Latent Class Mixture Models Research Articles

Effect of long-term blood pressure trajectory on the future development of chronic kidney disease: an analysis of data from the Korean National Insurance Health Checkup Study.

Chronic kidney disease (CKD) is a prevalent health issue that causes the irreversible loss of functioning nephrons, end-stage renal disease, cardiovascular disease, and premature mortality. Hypertension is the leading cause of CKD. However, the effect of long-term blood pressure (BP) changes on the development of CKD is still unknown. Therefore, the current study investigated the association between BP trajectory and the future development of CKD. In this study, 246,874 individuals aged ≥40 years who underwent health examinations during the screening period (2002-2009) were evaluated. The systolic BP (SBP) trajectory was determined using latent-class mixture modeling. New-onset CKD was identified during the follow-up period (2010-2019). The association between SBP trajectories and new-onset CKD was assessed. In total, 111,900 adults (53,420 women, 51.9  6.4 years old) presented with 2 SBP trajectory classes: class 1 (n = 66,935) and class 2 (n = 44,965). During the follow-up period, patients with SBP trajectory class 2 had an approximately 2.1-fold increased risk of developing CKD (unadjusted hazard ratio [HR], 2.114; 95% confidence interval [CI], 1.989-2.246, p<0.001). In the multivariate analysis adjusted for other significant variables, SBP trajectory class 2 was significantly associated with CKD in men (HR, 1.093; 95% CI, 1.005-1.189; p=0.037), but not in women (HR, 1.059; 95% CI, 0.947-1.184; p=0.321). An elevated longitudinal BP was associated with a higher incidence of CKD in male participants aged ≥40 years. Nevertheless, further studies are needed to validate the clinical significance of an elevated SBP trajectory on CKD development.

BMI trajectories from birth to young adulthood associate with distinct cardiometabolic profiles.

Numerous studies have investigated links between body mass index (BMI) trajectories and cardiovascular risk, yet discrepancies in BMI measurement duration and timing of the cardiovascular-related outcome evaluation have led to inconsistent findings. We included participants from the Swedish birth cohort (BAMSE) and applied latent class mixture modeling to identify BMI trajectories using data of multiple BMI measures (≥ 4 times) from birth until 24-year follow-up (n = 3204). Subsequently, we analyzed the associations of BMI trajectories with lipids (n = 1974), blood pressure (n = 2022), HbA1c (n = 941), and blood leukocytes (n = 1973) using linear regression. We also investigated the circulating levels of 92 inflammation-related proteins (n = 1866) across BMI trajectories. Six distinct BMI groups were identified, denoted as increasing-persistent high (n = 74; 2.3%), high-accelerated increasing (n = 209; 6.5%), increasing-accelerated resolving (n = 142; 4.4%), normal-above normal (n = 721; 22.5%), stable normal (n = 1608; 50.2%), and decreasing-persistent low (n = 450; 14.1%) BMI groups. The increasing-persistent high and high-accelerated increasing BMI groups had higher levels of total cholesterol [mean difference (95% confidence intervals): 0.30 (0.04-0.56) and 0.16 (0.02-0.31) mmol/L], triglyceride, low-density lipoprotein, hemoglobin A1C [3.61 (2.17-5.54) and 1.18 (0.40-1.98) mmol/mol], and low-density lipoprotein/high-density lipoprotein ratios, but a lower level of high-density lipoprotein than the stable normal BMI group. These two groups also had higher leukocyte cell counts and higher circulating levels of 28 inflammation-related proteins. No increased cardiometabolic markers were observed in the increasing-accelerated resolving BMI group. Participants with persistently high or accelerated increasing BMI trajectories from birth to young adulthood have elevated levels of cardiometabolic risk markers at young adulthood than those with stable normal BMI. However, a raised BMI in childhood may not be inherently harmful to cardiometabolic health, provided it does not persist into adulthood.

Body mass index trajectories from birth to early adulthood and lung function development.

Limited studies have investigated the influence of body mass index (BMI) trajectories on lung function covering the entire growth period. We conducted a prospective study utilizing data from the Swedish BAMSE birth cohort. Latent class mixture modelling was employed to examine the diversity in BMI z-scores from birth to 24 years of age. Participants with four or more BMI z-scores were included (n=3204, 78·4%). Pre-bronchodilator (BD) spirometry was tested at 8, 16, and 24 years, while post-BD spirometry, multiple-breath nitrogen washout (for lung clearance index, LCI), and urinary metabolomics data were assessed at 24 years. Six distinct BMI development groups were identified. Compared to the stable normal BMI group, the accelerated increasing BMI group exhibited reduced pre- and post-BD FEV1/FVC ratio z scores (β=-0·26, 95% CI=[-0·44, -0·08], and -0·22, [ -0·39, -0·05], respectively), along with elevated LCI (0·30, [0·22, 0·42]) at 24 years. The persistent high BMI group demonstrated lower FEV1, and FVC z scores growth between 16 and 24 years (-0.24, [-0.42, -0.05], and -0.27, [-0.45, -0.01], respectively), and elevated LCI (0·20, [0·03, 0·39]) at 24 years. However, those impairments were not observed in the accelerated resolving BMI group. Conversely, the persistent low BMI group displayed persistently decreased FEV1, and FVC from 8 to 24 years, as well as decreased lung function growth. Additionally, histidine-related metabolites were associated with pre- and post-BD FEV1 (hypergeometric FDR=0.008 and <0.001, respectively). Early interventions aiming for normal BMI during childhood may contribute to improved lung health later in life.

Trajectory modeling and response prediction in transcranial magnetic stimulation for depression

BackgroundRepetitive transcranial magnetic stimulation (rTMS) therapy could be improved by more accurate and earlier prediction of response. Latent class mixture (LCMM) and non-linear mixed effects (NLME) modeling have been applied to model the trajectories of antidepressant response (or non-response) to TMS, but it is not known whether such models are useful in predicting clinically meaningful change in symptom severity, i.e. categorical (non)response as opposed to continuous scores. MethodsWe compared LCMM and NLME approaches to model the antidepressant response to TMS in a naturalistic sample of 238 patients receiving rTMS for treatment resistant depression, across multiple coils and protocols. We then compared the predictive power of those models. ResultsLCMM trajectories were influenced largely by baseline symptom severity, but baseline symptoms provided little predictive power for later antidepressant response. Rather, the optimal LCMM model was a nonlinear two-class model that accounted for baseline symptoms. This model accurately predicted patient response at 4 weeks of treatment (AUC=0.70, 95 % CI=[0.52–0.87]), but not before. NLME offered slightly improved predictive performance at 4 weeks of treatment (AUC=0.76, 95 % CI=[0.58–0.94], but likewise, not before. ConclusionsIn showing the predictive validity of these approaches to model response trajectories to rTMS, we provided preliminary evidence that trajectory modeling could be used to guide future treatment decisions.

Trajectory of haemoglobin A1c and incidence of cardiovascular disease in patients with type 2 diabetes mellitus.

To evaluate the association between changes in haemoglobin A1c (HbA1c) and the concurrent incidence of cardiovascular disease (CVD) in type 2 diabetes mellitus (T2DM) patients. We conducted a retrospective cohort study among T2DM patients with HbA1c measurement after T2DM diagnosis between August 2009 and September 2010. The patients were classified into six subgroups based on baseline HbA1c (<7%; 7%-7.9%; ≥8%) and age (<65; ≥65 years), and then clustered into classes by HbA1c trajectory and CVD incidence over the 12-year follow-up period using joint latent class mixture models. We explored the HbA1c trajectories and CVD incidences in each latent class. Multinomial logistic regression was used to compare the baseline characteristics among different latent classes. A total of 128 843 T2DM patients were included with a median follow-up period of 11.7 years. Ten latent classes were identified in patients with baseline HbA1c ≥ 8% and age <65 years, while seven classes were identified in the other five groups. Among all the identified latent classes, patients with fluctuating HbA1c trajectories, characterized by alternating periods of increase and decrease, had higher CVD incidences. Male patients, and patients with higher baseline HbA1c and use of antidiabetic drugs were more likely to have a fluctuating HbA1c trajectory. More specifically, patients aged < 65 years with younger age or a smoking habit, and patients aged ≥ 65 years with a longer duration of T2DM were more likely to have a fluctuating HbA1c trajectory. We found that T2DM patients with fluctuating HbA1c trajectories could have a higher CVD risk. Different trajectory-associated characteristics in age subgroups highlight the need for individualized management of T2DM patients.

The CAREPAL-8: a short screening tool for multidimensional family caregiver burden in palliative care

BackgroundFamily caregivers of terminally ill and dying people do not only experience varying levels but also different dimensions of caregiver-related strain and burden. The aim of the study was to develop a short multidimensional screening tool for the detection of burden in family caregivers in palliative care.MethodsFamily caregivers of cancer patients newly admitted to specialist inpatient palliative care (N = 232) completed questionnaires on psychological burden, quality of life, social support and need fulfillment. A latent class mixture model was used to identify discrete classes of family caregivers related to their multidimensional caregiver burden. Multinomial logistic regression analyses were performed to identify the most predictive items from a set of established questionnaires.ResultsFour latent classes of family caregivers were identified: Currently stable caregivers (37%), Caregivers with unmet needs (20%), Psychologically burdened caregivers (30%), and High-risk caregivers (13%). Each of these classes describes a different risk profile of multidimensional family caregiver burden, although family caregivers exhibit high levels of distress across all classes. From a set of 48 items, we identified eight items that predicted the class membership best. These items represent the items of the novel multidimensional screening tool: The 8-item Screening Tool for Family Caregiver Burden in Palliative Care (CAREPAL-8). Except for social support, the items maintained fidelity to the conceptualization of multidimensional caregiver burden used in this study. A preliminary classification system was developed, which has yet to be validated.ConclusionsThis study represents the first step in the establishment of a practical, self-administered screening tool that might help healthcare providers to tailor caregiver care according to their burden in daily practice. Brevity of the 8-item tool might facilitate its use in routine clinical care.

Identifying longitudinal cognitive resilience from cross-sectional amyloid, tau, and neurodegeneration

BackgroundLeveraging Alzheimer’s disease (AD) imaging biomarkers and longitudinal cognitive data may allow us to establish evidence of cognitive resilience (CR) to AD pathology in-vivo. Here, we applied latent class mixture modeling, adjusting for sex, baseline age, and neuroimaging biomarkers of amyloid, tau and neurodegeneration, to a sample of cognitively unimpaired older adults to identify longitudinal trajectories of CR.MethodsWe identified 200 Harvard Aging Brain Study (HABS) participants (mean age = 71.89 years, SD = 9.41 years, 59% women) who were cognitively unimpaired at baseline with 2 or more timepoints of cognitive assessment following a single amyloid-PET, tau-PET and structural MRI. We examined latent class mixture models with longitudinal cognition as the dependent variable and time from baseline, baseline age, sex, neocortical Aβ, entorhinal tau, and adjusted hippocampal volume as independent variables. We then examined group differences in CR-related factors across the identified subgroups from a favored model. Finally, we applied our favored model to a dataset from the Alzheimer’s Disease Neuroimaging Initiative (ADNI; n = 160, mean age = 73.9 years, SD = 7.6 years, 60% women).ResultsThe favored model identified 3 latent subgroups, which we labelled as Normal (71% of HABS sample), Resilient (22.5%) and Declining (6.5%) subgroups. The Resilient subgroup exhibited higher baseline cognitive performance and a stable cognitive slope. They were differentiated from other groups by higher levels of verbal intelligence and past cognitive activity. In ADNI, this model identified a larger Normal subgroup (88.1%), a smaller Resilient subgroup (6.3%) and a Declining group (5.6%) with a lower cognitive baseline.ConclusionThese findings demonstrate the value of data-driven approaches to identify longitudinal CR groups in preclinical AD. With such an approach, we identified a CR subgroup who reflected expected characteristics based on previous literature, higher levels of verbal intelligence and past cognitive activity.

Trajectory Modeling and Response Prediction in Transcranial Magnetic Stimulation for Depression.

Repetitive transcranial magnetic stimulation (rTMS) therapy could be improved by better and earlier prediction of response. Latent class mixture (LCMM) and non-linear mixed effects (NLME) modelling have been applied to model the trajectories of antidepressant response (or non-response) to TMS, but it is not known whether such models can predict clinical outcomes. We compared LCMM and NLME approaches to model the antidepressant response to TMS in a naturalistic sample of 238 patients receiving rTMS for treatment resistant depression (TRD), across multiple coils and protocols. We then compared the predictive power of those models. LCMM trajectories were influenced largely by baseline symptom severity, but baseline symptoms provided little predictive power for later antidepressant response. Rather, the optimal LCMM model was a nonlinear two-class model that accounted for baseline symptoms. This model accurately predicted patient response at 4 weeks of treatment (AUC = 0.70, 95% CI = [0.52-0.87]), but not before. NLME offered slightly improved predictive performance at 4 weeks of treatment (AUC = 0.76, 95% CI = [0.58 - 0.94], but likewise, not before. In showing the predictive validity of these approaches to model response trajectories to rTMS, we provided preliminary evidence that trajectory modeling could be used to guide future treatment decisions.

Trajectories of anhedonia symptoms after traumatic injury

Anhedonia describes the inability or difficulty of experiencing or seeking pleasure. Previous research has demonstrated a relationship between posttraumatic stress disorder (PTSD) or experiencing trauma and anhedonia symptoms; however, little to no work has been done to understand the evolution of anhedonia symptoms after trauma. We aimed to identify anhedonia trajectories following traumatic injury. One hundred ninety-five participants were recruited from the emergency department of a Level-1 Trauma Center after experiencing a traumatic injury. To measure anhedonia symptoms, participants completed the Snaith-Hamilton Pleasure Scale (SHAPS) at 2-weeks, 3-months, and 6-months post-injury. Using latent class mixture modeling, we ran a trajectory analysis with three timepoints of SHAPS scores and compared mental and physical health outcomes across trajectories. Most of the sample fell in the resilient trajectory (85 %), while the remainder were in a remitting trajectory (7 %) where symptoms decreased over time, and a delayed (6 %) trajectory where symptoms did not emerge until 3-months after injury. In the resilient trajectory, there was consistently low levels of PTSD, pain, depression, and anxiety relative to the other trajectories. In the delayed trajectory, depression and PTSD were chronically elevated and pain levels were consistent but mild. In the remitting trajectory, PTSD and depression symptoms decreased over time. Identified anhedonia trajectories mirrored trajectories commonly reported for PTSD symptoms after injury. Evaluating anhedonia trajectories and how they relate to mental health outcomes may inform targeted interventions for traumatic injury patients.

20-year depressive symptoms, dementia, and structural neuropathology in older women.

Thecourse of depressive symptoms and dementia risk is unclear, as are potential structural neuropathological common causes. Utilizing joint latent class mixture models, we identified longitudinal trajectories of annually assessed depressive symptoms and dementia risk over 21 years in 957 older women (baseline age 72.7 years old) from the Women's Health Initiative Memory Study. In a subsample of 569 women who underwent structural magnetic resonance imaging, we examined whether estimates of cerebrovascular disease and Alzheimer's disease (AD)-related neurodegeneration were associated with identified trajectories. Five trajectories of depressive symptoms and dementia risk were identified. Compared to women with minimal symptoms, women who reported mild and stable and emerging depressive symptoms were at the highest risk of developing dementia and had more cerebrovascular disease and AD-related neurodegeneration. There are heterogeneous profiles of depressive symptoms and dementia risk. Common neuropathological factors may contribute to both depression and dementia. Highlights The progression of depressive symptoms and concurrent dementia risk is heterogeneous. Emerging depressive symptoms may be a prodromal symptom of dementia. Cerebrovascular disease and AD are potentially shared neuropathological factors.

Informal Family Care Partner Well-Being Is Diminished in End-Stage Liver Disease.

Chronic liver disease is a significant global neglected public health problem. End-stage liver disease is associated with substantial symptom complexity, disability, and care needs that require assistance from informal family care partners. Research on these care partners' caregiver burden or strain, symptoms, and quality of life is sparse and has not focused on these variables as co-occurring or in the context of the quality of the relationship care partners have with the patients. The purpose of this study was to provide a collective presentation of patterns and determinants of well-being as measured by caregiver strain, depression, sleep, and quality of life in a cohort of informal family care partners for adult outpatients with end-stage liver disease. Care partners (aged >18 years) were recruited from two liver clinics within two tertiary healthcare systems and invited to complete a cross-sectional survey. They completed the Multidimensional Caregiver Strain Index, Patient Health Questionnaire, Pittsburgh Sleep Quality Index, Short Form Health Survey, and Mutuality Scale. Descriptive statistics and latent class mixture modeling were used to analyze these data. The sample was predominantly female and White. The well-being of care partners was diminished. Three distinct classes of well-being were identified: mildly diminished (53.2%), moderately diminished (39.0%), and severely diminished (7.8%). Those at a greater risk of worse well-being were younger and spouses and had poorer relationship quality with the patients. To improve the well-being of care partners in moderately and severely diminished classes, assessing and addressing caregiver strain and co-occurring symptoms is essential. Addressing the strain and symptoms has the potential to maintain or optimize care partners' ability to provide care to patients. Future researchers should include longitudinal and dyadic studies to examine how patients' disease progression and symptoms may affect family care partners' well-being and vice versa.

Classes and predictors of reversal in male patients with congenital hypogonadotropic hypogonadism: a cross-sectional study of six international referral centres

Although some male patients with congenital hypogonadotropic hypogonadism (CHH) undergo spontaneous reversal following treatment, predictors of reversal remain elusive. We aimed to assemble the largest cohort of male patients with CHH reversal to date and identify distinct classes of reversal. This multicentre cross-sectional study was conducted in six international CHH referral centres in Brazil, Finland, France, Italy, the UK, and the USA. Adult men with CHH (ie, absent or incomplete spontaneous puberty by age 18 years, low serum testosterone concentrations, and no identifiable cause of hypothalamic-pituitary-gonadal [HPG] axis dysfunction) were eligible for inclusion. CHH reversal was defined as spontaneous recovery of HPG axis function off treatment. Centres provided common data elements on patient phenotype, clinical assessment, and genetics using a structured, harmonised data collection form developed by COST Action BM1105. Latent class mixture modelling (LCMM) was applied to establish whether at least two distinct classes of reversal could be identified and differentially predicted, and results were compared with a cohort of patients without CHH reversal to identify potential predictors of reversal. The primary outcome was the presence of at least two distinct classes of reversal. A total of 87 male patients with CHH reversal and 108 without CHH reversal were included in the analyses. LCMM identified two distinct reversal classes (75 [86%] in class 1 and 12 [14%] in class 2) on the basis of mean testicular volume, micropenis, and serum follicle-stimulating hormone (FSH) concentration. Classification probabilities were robust (0·998 for class 1 and 0·838 for class 2) and modelling uncertainty was low (entropy 0·90). Compared with class 1, patients in class 2 had significantly larger testicular volume (p<0·0001), no micropenis, and higher serum FSH concentrations (p=0·041), consistent with the Pasqualini syndrome (fertile eunuch) subtype of CHH. Patients without CHH reversal were more likely to have anosmia (p=0·016), cryptorchidism (p=0·0012), complete absence of puberty (testicular volume <4 cm³; p=0·0016), and two or more rare genetic variants (ie, oligogenicity; p=0·0001). Among patients who underwent genetic testing, no patients (of 75) with CHH reversal had a rare pathogenic ANOS1 variant compared with ten (11%) of 95 patients without CHH reversal. Individuals with CHH reversal had a significantly higher rate of rare variants in GNRHR than did those without reversal (nine [12%] of 75 vs three [3%] of 95; p=0·025). Applying LCMM to a large cohort of male patients with CHH reversal uncovered two distinct classes of reversal. Genetic investigation combined with careful clinical phenotyping could help surveillance of reversal after withdrawing treatment, representing the first tailored management approach for male patients with this rare endocrine disorder. National Institutes of Health National Center for Advancing Translational Sciences; Ministry of Health, Rome, Italy; Ministry of University, Rome, Italy; National Institutes of Health Eunice Kennedy Shriver National Institute of Child Health and Human Development; and the Josiah Macy Jr Foundation. For the Italian translation of the abstract see Supplementary Materials section.

An evaluation of emotion recognition, emotion reactivity, and emotion dysregulation as prospective predictors of 12-month trajectories of non-suicidal self-injury in an adolescent psychiatric inpatient sample

BackgroundLittle is known about trajectories of NSSI. We aimed to identify NSSI trajectories in adolescent psychiatric inpatients and emotional processes that differentiate between trajectories. MethodsParticipants were 180 adolescents (71.7 % female; mean age of 14.89 years, SD = 1.35) from a psychiatric inpatient facility. NSSI was assessed at their index hospitalization, as well as 6, and 12 months after discharge. Emotion recognition, emotion reactivity, and emotion dysregulation were assessed at baseline. Latent class mixture modeling was used to identify different NSSI trajectories and ANOVAs were used to evaluate predictors of the trajectories. ResultsAnalyses yielded three NSSI trajectories. These included a stable low-frequency class (90.53 % of sample), a stable moderate-frequency class, and a class characterized by high-frequency NSSI at baseline but that largely resolves by 6-month follow-up. After adjustments for multiple comparisons were made, only emotion regulation at baseline differentiated between the trajectories, with greater overall emotion dysregulation and greater emotional non-acceptance (a facet of emotion dysregulation) characterizing the initially high-frequency class and the stable moderate-frequency class more than the stable low-frequency class (ps < .05). Difficulties engaging in goal-directed behavior when distressed characterized the stable moderate-frequency NSSI class more than the stable low-frequency class (p < .05).LimitationsThe study sample consists predominantly of female and White adolescents and thus may not generalize to other demographic groups. ConclusionsThe current findings suggest that interventions involving emotion regulation with adolescents who engage in NSSI would particularly benefit from a focus on increasing acceptance of emotional experiences.

Development and validation of LCMM prediction algorithms to estimate recovery pattern of postoperative AKI in type A aortic dissection: a retrospective study.

Postoperative acute kidney injury (PO-AKI) is a prevalent complication among patients with acute type A aortic dissection (aTAAD) for which unrecognized trajectories of renal function recovery, and their heterogeneity, may underpin poor success in identifying effective therapies. This was a retrospective, single-center cohort study in a regional Great Vessel Center including patients undergoing aortic dissection surgery. Estimated glomerular filtration rate (eGFR) recovery trajectories of PO-AKI were defined through the unsupervised latent class mixture modeling (LCMM), with an assessment of patient and procedural characteristics, complications, and early-term survival. Internal validation was performed by resampling. A total of 1,295 aTAAD patients underwent surgery and 645 (49.8%) developed PO-AKI. Among the PO-AKI cohort, the LCMM identified two distinct eGFR trajectories: early recovery (ER-AKI, 51.8% of patients) and late or no recovery (LNR-AKI, 48.2% of patients). Binary logistic regression identified five critical determinants regarding poor renal recovery, including chronic kidney disease (CKD) history, renal hypoperfusion, circulation arrest time, intraoperative urine, and myoglobin. LNR-AKI was associated with increased mortality, continuous renal replacement therapies, mechanical ventilation, ICU stay, and hospital stay. The assessment of the predictive model was good, with an area under the curve (AUC) of 0.73 (95% CI: 0.69-0.76), sensitivity of 61.74%, and specificity of 75.15%. The internal validation derived a consistent average AUC of 0.73. The nomogram was constructed for clinicians' convenience. Our study explored the PO-AKI recovery patterns among surgical aTAAD patients and identified critical determinants that help to predict individuals at risk of poor recovery of renal function.

Analysis of the Association Between Changes in the GGT/HDL-C Ratio and the Risk of Diabetes Mellitus Based on a Latent Class Growth Mixed Modeling: A Longitudinal Cohort Study of Adults in China.

Longitudinal cohort analysis was performed to identify the association between changes in the gamma-glutamyl transferase (GGT)/high-density lipoprotein cholesterol (HDL-C) ratio trajectory and the risk of developing diabetes mellitus. This was a retrospective cohort study. We analyzed the latent trajectory classes of changes in the GGT/HDL-C ratio by applying a latent class mixture model with healthy individuals who underwent medical checkups from January 2017 to December 2021 as the study subjects. To analyze the effect of the GGT/HDL-C ratio trajectory classes on new-onset diabetes mellitus, we then applied a multivariate Cox proportional risk regression model. Statistical analysis was performed using the R-software with the LCMM package. The study cohort comprised 3410 participants. All participants were followed up for 5 years, and 95 developed diabetes (4-year incidence of 2.78%). By applying the latent class mixed model, we categorized participants into three trajectory groups: low-stability group (n = 2253), medium-increase group (n = 941), and high-increase group (n = 216). The Cox proportional risk regression model analysis showed that the hazard ratio (95% confidence interval) for the incidence of diabetes mellitus was 1.73 (1.04-2.87) in the medium-increase group and 3.96 (2.11-7.44) in the high-increase group. Moreover, we calculated the estimated model-based levels and linear slopes of the GGT/HDL-C ratios for each age group between 26 and 85 years at 10-year intervals, respectively. The results indicated the strongest correlation between the GGT/HDL-C ratio slope and diabetes in the 46-55 year age group, with an odds ratio of 1.51 (1.25-1.83). A large increase in the GGT/HDL-C ratio was highly associated with the risk of developing diabetes mellitus. This result suggests that vigilance for changes in the GGT/HDL-C ratio trajectory during community health screening can help identify potential patients with diabetes, enabling early intervention and treatment.

Trajectories of Remnant Cholesterol are Associated with Diabetic Foot Ulcer in Adult Patients with Type 2 Diabetes: A Retrospective Cohort Study.

This study aimed to investigate the potential association between long-term variations in remnant cholesterol (RC) levels and the development of diabetic foot ulcers (DFU) in participants with type 2 diabetes (T2D). This was a retrospective cohort study. Variation in RC was assessed by the following metrics: mean, standard deviation (SD), coefficient of variation (CV) and trajectories pattern of RC. To identify RC trajectories, we employed the latent class mixture model. The primary endpoint was the development of DFU, and the time-to-event data were analyzed using Cox regression. A total of 1874 patients with T2D were included, with a median follow-up duration of 4.7 years. Among them, 129 individuals (6.9%) developed DFU. The proportion of DFU was significantly higher in the U-shaped group compared to the median group (P for trend < 0.001). Upon adjustment for confounding variables, the U-shaped trajectory correlated with a higher risk of DFU, demonstrating a hazard ratio (HR) of 2.57 (95% CI, 1.54-4.27). Subgroup analysis showed the U-shaped trajectory had a higher DFU risk regardless of gender (HR=2.40 and 2.81, respectively), glycemic control (HR=1.89 and 7.41, respectively), smoking (HR=2.36 and 2.93, respectively), or hypertension (HR=2.30 and 2.97, respectively). No association was found between mean, SD and CV of RC and DFU. A U-shape trajectory of RC was independently associated with an elevated risk of DFU among patients with T2D.

Tax Avoidance and Firm Risk: New Insights from a Latent Class Mixture Model

ABSTRACT We examine the relations between tax avoidance and two components of firm risk (priced risk and idiosyncratic risk) using latent class mixture models. OLS regressions suggest that tax avoidance is negatively related to both priced risk and idiosyncratic risk. However, results from latent class mixture models reveal that 35.6 (58) percent of the sample exhibits a significant positive association between tax avoidance and priced (idiosyncratic) risk and firm characteristics are strikingly different across the latent classes. We use these differences to develop parsimonious models to predict latent class assignments, which we then use to demonstrate that firms predicted to have a positive relation between tax avoidance and priced risk or idiosyncratic risk have more negative future tax outcomes, including higher relative cash tax rate volatility, larger additions to tax reserves, larger payouts to tax authorities, and more negative tax-related news coverage over the following three years. Data Availability: Data are available from the public sources cited in the text. JEL Classifications: M41; G12; D81.

Patterns of self-care decision-making and associated factors: A cross-sectional observational study

ObjectiveThe aim of this study was to identify for the first time patterns of self-care decision-making (i.e. the extent to which participants viewed contextual factors influencing decisions about symptoms) and associated factors among community-dwelling adults with chronic illness. MethodsThis was a secondary analysis of data collected during the development and psychometric evaluation of the 27-item Self-Care Decisions Inventory that is based on Naturalistic Decision-Making (n = 430, average age = 54.9 ± 16.2 years, 70.2 % female, 87.0 % Caucasian, average number of chronic conditions = 3.6 ± 2.8). Latent class mixture modeling was used to identify patterns among contextual factors that influence self-care decision-making under the domains of external, urgency, uncertainty, cognitive/affective, waiting/cue competition, and concealment. Multivariate multinomial regression was used to identify additional socio-demographic, clinical, and self-care behavior factors that were different across the patterns of self-care decision-making. ResultsThree patterns of self-care decision-making were identified in a cohort of 430 adults. A ‘maintainers’ pattern (48.1 %) consisted of adults with limited contextual influences on self-care decision-making except for urgency. A ‘highly uncertain’ pattern (23.0 %) consisted of adults whose self-care decision-making was largely driven by uncertainty about the cause or meaning of the symptom. A ‘distressed concealers’ pattern (28.8 %) consisted of adults whose self-care decision-making was highly influenced by external factors, cognitive/affective factors and concealment. Age, education, financial security and specific symptoms were significantly different across the three patterns in multivariate models. ConclusionAdults living with chronic illness vary in the extent to which contextual factors influence decisions they make about symptoms, and would therefore benefit from different interventions.

Abstract 16940: Nutrient Intake of Patients With Heart Failure: Different Profiles Reveal Different Intervention Targets

Introduction: Diet quality is an important, but often underappreciated part of heart failure (HF) care. Tailoring of dietary interventions must take into consideration how patients with HF eat. Purpose: To identify distinct dietary profiles representing eating patterns in patients with HF. Methods: This was a secondary analysis of data from a randomized controlled double-blind clinical trial of a nutrition intervention with patients with HF. Baseline dietary data were collected using 24-hour dietary recall that included a weekend and two weekdays. Nutritional data were generated from the Nutrition Data Systems for Research. Nine unique diet components were quantified (along with a total Healthy Eating Index) and included in modeling based on theoretical and empirical knowledge: sodium, added sugars, trans-fat, vitamin C, vitamin D, vitamin E, folate, beta-cryptoxanthin, and lutein + zeaxanthin. Latent class mixture modeling was used to identify distinct dietary profiles among these 9 components representing different eating patterns that could serve as the bases of future interventions. Results: There were 112 participants with an average age 63 ± 12, 59.8% male, and 62.5% with class III or IV HF. Two dietary profiles were identified and were labeled by the prominent dimensional characteristic. The “higher intake” profile (n=79 (69.9% of the sample)) had significantly higher intake levels of sodium, vitamin D, vitamin E, vitamin C, beta-cryptoxanthin, lutein + zeaxanthin, and folate, and a “lower intake” profile (n=34 (30.1% of the sample) had significantly lower intake of these same nutrients. Patients with HF in the lower intake profile also had significant higher quality diets (Healthy Eating Index = 63.7± 13.2) compared with those in the higher intake profile (Healthy Eating Index = 53.9 ± 14.2, p&lt;0.001). Conclusion: Two different nutrient patterns were identified that would be amenable to different interventions; one focused on total calorie reduction and the other focused on optimizing nutrient intake.

73 Identification of 24-Month Cognitive Trajectories Among Clinical High Risk for Psychosis (CHR-P) Using Latent Class Mixture Modeling

Objective:Cohort studies demonstrate that people who later develop schizophrenia, on average, present with mild cognitive deficits in childhood and endure a decline in adolescence and adulthood. Yet, tremendous heterogeneity exists during the course of psychotic disorders, including the prodromal period. Individuals identified to be in this period (known as CHR-P) are at heightened risk for developing psychosis (~35%) and begin to exhibit cognitive deficits. Cognitive impairments in CHR-P (as a singular group) appear to be relatively stable or ameliorate over time. A sizeable proportion has been described to decline on measures related to processing speed or verbal learning. The purpose of this analysis is to use data-driven approaches to identify latent subgroups among CHR-P based on cognitive trajectories. This will yield a clearer understanding of the timing and presentation of both general and domain-specific deficits.Participants and Methods:Participants included 684 young people at CHR-P (ages 12–35) from the second cohort of the North American Prodromal Longitudinal Study. Performance on the MATRICS Consensus Cognitive Battery (MCCB) and the Wechsler Abbreviated Scale of Intelligence (WASI-I) was assessed at baseline, 12-, and 24-months. Tested MCCB domains include verbal learning, speed of processing, working memory, and reasoning &amp; problem-solving. Sex- and age-based norms were utilized. The Oral Reading subtest on the Wide Range Achievement Test (WRAT4) indexed pre-morbid IQ at baseline. Latent class mixture models were used to identify distinct trajectories of cognitive performance across two years. One- to 5-class solutions were compared to decide the best solution. This determination depended on goodness-of-fit metrics, interpretability of latent trajectories, and proportion of subgroup membership (&gt;5%).Results:A one-class solution was found for WASI-I Full-Scale IQ, as people at CHR-P predominantly demonstrated an average IQ that increased gradually over time. For individual domains, one-class solutions also best fit the trajectories for speed of processing, verbal learning, and working memory domains. Two distinct subgroups were identified on one of the executive functioning domains, reasoning and problem-solving (NAB Mazes). The sample divided into unimpaired performance with mild improvement over time (Class I, 74%) and persistent performance two standard deviations below average (Class II, 26%). Between these classes, no significant differences were found for biological sex, age, years of education, or likelihood of conversion to psychosis (OR = 1.68, 95% CI 0.86 to 3.14). Individuals assigned to Class II did demonstrate a lower WASI-I IQ at baseline (96.3 vs. 106.3) and a lower premorbid IQ (100.8 vs. 106.2).Conclusions:Youth at CHR-P demonstrate relatively homogeneous trajectories across time in terms of general cognition and most individual domains. In contrast, two distinct subgroups were observed with higher cognitive skills involving planning and foresight, and they notably exist independent of conversion outcome. Overall, these findings replicate and extend results from a recently published latent class analysis that examined 12-month trajectories among CHR-P using a different cognitive battery (Allott et al., 2022). Findings inform which individuals at CHR-P may be most likely to benefit from cognitive remediation and can inform about the substrates of deficits by establishing meaningful subtypes.