Scott T. Shimotsu

Capturing the Spectrum of Household Food and Beverage Purchasing Behavior: A Review

The household setting may be the most important level at which to understand the food choices of individuals and how healthful food choices can be promoted. However, there are few available measures of the food purchase behaviors of households and little consensus on the best way to measure it. This review explores the currently available measures of household food purchasing behavior. Three main measures are described, evaluated, and compared: home food inventories, food and beverage purchase records and receipts, and Universal Product Code bar code scanning. The development of coding, aggregation, and analytical methods for these measures of household food purchasing behavior is described. Currently, annotated receipts and records are the most comprehensive, detailed measure of household food purchasing behavior, and are feasible for population-based samples. Universal Product Code scanning is not recommended due to its cost and complexity. Research directions to improve household food purchasing behavior measures are discussed.

Hennepin Health: A Safety-Net Accountable Care Organization For The Expanded Medicaid Population

Health care payment and delivery models that challenge providers to be accountable for outcomes have fueled interest in community-level partnerships that address the behavioral, social, and economic determinants of health. We describe how Hennepin Health--a county-based safety-net accountable care organization in Minnesota--has forged such a partnership to redesign the health care workforce and improve the coordination of the physical, behavioral, social, and economic dimensions of care for an expanded community of Medicaid beneficiaries. Early outcomes suggest that the program has had an impact in shifting care from hospitals to outpatient settings. For example, emergency department visits decreased 9.1 percent between 2012 and 2013, while outpatient visits increased 3.3 percent. An increasing percentage of patients have received diabetes, vascular, and asthma care at optimal levels. At the same time, Hennepin Health has realized savings and reinvested them in future improvements. Hennepin Health offers lessons for counties, states, and public hospitals grappling with the problem of how to make the best use of public funds in serving expanded Medicaid populations and other communities with high needs.

Annotated receipts capture household food purchases from a broad range of sources

BackgroundAccurate measurement of household food purchase behavior (HFPB) is important for understanding its association with household characteristics, individual dietary intake and neighborhood food retail outlets. However, little research has been done to develop measures of HFPB. The main objective of this paper is to describe the development of a measure of HFPB using annotated food purchase receipts.MethodsHouseholds collected and annotated food purchase receipts for a four-week period as part of the baseline assessment of a household nutrition intervention. Receipts were collected from all food sources, including grocery stores and restaurants. Households (n = 90) were recruited from the community as part of an obesity prevention intervention conducted in 2007–2008 in Minneapolis, Minnesota, USA. Household primary shoppers were trained to follow a standardized receipt collection and annotation protocol. Annotated receipts were mailed weekly to research staff. Staff coded the receipt data and entered it into a database. Total food dollars, proportion of food dollars, and ounces of food purchased were examined for different food sources and food categories. Descriptive statistics and correlations are presented.ResultsA total of 2,483 receipts were returned by 90 households. Home sources comprised 45% of receipts and eating-out sources 55%. Eating-out entrees were proportionally the largest single food category based on counts (16.6%) and dollars (

Worksite environment physical activity and healthy food choices: measurement of the worksite food and physical activity environment at four metropolitan bus garages.

106 per month). Two-week expenditures were highly correlated (r = 0.83) with four-week expenditures.ConclusionReceipt data provided important quantitative information about HFPB from a wide range of sources and food categories. Two weeks may be adequate to reliably characterize HFPB using annotated receipts.

Adopting local alcohol policies: a case study of community efforts to regulate malt liquor sales

BackgroundThe present research describes a measure of the worksite environment for food, physical activity and weight management. The worksite environment measure (WEM instrument) was developed for the Route H Study, a worksite environmental intervention for weight gain prevention in four metro transit bus garages in Minneapolis-St. Paul.MethodsTwo trained raters visited each of the four bus garages and independently completed the WEM. Food, physical activity and weight management-related items were observed and recorded on a structured form. Inter-rater reliability was computed at the item level using a simple percentage agreement.ResultsThe WEM showed high inter-rater reliability for the number and presence of food-related items. All garages had vending machines, microwaves and refrigerators. Assessment of the physical activity environment yielded similar reliability for the number and presence/absence of fitness items. Each garage had a fitness room (average of 4.3 items of fitness equipment). All garages had at least one stationary bike and treadmill. Three garages had at least one weighing scale available. There were no designated walking areas inside or outside. There were on average < 1 food stores or restaurants within sight of each garage. Few vending machine food and beverage items met criteria for healthful choices (15% of the vending machine foods; 26% of the vending machine beverages). The garage environment was perceived to be not supportive of healthy food choices, physical activity and weight management; 52% reported that it was hard to get fruits and vegetables in the garages, and 62% agreed that it was hard to be physically active in the garages.ConclusionThe WEM is a reliable measure of the worksite nutrition, physical activity, and weight management environment that can be used to assess changes in the work environment.

Neighborhood socioeconomic characteristics, the retail environment, and alcohol consumption: A multilevel analysis

Purpose. To learn how the local context may affect a citys ability to regulate alcohol products such as high–alcohol-content malt liquor, a beverage associated with heavy drinking and a spectrum of nuisance crimes in urban areas. Approach. An exploratory, qualitative case study comparing cities that adopted policies to restrict malt liquor sales with cities that considered, but did not adopt policies. Setting. Nine large U.S. cities in seven states. Participants. City legislators and staff, alcohol enforcement personnel, police, neighborhood groups, business associations, alcohol retailers, and industry representatives. Method. Qualitative data were obtained from key informant interviews (n = 56) and media articles (n = 360). The data were coded and categorized. Similarities and differences in major themes among and across Adopted and Considered cities were identified. Results. Cities faced multiple barriers in addressing malt liquor–related problems, including a lack of enforcement tools, alcohol industry opposition, and a lack of public and political will for alcohol control. Compared to cities that did not adopt malt liquor sales restrictions, cities that adopted restrictions appeared to have a stronger public mandate for a policy and were less influenced by alcohol industry opposition and lack of legislative authority for alcohol control. Strategies common to successful policymaking efforts are discussed. Conclusion. Understanding the local context may be a critical step in winning support for local alcohol control policies.

Medicaid expansion and mental health: A Minnesota case study.

BACKGROUND The neighborhoods where people live can influence their drinking behavior. We hypothesized that living in a neighborhood with lower median income, higher alcohol outlet density, and only liquor stores and no grocery stores would be associated with higher alcohol consumption after adjusting for individual demographic and lifestyle factors. METHODS We used two self-report measures to assess alcohol consumption in a sample of 9959 adults living in a large Midwestern county: volume of alcohol consumed (count) and binge drinking (5 or more drinks vs.<5 drinks). We measured census tract median annual household income based on U.S. Census data. Alcohol outlet density was measured using the number of liquor stores divided by the census tract roadway miles. The mix of liquor and food stores in census tracts was assessed using a categorical variable based on the number of liquor and number of food stores using data from InfoUSA. Weighted hierarchical linear and Poisson regression were used to test our study hypothesis. RESULTS Retail mix was associated with binge drinking. Individuals living in census tracts with only liquor stores had a 46% higher risk of binge drinking than individuals living in census tracts with food stores only after controlling for demographic and lifestyle factors. CONCLUSION Census tract characteristics such as retail mix may partly explain variability in drinking behavior. Future research should explore the mix of stores, not just the over-concentration of liquor stores in census tracts.

Increased Likelihood of Missed Appointments (“No Shows”) for Racial/Ethnic Minorities in a Safety Net Health System

INTRODUCTION The health status and psychosocial needs of the Medicaid expansion population have been estimated but not measured. This population includes childless adults predicted to have high rates of mental illness, especially among the homeless. Given limitations in access to mental health services, it is unclear how prepared the U.S. health care system is to care for the needs of the expansion population. METHOD Using enrollment and claims data from the Minnesota Department of Human Services, this study presents prevalence rates of mental illness diagnoses and measures of unstable housing in Minnesotas childless-adult early Medicaid expansion population. Rates are compared with prior predictions of serious psychological distress and mental illness constructed from the National Survey on Drug Use and Health (NSDUH) using χ2 and t tests. RESULTS Diagnoses of mental illness in Minnesotas childless-adult early Medicaid expansion population were more than 15% higher than prevalence measures of mental illness/distress for the current Medicaid population. Diagnosis rates fell within confidence intervals of estimates of mental illness for Minnesotas Medicaid expansion population. Almost 1 in 3 enrollees had a marker of unstable housing; of this group, half had mental illness and/or distress. DISCUSSION Findings support predictions of the high burden of mental illness and unstable housing among the Medicaid expansion population. Minnesota offers lessons to other regions working to care for such populations: (a) the use of flexible financing structures to build integrated care systems and (b) passage of legislation to allow data sharing among mental health, social services, and medical care.

The Relationships Among Socioeconomic Status, Fruit and Vegetable Intake, and Alcohol Consumption

Missed appointments have been linked to adverse outcomes known to affect racial/ethnic minorities. However, the association of missed appointments with race/ethnicity has not been determined. We sought to determine the relationships between race/ethnicity and missed appointments by performing a cross-sectional study of 161 350 patients in a safety net health system. Several race/ethnicity categories were significantly associated with missed appointment rates, including Hispanic/Latino patients, American Indian/Alaskan Native patients, and Black/African American patients, as compared with White non-Hispanic patients. Other significant predictors included Mexico as country of origin, medical complexity, and major mental illness. We recommend additional research to determine which interventions best reduce missed appointments for minority populations in order to improve the care of vulnerable patients.

Capturing the Spectrum of Household Food and Beverage Purchasing Behavior

Purpose. We examined the relationships among fruit and vegetable intake, alcohol consumption, and socioeconomic status (SES). We hypothesized that fruit and vegetable consumption would be inversely associated with alcohol consumption and the relationship would differ by SES. Design. A cross-sectional analysis. Setting. Large, urban Midwestern county. Subjects. A unique, racially/ethnically diverse sample of 9959 adults (response rate: 66.3%). Measures. Fruit and vegetable intake was measured using two items that assessed servings per day. Alcohol consumption was measured in terms of volume of alcohol consumed and binge drinking. Individual measures of SES included education and household income. Analysis. Weighted multivariate linear and Poisson regression were used to estimate effects. Results. The relationship between fruit and vegetable intake and alcohol consumption varied by SES. Those with lower household incomes who consumed five or more servings of fruits and vegetables per day were less likely to engage in binge drinking relative to those consuming zero to one servings of fruits and vegetables per day (risk ratio = .66; 95% confidence interval: .46, .95). No association was observed for higher–household-income individuals. Conclusion. We observed an inverse relationship between fruit and vegetable consumption and alcohol intake in those with lower household incomes but not in those with higher household incomes. Results suggest that the relationship between diet and alcohol consumption may be more relevant in populations with more restricted economic choices. Results are, however, based on cross-sectional data.