Category: Statistics

JAMA Netw Open. 2023 Jun 1;6(6):e2318132. doi: 10.1001/jamanetworkopen.2023.18132.

ABSTRACT

IMPORTANCE: The clinical value of current multifactorial algorithms for individualized assessment of dementia risk remains unclear.

OBJECTIVE: To evaluate the clinical value associated with 4 widely used dementia risk scores in estimating 10-year dementia risk.

DESIGN, SETTING, AND PARTICIPANTS: This prospective population-based UK Biobank cohort study assessed 4 dementia risk scores at baseline (2006-2010) and ascertained incident dementia during the following 10 years. Replication with a 20-year follow-up was based on the British Whitehall II study. For both analyses, participants who had no dementia at baseline, had complete data on at least 1 dementia risk score, and were linked to electronic health records from hospitalizations or mortality were included. Data analysis was conducted from July 5, 2022, to April 20, 2023.

EXPOSURES: Four existing dementia risk scores: the Cardiovascular Risk Factors, Aging and Dementia (CAIDE)-Clinical score, the CAIDE-APOE-supplemented score, the Brief Dementia Screening Indicator (BDSI), and the Australian National University Alzheimer Disease Risk Index (ANU-ADRI).

MAIN OUTCOMES AND MEASURES: Dementia was ascertained from linked electronic health records. To evaluate how well each score predicted the 10-year risk of dementia, concordance (C) statistics, detection rate, false-positive rate, and the ratio of true to false positives were calculated for each risk score and for a model including age alone.

RESULTS: Of 465 929 UK Biobank participants without dementia at baseline (mean [SD] age, 56.5 [8.1] years; range, 38-73 years; 252 778 [54.3%] female participants), 3421 were diagnosed with dementia at follow-up (7.5 per 10 000 person-years). If the threshold for a positive test result was calibrated to achieve a 5% false-positive rate, all 4 risk scores detected 9% to 16% of incident dementia and therefore missed 84% to 91% (failure rate). The corresponding failure rate was 84% for a model that included age only. For a positive test result calibrated to detect at least half of future incident dementia, the ratio of true to false positives ranged between 1 to 66 (for CAIDE-APOE-supplemented) and 1 to 116 (for ANU-ADRI). For age alone, the ratio was 1 to 43. The C statistic was 0.66 (95% CI, 0.65-0.67) for the CAIDE clinical version, 0.73 (95% CI, 0.72-0.73) for the CAIDE-APOE-supplemented, 0.68 (95% CI, 0.67-0.69) for BDSI, 0.59 (95% CI, 0.58-0.60) for ANU-ADRI, and 0.79 (95% CI, 0.79-0.80) for age alone. Similar C statistics were seen for 20-year dementia risk in the Whitehall II study cohort, which included 4865 participants (mean [SD] age, 54.9 [5.9] years; 1342 [27.6%] female participants). In a subgroup analysis of same-aged participants aged 65 (±1) years, discriminatory capacity of risk scores was low (C statistics between 0.52 and 0.60).

CONCLUSIONS AND RELEVANCE: In these cohort studies, individualized assessments of dementia risk using existing risk prediction scores had high error rates. These findings suggest that the scores were of limited value in targeting people for dementia prevention. Further research is needed to develop more accurate algorithms for estimation of dementia risk.

PMID:37310738 | DOI:10.1001/jamanetworkopen.2023.18132

J Women Aging. 2023 Jun 13:1-19. doi: 10.1080/08952841.2023.2222638. Online ahead of print.

ABSTRACT

AIMS: The overarching goal of the present study was to examine the contribution of various sources of empowerment to mental health during retirement with an added emphasis on gender differences. The empowerment sources that were examined corresponded with three distinct ecological systems: (1) Chronosystem-resource gains from the pre- to post-retirement period and satisfaction with the prior working period; (2) Microsystem-marital power dynamics (measured by the division of household labor and decision-making in the marital relationship) and the presence of a confidant; (3) Ontogenetic system-a sense of meaning in one’s life during the retirement period and an assessment of absolute resources.

METHOD: The research sample consisted of 160 Israeli retirees (78 women and 82 men) who had retired within the previous eight years. Data were collected by the Panels Research Institute in Israel using the institute’s database of members. Participants completed an online questionnaire accessible via a website link. Statistical processing was performed using ANOVA and hierarchical regression analysis.

RESULTS: The results indicated that retirees’ reports of resource gains after retirement, their sense of meaning in life, their satisfaction with their working period prior to retirement, and their perceived level of absolute resources were all associated with mental health. Additionally, the more participants (both men and women) rated that the husband was involved in household labor, the better retirees reported their mental health to be. Gender differences were found in regard to some empowerment sources during retirement: retired women reported lower levels of mental health and prior work satisfaction compared to retired men, and men’s assessments of their participation in household labor and decision-making were higher than women’s assessments of their husbands’ involvement. The proportion of men who reported that their wives were their confidants was higher than the proportion of women who reported that their husbands were their confidants.

SUMMARY AND CONCLUSIONS: Overall, men experienced more sources of empowerment than women during retirement, but findings suggest that men’s emotional dependence on their wives is greater than women’s emotional dependence on their husbands. Based on the study’s findings, recommendations are offered to professionals who work with retirees.

PMID:37310726 | DOI:10.1080/08952841.2023.2222638

Fam Med. 2023 May;55(5):339-344. doi: 10.22454/FamMed.2023.788985. Epub 2023 Feb 21.

ABSTRACT

BACKGROUND AND OBJECTIVES: The family medicine residency application process is arduous and competitive. An important part of the application is the in-person interview process, which was affected during the last two interview cycles (2021-2022) due to restrictions imposed by the COVID-19 pandemic. Virtual interviews eliminate traveling costs associated with applications, potentially improving underrepresented minorities’ access to interviewing opportunities. Our goal was to assess if virtual interviews at our institution have negative or positive effects on access for underrepresented in medicine (URiM) applicants and our residency match results. Methods: We analyzed data from 2019-2022 to compare application volumes, applicant demographics, and match results between two in-person cycles (2019 and 2020) and two virtual cycles (2021-2022). Data were analyzed using Pearson χ2 criteria and P=.05 defined significance. Differences between years for expected counts were determined using single sample χ2 tests. Results: No statistical significance was noted on number of applications by URiM to our program despite decrease costs associated with virtual interview process. The number of URiM applicants matching to our program did not improve by simply implementing virtual interviews when compared to in-person interview seasons in the past.

CONCLUSIONS: Virtual interviews at our institution did not increase URiM applications to our program from substantial equivalent medical schools. Further research in this area from programs in other states may enhance our understanding of the impact of virtual interviews on URiM applications to residency and match results.

PMID:37310680 | DOI:10.22454/FamMed.2023.788985

Ther Innov Regul Sci. 2023 Jun 13. doi: 10.1007/s43441-023-00541-1. Online ahead of print.

ABSTRACT

Artificial Intelligence (AI) and Machine Learning (ML) are making headlines in medical research, especially in drug discovery, digital imaging, disease diagnostics, genetic testing, and optimal care pathway (personalized care). However, the potential uses and benefits of AI/ML applications need to be distinguished from hype. In the 2022 American Statistical Association Biopharmaceutical Section Regulatory-Industry Statistical Workshop, we convened a panel of experts from FDA and industry to talk about the challenges of successfully applying AI/ML in precision medicine and how to overcome those challenges. This paper provides a summary and expansion on the topics discussed in the panel: the application of AI/ML, bias, and data quality.

PMID:37310669 | DOI:10.1007/s43441-023-00541-1

Clin Exp Med. 2023 Jun 13. doi: 10.1007/s10238-023-01114-0. Online ahead of print.

ABSTRACT

To investigate the value of the peripheral blood lymphocyte count (LYM) combined with interleukin-6 (IL-6) in predicting disease severity and prognosis in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pneumonia. This was a prospective observational cohort study. A total of 109 patients with SARS-CoV-2 pneumonia who were admitted to Nanjing First Hospital from December 2022 to January 2023 were enrolled. The patients were divided into two groups based on disease severity: severe (46 patients) and critically ill (63 patients). The clinical data of all patients were collected. The clinical characteristics, sequential organ failure assessment (SOFA) score, peripheral blood lymphocyte count, IL-6 level and other laboratory test results were compared between the two groups. A receiver operating characteristic (ROC) curve was plotted to evaluate the predictive value of each index for SARS-CoV-2 pneumonia severity; patients were regrouped using the optimal cut-off value of the ROC curve, and the relationship between different LYM and IL-6 levels and the prognosis of patients was analysed. Kaplan‒Meier survival curve analysis was performed; in the different LYM and IL-6 groups, the patients were regrouped based on whether thymosin was used, and the effect of thymosin on patient prognosis was compared between the groups. Patients in the critically ill group were significantly older than patients in the severe group (age: 78 ± 8 vs. 71 ± 17, t = 2.982, P < 0.05), and the proportion of patients with hypertension, diabetes and cerebrovascular disease was significantly higher in the critically ill group than in the severe group (69.8% vs. 45.7%, 38.1% vs. 17.4%, 36.5% vs. 13.0%; χ² values, 6.462, 5.495, 7.496, respectively, all P < 0.05). Compared with the severe group, the critically ill group had a higher SOFA score on admission (score: 5.4 ± 3.0 vs. 1.9 ± 1.5, t = 24.269, P < 0.05); IL-6 and procalcitonin (PCT) in the critically ill group were significantly higher than those in the severe group on the first day of admission [288.4 (191.4, 412.9) vs. 513.0 (288.2, 857.4), 0.4 (0.1, 3.2) vs. 0.1 (0.05, 0.2); Z values, 4.000, 4.456, both P < 0.05]. The lymphocyte count continued to decline, and the lymphocyte count on the 5th day (LYM-5d) was still low (0.6 ± 0.4 vs. 1.0 ± 0.4, t = 4.515, both P < 0.05), with statistically significant differences between the two groups. ROC curve analysis indicated that LYM-5d, IL-6 and LYM-5d + IL-6 all had value for predicting SARS-CoV-2 pneumonia severity; the areas under the curve (AUCs) were 0.766, 0.725, and 0.817, respectively, and the 95% confidence intervals (95% CI) were 0.676-0.856, 0.631-0.819, and 0.737-0.897, respectively. The optimal cut-off values for LYM-5d and IL-6 were 0.7 × 10⁹/L and 416.4 pg/ml, respectively. LYM-5d + IL-6 had the greatest value in predicting disease severity, and LYM-5d had higher sensitivity and specificity in predicting SARS-CoV-2 pneumonia severity. Regrouping was performed based on the optimal cut-off values for LYM-5d and IL-6. Comparing the IL-6 ≥ 416.4 pg/ml and LYM-5d < 0.7 × 10⁹/L group with the other group, i.e., patients in the non-low-LYM-5d and high-IL-6 group, patients in the low-LYM-5d and high-IL-6 group had a higher 28-day mortality rate (71.9% vs. 29.9%, χ² value 16.352, P < 0.05) and a longer hospital stay, intensive care unit (ICU) stay and mechanical ventilation time (days: 13.7 ± 6.3 vs. 8.4 ± 4.3, 9.0 (7.0, 11.5) vs. 7.5 (4.0, 9.5), 8.0 (6.0, 10.0) vs. 6.0 (3.3, 8.5); t/Z values, 11.657, 2.113, 2.553, respectively, all P < 0.05), as well as a higher incidence of secondary bacterial infection during the disease course (75.0% vs. 41.6%, χ² value 10.120, P < 0.05). Kaplan‒Meier survival analysis indicated that the median survival time of patients in the low LYM-5d and high-IL-6 group was significantly shorter than that of patients in the non-low LYM-5d and high-IL-6 group (14.5 ± 1.8 d vs. 22.2 ± 1.1 d, Z value 18.086, P < 0.05). There was no significant difference in the curative effect between the thymosin group and the nonthymosin group. LYM and IL-6 levels are closely related to SARS-CoV-2 pneumonia severity. The prognosis for patients with IL-6 ≥ 416.4 pg/ml at admission and a lymphocyte count < 0.7 × 10 9/L on the 5th day is poor.

PMID:37310657 | DOI:10.1007/s10238-023-01114-0

Can J Anaesth. 2023 Jun 13. doi: 10.1007/s12630-023-02457-8. Online ahead of print.

ABSTRACT

PURPOSE: Venous thromboembolism (VTE) is a common complication of critical illness. Sex- or gender-based analyses are rarely conducted and their effect on outcomes is unknown. We assessed for an effect modification of thromboprophylaxis (dalteparin or unfractionated heparin [UFH]) by sex on thrombotic (deep venous thrombosis [DVT], pulmonary embolism [PE], VTE) and mortality outcomes in a secondary analysis of the Prophylaxis for Thromboembolism in Critical Care Trial (PROTECT).

METHODS: We conducted unadjusted analyses using Cox proportional hazards analysis, stratified by centre and admission diagnostic category, including sex, treatment, and an interaction term. Additionally, we performed adjusted analyses and assessed the credibility of our findings.

RESULTS: Critically ill female (n = 1,614) and male (n = 2,113) participants experienced similar rates of DVT, proximal DVT, PE, any VTE, ICU death, and hospital death. In unadjusted analyses, we did not find significant differences in treatment effect favouring males (vs females) treated with dalteparin (vs UFH) for proximal leg DVT, any DVT, or any PE, but found a statistically significant effect (moderate certainty) favouring dalteparin in males for any VTE (males: hazard ratio [HR], 0.71; 95% confidence interval [CI], 0.52 to 0.96 vs females: HR, 1.16; 95% CI, 0.81 to 1.68; P = 0.04). This effect remained after adjustment for baseline characteristics (males: HR, 0.70; 95% CI, 0.52 to 0.96 vs females: HR, 1.17; 95% CI, 0.81 to 1.68; P = 0.04) and weight (males: HR, 0.70; 95% CI, 0.52 to 0.96 vs females: HR, 1.20; 95% CI, 0.83 to 1.73; P = 0.03). We did not identify a significant effect modification by sex on mortality.

CONCLUSIONS: We found an effect modification by sex of thromboprophylaxis on VTE in critically ill patients that requires confirmation. Our findings highlight the need for sex- and gender-based analyses in acute care research.

PMID:37310606 | DOI:10.1007/s12630-023-02457-8

Environ Sci Pollut Res Int. 2023 Jun 13. doi: 10.1007/s11356-023-28187-4. Online ahead of print.

ABSTRACT

Although transportation systems are an increasing necessity in today’s connected world, the over-reliance on internal combustion engine vehicles has caused an increase in air and noise pollution. Air and noise pollutions are among the negative environmental factors that contribute to disease occurrence due to their negative health consequences. Literature has showed that air and noise pollutions were responsible for thousands of premature deaths in Europe. This has pushed scientists to search for models to calculate the effect of traffic on air and noise pollution, to help predict future scenarios, and to work on ways to mitigate that increase in pollution. In this paper, a statistical model is done by using data measured for 25 speed bump locations in Kuwait, which included traffic flow data in the form of vehicle count and classification, as well as noise level measurements taken through an Amprobe SM20 sound meter, in addition to air pollutant data obtained from the Environment Public Authority (EPA) in Kuwait. The results of the multivariate linear regression model showed that high traffic counts resulted in significantly higher noise levels, reaching more than 70 decibels in certain locations, which is considered unhealthy for extended time periods. The model also showed that sulfur dioxide levels were impacted by both light and heavy vehicles, while particulate matter less than 10 µm was affected mainly by heavy vehicles. An online survey on speed bumps was completed by 803 participants to understand the behavior of people at speed bumps in Kuwait and understand if age and gender might be a predictor of how people would behave, and this was done by doing Pearson’s chi-squared correlation tests on the results obtained from the survey.

PMID:37310605 | DOI:10.1007/s11356-023-28187-4

Pain Ther. 2023 Jun 13. doi: 10.1007/s40122-023-00532-y. Online ahead of print.

ABSTRACT

INTRODUCTION: Occipital neuralgia (ON) is a disabling cephalalgia form with demanding diagnostic workflow. We report the description and reliability analyses of the occipital nerves-applied strain (ONAS) test for occipital neuralgia (ON) early-stage diagnosis in cephalalgia patients.

METHODS: In a retrospective and observational study, we evaluated, among n = 163 consecutive cephalalgia patients, the sensitivity, specificity, and prior probability [positive (PPV) and negative (NPV) predictive values] of the ONAS test against two reference tests (occipital nerve anesthetic block and the painDETECT questionnaire). Multinomial logistic regression (MLR) and χ² analyses verified the ONAS test outcome’s dependence upon independent variables (gender, age, pain site, block test, and painDETECT outcomes). We assessed inter-rater agreement with Cohen’s kappa statistic.

RESULTS: ONAS test showed sensitivity and specificity of 81 and 18%, respectively, against the painDETECT and of 94 and 46%, respectively, against the block test. PPV was > 70% against both tests, while NPV was 81% against the block test and 26% against the painDETECT. Interrater agreement Cohen’s kappa was excellent. Significant association (χ² analyses) and relationship (MLR) were found only between ONAS test and pain site but not with the other independent predictors.

CONCLUSIONS: The ONAS test showed satisfactory reliability among cephalalgia patients; thus, it might be considered a valuable early stage tool for ON diagnosis in these patients.

PMID:37310572 | DOI:10.1007/s40122-023-00532-y

Radiol Med. 2023 Jun 13. doi: 10.1007/s11547-023-01663-0. Online ahead of print.

ABSTRACT

PURPOSE: To compare conventional CT images and virtual monoenergetic images (VMI) at dual-layer dual-energy CT (dlDECT) in patients with colorectal cancer (CRC) through quantitative analysis and to investigate the added value of VMI.

MATERIAL AND METHODS: Sixty-six consecutive patients with histologically documented CRC and available VMI reconstructions were retrospectively investigated. Subsequently, forty-two patients, without any colonic disease at colonoscopy, were selected as control group. Conventional CT images and VMI reconstructions at energy levels ranging from 40 (VMI₄₀) to 100 keV (VMI₁₀₀) in 10 keV increments, were obtained from the late arterial phase. First, signal-to-noise (SNR) and contrast-to-noise (CNR) ratios were obtained to select the best VMI reconstruction. Finally, the diagnostic accuracy of conventional CT and VMI₄₀ in late arterial phase was evaluated.

RESULTS: On quantitative analysis, SNR and CNR were higher for VMI₄₀ (19.5 ± 7.7 and 11.8 ± 6.2, respectively) with statistically significant differences compared to conventional CT (P < 0.05) and all the other VMI reconstructions (P < 0.05), except for VMI₅₀ (P > 0.05). The addition of VMI₄₀ to conventional CT images significantly improved the area under the curve (AUC) for the diagnosis of CRC, increasing it from 0.875 to 0.943 for reader 1 (P < 0.05) and from 0.916 to 0.954 for reader 2 (P < 0.05). The improvement was greater in the less experienced radiologist (0.068) compared to the more experienced one (0.037).

CONCLUSION: VMI₄₀ has showed the highest quantitative image parameters. Furthermore, the use of VMI₄₀ can lead to a significant improvement in the diagnostic performance for detecting CRC.

PMID:37310558 | DOI:10.1007/s11547-023-01663-0

Bull Math Biol. 2023 Jun 13;85(7):68. doi: 10.1007/s11538-023-01167-y.

ABSTRACT

Homogeneity across lineages is a general assumption in phylogenetics according to which nucleotide substitution rates are common to all lineages. Many phylogenetic methods relax this hypothesis but keep a simple enough model to make the process of sequence evolution more tractable. On the other hand, dealing successfully with the general case (heterogeneity of rates across lineages) is one of the key features of phylogenetic reconstruction methods based on algebraic tools. The goal of this paper is twofold. First, we present a new weighting system for quartets (ASAQ) based on algebraic and semi-algebraic tools, thus especially indicated to deal with data evolving under heterogeneous rates. This method combines the weights of two previous methods by means of a test based on the positivity of the branch lengths estimated with the paralinear distance. ASAQ is statistically consistent when applied to data generated under the general Markov model, considers rate and base composition heterogeneity among lineages and does not assume stationarity nor time-reversibility. Second, we test and compare the performance of several quartet-based methods for phylogenetic tree reconstruction (namely QFM, wQFM, quartet puzzling, weight optimization and Willson’s method) in combination with several systems of weights, including ASAQ weights and other weights based on algebraic and semi-algebraic methods or on the paralinear distance. These tests are applied to both simulated and real data and support weight optimization with ASAQ weights as a reliable and successful reconstruction method that improves upon the accuracy of global methods (such as neighbor-joining or maximum likelihood) in the presence of long branches or on mixtures of distributions on trees.

PMID:37310552 | DOI:10.1007/s11538-023-01167-y