Journal of Affective Disorders 167 (2014) 351–357
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Gender differences in depression severity and symptoms across depressive sub-types Gordon Parker a,b,n, Kathryn Fletcher a,b, Amelia Paterson a,b, Josephine Anderson b, Michael Hong b a b
School of Psychiatry, University of New South Wales, BDI Building, Hospital Road, Randwick, 2031 Sydney, Australia Black Dog Institute, Sydney, Australia
art ic l e i nf o
a b s t r a c t
Article history: Received 14 February 2014 Received in revised form 12 June 2014 Accepted 12 June 2014 Available online 19 June 2014
Background: Lifetime rates of depression are distinctly higher in women reflecting both real and artefactual influences. Most prevalence studies quantifying a female preponderance have examined severity-based diagnostic groups such as major depression or dysthymia. We examined gender differences across three depressive sub-type conditions using four differing measures to determine whether any gender differences emerge more from severity or symptom prevalence, reflect nuances of the particular measure, or whether depressive sub-type is influential. Methods: A large clinical sample was recruited. Patients completed two severity-weighted depression measures: the Depression in the Medically Ill 10 (DMI-10) and Quick Inventory of Depressive SymptomsSelf-Report (QIDS-SR) and two measures weighting symptoms and illness correlates of melancholic and non-melancholic depressive disorders - the Severity of Depressive Symptoms (SDS) and Sydney Melancholia Prototype Index (SMPI). Analyses were undertaken of three diagnostic groups comprising those with unipolar melancholic, unipolar non-melancholic and bipolar depressive conditions. Results: Women in the two unipolar groups scored only marginally (and non-significantly) higher than men on the depression severity measures. Women in the bipolar depression group, did however, score significantly higher than men on depression severity. On measures weighted to assessing melancholic and non-melancholic symptoms, there were relatively few gender differences identified in the melancholic and non-melancholic sub-sets, while more gender differences were quantified in the bipolar subset. The symptoms most commonly and consistently differentiating by gender were those assessing appetite/weight change and psychomotor disturbance. Conclusion: Our analyses of several measures and the minimal differentiation of depressive symptoms and symptom severity argues against any female preponderance in unipolar depression being contributed to distinctly by these depression rating measures. Our analyses indicated that gender had minimal if any impact on depression severity estimates. Gender differences in depressive symptoms and severity were more distinctive in bipolar patients, a finding seemingly not previously identified or reported. Limitations: The study had considerable power reflecting large sample sizes and thus risks assigning significant differences where none truly exist, although we repeated analyses after controlling for the type I error rate. & 2014 Elsevier B.V. All rights reserved.
Keywords: Depression Bipolar disorder Epidemiology Gender
1. Introduction As observed by Kessler (2005), one “of the most widely documented findings in psychiatric epidemiology is that women n Corresponding author at: School of Psychiatry, University of New South Wales, BDI Building, Hospital Road, Randwick, 2031 Sydney, Australia. Tel.: þ 61 2 93824372. E-mail address: [email protected] (G. Parker).
http://dx.doi.org/10.1016/j.jad.2014.06.018 0165-0327/& 2014 Elsevier B.V. All rights reserved.
have higher rates of depression than men.” He overviewed a number of studies quantifying a female preponderance ranging from 1.5 to 3 times higher rates of depression. Explanations reviewed by Kessler included artefactual influences as well as ‘real’ biological and psychosocial factors. The former explanation includes women simply being more likely to seek help, acknowledge and over-report depressive symptoms, and the tendency for women to ‘amplify’ and for men to ‘blunt’ reporting of any experienced depressive symptom (Nolen-Hoeksema et al., 1994).
352
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Kessler and most other reviewers favour ‘real’ factors such as hormones and social and gender-roles making greater contribution to the gender difference. However, the extent to which measures of depression include gender-weighted or genderdimorphic features may also influence rates, and at times have an artefactual impact on prevalence estimates. For example, if a depression measure includes depressive symptoms more likely to be experienced by depressed women than men (e.g. crying) then more women will exceed the measure's ‘caseness’ threshold and contribute to differential depression rates. If the female preponderance in depression rates simply and only reflects women being more likely to experience initial depressive episodes than men—which Kessler viewed as the ‘genuine’ explanation—we would expect symptom representation and severity to be comparable across men and women in depressed samples. If the preponderance also reflects differential experience and/or severity in certain symptoms then we would expect women to be more likely to report such symptoms and/or rate them more highly. Piccinelli and Wilkison (2000) reviewed a number of studies and concluded that several symptoms were over-represented in females (i.e. disturbances of appetite and sleep, fatigue, somatic anxiety and hypochondiasis). Most of these might, however, be viewed as indirect expressions or correlates of depression rather than mood-specific features, while Kornstein et al. (2000) also identified psychomotor disturbance—along with sleep changes and anxiety/somatisation—as more prevalent in women. Analysing data from the STARnD study, Marcus et al. (2005) quantified appetite and weight changes, anergia, psychomotor agitation and sympathetic arousal symptoms as over-represented in women. In a study of twins experiencing major depression, Khan et al (2002) quantified female twins as experiencing more fatigue, hypersomnia and psychomotor retardation, while male twins experienced more insomnia and agitation. Most references to a female preponderance in lifetime rates of ‘depression’ refer to severity-based depressive categories, especially major depression but also minor states such as dysthymia. We have long favoured a sub-typing model that seeks to weight categories (e.g. melancholic vs. a heterogeneous non-melancholic group; bipolar vs. unipolar depression—with bipolar depression being more likely to be melancholic in its clinical pattern of symptoms, Parker et al., 2013a). Bipolar disorder lacks any distinctive gender weighting (Goodwin and Jamison, 2007; Parker and Brotchie, 2010) while studies variably suggest that the prevalence of melancholia is comparable in men and women, or somewhat over-represented in women and even somewhat overrepresented in men (Hildebrandt et al., 2003), albeit with any gender weighting being relatively slight. If any depressive subtype has comparable or near comparable rates in men and women, then examining for gender differences in the prevalence and/or severity of constituent symptoms in such sub-types offers an advantage in exploring for any gender differences in their actual measurement. We therefore report a study examining gender differences across bipolar, unipolar melancholic and unipolar nonmelancholic depressive conditions. We hypothesised that if the former two conditions have comparable prevalences for men and women, then gender differentiation in symptoms would be less distinct than for the non-melancholic depressive conditions which we assume underpin the community studies quantifying a female preponderance. We analysed data from four differing depression rating measures, variably rating depression severity and subtyping features, and with multiple measures allowing for consistency of findings to be evaluated. We further hypothesised that women would be more likely to rate higher on depression severity as a consequence of mood amplification differences, but that such an artefactual influence would be less distinct in measures dominated by items examining sub-type symptoms.
2. Material and methods 2.1. General overview Over the 2005–2013 period, patients attending the Black Dog Institute Depression Clinic (a Sydney-based tertiary referral clinic for diagnostic and management advice for mood disorders) were invited to complete a series of questionnaires for research purposes. Written informed consent was obtained in accordance with University of New South Wales Ethics Committee requirements. As part of the routine clinical assessment prior to consultation with a psychiatrist, patients completed the Quick Inventory of Depressive Symptomatology-SelfReport (QIDS-SR; Rush et al., 2003), which quantifies depression severity and includes DSM-IV depression symptoms. Patients additionally completed the Mood Assessment Programme or MAP (Parker et al., 2008), a computerised tool assessing a range of features including socio-demographic, mood disorder and treatment history details. The MAP also contains a number of validated self-report measures of depressive symptomatology, including the Depression in the Medically Ill-10 measure (DMI-10; Parker et al., 2002), the Severity of Depressive Symptoms measure (SDS; Parker et al., 2009; 2010) and the Sydney Melancholia Prototype Index SMPI (Parker et al., 2012, 2013b). All measures are detailed shortly. Following consultation with a psychiatrist, diagnostic information was recorded. Inclusion criteria for the study were: age 15–85, a clinical diagnosis of a mood disorder and its sub-type (unipolar melancholic, unipolar non-melancholic, bipolar I or II disorder), no current psychosis or underlying organic issues, good comprehension of English, and ability to provide written informed consent. A clinical diagnosis of melancholia weighted features such as an anhedonic and non-reactive mood, psychomotor disturbance (and particularly impaired concentration), anergia and diurnal variation (with mood and energy worse in the morning), and was complemented by assessing illness correlates (e.g. a family history of depression, a superior response to any previous antidepressant medication as against any non-drug therapy). A clinical diagnosis of a bipolar disorder required distinct evidence of non-psychotic hypomanic episodes or psychotic manic episodes (assigned as bipolar II and I respectively), with oscillating periods of melancholic or psychotic symptoms during depressive episodes, and again complemented by assessing illness correlates (e.g. a family history of depression or bipolar disorder, a relatively or categorically distinct onset). 2.2. Overview of depression measures 2.2.1. Depression in the Medically Ill-10 (DMI-10) measure The DMI-10 (Parker et al., 2002) was designed to detect state depression severity in patients with medical disorders, and to respect and redress concerns that such patients might score falsely high on somatic symptoms such as insomnia, fatigue and weight loss (due to their medical illness rather than concomitants of depression), and so comprises 10 non-somatic depressed mood constructs. Scores for each item range from 0–3, with total scores ranging from 0–30 (a score of 9 or above being identified as indicative of ‘clinical depression’). In a subsequent study the measure demonstrated high internal consistency (α¼0.92; Hilton et al., 2006), and as it lacks somatic items and weights key depressive mood constructs it has been used as a general measure of depression severity. Any impact of gender on its scoring has not been previously examined. 2.2.2. Quick Inventory of Depressive Symptomatology-Self-Report (QIDS-SR) The QIDS-SR (Rush et al., 2003) assesses the presence and severity of 16 DSM-IV depression-related symptoms, is viewed as
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
353
3. Results
suitable for use with both unipolar and bipolar populations (Bernstein et al., 2010) and includes both melancholic and nonmelancholic symptoms. The measure demonstrates high internal consistency (α ¼86; Rush et al., 2003). Total scores range from 0 to 27, with scores of 11 or above indicative of moderate to severe depression.
3.1. Sample characteristics and measure scoring A total of 2865 participants received a clinical diagnosis (n ¼557 unipolar melancholic, n¼ 807 unipolar non-melancholic, n ¼1501 bipolar) and provided research data — but not all measures were completed by all subjects in this overall sample. Females were slightly over-represented (56.0%), and males were significantly older than females (40.4 vs. 37.2, t¼6.4, Po 0.001). Given the significant age difference, male and female participants were age-matched within each diagnostic category, yielding a subsequent total of 2308 participants (n ¼412 unipolar melancholic, n ¼690 unipolar non-melancholic, n ¼ 1055 bipolar). All subsequent analyses were based on the age-matched sample, and where the mean age was 39.3 (SD ¼ 12.7; age range 15–80). Item scores on each measure were contrasted for males and females via two methods: (i) scores for each item were re-coded to reflect the presence vs. absence of each symptom (0 vs. 1, 2 or 0 vs. 1, 2, 3), and (ii) raw scores for each item were summed to quantify severity of each symptom and, if relevant, total measure scores.
2.2.3. Severity of Depressive Symptoms (SDS) The self-report Severity of Depressive Symptoms (SDS) measure (Parker et al., 2009, 2010) comprises 32 clinical symptoms rated on a 0–3 scale, and again captures symptoms historically weighted to both melancholic and non-melancholic depression. Properties of the measure have been reported previously (Parker et al., 2009; 2010). 2.2.4. Sydney Melancholia Prototype Index (SMPI) The SMPI (previously known as the Self-Report of Depressive Experiences measure or SERDEX; (Parker et al., 2012, 2013a) is a 24-item measure comprising 12 melancholic and 12 nonmelancholic prototypic features (both symptoms and illness correlates) displayed in two columns. Individuals are invited to tick any item from either column that they regard as ‘characteristic’ in terms of their worst depressive episodes. The listed items assess symptoms historically favoured as most differentiating of melancholic and non-melancholic depression, in addition to assessing illness correlates (e.g. pre-morbid interpersonal functioning, distal and proximal stressors, the context and impact of proximal stressors on inducing and maintaining the depression, and trait emotional dysregulation levels) aligned with both melancholic and non-melancholic depression. Each item was selected by considering its utility in differentiating melancholic and nonmelancholic depression in previous studies (e.g., Parker and Hadzi-Pavlovic, 1996; Parker et al., 2010), and tested empirically to quantify their differentiating potential. Properties of the initial self-report measure have been reported previously (Parker et al., 2012, 2013b).
3.2. DMI-10 data Scores on this measure were available for 2130 subjects: 388 (195 male, 193 female) participants diagnosed with unipolar melancholic depression, 644 (323 male, 321 female) with unipolar non-melancholic depression, and 1098 (535 male, 563 female) with bipolar disorder. For this measure we tabulate the prevalence of affirming each item by each gender and across the three diagnostic groups. (Table 1). No gender difference was quantified across DMI items for the non-melancholic depressed patients, while mean total scores for the melancholic sample sub-set were also comparable for men and women. In the bipolar sub-set women were more likely to affirm three DMI items (i.e. feeling more vulnerable, self-critical and guilty). No comparisons remained significant after calculating FDR P values.
2.3. Statistical analyses 3.3. QIDS-SR data Differences in reporting between men and women were examined by undertaking t tests for quantitative (severity-based) data and chi square analyses for qualitative (prevalence) data. In light of the large number of analyses we also controlled for the type I error rate by adopting the ‘false discovery rate’ or FDR procedure SAS Institute Inc. (2013) and bolding analyses where significant differences remain after applying the FDR.
Scores on this measure were available for 1559 subjects: 273 (137 male, 136 female) diagnosed with unipolar melancholic depression, 434 participants (208 male, 226 female) with unipolar non-melancholic depression, and 852 participants (387 male, 465 female) with bipolar disorder. When total scores were examined (the QIDS-SR being assessed as a measure of severity), women
Table 1 DMI-10 item analyses by gender (presence/absence of symptom). DMI-10 item
Symptom present – % within gender indicating ‘yes’ Unipolar melancholic
1. Stewing or brooding negatively over things 2. Feel more vulnerable than usual 3. Being self-critical and hard on yourself 4. Feeling guilty about things in your life 5. Lost your core and essence 6. Feeling depressed 7. Feel less worthwhile 8. Feel hopeless or helpless 9. Feel more distant from other people 10. Nothing seems to be able to cheer you up
Po 0.05.
Bipolar
Male
Female
χ2
Male
Female
χ2
Male
Female
χ2
88.7 83.1 88.2 79.0 88.7 94.9 86.2 87.7 92.3 88.2
88.1 88.6 89.6 81.9 83.9 91.7 83.9 85.5 89.1 85.0
0.0 2.4 0.2 0.5 1.9 1.5 0.4 0.4 1.2 0.9
91.6 84.8 91.0 87.9 88.9 93.5 87.9 87.6 90.1 84.8
92.5 90.7 92.5 87.5 90.0 92.8 86.3 87.5 91.3 83.8
0.2 5.1n 0.5 0.0 0.2 0.1 0.4 0.0 0.3 0.1
88.4 82.8 87.9 84.3 83.9 88.4 83.9 80.0 90.3 80.2
90.1 88.1 92.4 88.6 85.1 88.3 86.3 82.8 88.3 78.7
0.8 6.2n 6.3n 4.4n 0.3 0.0 1.2 1.4 1.1 0.4
No significant differences in any analyses after FDR correction. n
Unipolar non-melancholic
354
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
gender differences identified for bipolar than for unipolar patients. Female bipolar patients rated, when depressed, and in comparison to males, higher on increased appetite, depressed mood and energy being worse in mornings, consummatory anhedonia, inability to be cheered by humour, being more indecisive, impaired concentration and slowed thinking, anergia, feeling physically slowed and somewhat ‘paralysed’ in doing things, feeling suicidal, being asocial, experiencing middle insomnia and weight gain (i.e. scoring higher than males on 15 of the 32 items in terms of severity and on 8 of the 32 items in terms of prevalence— essentially appetite/weight gain, energy and mood worse in the morning, psychomotor slowing and middle insomnia). After applying FDR many univariate analyses were no longer significant but such controlling made little difference to the analyses involving the bipolar patients.
with melancholia scored comparably to men (15.2 vs 14.3, ns), while women with non-melancholic depression scored slightly higher (15.1 vs 13.9, t ¼2.8, P o0.05) as did women with bipolar depression (15.5 vs 14.6, t¼ 2.7, Po 0.05). Whether individual symptoms were examined by prevalence or by severity, findings were generally consistent and therefore only symptom severity data are tabulated and further considered (see Table 2). In the unipolar melancholic sub-set men were more likely to report middle insomnia, whereas women were more likely to report hypersomnia, appetite increase and decreased energy levels as more severe. In the unipolar non-melancholic sub-set, women were more likely to report higher levels of impaired concentration and decreased energy levels. For bipolar patients, a larger number of gender differences were identified—with women more likely to report both appetite/weight gain and loss, hypersomnia, and greater impaired concentration. After applying FDR many univariate analyses were no longer significant but such controlling made little difference to the analyses involving the bipolar patients.
3.5. SMPI data Scores were available for 971 of the sample: 193 (98 male, 95 female) diagnosed with unipolar melancholic depression, 214 (103 male, 111 female) with unipolar non-melancholic depression; and 564 (265 male, 299 female) with bipolar disorder. As the SMPI generates only affirmed items judged as salient (but not their severity), we consider prevalence data only. In the unipolar melancholic sub-set men were more likely (P o0.05) than women to report mood and energy as worse in mornings (77.6% vs. 63.2%) and less likely to report feeling physically slowed and almost ‘paralysed’ (37.8% vs. 54.7%)—two of the 12 items assigning individuals to a melancholic diagnosis—while in relation to the 12 non-melancholic items, women were more likely to affirm being inclined to be emotional (57.9% vs. 29.6%), to worry even when not depressed (65.3% vs. 49.0%) and to judge that their depressive response could be explained by preceding stressful events and the impact on their personality style (34.7% vs. 19.4%). In the unipolar non-melancholic sub-set there were no gender differences across any of the 24 ‘melancholic’ and ‘non-melancholic’ items. In the bipolar sub-set, women were more likely than men to report four ‘melancholic’ items—anergia (87.6% vs. 74.7%), being physically slowed (57.9% vs. 43.0%), early years being no
3.4. SDS data Scores were available for 1770 subjects: 312 (156 male, 156 female) diagnosed with unipolar melancholic depression; 499 (244 male, 255 female) with unipolar non-melancholic depression, and 959 participants (452 male, 507 female) with bipolar disorder. In comparison to analyses involving the QIDS-SR measure, there was less consistency across symptom prevalence and symptom severity data, with the latter shown in Table 3. Women with unipolar melancholia rated higher than men on appetite decrease, depressed mood being worse in the evenings, anticipatory anhedonia, anergia, extreme guilt, feeling they deserved to be punished, feeling ‘paralysed’ in doing basic things and experiencing weight loss, while men did not rate higher than women on any of the 32 items. Women with unipolar non-melancholic depression rated higher than men on appetite increase, energy being worse both in mornings and evenings, finding it difficult to do basic things, feeling irritable, being more likely to wake in the middle of the night and in experiencing weight gain. Again, there were more Table 2 QIDS-SR item analyses by gender (severity of symptoms). QIDS-SR item
Symptom severity – mean (SD) Unipolar melancholic
1. Falling asleep delayed 2. Insomnia during night 3. Waking up too early 4. Hypersomnia 5. Feeling sad 6. Appetite decrease 7. Appetite increase 8. Weight decrease 9. Weight increase 10. Concentration/decision-making impaired 11. Negative view of self 12. Thoughts of death or suicide 13. Decreased interest in activities 14. Decreased energy levels 15. Feeling slowed down 16. Feeling restless QIDS-SR total score
P o 0.05. P o 0.001.
nn
Bipolar
Male
Female
t
Male
Female
t
Male
Female
T
1.4 (1.2) 1.9 (1.0) 0.8 (1.1) 1.0 (1.0) 1.8 (1.0) 0.7 (0.8) 0.9 (0.9) 0.6 (0.9) 1.1 (0.9) 1.5 (0.7) 1.6 (1.2) 1.0 (0.8) 1.6 (1.1) 1.5 (0.9) 1.0 (0.9) 0.8 (0.8) 14.3 (5.2)
1.5 (1.2) 1.6 (1.2) 0.9 (1.1) 1.3 (1.1) 1.9 (1.0) 0.9 (0.9) 1.4 (1.2) 0.7 (1.0) 1.3 (1.1) 1.6 (0.8) 1.7 (1.2) 1.0 (0.9) 1.8 (1.1) 1.8 (0.9) 1.0 (0.8) 0.8 (0.9) 15.2 (5.3)
0.7 2.1n 0.5 2.2n 1.2 1.2 2.6n 1.2 1.0 0.8 0.7 0.0 1.1 3.2n 0.5 0.3 1.7
1.4 (1.2) 1.7 (1.1) 1.0 (1.2) 0.7 (0.9) 1.7 (0.9) 0.7 (0.8) 1.3 (1.1) 0.9 (1.0) 1.2 (1.0) 1.4 (0.8) 1.8 (1.2) 0.9 (0.9) 1.4 (1.1) 1.3 (1.0) 0.8 (0.8) 0.8 (0.9) 13.9 (5.4)
1.6 (1.1) 1.8 (1.0) 1.2 (1.2) 0.8 (1.0) 1.9 (0.9) 0.9 (0.9) 1.4 (1.2) 0.8 (0.9) 1.2 (1.1) 1.6 (0.8) 1.9 (1.2) 0.9 (0.8) 1.4 (1.0) 1.7 (0.9) 1.0 (0.9) 1.0 (1.0) 15.1 (5.4)
1.7 1.7 1.5 1.7 1.9 1.2 0.6 0.5 0.2 2.1n 1.4 0.1 0.1 3.6nn 1.4 1.6 2.8n
1.6 (1.2) 1.8 (1.1) 1.1 (1.2) 0.8 (1.0) 1.7 (1.0) 0.9 (0.9) 1.2 (1.1) 0.7 (0.9) 1.2 (1.0) 1.5 (0.8) 1.8 (1.2) 1.0 (0.9) 1.5 (1.0) 1.5 (1.0) 1.0 (0.9) 1.3 (1.0) 14.6 (5.8)
1.8 (1.1) 1.8 (1.0) 1.0 (1.2) 1.0 (1.0) 1.8 (1.0) 1.2 (0.9) 1.6 (1.1) 0.9 (1.0) 1.4 (0.9) 1.7 (0.7) 1.9 (1.1) 1.0 (0.9) 1.5 (1.1) 1.6 (0.9) 1.0 (0.9) 1.3 (1.0) 15.5 (5.4)
1.6 0.5 0.9 2.8n 2.0 3.9nn 4.0nn 2.3n 1.8 3.6nn 1.4 0.1 0.7 1.8 1.1 0.5 2.7n
FDR-controlled significant variables are given in bold. n
Unipolar non-melancholic
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
355
Table 3 SDS item analyses by gender (severity of symptom). SDS item
Symptom severity – mean (SD) Unipolar melancholic Male
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
Appetite decrease Appetite increase Depressed mood worse in the morning Depressed mood worse in the evening Energy worse in the morning Energy worse in the evening Loss of interest in activities that you would normally enjoy Unable to obtain pleasure from activities Inability to be cheered up when something nice or pleasant occurred Inability to be cheered up by friends Loss of capacity to laugh at humorous things Not being able to look forward to things Difficulty making decisions Brain feeling foggy making concentration difficult Finding it difficult to do basic things (e.g. get out of bed, shower) Feeling physically slowed Find your thinking is slowed Being more agitated Feeling suicidal Feeling very guilty Feeling that you deserve to be punished Feeling irritable Feeling angry Feeling a need to stay away from people Being cheered up completely by being with friends Feeling a greater need to be close to people Feeling somewhat ‘paralysed’ when doing basic things Getting off to sleep takes much longer than usual Waking in the middle of the night more than usual Waking up much earlier than usual in the morning Experiencing distinct weight loss Experiencing distinct weight gain
0.9 (1.0) 0.6 (0.8) 1.9 (1.0) 1.0 (0.9) 2.0 (1.0) 1.0 (0.9) 2.3 (0.7) 2.3 (0.8) 2.1 (0.8) 2.0 (0.7) 1.8 (0.9) 2.3 (0.7) 2.1 (0.9) 2.2 (0.8) 1.9 (1.0) 1.7 (1.0) 1.9 (1.0) 1.4 (1.0) 1.3 (1.1) 1.5 (1.0) 0.7 (0.9) 1.9 (0.8) 1.5 (1.0) 2.1 (0.9) 0.6 (0.8) 0.6 (0.9) 1.7 (1.1) 1.6 (1.1) 1.5 (1.1) 1.1 (1.1) 0.5 (0.9) 0.6 (0.9)
Female 1.2 (1.2) 0.8 (1.0) 2.1 (1.0) 1.2 (1.1) 2.1 (1.0) 1.2 (1.0) 2.5 (0.6) 2.4 (0.7) 2.1 (0.8) 2.0 (0.8) 1.9 (0.9) 2.4 (0.7) 2.3 (0.8) 2.3 (0.9) 2.2 (0.9) 1.8 (1.0) 1.9 (1.0) 1.3 (1.2) 1.3 (1.1) 1.8 (1.1) 1.1 (1.1) 1.8 (1.0) 1.4 (1.0) 2.3 (0.9) 0.6 (0.8) 0.8 (1.0) 2.0 (1.0) 1.6 (1.2) 1.6 (1.1) 1.1 (1.2) 0.8 (1.1) 0.7 (1.0)
Unipolar non-melancholic
t n
2.4 1.4 1.2 2.2n 1.0 0.5 1.8 1.7 0.0 0.1 0.8 2.0n 1.8 1.1 3.1n 0.8 0.4 0.1 0.1 2.1n 2.7n 0.6 0.9 1.4 0.3 1.7 2.1n 0.2 0.6 0.3 2.2n 0.5
Bipolar
Male
Female
t
Male
Female
t
1.1 (1.0) 0.7 (0.9) 1.6 (1.1) 1.3 (1.0) 1.5 (1.0) 1.2 (0.9) 2.1 (0.9) 2.1 (0.9) 1.7 (0.9) 1.7 (0.8) 1.5 (1.0) 2.0 (0.8) 2.1 (0.8) 2.0 (0.9) 1.4 (1.0) 1.4 (1.0) 1.5 (0.9) 1.4 (1.0) 1.3 (1.1) 1.6 (1.0) 0.9 (1.0) 1.9 (0.9) 2.0 (0.9) 2.0 (0.9) 0.7 (0.7) 0.8 (0.9) 1.5 (1.1) 1.6 (1.1) 1.5 (1.1) 1.2 (1.1) 0.5 (0.8) 0.7 (1.0)
1.1 (1.0) 1.0 (1.1) 1.7 (1.1) 1.4 (1.1) 1.8 (1.1) 1.4 (1.0) 2.1 (0.9) 2.0 (1.0) 1.7 (0.9) 1.7 (0.9) 1.6 (1.1) 2.0 (0.9) 2.2 (0.9) 2.1 (0.9) 1.6 (1.1) 1.4 (1.0) 1.6 (1.0) 1.4 (1.1) 1.3 (1.1) 1.7 (1.1) 1.0 (1.2) 2.1 (0.9) 2.0 (1.0) 2.2 (0.9) 0.7 (0.9) 0.8 (0.9) 1.6 (1.1) 1.7 (1.1) 1.7 (1.0) 1.3 (1.1) 0.5 (0.9) 1.0 (1.1)
0.6 3.5nn 1.2 0.9 2.8n 2.8n 0.0 0.2 0.1 0.4 0.4 0.2 0.1 1.1 2.1n 0.7 1.2 0.7 0.7 1.2 1.5 2.8n 1.2 0.5 0.1 0.8 0.1 1.0 2.5n 0.9 0.2 2.9n
1.2 (1.1) 0.8 (1.1) 1.7 (1.1) 1.4 (1.0) 1.8 (1.0) 1.3 (1.0) 2.3 (0.7) 2.2 (0.8) 2.0 (0.8) 1.8 (0.9) 1.7 (0.9) 2.1 (0.8) 2.2 (0.9) 2.3 (0.8) 1.9 (1.0) 1.6 (1.0) 1.8 (1.0) 1.8 (1.0) 1.4 (1.1) 1.9 (1.0) 1.3 (1.1) 2.1 (0.8) 2.0 (1.0) 2.3 (0.8) 0.7 (0.8) 0.7 (1.0) 1.8 (1.0) 1.8 (1.1) 1.6 (1.1) 1.1 (1.1) 0.7 (1.0) 0.7 (1.0)
1.4 (1.1) 1.1 (1.1) 1.9 (1.0) 1.4 (1.0) 2.0 (1.0) 1.3 (1.0) 2.4 (0.8) 2.3 (0.8) 2.0 (0.9) 1.9 (0.9) 1.8 (0.9) 2.3 (0.8) 2.5 (0.7) 2.5 (0.7) 2.1 (0.9) 1.9 (1.0) 2.0 (1.0) 1.8 (1.0) 1.6 (1.1) 2.0 (1.0) 1.4 (1.1) 2.3 (0.9) 2.0 (1.0) 2.5 (0.8) 0.6 (0.7) 0.7 (0.9) 2.0 (1.0) 1.9 (1.1) 1.7 (1.1) 1.2 (1.1) 0.7 (1.0) 1.1 (1.1)
1.8 4.2nn 3.1n 0.7 2.9n 0.6 1.8 1.5 0.2 1.3 2.2n 2.7n 5.4nn 3.8nn 4.3nn 4.1nn 2.9n 0.2 2.6n 1.3 1.2 2.0 0.1 3.0n 1.3 0.1 3.1n 2.0 3.0n 1.0 0.0 5.0nn
FDR-controlled significant variables are bolded. n
P o 0.05. P o0.001.
nn
more difficult than most people (51.3% vs. 42.5%) and relationships and work performance being generally good when not depressed (71.3% vs. 62.2%)—and two ‘non-melancholic’ items: food cravings and increased appetite (50.2% vs. 34.0%) and, when not depressed, tending to worry more than most people (62.5% vs. 53.6%).
4. Discussion Women have been widely demonstrated to show higher levels and ‘caseness’ rates of depression reflecting both ‘real’ and artefactual effects. In relation to the latter, we sought to examine any impact that actual depression measures might contribute, whether by influencing severity or prevalence of item ratings by men and women (a response bias) or by elements of depression having differential salience to men or to women (a gender dimorphic weighting). We acknowledge that differences in prevalence rates by gender are generally assessed by ‘case’ measures rather than by our study measures of severity, but note (as stated in Section 1) that women have been shown to both return higher case rates and score higher on depression severity scores. Secondly, rather than study ‘clinical depression’ as an overall category, we considered three depressive sub-types: melancholic, nonmelancholic and bipolar depression. By studying both depression severity and prevalence in relation to depressive sub-types we were able to further determine where gender had its greatest differential impact, an approach not previously explored. By
examining several differing measures we could determine whether there was consistency in any identified gender differences. While not all patients completed all measures, the numbers completing the four measures under examination were substantive and reject any judgment that the study was under-powered. If anything, the large numbers risked according significant differences when mean scores or prevalence rates across the genders were actually slight, although we do also report statistical significance after controlling for the type I error rate. As noted, we examined one brief measure of depression severity (the DMI-10), with items weighting the phenomenological expression of depression and not including any of the somatic or vegetative features that commonly accompany clinical depression. Women scored comparably to men (i.e. only 1% higher in the melancholic, 3% higher in the non-melancholic and 5% higher in the bipolar sub-set, and with only the last being formally significant) in affirming the prevalence of individual items. That difference in the last group—if not due to chance and the large sample sub-set (as it was no longer significant after FDR analysis) —might reflect greater mood amplification by females (i.e. scoring higher on feeling vulnerable, self-criticism and guilt) or a sociological difference in the female experience of bipolar depression. Due to the large sample size and relatively small difference between male and female percentages, results would benefit from replication before firm conclusions are drawn. In essence, the DMI-10 appeared essentially uninfluenced by gender in those with unipolar depression.
356
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Our second measure—the QIDS-SR—generates both an estimate of severity and the prevalence of depressive sub-type symptoms. As a severity measure, women with melancholia scored nonsignificantly higher (6%) than men, while those with nonmelancholic depression and bipolar depression scored significantly higher (9% and 6% respectively). The essentially comparable percentage differences across the three depressive sub-types would suggest that formal statistical significance or nonsignificance was presumably influenced by sample sub-set sizes. When individual QIDS-SR items were examined, few differences were evident in the unipolar (melancholic and non-melancholic sub-sets) while numerous differences were established in the bipolar sub-set, and with such differential pattern across the three diagnostic groups remaining after controlling for multiple comparisons. In essence, on both measures of severity (i.e. DMI-10 and QIDS-SR) we observe a general trend for women to score minimally higher than men, with the DMI-10 difference being marginal. Again such a finding argues against any gender-specific response bias in rating depression severity, and that these measures do not appear sensitive to any gender bias in their indication of severity. Our remaining two measures (i.e. the SDS and SMPI) were designed to have representative melancholic and non-melancholic items and therefore allow symptom representation across men and women to be examined rather than depression severity levels. As for the QIDS-SR, the greatest number of gender differences across items were quantified in the bipolar patients (and still clearly evident after controlling for multiple comparisons), a key finding that will be returned to shortly. Our examination of three measures assessing clinical depressive symptoms (as against mood severity alone) allows consideration of consistency of item affirmation to be considered and thus allows some estimate of the validity of findings. For the unipolar melancholic sub-set, there was a consistent finding of women reporting greater weight loss and anergia (albeit only a trend for the SMPI) than men. For the unipolar non-melancholic sub-set, there were no significant gender differences across the 24 SMPI items and no consistent differences when QIDS-SR and SDS findings were compared. For the bipolar sub-set women appeared consistently more likely to report appetite and weight changes (most commonly increased and associated in the SMPI with food cravings), to have low energy and feel physically slowed and to have impaired concentration. Validity of findings in studies of this nature might be addressed by also administering an interviewerrated measure for examination against the self-report measures. In relation to the general finding that women have higher rates of lifetime depression, this study allows us to examine whether this might be contributed to by measurement response bias— whereby women might have a general tendency to affirm more depressive items and/or rate the severity of any experienced items in an amplified way when completing depression measures. We find little support for this proposition in our unipolar depressed patients (whether assigned as melancholic or non-melancholic) and therefore suggest that any such measurement error is unlikely to contribute to the general finding. We did, however, quantify greater severity of depression in women with a bipolar disorder— although when quantified by total scores the magnitude was low (i.e. 5% for DMI-10 and 6% for QIDS-SR). We suggest that such a difference was unlikely to reflect any response bias or amplification effect (as would have then expected to observe women rating higher on all depressive symptoms) and could reflect real gender differences in symptom prevalence and severity, particularly more substantive changes in appetite, psychomotor disturbance and other features weighted to a melancholic depression. Why women with a bipolar disorder (compared to those with a unipolar disorder) might rate such features as more prevalent or more
severe than men remains unclear and requires replication as we are not aware of any previous study examining for gender differences in depressive symptoms in bipolar disorder. Kessler (2005) overviewed data indicating that the female preponderance in clinical depression reflected women being more likely than men to have an initial episode (and that once depression had occurred, gender had little impact on repeat episodes). The latter finding would not be expected if the female preponderance was simply or somewhat underpinned by response biases in completing depression rating measures. This study similarly finds no support for postulating that gender-based response biases in completing depression measures contributes to the differential community rates. Study findings identify some clinical depressive symptoms that do appear to be differentially experienced by men and women when depressed. While greater variation in eating and weight symptoms was expected, the finding that many melancholic symptoms (e.g. anergia, feeling slowed down and impaired concentration) were rated as more prevalent and/or severe was unexpected. As the greatest differentiation in such symptoms was quantified in the bipolar sub-set (and where bipolar depression is seemingly most commonly melancholic in type), it would appear that women may experience melancholic depression more severely than men. Future research examining pre-menopausal and post-menopausal women with aged matched men would be of interest, as would other potentially explanatory studies.
Role of funding source All authors declare that the funding source had no impact on the study design, the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest All authors declare no conflict of interest.
Acknowledgements We acknowledge the assistance of Georgia McClure with database management and Dusan Hadzi-Pavlovic for statistical advice.. This study was supported by an Australian National Health and Medical Research Council (NHMRC) Program Grant 1037196.
References Bernstein, I.H., Rush, A.J., Suppes, T., Kyotoku, Y., Warden, D., 2010. The quick inventory of depressive symptomatology (clinician and self-report versions) in patients with bipolar disorder. CNS Spectr. 15, 367–373. Goodwin, F., Jamison, K., 2007. Manic Depressive IllnessOxford University Press, New York. Hildebrandt, M.G., Stage, K.B., Kragh-Soerensen, P., 2003. Gender differences in severity, symptomatology and distribution of melancholia in major depression. Psychopathology 36, 204–212. Hilton, T.M., Parker, G., McDonald, et al., 2006. A validation study of two brief measures of depression in the cardiac population: the DMI-10 and DMI-18. Psychosomatics 47, 129–135. Kessler, R.C., 2005. Gender differences in major depression. Epidemiological findings. In: Frank, E (Ed.), Gender and Its Effects on Psychopathology. American Psychiatric Press, Washington. Khan, A.A., Gardner, C.O., Prescott, C.A., Kendler, K.S., 2002. Gender differences in the symptoms of major depression in opposite-sex dizygotic twin pairs. Am. J. Psychiatry 159, 1427–1429. Kornstein, S.G., Schatzberg, A.F., Thase, M.E., 2000. Gender differences in chronic major and double depression. J. Affect. Disord. 60, 1–11. Marcus, S.M., Young, E.A., Kerber, K.B., 2005. Gender differences in depression: findings from the STARnD study. J. Affect. Disord. 87, 141–150. Nolen-Hoeksema, S., Parker, L.E., Larsen, J., 1994. Ruminative coping with depressed mood following loss. J. Personal. Soc. Psychol. 67, 92–104. Parker, G., Brotchie, H., 2010. Gender differences in depression. Int. Rev. Psychiatry 22, 429–436. Parker, G., Fletcher, K., Barrett, M., Synnott, M., Breakspear, M., Rees, A.M., HadziPavlovic, D., 2010. Inching toward Bethlehem: mapping melancholia. J. Affect. Disord. 123, 291–298.
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Parker, G., Fletcher, K., Hadzi-Pavlovic, D., 2012. Is context everything to the definition of clinical depression? A test of the Horwitz and Wakefield postulate. J. Affect. Disord. 136, 1034–1038. Parker, G., Fletcher, K., Hyett, M., 2008. The mood assessment program: a computerised diagnostic tool for deriving management plans for mood disorders. Med. J. Aust. 188 (12), S126–S128. Parker, G., Fletcher, K., Hyett, M., Hadzi-Pavlovic, D., Barrett, M., Synnott, H., 2009. Measuring melancholia: the utility of a prototypic symptom approach. Psychol. Med. 39, 989–998. Parker, G., Hadzi-Pavlovic, D., 1996. Melancholia: A Disorder of Movement and MoodCambridge University Press, New York. Parker, G., Hilton, T., Bains, J., Hadzi-Pavlovic, D., 2002. Cognitive-based measures screening for depression in the medically ill: the DMI-10 and the DMI-18. Acta Psychiatr. Scand. 105, 419–426.
357
Parker, G., McCraw, S., Blanch, B., Hadzi-Pavlovic, D., Synnott, H., Rees, A.M., 2013a. Discriminating melancholic and non-melancholic depression by prototypic clinical features. J. Affect. Disord. 25, 199–207. Parker, G., McCraw, S., Hadzi-Pavlovic, D., Hong, M., Barrett, M., 2013b. Bipolar depression: prototypically melancholic in its clinical features. J. Affect. Disord. 147, 331–337. Piccinelli, M., Wilkison, G., 2000. Gender differences in depression. Br. J. Psychiatry 177, 486–492. Rush, A.J., Trivedi, M.H., Ibrahim, et al., 2003. The 16-item quick inventory of depressive symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol. Psychiatry 54, 573–583. SAS Institute Inc., 2013. SAS/STAT 13.1 User's GuideSAS Institute Inc., Cary, NC.
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Gender differences in depression severity and symptoms across depressive sub-types Gordon Parker a,b,n, Kathryn Fletcher a,b, Amelia Paterson a,b, Josephine Anderson b, Michael Hong b a b
School of Psychiatry, University of New South Wales, BDI Building, Hospital Road, Randwick, 2031 Sydney, Australia Black Dog Institute, Sydney, Australia
art ic l e i nf o
a b s t r a c t
Article history: Received 14 February 2014 Received in revised form 12 June 2014 Accepted 12 June 2014 Available online 19 June 2014
Background: Lifetime rates of depression are distinctly higher in women reflecting both real and artefactual influences. Most prevalence studies quantifying a female preponderance have examined severity-based diagnostic groups such as major depression or dysthymia. We examined gender differences across three depressive sub-type conditions using four differing measures to determine whether any gender differences emerge more from severity or symptom prevalence, reflect nuances of the particular measure, or whether depressive sub-type is influential. Methods: A large clinical sample was recruited. Patients completed two severity-weighted depression measures: the Depression in the Medically Ill 10 (DMI-10) and Quick Inventory of Depressive SymptomsSelf-Report (QIDS-SR) and two measures weighting symptoms and illness correlates of melancholic and non-melancholic depressive disorders - the Severity of Depressive Symptoms (SDS) and Sydney Melancholia Prototype Index (SMPI). Analyses were undertaken of three diagnostic groups comprising those with unipolar melancholic, unipolar non-melancholic and bipolar depressive conditions. Results: Women in the two unipolar groups scored only marginally (and non-significantly) higher than men on the depression severity measures. Women in the bipolar depression group, did however, score significantly higher than men on depression severity. On measures weighted to assessing melancholic and non-melancholic symptoms, there were relatively few gender differences identified in the melancholic and non-melancholic sub-sets, while more gender differences were quantified in the bipolar subset. The symptoms most commonly and consistently differentiating by gender were those assessing appetite/weight change and psychomotor disturbance. Conclusion: Our analyses of several measures and the minimal differentiation of depressive symptoms and symptom severity argues against any female preponderance in unipolar depression being contributed to distinctly by these depression rating measures. Our analyses indicated that gender had minimal if any impact on depression severity estimates. Gender differences in depressive symptoms and severity were more distinctive in bipolar patients, a finding seemingly not previously identified or reported. Limitations: The study had considerable power reflecting large sample sizes and thus risks assigning significant differences where none truly exist, although we repeated analyses after controlling for the type I error rate. & 2014 Elsevier B.V. All rights reserved.
Keywords: Depression Bipolar disorder Epidemiology Gender
1. Introduction As observed by Kessler (2005), one “of the most widely documented findings in psychiatric epidemiology is that women n Corresponding author at: School of Psychiatry, University of New South Wales, BDI Building, Hospital Road, Randwick, 2031 Sydney, Australia. Tel.: þ 61 2 93824372. E-mail address: [email protected] (G. Parker).
http://dx.doi.org/10.1016/j.jad.2014.06.018 0165-0327/& 2014 Elsevier B.V. All rights reserved.
have higher rates of depression than men.” He overviewed a number of studies quantifying a female preponderance ranging from 1.5 to 3 times higher rates of depression. Explanations reviewed by Kessler included artefactual influences as well as ‘real’ biological and psychosocial factors. The former explanation includes women simply being more likely to seek help, acknowledge and over-report depressive symptoms, and the tendency for women to ‘amplify’ and for men to ‘blunt’ reporting of any experienced depressive symptom (Nolen-Hoeksema et al., 1994).
352
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Kessler and most other reviewers favour ‘real’ factors such as hormones and social and gender-roles making greater contribution to the gender difference. However, the extent to which measures of depression include gender-weighted or genderdimorphic features may also influence rates, and at times have an artefactual impact on prevalence estimates. For example, if a depression measure includes depressive symptoms more likely to be experienced by depressed women than men (e.g. crying) then more women will exceed the measure's ‘caseness’ threshold and contribute to differential depression rates. If the female preponderance in depression rates simply and only reflects women being more likely to experience initial depressive episodes than men—which Kessler viewed as the ‘genuine’ explanation—we would expect symptom representation and severity to be comparable across men and women in depressed samples. If the preponderance also reflects differential experience and/or severity in certain symptoms then we would expect women to be more likely to report such symptoms and/or rate them more highly. Piccinelli and Wilkison (2000) reviewed a number of studies and concluded that several symptoms were over-represented in females (i.e. disturbances of appetite and sleep, fatigue, somatic anxiety and hypochondiasis). Most of these might, however, be viewed as indirect expressions or correlates of depression rather than mood-specific features, while Kornstein et al. (2000) also identified psychomotor disturbance—along with sleep changes and anxiety/somatisation—as more prevalent in women. Analysing data from the STARnD study, Marcus et al. (2005) quantified appetite and weight changes, anergia, psychomotor agitation and sympathetic arousal symptoms as over-represented in women. In a study of twins experiencing major depression, Khan et al (2002) quantified female twins as experiencing more fatigue, hypersomnia and psychomotor retardation, while male twins experienced more insomnia and agitation. Most references to a female preponderance in lifetime rates of ‘depression’ refer to severity-based depressive categories, especially major depression but also minor states such as dysthymia. We have long favoured a sub-typing model that seeks to weight categories (e.g. melancholic vs. a heterogeneous non-melancholic group; bipolar vs. unipolar depression—with bipolar depression being more likely to be melancholic in its clinical pattern of symptoms, Parker et al., 2013a). Bipolar disorder lacks any distinctive gender weighting (Goodwin and Jamison, 2007; Parker and Brotchie, 2010) while studies variably suggest that the prevalence of melancholia is comparable in men and women, or somewhat over-represented in women and even somewhat overrepresented in men (Hildebrandt et al., 2003), albeit with any gender weighting being relatively slight. If any depressive subtype has comparable or near comparable rates in men and women, then examining for gender differences in the prevalence and/or severity of constituent symptoms in such sub-types offers an advantage in exploring for any gender differences in their actual measurement. We therefore report a study examining gender differences across bipolar, unipolar melancholic and unipolar nonmelancholic depressive conditions. We hypothesised that if the former two conditions have comparable prevalences for men and women, then gender differentiation in symptoms would be less distinct than for the non-melancholic depressive conditions which we assume underpin the community studies quantifying a female preponderance. We analysed data from four differing depression rating measures, variably rating depression severity and subtyping features, and with multiple measures allowing for consistency of findings to be evaluated. We further hypothesised that women would be more likely to rate higher on depression severity as a consequence of mood amplification differences, but that such an artefactual influence would be less distinct in measures dominated by items examining sub-type symptoms.
2. Material and methods 2.1. General overview Over the 2005–2013 period, patients attending the Black Dog Institute Depression Clinic (a Sydney-based tertiary referral clinic for diagnostic and management advice for mood disorders) were invited to complete a series of questionnaires for research purposes. Written informed consent was obtained in accordance with University of New South Wales Ethics Committee requirements. As part of the routine clinical assessment prior to consultation with a psychiatrist, patients completed the Quick Inventory of Depressive Symptomatology-SelfReport (QIDS-SR; Rush et al., 2003), which quantifies depression severity and includes DSM-IV depression symptoms. Patients additionally completed the Mood Assessment Programme or MAP (Parker et al., 2008), a computerised tool assessing a range of features including socio-demographic, mood disorder and treatment history details. The MAP also contains a number of validated self-report measures of depressive symptomatology, including the Depression in the Medically Ill-10 measure (DMI-10; Parker et al., 2002), the Severity of Depressive Symptoms measure (SDS; Parker et al., 2009; 2010) and the Sydney Melancholia Prototype Index SMPI (Parker et al., 2012, 2013b). All measures are detailed shortly. Following consultation with a psychiatrist, diagnostic information was recorded. Inclusion criteria for the study were: age 15–85, a clinical diagnosis of a mood disorder and its sub-type (unipolar melancholic, unipolar non-melancholic, bipolar I or II disorder), no current psychosis or underlying organic issues, good comprehension of English, and ability to provide written informed consent. A clinical diagnosis of melancholia weighted features such as an anhedonic and non-reactive mood, psychomotor disturbance (and particularly impaired concentration), anergia and diurnal variation (with mood and energy worse in the morning), and was complemented by assessing illness correlates (e.g. a family history of depression, a superior response to any previous antidepressant medication as against any non-drug therapy). A clinical diagnosis of a bipolar disorder required distinct evidence of non-psychotic hypomanic episodes or psychotic manic episodes (assigned as bipolar II and I respectively), with oscillating periods of melancholic or psychotic symptoms during depressive episodes, and again complemented by assessing illness correlates (e.g. a family history of depression or bipolar disorder, a relatively or categorically distinct onset). 2.2. Overview of depression measures 2.2.1. Depression in the Medically Ill-10 (DMI-10) measure The DMI-10 (Parker et al., 2002) was designed to detect state depression severity in patients with medical disorders, and to respect and redress concerns that such patients might score falsely high on somatic symptoms such as insomnia, fatigue and weight loss (due to their medical illness rather than concomitants of depression), and so comprises 10 non-somatic depressed mood constructs. Scores for each item range from 0–3, with total scores ranging from 0–30 (a score of 9 or above being identified as indicative of ‘clinical depression’). In a subsequent study the measure demonstrated high internal consistency (α¼0.92; Hilton et al., 2006), and as it lacks somatic items and weights key depressive mood constructs it has been used as a general measure of depression severity. Any impact of gender on its scoring has not been previously examined. 2.2.2. Quick Inventory of Depressive Symptomatology-Self-Report (QIDS-SR) The QIDS-SR (Rush et al., 2003) assesses the presence and severity of 16 DSM-IV depression-related symptoms, is viewed as
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
353
3. Results
suitable for use with both unipolar and bipolar populations (Bernstein et al., 2010) and includes both melancholic and nonmelancholic symptoms. The measure demonstrates high internal consistency (α ¼86; Rush et al., 2003). Total scores range from 0 to 27, with scores of 11 or above indicative of moderate to severe depression.
3.1. Sample characteristics and measure scoring A total of 2865 participants received a clinical diagnosis (n ¼557 unipolar melancholic, n¼ 807 unipolar non-melancholic, n ¼1501 bipolar) and provided research data — but not all measures were completed by all subjects in this overall sample. Females were slightly over-represented (56.0%), and males were significantly older than females (40.4 vs. 37.2, t¼6.4, Po 0.001). Given the significant age difference, male and female participants were age-matched within each diagnostic category, yielding a subsequent total of 2308 participants (n ¼412 unipolar melancholic, n ¼690 unipolar non-melancholic, n ¼ 1055 bipolar). All subsequent analyses were based on the age-matched sample, and where the mean age was 39.3 (SD ¼ 12.7; age range 15–80). Item scores on each measure were contrasted for males and females via two methods: (i) scores for each item were re-coded to reflect the presence vs. absence of each symptom (0 vs. 1, 2 or 0 vs. 1, 2, 3), and (ii) raw scores for each item were summed to quantify severity of each symptom and, if relevant, total measure scores.
2.2.3. Severity of Depressive Symptoms (SDS) The self-report Severity of Depressive Symptoms (SDS) measure (Parker et al., 2009, 2010) comprises 32 clinical symptoms rated on a 0–3 scale, and again captures symptoms historically weighted to both melancholic and non-melancholic depression. Properties of the measure have been reported previously (Parker et al., 2009; 2010). 2.2.4. Sydney Melancholia Prototype Index (SMPI) The SMPI (previously known as the Self-Report of Depressive Experiences measure or SERDEX; (Parker et al., 2012, 2013a) is a 24-item measure comprising 12 melancholic and 12 nonmelancholic prototypic features (both symptoms and illness correlates) displayed in two columns. Individuals are invited to tick any item from either column that they regard as ‘characteristic’ in terms of their worst depressive episodes. The listed items assess symptoms historically favoured as most differentiating of melancholic and non-melancholic depression, in addition to assessing illness correlates (e.g. pre-morbid interpersonal functioning, distal and proximal stressors, the context and impact of proximal stressors on inducing and maintaining the depression, and trait emotional dysregulation levels) aligned with both melancholic and non-melancholic depression. Each item was selected by considering its utility in differentiating melancholic and nonmelancholic depression in previous studies (e.g., Parker and Hadzi-Pavlovic, 1996; Parker et al., 2010), and tested empirically to quantify their differentiating potential. Properties of the initial self-report measure have been reported previously (Parker et al., 2012, 2013b).
3.2. DMI-10 data Scores on this measure were available for 2130 subjects: 388 (195 male, 193 female) participants diagnosed with unipolar melancholic depression, 644 (323 male, 321 female) with unipolar non-melancholic depression, and 1098 (535 male, 563 female) with bipolar disorder. For this measure we tabulate the prevalence of affirming each item by each gender and across the three diagnostic groups. (Table 1). No gender difference was quantified across DMI items for the non-melancholic depressed patients, while mean total scores for the melancholic sample sub-set were also comparable for men and women. In the bipolar sub-set women were more likely to affirm three DMI items (i.e. feeling more vulnerable, self-critical and guilty). No comparisons remained significant after calculating FDR P values.
2.3. Statistical analyses 3.3. QIDS-SR data Differences in reporting between men and women were examined by undertaking t tests for quantitative (severity-based) data and chi square analyses for qualitative (prevalence) data. In light of the large number of analyses we also controlled for the type I error rate by adopting the ‘false discovery rate’ or FDR procedure SAS Institute Inc. (2013) and bolding analyses where significant differences remain after applying the FDR.
Scores on this measure were available for 1559 subjects: 273 (137 male, 136 female) diagnosed with unipolar melancholic depression, 434 participants (208 male, 226 female) with unipolar non-melancholic depression, and 852 participants (387 male, 465 female) with bipolar disorder. When total scores were examined (the QIDS-SR being assessed as a measure of severity), women
Table 1 DMI-10 item analyses by gender (presence/absence of symptom). DMI-10 item
Symptom present – % within gender indicating ‘yes’ Unipolar melancholic
1. Stewing or brooding negatively over things 2. Feel more vulnerable than usual 3. Being self-critical and hard on yourself 4. Feeling guilty about things in your life 5. Lost your core and essence 6. Feeling depressed 7. Feel less worthwhile 8. Feel hopeless or helpless 9. Feel more distant from other people 10. Nothing seems to be able to cheer you up
Po 0.05.
Bipolar
Male
Female
χ2
Male
Female
χ2
Male
Female
χ2
88.7 83.1 88.2 79.0 88.7 94.9 86.2 87.7 92.3 88.2
88.1 88.6 89.6 81.9 83.9 91.7 83.9 85.5 89.1 85.0
0.0 2.4 0.2 0.5 1.9 1.5 0.4 0.4 1.2 0.9
91.6 84.8 91.0 87.9 88.9 93.5 87.9 87.6 90.1 84.8
92.5 90.7 92.5 87.5 90.0 92.8 86.3 87.5 91.3 83.8
0.2 5.1n 0.5 0.0 0.2 0.1 0.4 0.0 0.3 0.1
88.4 82.8 87.9 84.3 83.9 88.4 83.9 80.0 90.3 80.2
90.1 88.1 92.4 88.6 85.1 88.3 86.3 82.8 88.3 78.7
0.8 6.2n 6.3n 4.4n 0.3 0.0 1.2 1.4 1.1 0.4
No significant differences in any analyses after FDR correction. n
Unipolar non-melancholic
354
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
gender differences identified for bipolar than for unipolar patients. Female bipolar patients rated, when depressed, and in comparison to males, higher on increased appetite, depressed mood and energy being worse in mornings, consummatory anhedonia, inability to be cheered by humour, being more indecisive, impaired concentration and slowed thinking, anergia, feeling physically slowed and somewhat ‘paralysed’ in doing things, feeling suicidal, being asocial, experiencing middle insomnia and weight gain (i.e. scoring higher than males on 15 of the 32 items in terms of severity and on 8 of the 32 items in terms of prevalence— essentially appetite/weight gain, energy and mood worse in the morning, psychomotor slowing and middle insomnia). After applying FDR many univariate analyses were no longer significant but such controlling made little difference to the analyses involving the bipolar patients.
with melancholia scored comparably to men (15.2 vs 14.3, ns), while women with non-melancholic depression scored slightly higher (15.1 vs 13.9, t ¼2.8, P o0.05) as did women with bipolar depression (15.5 vs 14.6, t¼ 2.7, Po 0.05). Whether individual symptoms were examined by prevalence or by severity, findings were generally consistent and therefore only symptom severity data are tabulated and further considered (see Table 2). In the unipolar melancholic sub-set men were more likely to report middle insomnia, whereas women were more likely to report hypersomnia, appetite increase and decreased energy levels as more severe. In the unipolar non-melancholic sub-set, women were more likely to report higher levels of impaired concentration and decreased energy levels. For bipolar patients, a larger number of gender differences were identified—with women more likely to report both appetite/weight gain and loss, hypersomnia, and greater impaired concentration. After applying FDR many univariate analyses were no longer significant but such controlling made little difference to the analyses involving the bipolar patients.
3.5. SMPI data Scores were available for 971 of the sample: 193 (98 male, 95 female) diagnosed with unipolar melancholic depression, 214 (103 male, 111 female) with unipolar non-melancholic depression; and 564 (265 male, 299 female) with bipolar disorder. As the SMPI generates only affirmed items judged as salient (but not their severity), we consider prevalence data only. In the unipolar melancholic sub-set men were more likely (P o0.05) than women to report mood and energy as worse in mornings (77.6% vs. 63.2%) and less likely to report feeling physically slowed and almost ‘paralysed’ (37.8% vs. 54.7%)—two of the 12 items assigning individuals to a melancholic diagnosis—while in relation to the 12 non-melancholic items, women were more likely to affirm being inclined to be emotional (57.9% vs. 29.6%), to worry even when not depressed (65.3% vs. 49.0%) and to judge that their depressive response could be explained by preceding stressful events and the impact on their personality style (34.7% vs. 19.4%). In the unipolar non-melancholic sub-set there were no gender differences across any of the 24 ‘melancholic’ and ‘non-melancholic’ items. In the bipolar sub-set, women were more likely than men to report four ‘melancholic’ items—anergia (87.6% vs. 74.7%), being physically slowed (57.9% vs. 43.0%), early years being no
3.4. SDS data Scores were available for 1770 subjects: 312 (156 male, 156 female) diagnosed with unipolar melancholic depression; 499 (244 male, 255 female) with unipolar non-melancholic depression, and 959 participants (452 male, 507 female) with bipolar disorder. In comparison to analyses involving the QIDS-SR measure, there was less consistency across symptom prevalence and symptom severity data, with the latter shown in Table 3. Women with unipolar melancholia rated higher than men on appetite decrease, depressed mood being worse in the evenings, anticipatory anhedonia, anergia, extreme guilt, feeling they deserved to be punished, feeling ‘paralysed’ in doing basic things and experiencing weight loss, while men did not rate higher than women on any of the 32 items. Women with unipolar non-melancholic depression rated higher than men on appetite increase, energy being worse both in mornings and evenings, finding it difficult to do basic things, feeling irritable, being more likely to wake in the middle of the night and in experiencing weight gain. Again, there were more Table 2 QIDS-SR item analyses by gender (severity of symptoms). QIDS-SR item
Symptom severity – mean (SD) Unipolar melancholic
1. Falling asleep delayed 2. Insomnia during night 3. Waking up too early 4. Hypersomnia 5. Feeling sad 6. Appetite decrease 7. Appetite increase 8. Weight decrease 9. Weight increase 10. Concentration/decision-making impaired 11. Negative view of self 12. Thoughts of death or suicide 13. Decreased interest in activities 14. Decreased energy levels 15. Feeling slowed down 16. Feeling restless QIDS-SR total score
P o 0.05. P o 0.001.
nn
Bipolar
Male
Female
t
Male
Female
t
Male
Female
T
1.4 (1.2) 1.9 (1.0) 0.8 (1.1) 1.0 (1.0) 1.8 (1.0) 0.7 (0.8) 0.9 (0.9) 0.6 (0.9) 1.1 (0.9) 1.5 (0.7) 1.6 (1.2) 1.0 (0.8) 1.6 (1.1) 1.5 (0.9) 1.0 (0.9) 0.8 (0.8) 14.3 (5.2)
1.5 (1.2) 1.6 (1.2) 0.9 (1.1) 1.3 (1.1) 1.9 (1.0) 0.9 (0.9) 1.4 (1.2) 0.7 (1.0) 1.3 (1.1) 1.6 (0.8) 1.7 (1.2) 1.0 (0.9) 1.8 (1.1) 1.8 (0.9) 1.0 (0.8) 0.8 (0.9) 15.2 (5.3)
0.7 2.1n 0.5 2.2n 1.2 1.2 2.6n 1.2 1.0 0.8 0.7 0.0 1.1 3.2n 0.5 0.3 1.7
1.4 (1.2) 1.7 (1.1) 1.0 (1.2) 0.7 (0.9) 1.7 (0.9) 0.7 (0.8) 1.3 (1.1) 0.9 (1.0) 1.2 (1.0) 1.4 (0.8) 1.8 (1.2) 0.9 (0.9) 1.4 (1.1) 1.3 (1.0) 0.8 (0.8) 0.8 (0.9) 13.9 (5.4)
1.6 (1.1) 1.8 (1.0) 1.2 (1.2) 0.8 (1.0) 1.9 (0.9) 0.9 (0.9) 1.4 (1.2) 0.8 (0.9) 1.2 (1.1) 1.6 (0.8) 1.9 (1.2) 0.9 (0.8) 1.4 (1.0) 1.7 (0.9) 1.0 (0.9) 1.0 (1.0) 15.1 (5.4)
1.7 1.7 1.5 1.7 1.9 1.2 0.6 0.5 0.2 2.1n 1.4 0.1 0.1 3.6nn 1.4 1.6 2.8n
1.6 (1.2) 1.8 (1.1) 1.1 (1.2) 0.8 (1.0) 1.7 (1.0) 0.9 (0.9) 1.2 (1.1) 0.7 (0.9) 1.2 (1.0) 1.5 (0.8) 1.8 (1.2) 1.0 (0.9) 1.5 (1.0) 1.5 (1.0) 1.0 (0.9) 1.3 (1.0) 14.6 (5.8)
1.8 (1.1) 1.8 (1.0) 1.0 (1.2) 1.0 (1.0) 1.8 (1.0) 1.2 (0.9) 1.6 (1.1) 0.9 (1.0) 1.4 (0.9) 1.7 (0.7) 1.9 (1.1) 1.0 (0.9) 1.5 (1.1) 1.6 (0.9) 1.0 (0.9) 1.3 (1.0) 15.5 (5.4)
1.6 0.5 0.9 2.8n 2.0 3.9nn 4.0nn 2.3n 1.8 3.6nn 1.4 0.1 0.7 1.8 1.1 0.5 2.7n
FDR-controlled significant variables are given in bold. n
Unipolar non-melancholic
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
355
Table 3 SDS item analyses by gender (severity of symptom). SDS item
Symptom severity – mean (SD) Unipolar melancholic Male
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
Appetite decrease Appetite increase Depressed mood worse in the morning Depressed mood worse in the evening Energy worse in the morning Energy worse in the evening Loss of interest in activities that you would normally enjoy Unable to obtain pleasure from activities Inability to be cheered up when something nice or pleasant occurred Inability to be cheered up by friends Loss of capacity to laugh at humorous things Not being able to look forward to things Difficulty making decisions Brain feeling foggy making concentration difficult Finding it difficult to do basic things (e.g. get out of bed, shower) Feeling physically slowed Find your thinking is slowed Being more agitated Feeling suicidal Feeling very guilty Feeling that you deserve to be punished Feeling irritable Feeling angry Feeling a need to stay away from people Being cheered up completely by being with friends Feeling a greater need to be close to people Feeling somewhat ‘paralysed’ when doing basic things Getting off to sleep takes much longer than usual Waking in the middle of the night more than usual Waking up much earlier than usual in the morning Experiencing distinct weight loss Experiencing distinct weight gain
0.9 (1.0) 0.6 (0.8) 1.9 (1.0) 1.0 (0.9) 2.0 (1.0) 1.0 (0.9) 2.3 (0.7) 2.3 (0.8) 2.1 (0.8) 2.0 (0.7) 1.8 (0.9) 2.3 (0.7) 2.1 (0.9) 2.2 (0.8) 1.9 (1.0) 1.7 (1.0) 1.9 (1.0) 1.4 (1.0) 1.3 (1.1) 1.5 (1.0) 0.7 (0.9) 1.9 (0.8) 1.5 (1.0) 2.1 (0.9) 0.6 (0.8) 0.6 (0.9) 1.7 (1.1) 1.6 (1.1) 1.5 (1.1) 1.1 (1.1) 0.5 (0.9) 0.6 (0.9)
Female 1.2 (1.2) 0.8 (1.0) 2.1 (1.0) 1.2 (1.1) 2.1 (1.0) 1.2 (1.0) 2.5 (0.6) 2.4 (0.7) 2.1 (0.8) 2.0 (0.8) 1.9 (0.9) 2.4 (0.7) 2.3 (0.8) 2.3 (0.9) 2.2 (0.9) 1.8 (1.0) 1.9 (1.0) 1.3 (1.2) 1.3 (1.1) 1.8 (1.1) 1.1 (1.1) 1.8 (1.0) 1.4 (1.0) 2.3 (0.9) 0.6 (0.8) 0.8 (1.0) 2.0 (1.0) 1.6 (1.2) 1.6 (1.1) 1.1 (1.2) 0.8 (1.1) 0.7 (1.0)
Unipolar non-melancholic
t n
2.4 1.4 1.2 2.2n 1.0 0.5 1.8 1.7 0.0 0.1 0.8 2.0n 1.8 1.1 3.1n 0.8 0.4 0.1 0.1 2.1n 2.7n 0.6 0.9 1.4 0.3 1.7 2.1n 0.2 0.6 0.3 2.2n 0.5
Bipolar
Male
Female
t
Male
Female
t
1.1 (1.0) 0.7 (0.9) 1.6 (1.1) 1.3 (1.0) 1.5 (1.0) 1.2 (0.9) 2.1 (0.9) 2.1 (0.9) 1.7 (0.9) 1.7 (0.8) 1.5 (1.0) 2.0 (0.8) 2.1 (0.8) 2.0 (0.9) 1.4 (1.0) 1.4 (1.0) 1.5 (0.9) 1.4 (1.0) 1.3 (1.1) 1.6 (1.0) 0.9 (1.0) 1.9 (0.9) 2.0 (0.9) 2.0 (0.9) 0.7 (0.7) 0.8 (0.9) 1.5 (1.1) 1.6 (1.1) 1.5 (1.1) 1.2 (1.1) 0.5 (0.8) 0.7 (1.0)
1.1 (1.0) 1.0 (1.1) 1.7 (1.1) 1.4 (1.1) 1.8 (1.1) 1.4 (1.0) 2.1 (0.9) 2.0 (1.0) 1.7 (0.9) 1.7 (0.9) 1.6 (1.1) 2.0 (0.9) 2.2 (0.9) 2.1 (0.9) 1.6 (1.1) 1.4 (1.0) 1.6 (1.0) 1.4 (1.1) 1.3 (1.1) 1.7 (1.1) 1.0 (1.2) 2.1 (0.9) 2.0 (1.0) 2.2 (0.9) 0.7 (0.9) 0.8 (0.9) 1.6 (1.1) 1.7 (1.1) 1.7 (1.0) 1.3 (1.1) 0.5 (0.9) 1.0 (1.1)
0.6 3.5nn 1.2 0.9 2.8n 2.8n 0.0 0.2 0.1 0.4 0.4 0.2 0.1 1.1 2.1n 0.7 1.2 0.7 0.7 1.2 1.5 2.8n 1.2 0.5 0.1 0.8 0.1 1.0 2.5n 0.9 0.2 2.9n
1.2 (1.1) 0.8 (1.1) 1.7 (1.1) 1.4 (1.0) 1.8 (1.0) 1.3 (1.0) 2.3 (0.7) 2.2 (0.8) 2.0 (0.8) 1.8 (0.9) 1.7 (0.9) 2.1 (0.8) 2.2 (0.9) 2.3 (0.8) 1.9 (1.0) 1.6 (1.0) 1.8 (1.0) 1.8 (1.0) 1.4 (1.1) 1.9 (1.0) 1.3 (1.1) 2.1 (0.8) 2.0 (1.0) 2.3 (0.8) 0.7 (0.8) 0.7 (1.0) 1.8 (1.0) 1.8 (1.1) 1.6 (1.1) 1.1 (1.1) 0.7 (1.0) 0.7 (1.0)
1.4 (1.1) 1.1 (1.1) 1.9 (1.0) 1.4 (1.0) 2.0 (1.0) 1.3 (1.0) 2.4 (0.8) 2.3 (0.8) 2.0 (0.9) 1.9 (0.9) 1.8 (0.9) 2.3 (0.8) 2.5 (0.7) 2.5 (0.7) 2.1 (0.9) 1.9 (1.0) 2.0 (1.0) 1.8 (1.0) 1.6 (1.1) 2.0 (1.0) 1.4 (1.1) 2.3 (0.9) 2.0 (1.0) 2.5 (0.8) 0.6 (0.7) 0.7 (0.9) 2.0 (1.0) 1.9 (1.1) 1.7 (1.1) 1.2 (1.1) 0.7 (1.0) 1.1 (1.1)
1.8 4.2nn 3.1n 0.7 2.9n 0.6 1.8 1.5 0.2 1.3 2.2n 2.7n 5.4nn 3.8nn 4.3nn 4.1nn 2.9n 0.2 2.6n 1.3 1.2 2.0 0.1 3.0n 1.3 0.1 3.1n 2.0 3.0n 1.0 0.0 5.0nn
FDR-controlled significant variables are bolded. n
P o 0.05. P o0.001.
nn
more difficult than most people (51.3% vs. 42.5%) and relationships and work performance being generally good when not depressed (71.3% vs. 62.2%)—and two ‘non-melancholic’ items: food cravings and increased appetite (50.2% vs. 34.0%) and, when not depressed, tending to worry more than most people (62.5% vs. 53.6%).
4. Discussion Women have been widely demonstrated to show higher levels and ‘caseness’ rates of depression reflecting both ‘real’ and artefactual effects. In relation to the latter, we sought to examine any impact that actual depression measures might contribute, whether by influencing severity or prevalence of item ratings by men and women (a response bias) or by elements of depression having differential salience to men or to women (a gender dimorphic weighting). We acknowledge that differences in prevalence rates by gender are generally assessed by ‘case’ measures rather than by our study measures of severity, but note (as stated in Section 1) that women have been shown to both return higher case rates and score higher on depression severity scores. Secondly, rather than study ‘clinical depression’ as an overall category, we considered three depressive sub-types: melancholic, nonmelancholic and bipolar depression. By studying both depression severity and prevalence in relation to depressive sub-types we were able to further determine where gender had its greatest differential impact, an approach not previously explored. By
examining several differing measures we could determine whether there was consistency in any identified gender differences. While not all patients completed all measures, the numbers completing the four measures under examination were substantive and reject any judgment that the study was under-powered. If anything, the large numbers risked according significant differences when mean scores or prevalence rates across the genders were actually slight, although we do also report statistical significance after controlling for the type I error rate. As noted, we examined one brief measure of depression severity (the DMI-10), with items weighting the phenomenological expression of depression and not including any of the somatic or vegetative features that commonly accompany clinical depression. Women scored comparably to men (i.e. only 1% higher in the melancholic, 3% higher in the non-melancholic and 5% higher in the bipolar sub-set, and with only the last being formally significant) in affirming the prevalence of individual items. That difference in the last group—if not due to chance and the large sample sub-set (as it was no longer significant after FDR analysis) —might reflect greater mood amplification by females (i.e. scoring higher on feeling vulnerable, self-criticism and guilt) or a sociological difference in the female experience of bipolar depression. Due to the large sample size and relatively small difference between male and female percentages, results would benefit from replication before firm conclusions are drawn. In essence, the DMI-10 appeared essentially uninfluenced by gender in those with unipolar depression.
356
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Our second measure—the QIDS-SR—generates both an estimate of severity and the prevalence of depressive sub-type symptoms. As a severity measure, women with melancholia scored nonsignificantly higher (6%) than men, while those with nonmelancholic depression and bipolar depression scored significantly higher (9% and 6% respectively). The essentially comparable percentage differences across the three depressive sub-types would suggest that formal statistical significance or nonsignificance was presumably influenced by sample sub-set sizes. When individual QIDS-SR items were examined, few differences were evident in the unipolar (melancholic and non-melancholic sub-sets) while numerous differences were established in the bipolar sub-set, and with such differential pattern across the three diagnostic groups remaining after controlling for multiple comparisons. In essence, on both measures of severity (i.e. DMI-10 and QIDS-SR) we observe a general trend for women to score minimally higher than men, with the DMI-10 difference being marginal. Again such a finding argues against any gender-specific response bias in rating depression severity, and that these measures do not appear sensitive to any gender bias in their indication of severity. Our remaining two measures (i.e. the SDS and SMPI) were designed to have representative melancholic and non-melancholic items and therefore allow symptom representation across men and women to be examined rather than depression severity levels. As for the QIDS-SR, the greatest number of gender differences across items were quantified in the bipolar patients (and still clearly evident after controlling for multiple comparisons), a key finding that will be returned to shortly. Our examination of three measures assessing clinical depressive symptoms (as against mood severity alone) allows consideration of consistency of item affirmation to be considered and thus allows some estimate of the validity of findings. For the unipolar melancholic sub-set, there was a consistent finding of women reporting greater weight loss and anergia (albeit only a trend for the SMPI) than men. For the unipolar non-melancholic sub-set, there were no significant gender differences across the 24 SMPI items and no consistent differences when QIDS-SR and SDS findings were compared. For the bipolar sub-set women appeared consistently more likely to report appetite and weight changes (most commonly increased and associated in the SMPI with food cravings), to have low energy and feel physically slowed and to have impaired concentration. Validity of findings in studies of this nature might be addressed by also administering an interviewerrated measure for examination against the self-report measures. In relation to the general finding that women have higher rates of lifetime depression, this study allows us to examine whether this might be contributed to by measurement response bias— whereby women might have a general tendency to affirm more depressive items and/or rate the severity of any experienced items in an amplified way when completing depression measures. We find little support for this proposition in our unipolar depressed patients (whether assigned as melancholic or non-melancholic) and therefore suggest that any such measurement error is unlikely to contribute to the general finding. We did, however, quantify greater severity of depression in women with a bipolar disorder— although when quantified by total scores the magnitude was low (i.e. 5% for DMI-10 and 6% for QIDS-SR). We suggest that such a difference was unlikely to reflect any response bias or amplification effect (as would have then expected to observe women rating higher on all depressive symptoms) and could reflect real gender differences in symptom prevalence and severity, particularly more substantive changes in appetite, psychomotor disturbance and other features weighted to a melancholic depression. Why women with a bipolar disorder (compared to those with a unipolar disorder) might rate such features as more prevalent or more
severe than men remains unclear and requires replication as we are not aware of any previous study examining for gender differences in depressive symptoms in bipolar disorder. Kessler (2005) overviewed data indicating that the female preponderance in clinical depression reflected women being more likely than men to have an initial episode (and that once depression had occurred, gender had little impact on repeat episodes). The latter finding would not be expected if the female preponderance was simply or somewhat underpinned by response biases in completing depression rating measures. This study similarly finds no support for postulating that gender-based response biases in completing depression measures contributes to the differential community rates. Study findings identify some clinical depressive symptoms that do appear to be differentially experienced by men and women when depressed. While greater variation in eating and weight symptoms was expected, the finding that many melancholic symptoms (e.g. anergia, feeling slowed down and impaired concentration) were rated as more prevalent and/or severe was unexpected. As the greatest differentiation in such symptoms was quantified in the bipolar sub-set (and where bipolar depression is seemingly most commonly melancholic in type), it would appear that women may experience melancholic depression more severely than men. Future research examining pre-menopausal and post-menopausal women with aged matched men would be of interest, as would other potentially explanatory studies.
Role of funding source All authors declare that the funding source had no impact on the study design, the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest All authors declare no conflict of interest.
Acknowledgements We acknowledge the assistance of Georgia McClure with database management and Dusan Hadzi-Pavlovic for statistical advice.. This study was supported by an Australian National Health and Medical Research Council (NHMRC) Program Grant 1037196.
References Bernstein, I.H., Rush, A.J., Suppes, T., Kyotoku, Y., Warden, D., 2010. The quick inventory of depressive symptomatology (clinician and self-report versions) in patients with bipolar disorder. CNS Spectr. 15, 367–373. Goodwin, F., Jamison, K., 2007. Manic Depressive IllnessOxford University Press, New York. Hildebrandt, M.G., Stage, K.B., Kragh-Soerensen, P., 2003. Gender differences in severity, symptomatology and distribution of melancholia in major depression. Psychopathology 36, 204–212. Hilton, T.M., Parker, G., McDonald, et al., 2006. A validation study of two brief measures of depression in the cardiac population: the DMI-10 and DMI-18. Psychosomatics 47, 129–135. Kessler, R.C., 2005. Gender differences in major depression. Epidemiological findings. In: Frank, E (Ed.), Gender and Its Effects on Psychopathology. American Psychiatric Press, Washington. Khan, A.A., Gardner, C.O., Prescott, C.A., Kendler, K.S., 2002. Gender differences in the symptoms of major depression in opposite-sex dizygotic twin pairs. Am. J. Psychiatry 159, 1427–1429. Kornstein, S.G., Schatzberg, A.F., Thase, M.E., 2000. Gender differences in chronic major and double depression. J. Affect. Disord. 60, 1–11. Marcus, S.M., Young, E.A., Kerber, K.B., 2005. Gender differences in depression: findings from the STARnD study. J. Affect. Disord. 87, 141–150. Nolen-Hoeksema, S., Parker, L.E., Larsen, J., 1994. Ruminative coping with depressed mood following loss. J. Personal. Soc. Psychol. 67, 92–104. Parker, G., Brotchie, H., 2010. Gender differences in depression. Int. Rev. Psychiatry 22, 429–436. Parker, G., Fletcher, K., Barrett, M., Synnott, M., Breakspear, M., Rees, A.M., HadziPavlovic, D., 2010. Inching toward Bethlehem: mapping melancholia. J. Affect. Disord. 123, 291–298.
G. Parker et al. / Journal of Affective Disorders 167 (2014) 351–357
Parker, G., Fletcher, K., Hadzi-Pavlovic, D., 2012. Is context everything to the definition of clinical depression? A test of the Horwitz and Wakefield postulate. J. Affect. Disord. 136, 1034–1038. Parker, G., Fletcher, K., Hyett, M., 2008. The mood assessment program: a computerised diagnostic tool for deriving management plans for mood disorders. Med. J. Aust. 188 (12), S126–S128. Parker, G., Fletcher, K., Hyett, M., Hadzi-Pavlovic, D., Barrett, M., Synnott, H., 2009. Measuring melancholia: the utility of a prototypic symptom approach. Psychol. Med. 39, 989–998. Parker, G., Hadzi-Pavlovic, D., 1996. Melancholia: A Disorder of Movement and MoodCambridge University Press, New York. Parker, G., Hilton, T., Bains, J., Hadzi-Pavlovic, D., 2002. Cognitive-based measures screening for depression in the medically ill: the DMI-10 and the DMI-18. Acta Psychiatr. Scand. 105, 419–426.
357
Parker, G., McCraw, S., Blanch, B., Hadzi-Pavlovic, D., Synnott, H., Rees, A.M., 2013a. Discriminating melancholic and non-melancholic depression by prototypic clinical features. J. Affect. Disord. 25, 199–207. Parker, G., McCraw, S., Hadzi-Pavlovic, D., Hong, M., Barrett, M., 2013b. Bipolar depression: prototypically melancholic in its clinical features. J. Affect. Disord. 147, 331–337. Piccinelli, M., Wilkison, G., 2000. Gender differences in depression. Br. J. Psychiatry 177, 486–492. Rush, A.J., Trivedi, M.H., Ibrahim, et al., 2003. The 16-item quick inventory of depressive symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol. Psychiatry 54, 573–583. SAS Institute Inc., 2013. SAS/STAT 13.1 User's GuideSAS Institute Inc., Cary, NC.
