FOR A-PLUS WRITER ONLY

For the Spanish version, a Spanish version of the S-GDS was used. Following successful translation, back translation, and pilot testing, this version has been used successfully in a statewide program in Florida for several years (Schonfeld et al., 2009). The vignette and other parts of the survey were translated into Spanish using back-translation methods to correct errors. The final translation, developed by a native Spanish-speaking mental health researcher, was reviewed by a second native Spanish-speaking research staff member. An interviewer fluent in Spanish was available to assist Spanish-speaking participants. A total of 64 (27.71%) participants completed the survey in Spanish, all but two of whom self-identified as Hispanic ethnicity (one White, one Other).

Participants completed the written survey in a group setting. For 12 (4.92%) participants unable to read or write, an interviewer administered the survey orally in a private area.

Data Analysis

Coding of outcome variables. The open-ended responses for the outcomes were coded separately by two of the authors (AG, SM), who met to review and resolve any discrepancies. Final categories and coding were decided by all four authors. All authors were unaware of study condition during the coding process. Four outcome variables were examined in relation to vignette type and covariates:

1. Identified depression (from open-ended question)—coded “yes” if the person wrote “depressed” or “depression,” coded “no” for all other responses;

2. Recognized depression (from checklist question)—coded “yes” if the person selected depression only, coded “no” for all others (including a small number of participants who selected multiple responses; they were judged to be unsure of the person’s problem);

3. Identified need for professional help (from open-ended question)— coded “yes” if the person specified that the person needed help from a doctor, other health or social service professional, or mental health professional, coded “no” for all others; and

4. Recognized need for professional help (from checklist question)— coded “yes” if the person selected “talk to doctor” or “see counselor”, coded “no” for all other responses (including a small number of participants who selected multiple responses; they were judged to be unsure of the best solution for the person’s problem).

Analyses. Descriptive statistics were calculated for all variables. Demographics were examined in bivariate analysis (chi-square for categorical variables, t test for continuous variables) with vignette condition to ensure the validity of the randomization process; there were no significant differences (p > .05), supporting the validity of the randomization process. For the first two hypotheses, chisquare was used to examine differences on the outcomes (identified depression [yes/no], recognized depression [yes/no], identified need for professional help [yes/no], recognized need for professional help [yes/no]) by vignette type. The open-ended questions were the primary outcomes of focus, because they more closely represent older adults’ spontaneous, natural impressions of the person’s problem and recommended solutions, as opposed to a checklist. The responses from the checklists were examined to explore participants’ ability to recognize depression and the need for professional help when presented to them.

To examine vignette type and other covariates together as covariates of participants’ knowledge (third and fourth hypotheses), logistic regression analyses were conducted for the two primary, open-ended outcomes: identified depression and identified need for professional help. Demographic variables (sex, race/ethnicity—White or minority, education— at least high school diploma or not) and vignette condition (with sadness, without sadness) were included in both logistic regression analyses. Race/ethnicity was collapsed into White or minority based on the small number of participants from minority groups other than Hispanic ethnicity (see Table 1); given the sample size, power was deemed insufficient to examine other racial or ethnic differences. Other covariates (ERA-12, S-GDS, any prior mental health experience—yes or no) were included only if they were significantly related to the outcome variable in bivariate logistic regression at p ≤ .05. Residuals were examined for multivariate outliers; no outliers were identified.

Results

Sample Description

A description of the sample is provided in Table 1. The average age was 77.76, most participants were female (75.00%), slightly more than half of the participants were White (51.64%), approximately a third were Hispanic (33.61%), and 57.79% had a high school diploma or higher. Approximately a quarter identified ever having a mental health problem (24.18%), but a third had discussed such a problem with their doctor (34.84%) or received antidepressant or anxiety medication (32.79%). Most participants did not have significant depressive symptoms according to the S-GDS (76.23%).

Table 1. Sample Description (N = 244)a

Note: S-GDS = Short Geriatric Depression Scale; ERA-12 = Expectations Regarding Aging-12–Mental Health subscale.

a. Missing data: sex (7), age (14), race/ethnicity (13), education (10), ever experienced mental health problem (3), ever talked to physician about mental health problem (8), ever talked to any professional about mental health problem (13), ever received antidepressant or antianxiety medication (10), mental health problem in family member (9), any mental health experience (6), S-GDS (8), and ERA-12 (8).

Depression Outcomes by Vignette Condition

Table 2 provides the coding from responses to the two open-ended questions. Overall, 40.98% correctly identified the person as depressed, 48.33% who received the sad vignette and 33.87% who received the anhedonia vignette,

Table 2. Participants’ Identification of the Problem and Recommended Solution From Open-Ended Questions

Sadness Anhedonia

(n = 120) (n = 124) Total (N = 244)

a. Percentage within vignette condition. Numbers exceed 100% because a small number of participants wrote multiple answers (e.g., “depressed and lonely”).

b. For don’t know/vague category, participants wrote that they did not know what the problem was or how to address it, or they wrote vague responses, such as “the person has a problem” or “the person needs help.”

c2(1) = 5.27, p < .05. Only 5.74% thought the person was experiencing normal aging, although this was higher for those in the anhedonia condition (8.87%) compared to those in the sad condition (2.50%). A total of 38.52% thought the person should seek some kind of professional help, including 40.00% for the sad condition and 37.10% for the anhedonia condition, c2(1) = 0.22, ns.

Table 3 provides the responses to the checklists. Here, 42.80% accurately checked depression only, 47.90% for the sad condition, and 37.90% for the anhedonia condition, c2(1) = 2.48, ns. Over half (52.48%) thought the person should see a doctor, and a quarter (25.21%) thought the person should see a counselor. In all, 62.30% checked either doctor or mental health professional,

61.67% who received the sad vignette and 62.90% who received the anhedonia vignette, c2(1) = 0.04, ns. Comparing descriptively across the open-ended responses in Table 2 and checked responses in Table 3, participants were more likely to check “see a doctor” or “see a counselor” than they were to spontaneously mention these solutions.

Using the open-ended responses, participants who identified the person in the story as depressed were much more likely to identify a need for professional help (n = 62, 62.00%) than those who did not identify depression (n = 32, 22.22%), c2(1) = 39.43, p < .001. A similar pattern was found for the checked responses; participants who checked depression only were more likely to check need for professional help (doctor or counselor; n = 80, 76.92%) than those who did not check depression (n = 72, 51.80%), c2(1) = 16.03, p < .001.

Covariates of Identification of Depression and Need for Professional Help

Due to missing data for some covariates, the sample size was significantly reduced for the multivariate logistic regression analyses. In the analysis for identification of depression (n = 207), there were no differences comparing retained and excluded participants in sex, age, education, mental health experience, S-GDS score, or vignette condition. Excluded participants were more likely to be minority, to not identify depression or need for professional help, and had lower ERA-12 scores. Similarly, in analysis for identification of need for professional help (n = 214), excluded participants were more likely to be minority and to not identify depression or need for professional help.

Table 4 presents the results of the bivariate and multivariate logistic regression for identification of depression. In bivariate analyses, participants were more likely to identify the person as depressed if they were White, had at least a high school diploma, had prior mental health experience, had more positive expectations regarding aging, and read the sad vignette. In the multivariate model, mental health experience, expectations regarding aging, and vignette condition remained statistically significant, as detailed in Table 4.

Table 5 presents the results for identification of need for professional help (doctor or counselor). In the bivariate analyses, participants were more likely to identify need for professional help if they were White, had a high school diploma, and identified the person in the vignette as depressed. The only significant covariate in the multivariate model was identification of depression; those who identified the person as depressed had 5.69 greater odds (95% CI = 3.03-10.71, p < .001) of identifying need for professional help.

Table 3. Participants’ Recognition of the Problem and Recommended Solution From Checklists

Sadness Anhedonia

(n = 120) (n = 124) Total (N = 244)

a. Percentage within vignette condition. Numbers exceed 100% because a small number of participants checked multiple answers.

Discussion

Using vignettes in an experimental design, our study demonstrated that symptom presentation had a significant impact on older adults’ identification of depression. They were more likely to identify depression in the presence of dysphoria or sadness compared to anhedonia without sadness, supporting our first hypothesis. Contrary to our second hypothesis, symptom presentation did not affect identification of the need to seek professional assistance from a doctor or mental health professional. Most participants did not know that the person was depressed or that the person should seek professional assistance. Overall, less than half (40.98%) of participants in the current study identified the person as depressed, 48.33% who read the sad vignette and 33.87% who read the vignette without sadness. This rate of depression identification is comparable to another study of older adults using the same vignette of depression with sadness (39.0%; Fisher &

Table 4. Logistic Regression of Participants’ Identification of Depression in the Vignette Character (n = 207)a

Note: The OR column displays results from bivariate analysis, and the adjusted OR column displays results from the final multivariate model including all covariates.

ERA-12 = Expectations Regarding Aging-12–Mental Health subscale; S-GDS = Short Geriatric Depression Scale; OR = odds ratio; CI = confidence interval.

a. Missing data: sex (7), minority (13), high school diploma (10), age (14), mental health experience (6), S-GDS (8), and ERA-12 (8). *p < .05. **p < .01. ***p < .001.

Table 5. Logistic Regression of Participants’ Recommendation That the Vignette Character Seek Professional Help From Doctor or Counselor (n = 214)

Note: The OR column displays results from bivariate analysis, and the adjusted-OR column displays results from the final multivariate model including all covariates. Missing data: sex (7), minority (13), high school diploma (10), age (14), mental health experience (6), S-GDS (8), ERA-12 (8). ERA-12 = Expectations Regarding Aging-12–Mental Health subscale; S-GDS = Short Geriatric Depression Scale; OR = odds ratio; CI = confidence interval.

*p < .05. **p < .01. ***p < .001.

Goldney, 2003) but lower than an all-ages national study using the same vignette (67.3%; slightly modified in our study as described in the methods; Jorm et al.,

2006). Thus, the current findings are consistent with these past studies of mental health literacy with older adults, finding similarly low rates for older adults, lower than for younger age groups (Fisher & Goldney, 2003; Jorm et al., 2006). Less than half of the current participants recommended professional help (38.52%). Fewer of our participants suggested that the person should see a doctor (29.51%), compared to 53.3% in another study of older adults (Fisher & Goldney, 2003), with similar numbers who thought the person should see a mental health professional (12.0% compared to 11.89% of our sample).

Descriptively, more participants recognized depression and the need for professional help from the checklist than the open-ended questions. More than half of the participants (57.6%) recognized the person as depressed in the checklist, and more than half thought the person should seek professional assistance from their doctor (52.5%) or counselor (25.2%). These responses to the checklists did not significantly differ by vignette condition. These findings suggest that older adults may be able to recognize depression more accurately as a problem requiring professional help when it is mentioned to them by others and that such education by others may eliminate the effects of symptom presentation. Thus, they may be more open to professional help if it is suggested to them, even if they do not spontaneously suggest it.

In addition to symptom presentation, older adults were more likely to identify depression if they had prior mental health experience and if they held more positive expectations regarding mental health of older adults. These findings are consistent with the third hypothesis and are similar to other research that has demonstrated the relationship between familiarity with mental health problems and their recognition (Lauber, Nordt, Falcato, & Rossler, 2003). Also, prior research has found that older adults who hold more negative views of mental health in later life and think that depression is a normal part of aging are less likely to seek services for depression (Sarkisian et al., 2003). Therefore, older adults without prior mental health experience likely require more extensive education, and education about depression in later life should emphasize that depression is not normal. Participants’ own depressive symptoms did not appear to make them more likely to identify the person in the vignette as depressed. This is consistent with another study of adults across ages, using the same vignette of depression (Goldney, Fisher, & Wilson, 2001). Thus, merely experiencing similar symptoms did not appear to enhance identification of the problem, unless the person had prior mental health treatment experience, suggesting that they had labeled their own past mental health issue as such. Alternatively, the majority of the sample was not depressed; so it is possible that there were insufficient numbers of depressed participants to detect a relationship between participants’ own depressive symptoms and accurate identification.

Related to the fourth hypothesis, participants were most likely to identify a need for professional help if they accurately identified the problem as depression.

Other covariates of identifying depression, mental health experience and attitudes about aging, were not significant. Therefore, the most important factor related to help seeking appears to be identifying the problem as depression; this finding parallels research of actual service utilization, which indicates that the strongest covariate of older adults’ use of mental health services is their perception of a mental health need (Klap et al., 2003). Although vignette condition did not directly relate to identified need, it was strongly related to identifying depression. So, even if older adults with depression without dysphoria or sadness do present for professional help to a doctor, they are not likely to describe their problem as depression, which complicates the professional’s task of diagnosing the problem (Gallo, Ryan, et al., 1999; Kessler, Lloyd, Lewis, & Gray, 1999). Participants were more likely to identify depression and need for professional assistance if they were White and had more education (in bivariate analyses only). Health disparities in mental health service utilization have been well documented across all ages, including for older adults, with lower-income and minority elders less likely to receive mental health services (Crystal et al., 2003). Older adults from different cultural and educational backgrounds have differing conceptualizations of depression and may not even view depression as a mental health issue (Marwaha & Livingston, 2002). A prior study found that Black older adults were more likely to present with depression without sadness than White older adults (Gallo, Cooper-Patrick, & Lesikar, 1998), which could be one contributing factor to their lower rate of depression diagnosis (Gallo et al., 1997) and treatment. These findings underscore the necessity of developing public health education materials that emphasize the variety of depressive symptoms in addition to sadness or depressed mood and that are culturally relevant and appropriate for older adults of varying educational levels. The large number of Spanishspeaking participants (27.71% of total sample) also indicates the need for Spanish-speaking professionals and Spanish-language educational materials.

One promising observation is that the vast majority of participants thought something was wrong and recommended that some action be taken (only 3.3% said nothing was wrong, only 2.9% said that nothing should be done). In cases of milder, subthreshold depressive symptoms, these other actions (e.g., socializing, seeking social support) may be sufficient to ameliorate symptoms. Also these activities may lead the person to seek professional treatment in cases of more severe depression, such as if someone encourages or assists the person to seek formal treatment.

Limitations of the study should be considered. These findings are from a convenience sample using a hypothetical vignette. Thus, it is difficult to determine how the findings would generalize to the general older adult population and impact their actual help-seeking behavior. Given that identifying a behavioral health need is one of the strongest predictors of older adults’ actual service use (Klap et al., 2003), it is important to understand conditions that affect how older adults identify need. The current findings suggest that symptom presentation, specifically the presence of sadness, is one important factor in identifying need. Also, utilization of the vignette methodology allowed for an experimental design, which is a strength of the study in demonstrating a causal effect of symptom presentation. Another limitation related to the multivariate regression analyses is that minority participants and participants who did not identify depression or need for help had more missing data in the surveys, which further limits the generalizability of the multivariate results.

In conclusion, these findings indicate that many older adults still need education about depression and when professional assistance is indicated. Education about depression has been shown to improve knowledge and some attitudes about depression for adults with depressive symptoms, although additional research is needed to optimize educational materials to produce behavioral and symptom change (Jorm et al., 2003). The current study suggests that public health education materials for older adults, such as educational pamphlets or websites, should note explicitly that depression is not only sadness but can also mean losing interest or pleasure in activities and that this form of depression is not normal and still warrants treatment.These materials and Web resources could be disseminated broadly in a range of medical, social service, and community settings. Development of materials should be conducted in collaboration with professionals and older adults from diverse racial and ethnic backgrounds to ensure that they describe anhedonia in culturally relevant terms and use culturally relevant examples of anhedonia. Such educational materials also would be valuable to disseminate across communitybased settings that involve older adults, such as volunteer organizations, faith-based organizations, or community or neighborhood organizations.

The findings also have implications for health care and social service professionals from a variety of backgrounds, who need to be aware of depression without sadness and the fact that older adults may not recognize it in themselves. In this case, the older person would be unlikely to describe their problem in psychological or mood-related terms. A more psychological presentation of symptoms is associated with better recognition of depression by a health care professional (Kessler et al., 1999), however, and primary care physicians have reported that depression without sadness is challenging to identify (Gallo, Ryan, et al., 1999). Thus, professionals need to look for depression without sadness and thoroughly educate their patients when it is identified.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The authors received no financial support for the research and/or authorship of this article.

References

Blazer, D. G. (2003). Depression in late life: Review and commentary. Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 58A, 249-265.

Crystal, S., Sambamoorthi, U., Walkup, J. T., Akincigil, A., McCall, N. T., Parks, P., et al. (2003). Diagnosis and treatment of depression in the elderly Medicare population: Predictors, disparities, and trends. Journal of the American Geriatrics Society, 51, 1718-1728.

Fisher, L. J., & Goldney, R. D. (2003). Differences in community mental health literacy in older and younger Australians. International Journal of Geriatric Psychiatry, 18(1), 33-40.

Gallo, J. J., Anthony, J. C., & Muthén, B. O. (1994). Age differences in the symptoms of depression: A latent trait analysis. Journal of Gerontology: Psychological Sciences, 49, P251-P264.

Gallo, J. J., Cooper-Patrick, L., & Lesikar, S. (1998). Depressive symptoms of Whites and African Americans aged 60 years and older. Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 53B, P277-P286.

Gallo, J. J., Rabins, P., & Anthony, J. (1999). Sadness in older persons: 13-year follow-up of a community sample in Baltimore, Maryland. Psychological Medicine, 29, 341-350.

Gallo, J. J., Rabins, P. V., Lyketsos, C. G., Tien, A. Y., & Anthony, J. C. (1997). Depression without sadness: Functional outcomes of nondysphoric depression in later life. Journal of the American Geriatrics Society, 45, 570-578.

Gallo, J. J., Ryan, S. D., & Ford, D. E. (1999). Attitudes, knowledge, and behavior of family physicians regarding depression in late life. Archives of Family Medicine, 8, 249-256.

Goldney, R. D., Fisher, L. J., & Wilson, D. H. (2001). Mental health literacy: An impediment to the optimum treatment of major depression in the community. Journal of Affective Disorders, 64, 277-284.

Jorm, A. F., Christensen, H., & Griffiths, K. M. (2006). The public’s ability to recognize mental disorders and their beliefs about treatment: Changes in Australia over 8 years. Australian and New Zealand Journal of Psychiatry, 40, 36-41.

Jorm, A. F., Griffiths, K. M., Christensen, H., Korten, A. E., Parslow, R. A., & Rodgers, B. (2003). Providing information about the effectiveness of treatment options to depressed people in the community: A randomized controlled trial of effects on mental health literacy, help-seeking and symptoms. Psychological Medicine, 33, 1071-1079.

Jorm, A. F., Korten, A. E., Jacomb, P. A., Christensen, H., Rodgers, B., & Pollitt, P. (1997). “Mental health literacy”: A survey of the public’s ability to recognise mental disorders and their beliefs about the effectiveness of treatment. Medical Journal of Australia, 166, 182-186.

Kessler, D. A., Lloyd, K., Lewis, G., & Gray, D. P. (1999). Cross sectional study of symptom attribution and recognition of depression and anxiety in primary care. British Medical Journal, 318, 436-440.

Klap, R., Unroe, K. T., & Unützer, J. (2003). Caring for mental illness in the United States: A focus on older adults. American Journal of Geriatric Psychiatry, 11, 517-524.

Lauber, C., Nordt, C., Falcato, L., & Rossler, W. (2003). Do people recognize mental illness? Factors influencing mental health literacy. European Archives of Psychiatry and Clinical Neuroscience, 253, 248-251.

Marwaha, S., & Livingston, G. (2002). Stigma, racism or choice. Why do depressed ethnic elders avoid psychiatrists? Journal of Affective Disorders, 72, 257-265.

Sarkisian, C. A., Hays, R. D., Berry, S., & Mangione, C. M. (2002). Development, reliability, and validity of the expectations regarding aging (ERA-38) survey. The Gerontologist, 42, 534-542.

Sarkisian, C. A., Lee-Henderson, M. H., & Mangione, C. M. (2003). Do depressed older adults who attribute depression to “old age” believe it is important to seek care? Journal of General Internal Medicine, 18, 1001-1005.

Sarkisian, C. A., Steers, N., Hays, R. D., & Mangione, C. M. (2005). Development of the 12-Item Expectations Regarding Aging Survey. The Gerontologist, 45, 240-248.

Schonfeld, L., King-Kallimanis, B., Duchene, D., Etheridge, R., Herrera, J., Lawton Barry, K., et al. (2009). The Florida BRITE Project: Screening and brief intervention for substance misuse in older adults. American Journal of Public Health.

Early view online, published May 14, 2009, http://www.ajph.org/cgi/reprint/ AJPH.2008.149534v1

Sheikh, J. I., & Yesavage, J. A. (1986). Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinical Gerontologist, 5, 165-174.

Wang, P. S., Lane, M., Olfson, M., Pincus, H. A., Wells, K. B., & Kessler, R. C. (2005). Twelve-month use of mental health services in the United States: Results from the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 629-640.

Bios

Amber M. Gum, PhD, is assistant professor in the Department of Aging and Mental Health Disparities, Louis de la Parte Florida Mental Health Institute, University of South Florida. She conducts research on older adults’ utilization of mental health services and mental health interventions for underserved older adults.

Sarah J. McDougal, BA, is an MPH candidate in the School of Public Health at the University of Washington, Seattle. Her current research interests include risks of sexually transmitted infections (STI) among male victims and/or perpetrators of intimate partner violence, partner concurrency and risk of STIs, and pathogens associated with non-gonococcal urethritis. She was formerly research assistant in the Department of Aging and Mental Health Disparities, Louis de la Parte Florida Mental Health Institute, University of South Florida.

Jessica M. McIlvane, PhD, is assistant professor in the School of Aging Studies at the University of South Florida, Tampa. Her research interests include coping with stress and chronic illness in older adulthood, racial/ethnic health disparities in arthritis, and social relationships.

Chivon A. Mingo, MA, is a doctoral candidate in the School of Aging Studies at the University of South Florida, Tampa. Her research focus is on health disparities in older adults, specifically osteoarthritis (OA) health disparities between Blacks and Whites.