Abstract
Expectations about treatment and beliefs about illness influence adherence in physical disorders, but the extent to which this occurs in mood disorders is unknown. Identifying modifiable factors, such as beliefs, may improve adherence to mood disorder medications. Data from the Collaborative Psychiatric Epidemiology Surveys were used to examine relationships among perceived symptom targets of medication (mood only, non-mood only, mood and non-mood) and self-reported adherence to antidepressants, anxiolytics, and sedatives. The sample included 807 community-dwelling individuals with and without depression and anxiety who regularly took one of these medications in the year before the survey. Slightly over half (53.2%) of respondents were adherent. Perceived medication purpose was only significantly related to adherence among Latino respondents. Latino respondents who viewed their symptom target as non-mood only were the most adherent. Perceived symptom targets of medications were not associated with most patients’ adherence behaviors for antidepressants, anxiolytics, and sedatives.
Keywords: Adherence, perceptions, beliefs, mood
Introduction
Of individuals prescribed medications for affective disorders, up to 60% are nonadherent to these medications.1 Nonadherence to medication refers to a spectrum of behaviors in which a patient is not taking medication as instructed. One may take a greater or lesser quantity of a medication than intended, not take a medication at all, or take it fewer days or less often than instructed.2 Nonadherence is of special concern for patients prescribed medication for depression and anxiety, as depression and anxiety are associated with lower quality of life, poorer management of comorbid physical illnesses, and higher healthcare costs.3-7 It is important to identify modifiable factors that may improve adherence for individuals with mood disorders.8
Beliefs about illness and treatment are one set of factors that may be changed9 and that may influence adherence to antidepressants and anxiolytics/sedatives. According to the Common Sense Model of Illness Representations (CSM), individuals form representations of illness that consist of five dimensions: identity of symptoms, causes, consequences, timeline (e.g. chronic or acute), and controllability.10 In both physical and mental illnesses, differences in illness representations explain a large proportion of variance in adherence rates.11-12
There have been calls to understand how differences in illness representations in depression are associated with adherence and treatment outcomes.13 When individuals describe depression, their descriptions most often fall into the identity (listing symptoms) and consequence (describe impact on social functioning and self-image) dimensions.14 Adherence to medications for physical illnesses is less likely when beliefs about symptom identity do not match beliefs about targets of treatment. Individuals taking maintenance medications for chronic illnesses such as hypertension and diabetes will be less adherent if they expect an immediate change in acute subjective symptoms.15 For example, individuals who mistakenly believe they can rely solely on subjective symptoms to monitor blood pressure are less likely to adhere to antihypertensive medication.16
The same relationship may occur in patients with depression and/or anxiety. Beliefs about whether symptoms should be attributed to depression or to some other condition may influence adherence to medications for mood disorders.8 For example, a person might not associate fatigue with depression, so if they feel they are taking an antidepressant to mitigate fatigue, they may be less adherent than if they feel they are taking an antidepressant to mitigate depressed mood or anhedonia. A similar relationship between symptom identity and medication adherence might occur in anxiety, as it also encompasses mood and physical symptoms that could be attributed to a variety of causes. In addition, anxiety and depression often co-occur, and a mix of antidepressants and anxiolytics/sedatives are prescribed to people with anxiety and depression.4, 17
The purpose of this study is to explore whether perceived symptom targets of medication are related to adherence in a group of individuals taking antidepressants, sedatives, and anxiolytics. Because depression and anxiety are characterized as mood disorders, the hypothesis is that individuals who report taking these medications for mood reasons will be more likely to be adherent than those who report taking the medications for non-mood reasons, including physical and social functioning. Symptom identity is thus split into three groups: mood only, non-mood only (physical, role functioning, cognitive, or other), or both mood and non-mood.
To understand the relationship between perceived symptom targets and adherence, it is important to also consider clinical and sociodemographic factors that may be associated with adherence and/or perceptions. Nonadherence is higher for those who are unable to purchase medications.18 Depression and anxiety are associated with reduced adherence to medications in general, possibly due to the negative thoughts and fatigue that are part of these disorders.4 Care from mental health specialists may be associated with improved adherence to antidepressants.19 Perceived symptom targets also may differ by whether mental health care and/or prescriptions were from primary care providers, whose focus is physical health, or from mental health specialists. Adults over the age of 65 are more likely to report physical than mood symptoms of mood disorders;20 this might be associated with perceived symptom targets. While the degree to which members of racial/ethnic minority groups differentially report mood or non-mood symptoms of mood disorders is unclear, treatment recommendations for mood disorders in minority groups advise clinicians to be especially cognizant of physical symptoms.21-22
In addition, because older adults and members of racial/ethnic minority groups are less likely to access mental health care in general,23 understanding how treatment beliefs are associated with adherence in these subgroups is especially important. Understanding how patients view their prescribed treatments may help clinicians identify modifiable beliefs that influence adherence and improve care.
Methods
The CPES is a nationally representative survey of 20,013 community-dwelling adults collected between 2001-2003 that focuses on mental illness symptoms and treatment. Further details about the CPES are available elsewhere.24 Survey coordinators presented respondents with a list of medication names, and respondents were asked: “Which of the medicines on this list did you take in the past 12 months for any of the following problems: problems with your emotions, nerves, mental health, substance use, energy, concentration, sleep, or ability to cope with stress?” Respondents were prompted to consult their medication bottles to obtain medication names. Detailed information about the medication can be linked to medication name only for those reporting a single medication. For that reason, the sample was restricted to 807 individuals who reported regularly taking one antidepressant, anxiolytic, or sedative under the supervision of a healthcare provider in the year prior to the survey.
Outcome Variable
Respondents were included if they were taking the medication at the time of the survey or had stopped the medication with a healthcare provider's agreement within the last year. A respondent was defined as adherent if, while taking the medication, he/she remembered to take the medication and took the amount instructed every day in a typical month. The exact question asked of respondents was: “People do not always take their medicine as they are supposed to. Think of a typical month when you took [name of medication] in the past 12 months. How many days out of 30 did you typically either forget to take it or take less of it than you were supposed to take?” (Codebook available at http://www.icpsr.umich.edu/icpsrweb/CPES/files/cpes).
Main Explanatory Variable
Respondents were asked the medication's purpose: “mood (sadness/depression/crying, manic mood, anger/irritability, nerves/anxiety, panic, suicidal thoughts); physical symptoms (low energy, poor appetite, poor sleep, physical pain); cognitive symptoms (poor concentration, poor memory); role functioning (little or no sexual functioning, marital problems, not getting along with others, poor work performance); or other (alcohol/drug problems; other volunteered response)”. It was subsequently categorized as mood only, non-mood only (physical/cognitive/role functioning/other) or mood and non-mood.
Other Control Variables
Respondent sociodemographics and physical health
Respondent sociodemographics were recorded, including age, race/ethnicity (Non-Latino White, African American / Afro-Caribbean, Asian, Latino, or other), education (high school education or higher versus less than a high school education), sex, poverty level (above 200% of the poverty threshold versus at or below 200% of the poverty threshold), and possession of private insurance. The number of chronic physical conditions ever experienced by respondents was also recorded.
Other mental health conditions and care
Variables included: type of provider that prescribed the medication, any other mental health care (MHC) use in the past year (none, from a primary care provider only, from a mental health specialist only, from both types of providers), past-year or history of probable major depressive disorder or anxiety (generalized anxiety disorder, panic disorder, social phobia, agoraphobia, posttraumatic stress disorder); and medication class (antidepressant or anxiolytic/sedative). All variables were based on respondent self-report; more detail is available elsewhere.25-26
Analyses
The relationship between adherence and perceived purpose was analyzed with Rao-Scott chi-square tests and multivariate logistic regression for the entire sample. Unadjusted chi-square tests were performed for subgroups of older adults and racial/ethnic minorities; small sample size in subgroups precluded adjusted analyses. Analyses were performed with SAS 9.2 and SUDAAN 10.0.1 and accounted for sampling weights. Unweighted n's and weighted frequencies are reported for all variables. This study was deemed exempt from review by the James J Peters VA Medical Center Institutional Review Board.
Results
The majority of respondents reported that their antidepressant, anxiolytic, or sedative was prescribed for mood improvement only (n= 481, 60.29%). In contrast, about one-fifth (n = 204, 22.83%) reported taking the medication for a non-mood purpose only; and one-sixth reported both mood and non-mood purposes (n=120, 16.88%) (see Table 1). Slightly over half the sample reported medication adherence (n = 416, 53.22%).
Table 1.
Descriptive statistics and bivariate testsa examining association between perceived symptom target and adherence to medication
| Variable | Entire Sample (n = 807) N(%)b or Mean (SE) | Adherentc N(%)b or Mean (SE) | |
|---|---|---|---|
| No (n = 391, 46.78%) | Yes (n = 416, 53.22%) | ||
| Problems for which respondent took medication | |||
| Mood improvement only | 481 (60.29) | 236 (47.00) | 245 (53.00) |
| Non-mood only | 204 (22.83) | 84 (43.64) | 120 (56.36) |
| Mood and non-mood | 120 (16.88) | 70 (50.61) | 50 (49.39) |
| Medication class | |||
| Antidepressant | 587 (75.98) | 299 (49.31) | 288 (50.69) |
| Sedative/Anxiolytic | 220 (24.02) | 92 (38.77) | 128 (61.23) |
| From whom did respondent receive prescription? | |||
| General or family doctor | 533 (67.25) | 255 (48.70) | 278 (51.30) |
| Psychiatrist | 150 (15.62) | 80 (49.46) | 70 (50.54) |
| Other health professional | 124 (17.14) | 56 (36.81) | 68 (63.19) |
| MHC in past 12 months | |||
| None | 389 (45.47) | 174 (42.97) | 215 (57.03) |
| From a primary care provider only | 205 (31.15) | 105 (50.14) | 100 (49.86) |
| From a mental health specialist only | 108 (11.71) | 53 (41.27) | 55 (58.73) |
| From a primary care provider and a specialist | 105 (11.66) | 59 (58.18) | 46 (41.82) |
| MDD (past 12 months) | |||
| Yes | 190 (20.21) | 103 (54.53) | 87 (45.47) |
| No | 617 (79.79) | 288 (44.82) | 329 (55.18) |
| MDD (history) | |||
| Yes | 172 (21.57) | 84 (48.71) | 88 (51.29) |
| No | 635 (78.43) | 307 (46.25) | 328 (53.75) |
| Anxiety (past 12 months) | |||
| Yes | 349 (39.70) | 178 (51.31) | 171 (48.69) |
| No | 458 (60.30) | 213 (43.80) | 245 (56.20) |
| Anxiety (History) | |||
| Yes | 179 (21.00) | 91 (51.38) | 88 (48.62) |
| No | 628 (79.00) | 300 (45.55) | 328 (54.44) |
| Number of comorbid physical conditions | 1.03 (0.05) | 0.94 (0.09) | 1.10 (0.08) |
| Race | |||
| Non-Latino White | 462 (85.19) | 214 (45.87) | 248 (54.13) |
| African American or Afro-Caribbean | 172 (5.49) | 91 (49.43) | 81 (50.57) |
| Asian | 32 (1.19) | 16 (55.69) | 16 (44.31) |
| Latino | 122 (5.73) | 59 (51.56) | 63 (48.44) |
| Other | 19 (2.40) | 11 (56.99) | 8 (43.01) |
| Education | |||
| High school education or higher | 645 (84.29) | 317 (48.54) | 328 (51.46) |
| Less than high school | 162 (15.71) | 74 (37.33) | 88 (62.67) |
| Sex | |||
| Male | 199 (30.86) | 100 (48.97) | 99 (51.03) |
| Female | 608 (69.14) | 291 (45.80) | 317 (54.20) |
| Poverty threshold | |||
| Less than or equal to 200% | 351 (39.11) | 169 (46.22) | 182 (53.78) |
| Over 200% | 456 (60.89) | 222 (47.14) | 234 (52.86) |
| Private insurance | |||
| Yes | 546 (74.71) | 262 (48.07) | 284 (51.93) |
| No | 260 (25.29) | 128 (42.92) | 132 (57.08) |
| Age | |||
| Under 65 | 690 (84.94) | 354 (50.27) | 336 (49.73)* |
| 65 and older | 117 (15.06) | 37 (27.07) | 80 (72.93) |
MHC = mental health care; MDD = major depressive disorder; SE = standard error
Rao-Scott chi-square or t-test
Unweighted n's and weighted percentages are presented; Numbers may not sum to total sample size in the case of missing values on some variables
Weighted percentages reflect occurrence of adherence within each level of each variable.
Unadjusted Rao-Scott chi-square test significant at p < .001
Entire Sample
Adults 65 years of age or older had higher adherence (72.93%) than those under the age of 65 (49.73%) in unadjusted bivariate tests (χ2 = 12.64, df = 1, p < .001). Perceived medication purpose was not significantly associated with medication adherence in unadjusted analyses.
In adjusted logistic regression, older age (Odds Ratio [OR] / year = 1.02, 95% Confidence Interval [CI] = 1.01-1.04; Table 2) remained significantly associated with increased adherence. History of an anxiety disorder was associated with lower odds of adherence (OR = 0.61, CI = 0.38-0.99). Individuals who possessed private insurance also had lower odds of adherence than those who did not have private insurance (OR = 0.64, CI = 0.41-0.99). Perceived medication purpose was not associated with medication adherence in adjusted analyses.
Table 2.
Adjusted logistic regression examining association between perceived symptom target and adherence to medication in entire sample
| Variable | Odds Ratio (95% Confidence Interval) |
|---|---|
| Problems for which respondent took medication | |
| Mood improvement only | 1.00 |
| Non-mood only | 0.77 (0.44, 1.32) |
| Mood and non-mood | 0.91 (0.55, 1.51) |
| Medication class | |
| Sedative/Anxiolytic | 1.00 |
| Antidepressant | 0.71 (0.41, 1.23) |
| From whom did respondent receive prescription? | |
| General or family doctor | 1.00 |
| Psychiatrist | 0.85 (0.41, 1.76) |
| Other health professional | 1.59 (0.99, 2.56) |
| MHC in past 12 months | |
| None | 1.00 |
| From a primary care provider only | 0.79 (0.48, 1.30) |
| From a mental health specialist only | 1.53 (0.71, 3.28) |
| From a primary care provider and a specialist | 0.79 (0.43, 1.45) |
| MDD (past 12 months) | 0.82 (0.49, 1.35) |
| MDD (history) | 0.90 (0.63, 1.28) |
| Anxiety (past 12 months) | 0.79 (0.49, 1.29) |
| Anxiety (History) | 0.61 (0.38, 0.99) |
| Number of comorbid physical conditions | 1.00 (0.79, 1.26) |
| Race | |
| Non-Latino White | 1.00 |
| African American or Afro-Caribbean | 0.79 (0.49, 1.27) |
| Asian | 0.56 (0.20, 1.61) |
| Latino | 0.73 (0.42, 1.25) |
| Other | 0.80 (0.29, 2.24) |
| High school education or higher | 0.77 (0.34, 1.71) |
| Male | 0.85 (0.59, 1.24) |
| Poverty threshold less than or equal to 200% | 0.79 (0.44, 1.41) |
| Possessed private insurance | 0.64 (0.41, 0.99) |
| Age | 1.02 (1.01, 1.04) |
MHC = mental health care; MDD = major depressive disorder; SE = standard error
Subgroup Analyses
In unadjusted chi-square analyses, perceived purpose and adherence were significantly related for one racial/ethnic group: Latino respondents. Over half of Latino respondents (n=58, 53.87%) reported a mood reason only, 50 (38.77%) reported a non-mood reason only, and 14 (7.4%) reported both mood and non-mood reasons. Latino respondents who reported only non-mood reasons were more likely to be adherent than those who reported only mood reasons or mood and non-mood reasons (64.50%, 38.02%, 39.68% were adherent, respectively, χ2 = 6.61, df = 2, p =.04; Table 3). There was no significant relationship between perceived purpose and adherence for respondents 65 years of age or older or for African American/Afro-Caribbean respondents.
Table 3.
Subgroup analyses (Rao-Scott chi-squarea) of perceived symptom targets and adherence in older adults and in racial/ethnic minorities
| Older Adults (n = 117) | African-American / Afro-Caribbean (n = 172) | Latino (n = 122) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Not adherent N(%)b | Adherent N(%)b | χ2, p-value | Not adherent N(%)b | Adherent N(%)b | χ2, p-value | Not adherent N(%)b | Adherent N(%)b | χ2, p-value | |
| Mood only | 17 (31.27) | 37 (68.73) | 1.28, p = .53 | 59 (50.87) | 49 (49.13) | 0.34, p = .84 | 33 (61.98) | 25 (38.02) | 6.61, p = .04 |
| Non-mood only | 16 (25.42) | 34 (74.58) | 22 (45.87) | 26 (54.13) | 20 (35.41) | 30 (64.59) | |||
| Mood and non-mood | 4 (13.25) | 8 (86.75) | 10 (53.00) | 6 (47.00) | 6 (60.32) | 8 (39.68) | |||
Rao-Scott chi-square, 2 degrees of freedom for each test
Unweighted n's and weighted percentages are presented
Discussion
This study aimed to understand the relationship between perceived symptom targets of antidepressants and anxiolytics/sedatives and adherence to these medications in a community sample. In physical illnesses, concordance between symptoms and perceived symptom targets of medication is associated with improved adherence.15-16 In this sample, however, perceived symptom targets were only associated with medication adherence among Latino respondents. Contrary to the hypothesis, Latino respondents who perceived their antidepressant/sedative/anxiolytic medication to be for only non-mood reasons had higher adherence than those who perceived their medication to be for mood improvement. One possibility is that Latino respondents may be better educated on the relationship between somatic symptoms and mood disorders than other respondents, given that treatment guidelines for Latino patients stress the need for providers to recognize somatic symptoms.22 However, Latino is a broad ethnic category, and the number of Latino respondents in this sample was fairly small (n=122); these results should be replicated in larger samples with more detailed measures of ethnicity.27
Adherence was related to some clinical and sociodemographic factors in the larger sample in multivariate analyses. Worse adherence was seen in respondents with a history of anxiety and those who were younger, which is consistent with other findings in the literature.2, 4 Worse adherence was also seen in individuals with private insurance, although this relationship was not robust to different specifications of the model (results available from authors) and may reflect a statistical artifact.
From these results, it does not appear that the distinction between mood and non-mood targets of antidepressants and anxiolytics/sedatives are important to patients’ adherence behaviors. While this study did not identify a modifiable belief associated with medication adherence, these results suggest that clinicians may mention a range of benefits of these medications without decreasing the patients’ likelihood of adherence. While providers should be careful to describe the full range of benefits and risks of any medication, they do not need to be concerned about lower adherence in the general patient population if a patient is more interested in non-mood benefits of the medication. Future research should determine whether discussing these benefits (in addition to mood-related benefits) with Latino patients improves adherence, or whether this pattern reflects a subgroup of patients who have been better educated in non-mood rather than mood benefits of antidepressants, anxiolytics, and sedatives.
A strength of this study is the use of CPES data, which are unique in that they include detailed information about antidepressant, anxiolytic, and sedative use within a large, diverse community sample. However, the study has the following limitations: 1) The sample was restricted to those taking one medication in the past year, and most respondents had been taking the medication for at least 30 days; findings may not apply to individuals beginning new medications or taking multiple medications for mood disorders. 2) Adherence was by patient self-report and may have been over-reported. 3) The diagnostic indication for each prescribed medication is unknown. 4) The small sample size in subgroup analyses precluded multivariate analysis.
Implications for Behavioral Health
Perceived symptom targets of antidepressants, anxiolytics, and sedatives do not appear to be related to adherence in the general patient population. Healthcare providers can discuss the non-mood benefits of medications for mood and anxiety disorders with patients without decreasing adherence.
Acknowledgements
The authors received funding from: Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research & Development Service CDA 11-201/CDP 12-255 and IIR 10-146; the Greenwall Foundation; National Institute of Mental Health T32 MH16242-29; and the National Palliative Care Research Center. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.
Footnotes
Conflicts of Interest
'The authors have no conflicts of interest to report.
References
- 1.Lingam R, Scott J. Treatment non-adherence in affective disorders. Acta Psychiatrica Scandinavica. 2002;105:164–172. doi: 10.1034/j.1600-0447.2002.1r084.x. [DOI] [PubMed] [Google Scholar]
- 2.Julius RJ, Novitsky MA, Dubin WR. Medication adherence: A review of the literature and implications for clinical practice. Journal of Psychiatric Practice. 2009;15:34–44. doi: 10.1097/01.pra.0000344917.43780.77. [DOI] [PubMed] [Google Scholar]
- 3.Beekman ATF, Deeg DJH, Braam AW, et al. Consequences of major and minor depression in later life: A study of disability, well-being and service utilization. Psychological Medicine. 1997;27:1397–1409. doi: 10.1017/s0033291797005734. [DOI] [PubMed] [Google Scholar]
- 4.DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for noncompliance with medical treatment: Meta-analysis of the effects of anxiety and depression on patient adherence. Archives of Internal Medicine. 2000;160:2101–2107. doi: 10.1001/archinte.160.14.2101. [DOI] [PubMed] [Google Scholar]
- 5.Spencer R, Nilsson M, Wright A, et al. Anxiety disorders in advanced cancer patients. Cancer. 2010;116:1810–1819. doi: 10.1002/cncr.24954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Manning WG, Wells KB. The effects of psychological distress and psychological well-being on use of medical services. Medical Care. 1992;30:541–553. doi: 10.1097/00005650-199206000-00007. [DOI] [PubMed] [Google Scholar]
- 7.Olfson M, Gameroff MJ. Generalized anxiety disorder, somatic pain and health care costs. General Hospital Psychiatry. 2007;29:310–316. doi: 10.1016/j.genhosppsych.2007.04.004. [DOI] [PubMed] [Google Scholar]
- 8.Brown C, Dunbar-Jacob J, Palenchar DR, et al. Primary care patients’ personal illness models for depression: A preliminary investigation. Family Practice. 2001;18:314–320. doi: 10.1093/fampra/18.3.314. [DOI] [PubMed] [Google Scholar]
- 9.Halm E, Mora P, Leventhal H. No symptoms, no asthma: The acute episodic belief is associated with poor self-management among inner-city adults with persistent asthma. Chest. 2006;129:573–580. doi: 10.1378/chest.129.3.573. [DOI] [PubMed] [Google Scholar]
- 10.Leventhal H, Diefenbach M, Leventhal EA. Illness cognition: Using common sense to understand treatment adherence and affect cognition interactions. Cognitive Therapy & Research. 1992;16:143–163. [Google Scholar]
- 11.Horne R, Weinman J. Self-regulation and self-management in asthma: Exploring the role of illness perceptions and treatment beliefs in explaining non-adherence to preventer medication. Psychology and Health. 2002;17:17–32. [Google Scholar]
- 12.Lynch J, Kendrick T, Moore M, et al. Patients’ beliefs about depression and how they relate to duration of antidepressant treatment. Use of a US measure in a UK primary care population. Primary Care Mental Health. 2006;4:207–217. [Google Scholar]
- 13.Lynch J, Moore M, Moss-Morris R, et al. Are patient beliefs important in determining adherence to treatment and outcome for depression? Development of the beliefs about depression questionnaire. Journal of Affective Disorders. 2011;133:29–41. doi: 10.1016/j.jad.2011.03.019. [DOI] [PubMed] [Google Scholar]
- 14.Fortune G, Barrowclough C, Lobban F. Illness representations in depression. British Journal of Clinical Psychology. 2004;43:347–364. doi: 10.1348/0144665042388955. [DOI] [PubMed] [Google Scholar]
- 15.McAndrew LM, Muscumeci-Szabó TJ, Mora PA, et al. Using the common sense model to design interventions for the prevention and management of chronic illness threats: From description to process. British Journal of Health Psychology. 2008;13:195–204. doi: 10.1348/135910708X295604. [DOI] [PubMed] [Google Scholar]
- 16.Meyer D, Leventhal H, Gutmann M. Common-sense models of illness: The example of hypertension. Health Psychology. 1985;4:115–135. doi: 10.1037//0278-6133.4.2.115. [DOI] [PubMed] [Google Scholar]
- 17.Horwitz A. How an age of anxiety became an age of depression. The Milbank Quarterly. 2010;88:112–138. doi: 10.1111/j.1468-0009.2010.00591.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Donohue JM, Zhang Y, Men A, et al. Impact of Medicare Part D on antidepressant treatment, medication choice and adherence among older adults with depression. American Journal of Geriatric Psychiatry. 2011;19:989–997. doi: 10.1097/JGP.0b013e3182051a9b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Stein MB, Cantrell CR, Sokol MC, et al. Antidepressant adherence and medical resource use among managed care patients with anxiety disorders. Psychiatric Services. 2006;57:673–680. doi: 10.1176/ps.2006.57.5.673. [DOI] [PubMed] [Google Scholar]
- 20.Gallo JJ, Anthony JC, Muthén BO. Age differences in the symptoms of depression: A latent trait analysis. Journals of Gerontology: Psychological Sciences. 1994;49:251–264. doi: 10.1093/geronj/49.6.p251. [DOI] [PubMed] [Google Scholar]
- 21.Uebelacker LA, Strong D, Weinstock LM, et al. Use of item response theory to understand differential functioning of DSM-IV major depressive symptoms by race, ethnicity, and gender. Psychological Medicine. 2009;39:591–601. doi: 10.1017/S0033291708003875. [DOI] [PubMed] [Google Scholar]
- 22.Lewis-Fernández R, Das AK, Alfonso C, et al. Depression in US Hispanics: Diagnostic and management considerations in family practice. The Journal of the American Board of Family Medicine. 2005;18:282–296. doi: 10.3122/jabfm.18.4.282. [DOI] [PubMed] [Google Scholar]
- 23.Wang PS, Lane M, Olfson M, et al. Twelve-month use of mental health services in the United States. Archives of General Psychiatry. 2005;62:629–640. doi: 10.1001/archpsyc.62.6.629. [DOI] [PubMed] [Google Scholar]
- 24.Heeringa S, Wagner J, Torres M, et al. Sample designs and sampling methods for the Collaborative Psychiatric Epidemiology Studies (CPES). International Journal of Methods in Psychiatric Research. 2004;13:221–240. doi: 10.1002/mpr.179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Garrido MM, Kane RL, Kaas M, et al. Perceived need for mental health care among community-dwelling older adults. Journals of Gerontology: Psychological Sciences. 2009;64B:704–712. doi: 10.1093/geronb/gbp073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Garrido MM, Kane RL, Kaas M, et al. Use of mental health care by community-dwelling older adults. Journal of the American Geriatric Society. 2011;59:50–56. doi: 10.1111/j.1532-5415.2010.03220.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Alegría M, Takeuchi D, Canino G. Considering context, place, and culture: The National Latino and Asian American Study. International Journal of Methods in Psychiatric Research. 2004;13:208–220. doi: 10.1002/mpr.178. [DOI] [PMC free article] [PubMed] [Google Scholar]