Health Information Seeking in the Digital Age: A National Survey of Women with Disabilities

Abstract

Purpose:

Access to high quality and accessible online health information (OHI) is critical for reducing disparities, overcoming barriers, and improving the health of women with disabilities. This study aimed to understand women with physical disabilities’ use of the Internet to access OHI, most often searched health topics, perceived usefulness of OHI, and self-reported eHealth literacy and challenges in OHI seeking.

Methods:

We conducted a national online survey with 508 women with physical disabilities who used the Internet.

Results:

Respondents utilized a wide variety of OHI resources. They searched a broad array of health and disability-related topics, with bowel/bladder and finding a physician the most highly searched topics. They generally had confidence in their eHealth literacy skills and ability to understand statistics in OHI. Nevertheless, although our sample consisted of a majority of highly educated internet-users, a sizeable percentage found OHI seeking difficult and frustrating, did not find the information very helpful, and had concerns about the quality of information.

Conclusions:

This study serves as a call to action to disability and rehabilitation scientists, health care providers, and other health professionals to enhance the availability and accessibility of OHI critical to empowering women with physical disabilities to make well-informed health decisions.

Introduction

An estimated 36 million women in the United States – or one in four adult women – live with a disability [1]. Because disability increases with age, the number of women with disabilities will increase as the population ages [2,3]. Women more commonly experience disability than men [4]. The most common types of disabling conditions are those that affect mobility (i.e., ambulation difficulty or physical disability), and these conditions are more prevalent in women compared to men [2].

Research shows serious health disparities between women with and without disabilities in critical areas such as preventive screening [5,6], reproductive health [7,8], and cancer [9]. Access to health information is critical for reducing disparities, overcoming barriers, and improving the health and rehabilitation outcomes of women with disability. Women with disabilities encounter multiple barriers to obtaining health information and healthcare due, in part, to a lack of providers with disability-sensitive knowledge and limited access to technology and facilities [10,11].

The Health Information Technology objective of Healthy People 2030 [12] highlights the importance of access to electronic health information for tracking and managing one’s health care. Moreover, the National Council on Disability [13] frames access to information and communication in terms of civil rights. Despite significant increases in online health information (OHI) seeking, the digital divide by disability status remains a vast and persistent reality [14]. People with disabilities face significant infrastructure and access barriers [15]. They are less likely than people without disabilities to use the Internet, to have broadband at home, and to own a desktop or laptop computer [16].

The COVID-19 pandemic has further highlighted significant inequalities in digital accessibility among people with disabilities [15], exposing prominent gaps in their access to electronic and telecommunications technologies [17]. The social restrictions of the pandemic required people with disabilities to access telemedicine in lieu of traditional in-person healthcare delivery [15]. People with disabilities may have experienced benefits such as reduced exposure to COVID-19 and transportation costs [15]. However, their access to telemedicine may have been limited due to the lack of internet services, personal computers, or other telecommunication technologies [17,18]. Other factors potentially reducing their access include limited technology and computer skills [18–20] and inaccessible digital platforms [17]. COVID-19 may also have exacerbated telemedicine access problems for people with sensory, speech, manual dexterity, and cognitive limitations who may experience significant challenges communicating remotely with their providers [18–20].

There is a growing literature base regarding OHI seeking among women in the general population [e.g., 21,22], and a small emerging literature addressing OHI seeking among people with disabilities although without reporting on gender differences [23,24]. Although one study addressed OHI seeking among older women with potentially disabling chronic illnesses [25], no known research has investigated OHI seeking among a large, national sample of women with cross-cutting physical disabilities.

We elected to use an accessible, online survey in order to reach women with disabilities who by virtue of accessing and enrolling in the survey study online also had access to the Internet. They could thereby report on health-related topics searched online, perceived usefulness of health information found, and self-report on their eHealth literacy and challenges to OHI seeking. Clearly, access to online health-related resources can benefit women with disabilities on multiple levels. A recent qualitative study [26] revealed that women with physical disabilities preferred using Internet for obtaining general health as well as sexual and reproductive health information. We are aware of only one other survey of women with physical disabilities (N=63) that was limited to online social support during pregnancy and was not designed to evaluate broad-based health information seeking [27]. Thus, there is a need to examine online health information seeking among women with disabilities.

The Internet can serve as a significant catalyst for improving access to health information and ultimately improving health and rehabilitation outcomes for women with disabilities. Although there is growing research attention to OHI seeking by people with disabilities [23,24], no such research is known to exist specific to women with disabilities. To fill this gap, we conducted the National Health Information Survey for Women with Physical Disabilities (N=508). Drawing on data collected for that survey, our goal in the current study is to examine women with physical disabilities’ 1) use of the Internet to access OHI, 2) most often searched health topics, 3) perceived usefulness of OHI, and 4) self-reported eHealth literacy and challenges in health information seeking.

Methods

To address these questions, we conducted an online survey of women with physical disabilities in January and February 2020. Before releasing our survey, we conducted in-depth focus group interviews with 21 women with physical disabilities who identified gaps and challenges to health information seeking by women with disabilities [26]. We drew upon information shared by the focus group participants when developing our survey. Adhering to inclusive Community-based Participatory Research principles [28], we partnered with a nine-member Community Advisory Board (CAB) of women with physical disabilities who collaborated in all phases of the study and ensured the survey protocol was understandable and had direct relevance to the population. The researchers convened and facilitated the monthly CAB meetings via videoconference. Our CAB members also served as beta-testers of the survey created in Qualtrics, a HIPAA-compliant web-based survey tool.

Using previously successful recruitment strategies, we worked closely with five centers for independent living (CILs) as well as other organizations that serve women with disabilities. We selected the CILS for their capacity to help ensure a diverse sample in terms of geography and racial and ethnic minority representation. We worked with our extensive network to distribute recruitment information via existing listservs and distribution lists, and we posted recruitment flyers to multiple disability-related Facebook groups. Our established network includes the National Council of Independent Living, the Association of Programs for Rural Independent Living, and the Independent Living Resource Utilization. Our CAB members, who are active in their disability communities, also supported our efforts by recruiting through word-of-mouth and social media.

Recruitment materials provided a link to study information and a brief eligibility screen. Respondents who advanced through the eligibility questions were provided informed consent information followed by a series of comprehension of consent questions. The four comprehension of consent questions assessed respondents’ understanding of the study and what participation entailed. Only after correctly answering three of the four questions and confirming their consent were they directed to the online survey.

To be eligible, respondents had to identify as a woman; be at least 21 years of age; self-identify as having a mobility impairment due to a chronic, physically disabling condition; have had their disabling condition for at least one year; and be able to understand English well enough to answer survey questions in the English language. Respondents also had to have access to a computer or mobile device with an internet connection as determined by their ability to access the study link to complete the screening and enrollment process. Women who completed the survey were given the option of providing their name and contact information to be entered in a drawing for a gift card. Respondents’ names and contact information were not linked to their survey data.

Measures.

Survey questions were designed to address the primary aims of the study, namely, to examine women with physical disabilities’ 1) use of the Internet to access OHI, 2) most often searched health topics, 3) perceived usefulness of OHI, and 4) self-reported eHealth literacy and challenges in health information seeking.

Use of the Internet to Access OHI.

To explore women with disabilities’ use of the Internet to access OHI, we examined several aspects of their internet use. We first asked respondents how often they had accessed health information on the Internet in the past 12 months (3-point response scale: never, less than once a month, once a month or more).

Devices used to access the Internet and assistive technologies required were also assessed. Respondents were asked to indicate which of the following devices they used to get on the Internet: home desktop, laptop or tablet; computer at work or school; public computer (e.g., library); mobile phone; gaming device; or smart TV. Respondents were asked to check all that applied. Similarly, respondents were provided a list of assistive technologies (e.g., mouth stick, voice recognition program) and asked to select all devices they used to help them get on the Internet.

How the Internet is used for health-related purposes was assessed using items modeled after the Health Information National Trends Survey (HINTS) [29], a national survey assessing the public’s knowledge of, attitudes toward, and use of cancer and health-related information. Respondents were asked if (yes or no) they had used the Internet in the past 12 months for a variety of health-related activities (e.g., participating in online support groups, looking for a health care provider, watching a health-related video, tracking personal health information). Finally, respondents were also asked to indicate how often (5-point response; always to never) they “use information found on the Internet to help them communicate better with their doctor or health provider.”

OHI Topics Searched.

To assess the health-related topics most often searched by study respondents, we included a series of questions asking (yes or no) “Have you ever looked for information on the Internet about how your disability affects ________?” The list of 13 conditions included in the survey was broad and included menstruation; fertility; contraception; family planning; pregnancy, labor, and delivery; bladder and bowel problems; pelvic floor problems; menopause; cancers specific to women; cancer screenings; well-woman exams; finding a physician; and insurance coverage related to women’s health. Note that COVID-19 was not included in the list as data were collected in January and February, 2020, which was prior to the World Health Organization’s declaration of the COVID-19 outbreak as a pandemic in March, 2020 [30]. The list of health conditions was generated by the research team in partnership with our community advisors.

Perceived Usefulness of OHI.

For each of the health information topics listed above, respondents who endorsed looking for such information were asked how helpful the information was that they found using a 3-point response scale (not at all, a little, or very helpful). They also were asked to rate the extent to which the Internet was useful in helping make decisions about their health using a 5-point response scale ranging from “not useful at all” to “very useful.” Finally, they rated how important it was to them to be able to access health resources on the Internet using a 5-point response scale ranging from “not at all” to “very important.”

Self-reported eHealth Literacy and Challenges in Health Information Seeking.

Electronic health literacy was measured with the 8-item, self-report eHealth Literacy Scale (eHEALS) [31]. Respondents rated items assessing knowledge, comfort, and perceived skills at finding, evaluating, and applying e-health information to health problems using a 5-point response scale (strongly disagree to strongly agree). The items are summed with higher scores reflecting greater literacy. In our sample, the Cronbach’s alpha of the 8-item eHEALS was .87, which indicates satisfactory internal consistency reliability [32].

Challenges encountered in seeking health information were assessed using four items from the Information Seeking Experience Scale (ISEE) [33], with two items assessing the process of information seeking (i.e., effort, frustration) and two items assessing the information found (i.e., quality, difficulty in understanding). Considering their most recent health information search, respondents rated the items using a 4-point response scale (strongly agree to strongly disagree) with higher scores reflecting more negative information-seeking experiences. The Cronbach’s alpha of the 4-item ISEE in this sample was .78, which is regarded as satisfactory internal consistency reliability [32].

Challenges in understanding statistics in health information were assessed using two items patterned after the STAT-Confidence Scale [34]. The items included one question, “In general, how easy or hard do you find it to understand medical statistics?” and one statement, “In general, I feel uncomfortable with health information that has a lot of numbers and statistics.” Both items used a 4-point response scale ranging from “very easy” to “very hard” and “strongly agree” to “strongly disagree,” respectively.

Finally, confidence in health information seeking was assessed using a single item modeled after an item from the HINTS [29] that asked, “Overall, how confident, or sure, are you that you could get information about your health if you needed it?” Respondents rated their health information seeking confidence using a 5-point response scale ranging from “completely confident” to “not at all confident.”

Demographic and Disability Characteristics.

Demographic and disability characteristics were assessed through a series of questions placed at the end of the survey. Demographic questions included respondents’ age, educational level, race/ethnicity, relationship status, sexual orientation, household income, and employment status. In addition, respondents indicated if they lived in an urban (i.e., city, large town, or suburb just outside a city or large town) or rural (i.e., small town or city with a population of less than 50,000; in the country or a long way from town; on a reservation or tribal lands) community. Primary disability was determined by asking respondents to identify from a list of physically disabling conditions (e.g., amputation, cerebral palsy, fibromyalgia, multiple sclerosis, “other”) the cause of their mobility limitation. If they reported more than one disability, they were asked which disability/condition was most limiting. To determine disability duration, respondents were asked how many years they had been living with their most limiting disability. Disability severity was measured by 1) self-reported regular use of assistive devices, and 2) the 10-item Physical Function subscale of the Medical Outcomes Survey-Short Form (SF-36) [35]. A modified version of the Physical Function scale was used with respondents rating the extent to which their disability, rather than their health, limited their ability to engage in 10 activities (e.g., walking one block). Respondents rated each item on a 3-point scale (a lot, a little, or not at all), and scores were transformed to a 0 to 100 scale, with higher scores reflecting greater physical function. The Cronbach’s alpha for the scale in this sample was .88, which is regarded as satisfactory internal consistency reliability [32]. Finally, respondents were asked to rate their health using a 5-point response scale (excellent to poor).

Planned Analyses.

Study objectives required descriptive analyses consisting primarily of standard measures of central tendency and variability: frequencies, range, mode, median, mean, and standard deviation. Given the widespread distribution of recruitment materials and the lack of a guaranteed financial incentive for completing the survey (a random drawing for a gift card), we anticipated that many individuals would open the link and begin the survey without completing it. Recognizing the study results can only be appreciated with an understanding of the sample, we decided to limit our planned analyses to respondents who advanced to the end of the survey and answered demographic and disability questions. Analyses were conducted using SAS statistical software (SAS System for Windows, version 9.4; SAS Institute Inc., Cary, NC).

Results

Sample.

A total of 908 individuals accessed the study link and answered the initial screening question, with 842 confirming eligibility on all five criteria. A total of 652 indicated interest in completing the survey and answered the comprehension of consent (CoC) questions. Of these, 617 correctly answered at least three of the four CoC questions and began the survey. In addition to reviewing consent information and answering CoC questions, respondents were informed that opening the survey and answering the questions served to document their consent to participate. The number of respondents decreased with each successive item, resulting in a sample of 508 women who advanced to the end of the survey and completed demographic and disability questions.

The demographic and disability characteristics of the sample are shown in Table 1. [Table 1 near here] The sample was predominantly White, non-Hispanic, and highly educated (71% had college degrees) but was diverse in most other demographic characteristics. The sample was particularly diverse regarding sexual orientation, with over 25% of women who identified as members of sexual minority groups. A wide range of mobility-limiting conditions were represented in the sample with more than one-third of the women (36%) reporting more than one disabling condition. Primary disabling conditions endorsed by at least 1% of the sample are listed in Table 1. The vast majority (94%) reported using assistive devices, with 69% regularly using a manual or power wheelchair. The sample had significantly limited health and function as reflected by low scores on the SF-36 Physical Function Scale (Mean = 18.31, SD = 19.08; range 0-100) and relatively poor self-rated health (39% rated as fair or poor).

Table 1.

Sample characteristics (N=508)

Variable	Mean±SD or N (%)
Age (y)	42.7±14.0
Race
White	435 (85.6)
Black	20 (3.9)
Asian	9 (1.8)
American Indian/Alaskan Native	5 (1.0)
Native Hawaiian/Pacific Islander	3 (0.6)
Multiracial	21 (4.1)
Do not wish to say/Missing	15 (3.0)
Ethnicity
Not Hispanic	463 (91.1)
Hispanic	33 (6.5)
Do not wish to say/Missing	12 (2.4)
Level of Education
Less than high school	7 (1.4)
High school grad, GED	41 (8.1)
High School plus some college/Technical school	100 (19.7)
College degree	227 (44.8)
Master’s/Doctoral degree	132 (26.0)
Missing	1 (0.2)
Employment Status
Full-time	138 (27.2)
Part-time	99 (19.5)
Unemployed	271 (53.3)
Household Income
<$10,000	75 (14.8)
$10,000 – $19,999	77 (15.2)
$20,000 – $49,999	118 (23.2)
$50,000 – $99,999	92 (18.1)
$100,000 +	76 (15.0)
Don’t know/Don’t want to say/Missing	70 (13.8)
Community Environment
City or large town	223 (43.9)
Suburb or just outside a city or large town	160 (31.5)
Small town	88 (17.3)
The country or a long way from town	34 (6.7)
Missing	3 (0.6)
Relationship Status
Partnered	285 (56.1)
Not Partnered	223 (43.9)
Sexual Orientation
Heterosexual/Straight	369 (72.6)
Gay/Lesbian	31 (6.1)
Bisexual	68 (13.4)
Other	20 (3.9)
Do not wish to say/Missing	20 (3.9)
Most Limiting Condition
Joint and connective tissue disease	111 (21.9)
Spinal cord injury	91 (17.9)
Neuromuscular disorder	73 (14.4)
Spina bifida	61 (12.0)
Cerebral palsy	52 (10.2)
Back problem	21 (4.1)
Muscular sclerosis	17 (3.3)
Post-polio	10 (2.0)
Amputation	8 (1.6)
Brain injury	7 (1.4)
Complex regional pain syndrome	7 (1.4)
Other	50 (9.8)
Years Living with Disability*	26.9±16.7
Age of Primary Disability Onset*	15.8±18.2
Assistive Devices Regularly Used*
Power wheelchair	204 (40.2)
Manual wheelchair	214 (42.2)
Scooter (3 or 4 wheel mobility device)	52 (10.3)
Walker	98 (19.3)
Cane	116 (22.9)
Crutches	59 (11.6)
Brace or other orthotic device	137 (27.0)
Artificial limb or other prosthetic device	9 (1.8)
Ventilator	18 (3.6)
SF-36 Physical Function Scale	18.3±19.1
Self-rated health
Excellent	19 (3.7)
Very good	112 (22.0)
Good	181 (35.6)
Fair	141 (27.8)
Poor	55 (10.8)

^*Data was missing on one respondent; thus, N=507.

Use of the Internet to Access OHI.

All but two respondents indicated that they had accessed health information on the Internet in the past year, with 82% stating that they did so at least once a month. Respondents accessed the Internet using a variety of devices with 92% accessing the Internet using a desktop, laptop, or tablet computer from home, and 91% doing so using a mobile phone. More than one-quarter (28%) had access to the Internet via a computer at school or work. Although less common, some accessed the Internet via a smart TV (11%), gaming device (6%), and a public library or community center (4%). About half of the sample (55%) did not use any assistive technology to access the Internet, but the remainder used a wide range of assistive devices or aids. As shown in Table 2, the most frequently reported technology utilized were voice recognition programs, used by 16% of respondents, and closed captioning, used by 13% of respondents.

Table 2.

Number and Percentage of Sample Using Assistive Technology to Access the Internet, Engaging in Health-Related Internet Activities, and Engaging in Health Information Searches by Topic

Variable	N (%)
Assistive Technology / Adaptations to Use Internet (N=507)
Voice recognition program	83 (16.4)
Mouth stick, head pointer, stylus	22 (4.3)
Alternative mouse (track ball, joystick)	48 (9.5)
Ergonomic keyboard	66 (13.0)
Screen reader	22 (4.3)
Closed captioning	67 (13.2)
Magnifier	37 (7.3)
Eye tracking device	3 (0.6)
Personal Assistant	32 (6.3)
Onscreen keyboard	65 (12.8)
Other	17 (3.4)
No adaptations required	278 (54.8)
Health-Related Internet Activities in the Past 12 Months (N=508)
Attended online support group	340 (66.9)
Shared health information on social media sites	351 (69.1)
Looked for a health care provider	357 (70.3)
Downloaded your health information	243 (47.8)
Tracked personal health information	408 (80.3)
Watched health-related video	359 (70.7)
Listened to health-related podcast	153 (30.1)
Purchased medicine, vitamins, or supplements	311 (61.2)
Topics Searched (N=508)
Menstruation	192 (37.8)
Fertility	138 (27.2)
Contraception	150 (29.5)
Pre-pregnancy, family planning	94 (18.5)
Pregnancy, labor, and delivery	149 (29.3)
Bladder/Bowel problems	356 (70.1)
Pelvic floor problems	147 (28.9)
Menopause	137 (27.0)
Cancers specific to women	104 (20.5)
Cancer screening (e.g., mammograms)	121 (23.8)
Well-woman exams	125 (24.6)
Finding a physician	291 (57.3)
Insurance coverage	173 (34.1)

As shown in Table 2, when asked how the Internet was used for health-related purposes over the past 12 months, respondents reported using the Internet to support their health and health-information needs in a wide variety of ways. Two-thirds (67%) had used the Internet to participate in an online support group for people with a similar health or medical condition in the past 12 months. Other activities reported by a majority of respondents were tracking personal health information, watching a health-related video, looking for a health care provider, sharing health information on social media sites, and purchasing medicine, vitamins, or supplements online. On the additional item asking how often they use information on the Internet to help them communicate better with their doctor, the vast majority of respondents (80%) reported using the information to better communicate with their doctor as least sometimes (always, 11%; most of the time, 23%, sometimes, 46%; rarely, 14%; never, 6%).

OHI Topics Searched.

Respondents were asked if they had ever looked for information on the Internet on 13 health-related topics. As shown in Table 2, the two most highly endorsed topics, searched by over 50% of respondents, were bladder and bowel problems (70%) and finding a physician (57%). Topics searched by at least one-third of respondents included information about insurance coverage and menstruation. Most of the remaining topics (fertility; contraception; pregnancy, labor, and delivery; pelvic floor problems; menopause; and well woman exams) had been searched by at least one-quarter of all respondents.

Perceived Usefulness of OHI.

There was variability in respondents’ ratings of the helpfulness of the information found as shown in Table 3. [Table 3 near here] For 11 of the 13 health-related topics, the most frequently endorsed response was “a little helpful.” There were two topics for which the modal response was “not at all helpful”: 1) pregnancy, labor, and delivery and 2) insurance coverage related to women’s health. At least one-third of respondents rated the disability-specific information found pertaining to fertility, pre-pregnancy/family planning, cancer screening, and well woman exams as “not at all helpful.” In addition, at least 25% rated the disability-specific information found on menstruation, pelvic floor problems, menopause, and finding a doctor as “not at all helpful.” More highly rated, relative to other topics, was contraception and bladder/bowel problems. These were the only two topics for which one-quarter of the sample rated the information they found as being “very helpful.”

Table 3.

Helpfulness Ratings of Online Health Information (OHI)

		Very helpful	A little helpful	Not at all helpful
Variable	N*	N (%)	N (%)	N (%)
Did You Find Information That Was Helpful?
Menstruation	192	32 (16.7)	112 (58.3)	48 (25.0)
Fertility	137	15 (10.9)	75 (54.7)	47 (34.3)
Contraception	139	38 (27.3)	73 (52.5)	28 (20.1)
Pre-pregnancy, family planning	91	11 (12.1)	44 (48.4)	36 (39.6)
Pregnancy, labor, and delivery	145	29 (20.0)	51 (35.2)	65 (44.8)
Bladder/Bowel problems	350	87 (24.9)	204 (58.3)	59 (16.9)
Pelvic floor problems	137	15 (10.9)	81 (59.1)	41 (29.9)
Menopause	133	22 (16.5)	74 (55.6)	37 (27.8)
Cancers specific to women	101	22 (21.8)	56 (55.4)	23 (22.8)
Cancer screening (e.g., pap smears)	115	25 (21.7)	50 (43.5)	40 (34.8)
Well woman exams	122	24 (19.7)	52 (42.6)	46 (37.7)
Finding a physician	285	56 (19.6)	145 (50.9)	84 (29.5)
Insurance coverage	173	32 (18.5)	70 (40.5)	71 (41.0)

^*N differs for each variable. Survey respondents were first asked if they had ever looked for OHI on each topic. They were only asked to provide helpfulness ratings if they acknowledged having looked for such information.

In rating the extent to which the Internet was useful in helping make decisions about their health, more than two-thirds (69%) said that it was “useful” (53%) or “very useful” (16%) with nearly one-quarter (24%) indicating “unsure.” Only 7% of respondents reported that the Internet was “not useful” (6%) or “not at all useful” (1%) in making decisions about their health. Finally, when asked how important it was to be able to access health resources on the Internet, respondents almost universally indicated that it was “important” (35%) or “very important” (62%) for them.

Self-reported eHealth Literacy and Challenges in Health Information Seeking.

Electronic health literacy was relatively high in our sample with an average eHEALS score of 30.63 (SD = 5.01). Examining individual items on the eHEALS, shown in Table 4, a majority of respondents expressed agreement with the health literacy items demonstrating a general sense of confidence in their eHealth literacy skills. [Table 4 near here] The greatest confidence was demonstrated on Item 6 (“I have the skills I need to evaluate, or judge, the health resources I find on the Internet”) with more than 85% of respondents agreeing with this statement. The lowest rated item, the only item on which fewer than half of respondents expressed agreement, was Item 8 (“I feel confident, or sure, about making health decisions based on the information that I find on the Internet”).

Table 4.

eHealth Literacy Scale Item Ratings (N=508)

eHEALS	Strongly Disagree N (%)	Disagree N (%)	Undecided N (%)	Agree N (%)	Strongly Agree N (%)
I know what health resources are available on the internet.	4 (0.8)	60 (11.8)	97 (19.1)	276 (54.3)	71 (14.0)
I know where to find helpful resources on the internet.	7 (1.4)	72 (14.2)	91 (17.9)	274 (53.9)	64 (12.6)
I know how to find helpful health resources on the internet.	8 (1.6)	46 (9.1)	69 (13.6)	297 (58.5)	88 (17.3)
I know how to use the internet to answer my questions about health.	1 (0.2)	24 (4.7)	52 (10.2)	301 (59.3)	130 (25.6)
I know how to use the health information that I find on the internet.	2 (0.4)	21 (4.1)	66 (13.0)	311 (61.3)	107 (21.1)
I have the skills I need to evaluate, or judge, the health resources I find on the internet.	1 (0.2)	30 (5.9)	43 (8.5)	259 (51.0)	175 (34.4)
I can tell high quality health resources from low quality health resources on the internet.	5 (1.0)	36 (7.1)	88 (17.3)	222 (43.7)	157 (30.9)
I feel confident, or sure, about making health decisions based on the information I find on the internet.	9 (1.8)	104 (20.5)	144 (28.3)	193 (38.0)	58 (11.4)

Despite relatively strong confidence in their eHealth literacy, study respondents reported challenges in seeking health information, as reflected on the ISEE items. The average score on the 4-item ISEE was 57.84 (n=506; SD=21.98). Examining individual item responses (Table 5) revealed that a majority of respondents indicated that it “took a lot of effort to get the information they needed” (75%), that they “felt frustrated” during their search for information (72%), and that they were “concerned about the quality of the information” (73%). [Table 5 near here] In contrast, only 34% described that the information they found was “hard to understand.”

Table 5.

Information Seeking Experience Scale Ratings (N=506)

ISEE	Strongly agree N (%)	Agree N (%)	Disagree N (%)	Strongly disagree N (%)
It took a lot of effort to get the information you needed.	125 (24.7)	255 (50.4)	97 (19.2)	29 (5.7)
You felt frustrated during your search for the information.	149 (29.4)	215 (42.5)	103 (20.4)	39 (7.7)
You were concerned about the quality of the information.	144 (28.5)	225 (44.5)	108 (21.3)	29 (5.7)
The information you found was hard to understand.	18 (3.6)	154 (30.4)	198 (39.1)	136 (26.9)

The sample also did not report major challenges understanding statistics. More than two-thirds reported they found it easy (50%) or very easy (18%) to “understand medical statistics” compared to less than one-third who found it hard (28%) or very hard (3%). Responses were evenly split on the item, “I feel uncomfortable with health information that has a lot of numbers and statistics.” On this item, only a few more respondents agreed with the item (strongly agreed, 12%; somewhat agreed, 38%) than disagreed with the item (strongly disagreed 17%, somewhat disagreed 32%). Results of the overall confidence in health information seeking question indicated that more than half of the survey respondents were “very” (32%) or “completely” (24%) confident that they “could get information about their health if they needed it?” However, the remaining 44% were only “somewhat confident” at best that they could get information if they needed it (somewhat, 35%; a little 6%; not at all, 2%).

Discussion

Our national, cross-sectional, anonymous online survey study is the first to investigate OHI seeking by a sample of digitally competent women with physical disabilities. This community-based participatory research study was informed and conducted by many people with physical disabilities including investigators, consultants, a nine-member community advisory board, a medical advisory board experienced with our population, and formal collaborations with five disability organizations throughout the country. Our evidence contributes to the emerging literature on health information seeking among people with disabilities in the digital age.

Like many people in the general population [36], women with disabilities in the current study utilized a wide range of OHI sites and resources. Most respondents had connected online with peers about their health and disability by participating in online support groups and sharing health information through social media sites. We have learned from our own online group intervention studies that women with disabilities appreciate the opportunity to connect with other women with similar disability types and life experiences [37,38]. Authors of a literature review on using social media to seek health information [39] concluded that social media can provide health information as well as social and emotional support for its consumers. They found evidence, however, that benefits may be offset by concerns regarding the quality and authoritative source of information which, in turn, can lessen engagement. As highlighted by Chou et al. [40], an effective use of social media can help reduce health disparities, which is an important consideration for people with disabilities who have been characterized as an unrecognized health disparity population [41]. The respondents in our study had also accessed health information using multiple forms of other media. Interestingly, more than twice as many had watched health-related videos than had listened to health-related podcasts. This is not surprising since such videos, which offer both visual and audio information sometimes accompanied by demonstrations, represent an increasingly utilized platform for disseminating health information. Additionally, 80% of the women reported tracking their own health information online.

Survey respondents searched a wide range of health and disability-related topics. The most often searched topics, searched by more than half of all respondents, were “bowel and bladder problems,” and “finding a physician.” Not surprisingly, these two topics are relevant to women with physical disabilities of all ages, whereas many of the other topics listed were more relevant to either younger (e.g., menstruation, pregnancy) or older women (e.g., menopause).

Most women reported the health information they found online was only a little helpful with more women providing strong negative versus strong positive ratings of the information. The exception was information on bowel and bladder problems and contraception. These were the only two topics, out of 13, on which a greater number of women said the information was very helpful than said the information as not helpful at all. The topic rated most poorly in terms of the helpfulness of the information found was pregnancy, labor, and delivery. Despite the fact that decades have passed since the American with Disabilities Act [42] was enacted, women with disabilities in the U.S. continue to lack access to helpful and medically appropriate reproductive health services and information. Clearly, there is a need to develop and disseminate sound scientific reproductive health information for women with physical disabilities.

Our relatively well-educated sample generally had confidence in their health literacy skills. Most did not report difficulty understanding numbers and statistics in health information; however, 31% reported that it was hard to understand medical statistics, and 51% reporting being uncomfortable with health information with a lot of numbers and statistics. This was particularly notable given that 99% of the sample had at least a high school degree and 70% had at least a college degree. In addition, a majority described it requiring a lot of effort to get the information they needed and described feeling frustrated searching for it. A majority of the women also were concerned with the quality of the information available. Nonetheless, most reported the Internet was useful in helping with health-related decisions. While a majority stated that it was very important to be able to access OHI, a disturbing number were not sure they could get information about their health if they needed it. We anticipate this number would be much larger among a less advantaged population.

Although our study reveals the first known findings about health information seeking by women with disabilities, we acknowledge several limitations. First, our sample was not representative in that it was predominantly White whereas it is widely known that disability disproportionately impacts people of color [43]. As this was an online study, the racial disparity is not surprising given that Black and Hispanic adults in the U.S. are less likely than their White counterparts to own a desktop or laptop computer and have high-speed Internet at home [44]. Although translating the survey into Spanish could have increased participation among Latinas with disabilities, we did not propose to do so in the grant application. We have implemented other surveys and questionnaires in both English and Spanish [45,46], and our experience informs us of the costs associated with survey translation and back translation of the introductions, survey instructions, and informed consent information in addition to the content of the survey itself. Unfortunately, our funding for this project did not allow translation services.

Although women with disabilities are known to have low education rates [2], our sample was highly educated with 71 percent having at least a bachelor’s degree which is approximately twice the national average of women in the U.S. [47]. Since both race and education are linked with income, it is not surprising that our sample of predominately White, highly educated women would have the necessary resources and opportunities for accessing the Internet from their homes or elsewhere, thus allowing participation in this study. The lack of a representative sample narrows our focus, limiting the generalizability of our findings to the larger population of women with physical disabilities.

Additionally, as there are no existing health information seeking measures inclusive of issues salient to women with disabilities, it was necessary to seek the guidance of our community and medical advisors to generate many of the questions used in our survey. As a result, our analyses predominantly consisted of examining frequencies of individual, study-generated items which prohibited our making comparisons with other populations. Survey questions consisted of self-report items, some of which asked the respondent to recall and report on Internet activities engaged in within the past year. The accuracy of such recall cannot be guaranteed and an additional study limitation. Finally, although our choice of using an online survey methodology provided fast data collection at low cost and other advantages, it also was fraught with commonly observed limitations [48]. These limitations included the inability to track non-response rates as well as establish an accurate sampling frame and estimation of the characteristics of the population as a whole.

Our findings are consistent with previous research indicating that many people with disabilities are digitally engaged and find the Internet an important source of information about health and medical treatment [23,24]. As such, future research should attempt a broader reach to include women with physical disabilities who are more racially diverse and socioeconomically disadvantaged than those in the current study. To be inclusive of the wider population of women with physical disabilities, greater efforts should be taken to recruit through rehabilitation and disability, faith-based, and other community-based organizations in socially disadvantaged settings as well as encourage women to participate using their mobile phones and other mobile technologies.

Most importantly, our findings reveal that, even among this sample of highly educated frequent internet-users, a sizeable percentage found searching for information difficult and frustrating, did not find the information very helpful, or had concerns about the quality of the information. Having accurate health information is critically important to making well-informed decisions about one’s health. This is especially true for women with physical disabilities who may face numerous health challenges including secondary health conditions [45]. Given the potential utility of the Internet to provide such valuable information, it is concerning that nearly one-third of the sample were either unsure that the Internet was useful in helping them make decisions about their health or explicitly stated that it was not helpful. Similarly, the finding that over 40% of the women were only somewhat confident, at best, that they could get information about their health if they needed it, is unacceptable.

Researchers and healthcare providers alike can do more to make high quality, easy to understand, disability-sensitive health information available to women with physical disabilities. Attention needs to be given to making sure that numbers and statistics included in health information are clearly presented using plain language and written in a way that promotes understanding of its relevance to the individual consumer. Research findings must be translated into materials designed for and developed in partnership with consumers with lived experience. Consumer input will help ensure that materials meet the accessibility and accommodation needs of the population - nearly 50% of whom, in our sample, required personal assistance or assistive technology to access the Internet.

Finally, OHI must be easier to locate to be useful to consumers. Currently, much health information pertaining to the issues and topics of interest to women with physical disabilities is fragmented, requiring women to search many different sites and online resources to try and pool relevant information. There is a great need for a national resource and repository for health-related information for women with disabilities. This study serves as a call to action to disability and rehabilitation scientists, health care providers, public health and health policy professionals to develop and implement methods for enhancing the availability and accessibility of OHI critical to empowering women with physical disabilities to make well-informed health decisions and attain an optimal quality of life.

Implications for Rehabilitation.

• Access to high quality online health information (OHI) is critical for reducing disparities, overcoming barriers, and improving the health of women with disabilities.

• Many of the women with disabilities in our study found OHI seeking difficult and frustrating, did not find the information very helpful, and had concerns about the quality of the information.

• Disability and rehabilitation scientists, health care providers, and public health and health policy professionals need to do more to enhance the availability and accessibility of OHI and resources critical to empowering women with physical disabilities to make well-informed health decisions.

• Physical medicine and rehabilitation scientists are encouraged to develop and improve assistive technologies needed for accessing OHI, which in turn can promote the independent functioning of people with disabilities.