There is a dearth of existing research on the treatment of reproductive cancers among women with intellectual and developmental disabilities (IDD). This study analyzed the 2010 Healthcare Cost and Utilization Project Nationwide Inpatient Sample and compared the prevalence of reproductive cancer treatment hospitalization discharges among women with and without IDD. Discharges linked to women with IDD had higher incidences of cancer of the uterus and lower prevalence of cancer of the cervix. Moreover, discharges linked to women with IDD indicated these women were younger, had longer hospital stays, and were more likely to have public insurance coverage. Therefore, further research and targeted interventions to increase cancer prevention and screening are urgently needed.
Through the development of technology and improved supports, people with intellectual and developmental disabilities (IDD) are living longer than ever before (Emerson & Hatton, 2008). At the same time, there is a growing incidence of chronic illness, such as cancer, among adults with IDD (Hogg & Tuffrey-Wijne, 2008; Tuffrey-Wijne, 2003). Indeed, cancer is often associated with aging (McPherson, Steel, & Dixon, 2000). This is particularly worrisome because people with IDD have greater health care needs than the general population and these needs are often unmet (Backer, Chapman, & Mitchell, 2009).
In the past decade, a growing body of research has emerged regarding cancer and people with IDD, most of which has focused on screening and prevention. Numerous studies have examined access to preventive cancer screening for people with IDD, especially access to mammography and Pap tests (e.g., Greenwood, Wang, Bowen, & Wilkinson 2014; Merten, Pomeranz, King, Moorhouse, & Wynn, 2015; Parish, Swaine, Son, & Luken, 2013a; 2013b; Wilkinson, Lauer, Freund, & Rosen, 2011). Compared to the general population, people with IDD are significantly less likely to receive appropriate cancer screening (Merten et al., 2015; Osborn et al, 2012; Sullivan et al., 2003; Wilkinson, Culpepper, & Cerreto, 2007). Indeed, a study in Canada found that compared to other women, women with IDD are nearly half as likely to be screened for cervical cancer and 1.5 times less likely to receive mammography (Cobigo et al., 2013). Another study found that despite being known to be sexually active, women with IDD are less likely than women without IDD to be screened for cervical cancer (Brown, Plourde, Ouellette-Kuntz, Vigod, & Cobigo, 2016).
The disparities in cancer screening among women with IDD are likely due to a number of factors, such as poor understanding of the benefits of screening by the women (Swaine, Parish, & Luken, 2013) and their family caregivers (Swaine, Dababnah, Parish, & Luken, 2013); relatively low literacy (McIlfatrick, Taggart, & Truesdale-Kennedy, 2011); or prior sexual assault histories (Brown & Gill, 2009; Swaine, Dababnah, et al., 2013; Swaine, Parish, et al., 2013). Moreover, physicians may not offer Pap tests to women with IDD because of the time and preparation required to prepare these women to understand the procedure or because of incorrect assumptions that these women are not sexually active (Parish et al., 2013a). Similarly, barriers to accessing breast cancer screening include issues related to informed consent and lack of appropriate transportation (McIlfatrick et al., 2011). Additionally, support staff and caregivers often lack awareness and training of the importance of cancer screening (Hanna, Taggart, & Cousins, 2011; McIlfatrick et al., 2011; Truesdale-Kennedy, Taggart, McIlfatrick, 2011).
Sociodemographic factors also lead to disparate rates of cancer screening among women with IDD. For example, the relative poverty most women with IDD often experience results in lower rates of cancer screening (Wilkinson, Lauer, Freund, & Rosen, 2011). Moreover, differences between racial groups exist. A 2013 study in the United States found that African American women with IDD receive mammography at significantly lower rates than White women with IDD (Parish et al., 2013b).
In addition, there is a growing body of research concerning the prevalence of cancer among people with IDD. Generally, studies have found that women with intellectual disability (ID) are at reduced risk of cervical cancer but increased risk of cancers of the corpus uteri and ovaries (Hogg & Tuffrey-Wijne, 2008; Patja, Eero, & Iivanainen, 2001). Some studies indicate breast cancer incidence is similar in women with and without ID (Evenhuis, 1997; Patja, Eero, & Iivanainen, 2001; Sullivan et al., 2003). However, other evidence suggests women with IDD experience a greater number of risk factors for breast cancer compared to women without IDD (Taggart, Truesdale, & McIlfatrick, 2011). Additionally, incidence rates may vary depending on the etiology of women's IDD. Lower occurrence of breast cancer has been found for women with some disabilities, such as Down syndrome (Satgé & Sasco, 2002) and fragile X syndrome (Schultz-Pedersen, Hasle, Olsen, & Friedrich, 2001), whereas the incidence of breast cancer is elevated among women with cerebral palsy (Strauss, Cable, & Shavelle, 1999). Unfortunately, people with IDD tend be diagnosed at an advanced stage of cancer progression and the delayed diagnosis of breast cancer may result in markedly deleterious differences in disease progression (Satgé et al., 2014).
Despite the growing evidence about the prevalence of cancer among people with IDD, there is exceptionally limited research concerning treatment. An investigation of hospitalization trends of Australians with IDD found that those who had cancer were admitted to hospitals much more frequently than those without cancer and women were more likely to be hospitalized for cancer treatment than men. (Sullivan & Hussain, 2008). Further, a study in the United States found that compared with women who do not have disabilities, those with mental disorders and neurological conditions have significantly lower rates of breast conserving surgery and radiation therapy (Iezzoni et al., 2008). Unfortunately, research indicates that women with ID are more likely to die from their breast cancers than women without disabilities (Iezzoni et al., 2008; McCarthy et al., 2006; Satgé et al., 2014).
Notably, in the United States, there are no population-based studies of cancer treatment of women with IDD. In light of the scarce existing literature on reproductive cancer treatment among people with IDD, the aims of this study were to (1) determine the relative prevalence of discharges for reproductive cancer treatment hospitalizations linked to women with and without IDD; and (2) identify the characteristics of women who were discharged after receiving these treatments.
The present study used data from the 2010 Healthcare Cost and Utilization Project (HCUP; Agency for Healthcare Research and Quality, n.d.), which is nationally representative of hospital discharges. HCUP is sponsored by the Agency for Healthcare Research Quality in partnership with state-level data collection organizations. The aim of HCUP is to collect health care data including inpatient care, ambulatory surgery, and emergency department service utilization and costs in the United States annually. The HCUP Nationwide Inpatient Sample is a database of hospital inpatient stays and it is nationally representative of all community hospitals. It provides nationwide data on approximately eight million hospital stays each year from about 1,000 hospitals which approximate a 20% stratified sample of United States community hospitals (Agency for Healthcare Research & Quality, n.d.).
The HCUP Nationwide Inpatient Sample contains more than 100 clinical and nonclinical data elements for each hospital stay. For example, data on (a) primary and secondary diagnoses and procedures, (b) admission and discharge status, (c) patient demographic characteristics (e.g., sex, age, race, median household income for zip code), (d) hospital characteristics (e.g., ownership, size, teaching status), (e) expected payment source, (f) total charges, (g) discharge status, (h) length of stay, and (i) severity and comorbidity measures were included in the Nationwide Inpatient Sample. The Nationwide Inpatient Sample is drawn from a sampling frame that contains hospitals comprising about 90% of all discharges in the United States. The 2010 HCUP included 45 states and comprised over 96% of all U.S. hospital discharges (Houchens & Elixhauser, 2006).
For this study, data were derived from the 2010 Nationwide Inpatient Sample, including 1,051 hospitals from 45 states. The annual number of total discharges reported in the database was 7.8 million, and after weighting, these represented 39.0 million discharges. Additional methodological details of the Nationwide Inpatient Sample dataset are available elsewhere (Steiner, Elixhauser, & Schnaier, 2002).
The present study's sample was restricted to hospital discharges with diagnoses of one of five reproductive cancers (uterus, breast, cervix, ovaries, other female genital organs). Reproductive cancer treatment hospitalizations were identified using the International Classification of Disease, Ninth Revision, Clinical Modifications (ICD-9-CM) codes (Centers for Disease Control and Prevention, n.d.). Discharges with a principal diagnosis of any of these five reproductive cancers were used for the analyses. Discharges linked to women with IDD were identified from ICD-9 codes (see Table 1 for complete listing) and were compared to discharges of women without such ICD-9 codes.
Notably, the unit of analysis for the present study, consistent with HCUP records, was the hospital discharge, which is not a person or patient in the Nationwide Inpatient Sample dataset. In other words, if a woman was admitted to the hospital multiple times in the same year, she would be counted each time as a separate “discharge” from the hospital. We identified 456 hospital discharges of women with reproductive cancer who had ICD-9 codes for IDD. The comparison group was formed by the 155,914 hospital discharges of women with reproductive cancer who did not have an ICD-9 code for IDD. When weighted, this analytic sample represented 2,259 hospital discharges of women with IDD and 780,789 hospital discharges of women without IDD.
Independent variable of interest
The study's key independent variable was whether or not the discharge was for a woman with IDD. Hence, the unit of analysis is the hospital discharge rather than the individual patient. This independent variable was created based on ICD-9 diagnoses and procedure codes (see Table 1) and was derived from single-level Clinical Classifications Software (Agency for Healthcare Research and Quality, 2012). Clinical Classifications Software is a tool developed as part of HCUP for clustering diagnoses and procedures into a manageable number of clinically meaning categories (Agency for Healthcare Research and Quality, 2012).
The dependent variables included five reproductive cancer treatment hospitalizations and their identifying Clinical Classifications Software codes based on ICD-9 diagnoses and procedure codes, as follows: (1) cancer of uterus (Clinical Classifications Software code 25), (2) cancer of breast (Clinical Classifications Software code 24), (3) cancer of cervix (Clinical Classifications Software code 26), (4) cancer of ovary (Clinical Classifications Software code 27), and (5) other female genital organs (Clinical Classifications Software code 28).
To explore characteristics of the discharged women, we compared those with and without IDD diagnoses across (a) age (i.e., women's age in years), (b) race (i.e., non-Hispanic White, non-Hispanic Black, Hispanic, or non-Hispanic other), (c) insurance payer (i.e., Medicare, Medicaid, private, or uninsured), (d) median household income for the patient's zip code (i.e., $1–$38,999, $39,000–$47,999, $48,000–$62,999, $63,000 or more), (e) length of hospital stay (in days), (f) whether or not the patient died in the hospital, and (g) total hospital charges (i.e., the amount the hospital billed for the entire hospital stay, excluding physician fees).
This descriptive, exploratory study sought to describe discharges of patients who are receiving reproductive cancer treatments and to examine whether or not there are differences in hospital discharges of women with and without IDD in the United States.
First, Chi-square statistics and univariate linear and logistic regression analysis with a significance level of 0.05 were used to test for differences in the characteristics among the two groups of discharges (i.e., women with and without IDD) who were receiving reproductive cancer treatments (n = 156,370). Second, to examine whether or not there were differences in hospital discharges of women with and without IDD, Chi-square test statistics were used to test bivariate associations between discharges linked to women with IDD and women without IDD on all cancer hospitalizations (n = 500,750).
Proper p value reflecting the complex design cannot be evaluated based on uncorrected Pearson Chi-square statistic, and can be evaluated from the design-based F statistic (Lee & Forthofer, 2006). As such, to obtain a proper p value from the design-based F statistic, the Pearson Chi-square test statistic was corrected for the survey design with the second-order correction of Rao and Scott (1984) and converted into an F statistic using Stata 13.0 (StataCorp, 2013).
Table 2 describes the characteristics of the study sample. Discharges linked to diagnoses of IDD were significantly younger (mean = 61 years) than discharges of women without IDD (mean = 68 years). Women with IDD discharged from hospitals were more likely than other women to have public health insurance coverage (Medicaid or Medicare). Additionally, women with IDD had longer mean hospital stays than women without IDD (5.6 vs. 4.3 days). Finally, women with IDD who were discharged, were much more likely to live in zip codes with lower median income. There were no differences between the groups of discharged women in terms of race or Latino ethnicity, death in hospital, or total hospital charges.
Table 3 reports the unadjusted, weighted prevalence comparison between the two groups of discharged women in terms of each type of the five reproductive cancers (i.e., breast, uterus, cervix, ovary, and other female genital organs) among hospitalizations principally for cancer treatments (n = 500,750). There were statistically significant differences between discharges of women with and without IDD on uterine and cervical cancer treatment hospitalizations. Cancer of the uterus was significantly more prevalent among discharges linked to a diagnosis of IDD than it was for other women (8% vs. 6% of cancer hospitalizations). Hospitalization for cancer of the cervix was less prevalent for discharges of women with IDD than it was for women without IDD (2% vs. 4% of cancer hospitalizations). There were no statistically significant differences in the prevalence of hospitalizations for cancers of the breast, ovary or other female genital organs. Breast cancer hospitalizations were most common for both groups of women.
Using data from the 2010 HCUP Nationwide Inpatient Sample, this study analyzed hospital discharges for reproductive cancer treatment hospitalization of women with and without IDD in the United States. Compared to other reproductive cancer treatment hospitalization discharges, those that were linked with women with IDD were more likely to be treated for cancer of the uterus and less likely to be treated for cancer of the cervix. Discharges linked to women with ID were significantly younger, more likely to have public insurance coverage, more likely to live in low-income neighborhoods, and have longer hospital stays. Notably, there were no statistically significant differences in discharges of women with and without ID in the prevalence of hospitalizations for cancers of the breast, ovary, or other female genital organs.
Limitations and Future Research
Several limitations to the present study highlight important areas of inquiry for future research. These limitations are primarily related to the dataset itself. First, the unit of analysis was the hospital discharge rather than the individual patient. Therefore, if a woman was admitted to the hospital multiple times in the same year, she would be counted each time as a separate discharge from the hospital. This is notable because research indicates that women with IDD who have cancer are more likely to have repeat hospitalizations (Sullivan & Hussain, 2008). Moreover, generally, people with IDD are more likely to be hospitalized for ambulatory care procedures (Balough, Hunter, & Ouellette-Kuntz, 2005). Second, as with all administrative datasets, some data in the HCUP may have been miscoded or absent and verification was not possible. Third, this study likely underestimates the number of women with IDD because some women may not have been assigned the ICD-9 code for their disability since they were receiving care for cancer and not for their disability. Fourth, the HCUP data does not include information on outpatient procedures. This is notable since research suggests a trend shift away from inpatient treatment toward the outpatient setting. For instance, some cancer patients now receive chemotherapy and radiation therapy in the outpatient clinic setting (Halpern & Yabroff, 2008; Rubenstein, 1994). Thus, future research should examine data that includes information on both inpatient and outpatient treatments. Likewise, future research should utilize nationally representative data to ascertain the incidence of reproductive cancers of people with IDD in the United States. Finally, as an administrative dataset, HCUP does not permit longitudinal analysis, which could elucidate further information on long-term health outcomes.
As a descriptive study, and one of the first to address this issue, this study raises important questions for future research. Because the current study used discharge rather than the individual as the unit analysis, future research should seek to use the individual as the unit of analysis. Future research should examine longitudinal data that includes information on both impatient and outpatient treatments as well as elucidate further information on long-term health outcomes. Additionally, findings from the present study do not permit analysis of the extent to which the hospitalizations of women with IDD are appropriate. For example, is the reduced likelihood of hospitalization for cervical cancer due to decreased incidence of this cancer, or to a lack of screening and detection? The existing evidence suggests cervical cancer screening rates are well below clinical guidelines for women with IDD (Parish et al., 2013a). In addition, future research should seek to ascertain why women with IDD are significantly younger when hospitalized. Moreover, research should examine reasons for the higher likelihood of treatment for uterine cancer for women with IDD who were hospitalized. Further, a greater understanding as to why women with IDD have longer cancer treatment hospital stays is critical for health care policy and practice. Finally, greater attention must be given to determining the prevalence of reproductive cancers among women with IDD in the United States. There are significant methodological challenges associated with deriving credible estimates. Notably, however, prior studies have found that women with IDD are at reduced risk of cervical cancer but increased risk of cancers of the corpus uteri and ovaries (Hogg & Tuffrey-Wijne, 2008; Patja, Eero, & Iivanainen, 2001). Thus, further research in this arena is warranted.
Despite these limitations, this study has some important strengths. The HCUP provides high-quality, nationally representative data and thus allows us to draw inferences about the entire United States population of women with IDD who were hospitalized for reproductive cancer treatments. Accordingly, this study is not constrained by selection bias or sampling bias arising from convenience samples derived from a single hospital, disability service organization, or single geographic region. Moreover, to the best of our knowledge, this study is the first to use nationally representative data to compare the reproductive cancer treatment hospitalization of women with and without IDD in the United States. Therefore, this study addresses some notable gaps in the existing research.
Implications for Policy and Practice
Today, thanks to technological advances, health care treatment improvements, and better supports, people with IDD are living longer than ever before (Emerson & Hatton, 2008). Concurrently, these longer life expectancies translate into increased incidence of cancer for adults with IDD (Hogg & Tuffrey-Wijne, 2008; Tuffrey-Wijne, 2003). This study reinforces the need for greater attention by policymakers and health care professionals to cancer prevention and screening for people with IDD as well as best practices for treating this population, because so little is known about these aspects of their health (Hanna, Taggart, & Cousins, 2011).
The present study is part of a larger movement that champions an evolving understanding of the notion of disability and health itself. As Rimmer (1999) argued, early efforts toward health promotion largely ignored people with disabilities, treating them as already sick. Rimmer characterized a new paradigm was emerging to refocus the emphasis in health care for people with disabilities from disability prevention to health promotion. However, empirically tested, evidence-based interventions remain scarce (Marks & Heller, 2003; Parish, Mitra & Iezzoni, 2016; Swaine et al., 2014). The current study's findings related to the high incidence of discharges linked to women with ID supports evidence that women with IDD need to engage in health promotion activities that reduce the incidence of cancer.
Surely, the best way to reduce the incidence of reproductive cancer among women with IDD, as well as ensure the best prognosis, is prevention and early detection. This is particularly noteworthy since prior studies suggest women with IDD are as likely, and for certain types, more likely, as women without IDD to have reproductive cancers (Evenhuis, 1997; Hogg & Tuffrey-Wijne, 2008; Patja et al., 2001; Sullivan et al., 2003). Moreover, women with IDD who have breast cancer are often diagnosed at an advanced stage and the delayed diagnosis often results in worse prognosis (Satgé et al., 2014). Despite these risks, women with ID are significantly less likely than women without IDD to receive regular Pap tests or mammograms (Merten, et al., 2015; Osborn et al, 2012; Sullivan et al., 2003; Wilkinson, Culpepper, & Cerreto, 2007). Accordingly, health care professionals and policymakers must implement measures, such as training and guidelines for best practices, to ensure all patients, regardless of disability status, receive regular and appropriate diagnostic tests. Likewise, women with IDD and their caregivers, who often play an important role in assisting them to access health care services, need increased education on the importance of regular Pap tests and breast cancer screening (Parish, Moss, & Richman, 2008). Such training and education must be based on a commitment to promoting better health among this population and be fully accessible to the individuals' needs.
Moreover, increased attention must be given to appropriately treating women with IDD who have cancer. Findings from the present study suggest that women with cancer who have IDD may be more likely to be treated in-patient and have longer hospital stays. These findings corroborate past studies that have found people with IDD who have cancer are admitted more frequently than people with ID without cancer and at risk of repeat hospitalizations because of lack of supportive environments or lack of compliance with treatment (Sullivan & Hussain, 2008). Despite the high incidence of hospitalization among people with IDD, physicians and other health care professionals lack standards of care and best practices specific to this population (Voelker, 2002). In addition, there is longstanding research regarding the attitudes of health care professionals towards this population that suggests attitudinal barriers to access to quality care are prevalent (Ouellette-Kuntz, Burge, Henry, Bradley, & Leichner, 2003). Indeed, physicians of patients with IDD report feeling unprepared to treat this population, as if they are “operating without a map.” (Wilkinson, Dreyfus, Cerreto, & Bokhour, 2012). As such, it is imperative that training for physicians and other health care professionals include specific information on how to appropriately and effectively treat people with IDD.
Finally, it is vital that there be better coordination between health care professionals and disability support providers. Staff working with people with IDD need increased training and awareness about health promotion and cancer prevention (Hanna, Taggart, & Cousins, 2011). Moreover, direct support staff, who typically have much greater understanding on how best to support people with IDD, must work directly with health care professionals to ensure they are provided appropriate information regarding their cancer diagnosis (Tuffrey-Wijne, Bernal, & Hollins, 2010).
This study analyzed nationally representative hospital discharge data from the 2010 HCUP National Inpatient Sample and offers new descriptive information about reproductive cancer treatment hospitalizations among women with and without IDD. The evidence suggests that reproductive cancer treatment hospital discharges linked to IDD were significantly younger, more likely to have public insurance coverage, more likely to live in low-income neighborhoods, and have longer hospital stays. Further, discharges linked to the diagnosis of IDD were more likely to be hospitalized for cancer of the uterus and less likely to be hospitalized for cancer of the cervix. More research is needed to understand whether these rates of hospitalization are appropriate – that is, are women with IDD receiving cancer diagnosis and treatment when it is needed?
Partial support for this article was provided by a grant from the NIDILRR (#H133 B080009). However, contents do not necessarily represent the policy of the funder, and endorsement by the federal government should not be presumed. Partial support was also provided by the Lurie Institute for Disability Policy at Brandeis University.