| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
a Department of Sociology, Northeastern University, Boston, Massachusetts
b Epidemiology Program, Jean Mayer U.S. Department of Agriculture Human Nutrition Research Center On Aging, Tufts University, Boston, Massachusetts
Luis M. Falcón, Northeastern University, Department of Sociology, 523 Holmes Hall, Boston, MA 02115 E-mail: Falcon{at}neu.edu.
 |
Abstract |
|---|
 TOP Abstract MethodResultsDiscussionReferences |
|---|
Methods. Participants included 715 Hispanic elders (429 Puerto Rican, 128 Dominican, and 149 other Hispanic) in Massachusetts and 238 NHW elders living in the same neighborhoods. Depressive symptomatology was measured with the Center for Epidemiologic Studies Depression Scale (CES-D). The associations between socioeconomic status, household arrangements, acculturation, health problems, and depression score were examined with multiple linear and logistic regression.
Results. Puerto Rican and Dominican elders had significantly greater prevalences of high depression scores when compared with neighborhood NHWs. After controlling for various social, demographic, and health characteristics, the scores of Puerto Ricans remained significantly greater than those of the NHWs, but not "Other Hispanic" groups. Higher CES-D scores were significantly associated with being female, living alone, and having a higher number of health problems, but not with income. Effects of acculturation were only found among Dominican elders.
Conclusions. Puerto Rican elders experience high rates of depressive symptomatology that are associated with, but not fully explained by, high numbers of chronic health conditions. Further investigation is needed to both understand and treat this prevalent problem.
THERE has been increasing interest in research on both the biological and social factors associated with depression among older adults (Black, Goodwin, and Markides 1998
; Hays et al. 1998; Okun and Keith 1998). With aging, there is an increase in the incidence of disease and in changes in household arrangements due to the departure of children or the loss of a spouse. For elders, increased frailty leads to functional disability with the potential for depression (Parmelee, Lawton, and Katz 1998). The loss of social relationships, particularly those rooted in marriage or close social ties, is also an important threat to mental well being (Guarnaccia, Angel, and Lowe Worobey 1991; Hays et al. 1998; Okun and Keith 1998).
The literature documenting differences in depressive symptomatology among and across Hispanic subgroups has grown noticeably over the last few years (Guarnaccia et al. 1991; Kemp, Staples, and Lopez Aqueres 1987; Krause and Goldenhar 1992; Lopez Aqueres, Kemp, Plopper, Staples, and Brummel Smith 1984; Moscicki, Locke, Rae, and Boyd 1989; Narrow, Rae, Moscicki, Locke, and Regier 1990; Vera et al. 1991). For many reasons, Hispanics in the United States may be more likely to suffer from mental health problems than are other groups. A large number of Hispanics are immigrants attempting to function in a society quite different linguistically and culturally than the societies they come from. This immigrant character presents Hispanics with pressures toward acculturation and adaptation that serve as sources of stress (Black, Markides, and Miller 1988; Canabal and Quiles 1995; Golding and Burnam 1990; Rogler, Cortes, and Malgady 1991). Hispanics are also more likely to be poor and to have more health problems, situations which have recently been shown to strongly associate with depression (Black et al. 1998).
Existing studies have focused primarily on the Mexican-American population and on those aged 20 to 64 years. For various reasons, elderly Hispanics—Puerto Ricans in particular—have received little attention from researchers. Still, the few studies available on Puerto Ricans suggest higher levels of symptomatology for Puerto Ricans in general (Canabal and Quiles 1995) and for older Puerto Ricans in particular (Vera et al. 1991), than for other Hispanic subgroups. The Hispanic Health and Nutrition Examination Survey (HHANES), which included Mexican Americans, Cubans, and Puerto Ricans, made a major contribution to describing patterns of health among Hispanics. Data from HHANES showed that Hispanics suffered from high prevalences of depression and that Puerto Ricans had the highest prevalence and Cubans the lowest, with Mexican Americans between these two groups (Moscicki et al. 1989; Narrow et al. 1990; Vera et al. 1991). Krause and Goldenhar 1992, using data from the 1988 National Survey of Hispanic Elderly People, also found that older Puerto Ricans experienced higher levels of depressive symptoms, social isolation, and financial strain than other groups.
Depression in old age is a major health problem. Because Hispanic elders are a growing segment of the elderly population, the sporadic evidence suggesting higher incidence of depression is of concern. Puerto Ricans, together with Dominicans, are among the most disadvantaged Hispanic groups in the United States. The elders among these groups present an even more disadvantaged position than the population at large (Falcon, Tucker, and Bermudez 1997). Poverty, lack of participation in community programs, and extremely poor health are all characteristic of elderly Puerto Rican and Dominican Hispanics (Falcon et al. 1997). All these factors place Hispanic elders at risk of stress and conditions associated with poor mental health. Few studies have examined differences in depression prevalence across Hispanic elder subgroups. We were, for example, unable to locate any publications that address the mental health condition of Dominican elders, a growing group in the United States. The state of Massachusetts has the second largest Dominican settlement in the United States after the New York/New Jersey area. The Massachusetts Hispanic Elderly Study (MAHES) is the first community survey to collect extensive data on Dominican elders.
Hispanics in Massachusetts are a recently arrived population. There were about 300,000 Hispanics in the state in 1990, but as recently as 1970, there were only 65,000 (Melendez 1993). In the early 1990s, Massachusetts ranked number one in Hispanic population growth among the top 10 states with the largest Hispanic populations in the country (Melendez 1993). This fast paced growth, along with an extremely high rate of Hispanic poverty, makes the Hispanic population in Massachusetts distinctive. Over the last three decades, Hispanics have engaged in a process of dispersion from older settlements like New York City, and Massachusetts has been a preferred destination for many of them. Hispanic poverty in Massachusetts reached 38% in 1989—twice the national average for Hispanics (Melendez 1993). Further, because of the recency of this settlement, most of these Hispanics have faced the disruptive experiences associated with migration and integration into a linguistically and culturally different society. It is important to document the prevalence of depression among this group and to identify factors associated with differentials in depressive symptoms across groups.
The aim of this paper is to provide information on the prevalence of depressive symptomatology in a representative sample of Hispanic elders in Massachusetts, where the majority are Puerto Ricans and Dominicans. In addition, we examine how socioeconomic status, living arrangements, and health conditions relate to depressive symptoms. In the analyses, we examine how Hispanic groups differ from non-Hispanic whites (NHWs) that live in the same neighborhoods, and how Hispanic subgroups differ from each other.
|
Method |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
). Once sample clusters were allocated to counties, samples of 25 subject blocks were selected, with probability of selection of a block proportionate to the number of eligible Hispanics in the block. Blocks drawn into the sample were enumerated and a respondent chosen randomly. A total of 1,033 face-to-face interviews were completed, which include 475 Puerto Ricans, 146 Dominicans, 160 Other Hispanics, and 252 neighborhood NHWs. Respondents were asked to complete detailed questionnaires, including modules on social and economic background, migration, health history, physical functioning, depression, cognitive function, dietary intake, and prescription drug use. In addition, body measurements of weight, height, blood pressure, and blood samples were also collected. Bilingual interviewers were trained in the administration of the questionnaires. Almost all of the Hispanic respondents completed these interviews in Spanish.
Dependent Variable
To measure symptoms of depression, we used the Center for Epidemiologic Studies Depression Scale (CES-D). This scale is commonly used in survey studies and has been shown to have good consistency and validity (Moscicki et al. 1989; Radloff 1977, Radloff 1986). The scale includes 20 items, of which 16 measure symptoms in the negative direction and 4 in the positive direction. The questions in the scale refer to symptoms during the week prior to the interview. Once the positive scores are reversed in value for consistency (0–3), the sum of the scale values for the depression score can range from 0 to 60. A cutoff of 16 points or more is commonly used as an indicator of depression caseness. That is, respondents with a score of 16 points or higher are seen as exhibiting symptoms of clinical depression.
There is substantial debate in the literature on the appropriateness of the CES-D for use with minority populations (Cho et al. 1993; Miller, Markides, and Black 1997). Some have argued that the expression of depression may differ across cultures, implying that the CES-D may measure different constructs for different ethnic groups (Miller et al. 1997; Rogler et al. 1991). In contrast, recent work by Mackinnon, McCallum, Andrews, and Anderson 1998 in five Asian countries suggested that the factor structure of the CES-D in those countries was comparable to those based on populations in western cultures. Further, they argued that their results support the use of single-scale measures of the CES-D rather than the use of subscales. A study by Mahard 1988 with a sample of elderly Puerto Ricans—half of whom had been diagnosed as being clinically depressed—examined the validity of the CES-D to diagnose clinical depression. She concluded that the CES-D scale showed high reliability and was able to discriminate between those who were diagnosed with depression and those who were not (Mahard 1988). Work by Cho et al. 1993 examined the adequacy of the CES-D as a screening instrument for major depresssion with a sample of Cuban and Puerto Ricans from the HHANES. They concluded that the CES-D may be useful in community-based surveys as a first-stage screening device (Cho et al. 1993). Because we are applying this to several groups, we examined the reliability of the scale by subgroup. The CES-D scale obtained from the MAHES data behaved in a highly reliable way as measured by Cronbach's alpha. For the sample as a whole, the alpha was 0.90. The alpha values did not differ by subgroup, ranging from 0.88 for the NHWs and Other Hispanics to 0.90 for Puerto Ricans and 0.91 for Dominicans. Exploratory analysis of the factor structure of the CES-D items showed that, for the most part, the same items loaded on the first and strongest factor for all groups; initial eigenvalues ranged from 6.5 for NHWs to 7.7 for Dominicans, with the next highest factor eigenvalue ranging from 1.4 for Other Hispanics to 1.7 for NHWs. After varimax rotation with a three-factor solution, the first factor included loadings greater than or equal to a threshold of 
for at least three of the four ethnic subgroups, for 13 of the 20 items. The total number of items loading greater than or equal to 
on Factor 1 by subgroup ranged from 13 for NHWs to 16 for Dominicans. Based on these results, and on suggestions from prior studies, we used the CES-D score as a single summated scale (Mackinnon et al. 1998), first as a continuous variable, but also to determine caseness of depressive symptomatology, using the standard cutoff point of 16.
Key Independent Variables
Health status.
As a measure of health status, we used a count of the number of existing chronic health conditions. Respondents were asked whether a doctor has ever told them that they had specific illnesses or conditions. The list included 20 conditions commonly affecting the elderly; in addition, it allowed the opportunity to enumerate additional conditions that were not included in the list. Respondents were also asked if the condition was a current health problem that was bothering them at the time. An additive scale was created that indicated the number of currently problematic existing health conditions listed by the respondents.
Socioeconomic status.
A measure of total household income, including the income from all individuals residing within the household, was obtained through a series of questions and is included as a measure of financial resources. Household income was calculated from questions about income from employment, public assistance, retirement pensions, and other sources, for all members of the household. About 13% of the respondents did not provide information from which to calculate their annual household income. In order to maximize the sample size, respondents who had missing information on income questions were assigned an income value using the expectation maximization procedure in the SPSS Missing Value Analysis Program (Hill 1997). Variables used as predictors included age, gender, ethnicity, education of participant and household head, employment status, job category for current or former occupation, number of income sources, relationship to household head, number of persons in the household, participation in needs-based programs, and home ownership. To validate the method, we set 15% of valid cases to missing and imputed them as well. The Pearson correlation coefficient between actual and imputed household income was 0.79 (p < .0001). A dummy variable, indicating whether the household income value was estimated, was included in preliminary analyses to determine any systematic associations between those with missing information and the outcomes, but no such association was found.
Household arrangements.
Living arrangements are important to the well-being of elderly adults. Many elders prefer to remain living by themselves as evidence of their ability to function independently. For others, choosing to live with others or to live alone is not an option, due to their health conditions or to the absence of nearby relatives. The type of arrangement elders live in affects the availability of sources of both formal and informal support. To examine how living arrangements relate to depressive symptoms, we used a measure that combines the marital status of the elder with the presence of other persons within their household. The resulting variable has three categories: married and lives with spouse, unmarried and lives with others, and unmarried and lives alone. In the analysis, we compared living alone to the other two categories: married and unmarried but living with others.
Use of English language.
For immigrants, English language use is an important indicator of acculturation. The ability to use the English language is affected by both the level of education and the amount of exposure that individuals may have had to mainstream culture in the United States. As such, it has been used in many studies as a proxy for the level of acculturation to the society. In our analysis, we used English usage as an indicator of the extent to which respondents are in contact with the overall English-speaking environment. The measure is based on a modified version of the Cuellar Acculturation Scale (Cuellar, Harris, and Jasso 1980). The scale included seven questions which asked respondents to indicate their use of English, Spanish, or both languages to watch TV, read newspapers or books, speak in the neighborhood, speak at work, listen to the radio, talk with friends, and talk at home. The resulting scale was coded to indicate the number of activities in which respondents used either English only or both English and Spanish. The scale ranges from 0 to 7 activities.
Percentage of life in the United States.
Another indicator of exposure to U.S. society is the proportion of their lives respondents have spent on the U.S. mainland. Recent immigrants are considered to have little awareness of American society and to be unacculturated, whereas more time spent in the United States is generally assumed to be correlated with more acculturation to U.S. society. Because almost all of the Hispanic respondents in the MAHES were foreign born, the commonly used measure of nativity is a constant in our data. In our analysis, we measure the time spent on the U.S. mainland as a ratio of the time spent in the U.S. to the age of the respondent. The resulting variable indicates the percentage of their lives spent in the U.S.
Control variables.
Age, education, and gender may be associated with both the key independent variables and the depression scores and were therefore included as control variables. Age is a continuous variable with a range of 60 to 98 years. Education is also continuous, with a range of 0 to 17 years. Gender was coded as a dichotomous variable, with men coded 1 and women, as the reference category, coded 0.
Statistical analysis.
Descriptive and multivariate methods were used to examine depressive symptoms, with the CES-D score as a dependent measure. The presentation of results includes a description of the mean scores for depression and prevalence of depression caseness, using the cutoff point described above, as well as descriptions of socioeconomic and demographic variables for the four ethnic groups: Puerto Ricans, Dominicans, Other Hispanics, and NHW controls. Prevalences are presented by the various social and demographic characteristics of the groups. Descriptive comparisons were tested with the Wald test for means and the PR test for proportions contained in the Stata software package (StataCorp 1999).
Multivariate models were used to examine associations with social, economic, and health characteristics and to determine whether the ethnic differences could be explained by differences in these factors. Linear models were estimated for the regression of the CES-D score on measures of income, living arrangements, and health variables, using the survey regression procedure in STATA. Similar models were run with the survey logistic regression, with the caseness measure as a dependent variable. We built regression models according to a hierarchical approach: Variable sets were added to the equation in stages. This allows for the observation of the contribution of these variables to explaining differences between the groups. We first controlled for demographic characteristics generally found to be associated with depression. We then added, sequentially, variables reflecting the social and economic context of the respondent and the existence of chronic health problems.
|
Results |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
|
Puerto Ricans reported significantly higher average numbers of chronic health conditions compared with NHWs (2.3 vs. 1.9). We saw a similar pattern for the depression indicators. Average CES-D scores for the Puerto Ricans (16.2) and Dominicans (13.2) were significantly higher than were those for the NHWs (10.00). Puerto Ricans also had the highest caseness prevalence at (CES-D 
16) at 44%, compared with 32% of Dominicans, 30% of Other Hispanics, and 22% of NHWs.
These descriptive results suggest large differences in depressive symptomatology across Hispanic subgroup and in comparison with neighboring NHWs. Furthermore, the size of both the average score and the caseness prevalence for Puerto Ricans suggest that this population may be experiencing serious mental health problems. It is important to clarify some of the underlying factors associated with these differences.
In Table 2 , the mean CES-D scores for the four groups are presented by gender and living arrangements. In general, men tended to have lower CES-D scores than did women. The gender difference in average depression scores was most accentuated among Dominicans, ranging from 6.1 for men to 16.6 for women. Elders who lived with a spouse or with others tended to have lower depression scores than did those who lived alone. Noticeably, at almost every level of the independent variables, Puerto Ricans reported a much higher average CES-D score than did the other groups.
|
|
|
After the addition of indicators for living arrangements, the effect of Puerto Rican ethnicity became larger, suggesting that Puerto Ricans might have an even higher average depression score if not for their higher proportion of marriage versus living alone. Being married was negatively associated with depression scores, but there was no significant difference in scores between those living with others and those living alone.
The final model included the indicator for the number of chronic health conditions, with minor effects on the CES-D gap between Puerto Ricans and NHWs. Puerto Rican ethnicity remained strongly associated, with a depression score 3.6 points higher than that for NHWs. Chronic health conditions were strongly and positively associated with CES-D score, with an increment of 2.2 points for every additional condition. When the health indicator was added, the variation in CES-D scores accounted for by the model almost doubled—from 12% to 23%—underscoring the importance of health on the depression scores of these subjects.
We conducted regression analyses with a set of models similar to those shown in Table 4 , but restricted only to the Hispanic groups (results not shown). In these models, we introduced variables indicating the use of English and the percentage of participants' lives in the U.S. Once the controls for social and demographic characteristics were included, English language usage and percentage of life in the U.S. did not account for any of the differences between the Hispanic groups. Ethnic differences in CES-D also disappeared once a control for the number of health problems was included. In addition, we conducted logistic regression analysis to determine the effects of the variables presented above on caseness, using the standard cutoff point of 16 to define caseness for all groups. The results were consistent with those found when examining continuous CES-D scores. After adjustment for the social, demographic, and health variables, Puerto Ricans were twice as likely to be at caseness level relative to NHWs (
; 95% confidence interval [CI] = 1.3–3.4; 
, data not shown). After adjustment for other variables, there were no significant differences in the likelihood of caseness between the Dominicans or the Other Hispanics when compared with the NHWs. As with the continuous score, other important factors in predicting caseness were gender (male vs. female, 
), income (per $1,000, 
), living with a partner (vs. living alone, 
), and the number of health problems (for each additional problem, 
).
In Table 5 , we present results from regression equations that examined the associations between our key independent variables and CES-D scores separately for the individual subgroups. The results suggest important differences across the groups in the factors that relate to level of depression scores. First, age, education, and household income were not significantly associated with depression scores within any of the groups when examined separately. Being male was significantly associated with lower depression scores among Dominicans and Puerto Ricans, but not among Other Hispanics or NHWs. Being married appeared to be most protective for the Other Hispanics, and living with others versus living alone was associated with lower scores for the Puerto Ricans but not for the other groups.
|
|
Discussion |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
; Tran and Dhooper 1997) and from the HHANES (Canabal and Quiles 1995). Although the difference in CES-D score between the Dominicans and the NHWs disappeared after control for socioeconomic variables, the difference between Puerto Ricans and NHWs remained strong and significant after adjustment for all other variables, including the presence of chronic health conditions.
We saw CES-D scores and prevalences for Puerto Rican elders that were much higher than those previously reported for Puerto Ricans or for other Hispanic elders in previous studies. Black et. al. (1998) in an analysis of Mexican-Americans in the Established Populations for the Epidemiologic Studies of the Elderly (EPESE), reported average CES-D scores of 12.2 and 8.3 for women and men, respectively, compared with our scores for Puerto Rican men and women of 18.5 and 13. Using HHANES data, Potter, Rogler, and Moscicki 1995 found an average CES-D score of 12.3 for Puerto Ricans aged 20 and older, whereas Canabal and Quiles 1995 reported an average CES-D score of 13.3 for the same age group. Similarly, the prevalence of caseness we saw for Puerto Ricans (44%) is much higher than that reported for Mexican American elders in the EPESE (26%; Black et al. 1998) or for the younger sample of Puerto Ricans in the HHANES (28%; Potter et al. 1995).
The high average CES-D score for Puerto Ricans (16.2) in the MAHES sample is of concern. Previous studies have found high average CES-D scores among Puerto Ricans (Canabal and Quiles 1995; Cho et al. 1993; Magni, Rossi, Rigatti Luchini, and Merskey 1992; Mahard 1988; Potter et al. 1995) but usually in the range of 13 to 14 points. A study by Vera et al. 1991 comparing depressive symptoms among Puerto Ricans on the island and those on the mainland showed a caseness rate of 35% among Puerto Ricans aged 55–64 in Puerto Rico and of 29% for Puerto Ricans aged 55–64 on the mainland. As we showed in the analysis, Puerto Ricans exhibit many characteristics generally associated with depressive symptomatology. They are more likely to have low education, lower family income, and to suffer from a higher number of health problems than are other groups. Puerto Rican elders have also been found to experience more social isolation (Krause and Goldenhar 1992).
Almost all of the Hispanic interviews in the MAHES were conducted in Spanish. Studies that have compared the effects of the language of interview on Hispanics' assessment of their own health and depressive symptoms have found that those interviewed in Spanish were more likely to rate their health as poor and more likely to have higher CES-D scores (Angel and Guarnaccia 1989; Vera et al. 1991). Still, we do not find such high CES-D scores among the Dominicans or the Other Hispanics in our sample. What makes Puerto Ricans stand apart from the other groups is their high rates of chronic conditions. Although we use self-reported health problems in these analyses, more objective data were collected on self-reported diabetes, which we then corroborated with blood glucose and glycosylated hemoglobin measures, showed high concordance. Similarly, a study by Jette, Crawford, and Tennstedt 1996 based on a sample of Puerto Rican elders in western Massachusetts found substantial agreement between self-reported disability limitations and observed functional tasks.
HHANES analyses (Narrow et al. 1990; Vera et al. 1991) identified female gender and low education as factors associated with depressive symptomatology. Our findings are consistent with these for gender differences; women had higher depression scores within all subgroups. Puerto Rican women were at greater risk, followed by Dominican women. The HHANES studies found that, among Puerto Ricans and Cubans, depression was greatest among those with disrupted marital unions; for Mexican Americans it was greatest among those never married (Moscicki et al. 1989). We also found that living alone was associated with greater risk but that being married was most protective for Other Hispanics, whereas unmarried Puerto Ricans living with others were at lowest risk of depression compared with those living alone. These differences across subgroup in the effects of marital status on depression deserve further attention.
Other studies (Guarnaccia et al. 1991) have suggested that socioeconomic status should be an important determinant of depression symptoms among Hispanics. Using HHANES data, Moscicki et al. 1989 work on Mexican-Americans found strong negative associations between income level and CES-D score. HHANES analyses found that annual household income of less than $5,000 was significantly associated with depressive symptomatology for Puerto Ricans aged 20 and older (Moscicki et al. 1989). Although income level and depressive symptomatology were significantly associated in bivariate correlations, we found that income was no longer associated with depression scores within Hispanic groups after adjustment for other variables. This may be due partially to the limited income range in this group of Hispanics and to the association between these indicators and others in the models. The Massachusetts Hispanic population is generally considered to have higher rates of poverty than Hispanics in other parts of the country (Falcon 1993; Falcon et al. 1997), and this is likely to contribute to the high prevalence of a variety of poor health indicators, including depression, in this population. It is likely that the greater number of chronic conditions seen among Puerto Ricans than in other groups may result, in part, from years of poverty, unemployment, or employment in jobs with poor health protection.
HHANES studies found that Mexican-Americans with depressive symptomatology were more likely to be foreign born and to have a Mexican cultural orientation, suggesting that immigration may be a relevant factor in explaining the greater prevalence of depression among Hispanics (Moscicki et al. 1989). Although our results were consistent with this finding for the Dominicans, we did not find this association among the Puerto Ricans or Other Hispanics in our sample, nor did acculturation measures explain differences in depression scores across Hispanic subgroups. On average, the Dominican elders had arrived in the U.S. at a later age and were less likely to use English than were other groups. As such, a higher proportion of them, compared with the other Hispanics in our population, may have been experiencing adjustment problems. The lack of effects of acculturation on depression for Puerto Ricans is also consistent with findings by Canabal and Quiles 1995 for Puerto Ricans aged 20 years and older in the HHANES.
The factor most consistently associated with depressive symptomatology in this group of Massachusetts elders was the presence of chronic health conditions. This is consistent with other findings of the association between health and depression, including a recent study by Black et al. 1998, which found significant associations between chronic health conditions, functional disability, and depression among Mexican-American Elders in the Hispanic EPESE. In addition to higher prevalence of depression, the Puerto Ricans in this sample had more chronic health conditions than the other ethnic groups, consistent with evidence in the literature of a larger number of health problems among Puerto Ricans (Angel and Guarnaccia 1989). The inclusion of chronic health conditions in the regression models reduced the differences across groups, reinforcing its strong contribution to differences in depressive symptomatology. Among Hispanics, the inclusion of this variable diminished the differences between the Puerto Ricans and the other Hispanic groups to a level at which they were no longer significant. However, the inclusion of this variable did not eliminate the large difference in depression scores between the Puerto Ricans and the NHWs. Further investigation is therefore indicated to identify additional factors that may explain these differences in depressive symptomatology across ethnic groups. The extremely high prevalence of depression among Puerto Ricans makes this a problem of serious magnitude and one deserving of much more attention—both to understanding the causal pathways and to approaching methods of treatment.
|
Acknowledgments |
|---|
Received for publication January 20, 1999. Accepted for publication October 15, 1999.
|
References |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. C. Brown, C. A. Mason, A. R. Spokane, M. C. Cruza-Guet, B. Lopez, and J. Szapocznik The Relationship of Neighborhood Climate to Perceived Social Support and Mental Health in Older Hispanic Immigrants in Miami, Florida J Aging Health, June 1, 2009; 21(3): 431 - 459. [Abstract] [PDF]
|
|
|
|
 |
|
 T. Perrino, C. A. Mason, S. C. Brown, A. Spokane, and J. Szapocznik Longitudinal Relationships Between Cognitive Functioning and Depressive Symptoms Among Hispanic Older Adults J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2008; 63(5): P309 - P317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Warnecke, A. Oh, N. Breen, S. Gehlert, E. Paskett, K. L. Tucker, N. Lurie, T. Rebbeck, J. Goodwin, J. Flack, et al. Approaching Health Disparities From a Population Perspective: The National Institutes of Health Centers for Population Health and Health Disparities Am J Public Health, September 1, 2008; 98(9): 1608 - 1615. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Merete, L. M. Falcon, and K. L. Tucker Vitamin B6 Is Associated with Depressive Symptomatology in Massachusetts Elders J. Am. Coll. Nutr., June 1, 2008; 27(3): 421 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Clark, C. DeCarli, D. Mungas, H. I. Chui, R. Higdon, J. Nunez, H. Fernandez, M. Negron, J. Manly, S. Ferris, et al. Earlier Onset of Alzheimer Disease Symptoms in Latino Individuals Compared With Anglo Individuals Arch Neurol, May 1, 2005; 62(5): 774 - 778. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Robison, C. Gruman, S. Gaztambide, and K. Blank Screening for Depression in Middle-Aged and Older Puerto Rican Primary Care Patients J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2002; 57(5): M308 - 314. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
