Development of a Social Risk Score in the Electronic Health Record to Identify Social Needs Among Underserved Populations: Retrospective Study.

BACKGROUND: Patients with unmet social needs and social determinants of health (SDOH) challenges continue to face a disproportionate risk of increased prevalence of disease, health care use, higher health care costs, and worse outcomes. Some existing predictive models have used the available data on social needs and SDOH challenges to predict health-related social needs or the need for various social service referrals. Despite these one-off efforts, the work to date suggests that many technical and organizational challenges must be surmounted before SDOH-integrated solutions can be implemented on an ongoing, wide-scale basis within most US-based health care organizations. OBJECTIVE: We aimed to retrieve available information in the electronic health record (EHR) relevant to the identification of persons with social needs and to develop a social risk score for use within clinical practice to better identify patients at risk of having future social needs. METHODS: We conducted a retrospective study using EHR data (2016-2021) and data from the US Census American Community Survey. We developed a prospective model using current year-1 risk factors to predict future year-2 outcomes within four 2-year cohorts. Predictors of interest included demographics, previous health care use, comorbidity, previously identified social needs, and neighborhood characteristics as reflected by the area deprivation index. The outcome variable was a binary indicator reflecting the likelihood of the presence of a patient with social needs. We applied a generalized estimating equation approach, adjusting for patient-level risk factors, the possible effect of geographically clustered data, and the effect of multiple visits for each patient. RESULTS: The study population of 1,852,228 patients included middle-aged (mean age range 53.76-55.95 years), White (range 324,279/510,770, 63.49% to 290,688/488,666, 64.79%), and female (range 314,741/510,770, 61.62% to 278,488/448,666, 62.07%) patients from neighborhoods with high socioeconomic status (mean area deprivation index percentile range 28.76-30.31). Between 8.28% (37,137/448,666) and 11.55% (52,037/450,426) of patients across the study cohorts had at least 1 social need documented in their EHR, with safety issues and economic challenges (ie, financial resource strain, employment, and food insecurity) being the most common documented social needs (87,152/1,852,228, 4.71% and 58,242/1,852,228, 3.14% of overall patients, respectively). The model had an area under the curve of 0.702 (95% CI 0.699-0.705) in predicting prospective social needs in the overall study population. Previous social needs (odds ratio 3.285, 95% CI 3.237-3.335) and emergency department visits (odds ratio 1.659, 95% CI 1.634-1.684) were the strongest predictors of future social needs. CONCLUSIONS: Our model provides an opportunity to make use of available EHR data to help identify patients with high social needs. Our proposed social risk score could help identify the subset of patients who would most benefit from further social needs screening and data collection to avoid potentially more burdensome primary data collection on all patients in a target population of interest.

The Impact of Various Risk Assessment Time Frames on the Performance of Opioid Overdose Forecasting Models.

BACKGROUND: An individual's risk for future opioid overdoses is usually assessed using a 12-month "lookback" period. Given the potential urgency of acting rapidly, we compared the performance of alternative predictive models with risk information from the past 3, 6, 9, and 12 months. METHODS: We included 1,014,033 Maryland residents aged 18-80 with at least 1 opioid prescription and no recorded death in 2015. We used 2015 Maryland prescription drug monitoring data to identify risk factors for nonfatal opioid overdoses from hospital discharge records and investigated fatal opioid overdose from medical examiner data in 2016. Prescription drug monitoring program-derived predictors included demographics, payment sources for opioid prescriptions, count of unique opioid prescribers and pharmacies, and quantity and types of opioids and benzodiazepines filled. We estimated a series of logistic regression models that included 3, 6, 9, and 12 months of prescription drug monitoring program data and compared model performance, using bootstrapped C-statistics and associated 95% confidence intervals. RESULTS: For hospital-treated nonfatal overdose, the C-statistic increased from 0.73 for a model including only the fourth quarter to 0.77 for a model with 4 quarters of data. For fatal overdose, the area under the curve increased from 0.80 to 0.83 over the same models. The strongest predictors of overdose were prescription fills for buprenorphine and Medicaid and Medicare as sources of payment. CONCLUSIONS: Models predicting opioid overdose using 1 quarter of data were nearly as accurate as models using all 4 quarters. Models with a single quarter may be more timely and easier to identify persons at risk of an opioid overdose.

Assessing the Population-Level Correlation of Medication Regimen Complexity and Adherence Indices Using Electronic Health Records and Insurance Claims.

BACKGROUND: Nonadherence to medication regimens can lead to adverse health care outcomes and increasing costs. OBJECTIVES: To (a) assess the level of medication complexity at an outpatient setting using population-level electronic health record (EHR) data and (b) evaluate its association with medication adherence measures derived from medication-dispensing claims. METHODS: We linked EHR data with insurance claims of 70,054 patients who had an encounter with a U.S. midwestern health system between 2012 and 2013. We constructed 3 medication-derived indices: medication regimen complexity index (MRCI) using EHR data; medication possession ratio (MPR) using insurance pharmacy claims; and prescription fill rates (PFR; 7 and 30 days) using both data sources. We estimated the partial correlation between indices using Spearman's coefficient (SC) after adjusting for age and sex. RESULTS: The mean age (SD) of 70,054 patients was 37.9 (18.0) years, with an average Charlson Comorbidity Index of 0.308 (0.778). The 2012 data showed mean (SD) MRCI, MPR, and 30-day PFR of 14.6 (17.8), 0.624 (0.310), and 81.0 (27.0), respectively. Patients with previous inpatient stays were likely to have high MRCI scores (36.3 [37.9], P < 0.001) and were less adherent to outpatient prescriptions (MPR = 50.3 [27.6%], P < 0.001; 30-day PFR = 75.7 [23.6%], P < 0.001). However, MRCI did not show a negative correlation with MPR (SC = -0.31, P < 0.001) or with 30-day PFR (SC = -0.17, P < 0.001) at significant levels. CONCLUSIONS: Medication complexity and adherence indices can be calculated on a population level using linked EHR and claims data. Regimen complexity affects patient adherence to outpatient medication, and strength of correlations vary modestly across populations. Future studies should assess the added values of MRCI, MPR, and PFR to population health management efforts. DISCLOSURES: No outside funding supported this study. The authors have nothing to disclose. The abstract of this work was presented at INFORMS Healthcare Conference, held on July 27-29, 2019, in Cambridge, MA.

Evaluating the Impact of Prescription Fill Rates on Risk Stratification Model Performance.

BACKGROUND: Risk adjustment models are traditionally derived from administrative claims. Prescription fill rates-extracted by comparing electronic health record prescriptions and pharmacy claims fills-represent a novel measure of medication adherence and may improve the performance of risk adjustment models. OBJECTIVE: We evaluated the impact of prescription fill rates on claims-based risk adjustment models in predicting both concurrent and prospective costs and utilization. METHODS: We conducted a retrospective cohort study of 43,097 primary care patients from HealthPartners network between 2011 and 2012. Diagnosis and/or pharmacy claims of 2011 were used to build 3 base models using the Johns Hopkins ACG system, in addition to demographics. Model performances were compared before and after adding 3 types of prescription fill rates: primary 0-7 days, primary 0-30 days, and overall. Overall fill rates utilized all ordered prescriptions from electronic health record while primary fill rates excluded refill orders. RESULTS: The overall, primary 0-7, and 0-30 days fill rates were 72.30%, 59.82%, and 67.33%. The fill rates were similar between sexes but varied across different medication classifications, whereas the youngest had the highest rate. Adding fill rates modestly improved the performance of all models in explaining medical costs (improving concurrent R by 1.15% to 2.07%), followed by total costs (0.58% to 1.43%), and pharmacy costs (0.07% to 0.65%). The impact was greater for concurrent costs compared with prospective costs. Base models without diagnosis information showed the highest improvement using prescription fill rates. CONCLUSIONS: Prescription fill rates can modestly enhance claims-based risk prediction models; however, population-level improvements in predicting utilization are limited.

Comparing Population-based Risk-stratification Model Performance Using Demographic, Diagnosis and Medication Data Extracted From Outpatient Electronic Health Records Versus Administrative Claims.

BACKGROUND: There is an increasing demand for electronic health record (EHR)-based risk stratification and predictive modeling tools at the population level. This trend is partly due to increased value-based payment policies and the increasing availability of EHRs at the provider level. Risk stratification models, however, have been traditionally derived from claims or encounter systems. This study evaluates the challenges and opportunities of using EHR data instead of or in addition to administrative claims for risk stratification. METHODS: This study used the structured EHR records and administrative claims of 85,581 patients receiving outpatient care at a large integrated provider system. Common data elements for risk stratification (ie, age, sex, diagnosis, and medication) were extracted from outpatient EHR records and administrative claims. The performance of a validated risk-stratification model was assessed using data extracted from claims alone, EHR alone, and claims and EHR combined. RESULTS: EHR-derived metrics overlapped considerably with administrative claims (eg, number of chronic conditions). The accuracy of the model, when using EHR data alone, was acceptable with an area under the curve of ∼0.81 for hospitalization and ∼0.85 for identifying top 1% utilizers using the concurrent model. However, when using EHR data alone, the predictive model explained a lower amount of variation in utilization-based outcomes compared with administrative claims. DISCUSSION: The results show a promising performance of models predicting cost and hospitalization using outpatient EHR's diagnosis and medication data. More research is needed to evaluate the benefits of other EHR data types (eg, lab values and vital signs) for risk stratification.

Change in claims-based diabetes medications is a diabetes improvement indicator.

BACKGROUND: Because laboratory test results are less available to researchers than claims data, a claims-based indicator of diabetes improvement would be valuable. OBJECTIVES: To determine whether a decrease in medication use for diabetes parallels clinical improvement in glycemic control. STUDY DESIGN: This was a retrospective cohort study using up to 3.5 years of pharmacy and laboratory data from 1 private insurer. Data included 104 patients with diabetes who underwent bariatric surgery and had at least 1 glycated hemoglobin (A1C) test before and after surgery. METHODS: We assigned each A1C test to a 90-day interval before or after surgery. Medication availability was noted for the midpoint of the interval (on insulin, on oral medications, count of medications). Each subject could contribute 1 presurgery and up to 3 postsurgery observations. We recorded the changes in A1C test results and medication use from the presurgery to the postsurgery period. Using the A1C test as the reference standard, positive and negative predictive values of medication-based indicators were calculated. RESULTS: After bariatric surgery, A1C test values decreased by more than 1% and the count of unique medications decreased by 0.6. All 3 medication-based indicators had high positive predictive values (0.85) and low negative predictive values (0.20), and count of medications had better performance than the other indicators. CONCLUSIONS: Without clinical information, a decrease in use of medications can serve as a proxy for clinical improvement. Validation of results in other settings is needed.

Impact of bariatric surgery on health care costs of obese persons: a 6-year follow-up of surgical and comparison cohorts using health plan data.

IMPORTANCE: Bariatric surgery is a well-documented treatment for obesity, but there are uncertainties about the degree to which such surgery is associated with health care cost reductions that are sustained over time. OBJECTIVE: To provide a comprehensive, multiyear analysis of health care costs by type of procedure within a large cohort of privately insured persons who underwent bariatric surgery compared with a matched nonsurgical cohort. DESIGN: Longitudinal analysis of 2002-2008 claims data comparing a bariatric surgery cohort with a matched nonsurgical cohort. SETTING: Seven BlueCross BlueShield health insurance plans with a total enrollment of more than 18 million persons. PARTICIPANTS: A total of 29 820 plan members who underwent bariatric surgery between January 1, 2002, and December 31, 2008, and a 1:1 matched comparison group of persons not undergoing surgery but with diagnoses closely associated with obesity. MAIN OUTCOME MEASURES: Standardized costs (overall and by type of care) and adjusted ratios of the surgical group's costs relative to those of the comparison group. RESULTS: Total costs were greater in the bariatric surgery group during the second and third years following surgery but were similar in the later years. However, the bariatric group's prescription and office visit costs were lower and their inpatient costs were higher. Those undergoing laparoscopic surgery had lower costs in the first few years after surgery, but these differences did not persist. CONCLUSIONS AND RELEVANCE: Bariatric surgery does not reduce overall health care costs in the long term. Also, there is no evidence that any one type of surgery is more likely to reduce long-term health care costs. To assess the value of bariatric surgery, future studies should focus on the potential benefit of improved health and well-being of persons undergoing the procedure rather than on cost savings.

Validating the adapted Diabetes Complications Severity Index in claims data.

OBJECTIVES: To test the validity of the adapted Diabetes Complications Severity Index (aDCSI), which does not include laboratory test results, as an indicator of diabetes severity. STUDY DESIGN: Retrospective cohort study using 4 years of claims data from 7 health insurance plans. METHODS: Individuals with diabetes mellitus and continuous enrollment were study subjects (N = 138,615). The 2 independent variables--the aDCSI score (sum of 7 diabetes complications graded by severity as 0, 1, or 2; range 0-13) and the aDCSI diabetes complication count (sum of 7 diabetes complications without severity grading; range 0-7)--were generated using only claims data. We evaluated the numbers of hospitalizations attributable to the aDCSI with Poisson regression models, both categorically and linearly. RESULTS: The aDCSI score (risk ratio 1.39 to 6.10 categorically and 1.41 linearly) and diabetes complication count (risk ratio 1.67 to 9.11 categorically and 1.65 linearly) were both significantly positively associated with the number of hospitalizations over a 4-year period. Risk ratios from the aDCSI score were very similar to the risk ratios previously reported for the Diabetes Complications Severity Index (DCSI); the absolute difference between risk ratios ranged from 0.01 to 1.6 categorically and was 0.05 linearly. CONCLUSIONS: The aDCSI is a good measure of diabetes severity, given its ability to explain hospitalizations and its similar performance to the DCSI.

Predicting costs with diabetes complications severity index in claims data.

OBJECTIVES: To test the usefulness of the Diabetes Complications Severity Index (DCSI) without laboratory test results in predicting healthcare costs, for potential use in disease management programs. STUDY DESIGN: Retrospective cohort study using up to 2 years of claims data from 7 health insurance plans. METHODS: Individuals with diabetes mellitus and continuous enrollment were study subjects. The DCSI (sum of 7 diabetes complications graded by severity as 0, 1, or 2; range 0-13) and count of diabetes complications (sum of 7 diabetes complications without severity grading; range 0-7) were the main independent variables and were generated using only diagnostic codes. We analyzed 5 types of healthcare costs (ie, total costs, inpatient costs, hospital other costs, pharmacy costs, and professional costs) attributable to the DCSI and the complication count with linear regression models, both concurrently and prospectively. RESULTS: The DCSI without laboratory data was a better predictor of costs than was complication count (adjusted R2 of total costs: 0.095 vs 0.080). The DCSI explained concurrent costs better than future costs (adjusted R2 of total costs: 0.095 vs 0.019). There were important differences in healthcare utilization among people stratifi ed by DCSI scores: 5-fold and 3-fold differences in concurrent and prospective total costs, respectively, across 4 DCSI groups. CONCLUSIONS: The DCSI without laboratory data may be useful for stratifying individuals with diabetes into morbidity groups, which can be used for selection into disease management programs or for matching in observational research.

Quantification of a methadone metabolite (EDDP) in urine: assessment of compliance.

OBJECTIVE: To investigate the possibility of utilizing the ratio of the methadone metabolite, 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP), to urine creatinine to develop a regression model that would predict drug adherence in patients prescribed methadone for either pain management or drug addiction. DESIGN: Retrospective study. SETTING: Marshfield Clinic-Lakeland Center, one of 41 regional centers that make up Marshfield Clinic, a large, private, multi-specialty healthcare institution in central Wisconsin. PARTICIPANTS: Patients receiving methadone treatment for substance abuse or chronic pain. Group 1 was an initial pilot group consisting of 7 patients who were followed for a 4-month period. Group 2 consisted of 33 patients who were followed over a 28-month period. METHODS: Age, gender, weight, height, methadone dosage, quantitative urine creatinine and EDDP levels, reported compliance/non-compliance, and relevant clinical cofactors were retrospectively abstracted from the patients' medical records. Log-log regression analyses were used to model EDDP and the EDDP/creatinine ratio from urine screening results as functions of methadone dose, and in the larger cohort (group 2), body size, gender and age. The coefficient of determination adjusted for the number of predictor terms (R(adj)(2)) was reported as a measure of model fit. RESULTS: For group 1 data, there was a significant positive relation (P<0.001) but also substantial variability (R(adj)(2) = 0.49). Adjustment for creatinine through the EDDP/creatinine ratio provided a tighter relation (R(adj)(2) = 0.95). Similarly, for group 2 data, there was a significant positive relation (P=0.001) and substantial variability (R(adj)(2) = 0.53). Adjustment for creatinine through EDDP/creatinine ratios provided a substantially stronger relation (R(adj)(2) = 0.73). Gender and age showed no evidence of association with the EDDP/creatinine ratio (P=0.60 and P=0.51, respectively). Body size was significant in the model, both when measured by body surface area and by lean body weight, and improved the prediction when added to our model (R(adj)(2) = 0.80). CONCLUSION: For the first time, urine analyses may be used to monitor methadone over- or under-use in a clinical setting, regardless of the state of patient hydration or the manipulation of a sample by addition of another substance, such as bleach, soap, or even methadone, which could render an appropriate sample inappropriate or an inappropriate sample appropriate. A similar approach may prove useful for other drug treatments, allowing for more accurate monitoring of commonly abused prescription medications.