* * * * *
Issues of Social Development in Adulthood

Yvonne Gonzalez
University of Phoenix
PSY/201
Foundations of Psychology
Tracy S. Ramos, Faculty
April 7th, 2013

* * * * * * * * * I found this article to be very interesting. Surprisingly, it was difficult to locate a credible article that discussed women in their midlife; which it is why I really liked this article. Life Course Transitions and Depressive Symptoms among Women in Midlife, discusses about different issues that has accrue in adult women through midlife. It also discusses how midlife in adult women can affect their social development. This article will provide you with transitioning stage of midlife in adult women, as well as some of the consequences that can occur as women transition into midlife. It gives you step by step as to how each transition will affect her. This study was done on women from the ages 50-59 to show midlife transition and depression will affect the women. It will also tell you things about the male midlife transition and how they will react differently than women in their midlife transition. * Yes, I find that this article would be a great choice as a main source for a research paper. I say this because; it provides you with credible information and statistics on social development in women transitioning through midlife. It also provides you with information on how to overcome the issues that may affect social development. This resource is very informative and it provides you with actual statistical data found through research studies. It explains how marital transition will be differently when you are a young adult as oppose to when you are in your midlife transition. I believe that social development can be summed up to the negative and positive perceptions within adults. The article gives you a better understanding of how adults will reach a wellness and gives an understanding of how social development can be an issue in women entering midlife.

INT’L. J. AGING AND HUMAN DEVELOPMENT, Vol. 58(4) 241-265, 2004
LIFE COURSE TRANSITIONS AND DEPRESSIVE
SYMPTOMS AMONG WOMEN IN MIDLIFE
M. JEAN TURNER
TIMOTHY S. KILLIAN
REBEKAH CAIN
University of Arkansas, Fayetteville
ABSTRACT
This study examined the relationship between three midlife transitions and depressive symptoms among 952 women 50 to 59 years of age. Using longitudinal data from women interviewed for the 1992 and 2000 Health and
Retirement Study, the study described changes in marital status, change to a parental caregiving role, and changes in perceived health across the eight years. Further, it examined the impact of these changes on mental health.
The findings indicate that becoming widowed, becoming a caregiver, and perceiving health declines significantly increased depressive symptoms in the year 2000, even when controlling for pre-transition levels of depressive symptoms. The findings are consistent with the lifecourse perspective that individual development occurs in context and across the lifespan. The findings confirm and add to current midlife research literature.
INTRODUCTION
Midlife is increasingly understood as a period of life that is characterized by transitions that may cause stress and have consequences for the mental health of adults (Aldwin & Levenson, 2001). Much of what we know about midlife transitions and their consequences is based on cross sectional research. This study used nationally representative longitudinal data to examine the relationship between midlife transitions and depressive symptoms among women.
241
2004, Baywood Publishing Co., Inc.
Specifically, we examined relationships between depressive symptoms and three common midlife transitions: transitions in marital status, the transition to caregiving, and perceived changes in physical health. Researchers, professionals, and policymakers must understand the relationship between common transitions in midlife and the mental health of adults in order to adequately respond to the needs of this growing population.
BACKGROUND
The conceptualization of midlife as an important period of development for adults is relatively recent (Skolnick, 1991). One of the reasons that little attention has been given to development during midlife is the perception that midlife is relatively free of social and psychological challenges (Lachman & James, 1997;
Staudinger & Bluck, 2001). The growth of the lifecourse perspective as a way of understanding human development has increased interest in adult development overall. One of the most basic themes of the lifecourse perspective is that development continues across the lifespan. Despite an increased interest in adult development, much of the research has focused on the development of older adults. The end result is the appearance that researchers have focused on the beginning and the end of the life course, with little attention to the middle. However, as the segment of the population comprised by those in midlife grows, and as the lifecourse perspective continues to provide fertile ground for research, researchers are finding that midlife is an interesting and important developmental period in the life course.
As research on midlife grows, it is becoming clear that transitions in midlife have both positive and negative consequences for mental health (Marks &
Lambert, 1998; Marks, Lambert, & Choi, 2002; Moen, Dempster-McClain, &
Williams, 1992). For example, it is relatively common for midlife adults to make the transition to providing care for older family members and to spend a significant amount of time providing that care (Brody, 1995; Stone, Cafferata, & Sangl,
1987). Life course transitions are often associated with increased stress, defined as any event that causes pressure to change (Boss, 1988; Kramer, 1993; McCubbin,
1995). Although stress is neither good nor bad, coping with stress is a process that includes a period of mental disorganization, often accompanied by an increase in depressive symptoms, before individuals are able to reorganize to their previous levels of functioning (McCubbin & Patterson, 1983). As a result, transitions experienced in midlife may have negative consequences for mental health, such as increased symptoms of depression.
The Lifecourse Perspective
The lifecourse perspective provides a framework that recognizes the complexity of life pathways, family relationships, and social and temporal contexts
242 / TURNER, KILLIAN AND CAIN through which development occurs (Moen & Wethington, 1999). A basic tenet of the lifecourse perspective is the interconnectedness of the individual life course and one’s developmental trajectory with others’ lives and developmental trajectories
(Elder, 1998). Life course transitions must be examined with an understanding of the role of personal and social history, family relationships, and the structural and social contexts that influence individuals’ adaptation to these transitions. Systematic differences in how well people cope with life course transitions can be partly accounted for by the contexts within which those transitions are experienced. Two kinds of contexts are particularly relevant for this study: the temporal context and the context of gender.
The Temporal Context
Although midlife is the subject of increased attention from researchers, there is little consensus about when midlife begins and ends. Early developmental theoretical perspectives proposed that midlife began somewhere between 25 and
40 years of age (Erikson, 1963; Levinson, 1978). However, the changing social context in which midlife is experienced has resulted in midlife being experienced later and lasting longer than proposed by early theorists. The extension of the lifespan has led some to conclude that midlife begins at around 40 years of age and lasts about 20 years (Staudinger & Bluck, 2001).
The lifecourse perspective draws attention to the importance of timing and the role of timing in how people experience and cope with life course transitions.
It is likely that some transitions are more relevant at some periods of the life course than at others, and, therefore, are experienced differently at different periods in life (Elder, 1998). For example, marital transitions may be experienced differently by young adults than by adults in midlife (Uhlenberg, Cooney, &
Boyd, 1990). Therefore, the theme of timing underlines the importance of examining the relationship between transitions and mental health at particular periods in the life course, such as midlife. Not only do transitions occur in the temporal context, but they also occur and are coped with in the context of gender.
The Gender Context
Men and women often experience transitions differently. For example, women are more likely to experience widowhood than men (Treas, 1995). Further, women experience widowhood differently than do men (Marks, 1996; Martin-
Matthews, 1999). Gender is also related to how people react to transitions in their relationships with their parents. For example, traditional gender roles suggest that “good” women are able, willing, and available to care for older parents and parents-in-law (Alford-Cooper, 1993; Hooyman & Gonyea, 1995).
In fact, Stone et al. (1987) found that women comprise 80% of all those who provide assistance to older family members. Further, Allen, Blieszner, and
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 243
Roberto (2000) reported that researchers have consistently found that daughter’s are three times more likely to assist aging parents with activities of daily living than are sons. Traditional gender roles have socialized women into the nurturing roles of mothering and caregiving. Therefore, they are more likely than men to be caught in the middle of conflicting obligations, and to experience psychological distress and increased symptoms of depression after the transition to caregiving (Desbiens, Mueller-Rizner, Viring, & Lynn, 2001; Faison,
Faria, & Frank, 1999; Gallicchio, Siddiqi, Langenberg, & Baumgarten, 2002;
Kramer & Kipnis, 1995; Martin, 2000; Yee & Schulz, 2000). In addition, women experience different health changes than men do (Spiro, 2001) and respond to them in different ways (Brandtstaedter, Wentura, & Greve, 1993;
Reboussin et al. 2000; Stewart & Ostrove, 1998). From this review, it is clear that the midlife experiences of women are substantively different from the midlife experiences of men. We believe these differences warrant separate investigations. Because of these differences and the need to clarify the complex midlife transitions specific to women, this study focuses only on women. Understanding the issues specific to women will enhance future comparative studies of the role of midlife transitions on the psychological wellbeing of both men and women.
In sum, timing and gender are important contexts within which people experience and cope with life course transitions. Not only are people likely to find some transitions more relevant at particular times in the life course than at other times, but women are also likely to react differently to those transitions than men. Moreover, gender and temporal contexts often interact as individuals experience midlife transitions. For example, becoming separated or divorced during midlife is likely to be more traumatic for women than for men (Marks & Lambert, 1998). One of the factors facilitating an easier adjustment for men is that men are likely to have significantly more opportunities to build a new relationship and get remarried in mid- and later life than are women.
Whereas some researchers have compared life course transitions and mental health across these contexts (e.g., Marks et al., 2002), we examined the relations between life course transitions within the contexts of midlife and gender. Marks and Lambert (1998) provided important new insights in their examination of marital transitions across the temporal context. In addition, Marks (1998), Marks and Lambert (1998), and Marks et al. (2002) were among the first to explore transitions rather than statuses. Building on their work, we examined the relationship between life course transitions and mental health within both gender and temporal contexts. Further, we explored three different transitions simultaneously within these contexts. Specifically, we examined three kinds of transitions that are likely to be relevant to women in midlife: transitions in marital status, transitions to parental caregiving, and transitions in self-perceived health, and the relationship of these transitions to mental health.
244 / TURNER, KILLIAN AND CAIN
Midlife Transitions
Marital Transitions
Although generally considered a late life event, the likelihood of becoming a widow increases dramatically for women during midlife (Aldwin & Levenson,
2001: Connidis, 2001). Nearly 16% of all women between 45 and 64 years old are widows, compared to 4% of men in that age group (Fields & Casper, 2001).
Not surprisingly, researchers have found that a period of psychological distress, including higher rates of depression, follows the transition to widowhood
(Arbuckle &de Vries, 1995; Lopata, 1996; Marks, 1996; Marks&Lambert, 1998;
Martin-Matthews, 1999).
Because people are more likely to experience divorce early in life, divorce is less frequently experienced during midlife than at earlier times in the life span
(Uhlenberg et al., 1990). Nevertheless, Uhlenberg et al. suggested that one in eight 40-year-old married women are likely to experience divorce at some time in their remaining lifespan. Although divorce may be accompanied by a sense of relief (Spanier & Thompson, 1983), it is often followed by lower levels of psychological functioning, including higher rates of depression (Cherlin, 1992).
Moreover, these effects are likely to be felt more strongly among women than among men (Connidis, 2001; Marks & Lambert, 1998).
Generally, previous research has found that marriage has been related to positive mental health outcomes for women (Gove, Style, & Hughes, 1990;
Marks, 1996; Marks & Lambert, 1998). Specifically, Marks (1996) suggested having an intimate confidant is likely one reason married people report higher levels of mental health than unmarried people. It may also be that the higher levels of income associated with being married contribute to higher levels of mental health (Umberson, Wortman, & Kessler, 1992). Although previous studies have reported that marriage is related to higher levels of mental health, few have examined the relationship between mental health and the transition to marriage, especially when the transition to marriage occurs during midlife. One of the few studies that did examine the transition to marriage during midlife found that getting married for the first time was related to positive mental health outcomes for midlife women (Marks & Lambert, 1998). However, the same study found no positive effect for getting remarried in midlife. In fact, the mental health of women getting remarried in midlife was similar to that of those who had been continuously married from young adulthood.
The Caregiving Transition
As their parents grow older, adult children are increasingly likely to provide them with assistance (Brody, 1995; Stone et al., 1987). Moreover, midlife adults may be providing assistance at levels that put their own wellbeing at risk
(Abel, 1991). DeVanzo and Rahman (1993) found that women are likely to spend
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 245 more years of their lives caring for parents over 65 years of age than for children under the age of 18 years. Stone et al. (1987) suggested that women caregivers spend an average of 18 years providing assistance to their aging parents. Although the link between the caregiving role and psychological distress has been well established (e.g., Desbiens et al., 2001; Faison et al., 1999; Gallagher-Thompson
& Powers, 1997; Lee, Walker, & Shoup, 2001; Martin, 2000; Schulz, O’Brien,
Bookwala, & Fleissner, 1995; Strawbridge, Wallhagen, Shema, & Kaplan, 1997;
Whitlatch & Feinberg, 1997), only a few researchers have examined the
“transition” to caregiving (e.g., Marks, 1998; Marks et al., 2002; Strawbridge et al., 1997).
Previous research has generally found that increased caregiving duties are related to psychological distress for caregivers (e.g., Desbiens et al., 2001;
Faison et al., 1999; Gallicchio et al., 2002; Marks, 1998). Strawbridge et al. (1997) found that the caregiving role was related to increased psychological distress, regardless of the relationship of the caregiver to the care recipient. Similarly, using longitudinal data from the National Survey of Families and Households
(NSFH), Marks et al. (2002) found that the experience of the transition to parental caregiving was ambiguous, leading to both psychological gains and losses, including higher rates of depressive symptoms for those who were providing care to a parent living either in or out of their household.
Transitions in Perceived Health Status
Health status has been found to be related to psychological wellbeing, including depression, during midlife (e.g., Aro, Nyberg, Absetz, Henriksson, & Loennqvist,
2001; Grant, Bartolucci, & Giger, 2000; Reboussin et al., 2000). As people grow older and experience declines in their health, they may be required to discontinue activities that are important to them (Whitbourne, 2001). Discontinuing important activities may be related to higher rates of depressive symptoms for women in midlife, especially if they have difficulty modifying or reducing the expectations that they have for themselves (Brandtstaedter et al., 1993; Reboussin et al., 2000).
In addition, health concerns may be particularly important for women during midlife. Although women are less likely to die during midlife than men, they are more likely to experience the onset of long-term, chronic disease than men
(Spiro, 2001). These factors, combined with declines in body image satisfaction
(Deeks & McCabe, 2001) and society’s valuing of youthfulness and vitality
(Stewart & Ostrove, 1998), contribute to increased psychological distress among midlife women.
Summary and Hypotheses
Although midlife is a rewarding time of life for most adults, some life course transitions may put the mental health of midlife adults, especially women, at risk.
Specifically, transitions in marital status, the transition to providing care for
246 / TURNER, KILLIAN AND CAIN parents, and perceived declines in health may be related to increased depressive symptoms among midlife women. The purpose of this study was to examine how these midlife transitions were related to symptoms of depression in a nationally representative sample of midlife women.
Because this study uses longitudinal data to address these issues, it improves on past research in two ways. First, we were able to control for participants’ earlier symptoms of depression. Secondly, we were able to examine changes over time to determine how the transitions in marital status, caregiving roles, and self-perceived health are related to mental health during midlife.
Based on previous research, we made the following global hypotheses:
1. Transitions in marital status during midlife will add significantly more explained variance in depressive symptoms than that explained by previous levels of depressive symptoms and other control variables.
2. The transition to the parent caregiver role will explain significantly more explained variance than that explained by previous levels of depressive symptoms and other control variables.
3. Changes in participants’ health status will add significantly more explained variance in symptoms of depression than that explained by earlier symptoms of depression and other control variables.
The following, more precise hypotheses were examined only for the global hypotheses that were supported:
4. The transition to marriage will be related to fewer symptoms of depression.
5. The transition to separation or divorce will be related to increased symptoms of depression.
6. The transition to widowhood will be related to increased symptoms of depression. 7. The transition to caregiving will be related to more depressive symptoms.
8. Perceived declines in participants’ health will be related to more depressive symptoms. METHODS
Data and Sample
The data for this study came from the Health and Retirement Study (HRS). In the initial HRS wave conducted in 1992, over 12,600 persons in 7,600 households were interviewed in their homes. In order to be eligible to participate in the
HRS, households had to have at least one member who was born between 1931 and 1941. Spouses of the eligible household members were also interviewed, regardless of their age. Therefore, although participants in the HRS varied widely in age, HRS data are only representative of those who were born between 1931 and 1941. HRS participants were contacted four additional times (i.e., in 1994,
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 247
1996, 1998, and 2000). This study used data from the HRS that were collected from female participants who were born between 1931 and 1941 and who participated in the initial (i.e., 1992) wave and the final (i.e., 2000) wave of the HRS.
Additionally, because one of the interests of this study was the transition to caregiving for older family members, only data from those participants who had at least one parent living during both HRS waves were examined.
In addition to using raw HRS data, this study also used data from the RAND corporation (RAND, 2002). RAND data included variables that were based on complex combinations of raw HRS variables. For example, the RAND data set included the variable marital status of participants for both 1992 and 2000. The
RAND marital status variable used two variables from the raw HRS variables to create a single and valid variable for participants’ 1992 marital status. RAND used information from 22 different raw HRS variables to create a single and valid variable for participants’ marital status during the final HRS wave. Because of the complexity of deriving seemingly simple variables from raw HRS data,
RAND data were consulted when possible. For variables that were not available in the RAND data set, raw HRS data were used and merged with RAND data to create a single data set for analyses. The RAND corporation provides detailed descriptions about how variables have been derived.
Data for the HRS were collected using a multi-stage area probability sampling design. There are two important characteristics of the sampling strategy used to collect HRS data. First, primary sampling units (PSUs) were chosen so that oversamples of special populations (i.e., minorities and residents of Florida) would be represented in the data. Second, the sampled unit was households. As a result of the complex sampling strategy used to collect these data, sampling weights are provided to allow researchers to control for oversamples by using the strata weights, and to allow researchers to do individual level analyses by using the individual unit of analyses weights. In this study, we conducted both weighted analyses using both strata and individual unit of analyses weights, and unweighted analyses of the data. Generally, both methods of analyses yielded similar results, and the few inconsistent findings between unweighted and weighted analyses are noted in the discussion of the results.
In 1992, there were 2,381 female participants between the ages of 51 and 61 who had at least one parent living. Of those, 1,891 respondents also participated in the final HRS wave. Therefore, nearly 80% of the women who participated in the first HRS wave also participated in the final HRS wave. Of those who participated in both HRS waves, only 984 individuals had at least one parent who was still living during the final HRS wave, making them eligible to participate in this study. Of the 984 eligible participants in both waves of the study, 32 women were dropped because of missing data. In the end, data from 952 participants was analyzed.
A comparison between the 952 participants who were included in the final sample and the 1,427 participants who were not included was conducted.
248 / TURNER, KILLIAN AND CAIN
Chi-square analyses indicated that those included in the final sample were not significantly different in terms of race or Hispanic origin from those who were not included. However, t-tests indicated significant differences between participants who were included in the final analyses and those who were not. Specifically, these analyses indicated that participants who were included in the final sample were younger, t(2379) = 7.09, p < .01, had completed more years of education, t(2379) = –3.11, p < .01, and had higher total household incomes, t(2379) = –3.38, p < .01 than participants who were not included in the final sample. Therefore, these variables were included in the final analyses as control variables.
Based on weighted analyses, nearly 87% of participants were white, about 10% were black, and nearly 3% were members of other racial or ethnic groups. Nearly
6% of the participants indicated that they had Hispanic origins. Of the participants with Hispanic origins, over 74% identified themselves as white, about
2.5% identified themselves as black, and about 23% identified with other racial categories. In 1992, participants ranged in age from 51 to 61 years, and their weighted mean age was 54.4 years (SE = 0.10). Years of completed education ranged from 0 to 17. Participants’ completed weighted average was 12.6 years of school (SE = 0.09). Participants’ total household incomes ranged from $0 to
$514,200 per year, with a weighted annual mean of $51,466 (SE = $1863.19).
Measures
Participants’ level of depressive symptoms in 2000 was the predicted variable in this study. Participants’ level of depressive symptoms in 1992 was regressed on 2000 levels of depressive symptoms as a control variable. Other predictor variables included life transition variables (i.e., marital transitions, transitions to caregiving, and transitions in perceived health).
Depressive Symptoms
RAND data were used to create two measures of depressive symptoms— one measure for the 1992 data and one measure for the 2000 HRS wave. These measures used eight items based on the Center for Epidemiological Studies
Depression Index (CES-D) developed by Radloff (1977). The wording and scoring of the items at the two data collection times were slightly different.
Specifically, in 2000 participants were asked to indicate, on a yes or no scale, whether or not they had experienced particular feelings during the past week. For example, one item was, “Now think about the past week and the feelings you have experienced. Please tell me if each of the following was true for you much of the time this past week. Much of the time during the past week, have you felt happy?”
In 1992, participants were asked to indicate how often during the past week they had experienced particular feelings on a 4-point scale. For example, one item was,
“Please tell me how often you have experienced the following feelings during the past week—all of the time, most of the time, some of the time, or none or almost
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 249 none of the time. During the past week, how often have you felt depressed?” To make the scales consistent across time, 1992 items were transformed by changing all of the time and most of the time to yes and changing some of the time and none of the time to no (RAND, 2002).
In 1992, participants responded to 16 items derived from the CES-D. In 2000, participants were asked to respond to only eight items. Therefore, only the eight items common to both HRS waves were used in this study. Higher scores indicated higher levels of depressive symptoms for six items. The remaining two items were reversed, and all eight items were summed to create a single score for depressive symptoms. The Cronbach alphas for the revised 1992 and 2000 scales were .79 and .78, respectively. The 1992 scale ranged from 0 to 8, with a weighted mean of 0.78 (SE = 0.05). The scale for the final wave had a weighted mean of 1.48
(SE = 0.07) and also ranged from 0 to 8.
The transformations of the full CES-D scale implemented by RAND are based on the suggestions of Kohout, Berkman, Evans, & Cornoni-Huntley
(1993) and used in the Established Populations for the Epidemiological Study of the Elderly (EPESE) survey to compare original CES-D scores with the shortened version of the CES-D. Moreover, using data from the AHEAD (Study of Asset and Health Dynamics Among the Oldest Old) sample, Turvey, Wallace,
& Herzog (1999) concluded that the yes/no format and the shortened length of the CES-D did not negatively affect the psychometric properties or factor structure of the scale.
Marital Status Transitions
Details about participants’ marital status and transitions in marital status are provided in Table 1. Marital status and transitions in marital status are based on RAND data. In 2000, 594 participants were married. Of those, three had never been married, 10 were separated or divorced, and three had been widowed in 1992. Therefore, 16 participants had made a transition to marriage between the two HRS waves. Transition to marriage was coded as 1 if participants had made the transition, and 0 if they had not made the transition to marriage.
There were 174 participants who were separated or divorced in 2000. Of those,
22 had been married in 1992. Also, of those separated or divorced in 2000, two participants had never been married and two participants had been widowed in 1992, suggesting that they had made a transition to marriage and to being separated or divorced in the eight years between the two data points. In the end,
26 participants had made the transition to being separated or divorced by the year 2000. The transition to separated or divorced was coded as 1 for participants who made the transition and 0 for all other participants.
In the year 2000, there were 154 participants who were widowed. Of those,
66 had been married in 1992. Also, 15 of those participants were separated or
250 / TURNER, KILLIAN AND CAIN divorced in 1992. It is not clear whether or not participants considered themselves widowed because the partner from whom they were separated or divorced had died, or if they had remarried and their new marriage partner had died before the 2000 data were collected. Overall, there were 81 participants who had made the transition to widowhood between the two waves of the study. The transition to widowhood was coded as 1 if participants had become widowed between the first and last HRS waves. Otherwise, the transition to widowhood was coded as 0.
All marital transitions were coded as 1 if participants had experienced the transition between 1992 and 2000. Otherwise, transitions were coded as 0. The non-coded comparison category, therefore, is not experiencing a marital transition between 1992 and 2000. Of all 952 participants, 123 experienced at least one marital transition between 1992 and 2000.
The Parental Caregiver Transition
Raw HRS data about caregiving were consulted and merged with RAND data to assess the caregiver transition. In 1992, participants were asked to respond to the following item, “Have you (or your husband/partner) spent 100 or more hours in the past 12 months helping your parent(s) (or stepparents) with basic personal needs like dressing, eating, and bathing?” Participants responded to a similar question in the final HRS wave. If participants had not provided assistance to at least one of their parents in 1992 and did provide assistance in
2000, the caregiving transition was coded as 1. Otherwise, the transition to caregiving was coded 0. Sixty-five participants were providing help to at least one parent in 1992. Of the 155 participants who were providing assistance to at least one parent in 2000, 130 had made the transition to the caregiver role between 1992 and 2000.
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 251
Table 1. Participants 1992 Marital Status by 2000 Marital Status
Participants’ 2000 marital status
Participants’ 1992 marital status Married
Never
married
Separated
or divorced Widowed
Married
Never married
Separated or divorced
Widowed
577
3
10
3
0
30
2
0
22
2
147
2
66
0
15
73
Note: It is not clear why two participants who indicated that they were separated or divorced in 1992 indicated that they had never been married in 2000. These participants were coded as not making a marital transition between 1992 and 2000.
Participants’ Perceived Health Status
Two RAND variables were used to assess changes in participants’ health.
First, participants in each HRS wave were asked to rate their overall health.
Participants responded on a 5-point scale that ranged from excellent to poor, with higher scores indicative of poorer health. To calculate a perceived change in health variable, participants’ responses in 1992 were subtracted from their responses in 2000. The result was a measure of change in their current health, with possible scores that ranged from –4 to 4. A score of 0 indicated no change in health between the two time periods. Positive scores indicated health deterioration, and negative scores indicated health improvement. Approximately
16% of the participants experienced health improvement, 47% did not perceive a health change, and 37% of the participants indicated that they had experienced a decline in health between the two data collection points. Participants’ scores ranged from –3 to 4 with a weighted mean of 0.25 (SE = 0.03).
A second measure of health change was based on an item that asked participants whether or not their health had improved or deteriorated in the last two years.
Participants responded on a 5-point scale that ranged from somewhat better to much worse, with higher scores indicating greater health deterioration. Responses to this item ranged from 1 to 5 with a weighted mean of 3.07 (SE = 0.03).
Both health variables were included in this study because each assesses health from a different perspective. The health transition variable is comparable to the other two transition variables in that it assesses changes across the eight-year period between the 1992 data collection and the 2000 data collection. The second health variable assessing perceived changes in health during the past two years allowed us to evaluate the impact of more recent perceived health changes on depressive symptoms. Using both variables allowed us to examine how perceived health changes were related to depressive symptoms among women in midlife.
RESULTS
Data were analyzed using multiple regression. First, the global hypotheses that marital status transitions, caregiving transitions, and perceived health status/ changes in health would be predictive of participants’ symptoms of depression after controlling for previous depressive symptoms (i.e., Hypotheses 1–3) were tested. Participants’ ages, years of completed education, and yearly total household incomes were also included as control variables. Hierarchical multiple regression was used to examine these hypotheses (Cohen & Cohen, 1983). First, the control variables, including participants’ 1992 depressive symptoms scores, were regressed on 2000 depressive symptoms scores. Next, marital status transition variables, the caregiving transition variable, and the health transition variables were entered as sets, and the change in R-square was observed. If the
252 / TURNER, KILLIAN AND CAIN changes in R-square were significant for any of the variable sets, the hypotheses were supported.
As can be seen in Table 2, Hypothesis 1 was supported. Specifically, Model 2 included participants’ transitions in marital status and explained a significantly greater amount of the variance than was explained by Model 1, which included only the control variables, F (3, 946) = 4.83, p .01. Therefore, participants’ transitions in marital status were related to depressive symptoms, even after controlling for their levels of depressive symptoms prior to their transitions in marital status.
Similarly, Hypothesis 2 was supported. Model 3 included participants’ transition to caregiving and explained significantly more of the variance than did
Model 1, F (1, 948) = 9.52, p < .01 (see Table 2). Participants’ transition to caregiving was related to their levels of depressive symptoms after controlling for previous levels of depressive symptoms and other control variables.
Hypothesis 3 was also supported. Model 4 included participants’ health variables, their previous symptoms of depression, and other control variables, and explained 26% of the variance in depressive symptoms that was measured in
2000. As can be seen in Table 2, this was an increase in R-square of .12, which is a statistically significant increase, F (2, 947) = 30.31, p < .01.
In sum, the results indicated that participants’ marital transitions, their transitions to caregiving, changes in their perceived health status, and recent changes in health were predictive of participant’s current levels of depressive symptoms even after controlling for participants’ previous levels of depressive symptoms, ages, completed years of education, and yearly total household income. Because
Hypotheses 1 through 3 were supported, all of the variables were entered into a full regression model to explain how specific predictor variables were related to participants’ symptoms of depression (i.e., Hypotheses 4 through 8).
Control Variables
As can be seen in Table 3, participants’ symptoms of depression in 2000 were positively related to their symptoms of depression in 1992, as expected. Women with more symptoms of depression in 1992 also reported more symptoms of depression in 2000. Moreover, the number of years of education completed by participants was negatively related to their symptoms of depression in 2000.
Neither participants’ ages nor their yearly total household incomes were related to participants’ symptoms of depression in 2000.
Marital Transitions
As can be seen in Table 3, the hypothesis that getting married would be related to fewer depressive symptoms (i.e., Hypothesis 4) was only partially supported.
Based on weighted analyses, participants who got married between the two data collection points experienced fewer depressive symptoms in 2000 than those not making a marital transition. However, the relationship was small and not
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 253
Table 2. Hierarchical Regression Analyses Predicting Depression: Weighted (bw), Unweighted (buw), and Standardized (Beta) Parameter Estimates
Model 1 Model 2 Model 3 Model 4 bw buw Beta bw buw Beta bw buw Beta bw buw Beta
Depression (1992)
Participants’ Age
Years of education
Annual household income
Marital Transition Variables
Transition to marriage
Transition to separation/ divorce Transition to widowhood
0.42**
0.00
–0.09**
–0.00
0.42**
0.01
–0.09**
–0.00
.33
.01
–.14
–.05
0.42**
–0.01
–0.08**
–0.00
–0.77*
–0.24
0.70**
0.43**
0.00
–0.09**
–0.00
–0.75
–0.69
0.75**
.34
.01
.13
–.05
–.05
–.06
.11
0.42**
–0.00
–0.09**
–0.00
0.42**
–0.01
–0.09**
–0.00
.33
–.01
–.13
–.05
0.41**
0.00
–0.08**
–0.00
0.41**
0.01
–0.09**
–0.00
.32
.02
–.13
–.04
254 / TURNER, KILLIAN AND CAIN
Caregiving Variable
Transition to caregiving
Health Variables
Transition in perceived health status
Perceived changes in health (in past 2 years)
Intercept
R-square(uw)
Change in R-square
2.23 1.88
.166
2.55 2.29
.178
.012*
0.53**
2.33
0.52**
2.04
.174
.008*
.09
0.25**
0.40**
0.83
0.23**
0.41**
0.43
.216
.050**
.12
.17
Note: All changes in R-squares are calculated by comparing the R-square in Model 1.
*p < .05. **p < .01.
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 255 statistically significant in the unweighted analyses. The hypothesis that getting separated or divorced would be related to more symptoms of depression (i.e.,
Hypothesis 5) was not supported. The hypothesis that making the transition to widowhood would be positively related to symptoms of depression (i.e.,
Hypothesis 6) was supported. Participants who had become widows between the two data collection points experienced more depressive symptoms than participants who had not experienced a marital transition during that time.
Caregiver Transition
Hypothesis 7 was supported. Participants who made the transition to caregiving reported more symptoms of depression than other participants. As can be seen in
256 / TURNER, KILLIAN AND CAIN
Table 3. Full Regression Model Predicting Depression during the
2000 HRS Wave: Weighted (bw), Unweighted (buw), and Standardized (Beta) Parameter Estimates
Variable bw buw Beta
Depressive symptoms (1992)
Participants’ age
Years of education
Annual household income
Transition to marriage
Transition to separation/divorce
Transition to widowhood
Transition to caregiving
Transition in perceived health
Perceived change in health
(in past 2 years)
Intercept
R-square
0.42**
–0.01
–0.08**
–0.00
–0.86*
–0.19
0.70**
0.46*
0.40**
0.24**
1.26
0.42**
–0.00
–0.08**
–0.00
–0.82
–0.41
0.64**
0.49**
0.40**
0.23**
1.02
.235
.33
–.00
–.12
–.04
–.05
–.04
.09
.09
.11
.23
*p < .05. **p < .01.
Table 3, participants who became caregivers between 1992 and 2000 had higher levels of depressive symptoms than other participants.
Participants’ Change in Self-Perceived Health
As hypothesized (i.e., Hypothesis 8), participants’ perceived health was significantly related to symptoms of depression (see Table 3). Therefore, the hypothesis that participants’ perceptions of declines in health and their recent changes in health would be positively related to symptoms of depression was supported.
DISCUSSION
In general, the results of this study point to the relevance of transitions for midlife women and the role of these transitions in mental health. Transitions in marital status, specifically to widowhood and possibly to marriage, the transition to the parental care role, and changes in perceived health status among midlife women are likely to have important implications for women’s mental health in midlife. The findings of this study largely confirm the findings of previous studies.
In addition, this research adds much to the literature on midlife transitions and mental health. Unlike many previous studies that are based on cross-sectional data, these findings provide a longitudinal comparison of depressive symptoms before and after the transitions have actually occurred during the eight years between data collection.
Control Variables
Not surprisingly, the level of depressive symptoms at the first data collection point was positively related to depressive symptoms at the last data collection point. This finding confirms much previous research suggesting that although life events and transitions are related to depressive symptoms in adulthood, depressive symptomology is relatively stable (Burns & Seligman, 1989; Kendler,
Gardner, & Prescott, 2002).
We made no hypotheses about how the age of the participants, the number of years of education that participants had completed, or participants’ yearly total household incomes would be related to depressive symptoms. However, we included these variables as control variables on the basis of our preliminary data analyses. Of these variables, only “number of years of education” was related to depressive symptoms. It is not clear why. It may be that increased levels of education provided participants with more resources to cope with the transitions they were experiencing during midlife (McCubbin &
Patterson, 1983).
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 257
Marital Transitions
It is unclear why the hypothesis that getting married in midlife is related to lower levels of depressive symptoms among women was not more strongly supported. Although the relationship was significant in the weighted analyses, it was not significant in the unweighted analyses, suggesting that the relationship between getting married and depressive symptoms of women in midlife is relatively weak. It may have been that getting married in midlife is off time and, therefore, not related to a decrease in depressive symptoms during midlife.
In addition, the women in this study who had gotten married during the eight years between data collection points may have been marrying for a second or third time. The research suggesting that mental health improves when midlife women marry specified that this benefit was only for those marrying for the first time (Marks & Lambert, 1998). Those who were remarrying did not report the same level of improvement in mental health as those marrying for the first time. It is possible that enough of the women in our study were remarrying to impact the statistical results of the study. Previous research also suggests that women who marry in midlife may benefit as a result of gaining an intimate partner in whom they can confide (Marks, 1996) or an increase in wealth associated with getting married (Umberson et al., 1992). This may be reflected in our relatively weak finding that getting married was related to lower rates of depressive symptoms. Future research should examine the processes through which getting married in midlife is related to other aspects of wellbeing for midlife women.
The hypothesis that women’s experience of divorce in midlife is related to more depressive symptoms was not supported. It is likely that the experience of divorce was not uniform across participants in this study. In other words, divorce may have been accompanied by a sense of relief for some participants, and served as a source of psychological distress for others (Spanier & Thompson, 1983).
If the final years of the marriage had been highly conflictual, relief may have been the dominant response to the divorce, much as adult children leaving home or retirement from a highly stressful job can be a relief for some rather than the distressing experience it is for others. Also, most previous studies have not focused on how divorce or separation impacts women during their midlife years. Most divorce research has been done with younger adults. However, as researchers begin to focus on the midlife years, some evidence suggests that midlife women do not experience divorce as negatively as younger women.
Specifically, Marks and Lambert (1998) found that women over 40 years of age reported fewer depressive symptoms after divorce or separation than did younger women. It is also possible that many of the women in our study had begun their adaptation to divorced status early in the eight years between the two waves of data collection. Research suggests that recovery from a divorce
258 / TURNER, KILLIAN AND CAIN takes from three to 10 years for women (Saxton, 1993). Therefore, those who started their adjustment earlier in the eight years between data points may have already significantly adjusted to their divorce status before the second data collection. Not surprisingly, the findings of this study confirm previous research suggesting that women who experience widowhood during midlife are likely to experience higher levels of depressive symptoms (Arbuckle & de Vries, 1995;
Lopata, 1996; Marks, 1996; Martin-Matthews, 1999). It should be noted that our method of operationalizing the transition to widowhood was based on participants’ experiences in the eight years prior to the second wave of data collection. Therefore, our second wave of data collection likely occurred when many participants were still negotiating bereavement and therefore in a period of relative disorganization. Other studies have suggested that, following a period of disorganization, widows are likely to reorganize and, perhaps, experience psychological growth after the transition to widowhood (Lieberman, 1996;
Wells & Kendig, 1997). It may be that the reorganization is still in process for many of the women in our study.
The Caregiver Transition
Although the operational definition of caregiving in this study was very broad, women in this study who became parental caregivers between the two data points were likely to report more depressive symptoms than women who did not become caregivers. This finding is consistent with previous research (Marks et al., 2002;
Strawbridge et al., 1997). Past research does not provide a consensus regarding the processes through which increased parental caregiving is related to increased psychological stress. It may be that women in midlife are likely to find themselves sandwiched between two generations to whom they are obligated to provide assistance (e.g., Brody, 1995). An alterative explanation is that being caught in the middle of competing family obligations is a relatively rare experience, and that it is more likely that midlife women are experiencing stress as the result of balancing increasing work responsibilities and caregiving (Abel, 1991; Gottlieb,
Kelloway, & Fraboni, 1994; Rosenthal, Martin-Matthews, & Matthews, 1996).
Although the processes through which the transition to caregiving is related to mental health are not resolved by the findings in our research, the findings do demonstrate the need for further research into these issues. Because our study examined those who had transitioned to the caregiving role in the previous eight years, it adds important information to the current literature about the mental health consequences of becoming a care provider. Examining the caregiving role among the same women after they have been at that role for a longer period of time would allow us to determine if adjustment and adaptation across time diminish the mental health consequences of caring for aging parents.
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 259
Health Status
The mental wellbeing of women in this study was related to participants’ perceived declines in health. This finding is consistent with previous research that has found that lower levels of health are related to higher rates of psychological distress (Shulz et al., 1995; Whitlatch & Feinberg, 1997). It may be that some of the women in this study have given up activities that are important to them because of perceived health limitations. Scaling back these activities may have contributed to higher rates of depressive symptoms. This conclusion would be consistent with earlier research that suggests physical limitations may prevent adults in midlife from engaging in activities that are important to them (Whitbourne, 2001) which may lead to increased depressive symptoms (Brandtstaedter et al., 1993). It may also have been that some of the midlife women in this study experienced more depressive symptoms because small and gradual declines in their physical health led to lower satisfaction with their body images. This explanation is also consistent with previous research (Deeks & McCabe, 2001; Reboussin et al., 2000). It is not surprising that self-perceived declines in health may be related to a negative view of oneself. This is particularly likely in Western cultures that value youthfulness and vitality. Because the aging of women is viewed more negatively than the aging of men (Stewart & Ostrove, 1998), the relationship between declining health and negative body image is particularly relevant for women.
Whereas other studies have found lower levels of perceived health status to be related to mental health, our study adds the transition component indicating that both longer-term and more recent changes in perceived health are significantly related to symptoms of depression among midlife women. Thus, not only does the perception of being in a lower health status contribute to depressive symptoms, but the process of becoming less healthy does as well.
Limitations
The findings of this study generally confirm previous research. Although the longitudinal component of this study adds to current literature, it is important to recognize some limitations of the study. First, most of the participants in this study were white. Although some minority participants were included, there were not enough to compare the relationship between midlife transitions and symptoms of depression across racial groups. Moreover, there are reasons to believe that midlife transitions, especially the transition to caregiving, are likely to be experienced differently across racial groups (Angel & Angel, 1997).
Clearly, more research is needed in this area.
Another limitation of this study is the retention of original participants in the
HRS between the first and second data collections. Although 80% is considered an acceptable retention rate, there were important differences between participants who were retained and those who were not retained. As earlier noted, participants who were not retained were more likely to be older, have fewer years
260 / TURNER, KILLIAN AND CAIN of education, and have lower levels of household income than those who were retained. Unfortunately, those not retained are the participants who are likely to have the fewest coping resources and, therefore, to be at the greatest risk for experiencing depressive symptoms in midlife. Although these variables were identified and included in the regression models as control variables, there is no way to know how participants who dropped out of the study between data collection points would have responded if they had been retained.
Other factors have been found to be related to depressive symptoms in midlife women that were not included in these analyses. For example, Bromberger and
Matthews (1996) found that trait anxiety, pessimism, and chronic stress were related to depressive symptoms in their study of 460 midlife women. It may also be that experiencing bereavement becomes more common as people age and may contribute to increasing numbers of depressive symptoms (Edelstein, Kalish,
Drozdick, &McKee, 1999). Although bereavement as the result of losing a spouse may be reflected in these analyses, bereavement as a result of losing friends or other family members may not be reflected in our analyses. Because this study focused on the relationship between three specific midlife transitions and depressive symptoms, it did not examine other potential sources of symptoms of depression. Future studies should examine a wider range of factors that may be related to depressive symptoms among midlife women.
A final cautionary note about these findings is related to the single outcome variable that was used in this research: a measure of depressive symptoms.
This study suggests that transitions in midlife have some negative implications for mental health. However, other researchers have noted that the relationship between midlife transitions and mental health is multidimensional (Marks et al.,
2002). Therefore, although this study focused on depressive symptomology as a potential outcome of midlife transitions, these transitions should not be considered as exclusively negative. Many theorists and researchers (Elder, 1998;
Marks, 1996; Marks & Lambert, 1998; Marks et al., 2002; Moen et al., 1992) have suggested that the life transitions at midlife have both positive and negative consequences. Longitudinal research with these participants over longer periods of time will allow researchers to examine how people negotiate the transitions of midlife and make adaptations over time. Expanded longitudinal studies will provide even more knowledge about the long-term relationships between experiencing each of these transitions and mental health in midlife. It may be that some of the transitions, such as caregiving, have long-term implications for mental health.
On the other hand, over time women in midlife may adapt to transitions in marital status or health, suggesting that these transitions may have few implications for the long-term mental health of women. Clearly, more research is needed.
In sum, the results of this study add to the already large amount of research literature that confirms the most basic proposition of the lifecourse perspective— that human development continues across the entire lifespan, including midlife. It
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 261 also confirms that much of adult development is related to the social, time, and gender contexts of life. Moreover, transitions that individuals experience in midlife significantly contribute to at least one aspect of successful aging— mental health. As a result, it is important for students, professionals, and policymakers to understand life course transitions and how they are related to mental health during midlife in order to appropriately respond to the needs of this rapidly growing age group.
REFERENCES
Abel, E. K. (1991). Who cares for the elderly: Public policy and the experiences of adult daughters. Philadelphia: Temple University Press.
Aldwin, C., & Levenson, M. (2001). Stress, coping, and health at midlife: A developmental perspective. In M. Lachman (Ed.), Handbook of midlife development (pp. 188-216).
New York: Wiley.
Alford-Cooper, F. (1993). Women as family caregivers: An American social problem.
Journal of Women and Aging, 5, 43-57.
Allen, K., Blieszner, R., & Roberto, K. (2000). Families in middle and later years:
A review and critique of research in the 1990s. Journal of Marriage and the Family, 62,
911-926.
Angel, R. J., & Angel, J. L. (1997) Who will care for us? Aging and long-term care in a multicultural America. New York: New York University Press.
Arbuckle, N. W., & de Vries, B. (1995). The long-term effects of later life spousal and parental bereavement on personal functioning. The Gerontologist, 35, 637-647.
Aro, A., Nyberg, N., Absetz, P., Henriksson, M., & Loennqvist, J. (2001). Depressive symptoms in middle-aged women are more strongly associated with physical health and social support than with socioeconomic factors. Nordic Journal of Psychiatry, 55, 191-198.
Boss, P. (1988). Family stress management. Thousand Oaks, CA: Sage Publications.
Brandtstaedter, J., Wentura, D., & Greve, W. (1993). Adaptive resources of the aging self: Outlines of an emergent perspective. International Journal of Behavioral Development,
16, 323-349.
Brody, E. (1995). Prospects for family caregiving: Responses to change, continuity and diversity. In R. Kane & J. Penrod (Eds.), Family caregiving in an aging society
(pp. 15-28). Thousand Oaks, CA: Sage Publications.
Bromberger, J. T., & Matthews, K. A. (1996). A longitudinal study of the effects of pessimism, trait anxiety, and life stress on depressive symptoms in middle-aged women.
Psychology and Aging, 11, 207-213.
Burns, M., & Seligman, M. (1989). Explanatory style across the life span: Evidence for stability over 52 years. Journal of Personality and Social Psychology, 56, 471-477.
Cherlin, A. (1992). Marriage, divorce, remarriage. Cambridge, MA: Harvard
University Press.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Connidis, I. (2001). Family ties and aging. Thousand Oaks, CA: Sage.
Deeks, A., & McCabe, M. (2001). Menopausal stage and age and perception of body image. Psychology and Health, 16, 367-379.
262 / TURNER, KILLIAN AND CAIN
Desbiens, N., Mueller-Rizner, N., Viring, B., & Lynn, J. (2001). Stress in caregivers of hospitalized oldest-old patients. Journal of Gerontology: Series A: Biological Sciences and Medical Sciences, 56A, M231-M235.
DeVanzo, J., & Rahman, O. (1993). American families: Trends and correlates. Population
Index, 59, 350-386.
Edelstein, B., Kalish, K. D., Drozdick, L. W., & McKee, D. R. (1999). Assessment of depression and bereavement in older adults. In P. A. Lichtenberg (Ed.), Handbook of assessment in clinical gerontology (pp. 11-58). New York: Wiley.
Elder, G. (1998). The life course and human development. In W. Damon (Series Ed.)
& R. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (5th ed., pp. 939-991). New York: Wiley.
Erikson, E. H. (1963). Childhood and society. New York: Norton.
Faison, K., Faria, S., & Frank, D. (1999). Caregivers of chronically ill elderly: Perceived burden. Journal of Community Health Nursing, 4, 243-254.
Fields, J., & Casper, L. N. (2001). America’s families and living arrangements: Population characteristics. U.S. Census Bureau.
Gallagher-Thompson, D., & Powers, D. (1997). Primary stressors and depressive symptoms in caregivers of dementia patients. Aging and Mental Health, 1, 248-265.
Gallicchio, L., Siddiqi, N., Langenberg, P., & Baumgarten, M. (2002). Gender differences in burden and depression among informal caregivers of demented elders in the community. International Journal of Geriatric Psychiatry, 17, 154-163.
Gottlieb, B. H., Kelloway, E. K., & Fraboni, M. (1994). Aspects of eldercare that place employees at risk. The Gerontologist, 24, 815-821.
Gove, W., Style, C., & Hughes, M. (1990). The effect of marriage on the well-being of adults. Journal of Family Issues, 11, 4-35.
Grant, J., Bartolucci, T., & Giger, J. (2000). Sociodemographic, physical, and psychosocial characteristics of depressed and non-depressed family caregivers of stroke survivors.
Brain Injury, 14, 1089-1100.
Hooyman, N. R., & Gonyea, J. (1995). Feminist perspectives on family care. Thousand
Oaks, CA: Sage.
Kendler, K., Gardner, C., & Prescott, C. (2002). Toward a comprehensive developmental model for major depression in women. American Journal of Psychiatry, 159,
1133-1145.
Kohout, F. J., Berkman, L. F., Evans, D. A., & Cornoni-Huntley, J. (1993). Two shorter forms of the CES-D depression symptoms index. Journal of Aging and Health, 5, 179-193.
Kramer, B. (1993). Expanding the conceptualization of caregiver coping: The importance of relationship focused coping strategies. Family Relations, 42, 383-391.
Kramer, B., & Kipnis, S. (1995). Eldercare and work-role conflict: Toward an understanding of gender differences in caregiving burden. The Gerontologist, 35, 340-351.
Lachman, M., & James, J. (1997). Charting the course of midlife development:
An overview. In M. Lachman & J. James (Eds.), Multiple paths of midlife development
(pp. 1-17). Chicago: University of Chicago Press.
Lee, J., Walker M., & Shoup, R. (2001). Balancing elder care responsibilities and work:
The impact on emotional health. Journal of Business and Psychology, 16, 277-289.
Levinson, D. (1978). The seasons of a man’s life. New York: Ballantine.
Lieberman, M. (1996). Doors close, doors open: Widows, grieving and growing.
New York: Putman.
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 263
Lopata, H. (1996). Current widowhood: Myths and realities. Thousand Oaks, CA:
Sage.
Marks, N. (1996). Flying solo at midlife: Gender, marital status, and psychological well-being. Journal of Marriage and Family, 58, 917-933.
Marks, N. (1998). Does it hurt to care? Caregiving, work-family conflict, and midlife well-being. Journal of Marriage and Family, 60, 951-966.
Marks, N.,&Lambert, J. (1998). Marital status continuity and change among young and midlife adults. Journal of Family Issues, 19, 652-687.
Marks, N., Lambert, J., & Choi, H. (2002). Transitions to caregiving, gender, and psychological well-being: A prospective U.S. national study. Journal of Marriage and
Family, 64, 657-668.
Martin, C. (2000). More than the work. Journal of Family Issues, 21, 986-1006.
Martin-Matthews, A. (1999). Widowhood: Dominant renditions, changing demography and variable meaning. In S. M. Neysmith (Ed.), Critical issues for future social works practice with aging persons (pp. 27-44). New York: Columbia University Press.
McCubbin, M. (1995). The typology model of adjustment and adaptation: A family stress model. Guidance and Counseling, 10, 31-37.
McCubbin, H., & Patterson, J. (1983). The family stress process: The double ABC-X model of adjustment and adaptation. Marriage and Family Review, 6, 7-37.
Moen, P., Dempster-McClain, D., & Williams, R. (1992). Successful aging: A lifecourse perspective on women’s multiple roles and health. American Journal of Sociology,
97, 1612-1638.
Moen, P., & Wethington, E. (1999). Midlife development in a life course context. In
S. Willis & J. Reid (Eds.), Life in the middle: Psychological and social development in middle age (pp. 3-23). San Diego, CA: Academic Press.
Radloff, L. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.
RAND Center for the Study of Aging (2002, Feb.). RAND HRS data. Version A. Santa
Monica, CA: The Rand Corporation.
Reboussin, B., Rejeski, W., Martin, K., Callahan, K., Dunn, A., King, A., & Sallis, J. F.
(2000). Correlates of satisfaction with body function and body appearance in middle and older aged adults: The Activity Counseling Trial (ACT). Psychology and Health,
15, 239-255.
Rosenthal, C. J., Martin-Matthews, A. M., & Matthews, S. H. (1996). Caught in the middle? Occupancy in multiple roles and help to parents in a national probability sample of Canadian adults. Journal of Gerontology: Series B: Social Sciences, 51B,
S274-S283.
Saxton, L. (1993). The individual, marriage, and the family (8th ed.). Belmont, CA:
Wadsworth.
Schulz, R., O’Brien, A., Bookwala, J., & Fleissner, K. (1996). Psychiatric and physical morbidity effects in dementia caregiving: Prevalence, correlates, and causes. The
Gerontologist, 35, 771-791.
Skolnick, A. (1991). Embattled paradise. New York: Basic.
Spanier, G. B., & Thompson, L. (1983). Relief and distress after marital separation.
Journal of Divorce, 7, 31-49.
Spiro, A., III (2001). Health in midlife: Toward a life-span view. In M. Lachman (Ed.),
Handbook of midlife development (pp. 156-187). New York: Wiley.
264 / TURNER, KILLIAN AND CAIN
Staudinger, V., & Bluck, S. (2001). A view on midlife development from a life-span theory. In M. Lachman (Ed.), Handbook of midlife development (pp. 3-39). New York:
Wiley.
Stewart, A., & Ostrove, J. (1998). Women’s personality in middle age: Gender, history, and midcourse correction. American Psychologist, 53, 1185-1194.
Stone, B., Cafferata, G. L., & Sangl, J. (1987). Caregivers of the frail elderly: A national profile. The Gerontologist, 27, 616-626.
Strawbridge, W., Wallhagen, M., Shema, S., & Kaplan, G. (1997). New burdens or more of the same? Comparing grandparent, spouse, and adult-child caregivers. The
Gerontologist, 37, 505-510.
Treas, J. (1995). Older Americans in the 1990s and beyond. Population Bulletin,
50(2), 1-46.
Turvey, C. L., Wallace, R. B., & Herzog, R. (1999). A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. International Psychogeriatrics, 11, 139-148.
Uhlenberg, P., Cooney, T., & Boyd, R. (1990). Divorce for women after midlife.
Journal of Gerontology: Series B: Social Sciences, 45, S3-S11.
Umberson, D., Wortman, C., & Kessler, R. (1992). Widowhood and depression:
Explaining long-term gender differences in vulnerability. Journal of Health and Social
Behavior, 33, 10-24.
Wells, Y. D., & Kendig, H. L. (1997). Health and well-being of spouse caregivers and the widowed. The Gerontologist, 37, 666-674.
Whitbourne, S. (2001). The physical aging process in midlife: Interactions with psychological and sociocultural factors. In M. Lachman (Ed.), Handbook of midlife development
(pp. 109-155). New York: Wiley.
Whitlatch, C., & Feinberg, L. (1997). Depression and health in family caregivers.
Journal of Aging and Health, 9, 222-244.
Yee, J., & Schulz, R. (2000). Gender differences in psychiatric morbidity among family caregivers: A review and analysis. The Gerontologist, 40, 147-164.
Direct reprint requests to:
M. Jean Turner, Ph.D.
118 HOEC, University of Arkansas
Fayetteville, AR 72701 e-mail: jturner@uark.edu
MIDLIFE WOMEN AND DEPRESSIVE SYMPTOMS / 265

Copyright of International Journal of Aging & Human Development is the property of Baywood Publishing
Company, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.