Category: Statistics

J Consult Clin Psychol. 2024 Apr;92(4):236-248. doi: 10.1037/ccp0000884.

ABSTRACT

BACKGROUND: Studies suggest that cognitive behavioral therapies (CBTs) may be efficacious in reducing symptoms of prolonged grief disorder (PGD), but no comprehensive overview and pooled estimate of CBTs’ effect on PGD in adulthood exist. We conducted a systematic review and meta-analysis of randomized controlled trials.

METHOD: Studies were selected independently by two researchers based on a systematic literature search in Pubmed, APA PsycInfo, Web of Science, and Embase. Meta-analyses provided pooled effect sizes for the effects of CBTs on PGD symptoms and secondary outcomes. We explored potential moderators of effect, risk of bias of included studies, and evaluated the quality of the meta-analytical evidence through the Grading of Recommendations, Assessment, Development, and Evaluation system.

RESULTS: The meta-analysis included 22 studies of 2,602 bereaved adults (averaged study M_age = 49 years). CBTs had a statistically significant medium effect on PGD symptoms at postintervention (K = 22, g = 0.65, 95% CI [0.49, 0.81]), and a large effect at follow-up (K = 7, g = 0.90, 95% CI [0.37, 1.43]). Statistically significant small-to-medium effects were found at postintervention on posttraumatic stress symptoms (K = 10, g = 0.74, 95% CI [0.49, 0.98]), depression (K = 19, g = 0.53, 95% CI [0.36, 0.71]), and anxiety (K = 9, g = 0.35, 95% CI [0.22, 0.49]). The effects on PGD remained unchanged when adjusted for possible outliers. None of the moderator analyses reached statistical significance.

CONCLUSION: This review suggests that CBTs are efficacious in reducing PGD symptoms in adulthood. Generalization of findings should be done with caution due to considerable inconsistency and indirectness of meta-analytic evidence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PMID:38573714 | DOI:10.1037/ccp0000884

Clin Cancer Res. 2024 Apr 4. doi: 10.1158/1078-0432.CCR-23-3952. Online ahead of print.

ABSTRACT

PURPOSE: Preventing Anthracycline Cardiovascular Toxicity with Statins (PREVENT) (NCT01988571) randomized breast cancer or lymphoma patients receiving anthracyclines to atorvastatin 40 mg daily or placebo. We evaluated the effects of atorvastatin on oxidative and nitrosative stress biomarkers, and explored whether these biomarkers could explain the lack of effect of atorvastatin on LVEF in PREVENT.

PATIENTS AND METHODS: Blood samples were collected and cardiac magnetic resonance imaging was performed prior to doxorubicin initiation and at 6 and 24 months. Thirteen biomarkers (arginine-nitric oxide metabolites, paraoxonase-1 [PON-1] activity, and myeloperoxidase) were measured. Dimensionality reduction using principal component analysis was used to define biomarker clusters. Linear mixed-effects models determined the changes in biomarkers over time according to treatment group. Mediation analysis determined if biomarker clusters explained the lack of effect of atorvastatin on LVEF.

RESULTS: Among 202 participants with available biomarkers, median age was 53 years; 86.6% had breast cancer; median LVEF was 62%. Cluster 1 levels, reflecting arginine methylation metabolites, were lower over time with atorvastatin, although this was not statistically significant (p=0.081); cluster 2 levels, reflecting PON-1 activity, were significantly lower with atorvastatin (p=0.024). There were no significant changes in other biomarker clusters (p>0.05). Biomarker clusters did not mediate an effect of atorvastatin on LVEF (p>0.05) Conclusions: Atorvastatin demonstrated very modest effects on oxidative/nitrosative stress biomarkers in this low cardiovascular risk population. Our findings provide potential mechanistic insight into the lack of effect of atorvastatin on LVEF in the PREVENT trial.

PMID:38573708 | DOI:10.1158/1078-0432.CCR-23-3952

Rehabil Psychol. 2024 Apr 4. doi: 10.1037/rep0000561. Online ahead of print.

ABSTRACT

PURPOSE: This study examines (a) whether disability registration has anticipatory, immediate, and delayed effects on depressive symptoms and (b) how these effects differ by gender.

RESEARCH METHOD/DESIGN: Using data from the Korea Welfare Panel Study spanning over 16 waves between 2005 and 2020, this study employed the individual-level fixed effects models to estimate the trajectories of depressive symptoms before and after the registration of physical disability, for a cohort of 20,054 individuals. Furthermore, gender-stratified fixed effects models were used to examine gender differences.

RESULTS: Compared to the preregistration reference period (i.e., 4 or more years before disability registration), there was a sustained rise in depressive symptoms leading up to the year of registration, indicating the presence of anticipatory effects. After disability registration, depressive symptoms consistently remained at a statistically higher level than during the initial reference period, with a gradual return to the baseline level of depressive symptoms over time. These anticipatory, immediate, and delayed effects of disability registration were notably more pronounced among men than women.

CONCLUSION/IMPLICATIONS: To develop more effective mental health interventions for people with disability, policymakers should consider gendered trajectories of depressive symptoms before and after disability registration. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PMID:38573669 | DOI:10.1037/rep0000561

Psychol Methods. 2024 Apr 4. doi: 10.1037/met0000657. Online ahead of print.

ABSTRACT

Latent moderated structural equation (LMS) is one of the most common techniques for estimating interaction effects involving latent variables (i.e., XWITH command in Mplus). However, empirical applications of LMS often overlook that this estimation technique assumes normally distributed variables and that violations of this assumption may lead to seriously biased parameter estimates. Against this backdrop, we study the robustness of LMS to different shapes and sources of nonnormality and examine whether various statistical tests can help researchers detect such distributional misspecifications. In four simulations, we show that LMS can be severely biased when the latent predictors or the structural disturbances are nonnormal. On the contrary, LMS is unaffected by nonnormality originating from measurement errors. As a result, testing for the multivariate normality of observed indicators of the latent predictors can lead to erroneous conclusions, flagging distributional misspecifications in perfectly unbiased LMS results and failing to reject seriously biased results. To solve this issue, we introduce a novel Hausman-type specification test to assess the distributional assumptions of LMS and demonstrate its performance. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PMID:38573667 | DOI:10.1037/met0000657

Psychol Methods. 2024 Apr 4. doi: 10.1037/met0000659. Online ahead of print.

ABSTRACT

Colliders, variables that serve as a common outcome of an independent and dependent variable, pose a major challenge in psychological research. Collider variables can induce bias in the estimation of a population relationship of interest when (a) the composition of a research sample is restricted by scores on a collider variable or (b) researchers adjust for a collider variable in their statistical analyses, as they might do for confounder variables. Both cases interfere with the accuracy and generalizability of statistical results. Despite their importance, however, collider effects remain relatively unknown in psychology. This tutorial article summarizes both the conceptual and the mathematical foundation for collider effects and their relevance to psychological research, and then proposes a method to correct for collider bias in cases of restrictive sample selection based on Thorndike’s Case III adjustment (1982). Two simulation studies demonstrated Thorndike’s correction as a viable solution for correcting collider bias in research studies, even when restriction on the collider variable was extreme and the selected sample size was as low as N = 100. Bias and relative bias results are reported to evaluate how well the correction equation approximates targeted population correlations under a variety of parameter conditions. We illustrate the application of the correction method to a hypothetical study of intelligence and conscientiousness, discuss the applicability of the method to more complex statistical models as a means of detection for collider bias, and provide code for researchers to apply to their own research. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PMID:38573665 | DOI:10.1037/met0000659

Psychol Methods. 2024 Apr 4. doi: 10.1037/met0000655. Online ahead of print.

ABSTRACT

Measurement invariance is an assumption underlying the regression of a latent variable on a background variable. It requires the measurement model parameters of the latent variable to be equal across the levels of the background variable. Item-specific violations of this assumption are referred to as differential item functioning and are ideally substantively explainable to warrant theoretically valid and meaningful results. Past research has focused on developing statistical approaches to explain differential item functioning effects in terms of item- or person-specific covariates. In this study, we propose a modeling approach that can be used to test if differences in item response times can be used to statistically explain differential item functioning. To this end, we operationalize a latent response process factor and test if item-specific group differences on this factor can account for the observed differences in item scores. We investigate the properties of the model in a simulation study, and we apply the model to a real data set. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PMID:38573663 | DOI:10.1037/met0000655

JAMA Ophthalmol. 2024 Apr 4. doi: 10.1001/jamaophthalmol.2024.0540. Online ahead of print.

ABSTRACT

IMPORTANCE: Greater understanding of the association between strabismus and mental health conditions across sociodemographic backgrounds may inform strategies to improve mental well-being in this population.

OBJECTIVE: To describe the association of strabismus with mental health conditions in a diverse cohort of US adults.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study used data from the National Institutes of Health’s All of Us Research Program, an ongoing program launched in 2015. The study included 3646 adults (aged ≥18 years) with strabismus and 3646 propensity score-matched controls. Statistical analysis was conducted from September 12, 2023, to January 29, 2024.

MAIN OUTCOMES AND MEASURES: Adults with strabismus were propensity score matched on age, gender, race and ethnicity, income, educational level, and insurance status in a 1:1 ratio with adults without strabismus. The prevalences of anxiety, depression, substance use and addiction, bipolar disorder, and schizophrenia spectrum disorder among adults with strabismus were compared with controls. Logistic regression was used to evaluate the association of mental health conditions with sociodemographic factors in each group.

RESULTS: This study included 3646 adults with strabismus (median age, 67 years [IQR, 53-76 years]; 2017 women [55%]) and 3646 propensity score-matched controls (median age, 67 years [IQR, 53-76 years]; 2017 women [55%]). Individuals with strabismus had higher prevalences of anxiety (1153 [32%] vs 519 [14%]; difference, 17%; 95% CI, 15%-19%; P < .001), depression (1189 [33%] vs 514 [14%]; difference, 19%; 95% CI, 17%-20%; P < .001), substance use and addiction (116 [3%] vs 51 [1%]; difference, 2%; 95% CI, 1%-3%; P < .001), bipolar disorder (253 [7%] vs 101 [3%]; difference, 4%; 95% CI, 3%-5%; P < .001), and schizophrenia spectrum disorder (103 [3%] vs 36 [1%]; difference, 2%; 95% CI, 1%-3%; P < .001) compared with individuals without strabismus. Among adults with strabismus, higher odds of mental health conditions were associated with younger age (odds ratio [OR], 1.11 per 10-year decrease; 95% CI, 1.06-1.16 per 10-year decrease), female gender (OR, 1.62; 95% CI, 1.41-1.85), Black or African American race and ethnicity (OR, 1.22; 95% CI, 1.01-1.48), low income (OR, 3.06; 95% CI, 2.56-3.67), and high school education or less (OR, 1.58; 95% CI, 1.34-1.85).

CONCLUSIONS AND RELEVANCE: In a diverse and nationwide cohort, adults with strabismus were more likely to have mental health conditions compared with adults without strabismus. Further investigation into the risk factors for poor mental health among adults with strabismus across sociodemographic backgrounds may offer novel opportunities for interventions to improve mental well-being in this population.

PMID:38573646 | DOI:10.1001/jamaophthalmol.2024.0540

JAMA Otolaryngol Head Neck Surg. 2024 Apr 4. doi: 10.1001/jamaoto.2024.0301. Online ahead of print.

ABSTRACT

IMPORTANCE: Oral tongue cancer (OTC) incidence has increased rapidly among young (<50 years) non-Hispanic White individuals in the US during the past 2 decades; however, it is unknown if age-associated trajectories have persisted.

OBJECTIVE: To examine US trends in OTC incidence and project future case burden.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional analysis of OTC incidence trends used the US Cancer Statistics Public Use Database, which covers approximately 98% of the US population, and included individuals with an OTC diagnosis reported to US cancer registries between January 1, 2001, and December 31, 2019.

EXPOSURES: Sex, race and ethnicity, and age.

MAIN OUTCOMES AND MEASURES: Estimated average annual percentage change in OTC incidence from 2001 to 2019. Given the substantial incidence rate increases among non-Hispanic White individuals compared with those of racial and ethnic minority groups, subsequent analyses were restricted to non-Hispanic White individuals. Forecasted OTC incidence trends and case burden among non-Hispanic White individuals to 2034.

RESULTS: There were 58 661 new cases of OTC identified between 2001 and 2019. Male individuals (57.6%), non-Hispanic White individuals (83.7%), those aged 60 years or older (58.0%), and individuals with localized stage disease at diagnosis (62.7%) comprised most cases. OTC incidence increased across all age, sex, and racial and ethnic groups, with marked increases observed among non-Hispanic White individuals (2.9% per year; 95% CI, 2.2%-3.7%). Increases among female individuals aged 50 to 59 years were most notable and significantly outpaced increases among younger non-Hispanic White female individuals (4.8% per year [95% CI, 4.1%-5.4%] vs 3.3% per year [95% CI, 2.7%-3.8%]). While all non-Hispanic White birth cohorts from 1925 to 1980 saw sustained increases, rates stabilized among female individuals born after 1980. Should trends continue, the burden of new OTC cases among non-Hispanic White individuals in the US is projected to shift more toward older individuals (from 33.1% to 49.3% among individuals aged 70 years or older) and female individuals (86% case increase vs 62% among male individuals).

CONCLUSIONS AND RELEVANCE: The results of this cross-sectional study suggest that the period of rapidly increasing OTC incidence among younger non-Hispanic White female individuals in the US is tempering and giving way to greater increases among older female individuals, suggesting a birth cohort effect may have been associated with previously observed trends. Recent increases among non-Hispanic White individuals 50 years or older of both sexes have matched or outpaced younger age groups. Continuing increases among older individuals, particularly female individuals, may be associated with a shift in the OTC patient profile over time.

PMID:38573630 | DOI:10.1001/jamaoto.2024.0301

Int Urol Nephrol. 2024 Apr 4. doi: 10.1007/s11255-024-04020-w. Online ahead of print.

ABSTRACT

BACKGROUND: Recent studies demonstrated that chronic prostatitis (CP) is closely related to the gut microbiota (GM). Nevertheless, the causal relationship between GM and CP has not been fully elucidated. Therefore, the two-sample Mendelian randomization (MR) analysis was employed to investigate this association.

METHODS: The summary data of gut microbiota derived from a genome-wide association study (GWAS) involving 18,340 individuals in the MiBioGen study served as the exposure, and the corresponding summary statistics for CP risk, representing the outcome, were obtained from the FinnGen databases (R9). The causal effects between GM and CP were estimated using the inverse-variance weighted (IVW) method supplemented with MR-Egger, weighted median, weighted mode, and simple mode methods. Additionally, the false discovery rate (FDR) correction was performed to adjust results. The detection and quantification of heterogeneity and pleiotropy were accomplished through the MR pleiotropy residual sum and outlier method, Cochran’s Q statistics, and MR-Egger regression.

RESULTS: The IVW estimates indicated that a total of 11 GM taxa were related to the risk of CP. Seven of them was correlated with an increased risk of CP, while the remained linked with a decreased risk of CP. However, only Methanobacteria (OR 0.86; 95% CI 0.74-0.99), Methanobacteriales (OR 0.86; 95% CI 0.74-0.99), NB1n (OR 1.16; 95% CI 1.16-1.34), Methanobacteriaceae (OR 0.86; 95% CI 0.74-0.99), Odoribactergenus Odoribacter (OR 1.43; 95% CI 1.05-1.94), and Sutterellagenus Sutterella (OR 1.33; 95% CI 1.01-1.76) still maintain significant association with CP after FDR correction. Consistent directional effects for all analyses were observed in the supplementary methods. Subsequently, sensitivity analyses indicated the absence of heterogeneity, directional pleiotropy, or outliers concerning the causal effect of specific gut microbiota on CP (p > 0.05).

CONCLUSION: Our study demonstrated a gut microbiota-prostate axis, offering crucial data supporting the promising use of the GM as a candidate target for CP prevention, diagnosis, and treatment. There is a necessity for randomized controlled trials to validate the protective effect of the linked GM against the risk of CP, and to further investigate the underlying mechanisms involved.

PMID:38573543 | DOI:10.1007/s11255-024-04020-w

Osteoporos Int. 2024 Apr 4. doi: 10.1007/s00198-024-07049-w. Online ahead of print.

ABSTRACT

Post hoc analysis of FRAME and ARCH revealed that on-study nonvertebral and vertebral fractures by Month 12 were less common in women initially treated with romosozumab versus placebo or alendronate. Recurrent fracture risk was also lower in romosozumab‑treated patients, and there were no fracture‑related complications. Results support continuing romosozumab treatment post‑fracture.

PURPOSE: Post hoc analysis evaluating efficacy and safety of romosozumab, administered in the immediate post‑fracture period, in the FRAME and ARCH phase 3 trials.

METHODS: In FRAME (NCT01575834) and ARCH (NCT01631214), postmenopausal women with osteoporosis were randomized 1:1 to romosozumab 210 mg monthly or comparator (FRAME, placebo; ARCH, alendronate 70 mg weekly) for 12 months, followed by antiresorptive therapy (FRAME, denosumab; ARCH, alendronate). In patients who experienced on-study nonvertebral or new/worsening vertebral fracture by Month 12, we report the following: fracture and treatment‑emergent adverse event (TEAE) incidence through 36 months, bone mineral density changes (BMD), and romosozumab timing. Due to the sample sizes employed, meaningful statistical comparisons between treatments were not possible.

RESULTS: Incidence of on-study nonvertebral and vertebral fractures by Month 12 was numerically lower in romosozumab- versus comparator-treated patients (FRAME, 1.6% and 0.5% versus 2.1% and 1.6%; ARCH, 3.4% and 3.3% versus 4.6% and 4.9%, respectively). In those who experienced on-study nonvertebral fracture by Month 12, recurrent nonvertebral and subsequent vertebral fracture incidences were numerically lower in patients initially treated with romosozumab versus comparator (FRAME, 3.6% [2/56] and 1.8% [1/56] versus 9.2% [7/76] and 3.9% [3/76]; ARCH, 10.0% [7/70] and 5.7% [4/70] versus 12.6% [12/95] and 8.4% [8/95], respectively). Among those with on-study vertebral fracture by Month 12, recurrent vertebral and subsequent nonvertebral fracture incidences were numerically lower with romosozumab versus comparator (FRAME, 0.0% [0/17] and 0.0% [0/17] versus 11.9% [7/59] and 8.5% [5/59]; ARCH, 9.0% [6/67] and 7.5% [5/67] versus 15.0% [15/100] and 16.0% [16/100], respectively). In patients with fracture by Month 12, no fracture‑related complications were reported in romosozumab-treated patients. BMD gains were numerically greater with romosozumab than comparators.

CONCLUSION: Data suggest support for the efficacy and safety of continuing romosozumab treatment following fracture.

TRIAL REGISTRATIONS: NCT01575834; NCT01631214.

PMID:38573517 | DOI:10.1007/s00198-024-07049-w