone surveys with patients. Factorial invariance testing was used to assess how consistently the instrument was interpreted across patient subgroups. RESULTS The RC scale generally met acceptable α statistic and confirmatory factor analysis thresholds for internal consistency. Factorial invariance results indicated that the scale also generally performed consistently across patient subgroups. CONCLUSIONS This study provides preliminary evidence that the RC instrument can be used for surveying diverse patient populations. Future use of this instrument with patients can better reflect their experiences as partners with professionals in improving their health.

Although frequently used to track health care disparities, patient race/ethnicity data collected by hospitals can be unreliable, particularly for smaller minority groups. We sought to determine whether the racial/ethnic distribution of hospitalized patients shifted after implementation of a statewide initiative to standardize data collection practices.We conducted a difference-in-differences analysis of the State Inpatient Databases to estimate changes in the proportion of patients identified as non-Hispanic white, non-Hispanic black, Hispanic, Asian/Pacific Islander, and "other," before (2005-2006) and after (2008-2009) standardized practices were implemented in New Jersey relative to New York, a state with similar demographics but no changes to data collection.Among 12,552,702 hospital discharges, modest relative changes were noted in the proportion of patients identified as non-Hispanic white [+1.1%; 95% confidence interval (CI): +0.9 to +1.2] and non-Hispanic black (+1.6%; 95% CI: +1.1 to +2.1) in New Jersey that were attributed to its use of standardized data collection practices as compared with New York. Larger relative changes were noted in the proportion of patients identified as Hispanic (-7.1%; 95% CI: -7.8 to -6.4), Asian/Pacific Islander (+26.5%; 95% CI: +25.1 to +27.9), and "other" (-24.6%; 95% CI: -26.4 to -22.8). This pattern was largely consistent in analyses stratified by sex, age, and Major Diagnostic Category.Measurement of health care disparities fundamentally depends on the racial/ethnic categorization of individuals. By redistributing substantial proportions of patients across smaller minority groups, standardized data collection could lead to shifts in estimates of health care disparities for these rapidly growing populations.

Objective To examine the effect of survey mode (mail vs. telephone) on the likelihood of reporting health care-related discrimination based on race, ethnicity, or nationality. Methods We use data from a mixed-mode, mail and telephone survey of public health care program enrollees (N=2807), including Somali, Hmong, African American, American Indian, and Latino populations. Self-reported discrimination was measured as the experience of unfair treatment by health care providers due to race, ethnicity, or nationality. We use propensity score matching to create exchangeable groups of phone and mail respondents based on the probability of completing the survey by phone. Results Overall, 33.1% of respondents reported having experienced discrimination in health care, but only 23.6% of telephone respondents reported discrimination compared with 36.8% of mail respondents. After matching phone and mail respondents based on probability of responding by telephone, all observable significant differences between respondents that were brought about by differential self-selection into mode were erased, allowing us to estimate the effect of survey mode on report of discrimination. Even after matching, the mode effect remains, where report of health care discrimination for telephone respondents would have been 12.6 percentage points higher had they responded by mail (22.6% vs. 35.2%). Conclusions Survey mode has a significant effect on report of discrimination. Respondents may be more willing to disclose experiences of discrimination in a mail survey than to a telephone interviewer. Findings have substantial policy and clinical significance as variation in report of discrimination based on mode may lead to underestimation of the extent of the problem.

OBJECTIVE We sought to build on the template-matching methodology by incorporating longitudinal comorbidities and acute physiology to audit hospital quality. STUDY SETTING Patients admitted for sepsis and pneumonia, congestive heart failure, hip fracture, and cancer between January 2010 and November 2011 at 18 Kaiser Permanente Northern California hospitals. STUDY DESIGN We generated a representative template of 250 patients in 4 diagnosis groups. We then matched between 1 and 5 patients at each hospital to this template using varying levels of patient information. DATA COLLECTION Data were collected retrospectively from inpatient and outpatient electronic records. PRINCIPAL FINDINGS Matching on both present-on-admission comorbidity history and physiological data significantly reduced the variation across hospitals in patient severity of illness levels compared with matching on administrative data only. After adjustment for longitudinal comorbidity and acute physiology, hospital rankings on 30-day mortality and estimates of length of stay were statistically different from rankings based on administrative data. CONCLUSIONS Template matching-based approaches to hospital quality assessment can be enhanced using more granular electronic medical record data.

Aims: To investigate the feasibility of discrete choice experiments for valuing EQ-5D-5L states using computer-based data collection, the consistency of the estimated regression coefficients produced after modeling the preference data, and to examine the similarity of the values derived across countries.Methods: Data were collected in Canada, England, The Netherlands, and the United States (US). Interactive software was developed to standardize the format of the choice tasks across countries, except for face-to-face interviewing in England. The choice task required respondents to choose between 2 suboptimal health states. A Bayesian design was used to generate 200 pairs of states that were randomly grouped into 20 blocks. Each respondent completed 1 block of 10 pairs. A main-effects probit model was used to estimate regression coefficients and to derive values.Results: Approximately 400 respondents participated from each country. The mean time to perform 1 choice task was between 29.2 (US) and 45.2 (England) seconds. All regression coefficients were statistically significant, except level 2 for Usual Activities in The Netherlands (P=0.51). Predictions for the complete set of 3125 EQ-5D-5L health states were similar for the 4 countries. Intraclass correlation coefficients between the countries were high: from 0.80 (England vs. US) through 0.98 (Canada vs. US).Conclusions: Derivation of value sets from the general population using computer-based choice tasks for the EQ-5D-5L is feasible. Parameter estimates were generally consistent and logical, and health-state values were similar across the 4 countries.

Epidemiologic studies of prescription medications increasingly rely on large administrative health care databases. These data do not capture patients' use of medication samples. This could potentially bias studies of short-term effects where date of initiation may be inaccurate.To assess the extent of sample use among patients initiating statin therapy.Retrospective cohort of patients who filled a first prescription for a statin after at least 6 months of statin-free period in 2007-2010. Low-density lipoprotein (LDL) values obtained within the 15 days preceding the first prescription were analyzed using a 2-component Gaussian mixture model to look for evidence of prior treatment.A total of 26,033 statin initiators with at least 1 LDL laboratory result within the 15 days preceding the prescription fill.Estimators for the proportion of patients filling a new prescription already on treatment.Among 9256 patients filling a branded statin, LDL distribution was bimodal, consisting of 2 Gaussian distributions: one, which made up 13.4% of the total population, had much lower LDL values (mean=71.8 mg/dL) compared with the second (mean=148.0 mg/dL), suggesting drug use before first dispensed prescription. Among 16,777 patients filling a generic statin, LDL levels were substantially higher with no evidence of bimodality that would suggest prior sample use.These results provide indirect evidence that the initial period of branded medication use may often be missed when using pharmacy claims data to define drug initiation. Further research is needed to examine approaches to better identify incident medication use when assessing short-term effects.

Objective: Emergency department (ED) crowding is linked with poor quality of care and worse outcomes, including higher mortality. With the growing emphasis on hospital performance measures, there is additional concern whether inadequate care during crowded periods increases a patient's likelihood of subsequent inpatient admission. We sought to determine if ED crowding during the index visit was associated with these "bounceback" admissions. Methods: We used comprehensive, nonpublic, statewide ED and inpatient discharge data from the California Office of Statewide Health Planning and Development from 2007 to identify index outpatient ED visits and bounceback admissions within 7 days. We further used ambulance diversion data collected from California local emergency medical services agencies to identify crowded days using intrahospital daily diversion hour quartiles. Using a hierarchical logistic regression model, we then determined if patients visiting on crowded days were more likely to have a subsequent bounceback admission. Results: We analyzed 3,368,527 index visits across 202 hospitals, of which 596,471 (17.7%) observations were on crowded days. We found no association between ED crowding and bounceback admissions. This lack of relationship persisted in both a discrete (high/low) model (OR, 1.01; 95% CI, 0.99, 1.02) and a secondary model using ambulance diversion hours as a continuous predictor (OR, 1.00; 95% CI, 1.00, 1.00). Conclusions: Crowding as measured by ambulance diversion does not have an association with hospitalization within 7 days of an ED visit discharge. Therefore, bounceback admission may be a poor measure of delayed or worsened quality of care due to crowding. © 2013 by Lippincott Williams and Wilkins.

Objective To investigate differences in self-rated health (SRH) between web and mail questionnaires in a mixed mode survey and to provide a model that explains those differences. Subjects A total of 15,200 mail respondents and 17,829 web respondents from the 2008 US National Health Survey conducted by the Gallup Panel. Research design Respondents were recruited using random digit dialing and assigned to one of the two survey modes (web or mail). Respondents with household Internet connection and frequent Internet usage were invited to complete the survey through the web mode. Respondents who had no Internet connection or who used the Internet infrequently were invited to the mail mode. Thus, respondents with better Internet access used the web mode. Measures Respondents completed a questionnaire that asked about SRH status, objective health conditions, health behaviors, and other socioeconomic variables. Statistical associations were analyzed with ordered Logit and negative binomial models. Results Web respondents reported better SRH than mail respondents. This difference is in part reflective of variability in objective health status between these two groups, and in part attributable to the effects of survey mode. These results maintained with age controlled. Conclusions The alignment between survey mode selection, Internet access, and health disparities, as well as genuine survey mode characteristics, leads to web-mail differences in SRH. Unless the digital divide and its influences on survey mode selection are resolved and differential genuine mode effects are fully comprehended, we recommend that both modes be simultaneously used on a complementary basis.

BACKGROUND As clinical data are increasingly collected and stored electronically, their potential use for comparative effectiveness research (CER) grows. Despite this promise, challenges face those wishing to leverage such data. In this paper we aim to enumerate some of the knowledge management and informatics issues common to such data reuse. DESIGN After reviewing the current state of knowledge regarding biomedical informatics challenges and best practices related to CER, we then present 2 research projects at our institution. We analyze these and highlight several common themes and challenges related to the conduct of CER studies. Finally, we represent these emergent themes. RESULTS The informatics challenges commonly encountered by those conducting CER studies include issues related to data information and knowledge management (eg, data reuse, data preparation) as well as those related to people and organizational issues (eg, sociotechnical factors and organizational factors). Examples of these are described in further detail and a formal framework for describing these findings is presented. CONCLUSIONS Significant challenges face researchers attempting to use often diverse and heterogeneous datasets for CER. These challenges must be understood in order to be dealt with successfully and can often be overcome with the appropriate use of informatics best practices. Many research and policy questions remain to be answered in order to realize the full potential of the increasingly electronic clinical data available for such research.

To document the strengths and challenges of using electronic health records (EHRs) for comparative effectiveness research (CER).A replicated case study of comparative effectiveness in hypertension treatment was conducted across 4 health systems, with instructions to extract data and document problems encountered using a specified list of required data elements. Researchers at each health system documented successes and challenges, and suggested solutions for addressing challenges.Data challenges fell into 5 categories: missing data, erroneous data, uninterpretable data, inconsistencies among providers and over time, and data stored in noncoded text notes. Suggested strategies to address these issues include data validation steps, use of surrogate markers, natural language processing, and statistical techniques.A number of EHR issues can hamper the extraction of valid data for cross-health system comparative effectiveness studies. Our case example cautions against a blind reliance on EHR data as a single definitive data source. Nevertheless, EHR data are superior to administrative or claims data alone, and are cheaper and timelier than clinical trials or manual chart reviews. All 4 participating health systems are pursuing pathways to more effectively use EHR data for CER.A partnership between clinicians, researchers, and information technology specialists is encouraged as a way to capitalize on the wealth of information contained in the EHR. Future developments in both technology and care delivery hold promise for improvement in the ability to use EHR data for CER.

Electronic health information routinely collected during health care delivery and reimbursement can help address the need for evidence about the real-world effectiveness, safety, and quality of medical care. Often, distributed networks that combine information from multiple sources are needed to generate this real-world evidence.We provide a set of field-tested best practices and a set of recommendations for data quality checking for comparative effectiveness research (CER) in distributed data networks.Explore the requirements for data quality checking and describe data quality approaches undertaken by several existing multi-site networks.There are no established standards regarding how to evaluate the quality of electronic health data for CER within distributed networks. Data checks of increasing complexity are often used, ranging from consistency with syntactic rules to evaluation of semantics and consistency within and across sites. Temporal trends within and across sites are widely used, as are checks of each data refresh or update. Rates of specific events and exposures by age group, sex, and month are also common.Secondary use of electronic health data for CER holds promise but is complex, especially in distributed data networks that incorporate periodic data refreshes. The viability of a learning health system is dependent on a robust understanding of the quality, validity, and optimal secondary uses of routinely collected electronic health data within distributed health data networks. Robust data quality checking can strengthen confidence in findings based on distributed data network.

BackgroundProviding culturally competent care shows promise as a mechanism to reduce health care inequalities. Until the recent development of the Consumer Assessment of Healthcare Providers and Systems Cultural Competency Item Set (CAHPS-CC), no measures capturing patient-level experiences with culturally competent care have been suitable for broad-scale administration.MethodsWe performed confirmatory factor analysis and internal consistency reliability analysis of CAHPS-CC among patients with type 2 diabetes (n=600) receiving primary care in safety-net clinics. CAHPS-CC domains were also correlated with global physician ratings.ResultsA 7-factor model demonstrated satisfactory fit (χ²₂₃₁=484.34, P

The complexity of health information often exceeds patients' skills to understand and use it.To develop survey items assessing how well healthcare providers communicate health information.Domains and items for the Consumer Assessment of Healthcare Providers and Systems (CAHPS) Item Set for Addressing Health Literacy were identified through an environmental scan and input from stakeholders. The draft item set was translated into Spanish and pretested in both English and Spanish. The revised item set was field tested with a randomly selected sample of adult patients from 2 sites using mail and telephonic data collection. Item-scale correlations, confirmatory factor analysis, and internal consistency reliability estimates were estimated to assess how well the survey items performed and identify composite measures. Finally, we regressed the CAHPS global rating of the provider item on the CAHPS core communication composite and the new health literacy composites.A total of 601 completed surveys were obtained (52% response rate). Two composite measures were identified: (1) Communication to Improve Health Literacy (16 items); and (2) How Well Providers Communicate About Medicines (6 items). These 2 composites were significantly uniquely associated with the global rating of the provider (communication to improve health literacy: P0.001, b=0.28; and communication about medicines composite: P=0.02, b=0.04). The 2 composites and the CAHPS core communication composite accounted for 51% of the variance in the global rating of the provider. A 5-item subset of the Communication to Improve Health Literacy composite accounted for 90% of the variance of the original 16-item composite.This study provides support for reliability and validity of the CAHPS Item Set for Addressing Health Literacy. These items can serve to assess whether healthcare providers have communicated effectively with their patients and as a tool for quality improvement.

Background:Primary data collection is a critical activity in clinical research. Even with significant advances in technical capabilities, clear benefits of use, and even user preferences for using electronic systems for collecting primary data, paper-based data collection is still common in clinical

Objectives Using data from the national Health and Retirement Study, we sought to: (a) estimate the proportion of the US adults with diabetes above the age of 50 who do not meet physical activity guidelines but believe they are sufficiently active; and (b) examine demographic and health-related correlates of such "overestimation." Research design Respondents who were classified as underactive according to a detailed activity inventory but reported exercising at least the "right amount," were designated as overestimating their physical activity. Multiple logistic regression was used to examine the association of demographic and health-related correlates with the odds of overestimation. Results Fifty-four percent of the survey sample did not meet physical activity guidelines, and one quarter of this underactive group overestimated their physical activity. The adjusted odds of overestimation were higher among respondents who held the perception that they were about the right weight or underweight [odds ratio (OR)=2.42; 95% confidence interval (CI), 1.49-3.94), who had good or better self-assessed diabetes control (OR=1.84; 95% CI, 1.12-3.04), and who were Black or Hispanic (OR=1.89; 95% CI, 1.13-3.16). Experiencing shortness of breath reduced the odds of overestimation (OR=0.34; 95% CI, 0.19-0.61). Conclusions Overestimation of physical activity is common among adults with diabetes, and is associated with the perceptions that one is about the right weight and that one has good control of diabetes, and with being Black or Hispanic. Clinicians should be aware that these factors may affect their patients' beliefs about how much physical activity is adequate.

Background: The patient-centered medical home (PCMH) has become a widely cited solution to the deficiencies in primary care delivery in the United States. To achieve the magnitude of change being called for in primary care, quality improvement interventions must focus on whole-system redesign, and not just isolated parts of medical practices. Methods: Investigators participating in 9 different evaluations of Patient Centered Medical Home implementation shared experiences, methodological strategies, and evaluation challenges for evaluating primary care practice redesign. Results: A year-long iterative process of sharing and reflecting on experiences produced consensus on 7 recommendations for future PCMH evaluations: (1) look critically at models being implemented and identify aspects requiring modification; (2) include embedded qualitative and quantitative data collection to detail the implementation process; (3) capture details concerning how different PCMH components interact with one another over time; (4) understand and describe how and why physician and staff roles do, or do not evolve; (5) identify the effectiveness of individual PCMH components and how they are used; (6) capture how primary care practices interface with other entities such as specialists, hospitals, and referral services; and (7) measure resources required for initiating and sustaining innovations. Conclusions: Broad-based longitudinal, mixed-methods designs that provide for shared learning among practice participants, program implementers, and evaluators are necessary to evaluate the novelty and promise of the PCMH model. All PCMH evaluations should as comprehensive as possible, and at a minimum should include a combination of brief observations and targeted qualitative interviews along with quantitative measures.

Debate over reforming the nation's healthcare system has stimulated a need for health services research (HSR) models that are nationally applicable. Toward this end, the authors identify the Military Health System (MHS) as America's "undiscovered" laboratory for HSR. Although many may confuse the MHS with the Department of Veterans Affairs (VA), the 2 systems vary dramatically with respect to their beneficiary populations, access to care, and other important attributes. In this article, the authors describe key characteristics of the MHS including its large beneficiary base, its direct care operating environment, its dedicated medical school and graduate education programs, and its fully operational integrated health information system. Although a few health systems (eg, Kaiser Permanente, Partners Healthcare, and Department of Veterans Affairs) possess some characteristics, no other has all of these components in place. This article sets the stage for contemporary HSR studies with broad applicability to current issues in American healthcare that could be performed within the MHS. Inclusion of the MHS environment in HSR studies of health services delivery modalities, adoption of health information technology, access to care, relationship of medical education to effective safe care delivery, health disparities, child health, and behavioral health would provide strong underpinnings for proposed changes in American healthcare delivery. Finally, the article highlights current regulatory barriers to research within the MHS whereas suggesting steps to minimize their impact in conducting HSR.

The Distributed Ambulatory Research in Therapeutics Network (DARTNet) is a federated network of electronic health record (EHR) data, designed as a platform for next-generation comparative effectiveness research in real-world settings. DARTNet links information from nonintegrated primary care clinics that use EHRs to deliver ambulatory care to overcome limitations with traditional observational research.Test the ability to conduct a remote, electronic point of care study in DARTNet practices by prompting clinic staff to obtain specific information during a patient encounter.Prospective survey of patients identified through queries of clinical data repositories in federated network organizations. On patient visit, survey is triggered and data are relinked to the EHR, de-identified, and copied for evaluation.Adult patients diagnosed with diabetes mellitus that scheduled a clinic visit for any reason in a 2-week period in DARTNet primary care practices.Survey on hypoglycemic events (past month) and over-the-counter and herbal supplement use.DARTNet facilitated point of care data collection triggered by an electronic prompt for additional information at a patient visit. More than one-third of respondents (33% response rate) reported either mild (45%) or severe hypoglycemic events (5%) in the month before the survey; only 3 of those were also coded using the ICD-9 (a significant difference in detection rates 37% vs. 1%). Nearly one-quarter of patients reported taking an OTC/herbal, 4% specifically for the treatment of symptoms of diabetes.Prospective data collection is feasible in DARTNet and can enable comparative effectiveness and safety research.

Large administrative datasets such as Medicare and Medicaid claims have much potential utility in clinical and comparative effectiveness (CE) studies. Among their advantages are the inclusion of clinically heterogeneous populations, without exclusions typical in clinical trials; the ability to study extremely large study populations with power to examine differential outcomes across individual drugs, treatment effect modification, and the risk of uncommon outcomes. However, claims data by themselves are subject to many limitations, notably, in their lack of information on such clinical characteristics as functional status, behaviors, and symptoms, which are important both as outcomes and as covariates.We describe data from multiple sources including standardized, electronically recorded clinical and functional data from the Nursing Home (NH) Minimum Data Set; prescription drug data from Medicaid and Medicare claims; and facility data. We present the strengths and challenges of using merged data about the NH population to study prescription drug exposures and outcomes in the frail elderly, and suggest strategies to address methodological difficulties.Merged data from NH sources can support unique study designs in CE research and provide great power. However, given the differing longitudinal structure, timing of observations, and other complex features of the underlying data sources, such studies pose many challenging design and analytic issues.Integrated data on the NH population have great potential for CE research among frail elderly persons, if methodological and measurement challenges can be adequately addressed.

Measuring spending on diseases is critical to assessing the value of medical care.To review the current state of cost of illness estimation methods, identifying their strengths, limitations, and uses. We briefly describe the current National Health Expenditure Accounts, and then go on to discuss the addition of cost of illness estimation to the National Health Expenditure Accounts.Recommendations are made for future research aimed at identifying the best methods for developing and using disease-based national health accounts to optimize the information available to policy makers as they struggle with difficult resource allocation decisions.

Objectives: To show the importance of patients’ time as a cost of health and medical care and to explain how to include it in costing studies without greatly increasing the work required for such studies. Background: Despite the decade-old recommendation of the Panel on Cost-Effectiveness in Health and Medicine, patients’ time is rarely included in costing or cost-effectiveness analyses (CEAs). Studies of cancer care, smoking cessation, and diabetes self-management show that it can be a large part of an intervention’s costs, sometimes larger than direct medical costs, and can potentially affect patients’ willingness to undertake the intervention. Measuring and Valuing Time: Good costing practice follows 2 principles: measure all important uses of a resource; and value it appropriately and in a way that is consistent with the valuation of other resources. Counts of formal medical services, already measured in most studies, can serve as the starting point for valuing patients’ time, and would be a major step toward recognizing time costs, even when analysts cannot measure other uses of time. The concept of opportunity cost, often approximated by a market price, is the basis for valuing all resources. The reasons why the wage is a reasonable proxy for the value patients place on their own time are explained. Wage data are well measured and readily available. Conclusions: Ignoring patients’ time underestimates disease burden and biases cost-effectiveness results toward interventions that use more time. The tools and data to include patients’ time are available and will improve if they are routinely used.

this issue, Birkmeyer et al ask 2 new questions regarding surgical quality variations that exist between hospitals.1 First, when it comes to lower quality surgical care, are the poor disproportionately affected by it? Second, are the gaps in quality between hospitals sufficient to explain the overall outcome gap that exists between poor and more affluent patients? Their results seem to answer both questions affirmatively. The authors analyzed Medicare data from 1999 to 2003 to generate their findings. They assessed near term death rates of Medicare beneficiaries after 6 major surgical procedures. A subject was considered to have died if he or she died within 30 days of the surgery, or died at any point during the surgical admission (ie, a death occurring past 30 days for a patient who had never been discharged). Socioeconomic status (SES) of the treated Medicare beneficiaries was determined from 2000 US Census data, using a composite variable based on zone improvement plan (ZIP) code measures of income, education, and housing. In unadjusted analyses, stratified analyses, and multi variable logistic regression analyses adjusting for patient characteristics such as age, sex, and comorbidity, the authors found that lower SES was strongly associated with a higher risk of postoperative death for all 6 procedures. In other words, in answer to the authors' first question, a Medicare beneficiary's chance of dying after surgery is higher if he or she is poor. To address their second question, they statistically explored the link between SES and outcome within and between individual hospitals, further stratifying the hospitals based on the SES of the Medicare patients they treated. Through these analyses, they found evidence that the relation between patient SES and surgical outcome is mediated by differences in the quality of care different hospitals provide. This was because the poor were more likely to be treated at lower quality of care hospitals, explaining, to a fairly large degree, why they have poorer outcomes. The actual extent to which the outcome gap is explained by the hospital quality gap is explored in 3 types of analyses. First, the authors present the features of hospitals that treat low and high SES patients. These findings show that relatively impoverished beneficiaries are less likely to receive care in either "high-volume hospitals," or in hospitals categorized as "teaching hospitals." In both cases, one would expect that this pattern would worsen the outcomes for poorer beneficiaries an abundant literature supports the link between both "high volume" and being a "teaching hospital" and delivering better outcomes.2"5 Next, they present "within stratum" analyses, which convincingly demonstrate that the wealth gap in outcomes is far less apparent within hospitals than between hospitals. For instance, there is a large absolute mortality gap, totaling 3%, between low and high SES Medicare patients undergoing gastrectomy in the United States. Yet, there is no such gap between the 2 groups of patients within hospital stratum. Outcomes are equally poor for the low and high SES patients treated in the hospitals that mostly serve the poor, and quite a bit better (and similar) in hospitals that serve mostly wealthier patients. Consistent with this finding, in analyses controlling for hospital "fixed effects," in which the hospital's ambient quality becomes a predictor of outcome, the association between patient SES and outcome is either reduced or eliminated for all 6 procedures examined.

Background: Fewer than half of Americans have been screened for colorectal cancer (CRC), a largely preventable disease. Methods: All physician members (n = 1030) of the Iowa Academy of Family Physicians were mailed a 3-page investigator-developed survey about their attitudes, barriers, and practices regarding CRC screening. Results: The usable response rate was 29%. Forty-three percent practiced in rural settings. Ninety-five percent felt that they were well informed about American Cancer Society guidelines and 90% tried to follow the guidelines. Most doctors (88%) disagreed with the statement that there was "no time to adequately discuss screening," but they would like more time to discuss screening. Only 40% felt their medical records were organized to easily determine screening status, 40% encouraged office staff to participate in screening, and 16% had a written policy regarding CRC screening. Physicians estimated that they recommend screening to 78% of their patients and that 54% of their patients were actually up-to-date. Discussion of CRC screening was strongly dependent on visit type, with physicians estimating that CRC screening is discussed at 11% of acute visits, 42% of chronic visits, and 87% of health maintenance visits. Several office system factors were associated with a recommendation for screening in a multivariable linear regression model (R 2 = 0.33). Conclusions: Although nearly all physicians felt that they were well informed about American Cancer Society guidelines and tried to follow guidelines for CRC screening, few had office systems to facilitate screening. Physicians would like more time to discuss screening. Office systems likely have the most potential to improve CRC screening among patients attending primary care practices.