Clarifying hierarchical age-period-cohort models: A rejoinder to Bell and Jones

Trends in activities of daily living disability among Chinese older adults from 1998 to 2018: an age-period-cohort analysis

This study aims to investigate the age, period, and cohort effects on trends in activities of daily living (ADL) disability among Chinese older adults; and to explore these three temporal effects on gender and residence disparities in disability. We utilized multiple cross-sectional waves of the Chinese Longitudinal Healthy Longevity Survey data (1998-2018), including 89,511 participants aged above 65 years old. Our measurement of disability is the number of ADL items (dressing, bathing, indoor transferring, toileting, eating, and continence) participants can't perform independently. Hierarchical age-period-cohort cross-classified random effects models were conducted to investigate age, period and cohort trends in ADL disability. Results showed that ADL disability increased with age at an increasing rate. A V-shaped cohort trend and a fluctuated period trend were identified. Females and urban residents were associated with more ADL limitations. When age increased, the gender and residence gaps in disability further increased. The cohort-based gender and residence inequalities in ADL limitations converged with successive cohorts. The period-based residence gap in ADL limitations diverged throughout the 20-year period, while the corresponding period-based change in gender disparity was not significant. These findings suggested that age, period, and cohort had different and independent effects on ADL disability among Chinese older adults. The age effect on trends in ADL is stronger compared to period and cohort effects. The gender and residence disparities in disability increased with age and decreased with successive cohorts. These patterns might help inform healthcare planning and the priorities for medical resource allocation accordingly.

Adolescent Marijuana Use in the United States and Structural Breaks: An Age-Period-Cohort Analysis, 1991-2018

To investigate temporal patterns, sociodemographic gradients, and structural breaks in adolescent marijuana use in the United States from 1991 to 2018, we used hierarchical age-period-cohort logistic regression models to distinguish temporal effects of marijuana use among 8th, 10th, and 12th graders from 28 waves of the Monitoring the Future survey (1991-2018). Structural breaks in period effects were further detected via a dynamic-programing-based method. Net of other effects, we found a clear age-related increase in the probability of marijuana use (10.46%, 23.17%, and 31.19% for 8th, 10th, and 12th graders, respectively). Period effects showed a substantial increase over time (from 16.23% in 2006 to 26.38% in 2018), while cohort effects remained stable throughout the study period. Risk of adolescent marijuana use varied by sex, racial group, family status, and parental education. Significant structural breaks during 1995-1996, 2006-2008, and 2011-2013 were identified in different subpopulations. A steady increase in marijuana use among adolescents during the latter years of this time period was identified. Adolescents who were male, were non-Black, lived in nonintact families, and had less educated parents were especially at risk of marijuana usage. Trends in adolescent marijuana use changed significantly during times of economic crisis.

Trends in cannabis use and attitudes toward legalization and use among Australians from 2001-2016: an age-period-cohort analysis

BACKGROUND AND AIMS

Changes in cannabis legalization and availability in Australia necessitate monitoring use and attitudes. We estimated age-period-cohort effects of past-year cannabis use and attitudes toward criminalization and legalization.

DESIGN

Analysis of six waves of the Australian National Drug Strategy Household Survey (NDSHS) every 3 years (2001-2016).

SETTING AND PARTICIPANTS

The study was nationally representative of Australian households using multistage random sampling, totaling 145 168 respondents 18-79 years old. Data were collected using telephone, face-to-face, and drop-and-collect. Sample sizes per analysis varied based on data availability (~107 000-127 000 per model).

MEASUREMENTS

Six waves of data for past-year cannabis use (by sex and education), attitudes toward criminalization and legalization.

FINDINGS

Past-year cannabis use decreased in young adults ages 18-35 from 2001-2016 (25.1%-18.6%) and increased in middle adults ages 36-55 (8.6%-10.1%) and older adults ages 56-79 (0.6%-3.0%). We observed a positive period effect and negative cohort effect for recent cohorts for past-year use (e.g. 1955 cohort had 1.41 (95% CI: [1.11, 1.70]) increased log odds vs. 1998 cohort had -2.86 (95% CI: [-3.17, -2.55]) increased log odds) compared with the mean across years. Results were consistent by sex and varied by education. We observed a negative period effect for criminalization favorability (0.14 (95% CI: [0.003, 0.28]) increased log odds in 2001 vs. -0.31 (95% CI: [-0.45, -0.17]) increased log odds in 2016) and positive cohort effect for recent cohorts. Last, we observed a positive period effect for legalization support (-0.03 (95% CI: [-0.20, 0.14]) increased log odds in 2001 vs. 0.38 (95% CI: [0.22, 0.55]) increased log odds in 2016) and negative cohort effect for recent cohorts.

CONCLUSION

Cannabis use appears to be increasing in Australia among adults over 35, while decreasing among adolescents and young adults. Legalization support also appears to have been increasing since 2007, signaling discordance between use and attitudes among adolescents and young adults, and potentially predicting increases in use over time.

Long-term trends in the body mass index and obesity risk in Estonia: an age-period-cohort approach

OBJECTIVES

To analyse the age, period and cohort effects on the mean body mass index (BMI) and obesity over the past two decades in Estonia.

METHODS

Study used data from nationally representative repeated cross-sectional surveys on 11,547 men and 16,298 women from 1996 to 2018. The independent effects of age, period and cohort on predicted mean BMI and probability of obesity (BMI ≥ 30 kg/m²) were modelled using hierarchical age-period-cohort analysis.

RESULTS

Curvilinear association between age and mean BMI was found for men, whereas the increase in mean BMI was almost linear for women. The predicted mean BMI for 40-year-old men had increased by 6% and probability of obesity by 1.8 times over 1996-2018; the period effects were slightly smaller for women. Men from the 1970s birth cohort had higher mean BMI compared to the average, whereas no significant cohort effects were found for obesity outcome.

CONCLUSIONS

Population-level BMI changes in Estonia during 1996-2018 were mostly driven by period rather than cohort-specific changes.

Recent increases in depressive symptoms among US adolescents: trends from 1991 to 2018

BACKGROUND

Mental health problems and mental health related mortality have increased among adolescents, particularly girls. These trends have implications for etiology and prevention and suggest new and emerging risk factors in need of attention. The present study estimated age, period, and cohort effects in depressive symptoms among US nationally representative samples of school attending adolescents from 1991 to 2018.

METHODS

Data are drawn from 1991 to 2018 Monitoring the Future yearly cross-sectional surveys of 8th, 10th, and 12th grade students (N = 1,260,159). Depressive symptoms measured with four questions that had consistent wording and data collection procedures across all 28 years. Age-period-cohort effects estimated using the hierarchical age-period-cohort models.

RESULTS

Among girls, depressive symptoms decreased from 1991 to 2011, then reversed course, peaking in 2018; these increases reflected primarily period effects, which compared to the mean of all periods showed a gradual increase starting in 2012 and peaked in 2018 (estimate = 1.15, p < 0.01). Cohort effects were minimal, indicating that increases are observed across all age groups. Among boys, trends were similar although the extent of the increase is less marked compared to girls; there was a declining cohort effect among recently born cohorts, suggesting that increases in depressive symptoms among boys are slower for younger boys compared to older boys in recent years. Trends were generally similar by race/ethnicity and parental education, with a positive cohort effect for Hispanic girls born 1999-2004.

CONCLUSIONS

Depressive symptoms are increasing among teens, especially among girls, consistent with increases in depression and suicide. Population variation in psychiatric disorder symptoms highlight the importance of current environmental determinants of psychiatric disorder risk, and provide evidence of emerging risk factors that may be shaping a new and concerning trend in adolescent mental health.

Healthy Eating among Mexican Immigrants: Migration in Childhood and Time in the United States

Past research on immigrant health frequently finds that the duration of time lived in the United States is associated with the erosion of immigrants' health advantages. However, the timing of U.S. migration during the life course is rarely explored. We draw from developmental and sociological perspectives to theorize how migration during childhood may be related to healthy eating among adult immigrants from Mexico. We test these ideas with a mechanism-based age-period-cohort model to disentangle age, age at arrival, and duration of residence. Results show that immigrants who arrived during preschool ages (2-5 years) and school ages (6-11 years) have less healthy diets than adult arrivals (25+ years). After accounting for age at arrival, duration of residence is positively related to healthy eating. Overall, the findings highlight the need to focus more research and policy interventions on child immigrants, who may be particularly susceptible to adopting unhealthy American behaviors during sensitive periods of childhood.

Drug-related deaths in Scotland 1979-2013: evidence of a vulnerable cohort of young men living in deprived areas

BACKGROUND

Even after accounting for deprivation, mortality rates are higher in Scotland relative to the rest of Western Europe. Higher mortality from alcohol- and drug-related deaths (DRDs), violence and suicide (particularly in young adults) contribute to this 'excess' mortality. Age-period and cohort effects help explain the trends in alcohol-related deaths and suicide, respectively. This study investigated whether age, period or cohort effects might explain recent trends in DRDs in Scotland and relate to exposure to the changing political context from the 1980s.

METHODS

We analysed data on DRDs from 1979 to 2013 by sex and deprivation using shaded contour plots and intrinsic estimator regression modelling to identify and quantify relative age, period and cohort effects.

RESULTS

The peak age for DRDs fell around 1990, especially for males as rates increased for those aged 18 to 45 years. There was evidence of a cohort effect, especially among males living in the most deprived areas; those born between 1960 and 1980 had an increased risk of DRD, highest for those born 1970 to 1975. The cohort effect started around a decade earlier in the most deprived areas compared to the rest of the population.

CONCLUSION

Age-standardised rates for DRDs among young adults rose during the 1990s in Scotland due to an increased risk of DRD for the cohort born between 1960 and 1980, especially for males living in the most deprived areas. This cohort effect is consistent with the hypothesis that exposure to the changing social, economic and political contexts of the 1980s created a delayed negative health impact.

Decomposing the effects of time on the social acceptability of biotechnology using age-period-cohort-country models

The study of European attitudes toward biotechnologies underlines a situation that is relatively contrasting in Europe. However, as different effects of time can influence the social attitudes (a life-cycle effect, a generational effect, and an exogenous temporal effect potentially affecting the entire population), an appropriate methodology should be used. To this end, age-period-cohort-country models have thus been estimated based on Eurobarometer data from 1991 onward. Applied to different data subsets, these models give similar results underlining the importance of the life-cycle effects as well as the heterogeneity of the link between political affiliation and biotechnologies attitudes across the European countries.

The Importance of the Baby Boom Cohort and the Great Recession in Understanding Age, Period, and Cohort Patterns in Happiness

Twenge, Sherman, and Lyubomirsky (TSL) claim that long-term cultural changes have increased young adults' happiness while reducing mature adults' happiness. To establish their conclusion, TSL use trend analyses, as well as more sophisticated mixed-effects models, but their analyses are problematic. In particular, TSL's trend analyses ignore a crucial cohort effect: well-known lower happiness among baby boomers. Furthermore, their data aggregation obscures the ephemerality of a recent period effect: the Great Recession. Finally, TSL overlook a key finding of their mixed-effects models that both pre- and post-Boomer cohorts became happier as they aged from young to mature adults. Our reanalyses of the data establish that the Baby Boomer cohort, the short-lived Great Recession, and unfortunate data aggregation account for TSL's results. The well-established, long-term relationship between age and happiness remains as it has been for decades despite any cultural shifts that may have occurred disfavoring mature adults.

The hierarchical age-period-cohort model: Why does it find the results that it finds?

It is claimed the hierarchical-age–period–cohort (HAPC) model solves the age–period–cohort (APC) identification problem. However, this is debateable; simulations show situations where the model produces incorrect results, countered by proponents of the model arguing those simulations are not relevant to real-life scenarios. This paper moves beyond questioning whether the HAPC model works, to why it produces the results it does. We argue HAPC estimates are the result not of the distinctive substantive APC processes occurring in the dataset, but are primarily an artefact of the data structure—that is, the way the data has been collected. Were the data collected differently, the results produced would be different. This is illustrated both with simulations and real data, the latter by taking a variety of samples from the National Health Interview Survey (NHIS) data used by Reither et al. (Soc Sci Med 69(10):1439–1448, 2009) in their HAPC study of obesity. When a sample based on a small range of cohorts is taken, such that the period range is much greater than the cohort range, the results produced are very different to those produced when cohort groups span a much wider range than periods, as is structurally the case with repeated cross-sectional data. The paper also addresses the latest defence of the HAPC model by its proponents (Reither et al. in Soc Sci Med 145:125–128, 2015a). The results lend further support to the view that the HAPC model is not able to accurately discern APC effects, and should be used with caution when there appear to be period or cohort near-linear trends.

Recent cohort effects in suicide in Scotland: a legacy of the 1980s?

BACKGROUND

Mortality rates are higher in Scotland relative to England and Wales, even after accounting for deprivation. This 'excess' mortality is partly due to higher mortality from alcohol-related and drug-related deaths, violence and suicide (particularly in young adults). This study investigated whether cohort effects from exposure to neoliberal politics from the 1980s might explain the recent trends in suicide in Scotland.

METHODS

We analysed suicide deaths data from 1974 to 2013 by sex and deprivation using shaded contour plots and intrinsic estimator regression modelling to identify and quantify relative age, period and cohort effects.

RESULTS

Suicide was most common in young adults (aged around 25-40 years) living in deprived areas, with a younger peak in men. The peak age for suicide fell around 1990, especially for men for whom it dropped quickly from around 50 to 30 years. There was evidence of an increased risk of suicide for the cohort born between 1960 and 1980, especially among men living in the most deprived areas (of around 30%). The cohort at highest risk occurred earlier in the most deprived areas, 1965-1969 compared with 1970-1974.

CONCLUSIONS

The risk of suicide increased in Scotland for those born between 1960 and 1980, especially for men living in the most deprived areas, which resulted in a rise in age-standardised rates for suicide among young adults during the 1990s. This is consistent with the hypothesis that exposure to neoliberal politics created a delayed negative health impact.