Social and Medical Determinants of Diabetes: A Time-Constrained Multiple Mediator Analysis.

Background A number of studies have shown an association between social determinants of health and the emergence of obesity and diabetes, but whether the relationship is causal is not clear. Objective To test whether social, environmental, and medical determinants directly or indirectly affect population-level diabetes prevalence after controlling for mediator-mediator interactions. Methods Data were obtained from the CDC and supplemented with nine other data sources for 3,109 US counties. The dependent variable was the prevalence of diabetes in 2017. Independent variables were a given county's 30 social, environmental, and medical characteristics in 2015 and 2016. A network multiple mediation analysis was conducted. First, we used Least Absolute Shrinkage and Selection Operator (LASSO) regression to relate the 2017 diabetes rate in each county to 30 predictors measured in 2016, identifying statistically significant and robust predictors as the mediators within the network model and as direct determinants of 2017 diabetes. Second, each of the direct causes of diabetes was taken as a new response variable and LASSO-regressed on the same 30 independent variables measured in 2015, identifying the indirect (mediated) causes of diabetes. Subsequently, these direct and indirect predictors were used to construct a network model. The completed network was then employed to estimate the direct and mediated impact of variables on diabetes. Results For 2017 diabetes rates, 63% of the variation was explained by five variables measured in 2016: the percentage of residents who were (1) obese, (2) African American, (3) physically inactive, (4) in poor health condition, and (5) had a history of diabetes. These five direct predictors, measured in 2016, mediated the effect of indirect variables measured in 2015, including the percentage of residents who were (1) Hispanic, (2) physically distressed, (3) smokers, (4) living with children in poverty, (5) experiencing limited access to healthy foods, and (6) had low income. Conclusion All of the direct predictors of diabetes prevalence, except the percentage of residents who were African American, were medical conditions potentially influenced by lifestyles. Counties characterized by higher levels of obesity, inactivity, and poor health conditions exhibited increased diabetes rates in the following year. The impact of social determinants of illness, such as low income, children in poverty, and limited access to healthy foods, had an indirect effect on the health of residents and, consequently, increased the prevalence of diabetes.

Emerging approaches to multiple chronic condition assessment.

Older adults experience a higher prevalence of multiple chronic conditions (MCCs). Establishing the presence and pattern of MCCs in individuals or populations is important for healthcare delivery, research, and policy. This report describes four emerging approaches and discusses their potential applications for enhancing assessment, treatment, and policy for the aging population. The National Institutes of Health convened a 2-day panel workshop of experts in 2018. Four emerging models were identified by the panel, including classification and regression tree (CART), qualifying comorbidity sets (QCS), the multimorbidity index (MMI), and the application of omics to network medicine. Future research into models of multiple chronic condition assessment may improve understanding of the epidemiology, diagnosis, and treatment of older persons.

Constructing Causal Networks Through Regressions: A Tutorial.

BACKGROUND: Significant progress has been made in the practice of conducting causal analysis using network models. Despite this progress, there is limited evidence that hospital risk managers are using these analytical models. OBJECTIVE: This article introduces the causal network, its related concepts, and methods of analysis. The article demonstrates how hospital risk managers can use existing regression software to construct a causal network and identify root causes of an adverse event. METHODS: Causal networks depict cause and effect in a set of variables. In this context, causes are strong correlations that meet 3 additional criteria: (1) causes occur prior to effects, (2) there is an articulated mechanism for how causes lead to effects, and (3) the association between cause and effect is not spurious, meaning the association persists even after other variables are statistically controlled for (a method of analysis called counterfactual). A causal network can be constructed through repeated use of least absolute shrinkage and selection operator (LASSO) regression. In the proposed regressions, the response variable is any variable in the data. The independent variables are variables that occur prior to the response variable. By design, the statistically significant coefficients in the time-constrained LASSO regressions identify "direct" causes of the response variables. When direct causes of all variables are identified, then the entire network model, including root causes, has been specified. In the final step, the parameters of the network model (ie, strength of causal associations) are estimated by fitting the network structure to the available data. We demonstrate these concepts through fitting a network model to simulated data for causes of excessive boarding in emergency departments. RESULTS: The network (involving 12 causes, over 4 periods, and 1 sentinel event) was accurately recovered from the simulated case reports. The recovered network did not differ from the original network used to simulate the data in any of the 156 possible links. The recovered network allowed the identification of root and direct causes. It showed that hospital occupancy rate, and not emergency department efficiency, was root cause of excessive emergency department boarding. DISCUSSION: Causal networks can provide insights into root, and direct, causes of an adverse event. These models provide empirical tests of causes of adverse events. We encourage the use of these methods by hospital risk managers.

Relative accuracy of social and medical determinants of suicide in electronic health records.

OBJECTIVE: This paper compares the accuracy of predicting suicide from Social Determinants of Health (SDoH) or history of illness. POPULATION STUDIED: 5 313 965 Veterans who at least had two primary care visits between 2008 and 2016. STUDY DESIGN: The dependent variable was suicide or intentional self-injury. The independent variables were 10 495 International Classification of Disease (ICD) Version 9 codes, age, and gender. The ICD codes included 40 V-codes used for measuring SDoH, such as family disruption, family history of substance abuse, lack of education, legal impediments, social isolation, unemployment, and homelessness. The sample was randomly divided into training (90 percent) and validation (10 percent) sets. Area under the receiver operating characteristic (AROC) was used to measure accuracy of predictions in the validation set. PRINCIPAL FINDINGS: Separate analyses were done for inpatient and outpatient codes; the results were similar. In the hospitalized group, the mean age was 67.2 years, and 92.1 percent were male. The mean number of medical diagnostic codes during the study period was 37; and 12.9 percent had at least one SDoH V-code. At least one episode of suicide or intentional self-injury occurred in 1.89 percent of cases. SDoH V-codes, on average, elevated the risk of suicide or intentional self-injury by 24-fold (ranging from 4- to 86-fold). An index of 40 SDoH codes predicted suicide or intentional self-injury with an AROC of 0.64. An index of 10 445 medical diagnoses, without SDoH V-codes, had AROC of 0.77. The combined SDoH and medical diagnoses codes also had AROC of 0.77. CONCLUSION: In predicting suicide or intentional self-harm, SDoH V-codes add negligible information beyond what is already available in medical diagnosis codes. IMPLICATIONS FOR PRACTICE: Policies that affect SDoH (eg, housing policies, resilience training) may not have an impact on suicide rates, if they do not change the underlying medical causes of SDoH.

Covariate Balancing through Naturally Occurring Strata.

OBJECTIVE: To provide an alternative to propensity scoring (PS) for the common situation where there are interacting covariates. SETTING: We used 1.3 million assessments of residents of the United States Veterans Affairs nursing homes, collected from January 1, 2000, through October 9, 2012. DESIGN: In stratified covariate balancing (SCB), data are divided into naturally occurring strata, where each stratum is an observed combination of the covariates. Within each stratum, cases with, and controls without, the target event are counted; controls are weighted to be as frequent as cases. This weighting procedure guarantees that covariates, or combination of covariates, are balanced, meaning they occur at the same rate among cases and controls. Finally, impact of the target event is calculated in the weighted data. We compare the performance of SCB, logistic regression (LR), and propensity scoring (PS) in simulated and real data. We examined the calibration of SCB and PS in predicting 6-month mortality from inability to eat, controlling for age, gender, and nine other disabilities for 296,051 residents in Veterans Affairs nursing homes. We also performed a simulation study, where outcomes were randomly generated from treatment, 10 covariates, and increasing number of covariate interactions. The accuracy of SCB, PS, and LR in recovering the simulated treatment effect was reported. FINDINGS: In simulated environment, as the number of interactions among the covariates increased, SCB and properly specified LR remained accurate but pairwise LR and pairwise PS, the most common applications of these tools, performed poorly. In real data, application of SCB was practical. SCB was better calibrated than linear PS, the most common method of PS. CONCLUSIONS: In environments where covariates interact, SCB is practical and more accurate than common methods of applying LR and PS.

Budget Impact Analysis of Veterans Affairs Medical Foster Homes versus Community Living Centers.

The objectives were to determine whether and by what amounts the US Department of Veterans Affairs (VA) use of Medical Foster Homes (MFH) rather than Community Living Centers (CLC) reduced budget impacts to the VA. This was a retrospective, matched, case-control study of veterans residing in MFH or CLC in the VA health care system from 2008 to 2012. Administrative data sets, nearest neighbor matching, generalized linear models, and a secondary analysis were used to capture and analyze budget impacts by veterans who used MFH or CLC exclusively in 2008-2012. Controls of 1483 veterans in CLC were matched to 203 cases of veterans in MFH. Use of MFH instead of CLC reduced budget impacts to the VA by at least $2645 per veteran per month. A secondary analysis of the data using different matching criteria and statistical methods produced similar results, demonstrating the robustness of the estimates of budget impact. When the average out-of-pocket payments made by MFH residents, not made by CLC residents, were included in the analysis, the net reduction of budget impact ranged from $145 to $2814 per veteran per month or a savings of $1740 to $33,768 per veteran per year. Even though outpatient costs of MFH are higher, much of the reduced budget impact of MFH use arises from lower inpatient or hospital costs. Reduced budget impacts on the VA system indicate that expansion of the MFH program may be cost-effective. Implications for further research are suggested.

The VA Medical Foster Home Program, Ambulatory Care Sensitive Conditions, and Avoidable Hospitalizations.

This quality control study analyzes whether the Veterans Administration Medical Foster Home (VA MFH) program has been successful in improving access and effectiveness of ambulatory care. Individuals hospitalized for one or more of 22 adult ambulatory care sensitive conditions were identified. Pre and post comparisons of a specified population of participants in the program were conducted to determine rates of avoidable hospitalizations for 6 months prior to and following MFH enrollment. The overall rate of avoidable hospitalizations declined from 18.5 to 14.9 per 100 enrollees following enrollment. The number of bed days used declined by 39%, as did the cost associated with avoidable hospitalizations. Enrollment in the VA MFH program resulted in an overall reduction in the rate of avoidable hospitalizations, resource utilization, and costs. Studies are needed comparing these results with other matched cohorts of nursing home eligible veterans.

Automated monitoring to detect H1N1 symptoms among urban, Medicaid-eligible, pregnant women: a community-partnered randomized controlled trial.

In response to the H1N1 epidemic, we used community health workers to design and implement a randomized controlled trial to test the efficacy of a new automated call-monitoring system for second and third trimester predominantly Medicaid-eligible pregnant women in an urban free standing birth center to promptly detect symptoms of influenza and assure rapid treatment to prevent adverse outcomes from influenza. Daily automated telephone call to second and third trimester pregnant women asking if the woman experienced flu-like symptoms. Calls continued daily until 38 weeks gestation. A community health worker's voice was used for the automated call recording. Positive responses triggered an immediate referral to a nurse-midwife for prompt treatment with anti-viral medication. Fifty pregnant participants were randomized into daily-automated call group (n = 26) or health information group (n = 24). The automated call group participants ranged in age from 14 to 36 (mean = 23.5, SD = 6.3), 84.7 % identified their race/ethnicity as African-American Non-Hispanic, and 80.7 % were Medicaid-Eligible. In the automated call group, 11.5 % chose to be immunized against H1N1. The mean percent of patients reached daily was 45.1 % (SD = 3.2 %) and at least once every 3 days was 65.1 % (SD = 3.1 %). One pregnant woman in the automated call group contracted H1N1 influenza and received prompt anti-viral treatment without any serious outcomes. Participation in daily-automated telephone calls did not differ significantly between patients younger than 18 years old versus patients 18 years or older. There was also no difference in participation between patients with parity of 0 versus patients with parity ≥1. Participation in daily telephone calls significantly (alpha ≤0.05) increased when a community health worker provided personal follow-up of non-responsive participants. 93.3 % of surveyed pregnant women, who received automated daily calls, recommended to use a similar daily call system in response to a future health crisis. Automated daily phone calls, designed and produced by community health workers, is a feasible, well received strategy to provide urgent health information to an urban, Medicaid-eligible group of pregnant women, regardless of age or parity.

Rethinking satisfaction surveys: minute survey.

The Minute Survey uses 2 questions to assess patient's satisfaction. The first question asks the patient to rate overall satisfaction with the care. The second question asks the patient to explain what worked well and what needed improvement. The Minute Survey reduces cost of conducting satisfaction surveys by (1) reducing cost of printing, (2) reducing cost of handling and mailing, (3) increasing response rate and thus reducing the need for follow-up reminder, and (4) by relying on time to dissatisfied patient as opposed to percent of dissatisfied patients. We report response rate of 34% to 77% to minute surveys. The combination of Minute Survey and analysis of time to dissatisfied patient may reduce the cost of conducting satisfaction surveys by 89% compared with Consumer Assessment of Healthcare Providers and Systems survey suggested for use by Centers for Medicare and Medicaid.

An example of activity based costing of treatment programs.

We introduce a new tool that can be used for estimating number, length of time, and nature of services patient receive in drug treatment programs. While the field has made significant progress in standardizing the collection of expenditure data, little progress has been made on creating a standard measure for estimating program activities and census. We report on a method of estimating program activities.

Probabilistic master lists: integration of patient records from different databases when unique patient identifier is missing.

We show how Bayesian probability models can be used to integrate two databases, one of which does not have a key for uniquely identifying clients (e.g., social security number or medical record number). The analyst selects a set of imperfect identifiers (last visit diagnosis, first name, etc.). The algorithm assesses the likelihood ratio associated with the identifier from the database of known cases. It estimates the probability that two records belong to the same client from the likelihood ratios. As it proceeds in examining various identifiers, it accounts for inter-dependencies among them by allowing overlapping and redundant identifiers to be used. We test that the procedure is effective by examining data from the Medical Expenditure Panel Survey (MEPS) Population Characteristics data set, a publicly available data set. We randomly selected 1,000 cases for training data set--these constituted the known cases. The algorithm was used to identify if 100 cases not in the training data set would be misclassified in terms of being a case in the training set or a new case. With 12 fields as identifiers, all 100 cases were correctly classified as new cases. We also selected 100 known cases from the training set and asked the algorithm to classify these cases. Again, all 100 cases were correctly classified. Less accurate results were obtained when the training data set was too small (e.g., less than 100 records) or the number of fields used as identifiers was too small (e.g., less than seven fields). In a test of performance of the algorithm, when the ratio of testing to training data set exceeds 4 to 1, the accuracy of the algorithm exceeded 90% of cases. As the ratio increases, the accuracy of algorithm improves further. These data suggest the accuracy of our automated and mathematical procedure to merge data from two different data sets without the presence of a unique identifier. The algorithm uses imperfect and overlapping clues to re-identify cases from information not typically considered to be a patient identifier.

Costs and benefits of combining probation and substance abuse treatment.

AIMS OF THE STUDY: We compared seamless combination of probation and treatment (where the probation officer is co-located with treatment provider or is actively engaged in treatment) to traditional probation where treatment is left to the client's choice. METHODS: Clients were randomly assigned to either seamless or traditional probation. We used a decision analytic approach which had two advantages: First it separated estimation of probability of adverse events (e.g. hospitalization) from the daily cost of the adverse event, thereby allowing use of estimates of daily costs available within the literature. Second, the reliance on daily probability of various adverse events also had the benefit of reflecting both length of the event and its intermittent re-occurrence. Subjects were 272 clients on probation in Northern Virginia and Maryland in the United States. Clients were randomly assigned to seamless and traditional probation and were followed for an average of 2.75 years (arrest information was only available for 1 year); 77% of clients participated in the follow-up interviews. At baseline, there was no statistically significant difference among the clients. RESULTS: During the follow-up period, clients in the seamless probation had less recidivism but the cost savings from this component (dollar 2.31 per client per follow-up day) was not sufficient to overcome increased costs due to mental hospitalization of seamless clients (dollar 13.50 per client per follow-up day), cost of delivery of seamless probation (dollar 2.58 per client per follow-up day), more frequent use of jail/prison for clients in the seamless group (dollar 2.08 per client per follow-up day) and additional treatment costs (dollar 1.24 per client per follow-up day). The expected cost of seamless probation and its consequences was dollar 38.84 per follow-up day. The expected cost of traditional probation and its consequences was dollar 21.60 per follow-up day. Seamless probation was dollar 6,293 more expensive than traditional probation per client per year. DISCUSSION: Sensitivity analysis suggested that the analysis was not sensitive to small change in any single cost or probability estimate. Sensitivity analysis suggested that increased supervision intensity and use of sanctions had contributed to lower cost-effectiveness. IMPLICATIONS: One possible way of improving seamless probation is to improve the intensity of the substance abuse treatment while reducing the intensity of supervision to its traditional levels. This analysis was limited to 2.75 years follow-up period and does not address cost savings that might occur after this period.

Activity based costing of probation with and without substance abuse treatment: a case study.

BACKGROUND: Since many offenders have drug problems, investigators have proposed that drug testing and treatment should be an integral part of probation. In 1994, the Office of National Drug Control Policy (ONDCP) funded a demonstration project designed to integrate drug treatment with traditional supervision services. As part of this demonstration a new procedure called 'seamless' probation was set up in which treatment providers were co-located with probation officers and probation officers coordinated offenders' participation in treatment. AIMS OF THE STUDY: This study examines the cost of providing substance abuse treatment coordination through probation agencies. METHODS: We used Activity Based Costing (ABC) to examine the cost of probation with and without treatment coordination in one probation agency. Agency budget was analyzed and allocated to various programs. A questionnaire was developed to assess probation officer's activities. The cost of coordinating treatment for one offender was calculated by dividing the total cost of the program by units of various activities done by the probation officers. RESULTS: Preliminary test of reliability of the instrument showed that it was accurately portraying the probation officers time allocation. Probation officers spent 6.9% of their time in seamless supervision and 83.3% time in traditional supervision (83.83%). The seamless probation officers had more group meetings and more phone contact with their offenders than traditional probation officers. The average cost per offender per day was 12 dollars for seamless probation and 7 dollars for traditional probation. DISCUSSION: This study is limited because it focuses on one agency at one point in time. Results may not be relevant to other agencies or to the same agency as it makes its operation more efficient. This study provides a method of allocating budget cost to per client costs using survey of probation officer's activities -- a tool developed in this study. Comparison of seamless and traditional supervision activities showed major differences in terms of the probation officers' activities and costs. IMPLICATIONS: There are significant costs associated with asking probation officers to coordinate treatment. Studies should be undertaken to examine the relative benefits that can be derived from this increased cost.

Evaluating Medicaid HMOs when encounter data are missing: case of developmentally delayed children.

In evaluating Medicaid Health Maintenance Organizations (HMOs), crucial information regarding severity of illness of patients is often missing--in part because encounter data are not available. If we assume that patients are either in the HMO or in fee-for-service (FFS) plans (i.e., no in or out migration); then severity of HMO patients can be deduced from encounters of FFS patients. We applied this approach to effectiveness of HMO services for developmentally delayed children. Data supported the assumption of a closed system. Data also showed that over 12 months, severity of FFS patients declined. Therefore, we inferred that the HMO was attracting sicker patients. The HMO was paid less than FFS plan, despite the fact that it attracted sicker patients.

Unit cost of counseling and patients length of stay in a residential drug treatment setting.

BACKGROUND: Many published reports on cost of counseling give a fixed cost per hour of service. These estimates may be flawed. AIMS OF THE STUDY: The purpose of this study is to show, by way of an example, how cost of an hour of counseling depends on the nature of the patient, in general, and length of the patient s stay, in particular. Even though the health care professional provides the same hour of work, the cost of the hour is different for short-stay and long-stay patients. METHODS: We identified 5-short and 5 long stay patients in a residential treatment program. For each group, we asked the counselors to review the medical records and measure the patients utilization of various service units. We estimated the cost of a unit of service by dividing cost of an average patient by the program utilization of short and long-stay patients. RESULTS: The cost of an hour of counseling for long stay patients was 2/3 less than the cost of short-stay patients. Similar large changes in unit cost of treatment were observed for cost of group counseling or other components of substance abuse treatment. DISCUSSION: Our data was limited to one case study and may not indicate similar patterns in other treatment programs. The paper suggests that methods of studying cost of treatment should be adjusted to reflect case mix of patients and their expected length of stay. IMPLICATIONS FOR HEALTH POLICIES: Our analysis shows that higher rates should be set for patients at risk for short stays; conversely lower rates should be set for patients likely to complete treatment. Without adjusting the rate for the case mix of patients, health care institutions have an incentive to avoid the difficult cases and concentrate on long stay cases. IMPLICATIONS FOR FURTHER RESEARCH: A number of instruments that measure severity of illness or difficulty of treatment can be used to anticipate patients length of stay. Then the rate for units of treatment can be set based on patients expected length of stay. This paper presents a questionnaire that can be used to collect cost data and estimate cost per unit of treatment adjusted for expected length of stay.