Testing a Personal Narrative for Persuading People to Value and Use Comparative Physician Quality of Care Information: An Experimental Study.

Objective: This study tests whether a personal narrative can persuade people to value comparative data on physician quality. Method: We conducted an online experiment with 850 adults. One group viewed a cartoon narrative on physician quality variation, another saw text on physician quality variation, and there was a control group. Study participants hypothetically selected a physician from a display of four physicians. The top-quality physician was furthest away and most expensive. We conducted multivariate models examining the relationship between experimental group and choice of the top-quality physician. Results: There was no overall relationship between narrative or text information and choice of the highest quality physician. Among higher numerate participants, however, those who viewed the narrative had odds 2.7 times higher of selecting the top-quality physician compared with the control group. Discussion: Personal narratives can persuade higher numerate people to consider quality when selecting physicians.

Does the Race/Ethnicity or Gender of a Physician's Name Impact Patient Selection of the Physician?

This study's objective was to examine the extent to which individuals exhibit a preference for physicians based upon the race/ethnicity and gender of a physician's name. We conducted an online survey of 915 adults, who viewed a comparative display of four physicians' quality performance. We randomized the name of one physician, whose quality performance was equal to that of one physician and better than two other physicians, to be either typically African American male, African American female, white male, white female, or Middle Eastern (gender ambiguous). In regression models, participants more frequently selected the physician with the randomized name when displayed with a white male name, compared to when presented with an African American male, African American female, or Middle Eastern name (ORs ranging from .59 to .64). White and male study participants exhibited this pattern, while racial/ethnic minority participants did not. If the hypothetical choice bias observed here translates to people's actual selection of physicians, it could be a contributing factor for why women and racial/ethnic minority physicians have lower incomes than white male physicians.

Prenatal Development and Adolescent Obesity: Two Distinct Pathways to Diabetes in Adulthood.

OBJECTIVE: Higher body-mass index (BMI) and lower birth weight (BW) are associated with elevated risk of diabetes in adulthood, but the extent to which they compose two distinct pathways is unclear. METHODS: We used data from the National Longitudinal Study of Adolescent to Adult Health, a cohort of adolescents (1994-1995) followed for 14 years over four waves into adulthood (n = 13,413). Sex-stratified path analysis was used to examine pathways from BW [kg; linear (BW) and quadratic (BW2)] to latent trajectories in BMI from adolescence to adulthood to prevalent diabetes or prediabetes (pre/diabetes) in adulthood, adjusting for sociodemographic characteristics. RESULTS: Two pathways from BW to pre/diabetes were characterized: one from higher BW to elevated BMI and pre/diabetes and a second from lower BW, independent of BMI. In the BMI-independent pathway, greater BW was associated with marginally lower odds of pre/diabetes in women, but not men. Girls born at lower and higher BW exhibited elevated BMI in adolescence [coeff (95% CI): BW: -2.1 (-4.1, -0.05); BW2: 0.43 (0.09, 0.76)]; higher BW predicted marginally faster BMI gain and higher adolescent BMI and faster BMI gain were associated with pre/diabetes [coeff (95% CI): BMI intercept: 0.09 (0.06, 0.11); BMI slope: 0.11 (0.07, 0.15)]. In boys, BW was weakly associated with BMI intercept and slope; BMI slope, but not BMI intercept, was positively associated with pre/diabetes [coeff (95% CI): 0.29 (0.19, 0.39)]. CONCLUSIONS: Findings suggest that in girls, slowing BMI gain is critical for diabetes prevention, yet it may not address distinct pathology stemming from early life.

Presenting Cost and Efficiency Measures That Support Consumers to Make High-Value Health Care Choices.

OBJECTIVE: To identify approaches to presenting cost and resource use measures that support consumers in selecting high-value hospitals. DATA SOURCES: Survey data were collected from U.S. employees of Analog Devices (n = 420). STUDY DESIGN: In two online experiments, participants viewed comparative data on four hospitals. In one experiment, participants were randomized to view one of five versions of the same comparative cost data, and in the other experiment they viewed different versions of the same readmissions data. Bivariate and multivariate analyses examined whether presentation approach was related to selecting the high-value hospital. PRINCIPAL FINDINGS: Consumers were approximately 16 percentage points more likely to select a high-value hospital when cost data were presented using actual dollar amounts or using the word "affordable" to describe low-cost hospitals, compared to when the Hospital Compare spending ratio was used. Consumers were 33 points more likely to select the highest performing hospital when readmission performance was shown using word icons rather than percentages. CONCLUSIONS: Presenting cost and resource use measures effectively to consumers is challenging. This study suggests using actual dollar amounts for cost, but presenting performance on readmissions using evaluative symbols.

High birth weight modifies association between adolescent physical activity and cardiometabolic health in women and not men.

Recent evidence suggests that adverse prenatal development alters physiological response to physical activity, but longitudinal epidemiologic evidence is scant. This study tested the hypothesis that lower physical activity during adolescence and young adulthood is more strongly associated with later cardiovascular disease (CVD) risk and diabetes or prediabetes (DM/PDM) in women and men who were born with high or low birth weight (HBW, LBW), compared to normal birth weight (NBW). We analyzed data from the National Longitudinal Study of Adolescent to Adult Health, a cohort study of US adolescents followed into adulthood (1994-2009). Using sex-stratified multivariable regression, 30-year CVD risk score (calculated using objective measures; n=12,775) and prevalent DM/PDM (n=15,138) at 24-32years of age were each modeled as a function of birth weight category, self-reported moderate-to-vigorous physical activity frequency in adolescence (MVPA1) and young adulthood (MVPA3), and MVPA-birth weight interactions. Greater MVPA1 was associated with lower 30-year CVD risk score and DM/PDM risk in HBW women but not NBW or LBW women. Associations between MVPA1 and 30-year CVD risk or DM/PDM were not modified by HBW in men; or by LBW in women or men. Additionally, birth weight did not modify estimated effects of MVPA3. Findings suggest that frequent MVPA in adolescence may be a particularly important cardiometabolic risk reduction strategy in girls born HBW; however, we found no evidence that birth weight and MVPA interact in cardiometabolic disease risk in men, for MVPA in adulthood, or for LBW.

Childhood Maltreatment and BMI Trajectory: The Mediating Role of Depression.

INTRODUCTION: Childhood maltreatment is associated with later obesity, but the underlying mechanisms are unknown. The objective of this study was to estimate the extent to which depression mediates the associations between childhood maltreatment and BMI in adolescence through adulthood. METHODS: Data on a cohort of 13,362 adolescents in the National Longitudinal Study of Adolescent to Adult Health (Wave I [1994-1995] to Wave IV [2008-2009]) were analyzed in 2015-2016. Classes of maltreatment experienced prior to age 12 years were statistically identified using latent class analysis. Gender-stratified latent growth curve analysis was used to estimate total effects of maltreatment classes on latent BMI trajectory (aged 13-31 years) and indirect effects of maltreatment classes that occurred through latent depression trajectory (aged 12-31 years). RESULTS: Four latent maltreatment classes were identified: high abuse and neglect; physical abuse dominant; supervisory neglect dominant; and no/low maltreatment. In girls, compared with no/low maltreatment, supervisory neglect dominant (coefficient=0.3, 95% CI=0.0, 0.7) and physical abuse dominant (coefficient=0.6, 95% CI=0.1, 1.2) maltreatment were associated with faster gain in BMI. Change in depression over time fully mediated the association of BMI slope with physical abuse dominant maltreatment, but not with supervisory neglect dominant maltreatment. In boys, high abuse and neglect maltreatment was associated with marginally greater BMI at baseline (coefficient=0.7, 95% CI= -0.1, 1.5); this association was not mediated by depression. CONCLUSIONS: Although maltreatment was associated with depression and BMI trajectories from adolescence to adulthood, depression only mediated associations with physical abuse dominant maltreatment in girls.

Does patient activation predict the course of type 2 diabetes? A longitudinal study.

OBJECTIVE: To examine whether patient activation is predictive of the course of diabetes over a three year period among patients with and without diabetes. METHODS: Longitudinal analyses utilized electronic health record data from 2011 to 2014. We examined how the patient activation measure (PAM) was predictive of 2014 diabetes-related outcomes among patients with diabetes (n=10,071); pre-diabetes (n=1804); and neither diabetes nor pre-diabetes (n=46,402). Outcomes were clinical indicators (blood pressure, cholesterol, and trigylcerides), costly utilization, and progression from no diabetes to pre-diabetes or diabetes. RESULTS: Higher PAM level predicted better clinical indicator control in patients with diabetes. In patients with pre-diabetes, PAM level predicted better clinical indicator control, and those in the highest level of PAM in 2011 had lower odds of having a hospitalization compared to those in the lowest level. In patients without diabetes or pre-diabetes in 2011, higher PAM level was associated with lower odds of developing pre-diabetes. CONCLUSIONS: More activated patients with diabetes and pre-diabetes had better outcomes than less activated patients. More activated patients without diabetes or pre-diabetes were less likely to develop pre-diabetes over a three year period. PRACTICE IMPLICATIONS: Strategies to improve patient activation may be useful to help curb the diabetes epidemic.

Improving Population Health Management Strategies: Identifying Patients Who Are More Likely to Be Users of Avoidable Costly Care and Those More Likely to Develop a New Chronic Disease.

OBJECTIVE: To explore using the Patient Activation Measure (PAM) for identifying patients more likely to have ambulatory care-sensitive (ACS) utilization and future increases in chronic disease. DATA SOURCES: Secondary data are extracted from the electronic health record of a large accountable care organization. STUDY DESIGN: This is a retrospective cohort design. The key predictor variable, PAM score, is measured in 2011, and is used to predict outcomes in 2012-2014. Outcomes include ACS utilization and the likelihood of a new chronic disease. DATA: Our sample of 98,142 adult patients was drawn from primary care clinic users. To be included, patients had to have a PAM score in 2011 and at least one clinic visit in each of the three subsequent years. PRINCIPAL FINDINGS: PAM level is a significant predictor of ACS utilization. Less activated patients had significantly higher odds of ACS utilization compared to those with high PAM scores. Similarly, patients with low PAM scores were more likely to have a new chronic disease diagnosis over each of the years of observation. CONCLUSIONS: Assessing patient activation may help to identify patients who could benefit from greater support. Such an approach may help ACOs reach population health management goals.

How much do clinicians support patient self-management? The development of a measure to assess clinician self-management support.

BACKGROUND: Primary care provider (PCP) support of patient self-management may be important mechanism to improving patient health outcomes. In this paper we develop a PCP-reported measure of clinician strategies for supporting patient self-management, and we psychometrically test and validate the measure. METHODS: We developed survey items based upon effective self-management support strategies identified in a prior mixed methods study. We fielded a survey in the fall of 2014 with 139 Fairview Health Services PCPs, and conducted exploratory factor analysis and Cronbach's Alpha to test for scale reliability. To validate the measure, we examined the Self-Management Support (SMS) scale's relationship to survey items on self-management support, as well as clinicians' patient panel rates of smoking cessation and weight loss. RESULTS: Nine survey items clustered reliably to create a single factor (Cronbach's Alpha=0.73). SMS scores ranged from 2.1 to 4.9. The SMS was related to each of the validation variables. PCPs who reported spending 60% percent or more of their time counseling, educating, and coaching patients had a mean SMS score of 4.0, while those who reported spending less than 30% of their time doing so had mean SMS scores 15% lower. PCPs' SMS scores exhibited significant but modest associations with their patients' smoking cessation and weight loss (among obese patients) (r=0.21 and r=0.13 respectively). CONCLUSIONS: This study develops and tests a promising measure of PCPs' strategies to support patient self-management. It highlights variation across PCPs. Future work should examine whether increasing scores of PCPs low on the SMS improves chronic care quality outcomes.

Summarized Costs, Placement Of Quality Stars, And Other Online Displays Can Help Consumers Select High-Value Health Plans.

Starting in 2017, all state and federal health insurance exchanges will present quality data on health plans in addition to cost information. We analyzed variations in the current design of information on state exchanges to identify presentation approaches that encourage consumers to take quality as well as cost into account when selecting a health plan. Using an online sample of 1,025 adults, we randomly assigned participants to view the same comparative information on health plans, displayed in different ways. We found that consumers were much more likely to select a high-value plan when cost information was summarized instead of detailed, when quality stars were displayed adjacent to cost information, when consumers understood that quality stars signified the quality of medical care, and when high-value plans were highlighted with a check mark or blue ribbon. These approaches, which were equally effective for participants with higher and lower numeracy, can inform the development of future displays of plan information in the exchanges.

Mental health care among low-income pregnant women with depressive symptoms: facilitators and barriers to care access and the effectiveness of financial incentives for increasing care.

Access to mental health care is suboptimal for low-income pregnant women. Using in-depth interviews, we examined barriers and facilitators to accessing care among 42 low income pregnant women with depressive symptoms. To pilot whether financial incentives would increase utilization during pregnancy, half the women were randomized to receive $10 gift cards after mental health visits. Women reported external and internal barriers to accessing mental health care, and internal and interpersonal facilitators. Financial incentives did not impact how often the women visited mental health providers, suggesting that small incentives are not sufficient to catalyze mental health care use for this population.

How well do patient activation scores predict depression outcomes one year later?

BACKGROUND: This study examines the relationship between patient activation, a measure of individuals׳ knowledge, skill, and confidence for managing their health, and rates of depression remission and response among patients with depression. METHODS: Patients from Fairview Health Services in Minnesota with moderate to severe depression in 2011 and a PHQ-9 score in 2012 were included in the analysis (n=5253). Patient activation in 2011 and other health and demographic features were extracted from the electronic health record. We examined how patient activation predicted depression remission and response rates and changes in depression severity over one year using regression models. We also explored how activation predicted healthy behaviors among depressed patients. RESULTS: Higher baseline patient activation predicted lower depression severity and higher depression remission and response rates a year later. The most activated patients had PHQ-9 scores in 2012 two points lower than the lowest activated patients, and they had twice the odds of remission. Activation also predicted increase in healthy behaviors. LIMITATIONS: We were unable to examine the use of mental health services or control for the number of prior depressive episodes and duration of the current depressive episode in the analysis. CONCLUSIONS: We found that higher patient activation predicted better depression outcomes. While we are unable to explore the mechanism of this association, we observed that more activated patients are also engaged in more healthy behaviors, suggesting that the mechanism may be behavioral. Support of patient activation may be an effective approach for providers to reduce patients׳ depression severity.

The impact of tobacco dependence treatment coverage and copayments in Medicaid.

BACKGROUND: The Affordable Care Act (ACA) expands Medicaid's tobacco dependence treatment (TDT) coverage; however, these expansions differ in comprehensiveness based on Medicaid eligibility category. PURPOSE: To examine whether more generous Medicaid TDT coverage (in terms of cost-sharing requirements and treatments covered) is associated with greater likelihood of quit attempts and successful quit rates. METHODS: This study used repeated cross-sections from the Current Population Survey (2001-2011), linked to state-level survey data on Medicaid TDT coverage. The sample included 3,071 adult Medicaid recipients who reported smoking 12 months prior to the survey and resided in 28 states with consistent TDT coverage across Medicaid fee-for-service and managed care. Logistic regression models, conducted in October 2013, examined the relationship between state TDT coverage and Medicaid recipients' successful quits and attempted quit rates, controlling for individual and state characteristics. RESULTS: Forty-one percent of Medicaid recipients attempted to quit smoking in the prior year and 7% quit successfully. Medicaid recipients in states with the most generous coverage (counseling without copayment and pharmacotherapy with copayment) had the highest predicted successful quit rates (8.3%). Those living in states with no TDT or pharmacotherapy-only coverage had lower predicted successful quit rates (range=4.0%-5.6%). CONCLUSIONS: These findings suggest that the ACA will increase smoking quit rates among Medicaid recipients. Recipients who have more generous TDT coverage (such as the new Medicaid expansion population and pregnant women) will likely see greater increases in quit rates compared to existing adult Medicaid enrollees.