| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
1 Department of Sociology, Michigan State University, East Lansing.
2 Center for the Study of Aging and Human Development, Duke University, Durham, North Carolina.
3 Department of Sociology and the Population Research Center, University of Texas, Austin.
Address correspondence to Dr. Zhenmei Zhang, Department of Sociology, Michigan State University, East Lansing, MI 48824. E-mail: zhangz12{at}msu.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Data derived from two waves (1998–2000) of the Chinese Longitudinal Healthy Longevity Survey. We estimated logistic and multinomial regression models of cognitive impairment for a nationwide sample of people aged 80 to 105 (N = 8,444).
Results. Among both men and women, urban residence in early life as well as education was associated with lower odds of cognitive impairment at baseline. We found modest support for a protective effect of advantaged childhood background on the odds of cognitive impairment onset during the 2-year follow-up, especially among women.
Discussion. Our findings suggest that socioeconomic environment throughout the life course, early life in particular, can influence the risk of cognitive impairment in old age. Not only can public policy that targets illiteracy, hunger, and poverty improve the lives of tens of thousands of children, but ultimately such investments will pay significant dividends many decades later in enhancing the cognitive well-being of older persons.
Key Words: Early life conditions • Cognitive impairment • Oldest old • China
THE World Health Organization estimates that there will be about 29 million demented people worldwide by the year 2020 (Haan & Wallace, 2004
), and three quarters of this group will be in developing countries (Prince, 2000). Research over the past two decades in developed countries has emphasized adult risk factors for cognitive impairment (e.g., occupation, social networks, and leisure activities). Only recently has research begun to trace the origin of cognitive functioning in old age to early life conditions (Richards & Deary, 2005). For example, cross-sectional studies of the older population have reported that higher childhood socioeconomic status (SES) is associated with better cognitive functioning (Kaplan et al., 2001; Luo & Waite, 2005; Turrell et al., 2002). Overall, though, little research has examined early life adversity and cognitive impairment in late life, and even less research has investigated the effects of cumulative adversity over the life course on cognitive aging (Haan & Wallace, 2004).
In this study, we examined the association between early life conditions and cognitive impairment among oldest old persons in China. We add to the growing body of literature in this field in a number of ways. First, our focus on China provides a more global perspective on this association. Most existing research is based on data from developed countries, particularly European countries (Kaplan et al., 2001; Singh-Manoux, Richards, & Marmot, 2005; Turrell et al., 2002), with little attention to the childhood origins of adult health problems in developing countries where early life conditions were much worse than those in developed countries in the early part of the 20th century (Campbell, 1997). Many of China's current oldest old individuals suffered from hunger, wars, infectious diseases, and a severe lack of educational opportunities in their childhood (Zeng, Gu, & Land, 2007). It is still an open question whether these types of extremely adverse childhood conditions have left an indelible imprint on cognitive well-being at oldest old ages.
Second, we have taken advantage of a large population-based panel survey to assess whether childhood conditions are associated with both the prevalence and incidence of cognitive impairment at advanced ages. Previous work examining the effects of early life conditions on cognitive functioning has typically been based on small regional samples (Hall, Gao, Unverzagt, & Hendrie, 2000; Wilson et al., 2005) that focused on people younger than 80 (Luo & Waite, 2005; Turrell et al., 2002). And, except for two studies (Everson-Rose, Mendes de Leon, Bienias, Wilson, & Evans, 2003; Hall et al., 2000), previous research has relied on cross-sectional data, preventing an assessment of cognitive decline at advanced ages.
Third, we examined the effect of cumulative socioeconomic disadvantage on the odds of cognitive impairment at baseline and during follow-up. Current understanding of how childhood conditions combine with adult life circumstances to influence cognitive functioning is very limited. Do social disadvantages in early life, particularly those onerous conditions experienced by the Chinese oldest old population, create a life-long deficit in cognitive functioning? Or, do early life disadvantages affect cognitive decline by altering adult trajectories of achievement, which in turn affects cognitive decline in old age?
Increasingly, research on adult health has shown that morbidity, disability, and mortality are hinged to insults from negative events and exposures (e.g., poverty, family distress, unhealthy behaviors, and episodes of illness) that can accumulate over the life course (Kuh & Ben-Shlomo, 1997). Positive effects of experiences such as education and marriage can also accumulate and have lasting beneficial effects on health (Lillard & Waite, 1995; Lund, Holstein, & Osler, 2004; Ross & Wu, 1996). This so-called cumulative advantage/disadvantage model suggests that early advantages/disadvantages can compound over time by setting people onto different life trajectories (e.g., those from advantaged family backgrounds are more likely to enjoy higher levels of education, better jobs, higher income, and stable family relationships), which leads to increasing inequalities in health and well-being as people age (Crystal & Shea, 1990; Ferraro & Kelley-Moore, 2003; O'Rand & Hamil-Luker, 2005).
Although the life course approach is conceptually appealing, researchers' understanding of the specific mechanisms linking early life conditions and late-life cognition is fragmentary. Two perspectives, the latency model and the pathway model, potentially provide some insights into these mechanisms. The latency model suggests that early life circumstances have a direct association with late-life health outcomes such as cognitive impairment. Early life adversity can lead to fewer initial cognitive resources (i.e., less reserve) because the brain grows most during the prenatal period and the first 3 years of life and continues to grow in childhood and adolescence (Kim, Stewart, Shin, & Yoon, 2003). Impaired development of the brain due to early nutritional deprivation can result in less efficient brain function because of "less myelin, less branching of dendrites, and less developed connectivity patterns" (Moceri, Kukull, Emanuel, van Belle, & Larson, 2000, p. 415), although the brain can still function normally. The negative effects of the impaired brain development may be small until aggravated by the aging process (Moceri et al., 2000). A growing body of research has reported that early life nutritional deprivation is associated with small head size, short arm length and height, cognitive impairment, and dementia in later life (Jeong et al., 2005). If the latency model holds, early life adversity should be associated with poor cognitive functioning after one controls for what happens in adulthood.
The pathway model suggests, however, that early life circumstances may indirectly influence cognitive impairment via education attainment, occupational exposures and challenges, and illness. A large body of research across different countries has reported an inverse relationship between education and late-life cognition as measured by cognitive tests, cognitive decline, cognitive impairment, or dementia (e.g., Cagney & Lauderdale, 2002; M. Y. Zhang et al., 1990). The literature points to two explanations for the education–cognition link: (a) The rich environment in schools promotes brain growth in formative years and enables the brain network to operate more efficiently, thus providing a reserve against cognitive decline (Kaplan et al., 2001); and (b) education leads to mental stimulation throughout the life course because higher levels of education often lead to occupations that involve cognitive challenges and practice (Andel, Kareholt, Parker, Thorslund, & Gatz, 2007; Hauser & Roan, 2007; Schooler, 1987; Schooler, Mulatu, & Oates, 1999). In addition, poor early life conditions can increase the risk of several chronic conditions in adulthood, which in turn affects late-life cognitive functioning (Turrell et al., 2002). Those conditions that are frequently linked with cognitive impairment include hypertension, heart disease, and stroke (National Research Council, 2000). According to the pathway model, then, the association between early life adversity and cognitive impairment should be attenuated when adult SES and health are controlled. In sum, the origin of cognitive impairment in old age is most likely complex and multifactorial. Research over the past 20 years has pointed increasingly to the role of early life environments in the development of cognitive impairment and dementia.
In this article, we draw on two waves (1998–2000) of the Chinese Longitudinal Healthy Longevity Survey (CLHLS) to study the association between early life circumstances and cognitive impairment in oldest old Chinese. On the basis of cumulative advantage and disadvantage theory and results from previous research, we evaluated the following hypotheses. First, early life SES disadvantage would be associated with increased odds of cognitive impairment at baseline and onset over the 2-year follow-up. Second, the effect of early life environment on cognitive impairment would be partially mediated via adult SES and other adult factors. Third, those who were disadvantaged in both childhood and adulthood would be more likely to have cognitive impairment compared to those from advantaged backgrounds.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
).
This study focused on 8,805 respondents aged 80 to 105 in 1998. Those who reported being aged 106 or older were excluded because of insufficient information to validate their advanced age (Zeng & Vaupel, 2002). Out of the 8,805 baseline interviewees, 4,691 (53.3%) survived and were reinterviewed, whereas 3,264 (37.1%) died and 850 (9.6%) were lost to follow-up in the second wave in 2000. The CLHLS is the largest longitudinal survey of the oldest old population ever conducted in a developing country.
Measures
Cognitive impairment
The CLHLS measures cognitive function using the Chinese version of the Mini-Mental State Examination (MMSE). The MMSE was adapted from Folstein, Folstein, and McHugh (1975) and tests four aspects of cognitive functioning: orientation, calculation, recall, and language. On the whole, most items could be translated without modification. Several items were modified to meet the cultural and socioeconomic conditions in China and to make the questions easily understandable and practically answerable among oldest old Chinese whose cognitive function is normal. For example, the respondents were asked to name as many foods as possible instead of to write a sentence, which is an impossible task for members of this age group, the majority of whom are illiterate. The calculation question was also simplified (Zeng & Vaupel, 2002). Several similar versions of the Chinese MMSE, all of which were adapted from Folstein and colleagues (1975), are reliable and valid for use with Chinese older persons (Yu et al., 1989; Z. X. Zhang et al., 2006). The total possible score on the MMSE is 30, with lower scores indicating poor cognitive ability. Consistent with previous studies, the number of responses of "unable to answer" was much higher for the relatively difficult tasks compared with the relatively easy tasks. Based on recommendations in the literature, we counted responses of "unable to answer" as incorrect answers (Herzog & Wallace, 1997). For the current sample, the reliability of the MMSE scale was high (Cronbach's 
=.94). Respondents were defined as having moderate–severe cognitive impairment (hereafter cognitive impairment) in 1998 if their scores on the MMSE were less than 18 (Nguyen, Black, Ray, Espino, & Markides, 2003; Yu et al., 1989); respondents were considered to be experiencing the onset of cognitive impairment if their MMSE score fell below 18 during the 2-year follow-up among those whose initial MMSE scores were 18 or higher. The choice to use 18 as a cutoff point was due to previous research findings that those individuals who scored less than 18 had a very high probability of being cognitively impaired, regardless of education (Folstein et al., 1975; M. Y. Zhang et al., 1990). Previous clinical studies in China have also validated the use of 18 or 19 as a cutoff point for screening dementia among populations with little or no formal education (M. Zhang et al., 1998; Z. X. Zhang et al., 2006). We also found that our results were insensitive to varying the cutoff point up or down by one unit.
We examined the prevalence of cognitive impairment at baseline to document how early life and adult conditions affect the cognitive status of persons surviving to advanced ages. These persons, especially in a developing society such as China, represent the most robust of their birth cohorts. Prevalence reflects the stamp of the cognitive aging process that occurred prior to baseline on the surviving old-age population. The dependent variable in the prevalence model was a dichotomous variable, with cognitive impairment coded as 1. Onset between 1998 and 2000 refers to new events of cognitive impairment among those without impairment at baseline. To the extent that much of cognitive aging occurs prior to baseline, leaving behind a highly robust group of elders, examining incidence among oldest old persons offers a conservative approach to assessing the consequences of early life on the cognitive aging process. Admittedly, however, the number of events identifying the onset of cognitive impairment in our analysis was relatively small (247 events among men vs 528 events among women), which also points to the conservative evaluation of hypothesized effects. To take account of competing risks, the dependent variable in the incidence model had four outcomes: cognitive impairment, death, loss to follow-up, and cognitively normal (the base category).
Our analytic sample for the prevalence model included 8,444 respondents. This reflected the elimination of 361 respondents who did not participate in the cognitive tests due to various reasons including sickness, refusal, or hearing or vision impairment. There were 6,475 (76.7%) oldest old persons out of 8,444 respondents who did not suffer from cognitive impairment at baseline according to the above criterion (MMSE score 
18). Of 6,475 respondents in the incidence model, 775 (12.0%) experienced the onset of cognitive impairment, 1,886 (29.1%) died, 711 (11.0%) did not participate in the 2000 survey, and the remaining 3,103 (47.9%) maintained relatively normal cognitive functioning.
Early life SES
The CLHLS defines childhood as the period from birth to age 14. Early life conditions are indexed by three variables: place of birth, whether the respondent frequently went to bed hungry in childhood, and education. A proxy (i.e., spouse or a close family member) participated when the respondent was unable to answer some questions due to various reasons (e.g., cognitively impaired and could not understand the question). Approximately 20% of respondents had at least one proxy answer among the three childhood SES indicators. Place of birth was collected in the survey in terms of urban or rural area. Tremendous gaps in standard of living and health care existed between urban and rural areas, with rural areas disadvantaged in early 20th-century China where many of the oldest old participants were born and raised (Zeng et al., 2007). Rural/urban residence at birth thus reflected general individual as well as community socioeconomic conditions of the respondent's childhood. We dichotomized responses to the item about frequently going to bed hungry in childhood (went to bed without hunger = 1). This measure reflected both poverty as well as early nutritional deprivation.
We considered educational attainment mainly as an indicator of childhood SES, although it is arguably a measure that can reflect adult SES. We based our choice on the fact that whether one had schooling is a strong indicator of childhood SES among the oldest old Chinese because when most of them were children, the vast majority of schools in China were private (Kuo, 1915), and only those families who had enough resources could afford to send their children to schools (Zeng et al., 2007). In our sample, about 37% of men and 87% of women had no schooling, 46% of men and 10% of women had some elementary education, and less than 17% of men and 3% of women had a middle-school education or higher education. Therefore, we dichotomized the education variable into had schooling (coded 1) and no schooling.
Adult SES
Adult SES was measured by respondents' primary lifetime occupation before age 60 and current residence (urban = 1). We classified own occupation into two categories: professional/administration and other low-SES job. Researchers frequently use urban residence as a major indicator of SES among Chinese elders due to the sharp divide between urban and rural areas in income, health care, school quality, access to public goods such as housing, sanitation, and other dimensions of welfare (e.g., Liang et al., 2000; Zhu & Xie, 2007). Although current residence could potentially differ from the primary residence during adulthood, we expected a negligible difference because the household registration system established in 1955 limited rural–urban migration (Li, 2005). The pilot reform of the household registration system started only a couple of years ago, and the oldest old population is the least likely adult age group to migrate.
Other adult conditions
We included marital status (married = 1), number of children who frequently come to visit (0–5), involvement in leisure activities, and disability in 1998 in our incidence model; except for marital status, we did not include these measures in the prevalence model due to possible endogeneity. Measures of leisure activities (i.e., gardening, playing cards/mah-jongg, listening to radio/watching TV) were dichotomized, with 1 = everyday/sometimes and 0 = never. Disability was measured in terms of six functional problems: bathing, dressing, toileting, indoor transferring, maintaining continence, and eating. If respondents needed help in one or more items, we classified them as disabled.
Control variables
Age was measured as a continuous variable.
Analytic Strategies
We estimated gender-specific models because previous research on early life influences on late-life health has suggested that early life risk factors may operate differently for men and women (Best, Hayward, & Hidajat, 2005; Kim et al., 2003). In the prevalence model, we used binary logistic regression to model how early life conditions were associated with the probability of being cognitive impaired at baseline. We first estimated a reduced form model to identify the total effect of early life SES on the odds of having cognitive impairment and then estimated an expanded model to determine whether the effect was reduced after we introduced adult SES. If the effect of early life SES is substantially reduced, this would lend support to the pathway model whereby early life SES indirectly affects the probability of cognitive impairment. If the effect of early life SES is not reduced, this would lend support to the latency model and suggest that early life SES may have a direct effect on cognitive impairment.
We modeled the onset of cognitive impairment over a 2-year observation period using multinomial logistic regression. These models included death and attrition as competing risks, as death and attrition may preempt the observation of the onset of cognitive impairment, and these processes were likely to be linked to cognitive functioning problems. We adopted a nested modeling strategy to evaluate the spectrum of associations of early life and adult SES with the onset of cognitive impairment. We first examined how early life conditions (i.e., rural/urban residence at birth and hunger at bedtime in childhood) affected the onset of cognitive impairment controlling for age (Model 1), and then we added educational attainment (Model 2), adult SES (Model 3), and other adult risk factors (Model 4). Finally, we added respondents' MMSE scores at baseline as a covariate and examined whether cognitive functioning differences among nonimpaired persons affected our results (Model 5). Because the sampling weights in the data set are solely a function of major independent variables (i.e., age, gender, and rural/urban residence) used in the analysis, the unweighted estimates are preferred because they are unbiased, are consistent, and have smaller standard errors than weighted estimates (Winship & Radbill, 1994). We performed all analyses using Stata Version 10.0 (StataCorp, 2007). The tables reporting the results of the logistic regression models express the effects in terms of relative odds.
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Early Life Influences on the Prevalence of Cognitive Impairment
The results in Table 2 show that for both men and women, markers of early life SES advantage were significantly associated with lower odds of cognitive impairment. In Model 1, oldest old men born in urban areas had a 61% lower risk of being cognitively impaired; oldest old women born in urban areas had a 34% lower risk of being cognitively impaired compared to those born in rural areas, controlling for marital status and age. Freedom from frequent hunger appeared to have a slightly protective effect for women but was not associated with cognitive impairment for men. When education was added to Model 1 (see Model 2), the protective effect of birth residence was slightly attenuated; however, the change was not statistically significant. Comparing Models 1 and 2 we can conclude that (a) education lowered the risk of cognitive impairment by 40% for men and 38% for women, and (b) the protective effect of urban birth was not attributable to education. Instead, childhood birth place and education had independent effects on the odds of cognitive impairment. When adult SES was included in Model 3, the associations between early life SES and cognitive impairment continued to be robust. In addition, neither occupation nor urban residence in adulthood was associated with the odds of cognitive impairment among men. However, among women, adult urban residence reduced the odds of cognitive impairment, net of early life conditions and demographic controls. In additional analyses (results not shown), we estimated a model including only adult SES and demographic controls. We found that without controlling for childhood SES, adult occupation was significantly associated with the odds of cognitive impairment for both men and women. For example, men and women who had worked in professional/administrative jobs were 33% and 38% less likely, respectively, to have cognitive impairment than those employed in lower SES jobs. However, although adult SES was associated with the probability of being cognitively impaired, the association seemed spurious because early life SES influenced both adult SES and cognitive impairment. This result was largely consistent with findings from prior studies in developed countries (e.g., Cagney & Lauderdale, 2002; Karp et al., 2004).
|
Early Life Conditions and the Onset of Cognitive Impairment
The incidence models shown in Tables 3 (men) and 4 (women) show that better early life conditions also appeared to offer some modest protection against the onset of cognitive impairment, especially for women. Among men, the effects of early life SES were in the expected direction, although none were statistically significant at the p <.05 level (see Models 1 and 2 of Table 3). Results in Model 3 of Table 3 show that current residence was statistically associated with onset: Men living in urban areas were 33% less likely to experience onset compared to those in rural areas. Adult occupation was not significantly associated with onset of cognitive impairment. Among women, Model 1 of Table 4 shows that women born in urban areas were 32% (p <.05) less likely to experience onset than their counterparts born in rural areas. Freedom from hunger reduced the risk by 17% (p <.1). Model 2 shows that those who had received some schooling were 30% (p <.05) less likely to experience onset compared to those who had received no schooling. When adult SES was added in Model 3, the results showed very little attenuation of the protective effects of early life conditions for women.
|
|
To gain some substantive sense of how early life SES combines with adult SES to influence the onset of cognitive impairment, we calculated the odds ratios of having cognitive impairment for those who were in the most advantaged childhood and adult SES positions versus those who were in the most disadvantaged conditions, controlling for age only. The calculations were based on results from Model 3 of Tables 3 and 4. For men, the odds of onset during the 2-year follow-up were 61% (1.03 x 0.88 x 0.80 x 0.81 x 0.67 = 0.39) lower for those who were most advantaged in both childhood and adulthood than for those who were disadvantaged throughout life. Among oldest old women, the pattern was similar. The odds of onset were 53% (0.72 x 0.85 x 0.69 x 1.10 x 1.02 = 0.47) lower for those who were most advantaged in both childhood and adulthood than for those who were most disadvantaged all their lives.
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Moceri et al., 2000). Although our data did not allow us to examine the specific reasons behind the protective effects of childhood urban residence, we suspect that they relate to the higher standard of living in urban areas (including presence of health facilities and access to information) as compared to rural areas during the early 20th century in China. We also found that frequent hunger was associated with women's higher risk of cognitive impairment but not men's. The reason for this is not clear. Due to the low status of girls in traditional Chinese families, we suspect that if a family had both girls and boys (as most Chinese families did at that time), undernutrition would be more severe among girls than boys because sons were often favored and given better food (Watson, 1991). However, we found only modest support for the protective effect of advantaged childhood background on the odds of impairment onset, and the effect seemed to be stronger for women than men. Overall, our results suggest that having socioeconomic advantages over the entire life course offers the greatest protection from cognitive impairment. As for the mechanisms linking early life SES and the risk of cognitive impairment, we found relatively strong support for the latency model and only modest support for the pathway model. Our results suggest that early life conditions are directly associated with the risk of cognitive impairment in old age. For instance, in our prevalence models, the magnitude and significance of the effect of rural/urban residence at birth and education changed very little with the addition of adult SES and other risk factors. The support for the pathway model is very modest, as can be seen from the slight attenuation of the effect of early life SES when adult factors were included.
This study has several strengths, including a large nationwide sample of oldest old persons in a developing country, the use of longitudinal data, and an exploration of the mechanisms linking early life SES and cognitive impairment. However, readers should consider several limitations when interpreting the results. First, a brief epidemiological screening instrument (MMSE) for cognitive impairment rather than comprehensive clinical evaluations was used to detect cognitive impairment. Clinical evaluations are usually more accurate. Due to budget limitations, the CLHLS did not conduct clinical tests. Second, our measures of early life conditions were relatively crude. Although childhood residence, education, and frequent hunger before going to bed captured major aspects of economic hardship in childhood, we could not identify the extent of poverty and nutritional deprivation, or the timing and duration of hunger. Third, our measures of adult SES were also relatively crude. We lacked detailed information about the characteristics of the respondents' occupation and financial resources. Fourth, our measures of early life conditions were based mainly on self-reports, and as with other studies that have used self-reports, there is the potential problem of recall bias, especially among those who are cognitively impaired. Finally, a caveat in interpreting the results is that our sample of oldest old individuals was a select group of robust survivors of much political, economic, and social unrest that had occurred in 20th-century China. They had survived Japanese invasion in the mid-1930s, numerous internal wars, famines (e.g., the 1959–1961 famine), and the Cultural Revolution (1966–1976). We suspect that our estimates of the impact of early life SES may be conservative due to selective mortality, because previous research in China has found that people with poor early life SES and/or adult SES are more likely to die than their advantaged counterparts (Liang et al., 2000; Zeng et al., 2007).
Despite these limitations, our study contributes to the ongoing discussion on early life conditions and late-life cognitive well-being. We found support for the hypothesis that the onset of cognitive impairment in old age may have an early origin. From a policy perspective, our study suggests that interventions should focus on the entire life course, and the formative years in particular. We found that even for the robust oldest old Chinese, early life conditions can protect against cognitive impairment and thus significantly improve the quality of life in old age. Not only can public policy in developing countries that targets illiteracy, hunger, and poverty improve the lives of tens of thousands of children, but ultimately it appears that such investments will pay significant dividends many decades later in enhancing the cognitive well-being of older persons.
|
Acknowledgments |
|---|
Zhenmei Zhang initiated and designed the study, supervised the data analysis, and wrote and revised the article. Danan Gu drafted the Methods section, analyzed the data, and contributed to the study design and revising of the article. Mark D. Hayward contributed to the study design, data analysis and interpretation, and revision of the article.
|
Footnotes |
|---|
Received for publication May 9, 2007. Accepted for publication October 9, 2007.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Fors, C. Lennartsson, and O. Lundberg Childhood Living Conditions, Socioeconomic Position in Adulthood, and Cognition in Later Life: Exploring the Associations J Gerontol B Psychol Sci Soc Sci, May 6, 2009; (2009) gbp029v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
