Less depressed or less forthcoming? Self-report of depression symptoms in women preparing for in vitro fertilization.

321 of the 419 patients completed the PHQ-9 screening tool, giving a response rate of 77%. Table 1 shows the clinical characteristics of our IVF study group. There were no significant PHQ response differences within the IVF study group based on clinical characteristics, including duration of infertility, treatment history, infertility-related diagnosis, or whether infertility was primary or secondary, though there was a trend towards exclusive female infertility diagnosis being associated with higher scores ( P =.07). There were also no significant associations between PHQ responses and demographic characteristics including age, marital status, BMI, and race, within the IVF study group. Demographic data for the IVF study group and comparative groups can be found in Table 2 . Based on PHQ-9 diagnostic scoring, women in the IVF study group had significantly lower incidences of MDD (1% versus 10%, P <0.0001) and ODD (2% versus 6%, P <0.01) than the sample of primary care patients used in the original publication of the measure ( Spitzer et al. 1999 ). An initial run of raw score frequencies among the IVF study group indicated that 90% of PHQ-9 scores fell within the 0–4 range (minimal or no symptoms). Few women (2%) had scores greater than or equal to 10, a range suggesting moderate to severe depressive symptoms. We decided to look further into the distribution of scores within the 0–4 range and found that half of women (50%) in that range had scores of 0, indicating no depressive symptoms at all over the previous 2 weeks. In order to make meaningful population comparisons given the distribution of our data, scores of 0 were separated from conventionally used ranges, and higher scoring categories were collapsed together. This resulted in 4 categorical score groups of 0 (no depressive symptoms), 1–4 (mild), 5–9 (moderate), and ≥10 (severe). Using these categories, we found that the overall distribution of scores for the IVF group was significantly different than both the distribution of NHANES general population scores and the scores from Iowa postpartum women ( P <.0001) ( Table 3 ). There was a particularly notable difference between the number of women in the IVF group reporting no depressive symptoms and those reporting no symptoms in both comparison groups.

Our study sought to examine depression self-report response profiles of women prior to undergoing IVF. This study was unique in its high response rate and the type of comparative groups we used. Our patients initiating IVF treatment rarely endorsed any significant depressive symptoms, and almost never screened positive for MDD. This was a statistically significant difference compared to several large comparison groups, including patients in primary care and postpartum settings, and a community sample. Within our IVF study group, there were no significant response differences based on a woman’s demographic or clinical characteristics, including type of infertility diagnosis, duration of infertility or history of infertility treatment. The trend we found toward having an exclusively female infertility diagnosis and higher PHQ scores may indicate an increased psychological burden for infertile women whose partners are fertile. Most notably, IVF patients endorsed no depressive symptoms significantly more often than individuals in the general population or in a population of postpartum women in Iowa. These findings suggest that women are less depressed prior to undergoing IVF, potentially feeling optimistic about improved chances to conceive. However, this low level of distress is inconsistent with many findings in the literature and with the anecdotal accounts by some clinicians ( Peterson and Eifert 2011 ; Fassino et al. 2002 ; Greil 1997 ; Friedler et al. 2011 ). Further, researchers found that distraction-based strategies were effective ways for individuals anticipating a stressful medical procedure to cope, particularly suppressing emotions related to the procedure ( Bennett et al., 2007 ; Fauerbach et al., 2002 , Tschuschke et al., 2001 ). Perhaps women cope with their anticipatory distress related to IVF by suppressing their negative emotions. Nevertheless, some researchers have suggested that the IVF patient population may be more immune to distress or depression. Given the financial barriers to reproductive health care, and limited insurance options for IVF treatment specifically, patients who are diagnosed with infertility and receive IVF treatment tend to report higher incomes. However, researchers have found that socioeconomic status was not predictive of psychological distress and rates of depressive disorders ( Kosidou et al. 2011 ; Anderson et al. 2009 ). Our sample also included more married individuals than found in typical age-matched populations, and marriage has been shown to be protective for depression in fertile couples; though, infertile couples often report strains in their relationships, and research on marital discord has been associated with increased risk for depression ( Peterson et al. 2003 ; Whisman et al. 2006 ). Having children is a widely shared value and major quality of life determinant across cultures ( Lampic et al. 2006 ; Kisakye et al. 2010 ; Newport 2011). Quality of life, as defined by the World Health Organization, refers to “individuals’ perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” (The WHOQOL Group, 1995). Besides representing a failure to achieve a personal goal, infertility negatively impacts quality of life as it represents a failure to achieve a cultural value, and thus carries a degree of social stigma. This means that infertile individuals may face increased social pressures, including pressure to underreport other socially stigmatized conditions such as mental illness. The authors are thus cautious regarding a conclusion that women prior to undergoing IVF are less distressed than the general population. It may also be the case that the PHQ is not the best measure of distress in women preparing for IVF. Recently, researchers developed and tested the first quality of life measure for use specifically with infertile men and women ( Boivin et al, 2011 ). Potential items were generated through consultation with patients, individually and in focus groups, and with a multidisciplinary team of gynecologists, nurses, psychologists, and social workers. They identified 6 infertility-specific domains most impacting quality of life: anger, affection, fatigue, family understanding, patient interactions with clinicians, and quality of emotional services ( Boivin et al. 2011 ). Further testing, development, and implementation of such specific psychological screening measures for this population may be needed to gauge the mental health needs of this population. One limitation of our study was the small sample size given the distribution of scores – that is, the tendency of IVF responses to center around a score of 0, and few IVF patients receiving PHQ-9 scores greater than 10. Thus, we did not have enough power to make population comparisons for this higher range of scores. We acknowledge that the national population samples used for comparison were not ideal given the age, gender, and demographic differences between the samples and the Iowa IVF and postpartum groups. Yet we believe, in fact, that our results might have been more robust had these variables been controlled for. This is due to the fact that the national population samples include more individuals least at risk for depression, such as older individuals and men, than the IVF sample. Rates of depression generally decline throughout adulthood, and some researchers have suggested old age is protective for depression ( Jorm et al. 2005 ). Women are also twice as likely as men to receive a diagnosis of MDD ( Weissman et al. 1996 ). The emotional, social, and economic burden of infertility is one often misunderstood by unaffected individuals. As with many stigmatizing conditions, infertility is often a silent and isolating affliction. If infertility patients were routinely presented with screening measures more specific to their experiences, they may respond differently. Not only do psychological measures need to be administered routinely, and at successive treatment intervals, but they should also be used as an opportunity to normalize the distress infertile individuals may be experiencing. In order to provide optimal infertility care, clinicians need to ensure they are using the appropriate measures when necessary.

A retrospective chart review identified women who made initial IVF visits to the University of Iowa Center for Advanced Reproductive Care between April, 2009 and December, 2010. 419 women or couples were seen as new IVF patients at the program’s two sites during that time. During those visits, the women were routinely administered a paper version of the PHQ-9 by a medical assistant or nurse. This was not introduced as a screening measure for depression or infertility-related distress. The administrators only stated that a physician would review the woman’s responses to this questionnaire in order to make appropriate mental health referrals if necessary. Responses to all individual items on the PHQ-9 were collected. Through chart review, patient demographics, and diagnostic and treatment data were also obtained for all subjects. The University of Iowa Institutional Review Board approved this study. Our PHQ-9 data were compared to initial screening results obtained as part of a previous epidemiological study on 7,860 postpartum women in 7 Iowa counties. Women who had given birth in the 6 months prior to January 2008 were recruited via State of Iowa birth records and the University of Iowa Women’s Wellness and Counseling Service, and were screened over the phone using the PHQ-9. We also compared our PHQ-9 IVF data to normative data that were obtained from previously published studies of the diagnostic validity of the PHQ-9 for mental disorders in 3,000 primary care patients ( Spitzer et al. 1999 ). Finally, we compared our PHQ-9 IVF data to general population PHQ data obtained from the publically available 2005–2006 National Health and Nutrition Examination Survey (NHANES) results. These data are collected by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention to assess the general health and nutritional status of the U.S. population ( Centers for Disease Control and Prevention, 2005 ). The PHQ-9 is a 9-item self-report measure originally designed to assess for the presence of major depressive disorder (MDD) and other depressive disorder (ODD - a subthreshold diagnosis of MDD) in a primary care setting. On a 0–3 Likert Scale, patients rate the frequency over the past 2 weeks of symptoms associated with 9 Diagnostic and Statistical Manual for Mental Disorders (DSM-IV) (American Psychiatric Association, 1994) criteria for MDD. The PHQ-9 can be scored to yield a diagnosis of MDD or ODD using an algorithm based on DSM-IV criteria, and as a depression severity rating based on raw scores categorized into score ranges. The possible score range is 0 to 27, with higher scores indicating greater depressive symptom severity. Typical reported categories of score ranges with descriptors are: 0–4 (Minimal or no symptoms), 5–9 (mild depression), 10–14 (minor depression or dysthymic disorder), 15–19 (moderately severe MDD), and 20–27 (severe MDD). The PHQ-9 measure has been shown to be psychometrically sound in a number of clinical populations, including gynecology outpatients in whom the PHQ-9 demonstrated good sensitivity (73%) and excellent specificity (98%) for MDD ( Spitzer et al. 1999 ; Spitzer et al. 2000 ). 1 Statistical analysis was performed using SAS software (SAS Institute, Cary, NC). A chi-square test was used to compare frequency distributions among collapsed categorical groups of score ranges. Nonparametric analysis was conducted to make comparisons in the overall distributions. A Bonferroni correction was used to calculate a significant p value based on the number of comparisons that were made. A P value of < .05 was considered statistically significant.