Modelling old‐age retirement: An adaptive multi‐outcome LAD‐lasso regression approach

PurposeMany international retirement migrants are amenity movers undertaking the first move in the late life course model of migration. The purpose of this paper is to examine second moves within the retirement destination community to test whether the model of late life course migration accurately portrays the motivations and housing choices local movers make after retiring to another country.Design/methodology/approachThe paper combines secondary data and survey results to examine the composition of the retiree migrant population in the Alicante province of Spain. The socioeconomic characteristics and housing choices of those who have made a second move since retiring to Spain are compared with those who have not moved through a series of t‐tests and chi‐square tests.FindingsThe paper finds that those who have made a second move within Spain are somewhat typical of second movers in the late life course. They are likely to cite mobility or health problems as a reason for moving and appear to recognize the need for a home that provides living area on one floor. Yet, they are choosing to move within an area that does not provide them with access to informal family care givers.Research limitations/implicationsThe data are restricted to retirees of two nationalities in one province of Spain. Further research is suggested in other locations and with retirees of other nationalities for comparison.Practical implicationsBecause many international retirees do not plan to return to their countries of origin, they will create demand for formal in‐home care services and supportive retiree housing in the near future in their retirement destination countries.Originality/valueThis paper provides understanding of a growing consumer housing segment in retirement destinations.

This study examines the transition from independent living to a coresidential living arrangement across the late life course among older unmarried (i.e., widowed, divorced/separated, and single) Mexican Americans. Using 18 years' worth of panel data from the Hispanic Established Populations for the Epidemiologic Study of the Elderly, event history analyses revealed that age at migration, physical disability, and cognitive impairment were strong predictors of the transition to a coresidential living arrangement. Importantly, a decline in physical and cognitive abilities heightened the risk of transition to a coresidential living arrangement, net of time-variant measures of disability and impairment. These findings provide evidence for incorporating a dynamic approach to examining living arrangements across the late life course for Mexican-origin Hispanics living in the United States, with implications for policy and service providers.

Due to the aging baby boom population, utilization rates of diagnostic imaging (i.e., X-ray, CT, and MRI scanning) have risen rapidly relative to other health services. The aim of this study is to investigate the utilization patterns of outpatient diagnostic imaging services (X-ray, CT, and MRI) across the late life course (65 years and older). A population-based retrospective cohort study was conducted for the period April 1, 2005, to March 31, 2006. All Ontario residents aged 65+ and eligible for government health insurance were included in the analysis. Utilization of diagnostic imaging followed an inverted U-pattern: increasing with advancing age, peaking in the 80-84 age group for CT scans and in the 70-74 age group for MRI and X-rays, and then declining in the later years. Overall, females received significantly more X-rays than males (p < .01), but males received significantly more CT and MRI scans (p < .01). A small proportion of high-users of radiology services accounted for a large proportion of overall utilization. Finally, our analysis revealed that a disproportionately large proportion of high-users of MRI services were in the highest SES quintile. No SES differences were observed for X-ray or CT scans. Population aging will lead to increased demand for healthcare services. Utilization of outpatient diagnostic imaging services is associated with age, gender, and SES. Given the increasing demand and the limited resources available, there may be a need for programs to target underserved populations to reduce remediable inequities. Whereas patient-level decisions regarding the use of diagnostic imaging are rightfully determined on the basis of clinical factors, allocation decisions should also be informed by the ethical principles of equity and fairness.

New ways of imagining, planning and living old age are actually emerging in the republic of Benin, West Africa. This process could be understood as the dissemination of an idea of retirement from the sector of formal labor and the corresponding social security system to a general notion of a good life in the late life course. It is preceded by emerging age-inscriptions which are contouring the new up to a point that it is settled and becoming a norm or a dominant pattern. It is also linked to the emergence of new a African middle class. It is going hand in hand with the emergence of other changes in the imaginaries of the life- course, for instance new ways of living and imagining youth. Additionally, it goes along with an accelerating process of social differentiation, since living old age as retirement is, for the moment, only possible for people who are more or less doing well and able to gain the necessary resources of self-maintenance during a time after work. Thus, retirement has become, beyond the sphere of formal work, a generalized notion of new pathways of old age. However, up to now, the desire to live old age as retirement is still an emerging age- inscription and has not become the dominant norm.

BackgroundThe appropriate handling of missing covariate data in prognostic modelling studies is yet to be conclusively determined. A resampling study was performed to investigate the effects of different missing data methods on the performance of a prognostic model.MethodsObserved data for 1000 cases were sampled with replacement from a large complete dataset of 7507 patients to obtain 500 replications. Five levels of missingness (ranging from 5% to 75%) were imposed on three covariates using a missing at random (MAR) mechanism. Five missing data methods were applied; a) complete case analysis (CC) b) single imputation using regression switching with predictive mean matching (SI), c) multiple imputation using regression switching imputation, d) multiple imputation using regression switching with predictive mean matching (MICE-PMM) and e) multiple imputation using flexible additive imputation models. A Cox proportional hazards model was fitted to each dataset and estimates for the regression coefficients and model performance measures obtained.ResultsCC produced biased regression coefficient estimates and inflated standard errors (SEs) with 25% or more missingness. The underestimated SE after SI resulted in poor coverage with 25% or more missingness. Of the MI approaches investigated, MI using MICE-PMM produced the least biased estimates and better model performance measures. However, this MI approach still produced biased regression coefficient estimates with 75% missingness.ConclusionsVery few differences were seen between the results from all missing data approaches with 5% missingness. However, performing MI using MICE-PMM may be the preferred missing data approach for handling between 10% and 50% MAR missingness.

Inter-connected trends in cognitive aging and depression: Evidence from the health and retirement study

BackgroundAging and rural-urban disparities are two major social problems in today’s ever-developing China. Much of the existing literature has supported a negative association between adverse community setting with the cognitive functioning of seniors, but very few studies have empirically investigated the impact of rural-urban community settings on cognitive decline in the late life course of the population in developing countries.MethodsData of seniors aged 65 or above (n = 1709) within CHARLS (The China Health and Retirement Longitudinal Study, a sister study of HRS), a nationally representative longitudinal cohort (2011–2015) in China, were analyzed using a multilevel modeling (MLM) of time within individuals, and individual within communities. Cognitive impairment was assessed with an adapted Chinese version of Mini-Mental State Examination.ResultsUrban community setting showed a significant protective effect (β = − 1.978, p < .000) on cognitive impairment in simple linear regression, and the MLM results showed it also had a significant lower cognitive impairment baseline (β = − 2.278, p < .000). However, the curvature rate of cognitive decline was faster in urban community setting indicated by a positive interaction between the quadratic time term and urban community setting on cognitive impairment (β = 0.320, p < .05). A full model adjusting other individual SES factors was built after model fitness comparison, and the education factor accounted for most of the within and between community setting variance.ConclusionsThe findings suggest that urban community setting in one’s late-life course has a better initial cognitive status but a potentially faster decline rate in China, and this particular pattern of senior cognitive decline emphasize the importance of more specific preventive measures. Meanwhile, a more holistic perspective should be adopted while construct a risk factor model of community environment on cognitive function, and the influence at society level needs to be further explored in future research.

BackgroundPopulation aging poses significant challenges to primary care providers and healthcare policy makers. Primary care reform can alleviate the pressures, but these initiatives require clinical benchmarks and evidence regarding utilization patterns. The objectives of this study is to measure older patients' use of health services, number of health conditions, and use of medications at the level of a primary care practice, and to investigate age- and gender-related utilization trends.MethodsA cross-sectional chart audit over a 2-year study period was conducted in the academic family practice clinic of Sunnybrook Health Sciences Centre in Toronto, Ontario, Canada. All patients 65 years and older (n = 2450) were included. Main outcome measures included the number of family physician visits, specialist visits, emergency room visits, surgical admissions, diagnostic test days, inpatient hospital admissions, health conditions, and medications.ResultsOlder patients (80-84 and 85+ age-group) had significantly more family physician visits (average of 4.4 visits per person per year), emergency room visits (average of 0.22 ER visits per year per patient), diagnostic days (average of 5.1 test days per person per year), health conditions (average of 7.7 per patient), and medications average of 8.2 medications per person). Gender differences were also observed: females had significantly more family physician visits and number of medications, while men had more specialist visits, emergency room visits, and surgical admissions. There were no gender differences for inpatient hospital admissions and number of health conditions. With the exception of the 85+ age group, we found greater intra-group variability with advancing age.ConclusionThe data present a map of greater interaction with and dependency on the health care system with advancing age. The magnitudes are substantial and indicate high demands on patients and families, on professional health care providers, and on the health care system itself. There is the need to create and evaluate innovative models of care of multiple chronic conditions in the late life course.

Age-related effects on ambulatory care service utilization are not well understood. We aim to measure the utilization patterns of ambulatory health care services (i.e. family physician visits, specialist physician visits and emergency room visits) in the late life course (65 years and older). A population-based retrospective cohort study was conducted for the period 1 April 2005 to 31 March 2006. All Ontario, Canada, residents aged 65+ and eligible for government health insurance were included in the analysis. This population-based cohort study demonstrates considerable increase in utilization rates and variability of ambulatory services as age increases. Variations in utilization were observed by gender as overall women were more likely to consult a family physician, and men more likely to visit specialists and the emergency room. A small group of high users, constituting 5.5% of the total population, accounted for 18.7% of total ambulatory visits. Finally, we report socio-economic status (SES) based disparity for specialist services in which high users were more likely to have higher SES. There is increasing utilization and variability in ambulatory service utilization with increase in age. Further research is required to explain the gender and SES differences reported in this study.

How to treat families within the German pay-as-you-go financed social insurance systems - this question is repeatedly discussed. A closer look on the statutory pension scheme as well as the statutory health insurance and the care insurance scheme reveals indeed, that people without children are treated to generously within these systems. This will place an additional burden on future generations. Therefore, reforms are necessary. In the statutory pension scheme benefits can be related to the number of children a person raised. In the statutory health and in the statutory care insurance scheme a second, capital funded pillar can be introduced.

Kidney disease and the cumulative burden of life course socioeconomic conditions: The Atherosclerosis Risk in Communities (ARIC) Study

Kidney Disease in Life-Course Socioeconomic Context: The Atherosclerosis Risk in Communities (ARIC) Study

Excess number of zeros (zero inflation, ZI) in count data is a common phenomenon which must be addressed in any analysis. The extra zeros may be a result of over-dispersion in the data. Ignoring zero-inflation can result in biased parameter estimates and standard errors. Over-dispersion is also associated with a zero-inflated data. Depending on the selected model, different results and conclusions may be reached. In this paper two commonly encountered models in count data are considered, namely, the Zero-Inflated Poisson (ZIP) and Zero-Inflated Negative Binomial (ZINB) probability distributions. Emphasis is placed on the Maximum Likelihood (ML) estimation of the model parameters. Specifically of interest was to es-timate the zero-inflation parameter and hence, the corrected frequencies. It was found that for the Poisson model, the zero-inflation parameter estimate was considerably higher than that from the Negative Binomial model. From the results however, it is suspected that the effectiveness of adjusting for the high number of zeros in both models might have been greatly affected by the inherent high variability between sites. It is then proposed that in future research, the problem of heterogeneity in count data be addressed before any further analysis.

BackgroundThere is no consensus on the most appropriate approach to handle missing covariate data within prognostic modelling studies. Therefore a simulation study was performed to assess the effects of different missing data techniques on the performance of a prognostic model.MethodsDatasets were generated to resemble the skewed distributions seen in a motivating breast cancer example. Multivariate missing data were imposed on four covariates using four different mechanisms; missing completely at random (MCAR), missing at random (MAR), missing not at random (MNAR) and a combination of all three mechanisms. Five amounts of incomplete cases from 5% to 75% were considered. Complete case analysis (CC), single imputation (SI) and five multiple imputation (MI) techniques available within the R statistical software were investigated: a) data augmentation (DA) approach assuming a multivariate normal distribution, b) DA assuming a general location model, c) regression switching imputation, d) regression switching with predictive mean matching (MICE-PMM) and e) flexible additive imputation models. A Cox proportional hazards model was fitted and appropriate estimates for the regression coefficients and model performance measures were obtained.ResultsPerforming a CC analysis produced unbiased regression estimates, but inflated standard errors, which affected the significance of the covariates in the model with 25% or more missingness. Using SI, underestimated the variability; resulting in poor coverage even with 10% missingness. Of the MI approaches, applying MICE-PMM produced, in general, the least biased estimates and better coverage for the incomplete covariates and better model performance for all mechanisms. However, this MI approach still produced biased regression coefficient estimates for the incomplete skewed continuous covariates when 50% or more cases had missing data imposed with a MCAR, MAR or combined mechanism. When the missingness depended on the incomplete covariates, i.e. MNAR, estimates were biased with more than 10% incomplete cases for all MI approaches.ConclusionThe results from this simulation study suggest that performing MICE-PMM may be the preferred MI approach provided that less than 50% of the cases have missing data and the missing data are not MNAR.

We examine the gender wealth gap with a focus on pension wealth and statutory pension rights. By taking into account employment characteristics of women and men, we are able to identify the extent to which the redistributive effect of pension rights reduces the gender wealth gap. The data for our analysis come from the German Socio-Economic Panel (SOEP), one of the few surveys that collects information on wealth and pension entitlements at the individual level. Pension wealth data are available in the SOEP for 2012 only. While the relative raw gender wealth gap is about 35% (or 31,000 euros) when analysing the standard measure of net worth, it shrinks to 28% when pension wealth is added. This reduction is due to redistributive elements such as caregiver credits provided through the statutory pension scheme. Results of a recentred influence functions (RIF) decomposition show that pension wealth reduces the gap substantially in the lower half of the distribution. At the 90th percentile, the gender wealth gap in net worth and in augmented wealth remains more stable at roughly 27–30%.