|Home | About | Journals | Submit | Contact Us | Français|
This study seeks to examine the relationship between socio-demographic characteristics and the utilization of fertility preservation services in reproductive age women diagnosed with cancer.
1041 women diagnosed with cancer between age 18 and 40 responded to our retrospective survey on demographic information and reproductive health history. Five cancer types were included: leukemia, Hodgkin’s disease (HD), Non-Hodgkin Lymphoma (NHL), breast cancer, and gastrointestinal cancer (GI). 918 women reported treatment with potential to affect fertility (chemotherapy, pelvic radiation, pelvic surgery, or bone marrow transplant). Student’s t-test, linear regression, and multivariate logistic regression were used where appropriate to determine the relationship between socio-demographic characteristics and the odds of utilizing fertility preservation services.
61% of women were counseled on the risk of cancer treatment to fertility by the oncology team. Overall, 4% of women pursued fertility preservation. In multivariate analysis, women who had not attained a bachelor’s degree (OR 0.7, 95%CI 0.5 – 0.9) were less likely to be counseled. Trends also suggested possible disparities in access to fertility preservation with age greater than 35 years old (OR 0.1, 95% CI 0.0 – 1.4) or previous children (OR 0.3, 95% CI 0.1 – 1.1) at diagnosis. Disparities in access to fertility preservation based on ethnicity and sexual orientation were also observed.
Socio-demographic health disparities likely affect access to fertility preservation services. Although awareness of fertility preservation has improved in the last decade, an unmet need remains for reproductive health counseling and fertility preservation in reproductive age women diagnosed with cancer.
According to 2006 Surveillance Epidemiology and End Results (SEER) statistics, approximately 120,000 women below the age of 50 develop cancer each year in the United States.1 Early screening techniques and cancer therapies have improved cancer survival significantly over the past twenty years - the medical community is now turning to issues related to life after cancer treatment.2 Young women face post-treatment reproductive concerns relating to infertility and premature menopause.3 Fertility preservation, saving eggs and/or embryos prior to treatment, appears to positively affect survivor’s quality of life.4,5 The American Society of Clinical Oncology recommends that all women be offered counseling and the opportunity to take action to protect or preserve their fertility prior to cancer therapy.6
Advances in reproductive technology may have outpaced access. A 2008 study from the Netherlands showed that as few as two percent of women preserved their fertility prior to cancer treatment.7 As recently as 2009, a survey of 249 oncologists at major academic centers in the United States reported that 82% of physicians had referred a patient to a reproductive endocrinologist at one point, but more than half rarely refer.8 The reasons given for lack of referral were: lack of knowledge about reproductive options, insufficient time to discuss the issue, perceptions that patients could not delay treatment, or the perception that if patients did not raise the issue themselves they were not interested.9
A focus on patient’s access to education and resources is paramount in the discussion of post-cancer treatment fertility issues. Utilizing a statewide cancer registry of reproductive aged women, we seek to identify whether barriers to access exist within various sociodemographic groups, and whether such barriers contribute to a low utilization of fertility preservation counseling and services prior to cancer therapy.
We performed an analytical cross-sectional study, using the California Cancer Registry (CCR) to sample women across the state. All study procedures were approved by University of California, San Francisco Committee on Human Research.
A computer-generated randomizer was used to sample reproductive age women from the cancer registry that had a history of leukemia, Hodgkin’s disease, non-Hodgkin lymphoma, breast cancer, or gastrointestinal (GI) cancer. These cancers were chosen for study because they are common, non-gynecologic cancer groups that can be treated with systemic chemotherapy. Patients were included in the sample if they were 18-40 years of age at diagnosis, and were diagnosed between 1993 and 2007. Among 6709 patients initially selected for the study, 4147 patients were excluded because their contact information in the cancer registry was outdated. Letters were sent to the primary physicians of each of the remaining women before we attempted to reach the patients. Additional women were excluded if their physician thought participation in the study would cause undue psychological burden (due to severe co-morbid mental illness, n=30 patients). After exclusions, 2532 patients were contacted for participation in the study. Analyses of access included only women who reported treatment with potential to compromise fertility (i.e., systemic chemotherapy, pelvic radiation, pelvic surgery, or bone marrow transplant).
The survey was created at UCSF and assessed for readability and content validity – the extent to which our survey accurately assessed reproductive health history and quality of life - by two independent experts in survey methodology. It was then tested on 20 patients from the UCSF Center for Reproductive Health for content and readability. Additional changes were made in response to this pilot study.
The survey was made available in both English and Spanish. A professional translation company (American Language Services, Los Angeles, CA) translated the study materials into Spanish using two independent translators. A third bilingual person checked translations. Paper surveys were created using Cardiff Teleform (Vista, CA). Patients could also complete the survey online through Surveymonkey.com (LLC, Palo Alto, CA).
Women were contacted between January 2010 and September 2010. A contact letter was sent to potential participants, explaining the purpose of the survey, the source of the individual’s personal contact information (the CCR). Women were able to opt out of further contact by reaching the UCSF study coordinator. After one week, a second mailing was sent that included the written survey, a link to the online survey, a written consent form, a postage-paid return envelope, and an optional refusal postcard. Women were asked to complete and return a written consent form by mail and to return the survey by mail or complete it online. Women who did not reply within 3 weeks received a reminder follow-up phone call. Those who did not reply within 2 weeks of the reminder call were sent a reminder post-card with a link to the electronic survey.
Two binary outcomes were examined with respect to several socio-demographic predictor variables. The first outcome, counseling, refers to whether women were counseled (yes/no) about the risk of their treatment to fertility by a member of the oncology team. The second, fertility preservation, refers to whether respondents pursued fertility preservation (yes/no) under the care of a reproductive endocrinologist.
We then sought to determine the yearly prevalence of each of our two primary outcomes. Prevalence was calculated as affirmative responses (e.g., counseled (yes) / all participants) for a given year of diagnosis (e.g., in 1993 14 out of 32 patients (44%) were counseled).
Disease type, stage, and treatment type were considered as potential confounders of our primary socio-demographic exposure variables.8,9 To define stage, summary staging was used. Summary staging is a California Cancer Registry index used for epidemiologic assessment of cancers (1=in situ; 2=localized; 3=regional by direct extension; 4=regional by lymph nodes; 5=regional by direct extension and lymph nodes; 6=metastatic). Treatment types included having received the following categories of treatment: chemotherapy, radiotherapy, and/or bone marrow transplant.
Several socio-demographic characteristics were assessed by survey. Some of these characteristics, such as ethnicity, gender, income, attainment of college education, population density, and age at diagnosis, were chosen based on previously observed disparities in general oncology and fertility care.10-14 Although parity and marital status have been previously associated with oncologists’ perceptions of their patient’s fertility concerns, their association with fertility counseling by oncologists or access to fertility preservation has not been determined.15
Parity at diagnosis was defined as whether respondents reported previously having at least one child. Marital status was defined as having a partner (spouse or significant other) versus being single at the time of diagnosis. Sexual orientation was defined as heterosexual, homosexual, bisexual, or other (the later three groups were combined at analysis for the purpose of forming dichotomous variables as “other than heterosexual”).
Annual income at the time of survey was divided into three groups: <$30,000; $30,001 - $100,000; >$100,001. Incomes of less than approximately $30,000 per year are believed to be insufficient to adequately meet the basic needs of an average American family.16 Those who have household incomes near and above $100,000 less likely to report difficulties in establishing medical care than women in lower income groups.17 Age groups at diagnosis were divided into mostly 5-year segments to achieve sufficient sample sizes for logistic regression (ages 18 – 25, 26 – 30, 31 – 35, and 36 – 40). Ethnicity was defined by the following: Caucasian, Asian/Pacific Islander, African American, Hispanic American or Latina, and Native American.
Several variables, including year of diagnosis and age at diagnosis, were derived from the cancer registry. Time since diagnosis was derived as the difference between date of diagnosis and date of recruitment. Population density was defined in the Cancer Registry as urban or rural. Urban areas were defined in the cancer registry as those with urban cores of 50,000 or more people, or areas involved in direct regional commerce with these urban cores. SES is a Cancer Registry socioeconomic index from based on income, education, and education for the census block group of residence at diagnosis.
Survey data was merged with CCR data with a unique, anonymous identifier code. Statistical analyses were performed using STATA Version 11 (College Station, TX). Statistical significance was defined by two-sided P-values less than 0.05.
Initially, Student’s t-tests were used to compare the characteristics of responders and non-responders. Several variables from the Cancer Registry were available for all patients recruited for the study. Age at diagnosis, time since diagnosis, summary staging, and SES were used to assess potential differences between those women who completed the survey and those who did not.
Logistic regression was used to determine the relationship between sociodemographic predictors and the odds of counseling for fertility preservation or pursuing fertility preservation. Univariate analyses were initially performed for each socio-demographic predictor to evaluate the magnitude of effect on counseling as well as fertility preservation. Multivariate analyses were then performed to adjust for the effect of potential confounding variables (disease type, stage, and treatment) and the effects of other socio-demographic variables whose univariate P-value was <0.2.
Finally, linear regression was used to determine the rate of increase in the prevalence of counseling and fertility preservation over time, while adjusting for the effects of sociodemographic, disease, and treatment characteristics.
Of the 2532 women who were contacted, 1378 women replied to our initial contact letter or survey. Of these 337 refused and 1041 (41%) completed our survey. 47% of participants completed it online and 53% on paper. Reasons for declining included: a request to be removed from all cancer registry studies (57%), no interest in further childbearing (10%), the topic of infertility was too emotionally difficult to discuss (2%), the survey was too long (3%), or no reason given (28%). The average time to complete the survey was 26 minutes. Of the 1041 women who completed the survey, 918 reported treatment with the potential to compromise their fertility and were included in our analyses of fertility counseling and preservation.
Table 1 shows comparisons of the 1041 responders and 1491 non-responders, based on disease and demographic data in the cancer registry. Patients who completed the survey were 1.4 years younger at diagnosis than those who did not (P<0.0001), and were diagnosed with more aggressive cancers, as indicated by a SEER summary stage index (range of 0 (in situ) to 7 (metastatic)) of 3.7 vs. 3.4 (P=0.0008). There were no differences between responders and non-responders in socioeconomic status (calculated from median income and education for the census block group of residence at diagnosis, P=0.8) or years since diagnosis (P=0.2).
The age and childbearing desires of the 918 participants who reported treatment with potential to affect fertility are listed in Table 2. Patients with a history of breast and gastrointestinal cancers tended to be oldest at diagnosis and most likely to have had children before treatment. Depending of the type of cancer, between 47-63% of respondents reported desiring to have children after treatment, with the highest rates among women with leukemia (59%) and Hodgkin’s (63%). These latter two groups were also composed of women with the lowest mean ages.
Overall, 61% of women were counseled on the risk of cancer treatment to fertility by the oncology team (Table 3). Age at diagnosis, the desire for future children at diagnosis, and education level were significantly associated with an increased odds of receiving fertility preservation counseling. No significant differences in access to counseling were noted with regard to: parity, marital status, household income, sexual orientation, population density, or ethnicity.
After adjustment for age, desire for future children, parity at diagnosis, household income, ethnicity, disease type, disease stage, and treatment type, achieving a higher education level remained associated with increased counseling (OR 1.4, 95% CI 1.0 - 2.1, Table 3). Women who had attained at least a bachelor’s degree were 20% more likely to be counseled than those who had not. Age at diagnosis and desire for future children at diagnosis were not associated with having received counseling. However, women whose household income was less than $30,000 per year may have been less likely to be counseled than those whose income was between $30,000 and $100,000 per year (OR 0.7, 95% CI 0.4 – 1.2), though this relationship did not reach statistical significance (P = 0.18). No significant differences in counseling were noted for women whose household income exceeded $100,000 per year versus those with between $30,000 and $100,000 per year.
Overall, 4% of women underwent fertility preservation (Table 4). Age at diagnosis, the desire for future children at diagnosis, parity at diagnosis, and education level were significantly associated with an increased odds of pursuing fertility preservation. There was a trend toward decreased access for Latina women versus Caucasian women. No significant differences in access to fertility preservation were noted with regard to: marital status, household income, or population density.
After adjustment for age, desire for future children, parity at diagnosis, education level ethnicity, disease type, disease stage, and treatment type, several differences persisted. For instance, women greater than 35 years old at diagnosis may have been approximately 90% less likely to preserve their fertility than their 18 to 25 year-old counterparts (OR 0.1, 95% CI 0.0 – 1.4, Table 4); however, this difference did not achieve statistical significance. A trend was seen with regard to pre-treatment parity, where women who already had at least 1 child at diagnosis were had 70% lower odds of pursuing fertility preservation than women without children (OR 0.3, 95% CI 0.1 – 1.1); however, this difference also did not achieve statistical significance. Latina women may have been 80% less likely to preserve fertility than Caucasian women (OR 0.2, 95% CI 0.0 – 1.3), though this difference did not achieve statistical significance.
Furthermore, no fertility preservation was noted among the 31 women identifying as African American, despite having no significant differences in childbearing from Caucasian women in our study. A similar pattern was seen among the 29 women who identified with a sexual orientation other than heterosexual, despite having no differences in childbearing desires compared to women identifying as heterosexual.
The prevalence of counseling by subjects’ oncology teams was 44% in 1993. After adjustment for age, desire for future children, parity at diagnosis, education, ethnicity, disease type, disease stage, and treatment type the prevalence of counseling increased significantly at a rate of approximately 0.9% per year from 1993 to 2007 (R-squared = 0.1, P < 0.05). The prevalence of fertility preservation was zero percent in 1993. By 2000, the prevalence of fertility preservation was two percent. After controlling for age, desire for future children, parity at diagnosis, education level ethnicity, disease type, disease stage, and treatment type, the prevalence of fertility preservation increased significantly at approximately 1.3% per year from 2000 to 2007 (R-squared = 0.11, P = 0.03) (Figure 1).
Though the prevalence of both counseling and of fertility preservation utilization has increased over time, there remains a large unmet need for fertility preservation. Across the state of California, it appears that certain groups of women are more likely to receive important information about reproductive health at the time of diagnosis, and are also more likely to preserve their fertility than others.
The prevalence of counseling and access to fertility preservation observed in this study is consistent with that previously reported by others. The 61% rate of counseling observed in our study is similar to previously reported rates of 34 to 72%.18 We observed that 4% of women underwent fertility preservation, which is consistent with the work of Jenninga and colleagues, whose data demonstrate a 2% rate of fertility preservation in the Netherlands between 2002 and 2008 as well as low utilization rates at academic medical centers in the United States.7,19,20 The observed 4% rate of utilization is also similar to the low utilization of infertility services in general.21 And, consistent with our observed rise in preservation rates starting in the year 2000, Lee and colleagues found over the last decade that women have become more likely to see a fertility doctor before cancer treatment to preserve fertility as opposed to seeing them afterward to treat infertility.22
Certain aspects of socioeconomic status appear to be associated with access to counseling about fertility loss and access to fertility preservation. For instance, women who graduate with a Bachelor’s degree are more likely to discuss the risks of cancer treatment to fertility than women who do not. This may reflect an assertiveness and proactive attitude on the part of patients with higher health literacy that may have been associated with patients being more likely to bring up the topic themselves. This is consistent with disparities seen in delivery of timely cancer care, where young women with lower levels of education are more likely to experience delays in diagnosis and treatment of cervical and breast cancers.11 Medical providers report their lack of awareness and comfort in discussing fertility, and their knowledge of available resources, as barriers to fertility discussions.23 A recent telephone survey of 282 French women, aged 20-44, who survived unspecified cancers indicated that, of the 37% of women who reported treatment-induced infertility, 30% of those women reported that they had not been informed of the risk of infertility before their treatment.24 Since oncologists often do not discuss fertility unless the patient raises the issue, patients may benefit from a certain education background in their ability to independently pursue further information.23
Household income may affect access. While we did not find a significant relationship between income and access, our relatively uneven distribution, with low sampling of women with incomes less than $30,000 may have limited our ability to detect true differences. Women whose household incomes are insufficient to adequately meet the basic needs of an average American family may be less likely to receive infertility counseling by their oncologists than women with higher incomes. Other studies including young women with cancer have showed that lower income is associated with poorer outcomes, including increased probability of delay in diagnosis.11 In a study of 314 cancer patients who were referred to a reproductive endocrinologist for fertility services, Lee and colleagues found no association between income and whether women were able to present in a timely fashion and preserve fertility versus whether they presented after cancer treatment with diagnosed infertility.22 Duffy and colleagues also showed in a study of 166 breast cancer survivors that income did not predict whether women were counseled about fertility issues by their oncologist.23 The possibility that income does impact one’s ability to access fertility preservation would not be surprising, given that fertility preservation is not routinely covered by insurance and that the costs of fertility preservation are significant.25 The average total cost of clinical services, consultations, procedures necessary to retrieve eggs and/or produce embryos, and freezing services is between $8,000 and $24,000.26,27
Women who understand the potential benefits of fertility preservation may be willing to make an extra effort to seek out additional resources. Despite the cost, it may be the case that women from lower income groups receive financial assistance from non-profit organizations or from family members. We also observed no difference in fertility preservation utilization between women who live in urban areas and those in rural areas, despite the need for women in rural regions to expend greater resources in order to travel further distances to specialized fertility centers.
We found no differences in counseling or fertility preservation with regard to partner status, but did show a trend toward decreased rates of fertility preservation for women who had previous children at the time of diagnosis. The finding that partner status does not predict counseling was also reported by Duffy and colleagues.23 Our study, like Duffy et al., also showed no difference with regard to counseling in women with previous children. However, there was a decrease in utilization of fertility preservation among women who were already parents at diagnosis, despite controlling for age and parenting desires. The emotional impact of not preserving fertility could be higher in women without previous children. In prior studies, women who experienced primary infertility (desired a child at diagnosis but were never able to have a child) experience greater distress than women with secondary infertility (had at least one child at diagnosis but were unable to have another).28,29
This study showed that women who are over 35 years old at diagnosis could less likely to be counseled than women in their early twenties, regardless of parenting desires or parity at diagnosis. This is consistent with the findings of Duffy and colleagues in young breast cancer patients.23 Age also predicted lower rates of fertility preservation in our study by univariate analysis, but the relationship did not retain significance in our multivariate model. Age at diagnosis has been associated increased risk of cancer recurrence, with difficulty in obtaining insurance, and in employment base discrimination of young women with cancer.30,31 Taken together with our findings, age could remain a focus of future study as a predictor of access to fertility preservation.
This is the first study to examine the association of sexual orientation and utilization. In a study of 249 oncologists across the United States, Forman et al. found that 1% of the reasons oncologists did not refer to a reproductive endocrinologist was because a patient identified as lesbian.8 Forman and colleagues study did not comment on childbearing desires, so it is difficult to know whether these women requested or declined referral based on parenting plans. This study demonstrates that there is no significant association between sexual orientation and oncologists counseling women about potential reproductive compromise after treatment. However, this study does show that, despite there being no significant differences in parenting desires based on sexual orientation, none of the women who identified with an orientation other than heterosexual preserved their fertility. Given the low overall utilization rates, our study may be underpowered, and possibly subject to sampling bias, with 31 women not identifying as heterosexual. Assessing the impact of sexual orientation on seeking fertility preservation should also be a focus of future research.
Caucasian women in our study appeared more likely to seek fertility preservation services than Latina or African American women, though our observation did not achieve statistical significance. This is in contrast to Lee and colleagues, who found that neither race nor ethnicity predicts whether women will present to fertility centers for fertility preservation before cancer treatment or for fertility care after cancer treatment.22 The data in our study are consistent with patterns seen with regard to health disparities in general infertility care, as women from many ethnic backgrounds are less likely to have access to any specialized fertility treatment, regardless of whether it be before or after a cancer diagnosis. African American women often seek fertility treatment after waiting a longer time with untreated infertility than Caucasian women.32,33 Low-income Latino immigrant patients, as well as African Americans and Arab Americans, face challenges in linguistic and cultural barriers as well as accessibility and affordability of treatment.34,35 Our findings are consistent with patterns seen in general infertility treatment in Massachusetts – a state with mandated insurance coverage for infertility services, where such services are disproportionately accessed by Caucasians, by the highly educated, and by the wealthy.33
This study has several important strengths and limitations. A significant proportion of the eligible participants in our study were not reached. However, the response was similar to that seen in several other registry-based studies of young female cancer survivors, with rates of 26-51%,3,15,36,37 And, given the importance of survivorship issues, the National Cancer Institute recently recommended the use of cancer registries as a means of conducting studies that could rapidly yield information about life after treatment.38,39 Despite the benefits of registry-based studies, the possibility of sampling bias remains a concern in this study. While women who responded were slightly younger at diagnosis than those who did not, there were no differences with regard to socioeconomic status. It is also likely, given the consistency of our counseling and preservation rates with other studies, that we achieved a relatively accurate sampling of the population, with regard to previous exposure to fertility preservation. However, other important confounding factors that were not controlled for in our study include the absence of information regarding current disease state (i.e., recurrence or not), and residual side effects.
We’ve observed that women who are childless, younger, Caucasian, heterosexual, and who graduated from college may be more likely than women of different backgrounds to be counseled about the risks of cancer treatment to fertility or to preserve fertility before cancer treatment. A focus on the quality of counseling and the quantity of patients seeing a fertility specialist has likely led to improved rates of access over the last decade. However, failing to focus on equity will likely result in continued unequal distribution of reproductive health counseling and fertility preservation. An opportunity lies ahead to explore educational and policy interventions to ameliorate health disparities that may exist in the growing field of fertility preservation.
This project was supported by National Institute of Health Grant Number TL1 RR024129. The National Institute of Health had no role in study design; in collection, analysis, and interpretation of data; in the writing of this report; or in the decision to submit this paper for publication.
Conflict of Interest Disclosures The authors have no conflicts of interest to declare.