Variability in Depressive Symptoms of Cognitive Deficit and Cognitive Bias During the First 2 Years After Diagnosis in Australian Men With Prostate Cancer

Abstract

The incidence and contribution to total depression of the depressive symptoms of cognitive deficit and cognitive bias in prostate cancer (PCa) patients were compared from cohorts sampled during the first 2 years after diagnosis. Survey data were collected from 394 patients with PCa, including background information, treatments, and disease status, plus total scores of depression and scores for subscales of the depressive symptoms of cognitive bias and cognitive deficit via the Zung Self-Rating Depression Scale. The sample was divided into eight 3-monthly time-since-diagnosis cohorts and according to depression severity. Mean scores for the depressive symptoms of cognitive deficit were significantly higher than those for cognitive bias for the whole sample, but the contribution of cognitive bias to total depression was stronger than that for cognitive deficit. When divided according to overall depression severity, patients with clinically significant depression showed reversed patterns of association between the two subsets of cognitive symptoms of depression and total depression compared with those patients who reported less severe depression. Differences in the incidence and contribution of these two different aspects of the cognitive symptoms of depression for patients with more severe depression argue for consideration of them when assessing and diagnosing depression in patients with PCa. Treatment requirements are also different between the two types of cognitive symptoms of depression, and several suggestions for matching treatment to illness via a personalized medicine approach are discussed.

Keywords

maturation development and aging depression mental health prostate cancer

It has been suggested that up to one fourth of all men who experience prostate cancer (PCa) may also experience depression, which may be associated with personal discomfort (Kronenwetter et al., 2005; Kunkel, Bakker, Myers, Oyesanmi, & Gomella, 2000; Sharpley, Bitsika, & Christie, 2008), greater likelihood of admission to emergency treatment, hospitalization, outpatient visits, death (Van Gastel, Schotte, & Maes, 1997), and suicide (Fang et al., 2010), most often during the first 2 years after diagnosis. However, despite the urgency of these findings, psychological (Carlson, Speca, Patel, & Goodey, 2003; Weber et al., 2004) and exercise-based (Monga et al., 2007) interventions aimed at reducing depression in patients with PCa generally report few significant effects and only minor (<10%) increases in quality of life, leading Chambers, Pinnock, Lepore, Highes, and O’Connell (2011) to conclude that there was “little evidence of quality improvement over time” (p. 87) from such intervention studies. These unimpressive outcomes might be improved by approaches that included more targeted diagnosis–treatment planning protocols based on consideration of the differences in depressive symptomatology reported by patients, plus individualized treatments that focus on that specific symptomatology, neurobiological pathways, and behavioral manifestations (Sharpley & Bitsika, 2013).

This perspective is supported by the finding that although there are nine diagnostic criteria for major depressive disorder (MDD) plus an impairment criterion (American Psychiatric Association, 2013), there are several alternatives within some of those criteria. This range of criteria can produce variations in symptom profiles between participants. In numerical terms, Ostergaard, Jensen, and Bech (2011) noted that there were 1,497 possible combinations of the diagnostic criteria for MDD that fulfill a diagnosis for that disorder. From that basis, there has been a recent suggestion to investigate different depression “subtypes” that represent discrete but coherent combinations of MDD symptoms (Insel, 2013). Baumeister and Parker (2012) reported 15 such depression subtypes that were based on MDD symptoms, etiology, time of onset, gender, and treatment resistance that were collected from 754 reviews of depression. The heterogeneity of MDD has been identified in other reviews (e.g., Ghaemi & Vöhringer, 2011; Luyten, Blatt, van Houdenhove, & Corveleyn, 2006; Parker, 2005), thus suggesting further considerations of how depression might be most effectively diagnosed as a series of subtypes of depression rather than a simple “MDD present/absent” categorization system. While there are many of these subtypes of depression, two which have the potential to disturb PCa patients’ decision making regarding treatment adherence and general self-management are those that are associated with the depressive symptoms of cognitive deficit and cognitive bias.

The specific depressive symptoms of cognitive deficit that are described in MDD are “diminished ability to think or concentrate,” plus an “inability to make decisions,” with associated symptoms of being easily distracted, difficulty remembering things, and impairments in completing complex intellectual tasks that were previously done successfully. It is not difficult to imagine how these cognitive deficits might adversely impinge on the depressed PCa patient’s ability to continue working, manage business and home finances, and make the most effective decisions regarding their treatment for PCa. In addition to this cognitive deficit, patients who suffer these symptoms of MDD may also often report a cognitive bias toward negative interpretation of information they receive (Ellis & Ashbrook, 1988).

These distinctions between the depressive symptoms of cognitive deficit and cognitive bias are relevant to treatment planning for depressed PCa patients because they can require different treatment approaches. For example, pharmacological treatments for cognitive deficits may be aimed at enhancing sleep quality and normalizing circadian rhythm in order to reestablish normal cognitive functioning and memory (Fu et al., 2004). Alternately, more psychotherapeutically and behaviorally focused treatments, such as “attentional bias modification” can be used to train depressed persons to reduce their cognitive bias toward negative stimuli and also overall depressive symptoms (Wells & Beevers, 2010).

Because each of these two sets of the depressive symptoms of cognitive deficit and cognitive bias requires different treatment approaches, it is important to match treatment to disease by careful monitoring of depressive symptoms of cognitive deficit and cognitive bias during the period when men with PCa are most vulnerable to depression and the aversive side effects described in the beginning of this article. That is, changes in the strength with which these two subtypes of depression influence overall depressive status can inform treatment choices during the first 2 years after diagnosis and may represent a pathway toward personalized medicine (Reaven et al., 2009) approaches for PCa patients. While that is a laudable goal, its rationale is based on data collected on the depressive symptoms of cognitive deficit and cognitive bias from the first 2 years after diagnosis. A search of the literature in June, 2014 indicated that no such data had been reported for men with PCa. This study was designed to compare the prevalence of the MDD depressive symptoms of cognitive deficit and cognitive bias in a sample of men with PCa during the first 2 years after diagnosis of PCa and to determine which of these sets of depressive symptoms contributed most to the overall depressive status of those men. Because this study is focused on the symptomatology for MDD, wider definitions and assessments of more general cognitive deficit and cognitive bias were excluded.

Method

Participants

A total of 394 complete and usable questionnaires were collected from a survey of patients with PCa in Brisbane, Australia, who had received their diagnosis of PCa within 25 months of the date of the survey. This response rate is 51%, and is similar to those in several previous survey studies of this population (Sharpley, Bitsika, & Christie, 2012; Sharpley, Bitsika, & Christie, 2013; Sharpley & Christie, 2007). All participants had cancers limited to the primary site and regional draining lymph nodes using conventional staging investigations. Treatments included radiotherapy, plus hormone therapy and surgery when required.

Materials

Background Questionnaire

The background questionnaire related to age, time since diagnosis (calculated from month and year of first diagnosis), treatments previously received and currently received, and present status of their cancer.

Depression

The Zung Self-Rating Depression Scale (SDS; Zung, 1965) is a standardized paper-and-pencil test of depression that has been used in studies of depression in patients with PCa (Sharpley, Bitsika, & Christie, 2009a, 2009b). The SDS includes all of the current criteria for MDD (Zung, 1965, 1973) and respondents indicate the frequency for each of the depressive symptoms contained in 20 items by choosing one response from: “None or a little of the time,” “Some of the time,” “Good part of the time,” or “Most or all of the time.” Several positively worded items act as lie scales and these items are reverse-scored when calculating the total scale score. Raw scores range from 20 to 80, with higher scores being indicative of more severe depression. The SDS has demonstrated split-half reliability of .81 (Zung, 1965), .79 (DeJonge & Baneke, 1989), and .94 (Gabrys & Peters, 1985) and internal consistency (alpha) of .88 for depressed patients and .93 for nondepressed patients (Schaefer et al., 1985), and .84 and .83 for previous Australian PCa samples (Sharpley et al., 2009b; Sharpley & Christie, 2007). The SDS is one of the four most widely used scales of depression (Shafer, 2006) and has been reported to be superior to the MMPI Depression Scale and the Beck Depression Inventory for assessing depression in male psychiatric inpatients (Schaefer et al., 1985). It has sensitivity of 93% in predicting depression validated via SCID clinical interview (Agrell & Dehlin, 1989), and raw scores of 40 or above indicate the presence of “clinically significant depression” (Zung, 1973, p. 335). SDS raw scores were used in this study.

Although the SDS total score represents the overall depressive status of respondents, Items 11 (“My mind is as clear as it used to be”) and 12 (“I find it easy to do the things I used to do”) assess the presence of depressive symptoms of cognitive deficit; Items 14 (“I feel hopeful about the future”), 17 (“I feel that I am useful and needed”), and 18 (“My life is pretty full”) measure the respondent’s depressive symptoms of cognitive bias. The mean scores on each of these two subsets of SDS items may be used to determine a particular patient’s status on these two aspects of the depressive symptoms of MDD. The mean scores for each of the two subsets of items that comprised the two key MDD criteria had a range of 1 to 4.

Ethical approval for this study was obtained from the Wesley Human Research Ethics Committee, Brisbane.

Statistical Analysis

Data were analyzed via SPSS 20. After checking for normality, the distribution of background, disease, and treatment variables was undertaken. Associations between background and other variables with the two sets of the depressive symptoms of cognitive deficit and total SDS score were tested via Spearman and Pearson correlations. The presence of significant differences in mean scores for the two sets of the depressive symptoms was determined via t tests. Linear regression was used to measure the relative contributions made by the two sets of cognitive symptoms of depression to total SDs score for each of eight time-since-diagnosis cohorts described below.

Results

Demographic Data

Normality was acceptable for SDS data. The distributions of demographic, treatment, and disease status variables are reported in Table 1. There were neither any significant Pearson correlations between age or time since diagnosis and total SDS score or either of the two cognitive subsets nor any significant Spearman correlations between past treatment or disease state and SDS total score for the entire sample, nor within any of the eight time-since-diagnosis cohorts described below. Although past hormone treatment was not significantly correlated to SDS total score (r = .070, ns) or either of the two sets of cognitive symptoms, current hormone treatment was significantly and positively correlated with SDS score (r = .191, p < .001) and with the depressive symptoms of cognitive deficit (r = .134, p < .01) but not with the depressive symptoms of cognitive bias (r = .091, ns). As expected, depressive symptoms of cognitive deficit and cognitive bias were significantly correlated with each other (r = .514, p < .001) and with SDS total score (cognitive deficit, r = .756, p < .001; cognitive bias, r = .771, p < .001). Internal consistency (Cronbach alpha) was .804 for the SDS.

Table 1.

Demographic Data.

Variable	Sample characteristics
Age in years (n = 394)	M = 67.46 (SD = 6.88), range = 27-87
Time since diagnosis in months	M = 10.01 (SD = 5.41), range = 1-24

Variable	n	%
Living situation (n = 394)
With wife/partner	348	88.3
Widowed	16	4.1
Separated/divorced	21	5.3
Never married	9	2.3
Past treatment (n = 395)^a
Radiotherapy	152	38.5
Surgery	75	19.0
Hormone therapy	129	32.7
No treatment	147	37.2
Current treatment (n = 364)
Radiotherapy	95	26.1
Surgery	9	2.6
Hormone therapy	137	37.6
No treatment	123	33.7
Present status (n = 378)
Cancer still present, undergoing treatment	250	66.2
In remission (no signs)	118	31.2
Cancer recurring after previous treatment	10	2.6

May have reported nil or multiple responses.

To test for possible variation over time since diagnosis, all participants were classified according to 3-monthly cohorts. Figure 1A reports the mean scores (x/4) for each of the two sets of depressive symptoms of cognitive deficit and cognitive bias over the 24-month period of the eight time-since-diagnosis cohorts. However, although there was a significant correlation between the two sets of depressive symptoms of cognitive deficit and cognitive bias, that relationship accounted for only 26.4% of the variance, not as strong as the relationships between each of the two sets of depressive cognitive symptoms and the total SDS score (which ranged from 57.2% to 59.4% of the variance). Furthermore, there was a significant difference between scores for the two sets of depressive cognitive symptoms—t(388) = 8.354, p < .01—arguing that, despite some level of agreement with each other and with total depression scores, participants’ scores on these two sets of depressive cognitive symptoms were not completely congruent (scores on the symptoms of cognitive deficit were significantly higher than those for cognitive bias). This suggests that these two cognitive depression variables were measuring aspects of depression that were related but not identical, with nearly 75% of the variance of each being unaccounted for by the other.

Figure 1.

(A) Mean scores and (B) standardized beta weights for cognitive subscales of the Self-Rating Depression Scale (SDS) over eight time-since-diagnosis cohorts: all patients.

Influence of Depressive Symptoms of Cognitive Deficit and Cognitive Bias on Total Depression

To assess the relative contribution made by the two sets of depressive symptoms of cognitive deficit and cognitive bias, a series of linear regressions was conducted, breaking the sample into eight 3-monthly time-since-diagnosis cohorts from Month 0 to Month 24 after diagnosis (sample sizes of these cohorts ranged from 15 to 96). Current hormone treatment was controlled for, and depressive symptoms of cognitive deficit and cognitive bias were entered into the regression equation on SDS total score within each time-since-diagnosis cohort. Figure 1B presents the standardized beta coefficients for the sets of depressive symptoms of cognitive deficit and cognitive bias and demonstrates that the influence of these two subscales of the SDS on the total SDS score varied over the eight time-since-diagnosis cohorts.

For the period from 1 to 3 months after diagnosis until 7 to 9 months after diagnosis, the influence of cognitive bias symptoms of depression on total SDS score was stronger than the influence of cognitive deficit symptoms of depression. However, while the former gradually decreased in power following that period, the latter gradually increased in power until they crossed over at the 10 to 12 months time-since-diagnosis cohort. Then, the two sets of depressive symptoms separated until 19 to 21 months after diagnosis when they again crossed over and remained close together until 24 months after diagnosis.

Effects of Severity of Depression Diagnosis

Although Figure 1B depicts the association between the two sets of cognitive symptoms of depressive and total depression scores for the sample, it does not provide any indication of the effects of depression severity on that association. Therefore, patients whose depression was classified as clinically significant (defined as SDS total raw scores >39; Zung, 1973, p. 335; n = 108) versus those whose SDS scores were indicative of less severe depression were compared by repeating the linear regressions for these subsamples of patients separately, with current hormone treatment controlled for in the analysis.

Figure 2A reports the standardized beta weights for the relative contributions of depressive symptoms of cognitive deficit and cognitive bias to total SDS score for patients with clinically significant depression and Figure 2B reports the same results for patients whose SDS scores were indicative of less severe depression. The two graphs in Figure 2 report quite different patterns of association between the two sets of depressive cognitive symptoms and total SDS scores. For those patients whose depression was more severe (Figure 2A), there was an ongoing trend for the two sets of cognitive depression symptoms to diverge over time, with depressive symptoms of cognitive bias increasing in their contribution to total depression score over the 18 months after diagnosis and depressive symptoms of cognitive deficit decreasing in their contribution to total SDS score over that period, although the period of comparison is less than that presented in Figure 1 because of low patient numbers. By comparison, for patients with less severe depression (Figure 2B), the contributions of the two sets of cognitive symptoms of depression to total SDS score ran fairly much in parallel for most of the period under scrutiny.

Figure 2.

Standardized beta weights for cognitive subscales of the Self-Rating Depression Scale (SDS) over six time-since-diagnosis cohorts: (A) patients with clinically significant depression and (B) patients without clinically significant depression.

Discussion

Although these results do not represent the depressive symptoms of cognitive deficit and cognitive bias collected from the same patients over time via a prospective model, they do refer to data on these aspects of depression collected from eight time-since-diagnosis cohorts of PCa drawn from the first 2 years after diagnosis and thus provide some comparisons of the differences between those cohorts. Overall, there were significantly higher mean scores for the depressive symptoms of cognitive deficit than for cognitive bias during the first 2 years after diagnosis, which is a new finding in this literature and suggestive that diagnosis and analysis of depression in patients with PCa might benefit from consideration of the relative occurrence of each of these two sets of cognitive symptoms of depression rather than simply referring to the total MDD-related score on a self-report inventory.

The second major finding was the way that these two sets of cognitive symptoms of depression were associated with the total depression score on the SDS. For the total sample and those depicted in Figure 1B, depressive symptoms of cognitive bias were more powerfully associated with total depression status during Months 1 to 9 and 13 to 18. However, it is notable that the overlaps that occurred at the 10- to 12- month and 19- to 24-month time periods reflected a decrease in the relative power of the depressive symptoms of cognitive bias as much as an increase in the relative power of the depressive symptoms of cognitive deficit. There appears to be a degree of synchrony in Figure 1B, with a possible reciprocal relationship between scores for cognitive deficit and cognitive bias, which will be discussed in more detail below.

The relationship between the two sets of cognitive symptoms of depression and total depression was much clearer when the sample was split according to depression severity. For PCa patients with clinically significant depression (Figure 2A), the depressive symptoms of cognitive bias had a much stronger association with total depression than those for cognitive deficit. By contrast, for less depressed patients (Figure 2B), cognitive deficit was more powerfully associated with total depression scores. There was also a difference in the nature of the association between the two sets of cognitive depression and total depression across the two depression severity subsamples. That is, although those relationships ran fairly parallel to each other during the first 18 months after diagnosis for PCa patients with less severe depression (Figure 2B), for patients with clinically significant depression, the association of cognitive bias gradually increased over time but the association of cognitive deficit gradually decreased over time. The latter finding is of most relevant for clinical treatment planning because it is these more severely depressed patients with PCa who will have greatest need for systematic and careful therapy, whereas the less depressed patients might benefit from less intensive treatments, such as bibliotherapy or group support meetings.

According to the author of the SDS, PCa patients whose SDS score is higher than 39 require treatment planning and delivery (Zung, 1973), and these results suggest that the association between total depression scores and the depressive symptoms of cognitive bias is stronger than that for the depressive symptoms of cognitive deficit in these more depressed PCa patients. These men would be expected to (as indicated by the relevant SDS items) feel hopeless about their future, believe that they are not useful to others and needed by them, and that their lives are empty. PCa patients who present with a strong cognitive bias toward negativity need assistance to help them alter that negative bias and develop a more optimistic perspective on their illness and its treatment. However, that therapy process may require treatment focused on cognitive deficit first.

It has been suggested that impairments in an individual’s ability to focus on positive stimuli and emotions (i.e., cognitive bias) may be explained by the “resource-allocation” hypothesis. That is, when they are already somewhat sad or depressed, individuals require more cognitive effort to focus on pleasant stimuli than on negative stimuli, thus leading them to be more likely to experience the consequent negative emotions that arise from focussing on negative stimuli (Ellis & Ashbrook, 1988). This imbalance in ability to focus on both negative and positive stimuli can result from impairment in the prefrontal cortex (PFC) activity of rational decision making, plus hyperactivation of the amygdala, which together lead to negative emotional bias in response to events (Siegle, Thompson, Carter, Steinhauer, & Thase, 2007). Initial treatments for these alterations in PFC and amygdala activity may include pharmacological agents, such as fluoxetine (Mayberg et al., 2000), paroxetine (Kennedy et al., 2001), and sertraline (Drevets, GBogers, & Raichle, 2002), which enhance sleep quality and normalize circadian rhythm, reversing the underactivation in PFC regions associated with cognitive function and helping to reduce overactivation of amygdala functions associated with emotional function. Nonpharmacological approaches for the initial stage of treatment for PCa patients with severe depression and high cognitive bias symptoms of depression could include attentional bias modification, which trains participants to attend to the position of a visually presented stimulus rather than the negative content of that visual stimulus and has been successful in reducing attentional bias toward negative stimuli and also overall depressive symptoms in a treatment versus a placebo comparison (Wells & Beevers, 2010). In a review of pharmacological treatments and attentional training, Browning, Holmes, and Harmer (2010) concluded that the former influenced the initial deployment of attention via the amygdala toward the negative stimulus, whereas the latter influenced later attention by the PFC on the negative stimulus. Other methods of reestablishing prefrontal and amygdala balance include functional magnetic resonance imaging live feedback (for a review, see Weiskopf et al., 2004), which has been used to teach self-control of blood flow to the hippocampus and limbo-thalamic-cortical pathway (Yoo, Lee, O’Leary, Panych, & Jolesz, 2008), amygdala, insula, and ventral striatum (Johnston, Boehm, Healy, Goebel, & Linden, 2010)), and EEG-based visual feedback to activate those regions of the PFC that are less active in depressed states and to simultaneously reduce activity in those PFC regions that are more active in depressed individuals (Hammond, 2005, 2011).

Limitations of this study include sampling from a single treatment center in a particular nation, use of a standardized test of depression rather than a clinical interview (although there are considerable data that support the validity of the former on the latter), inclusion of a sample of older men who may have other age-related factors that contributed to their depressive status, and testing of patients at a single point in their treatment regime rather than via a prospective-designed study. However, because patients in the sample failed to report any significant correlation between most of the background factors (age, time-since-diagnosis, cancer status, and past treatment type) and SDS scores or the cognitive depression subscale scores, the latter limitation may not represent a serious potential source of error. The significant correlation between current hormone treatment and SDS was controlled for in the regression analyses. The sample was self-selected from a mail-out invitation to participate and there are no data regarding the depressive status of nonparticipants. The Zung SDS items that measure depressive symptoms of cognitive deficit and cognitive bias represent those MDD symptoms as they are defined in the Diagnostic and Statistical Manual of Mental Disorders–Fifth Edition (DSM-5) but do not attempt to represent the wider definitions of cognitive deficit and cognitive bias that may appear in the psychological literature. Although data from measures of those entire constructs might be of interest, this investigation was purposely focused on the ways that cognitive deficit and cognitive bias manifested themselves within the overall MDD symptomatology rather than in the entire range of cognitive functioning per se. Finally, although the symptoms of MDD that were used in this study were those defined in the DSM-5, there has been some suggestion that men may experience depression differently to women and that the symptoms of irritability and risk taking may figure more largely in males’ depressive symptom profiles than in females’ (Innamorati et al., 2011; Rutz, von Knorring, Pihlgren, Rihmer, & Wålinder, 1995; Zierau, Bille, Rutz, & Bech, 2002). This aspect of the measurement process for MDD might be investigated in future studies.

In conclusion, PCa patients in this sample demonstrated variability in the contributions of depressive symptoms of cognitive bias versus the depressive symptoms of cognitive deficit to total depression in discrete cohorts sampled from the first 18 months after diagnosis. Because each of these subsets of the cognitive symptoms of depression has different MDD symptomatologies, neurological pathways, and treatment requirements, application of treatments for MDD that occurs in many PCa patients may benefit from consideration of the relative incidence and severity of the depressive symptoms of cognitive bias and cognitive deficit

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Agrell

Dehlin

(1989). Comparison of six depression scales in geriatric stroke patients. Stroke, 20, 1190-1194.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

Baumeister

Parker

(2012). Meta-review of depressive subtyping models. Journal of Affective Disorders, 139, 126-140.

Browning

Holmes

Harmer

(2010). The modification of attentional bias to emotional information: A review of the techniques, mechanisms, and relevance to emotional disorders. Cognitive, Affective, and Behavioral Neuroscience, 10, 8-20.

Carlson

Speca

Patel

Goodey

(2003). Mindfulness-based stress reduction in relation to quality of life, mood, symptoms of stress, and immune parameters in breast and prostate cancer outpatients. Psychosomatic Medicine, 65, 571-581.

Chambers

Pinnock

Lepore

Highes

O’Connell

(2011). A systematic review of psychosocial interventions for men with prostate cancer and their partners. Patient Education and Counseling, 85, e75-e88.

DeJonge

Baneke

(1989). The Zung Self-rating Depression Scale: A replication study on reliability, validity and prediction. Psychological Reports, 64, 833-834.

Drevets

GBogers

Raichle

(2002). Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism. European Neuropsychopharmacology, 12, 527-544.

Ellis

Ashbrook

(1988). Resource allocation model of the effects of depressed mood states upon memory. In Fiedler

Forgas

(Eds.), Affect, cognition, and social behavior: New evidence and integrative attempts (pp. 25-43). Gottingen, Germany: Hogrefe.

10.

Fang

Keating

Mucci

Adami

H-O.

Stampfer

Valdimarsdottir

Fall

(2010). Immediate risk of suicide and cardiovascular death after prostate cancer diagnosis: Cohort study in the United States. Journal of the National Cancer Institute, 102, 307-314.

11.

Williams

Cleare

Brammer

Walsh

Kim

. . .Bullmore

(2004). Attenuation of the neural response to sad faces in major depression by antidepressant treatment: A prospective, event-related functional magnetic resonance imaging study. Archives of General Psychiatry, 61, 877-889.

12.

Gabrys

Peters

(1985). Reliability, discriminant and predictive validity of the Zung Self-Rating Depression Scale. Psychological Reports, 57, 1091-1096.

13.

Ghaemi

S. N.

Vöhringer

P. A.

(2011). The heterogeneity of depression: An old debate renewed. Acta Psychiatrica Scandinavica, 124, 497.

14.

Hammond

(2005). Neurofeedback treatment of depression and anxiety. Journal of Adult Development, 12, 131-137.

15.

Hammond

(2011). What is neurofeedback: An update. Journal of Neurotherapy, 15, 305-336.

16.

Innamorati

Pompili

Gonda

Amore

Serafini

Niolu

. . .Giradi

(2011). Psychometric properties of the Gotland Scale for Depression in Italian psychiatric patients and its utility in the prediction of suicide risk. Journal of Affective Disorders, 132, 99-103.

17.

Insel

(2013). Transforming diagnosis. Retrieved from www.nimh.nih.gov/about/director/2013/transforming-diagnosis-shtml

18.

Johnston

S. J.

Boehm

S. G.

Healy

Goebel

Linden

D. E. J.

(2010). Neurofeedback: A promising tool for the self-regulation of emotion networks. NeuroImage, 49, 1066-1072.

19.

Kennedy

Evans

Kruger

Mayberg

Meyer

McCann

. . .Vaccarino

(2001). Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. American Journal of Psychiatry, 158, 899-905.

20.

Kronenwetter

Weidner

Pettengill

Marlin

Crutchfield

McCormac

. . .Ornish

(2005). A qualitative analysis of interview of men with early stage prostate cancer: The prostate cancer lifestyle trial. Nursing, 28(2), 99-107.

21.

Kunkel

E. J. S.

Bakker

J. R.

Myers

R. E.

Oyesanmi

Gomella

L. G.

(2000). Biopsychosocial aspects of prostate cancer. Psychosomatics, 41(2), 85-94.

22.

Luyten

Blatt

van Houdenhove

Corveleyn

(2006). Depression research and treatment: asre we skating to where the puck is going to be? Clinical Psychology Review, 26, 985-999.

23.

Mayberg

Brannan

Tekell

Silvab

Mahurina

McGinnisa

Jerabeka

(2000). Dysfunction in neural circuits involved in the pathophysiology of mood disorders: Regional metabolic effects of fluoxetine in major depression: Serial changes and relationship to clinical response. Biological Psychiatry, 48, 830-843.

24.

Monga

Garber

Thornby

Vallbona

Kerrigan

Monga

Zimmerman

(2007). Exercise prevents fatigue and improves quality of life in prostate cancer patients undergoing radiotherapy. Archives of Physical Medicine and Rehabilitation, 88, 1416-1422.

25.

Ostergaard

Jensen

Bech

(2011). The heterogeneity of the depressive syndrome: When numbers get serious. Acta Psychiatrica Scandinavica, 124, 495-496. doi:10.1111/j.1600-0447.2011.01744.x

26.

Parker

(2005). Beyond major depression. Psychological Medicine, 41, 467-474.

27.

Reaven

Blakeley-Smith

Nichols

Dasari

Flanigan

Hepburn

(2009). Cognitive-behavioral group treatment for anxiety symptoms in children with high-functioning autism spectrum disorders: A pilot study. Focus on Autism and Other Developmental Disabilities, 24, 27-37.

28.

Rutz

von Knorring

Pihlgren

Rihmer

Wålinder

(1995). Prevention of male suicides: Lessons from Gotland study. Lancet, 345, 524.

29.

Schaefer

Brown

Watson

Plenel

DeMotts

Howard

. . .Ballweg

(1985). Comparison of the validities of the Beck, Zung and MMPI depression scales. Journal of Consulting and Clinical Psychology, 53, 415-418.

30.

Shafer

(2006). Meta-analysis of the factor structures of four depression inventories: Beck, CES-D, Hamilton, and Zung. Journal of Clinical Psychology, 62, 123-146.

31.

Sharpley

Bitsika

(2013). Differences in neurobiological pathways of four “clinical content” subtypes of depression. Behavioural Brain Research, 256, 368-376.

32.

Sharpley

Bitsika

Christie

(2008). Psychological distress among prostate cancer patients: Fact or fiction? Clinical Medicine: Oncology, 2, 563-572.

33.

Sharpley

Bitsika

Christie

(2009a). Positive (but not negative) punishment predicts anxiety and depression among prostate cancer patients: An exploration of the behaviour analytic model of depression. Behaviour Change, 26, 225-234.

34.

Sharpley

Bitsika

Christie

(2009b). Understanding the causes of depression among prostate cancer patients: Development of the effects of prostate cancer on lifestyle questionnaire. Psycho-Oncology, 18, 162-168.

35.

Sharpley

Bitsika

Christie

(2012). The role of melancholia in prostate cancer patients’ depression. BMC Psychiatry, 12, 201.

36.

Sharpley

Bitsika

Christie

(2013). Do prostate cancer patients suffer more from depressed mood or anhedonia? Psycho-Oncology, 22, 1718-1723.doi:10.1002/pon.3203.

37.

Sharpley

Christie

(2007). An analysis of the psychometric structure and frequency of anxiety and depression in Australian men with prostate cancer. Psycho-Oncology, 16, 660-667.

38.

Siegle

Thompson

Carter

Steinhauer

Thase

(2007). Increased amygdala and decreased dorsolateral prefrontal BOLD responses in unipolar depression: Related and independent features. Biological Psychiatry, 61, 198-209.

39.

Van Gastel

Schotte

Maes

(1997). The prediction of suicide intent in depressed patients. Acta Psychiatrica Scandinavica, 96, 254-259.

40.

Weber

B. A.

Roberts

Resnick

Deimling

Zauszniewski

Musil

Yarandi

(2004). The effect of dyadic intervention on self-efficacy, social support, and depression for men with prostate cancer. Psycho-Oncology, 13, 47-60.

41.

Weiskopf

Scharnowski

Veit

Goebel

Birbaumer

Mathiak

(2004). Self-regulation of local brain activity using real-time functional magnetic resonance imaging (fMRI). Journal of Physiology, 98, 357-373.

42.

Wells

Beevers

(2010). Biased attention and dysphoria: Manipulating selective attention reduces subsequent depressive symptoms. Cognition & Emotion, 24, 719-728.

43.

Yoo

S.-S.

Lee

J-H.

O’Leary

Panych

L. P.

Jolesz

F. A.

(2008). Neurofeedback fMRI-mediated learning and consolidation of regional brain activation during motor imagery. International Journal of Imaging Systems and Technology, 18(1), 69-78.

44.

Zierau

Bille

Rutz

Bech

(2002). The Gotland Male Depression Scale: A validity scale in patients with alcohol use disorder. Nordic Journal of Psychiatry, 56, 265-271.

45.

Zung

(1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63-70.

46.

Zung

(1973). From art to science: The diagnosis and treatment of depression. Archives of General Psychiatry, 29, 328-337.