Latent Class Mixed Models Research Articles

Ambient Air Pollution and Long-Term Trajectories of Episodic Memory Decline among Older Women in the WHIMS-ECHO Cohort.

Background:Episodic memory decline varies by age and underlying neuropathology. Whether ambient air pollution contributes to the heterogeneity of episodic memory decline in older populations remains unclear.Objectives:We estimated associations between air pollution exposures and episodic memory decline according to pollutant, exposure time window, age, and latent class subgroups defined by episodic memory trajectories.Methods:Participants were from the Women’s Health Initiative Memory Study–Epidemiology of Cognitive Health Outcomes. Older women (; 74–92 years of age) completed annual (2008–2018) episodic memory assessments using the telephone-based California Verbal Learning Test (CVLT). We estimated 3-y average fine particulate matter [PM with an aerodynamic diameter of ()] and nitrogen dioxide () exposures at baseline and 10 y earlier (recent and remote exposures, respectively), using regionalized national universal kriging. Separate latent class mixed models were used to estimate associations between interquartile range increases in exposures and CVLT trajectories in women and , adjusting for covariates.Results:Two latent classes were identified for women (), “slow-decliners” { [95% confidence interval (CI): , ] and “fast-decliners” [ (95% CI: , )]}. In the slow-decliner class, but not the fast-decliner class, exposures were associated with a greater decline in CVLT scores over time, with a stronger association for recent vs. remote exposures [ (95% CI: , ) per and (95% CI: , 0.01) per , respectively]. Among women (), the largest latent class comprised “steady-decliners” [ (95% CI: , )], whereas the second class, “cognitively resilient”, had no decline in CVLT on average. was not associated with episodic memory decline in either class. A increase in recent was associated with nonsignificant acceleration of episodic memory decline in the -y-old fast-decliner class [ (95% CI: , 0.04)], and in the cognitively resilient class [ (95% CI: , 0.03)] and steady-decliner class [ (95% CI: , 0.05)]. Associations with recent exposure in women were stronger and statistically significant when 267 women with incident probable dementia were excluded [e.g., (95% CI: , ) for the cognitively resilient class]. In contrast with changes in CVLT over time, there were no associations between exposures and CVLT scores during follow-up in any subgroup.Discussion:In a community-dwelling U.S. population of older women, associations between late-life exposure to ambient air pollution and episodic memory decline varied by age-related cognitive trajectories, exposure time windows, and pollutants. https://doi.org/10.1289/EHP7668

Early pregnancy exposure to metals and maternal depressive symptom trajectories in Project Viva

BACKGROUND AND AIM: Exposure to neurotoxicant metals, hypothesized to alter dopaminergic function, have not been sufficiently explored as a risk factor for maternal depression. We evaluated the extent to which early pregnancy blood levels of essential and non-essential metals were prospectively associated with depressive symptom trajectories from mid-pregnancy to 7-years postpartum. METHODS: Participants were 872 women in Project Viva, a longitudinal Boston-area cohort recruited during pregnancy (1999-2002). We measured levels of 11 metal(loid)s (arsenic, barium, cadmium, cesium, copper, mercury, magnesium, manganese, lead, selenium, zinc) in maternal first trimester erythrocytes. We assessed depressive symptoms via the Edinburgh Postnatal Depression Scale at mid-pregnancy and four postpartum timepoints (6 months, 1, 3, 7 years). Utilizing latent class mixed modeling, we identified three depressive symptom trajectories: stable low (86%); elevated perinatal symptoms, then decreasing (9%); and moderate perinatal symptoms, then increasing (5%), among women with ≥3 repeat measures. Adjusting for maternal sociodemographics and dietary intake, we used multinomial logistic regression to test associations of individual metal levels with odds of having an increasing or decreasing trajectory. RESULTS:In our cohort of moderately high socioeconomic status participants (e.g., 75% college graduate), correlations between metal levels were mostly positive (Spearman: 0.01-0.60), except for negative correlations between barium and all other elements. Compared to having a stable low trajectory, a doubling of blood arsenic was associated with 1.27 (95% confidence interval: 1.01, 1.59) odds of an increasing depressive trajectory. Greater prenatal mercury blood level was associated with lower odds of having a decreasing trajectory [0.88 (95% confidence interval: 0.77, 1.00)]. All other first trimester metal levels were not associated with depressive symptom trajectory. CONCLUSIONS:In our study investigating associations of several early pregnancy metals with maternal depressive symptom trajectories, we did not observe strong evidence of associations. Residual confounding or multiple testing may explain observed associations with arsenic and mercury. KEYWORDS: Mental health outcomes, heavy metals, chemical exposures, female

Association Between Longitudinal Blood Pressure Trajectory and the Progression of Chronic Kidney Disease: Results From the KNOW-CKD.

[Figure: see text].

Oxygen-sensitive Magnetic Resonance Imaging: A Noninvasive Step Forward for Diagnosing Vasculopathy in the Cardiac Allograft.

Oxygen-sensitive Magnetic Resonance Imaging: A Noninvasive Step Forward for Diagnosing Vasculopathy in the Cardiac Allograft.

Disentangling the Role of HIV Exposure on Linear Growth Among Mother-Infant Dyads in Western Kenya: A Comparison of Three Growth Models

ObjectivesHIV-exposed and uninfected (HEU) infants may experience altered growth compared to HIV-unexposed and uninfected (HUU) infants. Most studies to date have used analytic techniques that do not reflect the dynamic trends in infant growth patterns. We therefore sought to evaluate growth patterns using 3 different analytic methods and examine the role of HIV on growth with each method. MethodsRepeated measures for infant anthropometrics were taken from 6 wk to 23 mo of age in the former Nyanza region, Kenya (n = 310, 52% HEU, 50% male). We used (1) Latent Class Mixed Modeling (LCMM) to identify length-for-age z-score (LAZ) trajectory classes and then multinomial logistic regression to assess how HIV exposure status predicted trajectory class membership, adjusting for covariates (e.g., maternal height, food insecurity); (2) SuperImposition by Translation and Rotation (SITAR) to estimate length in terms of size and velocity (rate of growth) differences by maternal HIV status; and (3) longitudinal regression, the conventional method for analyzing growth, to estimate differences in length and LAZ based on HIV exposure. ResultsAt 6 wks of age, HEU infants had a mean LAZ score of –1.03 ± 1.85 compared to –1.10 ± 1.83 for HUU infants. LCMM identified four LAZ trajectory classes (average 5.1 measurements/infant). Across time, class one LAZ scores remained near 1; class two declined 0 to –1; class three remained near –1; and class four fluctuated between –2 and –3. In logistic regression models, HEU infants were less likely to belong to classes one (RR = 0.3; 95% CI: 0.1,0.9) and two (RR = 0.4; 95% CI: 0.2,0.7) relative to class three. Similarly, SITAR estimated that HEU infants were on average 0.62 cm shorter than HUU infants across the study (95% CI: –1.3,0.1) but there were no differences in mean linear growth velocities. Longitudinal regression models predicted that mean stature for HEU infants was 0.8 cm (95% CI: –1.5, –0.1) shorter and that mean LAZ for HEU infants was 0.4 points (95% CI: –0.7, –0.1) lower compared to HUU infants. ConclusionsAcross the 3 methods, HEU infants were shorter than HUU infants during the first 23 mo of life. Compared to longitudinal regression, advanced modeling with LCMM and SITAR allows for a more flexible assessment of the altered growth patterns HEU infants experience. Funding SourcesNIH.

Prenatal Weight Change Trajectories Among Women With Obesity

ObjectivesCurrent gestational weight change (GWC) recommendations were established by the Institute of Medicine (IOM) with limited evidence for pattern and timing of weight change across pregnancy. Likewise, the recommendation of 5–9kg for women with obesity does not differentiate by severity of obesity. We sought to describe GWC trajectory patterns among a large, contemporary cohort of women with obesity. MethodsElectronic medical records were abstracted for 22,987 women with obesity (BMI ≥ 30.0) who delivered a singleton pregnancy at Kaiser Permanente Northern California between 2008–2013. We modeled GWC for women with a measured prepregnancy weight, at least three prenatal visit weights, and a normal oral glucose tolerance test at 20 wk gestation. Weight change trajectories were modeled using flexible latent class mixed modeling (package LCMM) in R. ResultsBefore pregnancy, 60.4% of women had class 1 obesity (BMI 30.0–34.9kg/m2), 24.8% class 2 (35.0–39.9), and 14.8% class 3 (≥40.0). Most women identified as White (37.3%) or Hispanic (37.3%), with 12.4% Black and 8.5% Asian. Five GWC trajectory groups across pregnancy were identified, each with a distinct pattern of weight change before 15 wk followed by weight gain thereafter (<15 wk/>15 wk): 12.2% exhibited HighLoss/SteadyGain with mean GWC at delivery of 0.9 ± 3.1kg (mean ± sd); 25.8% LowLoss/SteadyGain (6.4 ± 2.4); 30.4% LowGain/ModerateGain (11.4 ± 2.5); 23.0% ModerateGain/HighGain (16.8 ± 3.1); and 8.6% HighGain/HighGain (24.5 ± 5.0). Prepregnancy obesity classes were represented across GWC trajectory groups but were predominantly in two groups: those with class 1 obesity exhibited LowGain/ModerateGain (32.1%), class 2 was split between LowGain/ModerateGain (29.0%) and LowLoss/SteadyGain (28.8%), and class 3 exhibited LowLoss/SteadyGain (31.8%). Only the LowLoss/SteadyGain (25.8%) group demonstrated total GWC at delivery within the IOM recommendation. ConclusionsAmong women with obesity, GWC was not linear or uniform, and current recommendations do not reflect nuances of the prenatal period. Gestational weight change in women with obesity can be flexibly modeled using latent classes, an important next step for prenatal weight change guidance. Funding SourcesANDF, KPNC.

Trajectories of Systolic Blood Pressure in Children: Risk Factors and Cardiometabolic Correlates

To identify systolic blood pressure (SBP) percentile trajectories in children and to describe the early-life risk factors and cardiometabolic correlates of those trajectories. Using age-, sex-, and height-specific SBP percentiles based on the American Academy of Pediatrics reference, we examined SBP trajectories using latent class mixed models from ages 3 to 8years (n=844) from the Growing Up in Singapore Towards healthy Outcomes-study, a Singaporean mother-offspring cohort study. We analyzed associations between SBP trajectories and early-life risk factors using multinomial logistic regression and differences across trajectories in cardiometabolic outcomes using multiple linear regression. Children were classified into 1 of 4 SBP percentile trajectories: "low increasing" (15%), "high stable" (47%), "high decreasing" (20%), and "low stable" (18%). Maternal hypertension during early pregnancy was a predictor of the "high stable" and "low increasing" SBP trajectories. Rapid child weight gain in the first 2years of life was only associated with the "high stable" trajectory. Compared with children in the "low stable" trajectory, children in the "high stable" SBP trajectory had greater body mass index z scores, sum of skinfold thicknesses, waist circumference from ages 3 to 8years, and abdominal adipose tissue (milliliters) at 4.5years (adjusted mean difference [95% CI]: superficial and deep subcutaneous abdominal adipose tissue: 115.2 [48.1-182.3] and 85.5 [35.2-135.8]). Their fat mass (kilograms) (1.3 [0.6-2.0]), triglyceride levels (mmol/L) (0.10 [0.02-0.18]), and homeostasis model assessment of insulin resistance (0.28 [0.11 0.46]) at age 6years were also greater but not their arterial thickness and stiffness. Reducing maternal blood pressure during pregnancy and infant weight gain in the first 2years of life might help to prevent the development of high SBP.

Latent Class Analysis to Predict Outcomes of Early High-Intensity Physical Therapy After Total Knee Arthroplasty, Based on Longitudinal Trajectories of Walking Speed.

To (1) classify patients who are recovering from total knee arthroplasty (TKA) based on walking speed during an early physical therapy program, and (2) assess whether walking-speed trajectory predicts performance on the timed up-and-go (TUG) test. Cohort study. We included 218 patients from a 10-day physical therapy program after TKA. A latent class mixed model was used to classify patients according to their walking-speed trajectory during the program. We assessed the change in TUG test score from pre-TKA to 6 weeks and 1 year after TKA. The association between change in TUG test score and walking-speed trajectory was assessed using multivariable regression. There were 2 groups with distinct walking-speed trajectories: a high-gain group (46%) and a low-gain group (54%). There was no significant association between change in TUG test score and walking-speed trajectory after TKA and physical therapy. Function (based on TUG test performance) improved for all patients 1 year after TKA, irrespective of walking-speed trajectory (ie, high or low gain) early in postoperative physical therapy. Although we distinguished different groups based on functional outcomes during physical therapy, the clinical relevance of classifying patients based on walking speed remains unclear, as it did not predict short- and long-term functional outcomes. J Orthop Sports Phys Ther 2021;51(7):362-371. Epub 10 May 2021. doi:10.2519/jospt.2021.10299.

Mental health responses to the COVID-19 pandemic: a latent class trajectory analysis using longitudinal UK data

The mental health of the UK population declined at the onset of the COVID-19 pandemic. Convenience sample surveys indicate that recovery began soon after. Using a probability sample, we tracked mental health during the pandemic to characterise mental health trajectories and identify predictors of deterioration. This study was a secondary analysis of five waves of the UK Household Longitudinal Study (a large, national, probability-based survey that has been collecting data continuously since January, 2009) from late April to early October, 2020 and pre-pandemic data taken from 2018-19. Mental health was assessed using the 12-item General Health Questionnaire (GHQ-12). We used latent class mixed models to identify discrete mental health trajectories and fixed-effects regression to identify predictors of change in mental health. Mental health was assessed in 19 763 adults (≥16 years; 11 477 [58·1%] women and 8287 [41·9%] men; 3453 [17·5%] participants from minority ethnic groups). Mean population mental health deteriorated with the onset of the pandemic and did not begin improving until July, 2020. Latent class analysis identified five distinct mental health trajectories up to October 2020. Most individuals in the population had either consistently good (7437 [39·3%] participants) or consistently very good (7623 [37·5%] participants) mental health across the first 6 months of the pandemic. A recovering group (1727 [12·0%] participants) showed worsened mental health during the initial shock of the pandemic and then returned to around pre-pandemic levels of mental health by October, 2020. The two remaining groups were characterised by poor mental health throughout the observation period; for one group, (523 [4·1%] participants) there was an initial worsening in mental health that was sustained with highly elevated scores. The other group (1011 [7·0%] participants) had little initial acute deterioration in their mental health, but reported a steady and sustained decline in mental health over time. These last two groups were more likely to have pre-existing mental or physical ill-health, to live in deprived neighbourhoods, and be of Asian, Black or mixed ethnicity. Infection with SARS-CoV-2, local lockdown, and financial difficulties all predicted a subsequent deterioration in mental health. Between April and October 2020, the mental health of most UK adults remained resilient or returned to pre-pandemic levels. Around one in nine individuals had deteriorating or consistently poor mental health. People living in areas affected by lockdown, struggling financially, with pre-existing conditions, or infection with SARS-CoV-2 might benefit most from early intervention. None.

Joint Latent Class Analysis of Oral Anticoagulation Use and Risk of Stroke or Systemic Thromboembolism in Patients with Atrial Fibrillation.

Oral anticoagulation (OAC) is recommended to reduce the risk of stroke or systemic thromboembolism (TE) in atrial fibrillation (AF). In this study, we applied novel joint latent class mixed models to identify heterogeneous patterns of trajectories of OAC use and determined how these trajectories are associated with risks of thromboembolic outcomes. We used 2013-2016 claims data from a 5% random sample of Medicare beneficiaries, obtained from the Centers for Medicare and Medicaid Services. Our study sample included 16,399 patients newly diagnosed with AF in 2014-2015 who were followed for 12 months after the first AF diagnosis and filled at least one OAC prescription in this time period. OAC use was defined as the number of days covered with OACs every 30-day interval after the first AF diagnosis. We used a joint latent class mixed model to simultaneously evaluate the longitudinal patterns of OAC use and time to stroke or TE, while adjusting for age, race, CHAD2S2-VASc score and HAS-BLED score. Five classes of OAC use patterns were identified: late users (17.8%); late initiators (12.5%); early discontinuers (18.6%); late discontinuers (15.4%); and continuous users (35.6%). Compared with continuous users, the risk of stroke or TE was higher for participants in the late initiators (hazard ratio [HR] 1.73, 95% confidence interval [CI] 1.49-2.01) and late discontinuers (HR 1.23, 95% CI 1.04-1.45) classes. Late initiators and late discontinuers had a higher risk of stroke or TE than continuous users. Early initiation and continuous OAC use is important in preventing stroke and TE among patients diagnosed with AF.

Association of fourteen years diet quality trajectories and type 2 diabetes mellitus with related biomarkers.

Diet quality fluctuates throughout one’s adulthood, yet it remains unclear how long-term diet quality changes are related to type 2 diabetes mellitus (T2DM) and its biomarkers. We aimed to examine the association of long-term diet quality with T2DM and its biomarkers. Diet quality was assessed by the revised DBI-07, in which diet quality distance levels (DQD) represented the overall diet quality. Participants were clustered into classes sharing similar DQD levels using latent class mixed model. We used Cox regression and random effect linear regression to assess DQD trajectories’ association with T2DM and its biomarkers. Three DQD trajectories were derived: moderate to gradual decrease, high to moderate, high stable DQD level representing 2.14%, 6.18% and, 91.68% of the population. Compared to class 1, class 2 and 3 were associated with an increased risk of T2DM [HR=4.40; 95%CI: 2.02-9.59]; [HR=3.68; 95% CI: 2.11-6.43]. When class 3 was used as a reference, class 1 was also associated with an increased risk of TDM [HR=2.71; 95%CI: 1.55-4.73]. Our findings suggest that a long-term unhealthy diet is associated with an increased risk of T2DM. Gradually reducing DQD level may not make a difference, which establishes the importance of promoting healthy eating habits at early adulthood.

Identifying longitudinal-growth patterns from infancy to childhood: a study comparing multiple clustering techniques.

Most studies on children evaluate longitudinal growth as an important health indicator. Different methods have been used to detect growth patterns across childhood, but with no comparison between them to evaluate result consistency. We explored the variation in growth patterns as detected by different clustering and latent class modelling techniques. Moreover, we investigated how the characteristics/features (e.g. slope, tempo, velocity) of longitudinal growth influence pattern detection. We studied 1134 children from The Applied Research Group for Kids cohort with longitudinal-growth measurements [height, weight, body mass index (BMI)] available from birth until 12 years of age. Growth patterns were identified by latent class mixed models (LCMM) and time-series clustering (TSC) using various algorithms and distance measures. Time-invariant features were extracted from all growth measures. A random forest classifier was used to predict the identified growth patterns for each growth measure using the extracted features. Overall, 72 TSC configurations were tested. For BMI, we identified three growth patterns by both TSC and LCMM. The clustering agreement was 58% between LCMM and TS clusters, whereas it varied between 30.8% and 93.3% within the TSC configurations. The extracted features (n = 67) predicted the identified patterns for each growth measure with accuracy of 82%-89%. Specific feature categories were identified as the most important predictors for patterns of all tested growth measures. Growth-pattern detection is affected by the method employed. This can impact on comparisons across different populations or associations between growth patterns and health outcomes. Growth features can be reliably used as predictors of growth patterns.

Ascending Growth is Associated with Offspring Adiposity in Pregnancies Complicated with Obesity or Gestational Diabetes

Early growth is associated with childhood adiposity, but the influence of lifestyle remains unknown. This work aimed to investigate the association of growth profiles from high-risk pregnancies with adiposity at age 5 years, taking into account lifestyle and several antenatal/postnatal exposures. This prospective cohort study. 609 children born during the Finnish Gestational Diabetes Prevention Study (RADIEL), recruiting women with body mass index (BMI) greater than or equal to 30 and/or prior gestational diabetes mellitus (GDM) (2008-2013). Altogether 332 children attended the 5-year follow-up (2014-2017). Main outcome measures included growth profiles based on ponderal index (PI = weight/height3), investigated using latent class mixed models. Adiposity was assessed with anthropometrics and body composition (InBody720). We identified 3 growth profiles: ascending (n = 82), intermediate (n = 351), and descending (n = 149). Children with ascending growth had a higher body fat percentage, ISO-BMI, and waist circumference (P < .05) at age 5 years. Ascending (β 4.09; CI, 1.60-6.58) and intermediate (β 2.27; CI, 0.50-4.03) profiles were associated with higher fat percentage, even after adjustment for age, sex, gestational age, diet, physical activity, education, and prepregnancy BMI. Similar associations existed with ISO-BMI. After adjusting for age and education, ascending growth was associated with prepregnancy BMI (odds ratio [OR] 1.06; CI, 1.01-1.12), primiparity (OR 3.07; CI, 1.68-5.62), cesarean delivery (OR 2.23; CI, 1.18-4.21), and lifestyle intervention (OR 2.56; CI, 1.44-4.57). However, meeting the intervention goals and exclusive breastfeeding for 3 months or more were associated with lower odds of ascending growth. Accelerated early growth was associated with higher adiposity in 5-year-old children from high-risk pregnancies, even when adjusted for lifestyle. Reducing cesarean deliveries and promoting breastfeeding may be beneficial for postnatal growth.

Feeding patterns and BMI trajectories during infancy: a multi-ethnic, prospective birth cohort

BackgroundMilk feeding type (exclusive breastfeeding [EBF], formula feeding or mixed feeding) and timing of complementary feeding (CF) have been associated with infant growth. However, studies evaluating their combined role, and the role of ethnicity, are scarce. We examined associations of feeding patterns (milk feeding type combined with timing of CF) with infant body mass index (BMI) trajectories and potential ethnic-specific associations.MethodsInfant feeding and BMI data during the 1st year of life from 3524 children (Dutch n = 2880, Moroccan n = 404 and Turkish n = 240) from the Amsterdam Born Children and their Development (ABCD) cohort were used. Six feeding patterns were defined: EBF/earlyCF, EBF/lateCF (reference), formula/earlyCF, formula/lateCF, mixed/earlyCF and mixed/lateCF. A covariate adjusted latent class mixed model was applied to simultaneously model BMI trajectories and associations with feeding patterns. Potential ethnic differences in the associations were studied in a separate model where interactions between ethnicity and feeding patterns were included.ResultsFour distinct BMI trajectories (low, mid-low, mid-high and high) were identified. Feeding pattern of formula/earlyCF was associated with lower odds for low (OR: 0.43; 95% CI: 0.25, 0.76) or mid-high (0.28; 0.16, 0.51) (ref: high) trajectory compared with EBF/lateCF pattern (ref). An ethnic-specific model revealed that among Dutch infants, formula/earlyCF pattern was associated with lower odds for low trajectory (0.46; 0.24, 0.87), whereas among Turkish/Moroccan infants almost all feeding patterns were associated with lower odds for the low trajectory (ref: high).ConclusionInfant feeding patterns are associated with early BMI trajectories with specific ethnic differences. Future studies should take the role of ethnicity into account in the associations between infant feeding and growth.

Characterization of disease course and remission in early seropositive rheumatoid arthritis: results from the TACERA longitudinal cohort study.

Background:To characterise disease course and remission in a longitudinal observational study of newly diagnosed, initially treatment-naïve patients with seropositive rheumatoid arthritis (RA).Methods:Patients with early untreated seropositive RA were recruited from 28 UK centres. Multiple clinical and laboratory measures were collected every 3 months for up to 18 months. Disease activity was measured using the 28-joint Disease Activity Score with C-reactive protein (DAS28-CRP) and Simplified Disease Activity Index (SDAI). Logistic regression models examined clinical predictors of 6-month remission and latent class mixed models characterised disease course.Results:We enrolled 275 patients of whom 267 met full eligibility and provided baseline data. According to SDAI definition, 24.3% attained 6-month remission. Lower baseline Health Assessment Questionnaire (HAQ) and SDAI predicted 6-month remission (p = 0.013 and 0.011). Alcohol intake and baseline prescribing of methotrexate with a second disease-modifying antirheumatic drug (DMARD; vs monotherapy without glucocorticoids) were also predictive. Three distinct SDAI trajectory subpopulations emerged; corresponding to an inadequate responder group (6.5%), and higher and lower baseline activity responder groups (22.4% and 71.1%). Baseline HAQ and Short Form-36 Health Survey – Mental Component Score (SF-36 MCS) distinguished these groups. In addition, a number of baseline clinical predictors correlated with disease activity severity within subpopulations. Beneficial effects of alcohol intake were found across subpopulations.Conclusion:Three distinct disease trajectory subpopulations were identified. Differential effects of functional and mental well-being, alcohol consumption, and baseline RA medication prescribing on disease activity severity were found across subpopulations. Heterogeneity across trajectories cannot be fully explained by baseline clinical predictors. We hypothesise that biological markers collected early in disease course (within 6 months) may help patient management and better targeting of existing and novel therapies.

Trajectories of medication adherence in patients with Bipolar Disorder along 2 years-follow-up

BackgroundBipolar disorder (BD) is a chronic and severe mental illness. It requires a non-discontinued pharmacological treatment to prevent mood recurrences but nonadherence to medication is frequent. To this date, medication adherence in BD has been mostly evaluated in cross-sectional studies and often considered as a stable trait. We aimed to study medication adherence using a prospective person-oriented approach. Methods1627 BD patients were followed on a 2 years period and assessed every 6 months. Medication adherence was evaluated at each visit with the Medication Adherence Rating Scale (MARS). A latent class mixed model (LCMM) was used to identify trajectory classes of adherence over time. Regression analyses and linear mixed model were used to search for predictors and covariables of the trajectories. ResultsThree distinct and robust trajectories of medication adherence have been identified: one that starts poorly and keeps deteriorating (4.8%), one that starts poorly but improves (9%) and one that starts well and keeps improving (86.2%). A good tolerance to psychotropic medications, low depressive symptoms, the absence of comorbid eating disorders and anticonvulsant medication were associated to a better prognosis of adherence. Along the follow-up, the lower were the depressive symptoms, the better was the medication adherence (p < .001) LimitationsThe use of a single measure of medication adherence although it is a validated instrument and a possible positive selection bias that might limit the generalization of our findings. ConclusionsThis study demonstrates that medication adherence in BD patients is a heterogeneous and potentially variable phenomenon.

The neurodevelopmental nature of attention-deficit hyperactivity disorder in adults.

Population studies have suggested that most adults with attention-deficit hyperactivity disorder (ADHD) did not have the disorder in childhood, challenging the neurodevelopmental conceptualisation of ADHD. Arbitrary definitions of age at onset and lack of defined trajectories were accounted for the findings. The objective of this study was to assess the proportion of individuals presenting with either a neurodevelopmental trajectory or late-onset disorder, and to assess risk factors associated with them. Data of 4676 individuals from the 1993 Pelotas birth cohort at 11, 15, 18 and 22 years of age were used. Polythetic and latent class mixed model analyses were performed to define ADHD trajectories from childhood to adulthood, and characterise the neurodevelopmental or late-onset courses. Regression models were applied to assess factors associated with different trajectories. Classical polythetic analyses showed that 67% of those with ADHD at 22 years of age had a neurodevelopmental course of the disorder. Latent class mixed model analysis indicated that 78% of adults with ADHD had a trajectory of persistent symptoms, more common in males. The remaining adults with ADHD had an ascending symptom trajectory that occurred after puberty, with late-onset ADHD associated with female gender and higher IQ. Both polythetic and latent trajectories analyses provided empirical evidence supporting that the large majority of adults with ADHD had a neurodevelopmental disorder.

Identifying and predicting heterogeneity in cognitive decline among individuals with prodromal Alzheimer's disease using a latent class analysis

AbstractBackgroundIndividuals with prodromal Alzheimer’s disease (AD) show heterogeneity in their rates of cognitive decline, which is problematic for measuring (therapeutic) effects on cognition. We aimed to identify subgroups with homogeneous cognitive trajectories among individuals with prodromal AD in two large independent cohorts, and investigate which clinical and/or biological AD risk factors predict progression rate.MethodIndividuals with an abnormal amyloid marker (determined by either CSF Aβ‐42 levels or Florbetaben/Flutametamol/Florbetapir‐AV45 PET imaging) and mild cognitive impairment who had longitudinal cognitive data available were selected from the Amsterdam Dementia Cohort and ADNI (N=340, age =71.5±8.4, 43% female). We performed latent class mixed models (LCMM) to identify subgroups based on similarity of their trajectories on the Mini‐Mental State Examination (MMSE), and performed mixture modelling using individual MMSE slopes to validate those subgroups. We compared subgroups on baseline age, sex, years of education, ApoE‐e4 status, CSF tau and phosphorylated tau (p‐tau) and hippocampal volume using univariate logistic regression, and subsequently performed stepwise logistic regression with backward elimination to identify which baseline characteristics best predicted progression.ResultParticipants had 2.3±1.8 repeated assessments and a follow‐up of 2.3±1.6 [range=1‐8] years. LCMM identified three subgroups of MMSE progression, which were confirmed by mixture modelling: Group 1 (n=303, 88%) showing slowest progression (i.e. average 0.7 [range=0.4–1.9] points annual decline on MMSE); Group 2 (n=30, 10%) intermediate progression (average 2.6 [range=1.9–4.2] points annual decline on MMSE) and Group 3 (n=7, 2%) showing fastest progression (average 4.7 [range=3.3–6.2] points annual decline on MMSE) (Figure 1). No effects on progression were observed for age, sex, education and ApoE‐e4 (Table 1). Compared to individuals who slowly progressed, those with intermediate‐to‐fast progression had higher tau (OR=1.01, 95%CI=[1.01‐1.02], p&lt;.001) and p‐tau (OR=1.03, 95%CI=[1.02‐1.05], p&lt;.001) levels and lower hippocampal volume (OR=0.57, 95%CI=[.37‐.84], p=.006) at baseline. The model including both p‐tau (OR=1.03, 95%CI=[1.02‐1.05], p&lt;.001) and hippocampal volume (OR=0.55, 95%CI=[.34‐.85], p=.01) best predicted fast progression.ConclusionAmong individuals with prodromal AD, we identified three subgroups with different rates of decline on the MMSE, and demonstrated that baseline neurodegeneration markers were associated with faster decline. These findings have practical implications for clinical trial enrollment and predicting disease progression.

Body Mass Index Mediates the Association between Growth Trajectories and Cardiometabolic Risk in Children.

Background: Few studies have published mediation analyses to quantify the role concurrent BMI plays in the relationship between growth and cardiometabolic risk (CMR) outcomes. Methods: We used data from a longitudinal cohort study conducted in children aged 0-60 months through The Applied Research Group for Kids (TARGet Kids!) in Canada. Four age and sex standardized BMI (zBMI) trajectories were identified using latent class mixed models. CMR were assessed using a CMR score. Concurrent zBMI was the zBMI measured on the same visit as CMR. Mediation analyses were performed comparing each trajectory with the reference trajectory. Results: One thousand one hundred sixty-five children were included. On average, compared with being in the stable low trajectory, being in the catch-up trajectory was associated with an increased CMR score of 0.42, 0.28 of which was mediated through concurrent zBMI [95% confidence interval (CI) 0.17 to 0.41, p = 0.001]; being in the stable high trajectory was associated with an increased CMR score of 0.23, 0.24 through concurrent zBMI (95% CI 0.18 to 0.31, p < 0.001). Similarly, being in the rapid accelerating trajectory was associated with an increased CMR score of 1.43, 1.18 of which was through concurrent zBMI (95% CI 0.89 to 1.50, p < 0.001). Conclusions: There was a strong evidence that the effect of BMI trajectories on CMR was indirect via concurrent BMI. It is important for researchers to choose the most appropriate analytic method based on the study hypothesis to understand the total or direct effect of growth patterns on cardiometabolic disease risk in children.

Monoclonal Protein Response Trajectory and Survival in Newly Diagnosed Multiple Myeloma: A Retrospective Cohort Study

Background: Multiple myeloma is a heterogenous plasma cell cancer characterized by expression of a monoclonal protein (M-protein) or free light chains (FLC). Although survival has improved over the past 15 years, outcomes are variable and not fully predicted by conventional prognostic markers. Survival is associated with depth and duration of response to treatment but it is not known if the trajectory of response is associated with survival. We hypothesized that the trajectory of M-protein over the first six months of treatment in patients with newly diagnosed multiple myeloma can identify distinct clinical groups and is an independent predictor of overall survival (OS). The objectives of our study are to: 1) Identify groups of patients with different M-protein trajectories using trajectory analysis; 2) Evaluate patient and illness factors associated with different M-protein trajectories; and 3) Determine whether M-protein trajectory is an independent predictor of OS in newly diagnosed multiple myeloma. Methods: We conducted a retrospective cohort review of all patients with newly diagnosed multiple myeloma in Manitoba between 2012 and 2017 who had a measurable M-protein at presentation and who were initially treated with bortezomib-based chemotherapy. Baseline characteristics and monthly response assessments up to six months were collected and analyzed using trajectory analysis. We used latent class mixed models (LCMM) to produce trajectory groups of M-protein response over time. We used the Bayesian Information Criterion and clinical validity characteristics to select the optimal trajectory model. We developed a multinomial logistic model to evaluate the association of patient and illness characteristics with M-protein trajectories. We constructed a multivariable Cox proportional hazard model adjusted for trajectory group, ISS stage, transplant status, LDH, renal failure and platelet count to evaluate the association of M-protein trajectory on OS in multiple myeloma. Results: 266 patients were included in the trajectory analysis. 154 (57.9%) were male. Mean age was 66.4 years (SD 10.1). 29 (10.9%) were classified as ISS stage 1, 80 (30.1%) as ISS stage 2, and 109 (41.0%) as ISS stage 3. The best fit trajectory model identified three trajectory groups representing distinct M-protein trajectories over time (Figure 1). We plotted the trajectory of the absolute value for M-protein as the percentage change from baseline did not separate the population into distinct groups. Baseline bone marrow plasma percentage (p&amp;lt;0.01), albumin (p&amp;lt;0.01), and hemoglobin (p&amp;lt;0.01) were significantly associated with trajectory group membership. The median follow-up was 2.74 years for the OS analysis. The median OS was not reached for the low group, 4.3 years for the moderate group, and 5.4 years for the high group. In a multivariable Cox proportional hazard model, we found that trajectory group was not significantly associated with OS (moderate group HR 1.18, 95%CI 0.48-2.91, high group HR 1.10, 95%Cl 0.62-1.88). Variables significantly associated with OS included ISS stage (p&amp;lt;0.025), LDH (p&amp;lt;0.022), platelet count (p&amp;lt;0.001), and receipt of a stem cell transplant (p&amp;lt;0.001). Discussion and Conclusions: In this retrospective cohort study of patients with newly diagnosed multiple myeloma, we identified three distinct groups based on their M-protein trajectory over time. The identified groups have different mean baseline M-protein values and follow different evolutionary patterns over time. While M-protein trajectory group was not associated with differences in OS, the small sample size and short follow-up may limit interpretation of the results. Our analyses only looked at the M-protein trajectory over the first six months after treatment start, which indicates an early response to treatment. It is possible this early response does not play a large role in the OS of patients with myeloma given the mutliple lines of chemotherapy they may receive. Further studies are planned to assess the achievement of a complete remission at one year or at renal recovery for patients affected with renal involvement. Alternative uses for M-protein trajectory analysis include the trajectory of M-protein prior to relapse or in patients with smoldering myeloma to identify who may benefit from earlier treatment. Disclosures Kotb: Takeda: Honoraria; Merck: Honoraria, Research Funding; Sanofi: Research Funding; Celgene: Honoraria; Janssen: Honoraria; Amgen: Honoraria; Karyopharm: Current equity holder in publicly-traded company.