Abstract
Introduction: Much current literature reports a higher prevalence of chronic kidney disease (CKD) amongst women than men. Amongst patients that initiate renal replacement therapy (RRT) there is a preponderance of men. The reasons for this gender discrepancy are poorly reported.
Aim: To investigate gender differences in CKD prevalence and outcomes in a large Grampian cohort, focussing specifically on RRT initiation.
Methods: All creatinine values measured in the Grampian region from mid-1999 to 2009 were available. All individuals who had an eGFR of less than 60ml/min/1.73m2 in 2003 that could be demonstrated to be present for at least three months were defined to have CKD in 2003. These individuals’ records (as part of the Grampian Laboratory Morbidity and Mortality Study II (GLOMMS-II)) were linked to hospital episode data (SMR01), local and national renal registry data and the National Records for comorbidity, RRT and mortality information.
The prevalence of stage 3-5 CKD in Grampian in 2003 was calculated. Univariate analysis identified potential confounding factors between men and women with stage 3-5 CKD and those who did and did not initiate RRT during six years of follow-up. The rate of RRT initiation by gender was calculated. Gender rate ratios were calculated using poisson regression, adjusting for age, CKD stage, proteinuria and significant co-morbidities. Gender rate ratios were also stratified by age.
Results: There were 19635 individuals who had stage 3-5 CKD in Grampian in 2003 (64.9% women). Grampian women in 2003 had a higher prevalence of CKD than men, at 5.7% (95%CI 5.6, 5.8) vs 3.2% (95%CI 3.1, 3.3). Men with CKD had a 3.4-fold (95%CI 2.6, 4.4) increase in RRT initiation compared to women. This relationship between gender and RRT initiation persisted, after adjustment for potential confounders. However, gender rate ratios stratified by age showed age to be an important effect modifier – the highest gender rate ratio was seen in those aged 75+ years, where men had a 5.13-fold (95%CI 2.8, 9.5) increase in RRT initiation rate compared to women. The lower rates of women initiating RRT is unlikely to be due to death coming before RRT initiation since although 64.9% of the cohort at baseline were women, at the end of follow-up 67.0% of those still alive and not on RRT were women.
Conclusion: In Grampian, women had a higher prevalence of CKD than men. However, men with CKD initiated RRT at a faster rate than women. This imbalance was not explained by available confounders including, CKD stage comorbidity or baseline level of renal function but effect modification by age was shown to be important. Future research is required to investigate other explanations for this effect not explored here, including differences in interaction with health care.
Source of funding: The analysis for this work was done as part of a BMedSci intercalated degree, however a CSO grant (CZH/4/656) funded cohort set-up.
Conflict of Interests statement: None of the authors have any conflicting interests.
Aim: To investigate gender differences in CKD prevalence and outcomes in a large Grampian cohort, focussing specifically on RRT initiation.
Methods: All creatinine values measured in the Grampian region from mid-1999 to 2009 were available. All individuals who had an eGFR of less than 60ml/min/1.73m2 in 2003 that could be demonstrated to be present for at least three months were defined to have CKD in 2003. These individuals’ records (as part of the Grampian Laboratory Morbidity and Mortality Study II (GLOMMS-II)) were linked to hospital episode data (SMR01), local and national renal registry data and the National Records for comorbidity, RRT and mortality information.
The prevalence of stage 3-5 CKD in Grampian in 2003 was calculated. Univariate analysis identified potential confounding factors between men and women with stage 3-5 CKD and those who did and did not initiate RRT during six years of follow-up. The rate of RRT initiation by gender was calculated. Gender rate ratios were calculated using poisson regression, adjusting for age, CKD stage, proteinuria and significant co-morbidities. Gender rate ratios were also stratified by age.
Results: There were 19635 individuals who had stage 3-5 CKD in Grampian in 2003 (64.9% women). Grampian women in 2003 had a higher prevalence of CKD than men, at 5.7% (95%CI 5.6, 5.8) vs 3.2% (95%CI 3.1, 3.3). Men with CKD had a 3.4-fold (95%CI 2.6, 4.4) increase in RRT initiation compared to women. This relationship between gender and RRT initiation persisted, after adjustment for potential confounders. However, gender rate ratios stratified by age showed age to be an important effect modifier – the highest gender rate ratio was seen in those aged 75+ years, where men had a 5.13-fold (95%CI 2.8, 9.5) increase in RRT initiation rate compared to women. The lower rates of women initiating RRT is unlikely to be due to death coming before RRT initiation since although 64.9% of the cohort at baseline were women, at the end of follow-up 67.0% of those still alive and not on RRT were women.
Conclusion: In Grampian, women had a higher prevalence of CKD than men. However, men with CKD initiated RRT at a faster rate than women. This imbalance was not explained by available confounders including, CKD stage comorbidity or baseline level of renal function but effect modification by age was shown to be important. Future research is required to investigate other explanations for this effect not explored here, including differences in interaction with health care.
Source of funding: The analysis for this work was done as part of a BMedSci intercalated degree, however a CSO grant (CZH/4/656) funded cohort set-up.
Conflict of Interests statement: None of the authors have any conflicting interests.
| Original language | English |
|---|---|
| Pages | A4 |
| Publication status | Unpublished - Oct 2015 |
| Event | Scottish Renal Association: 2015 - Dundee, United Kingdom Duration: 1 Oct 2015 → 2 Oct 2015 |
Conference
| Conference | Scottish Renal Association |
|---|---|
| Country/Territory | United Kingdom |
| City | Dundee |
| Period | 1/10/15 → 2/10/15 |