Research Reports

Evaluation of the SHIFT-Depression® Inventory With a Sample of Australian Women, Demonstrating the Centrality of Gendered Role Expectations to Their Depression

Helen C. Vidler*a


To address the double fold incidence and prevalence rates of depression in women compared to men needs more than symptom management. In primary health care where time does not normally allow for thorough assessments of what causes or maintains depression, using a brief questionnaire is warranted. A means must be found to address system level barriers to care and the poor quality of life so many women experience. Using an online survey and mixed method design, initial evaluation of a measure developed to identify individual and contextual issues connected to depression in women, examined underlying factor structure. 266 depressed women aged 18 to 85yrs also completed the Centre for Epidemiological Studies Depression scale (CESD) and provided written commentary about any further issues. The SHIFT-Depression® Inventory analysis revealed five reliable underlying factors explaining 59% of the variance. The factors identified were; 'diminished self/feels powerless/focused on other's care'; 'financial problems, lacks access to healthcare, healthy food, housing, exercise’; 'relationship difficulties, loss/betrayal/abuse’; ‘women’s problems’ and 'lack of social support/feels isolated'. Analysis of participant’s comments found four groupings similar to the identified factors. This initial evaluation of the SHIFT-Depression® Inventory showed evidence of the multiple issues impacting on depressed women spanning the physical, intrapsychic, relational, social and contextual areas. The inventory provides preliminary assessment which can identify when more extensive questioning and referrals to various services may be appropriate. It could be used in primary health care settings when consultation times are brief, or as an adjunct to assessment in the mental health setting.

Keywords: women, depression, caring, relational, social, emotion-work, mixed-methods

Europe's Journal of Psychology, 2013, Vol. 9(2), doi:10.5964/ejop.v9i2.577

Received: 24 January 2013. Accepted: 17 April 2013. Published (electronic): 31 May 2013.

*Corresponding author at: School of Social Sciences and Psychology, University of Western Sydney, P.O. Box 112, ROZELLE NSW 2039, Australia. E-mail:

This is an open access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

‘I need a partner who takes the time to understand and support me. This is not going to happen. I have a young child and I’m not in a position to leave this relationship. I feel trapped.’

Participant’s comment

Introduction [TOP]

Background [TOP]

It is well recognized that depression is a primary cause of disability and a priority public health concern (Lopez, Mathers, Essati, Jamison, & Murray, 2006; WHO, 2001). Also, it is well recognized that there is consistent reporting of higher rates of depression for women compared to men (Australian Institute of Health and Welfare, 1998; McLennan, 1998; WHO, 2000, 2004). This poor quality of life for depressed individuals and for women in particular, warrants urgent attention to what causes, maintains and can help reduce it.

Most people with unipolar depression present to their general practitioner with a combination of physical complaints which makes it more likely the focus of treatment will be on physical issues with the subject of depression not raised. As well, time-poor general practitioners in primary health care settings, even when they do diagnose depression, frequently only have the time to focus on biomedical aspects such as exercise and/or prescribing anti-depressants.

Even though guidelines from the World Health Organisation (WHO) and Australia for the management of depression in primary health care recommend enquiry about recent losses or relationship difficulties and the provision of psycho-education, problem solving therapy or cognitive behaviour therapy (Ellis & Smith, 2002; WHO, 2010), the majority of busy general practitioners do not have the time or resources to focus on thorough assessments or provide such treatment. The option to refer to mental health practitioners for psychotherapeutic interventions varies depending upon a country’s funding arrangements. Referrals to mental health professionals in Australia are usually reserved for moderate to severe cases of depression.

Unfortunately therefore, many people who are depressed do not get diagnosed. As well, many who are diagnosed do not get adequate depression care management and this systemic problem increases the likelihood of an individual’s depression becoming chronic or reoccurring. This is of particular concern due to the higher rates of depression for women.

Estimates indicate that 50% of depression is undiagnosed and untreated according to Marcus, Yasamy, van Ommeren, Chisholm, and Saxena (2012). In Europe (Berghöfer, Hartwich, Bauer, Unützer, Willich, & Pfennig, 2012) as well as in the US (Katon & Ludman, 2003), it has been identified that there is a need for systematic management of depression in primary health care. Griffiths and Christensen (2008), from having conducted several systematic reviews of primary care models of depression management have identified the need for a reorganisation of the traditional model of general practice and, ‘in particular, the implementation of care management as well as enhanced care and guided self-help in these settings’ (Griffiths & Christensen, 2008, p. 81). These findings confirm the earlier work of Wagner (1997) and yet it remains the case that there are exceptionally high rates of depression reported for women which are left unaddressed.

Apart from understanding that general practitioners are expected to perform the impossible in short consultation times, we need to understand if there are other factors impacting upon primary health care delivery that may make the double fold incidence and prevalence rates of depression for women more understandable? We may ask whether cultural views regarding the essential characteristics of womanhood demonstrated by sex role stereotypes (Broverman, Vogel, Broverman, Clarkson, & Rosenkrantz, 1972), together with research from the biomedical area which attributes women’s fluctuating hormones, biochemistry and personality as causes for their depression, have ‘normalised’ depression in women. If health practitioners subscribe to sex role stereotypes that regard women as naturally emotional and sensitive (Nicolson, 1992), and there is no reason why they would not, then it is easy to see how the high incidence and prevalence rates of depression amongst women may be regarded as ‘normal’.

Certainly biomedical research does not have a history of seeking the beliefs and viewpoints of women regarding their depression (Dill & Anderson, 1999; Farmer, Owen, & McGuffin, 2000; Post, 1992). Therefore, if causes for depression in women are seen to be fluctuating hormones, malfunctioning biochemistry and ‘emotional’ or ‘sensitive’ personality traits, then the primary health care sector may see depression in women as inevitable. Unfortunately, such a ‘normalizing’ perspective of the experience of depression for women may preclude an inspection of the interpersonal, social and contextual issues that may be causing or maintaining depression.

If we look instead to research from the social sciences, it shows that during the reproductive years, when the rates of depression for women are highest, women are also more vulnerable and dependent due to their decreased ability to earn an income whilst also caring for children (Belle & Doucet, 2003). This makes women dependent upon their partner (if they have one) and necessitates that the relationship is stable and harmonious. Yet we know from data regarding domestic violence that there are many relationships which are violent, controlling, or otherwise difficult (Haggett, 2012; Koss & Heslet, 1992; Resnick, Acierno, & Kilpatrick, 1997). During the reproductive years, women are frequently financially dependent and can feel trapped and defeated within difficult relationships. (Brown, Harris, & Hepworth, 1995). This is hardly good for their mental health. Reported high prevalence rates, although dipping after the reproductive years, peak again in later life (Wu & Anthony, 2000). Later life, for many women who have not had the opportunity to accrue superannuation, or have been the recipients of lower wages due to gender pay gap issues (International Trade Union Confederation, 2008; Workplace Gender Equality Agency, 2012), means limited choice in their living situations due to having limited financial means. Once again, women may remain stuck in difficult relationships or affordable housing that they feel unsafe in. Surely, these social contexts have much to do with why a woman became depressed or stays depressed and need to be included as a matter of course in any assessment and treatment plan.

It is necessary to make a comprehensive assessment that includes not only individual biomedical and intrapsychic issues but also the psychosocial contextual areas, so that the circumstances impacting on a woman become clear. Then appropriate referrals can be made to a mental health practitioner, social worker, accommodation services, domestic violence support, financial or employment counselling. But if there is not enough time in the primary health care setting to do this, and there is also a disconnect between the primary care and specialist mental health sectors, there is no opportunity to do this assessment.

Therefore, there is an urgent need for an efficient and brief way to conduct an assessment which can identify each of these areas and can inform the direction that treatment should take.

This identified need underlies the purpose of the research reported here. This assessment would be helpful for all depression sufferers but particularly so for women when we consider their double fold incidence and prevalence rates of depression. There is nothing ‘normal’ about so many women experiencing such a poor quality of life.

Purpose of This Study [TOP]

Therefore, this research is based on an identified gap in the assessment and management of depression care for women that combines the individual and environmental issues. There is likely to also be a gap in mental health care settings where assessment and treatment may give preference to psychosocial/contextual issues and leave off enquiry about and referral for physical/ biomedical issues.

If depression is experienced as an interaction between poor physical health including; lack of sleep, the use of alcohol and other drugs, inadequate or unbalanced diet, and inability to exercise; as well as the intrapsychic, interpersonal, and psychosocial/contextual issues, then asking questions about each of these areas forms the basis of a comprehensive assessment.

Katzelnick et al. (2000) identified that controlled studies on depression care management programs designed specifically to optimize the treatment of depression in high utilizers of primary care are rare. For a review of studies that have used various measurements and tools or algorithms for the treatment of depression in primary care see Berghöfer et al. (2012).

Recent searching of the PubMed data base using the terms depression, assessment, management, treatment, tool, measure and women, returned no studies or measures that are concerned with depression management for women in particular (high utilizers of primary health care) that combine the physical, relational, social and contextual areas of women’s lives.

The current research builds upon an earlier mixed methods study with a sample of 22 women who were recruited from the Australian cohort of a large, longitudinal study as they exited that study. Participants in the large longitudinal study had been recruited whilst waiting for the general practitioner in urban community health centres and had all been diagnosed as clinically depressed (Herrman, Diehr, Fleck, Simon, & Patrick, 2002). The 22 women from the sample the current research builds upon, completed in-depth interviews as well as the Centre for Epidemiological Studies Depression Inventory (CESD) (Radloff, 1977), the Silencing the Self Scale (STSS) (Jack, 1991), and the preliminary version of the SHIFT-Depression® Inventory used in the current study. The inventory results showed there was a strong link between depression and the physical, intrapsychic, interpersonal, social, cultural and contextual factors for the women in that small sample.

The interview transcripts from the in-depth interviews were examined from a hermeneutic phenomenological understanding and themes were identified that summarized the essential issues connected to being depressed for that small sample of women. The qualitative findings showed that those women who sacrificed their own needs and preferenced the needs of others in their significant interpersonal relationships continued to be depressed whereas those women who decided to rebalance the care in those important relationships and to care for themselves more, recovered from depression. For more detail regarding the initial development and psychometric properties of the 32 item form of the inventory, along with the method used to locate themes from the in-depth interviews the reader is referred to (Vidler, 2005, 2006).

The current work also used a mixed, quantitative and qualitative method. Brannen (1992) argues for the combining of methods to best serve social groups or populations where one method alone may misrepresent them.

While the qualitative approach may overcome some of the problems of giving a voice and language to such groups through which they may better express their experiences, the quantitative approach would serve to indicate the extent and patterns of their inequality at particular historical junctures (Brannen, 1992, p. 22).

Using a mixed method approach in this way also enables a linking of the biomedical area, frequently associated with quantitative methods, with the social area usually associated with qualitative methods.

Design of the Study [TOP]

The overarching design of the study was informed by a transformative-emancipatory approach together with a constructivist/interpretative paradigm. According to Mertens (2007), the transformative-emancipatory approach provides a framework for a research process that can address social injustice and inequality because it makes apparent the research participant’s reality. It is acknowledged that this reality is constructed and shaped by social, political, cultural, economic and racial/ethnic values, power and privilege; which are important determinants of what will be privileged in the research context.

Because this study was concerned with bringing to light contextual issues that have been shown to be connected to depression for women, but are not routinely addressed in biomedical research, it was important to use an approach that preferences the research participant’s contributions. The transformative paradigm can also involve the use of both quantitative and qualitative methods and considers it a given that participant’s perspectives inform the work.

The 32 item multifactorial inventory developed for the earlier study to assist women to rank issues from the physical, stressful life experiences, thoughts and feelings, relationships with others, community and culture, and women’s problems areas (Vidler, 2005, 2006) has been further developed for the current study. Based upon comments women made during the in-depth interviews in the previous study, the 32 items were further modified and broadened into the current 38 item inventory. As a part of the development of the current inventory, the items were reviewed separately by five women who had knowledge of depression due to either, their work as a mental health professional, having suffered with depression in the past, or having had a family member who was depressed. The first of the five women was identified by the author and the remaining reviewers were identified by using a snowball sampling technique. It is important, according to Worthington and Whittaker (2006), to have an expert group with members who have similar characteristics to the group being studied to provide input about what methods or measures are being used. This ‘expert group’ provided valuable input regarding the meaning and grammar of various items. This resulted in the final version of the SHIFT-Depression® Inventory which asked for a response to each item using a five point Likert scale. This list of ‘things connected to your sadness or depression’ informed by depressed women, formed the basis for the online survey used in the current research.

The two research questions for this current study were:

  1. What is the underlying factor structure of the SHIFT-Depression® Inventory?

  2. Does the factor structure of the inventory provide evidence of physical, psychological (intrapsychic), relational (interpersonal), social and contextual issues connected to depression for the women in the sample?

The project had the approval of the Human Research Ethics Committee at the University.

Methods [TOP]

A decision was made to use an online survey method because it enabled immediate access to a large sample of women who self-identified as depressed that would otherwise be hard to reach. Another benefit of this method was that it provided the potential for privacy and anonymity that would not be available if the inventory had been completed in the presence of another person.

The Sample [TOP]

A convenience sample of 266 women was recruited to the project from the community by advertising on the University web site, local and national newspapers and from radio interview. Information was provided about the research, details of the HREC approval, and how to contact the principal investigator (PI) if they required further detail about the study. The two inclusion criteria imposed when recruiting women to the study were; they needed to be 18 years of age or older, and believe that they were sad or depressed.

Fricker (2008), in discussing various sampling methods for online surveys explained that unrestricted, self-selected surveys are useful because they leave it up to the individual to choose to participate or opt-in. They have advantages in their potential to reach a wide audience over a short time span and are very cost effective. Nevertheless, this type of sampling also posed limitations to this study due to the inability of generalizing the results to all women. The study participants may not represent depressed women in general, or may over or under represent certain age groups particularly those who are more familiar than others with computers and accessing the internet. It is also possible, due to the anonymity available to participants that their responses may have been under or over exaggerated or misrepresented, or may be from men as well as women. Therefore, this must only be considered an exploratory study.

The Online Survey [TOP]

The survey began with an introductory page providing information about how to contact the ethics committee who had approved the study, how the information they provided was confidential and private and how their privacy would be protected, plus the email address for the PI for further information about the study. There was also a statement making it clear that there was no ability to respond to any person who was in a crisis and that the survey was not intended to provide an individual diagnosis or treatment plan. Participants were informed that if they felt they may self-harm or felt suicidal that they should immediately go to their nearest hospital outpatient department by calling an ambulance or asking someone to take them there. Also provided was the telephone number for an all hours support line. It was very important that the participant’s safety was considered due to a slight risk that the online inventory, with questions relating to sadness or depression, may make them feel worse. It is important to consider the level of risk when using online methods according to Proudfoot et al. (2010), because the online survey lacks the level of control available when there is face to face contact.

Participants were also advised that if they provided a functional email address at the end of the survey, they would be emailed a copy of their completed survey and if they wished, they could take it to their health practitioner to assist with treatment planning. Along with their completed survey, they would also receive an information sheet containing the names of a range of community and non-government agencies providing support with depression, and information about how to access support from the private health sector, plus several links to web sites that also provided information regarding depression. Participants were advised that by clicking on ‘Begin the Survey’, they were providing their consent to participate. The survey was open between July and September 2009.

After the introductory page participants were presented with the 20 item version of the Centre for Epidemiological Studies Depression measure (CESD) (Radloff, 1977), using a four point Likert scale. The scoring ranges from 0 (rarely or none of the time – less than one day of the week), 1 (some or a little of the time – one to two days of the week), 2 (occasionally or a moderate amount of time – three to four days of the week), to 3 (most or all of the time – five to seven days of the week). The CESD is routinely used in a research setting to provide an indication of a person’s level of depressive symptoms. It is not a clinical diagnostic tool. Permission to adapt the measure into an online survey was gained from the copyright owners of the CESD, the National Institutes of Mental Health (NIMH).

This measure was included to gauge the level of depressive symptomology of women even though women themselves had self-assessed as being sad or depressed. Having a measure indicating the level of depressive symptoms provided a valuable cross check and a way of providing a level of data triangulation.

The CESD contains a list of 20 symptoms drawn from items used in previously validated longer scales of depression. The major components of depressive symptomatology were identified as depressed mood, feelings of guilt or worthlessness, feelings of helplessness and hopelessness, psychomotor retardation, loss of appetite and sleep disturbance. Several items are included in the scale to reflect each component. Response expectations for the general population suggest that many people would experience a few of the symptoms and a few people would experience many whereas expectations for a clinical sample are the reverse. With a clinical sample there is an expectation of higher item means, higher inter item correlations and very high internal consistency. It is common when using the full version of the CESD to use a cut point of 16, with a score of 16 or more indicating sufficient depressive symptomatology to suggest the presence of depression and a score of 15 or less indicating the absence of depression.

After participants completed the CESD, the 38 item SHIFT-Depression® Inventory followed with the question, ‘Please make a selection to indicate how much any of these things were connected with your sadness or depression over the past four weeks’. The relevance and level of importance to the participants of each item was measured using a five point scale. The possible scores ranged from 0 – 4 (0 = not at all, 1 = a little, 2 = moderately, 3 = quite a bit, and 4 = extremely). In this way, women indicated the relevance to them of the statements provided.

Immediately after completing the inventory, an open-ended question was asked to provide further information and some balance to the closed-ended questions in the questionnaires. This question had a text box provided for a response to ‘Is there anything you wish to add?’ This was then followed by another question and text box ‘Any further comments?’

The two text boxes gave women the opportunity to contribute information that was more detailed than the survey otherwise allowed for. Leaving room for women’s comments provided a way for women to build upon their answers to questions in the measures they may otherwise have felt were rigid and did not apply to them or were difficult to answer. This detailed qualitative information made it possible to bring greater insight than would have been available from only having the measures completed. The final page of the survey asked participants for their age, gender, relationship status, and email address. Finally, women were thanked for having taken part in the survey.

Data Procedures Quantitative [TOP]

The data from the inventory was deemed suitable for conducting an exploratory principal components analysis (PCA) because (a) there were 266 completed data sets for all 38 items and this number represents a fair to good number for factor analysis according to Henson and Roberts (2006), and, (b) inspection of the correlation matrix revealed the presence of many coefficients of .3 and above. As well, the strength of the inter-correlation amongst items was confirmed from the Kaiser-Meyer-Olkin value which was .868, exceeding the value of .6 recommended by Kaiser (1974). As well, Bartlett’s Test of Sphericity (Bartlett, 1954) reached statistical significance at p < .000 supporting the factorability of the correlation matrix.

Initial principal components analysis (PCA) revealed the presence of nine components with eigenvalues exceeding 1. A parallel analysis was performed on those components and, of the values exceeding the corresponding criterion values for a randomly generated data matrix of the same size (38 variables x 266 respondents), only six were retained. The oblique method was used to rotate the data and provide loadings on structure/pattern matrices together with communalities. Values of less than .40 in the communalities matrix resulted in eight variables being dropped and further PCA conducted. The final EFA grouped the 30 variables into six components with eigenvalues greater than 1. These eigenvalues were scrutinized using parallel analysis.

A test of the correlation between the two measures, the CESD and the SHIFT-Depression® Inventory was conducted using Pearson’s product moment correlation coefficient and the internal consistency of both was examined. The data analysis was conducted using SPSS Version 18.

Data Procedures Qualitative [TOP]

The naming of the factors after the final EFA was an inductive, subjective process whereby the collective verbs and nouns from the factor items were synthesised to develop the inherent meaning, a procedure similar to that outlined by Wei, Russell, Mallinckrodt, and Vogel (2007, p. 190).

There were 146 comments provided at the end of the measures in answer to (a)is there anything you wish to add?’ (n = 56) and (b)any further comments?’ (n = 90). Responses ranged from a few words to several hundred words. Because the two questions posed were essentially the same, a thematic content analysis was conducted on the combined comments (n = 146) to locate the meaning and the context for the participants, a procedure outlined by Grbich (1999).

Results [TOP]

Participants [TOP]

Of the 266 participants who began the survey and completed the measures, 222 participants completed the final page and provided information about gender, age, relationship status and email address. All 222 participants identified as female, 117 (53%) were either married or in a defacto relationship and living together; 12 (5%) were in a relationship but not living together; 23 (10%) were either divorced, separated or widowed, and 70 (32%) were single.

Of the 222 participants, 40 (18%) were aged 18 – 24yrs; 61 (27.5%) aged between 25 – 34yrs; 45 (20.25%) aged 35 – 44yrs; 45 (20.25%) aged 45 – 54yrs; 27 (12%) aged 55 – 64yrs; and 4 (2%) were aged 65yrs or more. The majority of the sample (65.75%) was aged 44yrs or less, the age period when most women who will have children do so, followed by aged 45yrs and older (34.25%).

Quantitative Data [TOP]

The mean CESD score for all participants was 31.79 (SD = 11.44) which shows a CESD score almost double the usual cut point of 16 generally applied in research settings. The mean inventory score for all participants was 55.66 (SD = 24.65).

The relationship between items on the SHIFT-Depression® Inventory and items on the CESD was examined using Pearson product-moment correlation coefficient. There was a strong, positive correlation between the two variables (r = .69, n = 266, p < .000), meaning a high score on the CESD was correlated to a high score on the SHIFT-Depression® Inventory. Cronbach’s Alpha for the CESD was .900 and for the SHIFT-Depression® Inventory .910. Both of these results exceed the recommended reliability standard of .7 recommended by DeVellis (2003) for an effective scale and indicates that the inventory has good internal consistency.

Exploratory Factor Analysis [TOP]

After conducting the final PCA on the 30 items of the SHIFT-Depression® Inventory, six components explained 59% of the variance with 29.5% applied to the first component followed by 7.5%, 6.6%, 5.7%, 5.3%, and 4.4%. Eigenvalues were compared using a parallel analysis showing that the sixth eigenvalue of 1.340 was weak when compared to the random eigenvalue of 1.3598. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy was .869 and Bartlett’s Test of Sphericity was significant at p < .000.

Examining the loadings in the structure/pattern matrices showed only one variable loading on the sixth component. Because it also loaded on the first component and the parallel analysis suggested that component was questionable, the six components were reduced to five. Only those variables with a value equal to or higher than .45 were retained to explain each factor. The five components or factors shown below are outlined with their factor loadings and the corresponding items from the SHIFT-Depression® Inventory in the Appendix.

  1. Diminished self, feels powerless, focused on other’s needs,

  2. Financial problems, lacks access to healthcare, healthy food, housing or exercise,

  3. Relationship difficulties, loss/betrayal/abuse,

  4. Women’s problems,

  5. Lack of social support, feels isolated.

Qualitative Data [TOP]

A thematic content analysis of the comments provided by women to the two questions at the end of the measures resulted in groupings that were similar to the factors identified above. Of the 134 comments provided, many were a ‘thank you’, or a comment about the ease or difficulty of completing the survey. Some participants provided a narrative regarding their particular life circumstances which appeared to show an appreciation of the opportunity to provide information in an anonymous way. From the 134 comments provided, 48 were retained because they related closely to one or more of the five factors identified above. These comments were able to be grouped into;

  1. Relationship difficulties, loss/betrayal/abuse, or discrimination (19 comments)

    For example; ‘I have had a lot of losses in the past couple of years and lost most of my family and friends within a short time of each other’, ‘I think one of the main reasons for feeling sad & depressed is a lack of an intimate, trusting and on-going romantic relationship’. ‘My current situation is related to my age and the feeling that society has branded me as useless solely because of my age (I am 60yrs)’.

  2. Diminished self, feels powerless, focused on other’s needs (17 comments)

    For example; ‘With two small children, a busy job four days a week, running a household, a husband who is lovely but also very busy, there is just not enough time in the day to get everything done and to also put away some time for yourself’. ‘I’m chronically frustrated and overwhelmed with the sheer amount of 24hr work three children under three years involves’.

  3. Lack of support and feeling isolated (8 comments)

    For example; ‘I love my partner very much but still feel lonely. I don’t have any friends close by and I avoid social situations as I am not good around people. I also won’t contact the friends I have’, ‘It would be good if there was some sort of support out there to help people before they get to the point of depression. This would be especially helpful to people who are isolated or have a small or non-existent social network’.

  4. Lack of money and access to health care (4 comments)

    For example; ‘Women are generally expected to do too much and often have poorer pay than men. This makes it difficult to get medicines’. ‘A doctor who cared and looked up from the computer screen would help’, ‘I felt like the doctor wasn’t interested’.

Discussion [TOP]

The final PCA conducted with 30 items remaining from the initial 38 items, produced five components or factors. The majority of the variance (29.5%), of a total variance of 59.26%, was explained by the first factor ‘Diminished self/feels powerless/focused on other’s needs’. Most of the items in this factor describe the very experience of being depressed for this sample of women. The feelings of powerlessness, the self-criticism, the negative thoughts and feelings, feeling worried and anxious, and feeling stuck in an unhappy situation. These all speak of the essence of feeling depressed and are not novel. Alongside of these items were also three items that showed that women’s behaviour was focused on the care of others or their responsibilities, and that this focus took precedence over their own self-care.

T6. Feeling overwhelmed and responsible for everything,

T7. I don’t take the time or do what I need for myself, and

R7. I feel I must attend to other’s needs before I look after myself.

The importance of these three items was confirmed by the comments provided by women later on in the survey which showed that women felt overwhelmed with their caring responsibilities and were feeling completely alone in this role.

This finding differs from a commonly held belief that a depressed person is passive or withdrawn, unable to fulfil their responsibilities or go to work, and unable to care for others. Rather than this suggesting a low level of depression, it suggests instead that it is expected of women to continue on even though they are unwell.

The second factor, ‘Financial problems/lack of access to healthcare, healthy food, housing, exercise’, explained 7.5% of the total variance. Depression was strongly linked to the experience of poor physical health and somatic complaints. As well, it was clear that many of these women also felt they could not access the care they needed or wanted. From women’s comments later on, the experience of financial hardship was linked to not being able to get the care they needed. They also experienced health professionals who seemed disinterested, or they experienced negative reactions to anti-depressants.

The third factor, accounting for 6.6% of the variance, was concerned with ‘Relationship difficulties, loss/betrayal/abuse’. The items here cover a broad range of situations where women experienced violations sexually, physically, or emotionally, or experienced loss or felt betrayed. Several of the comments women provided at the end of the survey demonstrated that they did not receive any care or support from their significant others or families but instead endured relationships that left them feeling worthless, uncared for and depleted.

The next factor ‘Women’s Problems’ explained 5.7% of the variance and contained two items ‘problems with my pregnancy’ and ‘problems with the birth of my baby’.

It is unknown just from these items whether the problems with a pregnancy or birth of a baby meant situational problems, emotional problems or physical health problems and as with all of the other items on the inventory, these items would need further inquiry if conducting an assessment so that they can be better understood. The result does not provide evidence for post natal depression produced by hormonal fluctuations. Going through a pregnancy and giving birth can be a difficult and life changing event and is enough to trigger a depressive episode by itself. Most women do not receive a risk assessment before pregnancy for comparison during pregnancy and again after childbirth. Depression after childbirth would be better understood generally if women were assessed for risk and monitored during the pregnancy.

The final and fifth factor ‘Lack of social support/feels isolated’ provided further evidence for the interpersonal and social elements of depression for the women in this sample. Being unable to get support, not knowing where to find it, feeling alone, feeling like an outsider in one’s community, not having social networks, lacking trustworthy, confiding relationships, or feeling people were judgmental or critical, would be felt by most as emotionally intolerable. This factor accounted for 4.4% of the variance and provided a window into the contexts in which these women lived their lives on a daily basis. Experiencing a lack of support and feeling isolated can trigger depression as well as maintain it.

There were eight items on the inventory which were dropped from the final analysis due to insufficient relevance for this sample of women. Five were related to physical issues (problems with my sleep, using alcohol or drugs to try to feel better, premenstrual tension or other problems with my menstrual cycle, problems with menopause [before/during/after], and other women’s problems – ovaries, breast, uterus), two were concerned with relating to others (I have difficulty with my relationships with people at work or in my family, I try to fit in and be what others expect of me), and one was concerned with negative or stressful life situations (I don’t have a job/study/sport/career/hobby or vocation).

Levels of depressive symptoms according to the CESD scores were high. With research studies generally using a cut point of 16 on the CESD, this sample reached twice that rate with a mean CESD score of 31.79 (SD = 11.44). The women in this sample, who self-assessed as sad or depressed, were experiencing considerable depressive symptomatology. Although the CESD was not designed for use as a clinical diagnostic tool, the scale is based on symptoms of depression as seen in clinical cases. Therefore it is sensitive to levels of severity of depressive symptomatology. Referring to the internal consistency estimate of reliability for the CESD scale in the original work of Radloff (1977), coefficient alphas ranging from .84 to .85 for the three community samples and .90 for the clinical sample were shown. The coefficient alpha demonstrated in the current study is higher than these earlier community samples and closer to the clinical samples reported in that work. This sample quite accurately self-selected as sad or depressed and resembles more closely a clinical sample.

Conclusion [TOP]

The limitations produced with the sampling method and use of an online survey render the findings unable to be generalized to all women and are therefore merely exploratory. Because of the need for a certain level of competence and familiarity with using the internet, the sample may over represent younger women or under represent older women and this may have skewed the results. The majority of the sample was of childbearing age. Also because of the anonymity available through online survey completion, responses may have been under or over exaggerated or misrepresented. Then again, a positive aspect of the online survey method could have been that the degree of privacy and anonymity offered, allowed women to feel safe and more likely to disclose sexual or relationship abuses in this way than if the questions were asked of them in person. By placing a difficult issue into an inventory, it may also make it easier for general practitioners and patients to have conversations about topics they may otherwise avoid.

There were also limitations with some of the items on the SHIFT-Depression® Inventory, as shown by the items regarding pregnancy and childbirth. If the inventory is being used as an assessment tool, the majority of the items require a deeper level of inquiry. There were also items on the inventory that did not prove to have a high relevance to the women in this sample, such as the item about use of alcohol, whereas many depressed individuals in the community do use alcohol to self-medicate.

Nevertheless, the results show that depression for this sample of women was connected to feeling overwhelmed with their responsibilities and caring for others, a lack of self-care, financial hardship, a lack of health care, enduring difficult relationships with significant others, isolation, and a lack of social support.

It appears that the women in the study felt they should prioritize their time to the care of others, for example, family and/or work, and this was connected to a lack of self-care. It is hard to know just from this result whether the life situations being experienced by these women were the precursors to then becoming depressed, a result of being depressed, or an interaction between these.

It is a social expectation that women prioritize the care of others (children, partners, elderly parents) and delay self-care. Neglecting to self-care provides fertile ground for the emergence of depression. As well, women are socialized to do the ‘emotion work’ of relationships (Fullagar, 2008; Minnotte, Pedersen, & Mannon, 2010). This can produce a pattern in a relationship where women give care and others take it with this continued through a self-reinforcing feedback loop.

The societal view, that it is in a woman’s nature to do the caring and the emotion work in relationships adds to its invisibility and it being taken for granted (Lee & Powers, 2002; Maclean, Glynn, & Ansara, 2004). The caring role therefore, is seen as inherently female and would explain why men in particular do not identify with it. Ways of behaving, expected because one is female or male are socially constructed and connected to identity (Broverman et al., 1972). It is probable that socially constructed beliefs regarding caring and emotion work maintain what women experience as their responsibility to carry the caring load in significant relationships, a role that is not highly valued. If a pattern of receiving little or no reciprocal care from a significant other becomes the norm, this can be experienced as unfair and very likely leads to feelings of resentment.

According to Jack (1991), women’s responses to unfair treatment can be anger, resentment, self-silencing, protest, despair and depression. This trajectory to depression looks similar to attachment distress observed in infants described by Bowlby (1969) and later described by Hazan and Shaver (1987) in regards to adult romantic relationships where one of the partners finds that they are unable to get their mate to respond or to acknowledge their need for emotional comfort.

When considering heterosexual relationships and acknowledging that there may be societal reasons (sex roles) that explain why many male partners are unable to reciprocate the care they receive from their female partners, there would also be many situations of dysfunctional relationships where reciprocating care was not a consideration.

For women who have experienced childhood neglect or abuse and then find themselves in a dysfunctional intimate relationship it may feel as if they are reliving earlier trauma and this reactivation would produce intolerable levels of emotional pain. If also facing a lack of social support and isolation, the pathway to depression may be inevitable.

Added to this, there is considerable research that shows links between childhood sexual abuse and depression in adult women (Beitchman et al., 1992). Without the opportunity to undergo healing therapy, these girls and women are vulnerable to depression in their adult attachments.

It is reasoned that no matter how well intentioned primary health care treatment geared to the symptoms of depression is, that treatment is only superficial if it does not include assessment of individual intrapsychic issues as well as other underlying causes such as the social and contextual inequities and relationship difficulties women are likely facing.

A diagnosis of depression for a woman should receive the benefit of the doubt that interpersonal, social and contextual issues are involved and need to be addressed. These issues could be the causes of, or the consequences of, depression but nevertheless must be included in a depression management plan.

This then would provide an avenue for treatment that is directed, according to Astbury (2010) ‘towards the rectification of any humiliation and devaluation’ and would honour a woman’s contextual and lived experience in society.

From the resulting factor structure of the SHIFT-Depression® Inventory, it has been shown that this could be a useful and efficient tool to use in both primary health and the mental health sectors because it does tap into the real world of depressed women.

Primary health practitioners don’t currently have the time to conduct assessments that include interpersonal and relationship difficulties, abuse, domestic violence, loss, negative life events and so on (Griffiths & Christensen, 2008). Using an inventory like this one, which can efficiently gather a woman’s rating of multiple issues, would address a current conundrum for the general practitioner who is expected to ask such questions but doesn’t have the time to unpack the often complex conversations that can result. The inventory could be handed to a woman at the end of a short consultation to take home to complete and return on a follow up visit. This would then provide an opportunity for a collaborative mental health treatment dialogue between women and their general practitioners, something that is currently missing. Using an inventory such as this one as a depression management tool in the primary care setting would be in accord with one of the recommendation of Griffiths and Christensen (2008) for guided self-help.

Braeken, Lechner, Houben, Van Gils, and Kempen (2011) have developed and tested a similarly time saving screening inventory for use with cancer patients in the primary care setting that quickly provides information about the experience of any psychosocial problems and whether or not there is a desire to discuss any identified issues with someone.

Whilst society continues to determine that it is a woman’s primary responsibility to provide the emotion work and care of children as well as other family members, and it follows from this that women continue to prioritize the care of others first and sacrifice self-care, the ground is fertile for women to become depressed. Combined with sex role stereotyping that regards women as naturally sensitive and emotional and therefore depression prone, the current observation that twice as many women as men will experience depression over the life cycle seems inevitable. If a depression management tool such as the one evaluated in this study could be introduced as part of a routine assessment in both primary care and mental health care settings, some inroad may be made to changing current practices that do not prioritize or promote a woman’s enjoyment of optimal health and quality of life.

Appendix [TOP]

Item and Factor Structure of the SHIFT-Depression® Inventory

Item no. Item text Factor loadings of items
Factor 1: Diminished self, feels powerless, focused on other´s needs
T3 Feeling negative and thinking negative things .710
T4 Not feeling in control of my life and feeling powerless .686
T6 Feeling overwhelmed and responsible for everything .680
T2 Being self-critical .661
T7 I don’t take the time or do what I need for myself .627
T5 Feeling I have disappeared or lost myself .603
R7 I feel I must attend to other’s needs before I look after myself .603
S4 I feel stuck in an unhappy situation .578
T1 Feeling worried or anxious .558
Factor 2: Financial problems, lacks access to healthcare, healthy food, housing, exercise
P3 Not being able to do exercise .628
S3 I have difficulty with my housing or transport issues .602
P8 Not being able to get the health care I need .554
P5 Reaction to or not having prescription medicines .544
P6 Reaction to or not having complementary medicines .544
P1 Ill health, injury, accident, headaches, virus, pain or disability .536
S5 I don’t have sufficient money or can’t manage on the money I have .517
P4 Not eating healthy or nutritious food .491
Factor 3: Relationship difficulties, loss/betrayal/abuse
R5 I’ve been betrayed by someone I trusted .681
S2 I’ve lost something special to me, job/house/way of life/dream .622
R4 I’ve lost someone special to me, partner/parent/child/sibling/friend/pet/other .570
R6 I’ve experienced emotional/physical/sexual abuse/bullying .547
R8 Others treat me disrespectfully or do what they want without concern for me .507
Factor 4: Women’s Problems
W2 Problems with the birth of my baby .739
W1 Problems with my pregnancy .673
Factor 5: Lack of social support, feels isolated
C1 I feel like an outsider or like I don’t belong in my community .709
C3 I don’t have any social networks or groups I belong to .582
R2 I cannot or do not seek support from others .562
C2 My neighbourhood/community is difficult/unfriendly/unsafe .534
R1 I’m isolated and don’t have a supportive confiding relationship/s .524
R3 Other people judge or criticize me .470

Note. Domains: P = Physical, T = Thoughts & Feelings, S = StressfulSocialSituations, R = RelationshipsWithOthers, C = Community&Neighbourhood, W = Women’sProblems.

Eight items dropped from final EFA: P2 problems with sleep, P7 Using alcohol or drugs to try to feel better, S1 I don’t have a job/study/sport/career/hobby or vocation, R9 I have difficulty with my relationships with people at work or in my family, C4 I try to fit in and be what others expect of me, W3 Premenstrual tension or other problems with my menstrual cycle, W4 Problems with menopause – before/after/during, W5 Other women/s problems – ovaries/breast/uterus.

Funding [TOP]

There was no funding to declare.

Competing Interests [TOP]

There are no competing interests to declare.

Acknowledgments [TOP]

I wish to thank the Director of the Social Justice Social Change Research Group (SJSC), Professor Rosemary Leonard, and Dr. Ron Brooker, Research Officer, SJSC, for statistical support. I would also like to thank the women who participated in the study for taking the time to provide information that will hopefully help other women.

References [TOP]

  • Astbury, J. (2010). The social causes of women’s depression: A question of rights violated? In D. C. Jack & A. Ali (Eds.), Silencing the self across cultures: Depression and gender in the social world (pp. 19-45). Oxford: Oxford University Press.

  • Australian Institute of Health and Welfare. (1998). Australia’s health. Canberra: Author.

  • Bartlett, M. S. (1954). A note on the multiplying factors for various chi square approximations. Journal of the Royal Statistical Society. Series B. Methodological, 16, 296-298.

  • Beitchman, J. H., Zucker, K. J., Hood, J. E., DaCosta, G. A., Akman, D., & Cassavia, E. (1992). A review of the long term effects of child sexual abuse. Child Abuse & Neglect, 16, 101-118. doi:10.1016/0145-2134(92)90011-F

  • Belle, D., & Doucet, J. (2003). Poverty, inequality, and discrimination as sources of depression among US women. Psychology of Women Quarterly, 27, 101-113. doi:10.1111/1471-6402.00090

  • Berghöfer, A., Hartwich, A., Bauer, M., Unützer, J., Willich, S. N., & Pfennig, A. (2012). Efficacy of a systematic depression management program in high utilizers of primary care: A randomized trial. BMC Health Services Research, 12, Article 298. doi:10.1186/1472-6963-12-298

  • Bowlby, J. (1969). Attachment: Attachment and loss (Vol. 1). London: Penguin.

  • Braeken, A. P. B. M., Lechner, L., Houben, R. M. A., Van Gils, F. C. J. M., & Kempen, G. I. J. M. (2011). Psychometric properties of the Screening Inventory of Psychosocial Problems (SIPP) in Dutch cancer patients treated with radiotherapy. European Journal of Cancer Care, 20, 305-314. doi:10.1111/j.1365-2354.2010.01182.x

  • Brannen, J. (1992). Combining qualitative and quantitative approaches: An overview. In J. Brannen (Ed.), Mixing methods: Qualitative and quantitative research (pp. 3-37). Aldershot: Avebury.

  • Broverman, I. K., Vogel, S. R., Broverman, D. M., Clarkson, F. E., & Rosenkrantz, P. S. (1972). Sex-role stereotypes: A current appraisal. The Journal of Social Issues, 28(2), 259-78. doi:10.1111/j.1540-4560.1972.tb00018.x

  • Brown, G. W., Harris, T. O., & Hepworth, C. (1995). Loss, humiliation and entrapment among women developing depression: A patient and non-patient comparison. Psychological Medicine, 25, 7-21. doi:10.1017/S003329170002804X

  • DeVellis, R. F. (2003). Scale development: Theory and applications (2nd ed.). Thousand Oaks, California: Sage.

  • Dill, J. C., & Anderson, G. A. (1999). Loneliness, shyness and depression: The etiology and interrelationship of everyday problems in living. In T. Joiner & J. C. Coyne (Eds.), The interactional nature of depression (pp. 93-126). Washingon, DC: American Psychological Association.

  • Ellis, P. M., & Smith, D. R. (2002). Treating depression: The beyondblue guidelines for treating depression in primary care. The Medical Journal of Australia, 176, S77-S83.

  • Farmer, A. E., Owen, M. J., & McGuffin, P. (2000). Bioethics and genetic research in psychiatry. The British Journal of Psychiatry, 176, 105-108. doi:10.1192/bjp.176.2.105

  • Fricker, R. D. (2008). Sampling methods for web and e-mail surveys. In N. Fielding, R. Lee, & G. Blank (Eds.), The Sage handbook of online research methods. (pp. 195-217). London: Sage Publications.

  • Fullagar, S. (2008). Leisure practices as counter-depressants: Emotion-work and emotion-play within women’s recovery from depression. Leisure Sciences, 30, 35-52. doi:10.1080/01490400701756345

  • Grbich, C. (1999). Qualitative research in health. Sydney: Allen & Unwin.

  • Griffiths, K. M., & Christensen, H. (2008). Depression in primary health care: From evidence to policy. The Medical Journal of Australia, 188(8), 881-84.

  • Haggett, A. (2012). Desperate housewives: Neuroses and the domestic environment 1945-1970 (pp. 1-10). London: Pickering & Chatto.

  • Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511-524. doi:10.1037/0022-3514.52.3.511

  • Henson, R. K., & Roberts, J. K. (2006). Use of exploratory factor analysis in published research: Common errors and some comment on improved practice. Educational and Psychological Measurement, 66, 393-416. doi:10.1177/0013164405282485

  • Herrman, H., Diehr, P., Fleck, M., Simon, G., & Patrick, D. L. (2002). Longitudinal Investigation of Depression Outcomes in primary care in six countries: The LIDO Study. 11: Functional status, health service use, and treatment of people with depressive symptoms. Psychological Medicine, 32, 889-902. doi:10.1017/S003329170200586X

  • International Trade Union Confederation. (2008). The global gender pay gap. Brussels: Author.

  • Jack, D. C. (1991). Silencing the self. New York: Harper Collins.

  • Kaiser, H. (1974). An index of factorial simplicity. Psychometrika, 39, 31-36. doi:10.1007/BF02291575

  • Katon, W. J., & Ludman, E. J. (2003). Improving services for women with depression in primary care settings. Psychology of Women Quarterly, 27, 114-120. doi:10.1111/1471-6402.00091

  • Katzelnick, D. J., Simon, G. E., Pearson, S. D., Manning, W. G., Helstad, C. P., Henk, H. J., Kobak, K. A. (2000). Randomized trial of a depression management program in high utilizers of medical care. Archives of Family Medicine, 9, 345-351. doi:10.1001/archfami.9.4.345

  • Koss, M. P., & Heslet, L. (1992). Somatic consequences of violence against women. Archives of Family Medicine, 1, 53-59. doi:10.1001/archfami.1.1.53

  • Lee, C., & Powers, J. R. (2002). Number of social roles, health and well-being in three generations of Australian women. International Journal of Behavioral Medicine, 9, 195-215. doi:10.1207/S15327558IJBM0903_03

  • Lopez, A. D., Mathers, C. D., Essati, M., Jamison, D. T., & Murray, C. J. L. (2006). Global and regional burden of disease and risk factors, 2001: Systematic analysis of population health data. Lancet, 367, 1747-1757. doi:10.1016/S0140-6736(06)68770-9

  • Maclean, H., Glynn, K., & Ansara, D. (2004). Multiple roles and women’s mental health in Canada. BMC Women’s Health, 4, (Suppl 1), S3.

  • Marcus, M., Yasamy, M. T., van Ommeren, M., Chisholm, D., & Saxena, S. (2012). Depression: A global public health concern. In World Federation for Mental Health (Ed.), Depression: A global crisis (p. 6-8).

  • McLennan, W. (1998). Mental health and wellbeing: Profile of adults, Australia 1997. Canberra: Australian Bureau of Statistics.

  • Mertens, D. (2007). Transformative paradigm: Mixed methods and social justice. Journal of Mixed Methods Research, 1, 212-225. doi:10.1177/1558689807302811

  • Minnotte, K. L., Pedersen, D., & Mannon, S. E. (2010). The emotional terrain of parenting and marriage: Emotion work and marital satisfaction. The Social Science Journal, 47, 747-761. doi:10.1016/j.soscij.2010.07.011

  • Nicolson, P. (1992). Towards a psychology of women’s health and health care. In P. Nicolson & J. Ussher (Eds.), The psychology of women’s health and health care (pp. 6-30). London: The Macmillan Press.

  • Post, R. M. (1992). Transduction of psychosocial stress into the neurobiology of recurrent affective disorder. The American Journal of Psychiatry, 149, 999-1010.

  • Proudfoot, J., Klein, B., Andersson, G., Carlbring, P., Kyrios, M., Munro, C., . . ., & Blankers, M. (2010). Guided CBT internet interventions: Specific issues in supporting clients with depression, anxiety and co-morbid conditions. In J. Bennett-Levy, D. A. Richards, P. Farrand, H. Christensen, K. M. Griffiths, D. J. Kavanagh, B. Klein, M. A. Lau, J. Proudfoot, L. Ritterband, J. White, & C. Williams (Eds.), Oxford guide to low intensity CBT interventions (pp. 253-264). London: Oxford University Press.

  • Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401. doi:10.1177/014662167700100306

  • Resnick, H. S., Acierno, R., & Kilpatrick, D. G. (1997). Health impact of interpersonal violence: 2. Medical & mental health outcomes. Behavioral Medicine, 23, 65-78. doi:10.1080/08964289709596730

  • Vidler, H. C. (2005). Women making decisions about self-care and recovering from depression. Women's Studies International Forum, 28(4), 4289-303. doi:10.1016/j.wsif.2005.04.014

  • Vidler, H. C. (2006). Improving treatment outcomes for depressed women: Use of a self-help inventory in counselling settings. Psychotherapy in Australia, 12(2), 274-81.

  • Wagner, E. H. (1997). Managed care and chronic illness: Health services research needs. Health Services Research, 32, 702-714.

  • Wei, M., Russell, D. W., Mallinckrodt, B., & Vogel, D. L. (2007). The experiences in Close Relationship Scale (ECR)-Short Form: Reliability, validity, and factor structure. Journal of Personality Assessment, 88, 187-204. doi:10.1080/00223890701268041

  • Workplace Gender Equality Agency (WGEA). (2012). Gender pay gap statistics. Retrieved from

  • World Health Organization. (2001). Mental Health: New understanding, new hope. The world health report 2001. Geneva: WHO.

  • World Health Organisation. (2000). Women’s mental health: An evidence based review. Geneva: WHO.

  • World Health Organisation. (2004). The global burden of disease 2004 update. Retrieved from

  • World Health Organization. (2010). mhGAP Intervention Guide: For mental, neurological and substance use disorders in non-specialized health settings 2010. Retrieved from

  • Worthington, R. L., & Whittaker, T. A. (2006). Scale development research: A content analysis and recommendations for best practice. The Counseling Psychologist, 34, 806-838. doi:10.1177/0011000006288127

  • Wu, L. T., & Anthony, J. C. (2000). The estimated rate of depressed mood in US adults: Recent evidence for a peak in later life. Journal of Affective Disorders, 60, 159-171. doi:10.1016/S0165-0327(99)00176-7

About the Author [TOP]

Dr Helen C. Vidler is a practising psychologist and family therapist, a member of the Australian Psychological Society, the Australian Association of Family Therapists, and registered with the Australian Health Practitioners Regulation Agency. She has over 30 years clinical experience using evidence based methods with depressed women. Over the past 12 years Helen has combined her role as a researcher together with her clinical practice. One of the intentions of this research is to bring to the foreground the perspectives of depressed women and thereby influence the management of depression in health settings.