Kenneth Nieser

All Publications

• Comparing methods for assessing the reliability of health care quality measures. Statistics in medicine Nieser, K. J., Harris, A. H. 2024

  Abstract

  Quality measurement plays an increasing role in U.S. health care. Measures inform quality improvement efforts, public reporting of variations in quality of care across providers and hospitals, and high-stakes financial decisions. To be meaningful in these contexts, measures should be reliable and not heavily impacted by chance variations in sampling or measurement. Several different methods are used in practice by measure developers and endorsers to evaluate reliability; however, there is uncertainty and debate over differences between these methods and their interpretations. We review methods currently used in practice, pointing out differences that can lead to disparate reliability estimates. We compare estimates from 14 different methods in the case of two sets of mental health quality measures within a large health system. We find that estimates can differ substantially and that these discrepancies widen when sample size is reduced.

  View details for DOI 10.1002/sim.10197

  View details for PubMedID 39145538

• Split-sample reliability estimation in health care quality measurement: Once is not enough. Health services research Nieser, K. J., Harris, A. H. 2024

  Abstract

  To examine the sensitivity of split-sample reliability estimates to the random split of the data and propose alternative methods for improving the stability of the split-sample method.Data were simulated to reflect a variety of real-world quality measure distributions and scenarios. There is no date range to report as the data are simulated.Simulation studies of split-sample reliability estimation were conducted under varying practical scenarios.All data were simulated using functions in R.Single split-sample reliability estimates can be very dependent on the random split of the data, especially in low sample size and low variability settings. Averaging split-sample estimates over many splits of the data can yield a more stable reliability estimate.Measure developers and evaluators using the split-sample reliability method should average a series of reliability estimates calculated from many resamples of the data without replacement to obtain a more stable reliability estimate.

  View details for DOI 10.1111/1475-6773.14310

  View details for PubMedID 38659301

• Quantifying and reducing inequity in average treatment effect estimation. BMC medical research methodology Nieser, K. J., Cochran, A. L. 2023; 23 (1): 297

  Abstract

  Across studies of average treatment effects, some population subgroups consistently have lower representation than others which can lead to discrepancies in how well results generalize.We develop a framework for quantifying inequity due to systemic disparities in sample representation and a method for mitigation during data analysis. Assuming subgroup treatment effects are exchangeable, an unbiased sample average treatment effect estimator will have lower mean-squared error, on average across studies, for subgroups with less representation when treatment effects vary. We present a method for estimating average treatment effects in representation-adjusted samples which enables subgroups to optimally leverage information from the full sample rather than only their own subgroup's data. Two approaches for specifying representation adjustment are offered-one minimizes average mean-squared error for each subgroup separately and the other balances minimization of mean-squared error and equal representation. We conduct simulation studies to compare the performance of the proposed estimators to several subgroup-specific estimators.We find that the proposed estimators generally provide lower mean squared error, particularly for smaller subgroups, relative to the other estimators. As a case study, we apply this method to a subgroup analysis from a published study.We recommend the use of the proposed estimators to mitigate the impact of disparities in representation, though structural change is ultimately needed.

  View details for DOI 10.1186/s12874-023-02104-2

  View details for PubMedID 38102563

  View details for PubMedCentralID PMC10722685

• Detection of differential depressive symptom patterns in a cohort of perinatal women: an exploratory factor analysis using a robust statistics approach. EClinicalMedicine Nieser, K. J., Stowe, Z. N., Newport, D. J., Coker, J. L., Cochran, A. L. 2023; 57: 101830

  Abstract

  Postpartum depression can take many forms. Different symptom patterns could have divergent implications for how we screen, diagnose, and treat postpartum depression. We sought to utilise a recently developed robust estimation algorithm to automatically identify differential patterns in depressive symptoms and subsequently characterise the individuals who exhibit different patterns.Depressive symptom data (N = 548) were collected from women with neuropsychiatric illnesses at two U.S. urban sites participating in a longitudinal observational study of stress across the perinatal period. Data were collected from Emory University between 1994 and 2012 and from the University of Arkansas for Medical Sciences between 2012 and 2017. We conducted an exploratory factor analysis of Beck Depression Inventory (BDI) items using a robust expectation-maximization algorithm, rather than a conventional expectation-maximization algorithm. This recently developed method enabled automatic detection of differential symptom patterns. We described differences in symptom patterns and conducted unadjusted and adjusted analyses of associations of symptom patterns with demographics and psychiatric histories.53 (9.7%) participants were identified by the algorithm as having a different pattern of reported symptoms compared to other participants. This group had more severe symptoms across all items-especially items related to thoughts of self-harm and self-judgement-and differed in how their symptoms related to underlying psychological constructs. History of social anxiety disorder (OR: 4.0; 95% CI [1.9, 8.1]) and history of childhood trauma (for each 5-point increase, OR: 1.2; 95% CI [1.1, 1.3]) were significantly associated with this differential pattern after adjustment for other covariates.Social anxiety disorder and childhood trauma are associated with differential patterns of severe postpartum depressive symptoms, which might warrant tailored strategies for screening, diagnosis, and treatment to address these comorbid conditions.There are no funding sources to declare.

  View details for DOI 10.1016/j.eclinm.2023.101830

  View details for PubMedID 36798754

  View details for PubMedCentralID PMC9925853

• Addressing heterogeneous populations in latent variable settings through robust estimation. Psychological methods Nieser, K. J., Cochran, A. L. 2023; 28 (1): 39-60

  Abstract

  Individuals routinely differ in how they present with psychiatric illnesses and in how they respond to treatment. This heterogeneity, when overlooked in data analysis, can lead to misspecified models and distorted inferences. While several methods exist to handle various forms of heterogeneity in latent variable models, their implementation in applied research requires additional layers of model crafting, which might be a reason for their underutilization. In response, we present a robust estimation approach based on the expectation-maximization (EM) algorithm. Our method makes minor adjustments to EM to enable automatic detection of population heterogeneity and to recognize individuals who are inadequately explained by the assumed model. Each individual is associated with a probability that reflects how likely their data were to have been generated from the assumed model. The individual-level probabilities are simultaneously estimated and used to weight each individual's contribution in parameter estimation. We examine the utility of our approach for Gaussian mixture models and linear factor models through several simulation studies, drawing contrasts with the EM algorithm. We demonstrate that our method yields inferences more robust to population heterogeneity or other model misspecifications than EM does. We hope that the proposed approach can be incorporated into the model-building process to improve population-level estimates and to shed light on subsets of the population that demand further attention. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

  View details for DOI 10.1037/met0000413

  View details for PubMedID 34694831

  View details for PubMedCentralID PMC9035483

• Statistical Methods to Examine Racial and Ethnic Disparities in the Surgical Literature: A Review and Recommendations for Improvement. Annals of surgery Harris, A. H., Eddington, H., Shah, V. B., Shwartz, M., Gurewich, D., Rosen, A. K., Quinteros, B., Wilcher, B., Nieser, K. J., Jones, G., Wu, J. T., Morris, A. M. 2024

  Abstract

  We characterized the quality of statistical methods for studies of racial and ethnic disparities in the surgical-relevant literature during 2021-2022.Hundreds of scientific papers are published each year describing racial and ethnic disparities in surgical access, quality, and outcomes. The content and design quality of this literature has never been systematically reviewed.We searched for 2021-2022 studies focused on describing racial and/or ethnic disparities in surgical or perioperative access, process quality, or outcomes. Identified studies were characterized in terms of three methodological criteria: 1) adjustment for variables related to both race/ethnicity and outcomes, including social determinants of health (SDOH); 2) accounting for clustering of patients within hospitals or other subunits ("providers") and; 3) distinguishing within- and between-provider effects.We identified 224 papers describing racial and/or ethnic differences. Of the 38 single institution studies, 24 (63.2%) adjusted for at least one SDOH variable. Of the 186 multisite studies, 113 (60.8%) adjusted for at least one SDOH variable, and 43 (23.1%) accounted for clustering of patients within providers using appropriate statistical methods. Only 10 (5.4%) of multi-institution studies made efforts to examine how much of overall disparities were driven by within versus between provider effects.Most recently published papers on racial and ethnic disparities in the surgical literature do not meet these important statistical design criteria and therefore may risk inaccuracy in the estimation of group differences in surgical access, quality, and outcomes. The most potent leverage points for these improvements are changes to journal publication guidelines and policies.

  View details for DOI 10.1097/SLA.0000000000006440

  View details for PubMedID 38979600

• Differences across race and ethnicity in the quality of antidepressant medication management. Health services research Harris, A. H., Liu, P., Breland, J. Y., Nieser, K. J., Schmidt, E. M. 2024

  Abstract

  To illustrate the importance of a multidimensional view of disparities in quality of antidepressant medication management (AMM), as well as discriminating "within-facility" disparities from disparities that exist between facilities.We used data from the Veterans Health Administration's (VA) Corporate Data Warehouse (CDW) which contains clinical and administrative data from VA facilities nationally.CDW data were used to measure five indicators of AMM quality, including the HEDIS Effective Acute-Phase and Effective Continuation-Phase measures. Mixed effects regression models were used to examine differences in quality indicators between racial/ethnic groups, controlling for other demographic and clinical factors. An adaptation of the Kitagawa-Blinder-Oaxaca (KBO) method was used to decompose mean differences in treatment quality between racial and ethnic groups into within- and between-facility effects.Demographic, clinical, and health service utilization data were extracted for patients in fiscal year 2017 with a diagnosis of depression and a new start of an antidepressant medication.The decomposition of the overall differences between White and Black patients on receiving an initial 90-day prescription (46.7% vs. 32.7%), Effective Acute-Phase (79.7% vs. 66.8%), and Effective Continuation-Phase (64.0% vs. 49.6%) HEDIS measures revealed that most of the overall effects were "within-facility," meaning that Black patients are less likely to meet these measures regardless of where they are treated. Although the overall magnitude of disparities between White and Hispanic patients on these three measures was very similar (46.7% vs. 32.7%; 79.7% vs. 69.2%; 64.0% vs. 53.6%), the differences were more attributable to Hispanic patients being treated in facilities with overall lower performance on these measures.Discriminating within- and between-facility disparities and taking a multidimensional view of quality are essential to informing efforts to address disparities in AMM quality.

  View details for DOI 10.1111/1475-6773.14347

  View details for PubMedID 38965913

• Childhood maltreatment exposure is differentially associated with Transdiagnostic perinatal depression symptoms. Journal of affective disorders Pingeton, B. C., Nieser, K. J., Cochran, A., Goodman, S. H., Laurent, H., Sbrilli, M. D., Knight, B., Newport, D. J., Stowe, Z. N. 2024

  Abstract

  History of childhood maltreatment (CM) is common and robustly associated with prenatal and postpartum (perinatal) depression. Given perinatal depression symptom heterogeneity, a transdiagnostic approach to measurement could enhance understanding of patterns between CM and perinatal depression.In two independently collected samples of women receiving care at perinatal psychiatry clinics (n = 523 and n = 134), we categorized longitudinal symptoms of perinatal depression, anxiety, stress, and sleep into transdiagnostic factors derived from the Research Domain Criteria and depression literatures. We split the perinatal period into four time points. We conducted a latent profile analysis of transdiagnostic factors in each period. We then used self-reported history of CM (total exposure and subtypes of abuse and neglect) to predict class membership.A three-class solution best fit our data. In relation to positive adaptive functioning, one class had relatively more positive symptoms (high adaptive), one class had average values (middle adaptive), and one class had fewer adaptive symptoms (low adaptive). More total CM and specific subtypes associated with threat/abuse increased an individual's likelihood of being in the Low Adaptive class in both samples (ORs: 0.90-0.97, p < .05).Generalizability of our results was curtailed by 1) limited racial/ethnic diversity and 2) missing data.Our results support taking a person-centered approach to characterize the relationship between perinatal depression and childhood maltreatment. Given evidence that increased exposure to childhood maltreatment is associated with worse overall symptoms, providers should consider incorporating preventative, transdiagnostic interventions for perinatal distress in individuals with a history of childhood maltreatment.

  View details for DOI 10.1016/j.jad.2024.05.021

  View details for PubMedID 38705531

• Differential Effects of an Emergency Department-to-Home Care Transitions Intervention in an Older Adult Population: A Latent Class Analysis. Medical care Green, R. K., Nieser, K. J., Jacobsohn, G. C., Cochran, A. L., Caprio, T. V., Cushman, J. T., Kind, A. J., Lohmeier, M., Shah, M. N. 2023; 61 (6): 400-408

  Abstract

  Older adults frequently return to the emergency department (ED) within 30 days of a visit. High-risk patients can differentially benefit from transitional care interventions. Latent class analysis (LCA) is a model-based method used to segment the population and test intervention effects by subgroup.We aimed to identify latent classes within an older adult population from a randomized controlled trial evaluating the effectiveness of an ED-to-home transitional care program and test whether class membership modified the intervention effect.Participants were randomized to receive the Care Transitions Intervention or usual care. Study staff collected outcomes data through medical record reviews and surveys. We performed LCA and logistic regression to evaluate the differential effects of the intervention by class membership.Participants were ED patients (age 60 y and above) discharged to a community residence.Indicator variables for the LCA included clinically available and patient-reported data from the initial ED visit. Our primary outcome was ED revisits within 30 days. Secondary outcomes included ED revisits within 14 days, outpatient follow-up within 7 and 30 days, and self-management behaviors.We interpreted 6 latent classes in this study population. Classes 1, 4, 5, and 6 showed a reduction in ED revisit rates with the intervention; classes 2 and 3 showed an increase in ED revisit rates. In class 5, we found evidence that the intervention increased outpatient follow-up within 7 and 30 days (odds ratio: 1.81, 95% CI: 1.13-2.91; odds ratio: 2.24, 95% CI: 1.25-4.03).Class membership modified the intervention effect. Population segmentation is an important step in evaluating a transitional care intervention.

  View details for DOI 10.1097/MLR.0000000000001848

  View details for PubMedID 37167559

  View details for PubMedCentralID PMC10176501

• Gene-set Enrichment with Mathematical Biology (GEMB). GigaScience Cochran, A. L., Nieser, K. J., Forger, D. B., Zöllner, S., McInnis, M. G. 2020; 9 (10)

  Abstract

  Gene-set analyses measure the association between a disease of interest and a "set" of genes related to a biological pathway. These analyses often incorporate gene network properties to account for differential contributions of each gene. We extend this concept further-defining gene contributions based on biophysical properties-by leveraging mathematical models of biology to predict the effects of genetic perturbations on a particular downstream function.We present a method that combines gene weights from model predictions and gene ranks from genome-wide association studies into a weighted gene-set test. We demonstrate in simulation how such a method can improve statistical power. To this effect, we identify a gene set, weighted by model-predicted contributions to intracellular calcium ion concentration, that is significantly related to bipolar disorder in a small dataset (P = 0.04; n = 544). We reproduce this finding using publicly available summary data from the Psychiatric Genomics Consortium (P = 1.7 × 10-4; n = 41,653). By contrast, an approach using a general calcium signaling pathway did not detect a significant association with bipolar disorder (P = 0.08). The weighted gene-set approach based on intracellular calcium ion concentration did not detect a significant relationship with schizophrenia (P = 0.09; n = 65,967) or major depression disorder (P = 0.30; n = 500,199).Together, these findings show how incorporating math biology into gene-set analyses might help to identify biological functions that underlie certain polygenic disorders.

  View details for DOI 10.1093/gigascience/giaa091

  View details for PubMedID 33034635

  View details for PubMedCentralID PMC7546080