Abstract
Introduction
The worldwide expansion of the Internet offers an important modality of disseminating medically accurate information about medication abortion. We chronicle the story of www.medicationabortion.com, an English-, Spanish-, Arabic-, and French-language website dedicated to three early abortion regimens.
Methods
We evaluated the website use patterns from 2005 through 2009. We also conducted a content and thematic analysis of 1,910 emails submitted during this period.
Results
The website experienced steady growth in use. In 2009, it received 35,000 visits each month from more than 20,000 unique visitors and was accessed by users in 208 countries and territories. More than half of all users accessed the website from a country in which abortion is legally restricted. Users from more than 40 countries sent emails with individual questions. Women often wrote in extraordinary detail about the circumstances of their pregnancies and attempts to obtain an abortion. These emails also reflect considerable demand for information about the use of misoprostol for self-induction.
Conclusion
The use patterns of www.medicationabortion.com indicate that there is significant demand for online information about abortion, and the findings suggest future priorities for research, collaboration, and educational outreach.
Keywords: Internet, abortion, reproductive health, misoprostol, mifepristone
Introduction
Subject: I NEED URGENT HELP [Nigeria, English]
Pleas i need your help im pregnant what can I do?….Iam from Nigeria 23 years old…the last sex i had was on febuary 19th I found out im pregnant by a test conducted on me […] can you prescribe some pills for me since i couldnt get acess to the mifepristone/misoprostol here in my country please?thanks!1
The worldwide expansion of the Internet offers an important modality of disseminating medically accurate information about medication abortion to both health service providers and women. However, a study published in 2002 revealed English-language online information about the mifepristone/misoprostol regimen was highly politicized and often medically inaccurate [1]. Further, few multilingual resources and virtually no Arabic-language online resources were dedicated to medication abortion in the early 2000s.
It was this context that motivated the development of www.medicationabortion.com, a project undertaken by Ibis Reproductive Health and the Office of Population Research at Princeton University. Our aim was to create a medically accurate, multilingual online resource dedicated to three early pregnancy termination regimens: mifepristone/misoprostol, methotrexate/misoprostol, and misoprostol-alone. We were especially interested in developing an Arabic-language website to expand information about medication abortion in the Arab world.2 The content went through extensive expert review, and the Arabic-language version went through a separate review process to ensure accessibility of materials across regional dialects. We launched the Arabic-, English-, and French-language versions in September 2003. In partnership with the Population Council’s Mexico City office we added a Spanish-language version of the website in September 2004.
Utilization of the website during the first year in which all four language versions were operational has been described previously [2]. Here we chronicle the history of www.medicationabortion.com focusing on the use patterns during the first five years of its operation and an evaluation of emails sent to website administrators over this period.
Methods
Description of the website
Each language version of the website contains a homepage, a section dedicated to each medication abortion regimen, 28 frequently asked questions (FAQs) linked to answers, sections dedicated to references, links, and educational resources. In May 2005, we launched an English-language database, searchable by country, with information about the legal status of abortion, the regulatory status of mifepristone, and a number of reproductive health indicators. Website content was updated regularly through the end of 2009. However, throughout the five-year evaluation period the overall structure of the website was maintained.
The website was not designed to be an interactive resource. As a result, we posted terms of use and privacy statements informing users that we do not reply to individual questions or provide individualized medical advice. However, in 2005, for administrative purposes, we also posted a statement, in small font, at the bottom of each page of the website reading: “Please contact us with suggestions, updates, or link requests at medicationabortion@ibisreproductivehealth.org.” Between 2005 and 2009, we received over 4,000 emails.3
Data analysis: Use patterns
We used Webusage 8.0 (2005) and SmarterStats6.2 (2006-2009) to analyze website utilization from January 1, 2005 through December 31, 2009. We present more detailed information about calendar year 2009, the final year in the evaluation period.
In order to evaluate country-level website utilization by the legal status of abortion, we developed a classification system based on the categories published by the Center for Reproductive Rights (CRR) in 2008 [3]. For this paper, we define countries as settings where first trimester abortion is generally legally restricted if they fall into CRR Categories I (abortion is only permitted to save the woman’s life or prohibited altogether), II (abortion is permitted to save the woman’s life or preserve her physical health), or III (abortion is permitted to save the woman’s life or preserve her physical or mental health). In all other countries, first trimester abortion is either legally permissible on socioeconomic, health, and life preservation grounds (Category IV) or without restriction as to reason (Category V) and we classify these countries as settings where first trimester abortion is generally legally permissible.
Data analysis: Emails
We reviewed the approximately 4,020 substantive (i.e., non-spam) emails received from 2005 through 2009 and found that 1,910 were sent by individuals writing about their personal situation or circumstances (“individual emails”) in English, Spanish, Arabic, or French.4 AF translated these into English and conducted content and thematic analyses using both a priori (predetermined) codes and inductive techniques. Translation and classification questions were discussed as a team and divergent opinions were resolved through consensus. In the results section we present the major themes that emerged and reproduce emails to illustrate findings. This study received a determination of exempt status by Princeton University’s Institutional Review Board.
Results
Overall use patterns
The website experienced a steady increase in traffic over the evaluation period, from 82,000 visits in 2005 to more than 421,000 visits in 2009 (Figure 1). In 2009, the website averaged more than 35,000 visits per month by approximately 20,000 unique visitors. Thus a significant proportion of users accessed the website more than once in a calendar month. The average visit included two page requests and lasted just shy of three minutes. The English-language version of the website was the most popular (52.3%), followed by the Spanish- (32.7%), Arabic- (12.0%) and French-language versions (4.0%).
Figure 1.
Visits to the website by year, 2005-2009
In 2009, more than half of all visits resulted from a Google search, nearly 30% originated from direct entry, and approximately 10% originated from other reproductive health websites. Users arrived at the website using thousands of search terms in a variety of languages. However, nearly 20% of all visits to the website were the result of a search including the term “misoprostol.” We present the most common search strategies on Figure 2.
Figure 2.
Ten most common search strategies, January 1, 2009-December 31, 2009*
* The 100 most commonly used search terms and phrases during the study period were reviewed and grouped by general topic. The top ten general topics are listed in this Figure.
Geographics5
In 2009, we were able to identify the country for 97% of the visits to the website, and the website received at least one visit from 208 countries and territories worldwide (Figure 3). Nearly a third of all visits came from the US (71,480) and Mexico (61,380) and about half of all visits originated from capital cities and major urban centers.6 Based on our binary legal classification and adjusting for the differential legal status of abortion in Mexico City (generally legally permissible) and the rest of the country (generally legally restricted), in 2009, 55% of visits originated from countries where abortion is generally legally restricted. Website users from the Arab world predominantly accessed the website from Saudi Arabia, Egypt, the UAE, and Morocco, all countries where first trimester abortion is generally legally restricted.
Figure 3.
Global map of the “geographics” of website visits, 2009
Emails: Overall
Subject: SOS [France, French]
I am a young [unmarried] Arab woman living in France and I am 15 to 16 weeks pregnant. I can no longer have an abortion in France and I cannot go abroad as I am monitored by my parents – can I appeal for your assistance in provoking a miscarriage at this stage of my pregnancy. Thanks.
Of the 1,910 individual emails written in English, Spanish, Arabic, or French, we were able to assign a country to about 60% (1,144) through the email content, email signatures, or the email address itself. We were able to identify emails originating from 42 countries and territories (Figure 4). Consistent with the overall use patterns of the website, the majority were sent from the US, Latin America (particularly Mexico, Colombia, and Peru), and the Arab world.
Figure 4.
Countries of individual emails sent from 2005-2009 (indicated in gray)
Many women wrote in extraordinary detail about the circumstances of their pregnancy, providing information about partners and children, chronicling experiences of rape, and relating attempts at self-induction. Some women wanted reassurance that the bleeding they were experiencing was “normal” or the regimen they used was accurate. Many wrote with questions about the side effects and long-term consequences of using medication abortion methods. Others asked where they could get mifepristone or misoprostol and how much it would cost. Notably, a sizeable minority of emails centered on unintended pregnancies and abortion needs after the first trimester. Although the emails were highly individualized, we were able to identify a number of additional country- and regional-specific themes.
Emails from the US (n=438)
Subject: abortion pill (US, English)
Is the pill safe forr a girl around 17 of age?.. Im 17 born january 26, 1992.. Would i work on me?.. Is it legal?.. Do i need my parents consent?.. And if i have a private plan like would it cover the price?.
Of the 438 emails sent from the US, nearly two thirds (n=285) centered on issues regarding the legal status of abortion and logistics of accessing services. Notably, these included minors seeking information about parental involvement and women seeking information about covering the cost of abortion care. Within this broad theme of legality and access were those showing confusion regarding the legal status of mifepristone. A subset of these writers indicated that they had been told by a clinician that medication abortion was not legal or available in the US.
A second major theme involved finding an abortion provider. In about a quarter of all emails (n=120), women (or their partners or loved ones) asked for information about where they could go to obtain a medication abortion. A third theme that emerged from emails in the US centered on parents (typically mothers) asking for information about medication abortion on behalf of their teen daughters. Although we received only about a dozen of these (roughly 3% of all US emails), this theme was notable because we did not receive comparable emails from individuals in other countries or regions. Finally, we received a number of emails from the US that sought information about “what happens” to the fetus during a medication abortion termination. These emails (about 5% of all US emails) often evinced a lack of understanding about what an abortion entails.
Emails from Latin America (n=330)
Subject: doubt (Colombia, Spanish)
I used misoprostol in the 3rd week of my pregnancy, the bleeding only lasted for three days, the first day there were like three clots, the next day I bled less and since there was almost nothing I inserted 4 more tablets and I expelled one more clot. All this happened in just 3 days, and how does this really work? Do i need to repeat it?
Of the 330 emails from Latin America, more than 80% (n=268) centered on the use of misoprostol alone for early pregnancy termination. These included questions about where to obtain misoprostol, the optimal regimen for use, efficacy, side effects, and cost. Women often provided detailed information about physical symptoms and bleeding patterns after the use of misoprostol. About two dozen of the misoprostol-related emails contained queries regarding teratogenicity in cases of continued pregnancy and about 12% of all emails from Latin America centered on the use of methotrexate, themes that did not emerge in emails from individuals in other countries or regions.
Emails from the Arab world (n=147)
Subject: None (Egypt, English)
WHICHE PLACES INEGYPT
The majority of the 147 emails from the Arab world centered on the theme of where to obtain an abortion. Many of these emails were extremely short and simply posed some permutation of the question “where do I go?” or “how do I get an abortion in country X?” About a quarter of all emails from the Arab world centered on the use of misoprostol-alone. Notably, this pattern emerged only in emails sent in 2008 and 2009 and almost all originated from Gulf countries such as Saudi Arabia, the UAE, and Kuwait. Finally, in over a fifth (n=31) of these emails the writer identified himself as being the male partner of a woman who needed an abortion, a dynamic we did not observe with frequency in the emails from other countries or regions.
Discussion
Of the more than 41 million abortions that occur worldwide each year, over 19 million are defined as unsafe [4]. The overwhelming majority of unsafe abortions take place in developing countries where abortion is generally legally restricted. In these contexts, complications from unsafe abortion constitute a major contributor to both maternal morbidity and mortality and are costly to health systems [4,5,6]. Research has repeatedly shown that the legal status of abortion does not correlate with the abortion rate. Liberalizing abortion laws, improving the quality of both post-abortion care and contraception services, and expanding access to safe, high-quality abortion care have been repeatedly identified as global priorities by reproductive health advocates [4,5].
Providing information about safe and effective medication abortion methods to both clinicians and women is a critical part of the effort to expand access to safe(r) abortion services. In those countries where abortion is generally legally permissible and mifepristone is registered for early pregnancy termination, providing multilingual evidence-based information about the regimen furthers the overarching goal of improving the quality of abortion services and serves to help women make an informed decision regarding abortion procedures.
But in those countries where abortion is generally legally restricted and mifepristone is not available, making information available about the use of misoprostol, including evidence-based information about timing, dosing schedules, side effects, complications necessitating additional medical intervention, and expected outcomes, may reduce women’s risk of abortion-related complications. Use of misoprostol for early pregnancy termination is certainly safer than many of the practices, such as sharp stick insertion, pummel massage, and caustic douches, used by women throughout the world who live in settings characterized by abortion laws that are generally restrictive [5]. A growing body of evidence suggests that in generally legally restrictive environments where “gold standard” modalities of pregnancy termination are not available, use of misoprostol has the potential to reduce complications associated with unsafe abortion [6,7,8]. Further, use of misoprostol to initiate pregnancy loss may facilitate women’s ability to obtain clinic-based post-abortion care services. However, studies in multiple countries have shown that women and many health care providers lack evidence-based information about how to best use misoprostol, and, until relatively recently, few multilingual resources about misoprostol for early pregnancy termination have been available [9,10].
The use patterns of www.medicationabortion.com and the emails submitted suggest that the website serves as a truly global resource and demonstrates high demand for online multilingual abortion information. There is undoubtedly demand for abortion information in additional languages, demand which is likely to increase as Internet use continues to rise dramatically, particularly in the developing world [11]. Although 40% of women of reproductive age live in countries where abortion is generally legally restricted, 55% of all visits to the website originate from these countries. Thus the website appears to be especially popular among users in these settings.
However, the findings from the analysis of emails sent to the website suggest that static online resources are not sufficient to meet fully users’ needs. The emails received from the US, Latin America, and the Arab world are consistent with broader trends around abortion in these respective countries and regions. In the US, for example, it is not surprising that there is significant confusion regarding the legal status of abortion and logistic issues surrounding access given the politicized nature of abortion and the complexity of both state-based regulations and insurance coverage. Users from Latin America, a region where abortion laws, overall, are generally legally restrictive but misoprostol is widely available without prescription, show considerable interest in the misoprostol-only abortion regimen. That emails from these users are overwhelmingly centered on misoprostol use (with a secondary interest in methotrexate) suggests a clear need for additional interactive resources and outreach efforts to help women better understand the optimal non-mifepristone regimens. In the Arab world there is considerable variability in both the legal status of abortion and the availability and accessibility of services, but the majority of emails originated from countries in which abortion is generally legally restricted. The recent uptick in questions about misoprostol from users in this region coincides with a trend of increased sales of misoprostol, efforts to increase information about misoprostol for pregnancy termination in certain parts of the region, and efforts by some governments to restrict misoprostol use [12,13]. Our findings suggest a budding need for additional outreach and educational efforts. Finally, the relatively high percentage of emails from male partners in this region is notable and may reflect the gendered nature of Internet access.
One striking aspect of the emails, overall, is the amount of revealing personal details volunteered by writers. This is likely related to the anonymous nature of online communication and is consistent with findings from other sexual and reproductive health websites [14,15]. Online resources, if carefully designed, could facilitate abortion research, enabling researchers to better understand global pregnancy and abortion experiences.
The unsolicited volunteering of so much personal and medical information also demonstrates the considerable need for online, interactive resources. The demand for www.womenonweb.com, a website that offers women online consultations for medication abortion, provides further evidence of this need [16,17]. Supporting and expanding these resources, identifying mechanisms to better link evidence-based online abortion resources, and developing additional online mechanisms for providing more individualized information, support, and services appear warranted.
Limitations
There are technical limitations in our ability to identify the country of users secondary to “routing” of Internet traffic, with the result that we likely underestimate visits from developing countries. In addition, service gaps in data collection software over the five-year period mean that all numbers are underestimated. There are also limitations in our ability to analyze emails received; some were in languages our study team does not read, and spam filters may have further prevented some messages from reaching us. Counts of emails received are therefore likely underestimates.
Conclusion
Findings from this research leave no doubt that there is considerable demand for online, multilingual resources dedicated to medication abortion and that there is a considerable need for additional interactive, individualized information. Bolstering the efforts and visibility of services like www.womenonweb.com appears warranted. But even in the absence of providing women with individualized information, websites like www.medicationabortion.com may also be able to create more dynamic capabilities. This might include drawing from the questions posed by women to develop country-specific case studies that walk the reader through situations. Moving forward, it will be valuable to explore ways in which online resources can further expand and develop additional features to better address users’ needs.
Acknowledgements
We are grateful to the William and Flora Hewlett Foundation and the Richard and Rhoda Goldman Fund for their support of www.medicationabortion.com. The Echo Endowed Chair in Women’s Health Research at the University of Ottawa (AF) is supported by Echo: Improving Women’s Health in Ontario, an agency of the Ministry of Health and Long-Term Care. This work was supported in part by the Eunice Kennedy Shriver National Institute of Child Health and Human Development grant for Infrastructure for Population Research at Princeton University, Grant R24HD047879 (JT). The authors would like to thank Aida Rouhana, Claudia Diaz-Olavarrieta, Kate Schaffer, and Kelsey Holt for their work on aspects of the website.
Footnotes
Identifying information from all quoted emails has been redacted. English language emails have been reproduced in otherwise unaltered form. Emails sent in Spanish, Arabic, and French have been translated to English.
Throughout this paper we use the standard definition of “Arab world” which refers to the 22 countries and territories of the Arab League.
Per our terms of use, we do not provide individualized medical information or advice in response to emails received. However, we do provide template information in several languages that directs users to the specific pages of www.medicationabortion.com and to the resources and links section of the website.
Of the other 2,110 emails received during the study period, approximately 1,800 contained administrative correspondence and link requests, requests from providers for resources and training information, queries from researchers and journalists, expressions of gratitude, and statements of opposition or hate. The remaining 300 were “individual emails” written in languages other than English, Spanish, Arabic, or French.
The geographics of website utilization is obtained by capturing geographic location information about an IP address or other logfile data, generally obtained through regional Internet registries.
At the other end of the spectrum, the website received only 2-3 visits from the Pacific Island nations of Palau, Vanuatu, and Kiribati. No visits were recorded from Chad, Comoros, Liechtenstein, Nauru, North Korea, San Marino, Sao Tome and Principe, Vatican City, or Greenland, although it is unclear if our analytics software captures IP addresses from Greenland independent from Denmark.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Mashiach R, Seidman GI, Seidman DS. Use of mifepristone as an example of conflicting and misleading medical information on the Internet. British Journal of Obstetrics and Gynaecology. 2002;109(4):437–442. doi: 10.1111/j.1471-0528.2002.t01-1-01124.x. [DOI] [PubMed] [Google Scholar]
- 2.Foster AM, Wynn L, Rouhana A, et al. Providing medication abortion information to diverse communities: Use patterns of a multi-lingual website. Contraception. 2006;73:264–271. doi: 10.1016/j.contraception.2006.03.013. [DOI] [PubMed] [Google Scholar]
- 3.Center for Reproductive Rights [Accessed 15 December 2011];The world’s abortion laws fact sheet. 2008 At: http://reproductiverights.org/sites/crr.civicactions.net/files/pub_fac_abortionlaws2008.pdf.
- 4.Singh S, Wulf D, Hussain R, et al. Abortion worldwide: A decade of uneven progress. Guttmacher Institute; New York: 2009. [Google Scholar]
- 5.Grimes DA, Benson J, Singh S, et al. Unsafe abortion: The preventable pandemic. Lancet. 2006;368(9550):1908–19. doi: 10.1016/S0140-6736(06)69481-6. [DOI] [PubMed] [Google Scholar]
- 6.Singh S. Hospital admissions resulting from unsafe abortion: estimates from 13 developing countries. Lancet. 2006;368(955):1887–1892. doi: 10.1016/S0140-6736(06)69778-X. [DOI] [PubMed] [Google Scholar]
- 7.Barbosa RM, Arilha M. The Brazilian experience with Cytotec. Studies in Family Planning. 1993;24(4):236–240. [PubMed] [Google Scholar]
- 8.Briozzo L, Vidiella G, Rodríguez F, et al. A risk reduction strategy to prevent maternal deaths associated with unsafe abortion. International Journal of Gynaecology and Obstetrics. 200695(2):221–226. doi: 10.1016/j.ijgo.2006.07.013. [DOI] [PubMed] [Google Scholar]
- 9.Cohen J, Ortiz O, Llaguno SE, et al. Reaching women with instructions on misoprostol use in a Latin American country. Reproductive Health Matters. 2005;13(26):84–92. doi: 10.1016/S0968-8080(05)26202-X. [DOI] [PubMed] [Google Scholar]
- 10.Espinoza H, Abuabara K, Ellertson C. Physicians’ knowledge and opinions about medication abortion in four Latin American and Caribbean region countries. Contraception. 2004;70:127–133. doi: 10.1016/j.contraception.2004.03.011. [DOI] [PubMed] [Google Scholar]
- 11. [Accessed 15 December 2011];Internet World Stats. At: http://www.internetworldstats.com/stats.htm.
- 12.Fernandez MA, Coeytaux F, Gomez Ponce de Léon R, et al. Assessing the global availability of misoprostol. International Journal of Gynecology and Obstetrics. 2009:180–186. doi: 10.1016/j.ijgo.2008.12.016. [DOI] [PubMed] [Google Scholar]
- 13.Foster A, Daoud F. Misoprostol in the West Bank: Navigating barriers to abortion access; Paper presented at the Middle East Studies Association annual meeting; San Diego, CA. November 2010. [Google Scholar]
- 14.Goold PC, Ward M, Carlin EM. Can the Internet be used to improve sexual health awareness in web-wise young people? Journal of Family Planning and Reproductive Health Care. 2003;29:28–30. doi: 10.1783/147118903101196864. [DOI] [PubMed] [Google Scholar]
- 15.Wynn LL, Foster AM, Trussell J. Would you say you had unprotected sex if…? Sexual health language in emails to a reproductive health website. Culture, health & sexuality. 2010;12(5):499–514. doi: 10.1080/13691051003653631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gomperts RJ, Jelinska K, Davies S, et al. Using telemedicine for termination of pregnancy with mifepristone and misoprostol in settings where there is no access to safe services. British Journal of Obstetrics and Gynaecology. 2008;115(9):1171–1175. doi: 10.1111/j.1471-0528.2008.01787.x. [DOI] [PubMed] [Google Scholar]
- 17.Gomperts R, Petow SAM, Jelinska K, et al. Regional differences in surgical intervention following medical termination of pregnancy provided by telemedicine. Acta Obstet Gynecol Scand. 2012;91:226–231. doi: 10.1111/j.1600-0412.2011.01285.x. [DOI] [PubMed] [Google Scholar]