Modelling survival in acute severe illness: Cox versus accelerated failure time models

Adipokines and adiposity among postmenopausal women of the Multi-Ethnic Study of Atherosclerosis

OBJECTIVE

We investigated whether the associations of serum adiponectin, leptin, and resistin with adiposity differ with menopausal age.

METHODS

In this cross-sectional study, we included 751 postmenopausal women from the Multi-Ethnic Study of Atherosclerosis (MESA) who reported their menopausal age (<45, 45-49, 50-54 and ≥55 y) and had anthropometrics, serum adipokines, and abdominal computed tomography measures of visceral and subcutaneous adipose tissue (VAT and SAT) obtained at MESA exam 2 or 3. Linear regression models were used for analysis.

RESULTS

The mean ± SD age was 65.1 ± 9.0 years for all participants. The median (interquartile range) values for serum adiponectin, leptin and resistin, VAT, and SAT were 21.9 (14.8-31.7) ng/L, 24.3 (12.5-42.4) pg/L, 15.3 (11.8-19.5) pg/L, 183.9 (130.8-251.1) cm2, and 103.7 (65.6-151.5) cm2, respectively. The mean ± SD values for body mass index, waist circumference, and waist-to-hip ratio were 28.3 ± 5.81 kg/m2, 96.6 ± 15.9 cm, and 0.91 ± 0.078, respectively. Adiponectin was inversely associated with all adiposity measures, with similar patterns across menopausal age categories. Leptin was positively associated with all adiposity measures, and the strength of associations varied across menopausal age categories for body mass index, waist circumference, and SAT (Pinteraction ≤ 0.01 for all). The associations of resistin with adiposity measures were mostly nonsignificant except in the 45- to 49-year menopausal age category.

CONCLUSIONS

Menopausal age category had no influence on the association of serum adiponectin with adiposity. The association of serum leptin and resistin differed according to menopausal age category for generalized adiposity but was inconsistent for measures of abdominal adiposity.

A prospective study on the relationship between polycystic ovary syndrome and age at natural menopause

OBJECTIVE

This study aimed to determine whether polycystic ovary syndrome (PCOS) was associated with age at menopause, compared with women without PCOS, after adjusting for potential confounders.

METHODS

A total of 1,696 reproductive-aged participants from the Tehran Lipid and Glucose Study were included in this population-based prospective study with a follow-up of approximately 20 years. Of these, 348 women with PCOS based on the Rotterdam criteria and 1,348 non-PCOS controls were followed to assess the age at which they reached menopause. An accelerated failure time survival regression model was used to identify the association between PCOS and the age at natural menopause (ANM), with and without adjustment for potential confounders.

RESULTS

The unadjusted accelerated failure time survival model revealed a significant positive association between PCOS and ANM; PCOS women experienced time to menopause by a factor of 1.05 than non-PCOS controls (95% confidence interval, 1.02-1.06; P < 0.001). After adjusting for age at baseline, menarche age, history of hypertension, history of type 2 diabetes mellitus, parity, oral contraceptive use, body mass index, education level, physical activity, and smoking, the results remained significant (time ratio: 1.03; 95% confidence interval, 1.01-1.06; P = 0.002).

CONCLUSIONS

This study indicates that ANM is significantly associated with PCOS in women. Our study findings may have implications for the fertility and reproductive health of women with PCOS. However, further large longitudinal studies on diverse populations accounting for other relevant confounders are still needed to provide data on the actual difference in age at menopause and to elucidate the underlying mechanisms of this association.

Anthropometric Indices and Age at Natural Menopause: A 15-Year Follow-up Population-Based Study

BACKGROUND

Following menopause, the risk of many diseases is increased, and this situation may be due to changes in anthropometric indices (AI), while the association between adiposity and age at natural menopause (ANM) is not clear yet.

OBJECTIVES

This longitudinal study was conducted to investigate the ability of AI in predicting ANM.

METHODS

For this purpose, a total of 3,876 women aged > 20 years old from participants of the Tehran lipid and glucose study (TLGS) met our eligibility criteria. The association between ANM and various AIs was assessed using the Accelerated Failure Time (AFT) model, and time ratio (TR) with 95% confidence intervals was reported in this longitudinal study.

RESULTS

The median [interquartile range (IQR) 25 - 75] of the survival time was 12.5 (7.9 - 15.8) years. At the end of the follow-up, 1,479 (38.2%) of the participants reached menopause. The median time to natural menopause was decreased by about 2% with one standard deviation (SD) increase of both a body shape index (ABSI) (time ratio (TR): 0.98, 95% CI: 0.97, 0.99) and lipid accumulation product (LAP) (TR: 0.98, 95% CI: 0.98, 0.99) z-scores; and this time was increased by about 1% (TR: 1.01, 95% CI: 1.00, 1.02) with one SD increase in body mass index (BMI) z-score.

CONCLUSIONS

The ABSI, LAP, and BMI were the most useful AIs for identification of the time to menopause onset, and ABSI and LAP were inversely associated with the ANM. Also, the BMI was directly associated with the ANM.

Predictive factors of survival time of breast cancer in kurdistan province of Iran between 2006-2014: a cox regression approach

BACKGROUND

Breast cancer is the most common cancer and the second most common cause of cancer-induced mortalities in Iranian women, following gastric carcinoma. The survival of these patients depends on several factors, which are very important to identify in order to understand the natural history of the disease.

MATERIALS AND METHODS

In this retrospective study, 313 consecutive women with pathologically-proven diagnosis of breast cancer who had been treated during a seven-year period (January 2006 until March 2014) at Towhid hospital, Sanandaj city, Kurdistan province of Iran, were recruited. The Kaplan-Meier method was used for data analysis, and finally those factors that showed significant association on univariate analysis were entered in a Cox regression model.

RESULTS

the mean age of patients was 46.10±10.81 years. Based on Kaplan-Meier method median of survival time was 81 months and 5 year survival rate was 75%±0.43. Tumor metastasis (HR=9.06, p=0.0001), relapse (HR=3.20, p=0.001), clinical stage of cancer (HR=2.30, p=0.03) and place of metastasis (p=0.0001) had significant associations with the survival rate variation. Patients with tumor metastasis had the lowest five-year survival rate (37%)and among them patients who had brain metastasis were in the worst condition (5 year survival rate= 11%±0.10).

CONCLUSIONS

Our findings support the observation that those women with higher stages of breast malignancies (especially with metastatic cancer) have less chance of surviving the disease. Furthermore, screening programs and early detection of breast cancer may help to increase the survival of those women who are at risk of breast cancer.

Data generation for the Cox proportional hazards model with time-dependent covariates: a method for medical researchers

The proliferation of longitudinal studies has increased the importance of statistical methods for time-to-event data that can incorporate time-dependent covariates. The Cox proportional hazards model is one such method that is widely used. As more extensions of the Cox model with time-dependent covariates are developed, simulations studies will grow in importance as well. An essential starting point for simulation studies of time-to-event models is the ability to produce simulated survival times from a known data generating process. This paper develops a method for the generation of survival times that follow a Cox proportional hazards model with time-dependent covariates. The method presented relies on a simple transformation of random variables generated according to a truncated piecewise exponential distribution and allows practitioners great flexibility and control over both the number of time-dependent covariates and the number of time periods in the duration of follow-up measurement. Within this framework, an additional argument is suggested that allows researchers to generate time-to-event data in which covariates change at integer-valued steps of the time scale. The purpose of this approach is to produce data for simulation experiments that mimic the types of data structures applied that researchers encounter when using longitudinal biomedical data. Validity is assessed in a set of simulation experiments, and results indicate that the proposed procedure performs well in producing data that conform to the assumptions of the Cox proportional hazards model.

Prognostic factors of survival time after hematopoietic stem cell transplant in acute lymphoblastic leukemia patients: Cox proportional hazard versus accelerated failure time models

Background

The aim of this study is to evaluate the prognostic factors of overall survival (OS) after haematopoietic stem cell transplant (HSCT) in acute lymphoblastic leukaemia (ALL) patients using accelerated failure time (AFT), Cox proportional hazard (PH), and Cox time-varying coefficient models.

Methods

206 patients were enrolled after HSCH in Shariati Hospital between 1993 and 2007. There was evidence of marked departures from the proportional hazards assumption with two prognostic factors, relapse and chronic graft-versus-host disease (cGVHD) (P < .001). Performance among AFT and Cox's models was assessed using explained variation and goodness of fit methods. Discrimination among the exponential, Weibull, generalized gamma (GG), log-logistic, and lognormal distributions was done using maximum likelihood and Akaike information criteria.

Results

The 5-year OS was 52% (95%CI: 47.3–56.7). Peak mortality hazard occurred at months 6–7 after HSCT followed by a decreasing trend. In univariate analysis, the data was better fitted by GG distribution than by other distributions. Univariate analysis using GG distribution showed a positive association between OS with acute graft-versus-host disease (aGVHD) (P = .021), no relapse (P < .001), cGVHD (P < .001), neutrophil recovery (P < .001) and platelet recovery (P < .001). Based on Cox PH models; however cGVHD and relapse were the predictive factors of OS (P < .001). Multivariate analysis indicated that, OS is related to relapse (P < .001) and platelet recovery (P = .037), where predictive power of Weibull AFT models was superior to Cox PH model and Cox with time-varying coefficient (R² = 0.46 for AFT, R² = .21 for Cox PH and R² = .34 for Cox time-varying coefficient). Cox-Snell residual shows Weibull AFT fitted to data better than other distributions in multivariate analysis.

Conclusion

We concluded that AFT distributions can be a useful tool for recognizing prognostic factors of OS in acute lymphoblastic leukemia patients.