Preconception Health Characteristics of Women with Disabilities in Ontario: A Population-Based, Cross-Sectional Study

Abstract

Background: There is growing recognition that preconception health, defined as the health of all reproductive-age individuals, impacts reproductive and perinatal outcomes. Although women with disabilities are becoming pregnant at increasing rates, little is known about their preconception health. Our objective was to describe the preconception health characteristics of women with physical, sensory, and intellectual/developmental disabilities and compare these characteristics with women without disabilities.

Materials and Methods: We conducted a population-based cross-sectional study of 15- to 44-year-old women with physical (n = 253,184), sensory (n = 93,170), intellectual/developmental (n = 8,986), and multiple disabilities (n = 29,868), and women without these disabilities (n = 2,307,822) using Ontario health administrative data (2017–2018). We described preconception health variables related to social determinants of health, physical health status, psychosocial well-being, history of assault, medication use, and continuity of primary care and compared women with and without disabilities in crude and age-standardized analyses, with standardized differences >0.10 indicating clinically meaningful results.

Results: Women with physical, sensory, intellectual/developmental, and multiple disabilities had poorer preconception health than women without disabilities. Disparities were pronounced for physical health status, psychosocial well-being, use of potentially teratogenic medications, and history of assault. Of all groups, women with intellectual/developmental disabilities had the greatest disparities.

Conclusion: Further research is needed to identify contributors to poor preconception health among women with disabilities and to develop tailored preconception health interventions to meet their unique needs and experiences.

Introduction

Preconception health, which describes the health of all reproductive-age individuals,¹ impacts reproductive and perinatal outcomes, including fertility, pregnancy complications, stillbirth, preterm birth, and congenital anomalies.¹ Because nearly half of pregnancies are unplanned,² the World Health Organization notes that health promotion activities for all reproductive-age individuals, regardless of their pregnancy intentions, are critical for reducing risks of these adverse outcomes.¹

Preconception interventions take a broad approach, addressing reproductive history, social determinants of health, health behaviors, physical health, psychosocial well-being, medication use, and access to primary care^1,3 and have primarily targeted women.¹ For women with elevated risks for poor perinatal outcomes, such as those with diabetes,⁴ specialized preconception interventions have been developed. Approximately 12% of 15- to 44-year-old North American women have a physical disability, affecting mobility flexibility, or dexterity; sensory disability affecting hearing or vision; or intellectual/developmental disability affecting cognitive or adaptive function.^5,6 However, few preconception interventions exist for women with disabilities.

Historically, childbearing rates in women with disabilities were low.⁶ With deinstitutionalization, repeal of involuntary sterilization laws, and medical advances, more women with disabilities are experiencing pregnancy.^6–8 Of note, however, existing studies show that compared with women without disabilities, women with disabilities have higher rates of unintended (unwanted or mistimed) pregnancies,⁹ are more likely to smoke, not exercise, be overweight, and have poor physical and mental health, and are less likely to access preventive health care.^10–12 However, these data are limited by use of community-based surveys that may not include the most marginalized disabled women. Moreover, they focus on physical health rather than using a comprehensive preconception health lens. Population-based data on a broad range of preconception health characteristics are needed to develop interventions tailored to the unique needs of women with disabilities.

Using linked population-based health administrative datasets in Ontario, Canada, the objective of this study was to describe the preconception health characteristics of women with physical, sensory, and intellectual/developmental disabilities and compare these characteristics with women without such disabilities.

Materials and Methods

Study design

We conducted a population-based, cross-sectional study in Ontario, Canada, which has 14 million residents, all of whom receive publicly funded health care. We studied women with physical, sensory, and intellectual/developmental disabilities, and those without these disabilities, who were 15–44 years at the midpoint of fiscal year 2017–2018. Consistent with public health approaches to preconception health,^1,3 we included all reproductive-age women, regardless of subsequent pregnancy status, excluding those who could not become pregnant because of a hysterectomy. Data use was authorized under section 45 of Ontario's Personal Health Information Protection Act,¹³ which does not require Research Ethics Board review.

Study population

We accessed and analyzed data at ICES (Toronto, Ontario), an independent, nonprofit organization that captures health data of Ontario residents who have used the health care system. We used databases on sociodemographics, physician visits, emergency department visits, and hospitalizations (Appendix Table A1), which are complete and accurate.¹⁴ The data were linked using a unique, encoded identifier and analyzed at ICES to preserve participants' confidentiality.

Table 3.

Most Common Potentially Teratogenic Medications Used by Women With and Without Disabilities, Ontario, 2017–2018

Types of potentially teratogenic medicationsa	Physical disabilities only (N = 48,022)	Sensory disabilities only (N = 14,391)	Intellectual/developmental disabilities only (N = 4,097)	Multiple disabilities (N = 11,767)	No disability (N = 231,823)
Anti-infectives and antiparasitic agents1	2,075 (4.3)	441 (3.1)	124 (3.0)	636 (5.4)	5,143 (2.2)
Anti-epileptic agents2	3,627 (7.6)	263 (1.8)	284 (6.9)	2,235 (19.0)	3,056 (1.3)
Antineoplastic agents3	703 (1.5)	46 (0.3)	6 (0.1)	169 (1.4)	415 (0.2)
Antithrombotic agents4	172 (0.4)	35 (0.2)	20 (0.5)	94 (0.8)	203 (0.1)
Dermatologicals5	297 (0.6)	100 (0.7)	27 (0.7)	66 (0.6)	1,683 (0.7)
Immunomodulating agents6	958 (2.0)	168 (1.2)	18 (0.4)	322 (2.7)	1,262 (0.5)
Pituitary, hypothalamic, and sex hormones7	529 (1.1)	179 (1.2)	55 (1.3)	182 (1.5)	1,795 (0.8)
Psycholeptic and psychoanaleptic agents8	946 (2.0)	208 (1.4)	125 (3.1)	241 (2.0)	2,672 (1.2)
Renin–angiotensin system agents9	1,313 (2.7)	433 (3.0)	126 (3.1)	582 (4.9)	2,815 (1.2)
Statins10	1,125 (2.3)	392 (2.7)	129 (3.1)	545 (4.6)	2,462 (1.1)
Other agents11	351 (0.7)	94 (0.7)	16 (0.4)	57 (0.5)	1,186 (0.5)

Data are presented as n (%).

^aExample of each potentially teratogenic medications group: 1. streptomycin; 2. phenobarbital; 3. cetuximab; 4. warfarin; 5. tetracycline; 6. penicillamine; 7. medroxyprogesterone and estrogen; 8. lithium; 9. ramipril; 10. pravastatin; 11. nicotine.

Measures

Following a systematic review to identify validated algorithms to measure physical, sensory, and intellectual/developmental disabilities in health administrative data,¹⁵ we developed an initial list of disability codes.^16,17 To these, we added diagnoses if they were judged by the Agency for Health Care Research and Quality Chronic Condition Indicator to be chronic,¹⁸ and rated by a group of 13 clinicians as being likely to result in activity limitations and participation restrictions, according to the International Classification of Functioning, Disability and Health definition of disability.¹⁹ Women were classified as having a disability if a diagnosis related to a physical, sensory, or intellectual/developmental disability was recorded in ≥2 physician visits or ≥1 emergency department visit or hospitalization from database inception to study entry. Women with ≥1 disability type were analyzed separately (“multiple disabilities”). As a comparison, we examined all other women without a recorded disability.

Although mental illness is often identified as a disability and women with mental illness experience many of the same risk factors for adverse reproductive and perinatal outcomes as women with other disabilities, we did not include mental illness in our definition of disability because (1) mental illness is also an indicator of psychosocial well-being captured in most conceptualizations of preconception health¹ and is therefore important to examine as an outcome, and (2) our focus on physical, sensory, and intellectual/developmental disabilities aligns with the National Institutes of Health's call for research in this area.²⁰

We described the age distribution of each group, and their preconception health characteristics, conceptualized based on established frameworks of risk factors for poor reproductive and perinatal outcomes, which follow a life course perspective.^1,3 These frameworks consider the social determinants of health, physical health status, psychosocial well-being, history of assault, medication use, and continuity of primary care, all of which are valuable targets for preconception interventions applied before pregnancy to optimize reproductive and perinatal health outcomes.^1,3

• (1)Social determinants of health were economic and social characteristics.^1,3 Neighborhood income quintile was measured by linking residential postal code with 2016 Census information on area-level median income. The Ontario Marginalization Index²¹ was used to measure area-level material deprivation (6 indicators, e.g., proportion of the population that is unemployed) and residential instability (7 indicators, e.g., proportion of dwellings that are not owned). Rural residence was measured with the Rurality Index of Ontario, which uses 10 indicators (e.g., presence of health care providers) to classify neighborhoods as rural (score ≥45) or urban (0–44).²² Although there are other important social determinants of health (e.g., martial status, race/ethnicity), these measures are not available in our health administrative data.
• (2)Physical health status related to common chronic medical conditions in reproductive-age women. We used validated disease registries to measure diabetes, hypertension, asthma, and HIV.^23–26 We also used the Johns Hopkins ACG^® System Version 9.0 to broadly classify comorbid conditions as stable or unstable (the latter based on greater acute health care use).²⁷
• (3)Psychosocial well-being was captured in diagnoses that indicate poor mental health. We measured psychotic disorders, mood/anxiety disorders, other mental illness, and substance use disorders in physician visits, emergency department visits, and hospitalizations, and self-harm in emergency department visits in the 2 years before cohort entry.
• (4)History of assault captured events related to victimization. We measured history of assault (i.e., physical or sexual assault or other maltreatment) in any emergency department visits or hospitalizations since database inception.
• (5)Medication use related to potentially teratogenic drugs that may be used to manage disability or other chronic condition symptoms,^28,29 but which may cause congenital anomalies when taken in pregnancy. We measured use of potentially teratogenic medications³⁰ in the year before cohort entry in those with available data, which was restricted to individuals eligible for Ontario's public drug benefits—with need based on income or disability, 11.5% of the cohort.
• (6)Continuity of primary care captured an indicator of regular access to a family physician or general practitioner. Continuity of primary care was measured using the Usual Provider Continuity Index, which calculates the proportion of visits made to the individual's usual primary care provider in a 2-year period,³¹ classified as low (≤50%), moderate (51%–80%), or high (>80%). Those with <3 visits were infrequent users.

Statistical analysis

Analyses were conducted in 2019 using SAS 9.4 (SAS Institute, Inc., Cary, NC). We first described the prevalence of physical, sensory, intellectual/developmental, and multiple disabilities in the cohort and the age distributions of each group. We then used frequencies and percentages to describe the preconception health characteristics of women with and without disabilities. We conducted crude and age-standardized analyses, because the average age of women with physical disabilities was older and that of women with intellectual/developmental disabilities were younger than women without disabilities. Differences between each disability group relative to women without disabilities were measured using standardized differences, which are independent of sample size.³² Standardized differences >0.10 are clinically meaningful.³²

Results

Cohort characteristics

There were 2,693,030 15- to 44-year-old women without a hysterectomy in Ontario in 2017–2018. The largest disability group was women with physical disabilities only (n = 253,184, 9.4% of women in Ontario), women with musculoskeletal disorders (e.g., disc disorders and rheumatoid arthritis) representing 46.4% of this group, neurological disorders (e.g., cerebral palsy and multiple sclerosis) representing 32.5%, permanent injuries (e.g., spinal cord injury and traumatic amputation) representing 17.2%, and congenital anomalies (e.g., reduction defects and spina bifida) representing 9.4%. (Disability subtypes may sum to >100% because women could have more than one subtype of physical, sensory, or intellectual/developmental disability.) Women with sensory disabilities were the second largest disability group (n = 93,170, 3.5%). Those who were deaf or had hearing loss represented 68.4% of this group, and those who were blind or had vision loss represented 33.2%. Only 0.3% of women had intellectual/developmental disabilities only (n = 8,986). Women with autism spectrum disorder represented 46.8% of this group, and women with intellectual disabilities represented 63.0%.

One percent (1.1%) of women had multiple disabilities (n = 29,868); women with physical and sensory disabilities were the largest group (73.3%), followed by physical and intellectual/developmental disabilities (15.2%), sensory and intellectual/developmental disabilities (6.1%), and all three types of disabilities (5.4%).

Preconception health characteristics

Tables 1 and 2 describe the crude and age-standardized preconception health characteristics of women with and without disabilities. Only clinically meaningful group differences (standardized differences >0.10) are reported here. Of all Ontario women 15–44 years of age, women with physical disabilities were the oldest group, with 40.4% being between 35 and 44 years (mean = 31.1, standard deviation [SD] = 8.14, vs. women with no disability, mean = 29.3, SD = 8.16). Conversely, women with intellectual/developmental disabilities were the youngest group, with 54.7% being between 15 and 24 years (mean = 25.2, SD = 7.86). Women with intellectual/developmental disabilities were the most disadvantaged; they were more likely than women without disabilities to live in neighborhoods with the lowest income quintiles and the highest material deprivation, even after age standardization.

Table 1.

Crude Comparison of Preconception Health Characteristics of Women With and Without Disabilities, Ontario, 2017–2018

	Physical disabilities only (N = 253,184)		Sensory disabilities only (N = 93,170)		Intellectual/developmental disabilities only (N = 8,986)		Multiple disabilities (N = 29,868)		No disability (N = 2,307,822)
Characteristic	n (%)	Std diffa	n (%)	Std diffa	n (%)	Std diffa	n (%)	Std diffa	n (%)
Age, years, mean (SD)	31.1 ± 8.14	0.21	29.5 ± 8.12	0.03	25.2 ± 7.86	0.52	30.1 ± 8.34	0.09	29.3 ± 8.16
15–19	27,959 (11.0)	0.12	127,41 (13.7)	0.04	2,703 (30.1)	0.36	3,923 (13.1)	0.06	351,047 (15.2)
20–24	36,044 (14.2)	0.07	15,814 (17.0)	0.01	2,211 (24.6)	0.19	5,056 (16.9)	0.00	387,211 (16.8)
25–29	42,237 (16.7)	0.03	18,642 (20.0)	0.05	1,563 (17.4)	0.02	5,382 (18.0)	0.00	415,122 (18.0)
30–34	44,690 (17.6)	0.02	16,577 (17.8)	0.02	1,045 (11.6)	0.19	4,929 (16.5)	0.05	428,386 (18.6)
35–39	53,030 (21.0)	0.08	15,261 (16.4)	0.04	847 (9.4)	0.25	5,213 (17.5)	0.01	414,512 (18.0)
40–44	49,224 (19.4)	0.16	14,135 (15.2)	0.05	617 (6.9)	0.22	5,365 (18.0)	0.12	311,544 (13.5)
Neighborhood income quintile
Lowest (Q1–Q2)	103,024 (40.7)	0.01	37,021 (39.7)	0.03	4,248 (47.3)	0.13	13,042 (43.7)	0.05	945,602 (41.0)
Highest (Q3–Q5)	149,659 (59.1)	0.01	55,958 (60.1)	0.03	4,698 (52.3)	0.13	16,752 (56.1)	0.05	1,356,695 (58.8)
Missing	501 (0.2)	0.01	191 (0.2)	0.01	40 (0.4)	0.04	74 (0.2)	0.00	5,525 (0.2)
Material deprivation quintile
Lowest (Q1–Q3)	152,024 (60.0)	0.04	57,588 (61.8)	0.00	4,714 (52.5)	0.19	16,788 (56.2)	0.12	1,429,083 (61.9)
Highest (Q4–Q5)	98,595 (38.9)	0.03	34,907 (37.5)	0.00	4,163 (46.3)	0.18	12,816 (42.9)	0.12	859,797 (37.3)
Missing	2,565 (1.0)	0.02	675 (0.7)	0.01	109 (1.2)	0.04	264 (0.9)	0.01	18,942 (0.8)
Residential instability quintile
Lowest (Q1–Q3)	144,273 (57.0)	0.01	53,498 (57.4)	0.01	4,741 (52.8)	0.08	16,329 (54.7)	0.04	1,308,522 (56.7)
Highest (Q4–Q5)	106,346 (42.0)	0.01	38,997 (41.9)	0.01	4,136 (46.0)	0.07	13,275 (44.4)	0.04	980,358 (42.5)
Missing	2,565 (1.0)	0.02	675 (0.7)	0.01	109 (1.2)	0.04	264 (0.9)	0.01	18,942 (0.8)
Rurality
Rural residence	13,397 (5.3)	0.08	3,641 (3.9)	0.01	435 (4.8)	0.06	1,473 (4.9)	0.06	85,015 (3.7)
Urban	236,606 (93.5)	0.08	88,741 (95.2)	0.00	8,410 (93.6)	0.08	28,067 (94.0)	0.06	2,200,530 (95.4)
Missing	3,181 (1.3)	0.03	788 (0.8)	0.01	141 (1.6)	0.05	328 (1.1)	0.01	22,277 (1.0)
Specific chronic medical conditions
Diabetes	9,922 (3.9)	0.12	4,168 (4.5)	0.14	469 (5.2)	0.18	2,515 (8.4)	0.29	45,294 (2.0)
Hypertension	12,851 (5.1)	0.14	3,929 (4.2)	0.10	252 (2.8)	0.02	2,288 (7.7)	0.24	55,883 (2.4)
Asthma	64,570 (25.5)	0.23	22,485 (24.1)	0.20	2,158 (24.0)	0.20	9,566 (32.0)	0.38	373,208 (16.2)
HIV	234 (0.1)	0.01	58 (0.1)	0.00	16 (0.2)	0.04	37 (0.1)	0.02	1,279 (0.1)
Stable chronic medical condition	68,754 (27.2)	0.22	22,628 (24.3)	0.16	2,249 (25.0)	0.17	10,511 (35.2)	0.40	412,777 (17.9)
Unstable chronic medical condition	48,773 (19.3)	0.22	15,109 (16.2)	0.14	1,418 (15.8)	0.12	8,297 (27.8)	0.42	265,407 (11.5)
Mental illness
Psychotic disorders	4,923 (1.9)	0.08	1,387 (1.5)	0.05	669 (7.4)	0.33	1,157 (3.9)	0.19	21,751 (0.9)
Mood/anxiety disorders	51,738 (20.4)	0.22	16,161 (17.4)	0.14	2,506 (27.9)	0.39	7,199 (24.1)	0.31	285,151 (12.4)
Other mental illness	1,891 (0.8)	0.06	447 (0.5)	0.03	272 (3.0)	0.21	483 (1.6)	0.13	7,297 (0.3)
Substance use disorders	5,558 (2.2)	0.10	1,060 (1.1)	0.02	254 (2.8)	0.14	625 (2.1)	0.09	22,114 (1.0)
Self-harm	2,376 (0.9)	0.06	614 (0.7)	0.03	305 (3.4)	0.21	472 (1.6)	0.11	10,790 (0.5)
History of assault	14,769 (5.8)	0.17	3,123 (3.4)	0.06	651 (7.2)	0.23	2,158 (7.2)	0.23	55,401 (2.4)
Teratogenic medication useb	9,619 (20.0)	0.34	1,853 (12.9)	0.15	739 (18.0)	0.29	4,012 (34.1)	0.67	19,116 (8.3)
Continuity of primary care
<3 primary care visits in 2 years	576,96 (22.8)	0.25	23,041 (24.7)	0.20	2,990 (33.3)	0.02	6,312 (21.1)	0.29	785,077 (34.0)
Low continuity (≤50%)	55,953 (22.1)	0.06	19,930 (21.4)	0.04	1,533 (17.1)	0.07	6,181 (20.7)	0.03	452,788 (19.6)
Moderate continuity (51%–80%)	55,176 (21.8)	0.07	20,324 (21.8)	0.07	1,604 (17.9)	0.03	6,432 (21.5)	0.06	441,692 (19.1)
High continuity (>80%)	84,359 (33.3)	0.13	29,875 (32.1)	0.11	2,859 (31.8)	0.10	10,943 (36.6)	0.20	628,265 (27.2)

Bold values represent standardized differences > 0.10.

Data are presented as n (%) unless otherwise specified.

^aStandardized difference (std diff) compares each disability group with the “no disability” group.

^bThe denominator for teratogenic medication use is women receiving medications under Ontario's publicly funded drug benefits program: physical (48,022), sensory (14,391), intellectual/developmental (4,097), multiple (11,767), and no disabilities (231,823).

Table 2.

Age-Standardized Comparison of Preconception Health Characteristics of Women With and Without Disabilities, Ontario, 2017–2018

Characteristic	Physical disabilities only (N = 253,184)		Sensory disabilities only (N = 93,170)		Intellectual/developmental disabilities only (N = 8,986)		Multiple disabilities (N = 29,868)		No disability (N = 2,307,822)
	%	Std diffa	%	Std diffa	%	Std diffa	%	Std diffa	%
Neighborhood income quintile
Lowest (Q1–Q2)	40.7	0.01	39.6	0.03	49.3	0.17	43.7	0.06	41.0
Highest (Q3–Q5)	59.1	0.01	60.2	0.03	50.4	0.17	56.1	0.06	58.8
Missing	0.2	0.01	0.2	0.01	0.4	0.02	0.3	0.00	0.2
Material deprivation quintile
Lowest (Q1–Q3)	60.0	0.04	61.9	0.00	50.8	0.23	56.2	0.12	61.9
Highest (Q4–Q5)	39.0	0.04	37.4	0.00	48.0	0.22	42.9	0.12	37.3
Missing	1.0	0.02	0.7	0.01	1.1	0.03	0.9	0.01	0.8
Residential instability quintile
Lowest (Q1–Q3)	57.1	0.01	57.6	0.02	50.7	0.12	54.6	0.04	56.7
Highest (Q4–Q5)	41.9	0.01	41.7	0.02	48.2	0.11	44.5	0.04	42.5
Missing	1.0	0.02	0.7	0.01	1.1	0.03	0.9	0.01	0.8
Rurality
Rural residence	5.3	0.08	3.9	0.01	5.2	0.07	5.0	0.06	3.7
Urban	93.4	0.08	95.2	0.01	93.4	0.09	93.9	0.06	95.4
Missing	1.3	0.03	0.9	0.01	1.4	0.04	1.1	0.02	1.0
Specific chronic medical conditions
Diabetes	3.4	0.09	4.4	0.14	7.3	0.26	8.0	0.28	2.0
Hypertension	4.2	0.10	4.1	0.10	4.7	0.12	7.1	0.22	2.4
Asthma	25.9	0.24	24.1	0.20	22.8	0.17	32.2	0.38	16.2
HIV	0.1	0.01	0.1	0.00	0.2	0.04	0.1	0.02	0.1
Stable chronic medical condition	25.4	0.18	24.2	0.15	28.5	0.25	34.6	0.39	17.9
Unstable chronic medical condition	18.5	0.20	16.1	0.14	17.5	0.17	27.4	0.41	11.5
Mental illness
Psychotic disorders	1.9	0.08	1.5	0.05	8.6	0.37	3.8	0.19	0.9
Mood/anxiety disorders	20.3	0.22	17.3	0.14	26.8	0.37	24.0	0.31	12.4
Other mental illness	0.8	0.06	0.5	0.03	2.7	0.20	1.7	0.14	0.3
Substance use disorders	2.2	0.10	1.1	0.02	2.6	0.12	2.1	0.09	1.0
Self-harm	1.0	0.07	0.7	0.03	2.5	0.17	1.6	0.12	0.5
History of assault	5.7	0.17	3.3	0.06	8.0	0.25	7.2	0.23	2.4
Teratogenic medication useb	18.9	0.32	12.7	0.15	19.6	0.33	33.3	0.65	8.3
Continuity of primary care
<3 primary care visits in 2 years	23.4	0.24	24.8	0.20	31.7	0.05	21.6	0.28	34.0
Low continuity (≤50)	22.4	0.07	21.3	0.04	16.7	0.08	20.7	0.03	19.6
Moderate continuity (51–80)	21.8	0.07	21.8	0.07	17.8	0.04	21.4	0.06	19.1
High continuity (>80)	32.4	0.11	32.1	0.11	33.8	0.14	36.3	0.20	27.2

Bold values represent standardized differences > 0.10.

Data presented as % unless otherwise specified.

^aStandardized difference (std diff) compares each disability group with the “no disability” group.

^bThe denominator for teratogenic medication use is women receiving medications under Ontario's publicly funded drug benefits program: physical (48,022), sensory (14,391), intellectual/developmental (4,097), multiple (11,767), and no disabilities (231,823).

In crude analyses, compared with women without disabilities, women with physical, sensory, intellectual/developmental, and multiple disabilities had higher rates of diabetes, asthma, and stable and unstable chronic medical conditions. All groups except women with intellectual/developmental disabilities had higher rates of hypertension. After age standardization, the diabetes rate was no longer higher in women with physical disabilities compared with those without disabilities, whereas the hypertension rate became higher in women with intellectual/developmental disabilities.

Women with physical (20.4%), sensory (17.4%), intellectual/developmental (27.9%), and multiple disabilities (24.1%) had higher rates of mood/anxiety disorders than women without disabilities (12.4%). Compared with women without disabilities, women with intellectual/developmental and multiple disabilities also had higher rates of psychotic disorders, self-harm, and other mental illness, and women with physical, intellectual/developmental, and multiple disabilities had higher rates of substance use disorders. Women with physical (5.8%), intellectual/developmental (7.2%), and multiple disabilities (7.2%) were more likely to have experienced assault compared with women without disabilities (2.4%).

All groups of women with disabilities had elevated use of potentially teratogenic medications in the previous year compared with women without disabilities (8.3%); women with multiple disabilities had the highest rate (34.1%). Table 3 describes the most commonly used potentially teratogenic medications among women with and without disabilities receiving public drug coverage in Ontario, with anti-epileptic agents being the most frequent of all. Women in all disability groups had higher rates of high continuity of primary care compared with women without disabilities. Age standardization did not change these findings substantially.

Appendix Table A2 describes women with multiple disabilities in more detail. Women with physical and sensory disabilities and those with all three types of disabilities had the highest rates of diabetes, hypertension, asthma, and unstable chronic conditions, whereas women with physical and intellectual/developmental disabilities had the highest rates of psychotic and mood/anxiety disorders, as well as the highest rates of substance use disorders and self-harm. Women with physical and intellectual/developmental disabilities and those with all three types of disabilities had the highest rates of potentially teratogenic medication use and history of assault. Women with physical and intellectual/developmental disabilities and those with all three types of disabilities had the highest rates of high continuity of primary care.

Discussion

This is the first Canadian study and the most comprehensive population-based study to date to describe the preconception health characteristics of women with physical, sensory, and intellectual/developmental disabilities. Across all indicators, women with disabilities had poorer preconception health compared with women without disabilities. Disparities were particularly pronounced for chronic medical conditions, mental illness, history of assault, and use of potentially teratogenic medications. Women with intellectual/developmental disabilities were the most socioeconomically marginalized group and had the greatest health disparities compared with women without disabilities, even after age standardization.

These findings are mostly consistent with limited research on the preconception health of US women with disabilities. Using the 2010 Behavioral Risk Factor Surveillance System (BRFSS) of nonpregnant women 18–44 years of age in all 50 states, the District of Columbia, and 3 US territories, Mitra et al.¹² found women with self-reported disabilities based on activity limitations and assistive device use had higher rates of mental distress and chronic medical conditions such as asthma, diabetes, and obesity. Using the 2003–2009 Washington State BRFSS, which used a similar measure of disability, Kim et al.¹¹ found 18- to 59-year-old women with disabilities who were living with children <18 years of age reported lower income and education levels, higher rates of mental distress, and an elevated prevalence of chronic medical conditions compared with those without disabilities. Women with disabilities in both studies also reported higher prevalence of adverse health behaviors, such as smoking and lack of exercise, and they were more likely to report their health status as “poor.” However, these studies are limited because of their reliance on self-report measures of disability and preconception health.

The use of administrative data for this study is its strength, eliminating risk of recall or reporting bias or burden to survey respondents. Our study is also novel because of its inclusion of a broad range of preconception health characteristics, including measures of self-harm, history of assault, and use of potentially teratogenic medications in addition to physical health status. To our knowledge, previous studies have not used this approach.

There are several possible explanations for why women with disabilities experience these disparities. They are more likely than women without disabilities to experience sexual and reproductive health education barriers.^9,33–35 Although preconception interventions have been developed for women with chronic medical conditions to address disease management, medication use, and family planning,⁴ none, to our knowledge, target women with disabilities.³⁶ Moreover, previous interventions related to reproductive health in women with disabilities have focused on contraceptive knowledge and use.³⁷ Narrowly focused and inaccessible sexual and reproductive health programs and information may contribute to poor preconception health among women with disabilities. Furthermore, providers may be unaware of or misunderstand important considerations in the preconception care of women with disabilities.

The exclusion of women with disabilities from preconception interventions may reflect their poorer access to preventive health care more broadly. Research from the United States indicates women with disabilities are less likely than their peers to receive preventive care,^9–11 including cancer screening and family planning.^38,39 Although we found women with disabilities had good continuity of primary care, our data did not allow us to elucidate the quality of their care or the amount of time devoted to preconception health care.

Limitations

Women with disabilities who did not have a recorded diagnosis in health administrative data were not deemed to have disabilities. This error may lead to some misclassification bias that could result in underestimation of disparities between women with and without disabilities. Although we excluded women with hysterectomy to obtain a cohort that could potentially become pregnant, we did not make exclusions based on infertility-related diagnoses; a small percentage of women in our cohort may be unable to become pregnant.

We could not determine the reproductive history of women without previous delivery hospitalization in Ontario, so this component of preconception health is not reported. We also had no data on health behaviors (e.g., smoking) or BMI. We only had area-level data on socioeconomic status, and data on other sociodemographic factors, such as race/ethnicity, are not collected in Ontario health records. Our data underestimate the true rate of assault because only assaults reported in hospital encounters were captured. For various reasons (e.g., financial dependence and previous poor interactions with clinicians), women with disabilities often do not report assault and some may not understand their experiences as assault.^40,41 Medication use data were only available for women with publicly funded drug benefits (∼11.5%); given the eligibility criteria for this program, these women may be more marginalized than the broader population.

Because the purpose of our study was to describe absolute differences between women with and without disabilities, we did not control for covariates beyond age. A potential area for future research is the examination of whether such differences are explained by other factors, including social disparities. Moreover, qualitative approaches may be useful for understanding the preconception health of women with disabilities, including the unique preconception heath needs and experiences of specific subgroups of women, the quality of care they receive, and factors contributing to disparities not available in administrative data (e.g., stigma).

Implications

Our findings, along with those of previous studies,^10–12,42 suggest more could be done to optimize the preconception health of women with disabilities. Supporting preconception health in this population is vital, as more women with disabilities are experiencing pregnancy,^6–8 and research following a life course perspective demonstrates that preconception health impacts perinatal outcomes.¹ Although preconception care programs for women with chronic medical conditions such as diabetes⁴ may be relevant to women with disabilities given their high rates of these and other conditions, many individual and system-level factors unique to women with disabilities warrant development of tailored preconception health programs and resources for this population. To improve the preconception health of women with disabilities, there is a need for multiple targets and levels of intervention within and beyond the health care system.

Based on our findings, potential targets for preconception interventions for women with disabilities should include management of chronic medical conditions, strategies to support mental health and address substance use, information about changes to medications if a woman decides to pursue pregnancy, and discussions about violence against women. Two of these findings are especially worth highlighting. The high rate of potentially teratogenic medication use among women with disabilities is important, as many disabilities or secondary conditions require the use of medication.^28,29 However, if taken during pregnancy, such medications may increase the risk of congenital anomalies. Women with disabilities, and particularly those with intellectual/developmental disabilities who often lack understanding of the purpose and adverse effects of their medications,⁴³ may benefit from more accessible information about medications and their consequences for pregnancy.²⁹

Other important topics include strategies to prevent violence against women, given the high rate of prior assault observed herein. Such strategies should consider the ways in which women with disabilities are vulnerable to violence and the unique challenges they face fleeing violence (e.g., inaccessible shelters and financial dependence on partners). Of note, although it is paramount that these issues be addressed before pregnancy, studies show that women with disabilities are more likely than their peers to experience adolescent pregnancy⁸ and unintended pregnancy.⁹ Preconception interventions, in the form of sexual and reproductive health education, should start early and include discussions about contraception, reproductive life plans (including the psychological and social impact of pregnancy), consent, and violence.^12,44

Individual-level interventions should explicitly address needs related to disability. For example, interventions should include consideration of how limitations to body function or mobility and secondary conditions (e.g., chronic pain and fatigue) may make it challenging for women with disabilities to modify health behaviors (e.g., exercise).¹¹ For women with physical disabilities, health care settings should be physically accessible. Appropriate communication supports should be provided for women who are blind or have vision loss (e.g., information in nonwritten formats) and those who are deaf or have hearing loss (e.g., ASL interpreters). Because many women with intellectual/developmental disabilities experience challenges in communication, literacy, memory, and organizational skills (including difficulty understanding and following medical advice^34,45–47), interventions for this group should be tailored to address these learning needs through the use of visual aids.^34,48

All disability groups may benefit from interventions that include or engage caregivers, although the individual woman should be at the center of communication, planning, and decisions regarding care.⁴⁹ Of note, however, although it is recommended that preconception interventions target both partners,^50,51 it may be advisable to also develop approaches specific to nonpartnered women or those with fewer social supports given that women with disabilities are less likely to be partnered,^11,12,52 and more likely to experience abuse.^40,53,54

At a system level, it is important to acknowledge that women with disabilities experience significant socioeconomic obstacles that may make it challenging for them to benefit from individual-level interventions that target modification of health behaviors (e.g., exercise and healthy eating) and other preconception health topics.¹¹ In describing their perinatal health framework for women with physical disabilities, Mitra et al.⁵⁵ argued that “efforts to improve the health and health care of women with disabilities must be embedded within a strategy of improving the economic and social opportunities available to women and men with disabilities overall” (p. 504). Interventions beyond health care, including policies and services related to housing, education, and employment, are needed to create a social context in which the individual preconception health disparities experienced by women with disabilities can be addressed.

In the context of care delivery, such adaptations may require longer and more frequent visits,^34,48,49 as well as collaboration among public health, primary care, obstetrics and gynecology, psychiatry, and other allied health professionals such as occupational and physical therapists, dieticians, and social workers.^29,44 Attending to the potentially complex preconception health needs of women with disabilities may therefore not only take more time but have implications for who is reimbursed to provide care and what is reimbursed,⁵⁶ which is difficult to change.

Arguably, women with disabilities in Canada, where there is universal health care, should have easier access to preconception care, compared with women with disabilities in the United States, who face insurance-related barriers including limits to the number of visits to a health care provider.¹¹ Yet, we found that preconception health disparities exist for women with disabilities even within a universal health care system. Our findings suggest that elimination of financial barriers to health services for women with disabilities, while essential, is not enough to address preconception health disparities. Efforts to improve preconception health for women with disabilities must therefore address various individual and system-level barriers within and outside the health care system.

In developing and implementing preconception interventions for women with disabilities, health care providers may benefit from accessing existing resources to facilitate preventive services receipt among women with disabilities,⁵⁷ such as the Women Be Healthy 2 curriculum for women with intellectual/developmental disabilities,⁵⁸ the Primary Care of Adults with Intellectual and Developmental Disabilities: 2018 Canadian Consensus Guidelines,⁴⁹ and the American College of Obstetrician and Gynecologists' Committee Opinion on Intimate Partner Violence (which includes screening questions specific to women with disabilities).⁵⁹

Educating health care providers about the unique needs and preconception health profiles of women with disabilities is imperative,¹² and may help eliminate other barriers to preconception health, such as negative attitudes toward people with disabilities or discomfort discussing sexual and reproductive health with women with disabilities.^55,58 Increasing the disability competency of health care providers through better training and inclusion of disability in public health drivers (e.g., Healthy People 2030 and Core Competencies for Public Health in Canada) may also help to address programmatic and system-level barriers to preconception care, such as inaccessible care settings.

Conclusions

Women with disabilities experience poorer preconception health than women without disabilities. Preconception health disparities are particularly pronounced for women with intellectual/developmental disabilities. Most preconception interventions emphasize individual-level factors and behavior change. However, such interventions may not be effective if women with disabilities continue to be marginalized in health care and other settings. To mitigate disparities, in addition to educating heath care providers about the unique needs of women with disabilities, preconception interventions for women with disabilities must aim to address the various ways in which they are marginalized in health care and other systems.

Disclaimers

This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by ICES or the MOHLTC is intended or should be inferred. Parts of this material are based on data and information compiled and provided by the Canadian Institute for Health Information (CIHI). However, the analyses, conclusions, opinions, and statements expressed herein are those of the authors, and not necessarily those of CIHI.

Previous Presentation

This article was presented at the Canadian Public Health Association Conference: “Public Health 2019” in Ottawa, Ontario, Canada, April 30 to May 2, 2019.

Appendix

Appendix Table A1.

Databases Accessed Through ICES to Create the Study Cohort and Variables

Database	Available information
Canadian Institute for Health Information Discharge Abstract Database	Diagnoses from discharges from acute care hospitals recorded using the Canadian Coding Standards for the International Classification of Diseases and Related Health Problems
Client Agency Program Enrolment Database	Information on enrolment in primary care programs
ICES Physician Database	Information on physician type
Ontario Drug Benefits Database	Prescriptions dispensed to women receiving Ontario Drug Benefits because of a disability, low income, or high medication costs
Ontario Health Insurance Plan (OHIP) Database	Physician billing claims for outpatient visits
Ontario Mental Health Reporting System	Diagnoses from discharges from hospitals with designated mental health beds recorded using the Diagnostic and Statistical Manual of Mental Disorders
National Ambulatory Care Reporting System	Diagnoses from emergency department visits
Registered Persons Database	Age, sex, postal code (linked to 2016 Census information to determine neighborhood income, residential instability, material deprivation, and rural residence)

Appendix Table A2.

Preconception Health Characteristics of Women with Multiple Disabilities, by Disability Subtype, Ontario, 2017–2018

Characteristic	Physical and sensory (N = 21,887), n (%)	Physical and intellectual/developmental (N = 4,546), n (%)	Sensory and intellectual/developmental (N = 1,814), n (%)	Physical, sensory, and intellectual/developmental (N = 1,621), n (%)
Age, years, mean (SD)	31.21 ± 8.19	27.56 ± 8.10	25.59 ± 7.33	26.52 ± 7.82
15–19	2,254 (10.3)	893 (19.6)	418 (23.0)	358 (22.1)
20–24	3,154 (14.4)	1,014 (22.3)	500 (27.6)	388 (23.9)
25–29	3,807 (17.4)	820 (18.0)	402 (22.2)	353 (21.8)
30–34	3,742 (17.1)	733 (16.1)	235 (13.0)	219 (13.5)
35–39	4,295 (19.6)	604 (13.3)	146 (8.1)	168 (10.4)
40–44	4,635 (21.2)	482 (10.6)	113 (6.2)	135 (8.3)
Neighborhood income quintile
Lowest (Q1–Q2)	9,447 (43.2)	2,120 (46.6)	783 (43.2)	692 (42.7)
Highest (Q3–Q5)	1,2403 (56.7)	2,404 (52.9)	1,022 (56.3)	923 (56.9)
Missing	37 (0.2)	22 (0.5)	9 (0.5)	6 (0.4)
Material deprivation quintile
Lowest (Q1–Q3)	12,430 (56.8)	2,423 (53.3)	1,034 (57.0)	901 (55.6)
Highest (Q4–Q5)	9,278 (42.4)	2,072 (45.6)	759 (41.8)	707 (43.6)
Missing	179 (0.8)	51 (1.1)	21 (1.2)	13 (0.8)
Residential instability quintile
Lowest (Q1–Q3)	11,999 (54.8)	2,422 (53.3)	995 (54.9)	913 (56.3)
Highest (Q4–Q5)	9,709 (44.4)	2,073 (45.6)	798 (44.0)	695 (42.9)
Missing	179 (0.8)	51 (1.1)	21 (1.2)	13 (0.8)
Rurality
Rural residence	1,030 (4.7)	262 (5.7)	95 (5.2)	86 (5.3)
Urban	20,637 (94.3)	4,219 (92.8)	1,696 (93.5)	1,515 (93.5)
Missing	220 (1.0)	65 (1.4)	23 (1.3)	20 (1.2)
Specific chronic medical conditions
Diabetes	1,933 (8.8)	329 (7.2)	104 (5.7)	149 (9.2)
Hypertension	1,901 (8.7)	217 (4.8)	63 (3.5)	107 (6.6)
Asthma	7,070 (32.3)	1,331 (29.3)	565 (31.2)	600 (37.0)
HIV	24 (0.1)	< 6	< 6	< 6
Stable chronic medical condition	7,790 (35.6)	1,430 (31.5)	659 (36.3)	632 (39.0)
Unstable chronic medical condition	6,266 (28.6)	1,124 (24.7)	412 (22.7)	495 (30.5)
Mental illness
Psychotic disorders	587 (2.7)	352 (7.7)	117 (6.5)	101 (6.2)
Mood/anxiety disorders	5,309 (24.3)	1,121 (24.7)	425 (23.4)	344 (21.2)
Other mental illness	241 (1.1)	152 (3.3)	34 (1.9)	56 (3.5)
Substance use disorders	446 (2.0)	130 (2.9)	23 (1.3)	26 (1.6)
Self-harm	220 (1.0)	157 (3.5)	45 (2.5)	50 (3.1)
History of assault	1,462 (6.7)	435 (9.6)	118 (6.5)	143 (8.8)
Teratogenic medication usea	1,824 (28.6)	1,477 (47.6)	181 (16.9)	530 (43.7)
Continuity of primary care
<3 primary care visits in 2 years	3,997 (18.3)	1,304 (28.7)	567 (31.3)	444 (27.4)
Low continuity (≤50%)	4,830 (22.1)	770 (16.9)	320 (17.6)	261 (16.1)
Moderate continuity (51%–80%)	5,035 (23.0)	781 (17.2)	315 (17.4)	301 (18.6)
High continuity (>80%)	8,025 (36.7)	1,691 (37.2)	612 (33.7)	615 (37.9)

Values < 6 suppressed to preserve patient confidentiality.

Data are presented as n (%) unless otherwise specified.

^aThe denominator for teratogenic medication use is women receiving medications under Ontario's publicly funded drug benefits program, physical and sensory (6,381), physical and intellectual/developmental (3,105), sensory and intellectual/developmental (1,069), and all three disabilities (1,212).

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health (NIH) (Award #5R01HD092326). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funding source had no role in the study design, data collection, data analysis, data interpretation, report writing, or decision to submit the article for publication.