Are Severe Depressive Symptoms Associated With Infertility-related Distress in Individuals and Their Partners?

Brennan D. Peterson; Camilla S. Sejbaek; Matthew Pirritano; Lone Schmidt


Hum Reprod. 2014;29(1):76-82. 

In This Article

Materials and Methods


This study is part of The Copenhagen Multi-centre Psychosocial Infertility (COMPI) Research Programme (Schmidt, 2006), a prospective longitudinal cohort study of infertile couples in fertility treatment. Patients who were consecutively referred at one of the four public hospital-based tertiary fertility clinics and one private clinic between January 2000 and August 2001 received a questionnaire for each spouse before attending their first treatment. The study complied with the Helsinki II Declaration was assessed by the Scientific Ethical Committee of Copenhagen and Frederiksberg Municipalities (KF 01-107/99), and was approved by the Danish Data Protection Agency (J. nr. 1999-1200-233; 2001-41-1486; 2005-41-5694).


The 5-item Mental Health Inventory 5 (MHI-5) from the 36-item Short-Form Health Survey (SF-36), Medical Outcomes Study, was used to measure severe depressive symptoms (Bjorner et al., 1998a,b; Strand et al., 2003). Previous studies comparing MHI-5 with other (validated) mental health scales have shown MHI-5 to be a good measure of severe depressive symptoms (Berwick et al., 1991; Strand et al., 2003). The five items in the MHI-5 measure the mood of the participants in the past 4 weeks (e.g. felt so down in the dumps that nothing could cheer you up, felt downhearted and blue). The response key was a 6-point scale ranging from (1) all of the time to (6) none of the time. The scores for two of the items were reversed and the answers were summed up to a raw score ranging from 5 to 30 (Bjørner et al., 1997). The raw score was then transformed to a scale ranging from 0 to 100. The scale was dichotomized with a cut-off point at 52 and participants scoring ≤52 were categorized as having severe depressive symptoms. This cut-off point was chosen in agreement with previous studies comparing other scales measuring depressive symptoms to the MHI-5, and investigating different cut-off points for the MHI-5 (Holmes, 1998; Strand et al., 2003). Studies in other research fields also use a cut-off point at ≤52 (Rugulies et al., 2012). Therefore, choosing the same cut-off point as other studies makes it possible to compare results. The MHI-5 has been widely used in studies assessing mental health, as well as general health, and has had a Cronbach's alpha of 0.82 (Strand et al., 2003).

Infertility-related distress was measured by the COMPI Fertility Problem Stress Scales, a 14-item instrument based on The Fertility Problem Stress Inventory (Abbey et al., 1991) and qualitative interviews of Danish infertile patients regarding the psychosocial consequences of infertility and fertility treatment (Schmidt, 1996). The measure includes questions regarding one's personal distress (six items, e.g. how much stress the individual felt in their life as a result of the childlessness), marital distress (four items, e.g. how much stress the childlessness placed on marriage and sexual relationship) and social distress (four items, e.g. how much stress the fertility problem placed on relationships with family, friends and workmates). The response key for 10 of the 14 items is based on a 4-point scale from (1) a great deal to (4) none at all, while four items are based on a 5-point scale from (1) strongly agree to (5) strongly disagree. An exploratory factor analysis produced a set of parsimonious factors (Schmidt et al., 2003). In the exploratory factor analysis, items with factor loadings >0.45 were assigned to the factor (personal, marital or social distress) for which they had the greatest loading. As mentioned above, the three domains were uncovered in accordance with Abbey et al. (1991) and the interviews conducted by Schmidt (1996). The range, mean and Cronbach alpha coefficients differed depending on the subscale: personal distress (range 0–20, mean 6.86, SD 4.48, Cronbach alpha 0.82); marital distress (range 0–14, mean 3.86, SD 3.16, alpha 0.73) and social distress (range 0–12, mean 1.87, SD 2.43, alpha 0.82) (Schmidt et al., 2003).

Data Analysis

This study was cross-sectional. Multilevel modelling using the APIM(Kenny et al., 2006; Kashy and Donnellan, 2008) was used to study the association between a partner's severe depressive symptoms and his or her partner's distress (personal, marital and social) (see Fig. 1). The APIM allows for the simultaneous estimation of actor effects (individual effects) and partner effects (the effects of another closely associated person) to shared stressors in dyads, thus providing a more complete picture of how severe depressive symptoms are related to distress in couples.

Figure 1.

General model of actor and partner effects of severe depressive symptoms on distress.

Data were analysed with the couple as the unit of analysis. This was done by conducting multilevel analyses using the SAS 9.2® mixed procedure. A multilevel analysis involves more than one regression model calculated at different levels of a nested design. In the current set of analyses, level 1 was the individual level that was nested within level 2, the couple. Multilevel analyses estimate the model independently at each of these levels. The design of the analysis is very similar to a multiple regression with one dependent variable and a set of predictors, or independent variables. Analyses provide unstandardized estimates of path coefficients for actor and partner effects.

The analyses were cross-sectional with severe depressive symptoms (dichotomous) as the independent variable, and personal, marital and social distress as the dependent variables. Three analyses were conducted, one for each of the three types of distress. Because previous research has found infertility diagnosis (i.e. male or female factor infertility) to be related to other psychosocial consequences of infertility (Peronace et al., 2007), infertility diagnosis based on five categories [i.e. male factor, female factor, both male–female factor, other causes (not specified), unknown] was therefore tested as a covariate in the analyses. However, no statistical association between infertility diagnosis and severe depressive symptoms and personal and marital distress was found, so it was not included in the final analyses.