Abstract
Background: The early diagnosis and treatment of depression are cancer care priorities. These priorities are critical for prostate cancer survivors because men rarely seek mental health care. However, little is known about the epidemiology of depression in this patient population. The goal of this study was to describe the prevalence and predictors of probable depression in prostate cancer survivors. Methods: The data were from a population-based cohort of North Carolinian prostate cancer survivors who were enrolled from 2004 to 2007 in the North Carolina–Louisiana Prostate Cancer Project (n = 1031) and were prospectively followed annually from 2008 to 2011 in the Health Care Access and Prostate Cancer Treatment in North Carolina study (n = 805). Generalized estimating equations were used to evaluate an indicator of probable depression (Short Form 12 mental composite score ≤48.9; measured at enrollment and during the annual follow-up) as a function of individual-level characteristics within the longitudinal data set. Results: The prevalence of probable depression fell from 38% in the year of the cancer diagnosis to 20% 6 to 7 years later. Risk factors for probable depression throughout the study were African American race, unemployment, low annual income, younger age, recency of cancer diagnosis, past depression, comorbidities, treatment decisional regret, and nonadherence to exercise recommendations. Conclusions: Depression is a major challenge for prostate cancer survivors, particularly in the first 5 years after the cancer diagnosis. To the authors' knowledge, this is the first study to demonstrate an association between treatment decisional regret and probable depression.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 3418-3427 |
| Number of pages | 10 |
| Journal | Cancer |
| Volume | 125 |
| Issue number | 19 |
| DOIs | |
| State | Published - Oct 1 2019 |
| Externally published | Yes |
Keywords
- depression
- health disparity
- predictors
- prostate cancer
- risk factors
ASJC Scopus subject areas
- Oncology
- Cancer Research
Access to Document
Other files and links
Fingerprint
Dive into the research topics of 'Prevalence and predictors of probable depression in prostate cancer survivors'. Together they form a unique fingerprint.Cite this
- APA
- Standard
- Harvard
- Vancouver
- Author
- BIBTEX
- RIS
Prevalence and predictors of probable depression in prostate cancer survivors. / Erim, Daniel O.; Bensen, Jeannette T.; Mohler, James L. et al.
In: Cancer, Vol. 125, No. 19, 01.10.2019, p. 3418-3427.Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - Prevalence and predictors of probable depression in prostate cancer survivors
AU - Erim, Daniel O.
AU - Bensen, Jeannette T.
AU - Mohler, James L.
AU - Fontham, Elizabeth T.H.
AU - Song, Lixin
AU - Farnan, Laura
AU - Delacroix, Scott E.
AU - Peters, Edward S.
AU - Erim, Theodora N.
AU - Chen, Ronald C.
AU - Gaynes, Bradley N.
N1 - Funding Information: The North Carolina–Louisiana Prostate Cancer Project and the Health Care Access and Prostate Cancer Treatment in North Carolina study were supported by the Department of Defense (contract DAMD 17-03-2-0052) and the American Cancer Society (award RSGT-08-008-01-CPHPS). Unequal access to mental health care may explain the association between race and probable depression. Evidence from studies in the general population have shown that the incidence of depression is identical in African Americans and white Americans and that African Americans have poorer access to mental health care in comparison with white Americans. Appropriate depression care promotes recovery and prevents relapse/recurrence of depression; thus, limited access to mental health care may make African American prostate cancer survivors more vulnerable to depression. However, little is known about access to mental health care among prostate cancer survivors, and this will be examined in another study. Up to 2 in 5 participants experienced probable depression in the first 2 years after their cancer diagnosis. This is consistent with findings from studies on patients with other types of cancer and suggests a high need for depression care in recently diagnosed survivors. In addition, the annual prevalence of probable depression between the fifth and seventh years (ie, 20%; see Fig.) is similar to the post–cancer treatment prevalence of depression in the prostate cancer literature (ie, 18%; 95% confidence interval [CI], 15%-22%). This finding suggests that the prevalence of depression among prostate cancer survivors remains stable from 5 years after the cancer diagnosis. Moreover, the initial downward trend in the annual prevalence of probable depression may be explained by developing or peaked psychological resilience, which has been shown to protect prostate cancer survivors from depression. The association between adherence to WHO's exercise recommendations and probable depression is consistent with the literature. The American Cancer Society's prostate cancer survivorship guideline promotes regular exercise and lists benefits that are expected to improve the survivorship experience (eg, lower risks of prostate cancer recurrence, fatigue, and anxiety). The survivorship guideline recommends regular patient-provider conversations about exercise. However, available evidence suggests that many providers fail to discuss exercise with their patients, and this inaction among cancer care providers should be discouraged. The American Cancer Society's prostate cancer survivorship guideline also encourages providers to screen for depression in survivors at risk for depression. Indicated risk factors include being unmarried, low education, advanced prostate cancer, low physical or cognitive functioning, younger age, medical comorbidities, psychiatric history, and poor coping skills. This study presents supportive evidence for some indicated risk factors (ie, young age, medical comorbidities, and psychiatric history). However, other risk factors identified in this study (ie, African American race, unemployment, low annual income, treatment decisional regret, and nonadherence to WHO's exercise recommendations) should be considered for inclusion in the guideline. An unemployed African American participant who earns less than $20,000 per year, has treatment decisional regret, and is nonadherent to exercise recommendations faces a 70% chance of probable depression (95% CI, 58%-80%) over a 12-month period. However, because of the low depression screening rate among men in the general population (4%-8%) and the rate of clinical recognition of depression among nonmental health providers (36%-47%), depression in this hypothetical participant is likely to remain undiagnosed. Lastly, to the best of our knowledge, this study is the first to demonstrate an association between treatment decisional regret and depression. Treatment decisional regret affects 4% to 18% of prostate cancer survivors in the near term, and emerging evidence suggests that its association with depression is due to repetitive negative thinking. Available evidence also suggests that treatment decisional regret is likely to occur in prostate cancer survivors who assume a passive role in cancer treatment decision making. Hence, preventing future depression may be an additional motivating factor for active participation in cancer treatment decision making. Unequal access to mental health care may explain the association between race and probable depression. Evidence from studies in the general population have shown that the incidence of depression is identical in African Americans and white Americans and that African Americans have poorer access to mental health care in comparison with white Americans. Appropriate depression care promotes recovery and prevents relapse/recurrence of depression; thus, limited access to mental health care may make African American prostate cancer survivors more vulnerable to depression. However, little is known about access to mental health care among prostate cancer survivors, and this will be examined in another study. Up to 2 in 5 participants experienced probable depression in the first 2 years after their cancer diagnosis. This is consistent with findings from studies on patients with other types of cancer and suggests a high need for depression care in recently diagnosed survivors. In addition, the annual prevalence of probable depression between the fifth and seventh years (ie, 20%; see Fig.) is similar to the post–cancer treatment prevalence of depression in the prostate cancer literature (ie, 18%; 95% confidence interval [CI], 15%-22%). This finding suggests that the prevalence of depression among prostate cancer survivors remains stable from 5 years after the cancer diagnosis. Moreover, the initial downward trend in the annual prevalence of probable depression may be explained by developing or peaked psychological resilience, which has been shown to protect prostate cancer survivors from depression. The association between adherence to WHO's exercise recommendations and probable depression is consistent with the literature. The American Cancer Society's prostate cancer survivorship guideline promotes regular exercise and lists benefits that are expected to improve the survivorship experience (eg, lower risks of prostate cancer recurrence, fatigue, and anxiety). The survivorship guideline recommends regular patient-provider conversations about exercise. However, available evidence suggests that many providers fail to discuss exercise with their patients, and this inaction among cancer care providers should be discouraged. The American Cancer Society's prostate cancer survivorship guideline also encourages providers to screen for depression in survivors at risk for depression. Indicated risk factors include being unmarried, low education, advanced prostate cancer, low physical or cognitive functioning, younger age, medical comorbidities, psychiatric history, and poor coping skills. This study presents supportive evidence for some indicated risk factors (ie, young age, medical comorbidities, and psychiatric history). However, other risk factors identified in this study (ie, African American race, unemployment, low annual income, treatment decisional regret, and nonadherence to WHO's exercise recommendations) should be considered for inclusion in the guideline. An unemployed African American participant who earns less than $20,000 per year, has treatment decisional regret, and is nonadherent to exercise recommendations faces a 70% chance of probable depression (95% CI, 58%-80%) over a 12-month period. However, because of the low depression screening rate among men in the general population (4%-8%) and the rate of clinical recognition of depression among nonmental health providers (36%-47%), depression in this hypothetical participant is likely to remain undiagnosed. Lastly, to the best of our knowledge, this study is the first to demonstrate an association between treatment decisional regret and depression. Treatment decisional regret affects 4% to 18% of prostate cancer survivors in the near term, and emerging evidence suggests that its association with depression is due to repetitive negative thinking. Available evidence also suggests that treatment decisional regret is likely to occur in prostate cancer survivors who assume a passive role in cancer treatment decision making. Hence, preventing future depression may be an additional motivating factor for active participation in cancer treatment decision making. This study has several strengths. Several clinically relevant factors (eg, depression history, comorbidities, and cancer stage) were controlled in all regression models. In addition, the application of sampling weights makes study findings generalizable to prostate cancer survivors in North Carolina. However, the generalizability of study findings to all prostate cancer survivors in the United States remains uncertain. The distributions of prostate cancer survivors by age and race during enrollment in the Surveillance, Epidemiology and End Results program (2011-2015) and PCaP (2004-2007) are similar in the 2 data sets (Table). Any differences may be driven by the relative sample sizes or an earlier age at cancer diagnosis for PCaP participants. Abbreviations: PCaP, North Carolina–Louisiana Prostate Cancer Project; SEER, Surveillance, Epidemiology, and End Results. P values were estimated with binomial tests of proportions. The identification strategy for depression (SF-12 MCS ≤48.9) is imperfect (sensitivity, 74%; specificity, 83%). Hence, the false-positives and false-negatives in the data set may bias regression estimates toward the null or increase variances and risks of type II errors in explanatory variables. This risk of a type II error may affect the expected association between employment (vs retirement) and probable depression (odds ratio, 1.28; P =.052; see Table). However, study findings are likely to remain robust if a diagnostic instrument such as Patient Health Questionnaire 9 is used to identify depressed study participants (sensitivity, 80%; specificity, 92%). New episodes of probable depression could not be teased apart from recurrence/relapses, nor could anxiety disorders be isolated from probable depression. These limitations preclude accurate measurement of the annual incidence of depression in the sample. Also, the study sample did not include prostate cancer survivors with late-stage cancer, so the study findings do not extend to late-stage disease. Lastly, the identification strategy for depression prevents the separation of anxiety disorders from probable depression or new cases from recurrences and relapses. These limitations prevent precise measurement of the annual incidence of probable depression among study participants, which could be used to simulate the natural history of depression in hypothetical prostate cancer survivors via Markov/microsimulation models. However, a conservative estimate was derived by the conversion of the 5-year cumulative incidence of probable depression between the third and seventh years after the prostate cancer diagnosis (ie, when the annual prevalence of probable depression appeared stable; see Fig.) into an annual incidence with a standard approach (ie, the proportion of incidental true-positive cases [n1 = 154] and incidental false-negative cases [n2 = 62] among at-risk study participants [N = 575] is (n1 + n2)/N or 216/575 or 37.6% over a 5-year period, which translates into 9.0% per year [95% CI, 7.9%-10.2%] under the constant incidence assumption). This conservative estimate of the annual incidence of probable depression may approximate the true annual incidence of depression because it is approximately 5 times the annual incidence of depression in Canadian men aged 65 years and older (ie, 1.8%), approximately 6 to 8 times the annual incidence of depression in Swedish men aged 70 to 85 years (ie, 1.2%), and consistent with the cancer literature (depression is up to 6 times more common in patients with cancer in comparison with the general population). However, the true annual incidence of depression in prostate cancer survivors may be lower than 9.0% because the estimated cumulative incidence may have inadvertently included a few recurrent cases. Conversely, the true annual incidence of depression may be higher than 9.0% because study participants who were lost to follow-up had fewer opportunities to be identified as true-positive cases. Nevertheless, 9.0% seems to be a more plausible estimate than 16% to 17%, which was obtained from inpatient samples of prostate cancer survivors with advanced disease. In conclusion, depression is a major challenge for prostate cancer survivors, particularly in the first 5 years after their cancer diagnosis. Risk factors for depression include African American race, unemployment, low annual income, relatively young age, recency of cancer diagnosis, past depression, comorbidities, treatment decisional regret, and nonadherence to WHO's exercise recommendations. Publisher Copyright: © 2019 American Cancer Society
PY - 2019/10/1
Y1 - 2019/10/1
N2 - Background: The early diagnosis and treatment of depression are cancer care priorities. These priorities are critical for prostate cancer survivors because men rarely seek mental health care. However, little is known about the epidemiology of depression in this patient population. The goal of this study was to describe the prevalence and predictors of probable depression in prostate cancer survivors. Methods: The data were from a population-based cohort of North Carolinian prostate cancer survivors who were enrolled from 2004 to 2007 in the North Carolina–Louisiana Prostate Cancer Project (n = 1031) and were prospectively followed annually from 2008 to 2011 in the Health Care Access and Prostate Cancer Treatment in North Carolina study (n = 805). Generalized estimating equations were used to evaluate an indicator of probable depression (Short Form 12 mental composite score ≤48.9; measured at enrollment and during the annual follow-up) as a function of individual-level characteristics within the longitudinal data set. Results: The prevalence of probable depression fell from 38% in the year of the cancer diagnosis to 20% 6 to 7 years later. Risk factors for probable depression throughout the study were African American race, unemployment, low annual income, younger age, recency of cancer diagnosis, past depression, comorbidities, treatment decisional regret, and nonadherence to exercise recommendations. Conclusions: Depression is a major challenge for prostate cancer survivors, particularly in the first 5 years after the cancer diagnosis. To the authors' knowledge, this is the first study to demonstrate an association between treatment decisional regret and probable depression.
AB - Background: The early diagnosis and treatment of depression are cancer care priorities. These priorities are critical for prostate cancer survivors because men rarely seek mental health care. However, little is known about the epidemiology of depression in this patient population. The goal of this study was to describe the prevalence and predictors of probable depression in prostate cancer survivors. Methods: The data were from a population-based cohort of North Carolinian prostate cancer survivors who were enrolled from 2004 to 2007 in the North Carolina–Louisiana Prostate Cancer Project (n = 1031) and were prospectively followed annually from 2008 to 2011 in the Health Care Access and Prostate Cancer Treatment in North Carolina study (n = 805). Generalized estimating equations were used to evaluate an indicator of probable depression (Short Form 12 mental composite score ≤48.9; measured at enrollment and during the annual follow-up) as a function of individual-level characteristics within the longitudinal data set. Results: The prevalence of probable depression fell from 38% in the year of the cancer diagnosis to 20% 6 to 7 years later. Risk factors for probable depression throughout the study were African American race, unemployment, low annual income, younger age, recency of cancer diagnosis, past depression, comorbidities, treatment decisional regret, and nonadherence to exercise recommendations. Conclusions: Depression is a major challenge for prostate cancer survivors, particularly in the first 5 years after the cancer diagnosis. To the authors' knowledge, this is the first study to demonstrate an association between treatment decisional regret and probable depression.
KW - depression
KW - health disparity
KW - predictors
KW - prostate cancer
KW - risk factors
UR - http://www.scopus.com/inward/record.url?scp=85068161713&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85068161713&partnerID=8YFLogxK
U2 - 10.1002/cncr.32338
DO - 10.1002/cncr.32338
M3 - Article
C2 - 31246284
AN - SCOPUS:85068161713
VL - 125
SP - 3418
EP - 3427
JO - Cancer
JF - Cancer
SN - 0008-543X
IS - 19
ER -