Three Critiques Of Text4baby And Some Ideas For Improvement – Jessica Turon

Introduction

Text4baby is a national educational program created and coordinated by the American non-profit Healthy Mothers, Healthy Babies (HMHB). The text4baby program is a joint effort by HMHB and many other local, state, and national partners, including corporate, non-profit, and governmental agencies. The program is publicly marketed as “a free mobile information service designed to promote maternal and child health” (1). More specifically, text4baby attempts to improve the US rates of preterm birth, low birth weight (LBW), and infant mortality (2). Women who sign up for the program provide their due date (if pregnant) or baby's birthday (if a mother of an infant under a year) and receive three text messages a week. The messages, which were developed by HMHB and then reviewed by the CDC and other governmental health agencies, cover a wide range of topics relating to prenatal health and infant health, and are coordinated with the month of pregnancy or baby's age. The underserved populations text4baby targets are more likely to have mobile phones than internet access; due to the unique cooperation of many mobile phone companies, the messages are free (no program charge or mobile plan text message charges) for 96% of mobile subscribers.
Text4baby is an ambitious program with an admirable goal and some major flaws. This paper provides a three-pronged criticism of the prenatal component of text4baby, encompassing philosophical and practical problems. It argues that text4baby: (I) is critically unaware of issues of gender and power in maternal health, (II) treats prematurity and LBW as though they result from or can be prevented by an individual's decisions and behaviors, and (III) fails to tailor the intervention to meet consumers' needs, treating women as a uniform market. The paper then provides suggestions for ameliorating these problems while retaining text4baby's primary innovation (the use of text messaging), sketching what a new model of the intervention might look like.

(I) Gender and Power

Pregnancy and childbirth are physiological processes unique to biological women, but are also socially constructed and gendered in culturally specific ways (3). In the United States, the dominant model views pregnancy and childbirth as medical events requiring expertise and equipment in a medical setting (4). Most women receive prenatal care in clinics, doctors' offices, or hospitals, from medical professionals. This care generally includes routine use of high-tech equipment and tests such as ultrasounds. When it comes to birth itself, 99% occur in hospitals, where "from the time [a woman] is admitted, decision-making power and responsibility for her state rest primarily with hospital personnel and the physician in charge" (3:46).
The history of childbirth in the US reflects a gradual transfer of power and responsibility from laboring women and midwives to physicians and specialists such as obstetricians, with a related shift away from birth at home to birth in a hospital (5). The implications of the medicalized view of pregnancy and childbirth are broad. Critics note that unlike other processes and events overseen by the medical system, these are not inherently pathological: "the history of obstetrics, when seen from the women's point of view, seeks to problematize, pathologize, and then control and rescue reproductive power" (6:18). In this system, pregnant women are passive patients, and obstetricians are authoritative actors who are responsible for the end result of a healthy baby.
The critique of text4baby from the standpoint of feminist thought about gender and power is twofold: first, in the tone and general content of the messages, and second, in the delivery method.
Mainstream care providers typically only value the kind of evidence and information acceptable in the medical arena, leading to a over-dependence on positivistic facts, measurements, and diagnoses. In text4baby, most messages are related to medical aspects of pregnancy, such as potential poor outcomes and their warning signs, the danger (to the fetus) of various activities, vaccinations, health care access, and nutrition. While text4baby does include messages coded as "emotional (encouragement)," even these are couched in medical terms. Women's "experiential knowledge is discounted throughout pregnancy" (7:32) by the medical characterization of pregnancy that text4baby invokes.
One major emphasis of text4baby is full participation in prenatal care. Text4baby attempts to address issues of access to care by providing phone numbers for free or low-cost health care and WIC programs. However, the messages do not acknowledge the deeper issues related to health care access for minority women. Personal history and institutional history are at issue in women's utilization of health care services (8). Undocumented immigrants may avoid accessing health care for fear that their autonomy and personal or family security may be sacrificed (9). African American women as a population have a troubled history with organized health care and may be hesitant to seek care from within this system (10). In fact, underutilization of health care services during pregnancy may for some marginalized women be a form of "voting with their feet," that is, rejecting the existing system of care rather than attending appointments only to be blamed and scolded for their poor health outcomes (6).
Another flaw of text4baby (as regards gender and power) is the strictly authoritarian nature of the intervention. There is no place for women to respond or even react to the texts. Despite the nature of text messaging as a back-and-forth, social, communicative technology, text4baby has ruled out any possibility for the woman to contribute knowledge, understanding, or experiences of her own pregnancy. In the book Women, Power, and Childbirth, the author cites evidence that “pregnant women are characterized by physicians as being irresponsible, childlike, and incompetent to make decisions regarding their own well-being." (7:25) Text4baby does nothing to work against this damaging pattern. By signing up for the text messages, a woman subjects herself to a strictly unidirectional flow of information, in which anonymous medical authorities deliver isolated edicts. If she has questions, problems, objections, or additions, there is no outlet or ability to respond. The communicative options available are to text STOP or UPDATE (to change a due date) – strictly take-it or leave-it. In this case, the idea that “information is power” fails to account for “the political character of the production and distribution of information itself" (11:16).

(II) Focus on Individual Factors

Text4baby focuses on the individual woman as the target for interventions related to LBW, preterm birth, and infant mortality. First, putting the onus on the woman and behavior factors for these outcomes is problematic, as research shows that a large proportion of risk is beyond the woman's scope of control. Second, a program consisting of a small number of 'tips' addressing the risk factors that are within an individual's control is of questionable efficacy; this paper discusses maternal smoking and nutrition.
In an article in the "GW Today," the PI for the study evaluating text4baby's efficacy says that "it’s strongly believed that a lot of the reason for [prematurity and LBW in low-income women] is behavioral" (12). It is difficult to interpret this comment charitably. In reality, both proximal and distal causes of poor pregnancy outcomes are difficult to determine, and there are a very large number of known and interacting risk factors (13). Overall, it is estimated that 40% of birthweight is due to heredity and the remainder due to environmental factors (14).
An Institute of Medicine (IOM) study of LBW categorized 41 "principal risk factors" for LBW into six categories. Of these 41, only a handful are factors that a woman could reasonably influence during her pregnancy. Others are demographic, such as race, socioeconomic status, or age; related to the woman's overall health and welfare prior to pregnancy; medical emergencies (e.g. first or second trimester bleeding, spontaneous premature rupture of membranes); and so on. Of particular importance is race: African-American women, for example, are three to four times more likely to have preterm births (15). A wide disparity remains when risk factors are controlled for: in one study of low SES white and black women, many of the risk factors were worse among white women, yet black infants weighed 200g less and were born 4 days earlier (16). Given text4baby's stated attempt to reduce health disparities among minorities, then, the emphasis on behavior is particularly unreasonable.
While the text4baby messages do not explicitly state that a woman's choices are responsible for her baby's health and birthweight, the tone is often scolding, cajoling, or threatening. Even in "encouragement" messages, the woman is responsible (praiseworthy or blameable) for how well her baby does: "Your healthy choices are helping your baby grow" (17). Many texts focus on danger signs of medical conditions. However, a single text message about a medical emergency (delivered asynchronously to the onset of the putative pathology) is unlikely to result in a net change of outcome. The value of including these 'tips' should be weighed against the anxiety that may result from constant worry about what could go wrong, especially given that in the vast majority of cases, birth outcomes are positive. Chronic stress is well established as correlating with preterm birth (18) and there is some research that points to “pregnancy anxiety” as an independent psychosocial stressor (19).
The fact that preventable and behavioral risks make up a minor influence on preterm births and LBW does not mean they are not worth targeting. For example, some 25% of women smoke during pregnancy and smoking has been blamed for 25% of LBW and 5% of infant mortality (14, 20). The text4baby messages do mention smoking more frequently than other topics (six times within the 118 pregnancy messages). However, it is highly unlikely that such a lightweight intervention will have a meaningful impact on maternal smoking cessation. In a Cochrane review, smoking cessation interventions during pregnancy resulted in an overall pooled reduction of 6% more women discontinuing smoking. Many women spontaneously quit when they learn they are pregnant, but their characteristics tend to be conducive to quitting; the pool of women who continue to smoke in pregnancy are often more addicted, have less social support, are poorer, and have less education. More than 50% have had or currently have psychological symptoms. Of the types of interventions studied, all were more involved than simple education, and the range of successes varied widely (21). While two of the text4baby messages do include the number for a toll-free quit-line, the chance that a few text messages will themselves help a pregnant smoker to quit is slim. It may be fairly argued, though, that the messages do no harm – if even a few women are prompted to quit smoking (or scale back) due to the messages, they may be worthwhile.
Nutrition and appropriate maternal weight gain are also preventable risk factors for preterm births and LBW babies. Text4baby sends about a dozen texts dealing with various aspects of nutrition, including prenatal vitamins and specific nutrients, getting enough calories, not getting too many calories, and eating breakfast (with specific food suggestions). Several provide a phone number to connect with a WIC program or the USDA food clearinghouse.
According to recent research, weight gain in pregnancy is still controversial among health care professionals, despite evidence confirming a set of IOM guidelines (22). The majority of women, in fact, gain either more or less than recommended, with only about 30% falling within the recommendations (23). A woman's pre-pregnancy weight, like other pre-pregnancy maternal health indicators, is a strong predictor of LBW (heavier women have heavier babies) and a low rate of gain during pregnancy is a strong predictor of preterm birth (24, 25). Some vulnerable groups, such as African American women, are twice as likely to gain less than 16 pounds during pregnancy (13). Maternal nutritional interventions that "[provide] information and advice alone" are less likely to be effective than those that actually supply food or vouchers (26:14). Even more bleakly, another study finds that “virtually
all nutritional interventions used to reduce low birth weight … have failed” (24).
Given the complicated recommendations in this area, undifferentiated text messages are a particularly inadequate way to offer medical suggestions or advice. The texts send mixed and vague messages about how much to eat. They do not provide information beyond the idea that the woman should be aware of – or possibly changing – her food intake. The breakfast menus that are suggested on a few occasions are made up of multiple items, including fresh dairy, produce, and healthy baked goods, which may also be difficult for women to procure.

(III) Lack of formative research and social marketing

A final criticism of text4baby is its disregard for the importance of marketing and formative research in public health interventions. Similar marketing principles to those used in corporate sales can be used in public health campaigns (27). Market segmentation, following formative research, is a powerful force for persuasion in public health social marketing (28, 29). Text4baby, however, is a strictly one-size-fits-all intervention. When signing up, the consumer requests English or Spanish, and is sent a uniform set of text messages, customized only by due date (the Spanish track is a translation of the English track). An obscure internal document shows "Audience Insight" in a flowchart at the beginning of the message writing process, but elsewhere, only the non-profits and governmental agencies are credited with message development (30).
One outcome of this unified approach is that text4baby has no 'awareness' of a woman's current health status or other characteristics. For example, because the intervention is simply a clock that unwinds with a woman's pregnancy, it continues vaguely to urge her to attend prenatal care – a pestering annoyance to a woman who is diligently or even exuberantly participating in prenatal care. Similarly, if a woman does not smoke (or live with a smoker), it is not necessary to send her repeated messages about the dangers of tobacco products.
The visible marketing for text4baby is generic. Pictures of smiling minority women are labeled "one smart mom!" and the intervention is described as providing "pregnant women and new moms with information they need to take care of their health and give their babies the best possible start in life” (1). Once the messages start arriving, however, they clearly target low-income women who may have trouble affording food or prenatal care. For a woman not in this situation, the messages will seem inappropriate, which may lead women to unsubscribe. Furthermore, the match between low income or low SES women and LBW or preterm babies is far from exact – multiple other factors are at stake. Suggesting that a targeted audience is poor is a marketing faux-pas and may also alienate women who could, in fact, take advantage of free or reduced cost products and services. Finally, it is also problematic to use minority women's images to signal low SES.
For women of all socioeconomic statuses, the rate at which their in-person doctor visits meets their desires for information about specific topics is low, and many women report difficulty in asking questions of their doctor; there are also systemic mismatches in the information doctors provide versus the information women want (4). One possible viewpoint is that text4baby could act as an equalizer, inasmuch as all women who sign up have access to the same information provided. However, lower-class and middle-class women have different amounts of knowledge about pregnancy and childbirth, different desires for that knowledge, and different degrees of ease in accessing the knowledge they want (31). From this point of view, text4baby is no panacea; instead, it is an even less personal and perceptive provider, guessing at what information pregnant women desire. One of the justifications for the SMS-based format of text4baby is the overlap between women at risk of LBW and preterm birth and women carrying mobile phones, versus women with internet access (2). Online, however, women seeking pregnancy and childbirth information can search anonymously (without fear of a negative reaction from a doctor) and specifically (for topics of their choosing, no matter how arcane). Search technology exists on mobile phones, in services such as Google SMS, ChaCha, and Mosio. The irony of text4baby, then, is that despite its forward-looking use of mobile phones, its execution fails to exploit any of the communicative power of these devices.
Pregnancy cuts across all demographic characteristics except sex and age - and even there, women's ages span some 6 decades. Clearly, women's experiences of pregnancy are radically different, and even information that seems to be applicable to all women may be inappropriate for those with unusual characteristics or conditions. The non-personalized, static nature of text4baby is perhaps its biggest flaw, as to a degree it subsumes the other problems mentioned.

Improving text4baby

A re-envisioning of text4baby first and foremost calls on the inherently communicative nature of text messaging and makes this a cornerstone of the campaign. The old text4baby is basically an SMS page-a-day calendar which often violates the Gricean maxims about productive communication. The new and improved text4baby solicits participation from the women who sign up for it, values their input, and incorporates it back into the intervention. Drawing from models of prenatal care such as Centering Pregnancy, the new text4baby facilitates “learning from others, community building, attitude change and insight development, mutual support, and problem-solving skill development” (32). This participatory model would right many of text4baby's wrongs as regards gender and power, the individual's (in)ability to affect outcomes, and marketing.
An important aspect of text4baby is its broad list of “outreach partners”. Local and state organizations – health care facilities, public health non-profits and governmental offices, private companies – are all eligible to become partners. Currently, about 180 outreach partners are listed on the text4baby website (33). The extent of the outreach partners' activities is to market text4baby to their own audiences. This is a serious underutilization of the partners as a resource. Many of the partners would probably be able to dedicate some work hours to text4baby in an attempt to make the program more personalized, interactive, and participatory. The following suggestions for improvements to text4baby are roughly ordered from easiest to most difficult to integrate or actualize. Most would require additional time, money, and labor to implement, but the outreach partners in many cases would be an excellent resource to use in this way.
One of the first changes that should be made to the text4baby program is a culturally competent re-write of the messages that avoids sounding impersonal and authoritative. The program could employ a spokesperson, perhaps a minority woman, whose likeness and name could accompany the marketing and whose unique voice could shape the tone of the messages. Psychological research shows that reactance can be reduced by making the communicator more similar to the message recipient (34); this would also moderate the medical perspective of the messages.
A next step would be to collect more information from women at signup: demographic, behavioral, and preferential characteristics. This would allow the message track to be personalized based on risk factors and their uneven distribution. A woman could opt out of providing any or all of this information and receive an undifferentiated message track, alleviating privacy concerns. The preference questions could ask whether she wanted to receive more or fewer texts on certain topics. The system could be built so that she could update her preferences at any time. This improvement could be completely computerized and would require no additional manpower other than the upfront programming of customized message tracks.
Another logical step forward in improving text4baby is to call on the outreach partners to send women locally relevant texts. For example, instead of national hotline numbers, a woman would receive the phone number, address, and hours of a nearby food bank, car-seat installation demonstration, or parenting group meetup. Outreach partners could take advantage of the opportunity to guide women to local resources throughout her pregnancy. This familiarity would pay dividends throughout her children's early life.
A further extension of this idea would be to connect individual health care providers into the text4baby system and allow them (with the woman's permission, of course) to communicate with her via text message. Many extant SMS-based health education programs follow this model; research has shown, for example, that outpatient failure-to-attend rates drop with text message appointment reminders (35). Medication reminders or other individual information could also be conveyed to the woman by her specific health care contacts, while still integrated into the text4baby interface.
The communicative aspect of text4baby could be improved by building in allowances for the women to respond to the text4baby number via text message. While this addition would require major infrastructure changes, it addresses a number of the criticisms. It is also possible to implement such a change to a greater or lesser degree, depending on what is economically and practically feasible for the sponsoring organizations. For example, periodically, text messages could provide options to women, such as: “Text keywords back for info on breastfeeding: HOW-TO, BENEFITS, IS IT HARD, TIPS, SAVE (learn how much $ you'll save), MYTHS (does it make breasts sag?)” Such an option not only empowers a woman to do her own research very conveniently, it engages her curiosity and puts her in charge of her own care., which is a central tenant of some successful prenatal care interventions (32). This method could also be used to avoid sending all women messages about dozens of extremely rare pregnancy complications. Such an improvement requires little additional work beyond message generation and the upfront programming to send the correct texts in response.
A more intensive option along these same lines could give women the ability to send her own questions to the text4baby number. Pre-written responses could be chosen by a person (volunteer or employee) from an available set of common question-answer pairs; more unique questions could be answered individually or referred to a professional. If necessary for financial reasons, a woman who signed up could have an allotment of question-answer exchanges for the duration of her pregnancy.
Finally, text4baby could serve as a venue for women's thoughts, dreams, hopes, fears, and emotions about pregnancy and her coming baby by occasionally inviting women to send texts on these topics. An employee could read through the messages and choose a few representative or resonant texts to send back out to subscribers. Women uninterested in this aspect of the service could opt out. Including such an option would make a statement to women that their experiences of pregnancy are important, valued, and worth sharing – that this information is on par with the what the medical experts say.
These improvements to text4baby answer many of the criticisms above. The problematic issues of gender and power are ameliorated in all the instances where women reply with their own questions and experiences, as well as by the culturally competent rewriting with women's words in a peer-to-peer, rather than authoritarian tone. In other places, these messages make it possible to sideline the medical model of pregnancy while highlighting the pleasures, anticipation, fears, and emotional experiences that women have while pregnant, and inviting them to respond in kind. Acknowledging that childbearing is more than a medical event, and valuing women's knowledge of it, may even make women more receptive to instruction or warnings when they are offered.
The overly-individualistic focus of the current text4baby intervention is addressed in several ways. Changing the voice and tone of the messages helps to avoid blaming women for risk factors beyond their control. Greater personalization of the texts, following the provision of demographic and personal information, allows messages to focus on the issues that are likely most important to a given woman. For example, if a woman is a smoker, more consistent motivational messages about smoking cessation or reduction could be sent. A woman who marked that she was interested in healthy meal ideas could receive more texts about versatile, healthy, cost-effective meal planning. Partnering with local agencies or even a woman's personal care provider also gives the new text4baby a more realistic chance of positively affecting behavioral risk factors. A text telling a woman the hours of a food bank and the location of the nearest bus stop removes just one more barrier to her accessing this resource.
The poor marketing of the old text4baby is addressed in each instance where the new text4baby solicits input about the woman and determines which texts are sent as a result to that input. This is both a one-time step, when she is asked for demographic and preference information at signup, and integrated over the course of her pregnancy, as she texts back keywords for more information on a particular topic. Furthermore, the local organizations and health care affiliates should ideally themselves have conducted formative research with the populations they target. Their portion of the intervention can be tailored not just by local facilities and events, but also with awareness of local norms, customs, and preferences.

Conclusion

Text4baby is a major health education initiative unlike anything that has been done in the United States. The width and character of its partnerships are unique and are a great foundation for a broad, strong, and well-funded intervention. Unfortunately, the program fails to exploit the technology that is its primary innovation by eliminating the possibility of communication with the women it seeks to reach and by sending identical texts to each subscriber. These texts overlook the breadth of the experience of pregnancy and childbirth and perpetuate the authoritative, sometimes misogynistic medical model. From this medical vantage point, they also focus on the individual level of 'responsibility' for health, suggesting that women can and should be held responsible for their pregnancy outcomes – to an extent far greater than research suggests is possible. Predetermined information is delivered on a schedule, not in accordance with women's questions or needs. Given this structure, it will be disappointing but unsurprising if the program as it stands fails to improve maternal and child health as hoped. Here and elsewhere, varied suggestions have been made about possible improvements to the program. It will be interesting to watch it evolve as it gains subscribers and publicity.

References

(1) Text4baby. About. Alexandria, VA: Healthy Mothers, Healthy Babies, 2010. text4baby.org/about.
(2) Text4baby. Text4baby Message Track. Alexandria, VA: Healthy Mothers, Healthy Babies, 2010. Downloaded at text4baby.ning.com/notes/Toolkits.
(3) Jordan B. Birth in Four Cultures. 4th ed. Long Grove, IL: Waveland Press, 1993.
(4) Shapiro MC, Najman JM, Chang A, Keeping JD, Morrison J, Western JS. Information Control and the Exercise of Power in the Obstetrical Encounter. Social Science & Medicine 1983; 17:139-46.
(5) Rooks, J. Midwifery and Childbirth in America. Philadelphia, PA: Temple University Press, 1997.
(6) Woliver, L. The Political Geographies of Pregnancy. Urbana, IL: University of Illinois Press, 2002.
(7) Turkel, K. Women, Power, and Childbirth: A Case Study of a Free-Standing Birth Center. Westport, CT: Bergin & Garvey, 1995.
(8) Thomas LW. A critical feminist perspective of the health belief model: implications for nursing theory, research, practice, and education. Journal of Professional Nursing 1995; 11:246-252.
(9) Berk M, Schur C. The Effect of Fear on Access to Care Among Undocumented Latino Immigrants. Journal of Immigrant Health 2001; 3:151-6.
(10) Gamble V. Under the Shadow of Tuskegee: African Americans and Health Care. American Journal of Public Health 1997; 87:1773-8.
(11) Gregg R. Pregnancy in a High-Tech Age: Paradoxes of Choice. New York, NY: New York University Press, 1995.
(12) Freedman D. Testing Texts: Can health-related text messages be a boon to expectant mothers and babies? GW Today. 2010 Apr 12. Accessed at www.gwu.edu/explore/gwtoday/learningresearch/testingtexts.
(13) Institute of Medicine. Preventing Low Birth Weight: Summary. Washington, D.C.: Division of Health Promotion and Disease Prevention, Institute of Medicine, National Academy Press, 1985.
(14) Bernabé J, Soriano T, Albaladejo R, Juarranz M, Calle M, Martínez D, Domínguez-Rojas V. Risk factors for low birth weight: a review. European Journal of Obstetrics & Gynecology and Reproductive Biology 2004; 116:3-15.
(15) Goldenberg R, Culhane J, Iams J, Romero R. Epidemiology and Causes of Preterm Birth. Lancet 2008; 371:75-84.
(16) Goldenberg R, Cliver S, Mulvihill F, Hickey C, Hoffman H, Klerman L, Johnson M. Medical, psychosocial, and behavioral risk factors do not explain the increased risk for low birth weight among black women. American Journal of Obstetrics and Gynecology 1996; 175:1317-24.
(17) Healthy Mothers, Healthy Babies. Pregnancy Messages English 3.5.10.FD.xls. Personal Communication via email. Accessed 2010 Mar 22.
(18) Hobel C, Goldstein A, Barrett E. Psychosocial Stress and Pregnancy Outcome. Clinical Obstetrics and Gynecology 2008; 51:333-48.
(19) Huizink A, Mulder E, Robles de Medino P, Visser G, Buitelaar J. Is pregnancy anxiety a distinctive syndrome? Early Human Development 2004: 79: 81-91.
(20) Iams J, Romero R, Culhane J, Goldenberg R. Primary, secondary, and tertiary interventions to reduce the morbidity and mortality of preterm birth. Lancet 2008; 371:164-75.
(21) Lumley J, Chamberlain C, Dowswell T, Oliver S, Oakley L, Watson L. Interventions for promoting smoking cessation during
pregnancy. Cochrane Database of Systematic Reviews 2009, Issue 3.
(22) Abrams B, Altman S, Pickett K. Pregnancy weight gain: still controversial. The American Journal of Clinical Nutrition 2000;71(suppl):1233S–41S.
(23) Caulfield L, Witter F, Stoltzfus R. Determinants of Gestational Weight Gain Outside the Recommended Ranges Among Black and White Women. Obstetrics & Gynecology 1996; 87:760-6.
(24) Goldenberg, R, Culhane, J. Low Birth Weight in the United States. The American Journal of Clinical Nutrition. 2007; 85(suppl):584S–90S.
(25) Wadsworth M. Early Life (pp.44-63). In: Marmot M and Wilkinson R, ed. Social Determinants of Health. Oxford, UK: Oxford University Press, 1999.
(26) Hallam A. The effectiveness of interventions to address health inequalities in the early years: a review of relevant literature. Edinburgh, UK: Scottish Government, Health Analytical Services Division, 2008.
(27) Siegel M. Marketing public health – an opportunity for the public health practitioner (pp. 127-52). In: Siegel M, Doner L. Marketing Public Health: Strategies to Promote Social Change (2nd edition). Sudbury,MA: Jones & Bartlett Publishers, Inc., 2007.
(28) Albrecht T, Bryant C. Advances in Segmentation Modeling for Health Communication and Social Marketing Campaigns. Journal of Health Communication 1996; 1:65-80.
(29) Siegel M. The importance of formative research in public health campaigns: an example from the area of HIV prevention among gay men (pp. 73-8). In: Siegel M, Doner L. Marketing Public Health: Strategies to Promote Social Change, 2nd edition. Sudbury, MA: Jones and Bartlett Publishers, 2007.
(30) Remick A. Text4baby powerpoint. Presentation at University of Maryland. Downloaded at http://dietintern.nfsc.umd.edu/Remick_handout.pdf.
(31) Lazarus E. What Do Women Want?: Issues of Choice, Control, and Class in Pregnancy and Childbirth. Medical Anthropology Quarterly 1994; 8:24-46.
(32) Rising, S. Centering Pregnancy: An Interdisciplinary Model of Empowerment. Journal of Nurse-Midwifery 1998; 43:46-54.
(33) Text4baby. Partnership Opportunities. Alexandria, VA: Healthy Mothers, Healthy Babies, 2010. text4baby.org/partner.
(34) Silvia PJ. Deflecting reactance: the role of similarity in increasing compliance and reducing resistance. Basic and Applied Social Psychology 2005; 27:277-84.
(35) Downer S, Meara J, Da Costa A. Use of SMS text messaging to improve outpatient attendance. Medical Journal of Australia 2005; 183:366-8.