Filipino Explanatory Models of Cleft Lip with or without Cleft Palate

Abstract

Objective

To describe beliefs about the cause, prevention, and treatment of cleft lip with or without cleft palate (CL±P) among working class people and health care workers in the Philippines.

Design and Setting

The study was a focused ethnography and took place in the province of Negros Occidental, Philippines.

Participants

Using a stratified purposeful recruitment strategy, 80 individuals were selected for study. They were categorized into three groups: those having a CL±P or a child with a CL±P, those who neither had a CL±P nor child with CL±P, and health care workers.

Methods

Kleinman's explanatory models of illness theory were used to elicit Filipino explanations for CL±P. Data were collected through guided individual and group informant interviews and analyzed through content analysis.

Results

Filipinos reported that inheritance, falls, cravings, environmental factors, and God's will were causes for CL±P. Beliefs about prevention of CL±P included limiting the number of children, being careful not to fall, and avoiding environmental factors. It was found that general causal explanations for CL±P were not always congruent with personal causal explanations, and general/personal causal explanations for CL±P were not always congruent with prevention explanations.

Conclusion

By eliciting patients’ explanations for CL±P and comparing these with their own, clinicians can find commonalities between divergent explanations and use these as a starting point from which to improve health outcomes. Findings from this study will be used to guide the design of health campaigns regarding CL±P in the Philippines.

Keywords

cleft lip cleft palate culture ethnography explanatory models Philippines

Cleft lip, with or without cleft palate (CL±P), is among the most common birth defects, occurring in about 1 in 700 births worldwide, with noted variability in prevalence depending on geographic origin and ethnicity (Global Strategies, 2002). In general, Asian and Amerindian populations are thought to have the highest birth prevalence (as high as 1/500), with Caucasian populations having intermediate prevalence and African populations having the lowest at 1/2500 (Murray, 2002; Cooper et al., 2006). Exceptions to these summaries are found in populations from specific geographic areas and may be explained by founder effects and/or environmental triggers (Global Strategies, 2002; Murray, 2002). Indigenous people of the Philippines have a high birth prevalence, with CL±P occurring in about 1 in 500 births (Murray et al., 1997), making orofacial clefting a significant health problem in the Philippines.

Biomedcially, nonsyndromic CL±P (cleft occurring in isolation without other structural or cognitive anomalies) is believed to have a complex etiology, with gene and gene-environment interactions contributing to its cause. A substantial amount of research has been reported on genetic and environmental factors that contribute to CL±P in the Philippines (Murray et al., 1997; Munger et al., 2004; Schultz et al., 2004; Zucchero et al., 2004; Riley et al., 2007). However, there is little research on attitudes and beliefs about cause, treatment, and prevention of CL±P specific to Filipino culture. Some Filipino cultural explanations about the cause of orofacial clefting are not congruent with biomedical gene and gene-environment explanations of etiology. This could be an important problem because understanding cultural beliefs about the cause and prevention of disease is a primary concern when developing culturally appropriate educational and prevention programs (Tripp-Reimer et al., 2001; Thackeray and Neiger, 2003; Haidet et al., 2008).

Traditional Filipino beliefs about cause, prevention, and modes of treatment for diseases can be categorized as natural and supernatural explanations according to Human Relations Area Files (HRAF) domain of sickness (Anderson, 1996). (The HRAF domain of sickness provides information related to health categories on hundreds of societies. Beginning in the late 1930s, the HRAF has categorized cause of illness as natural or supernatural based on reviews of ethnographic studies.) According to HRAF classification, supernatural explanations of illness include witches, sorcerers, devils, dead relatives or friends, and objects as conduits of harmful influences. Talismans, amulets, crosses, or prayers may be used to ward off or drive away supernatural forces (McKenzie and Chrisman, 1977; Williams, 1978). Natural explanations for illness include beliefs that wind and exposure to temperature extremes can disturb the equilibrium between hot and cold within the body and bring about illness. Other natural explanations for illness include heredity, falling, eating certain foods, stepping on dirt, excessive drinking, poor diet, lack of sleep, the moon, night air, and exposure to sick people (Guthrie et al., 1990; Nichter and Nichter, 1996; Miranda et al., 1998; Yamada et al., 1999; Galanti, 2000; Williams et al., 2000). Before seeking care from a medical doctor, some Filipinos may consult a traditional healer, hilot (traditional midwife), or folk doctor to treat illnesses or symptoms and be offered traditional Filipino remedies including herbs, teas, rubs, prayers, and rituals (McKenzie and Chrisman, 1977; Williams, 1978).

Only a limited number of published studies have examined traditional Filipino beliefs about the cause of birth defects; this includes studies concerning cleft lip and palate. One traditional Filipino belief is that a pregnant woman's food cravings should be satisfied, especially during the first trimester, to avoid harming the fetus; however, it is also believed that a baby will take on characteristics of the food craved and eaten by the mother (Miranda et al., 1998); in anthropology, this phenomenon is termed marking (Crocker and Crocker, 1970). Marking is believed to occur when a mother or father “marks” their unborn child in some visible and unnatural way. While theories of marking vary, most emphasize that the expectant parents, by looking at a certain person or object, transmit the visible characteristics to the susceptible fetus in a photographic process (Crocker and Croker, 1970).

Cheng (1990) provided a general overview of Asian American cultural beliefs on birth defects, with an emphasis on clefting, but her overview did not specifically document Filipino beliefs about the cause and prevention of orofacial clefting. Cheng did describe Asian beliefs of natural and supernatural phenomena causing disease and birth defects. For example, some Chinese believe that eating rabbit during pregnancy can lead to a harelip baby. People of the Buddhist faith believe that fate or bad karma causes illness, including deformity. Other Asian views maintain that disabling conditions could arise from retribution for wrongdoings, spells, and the imbalances of natural forces (Cheng, 1990). Because children with disabilities and deformities are viewed as punishments for sins, as monsters, or as foreign spirits, these children are more often abandoned or placed in special institutions (Cheng, 1990). The most specific reference to Filipinos’ beliefs concerning clefting was that both Filipino and Chinese cultures sometimes view children with a birth defect as a curse and ostracize them from society (Cheng, 1990). Recent studies have not examined Filipinos’ cultural attitudes and beliefs about cause, treatment, and prevention of orofacial cleft.

Explanatory models is a term used to describe the influence of culture and context on patient and health care provider perspectives about the illness experience (Haidet et al., 2008). The explanatory models of illness framework is based on the premise of the social construction of reality and was most fully developed by Arthur Kleinman and his colleagues as a means to define illness within a cultural context (Kleinman, 1978; Kleinman et al., 1978; Cohen et al., 1994). Kleinman contended that people construct their own explanatory models of illness (explanations for illness) based on their perceptions about the way they experience and cope with illness; he contended that explanations are situation dependent and shaped by an individual's understanding of biology as well as by psychological and sociocultural influences. Indeed, the way people define and cope with illness varies across ethnicity, class, and family. Examples of more recent research based on explanatory models of illness include studies of families living in California from varied socioeconomic, racial/ethnic, and linguistic backgrounds and their beliefs about attention deficit/hyperactivity disorder (ADHD) diagnosis in their children (Leslie et al., 2007); parental beliefs about ADHD in Goa, India (Wilcox et al., 2007); Spanish-speaking Hispanics diagnosed with diabetes and depression and their explanatory models and treatment experiences (Cabassa et al., 2008); Filipino Americans’ beliefs about the cause of hypertension (dela Cruz and Galang, 2008); rural Latino immigrant caregivers’ perception of their children's oral disease (Horton and Barker, 2008); and explanatory models of patients and general practitioners from the United Kingdom and The Netherlands for irritable bowel syndrome (Casiday et al., 2009a, 2009b).

While both biomedical and ethnomedical explanations provide a clinical reality (Kleinman, 1978; Haidet et al., 2008), divergence between the clinician's and patient's explanations of illness can impede progress toward improving health outcomes. Commonalities provide starting points from which to develop culturally innovative health campaigns and interventions for decreasing the prevalence of CL±P. The purpose of this study was to explore and describe current beliefs about the causes, prevention, and treatment of CL±P among working class Filipinos and health care workers before developing appropriate educational and prevention programs.

Methods

Design

The study was a focused ethnography in that it was topic oriented and targeted the subject of CL±P. Data were collected through focused informant interviews. To put the focused data into context within the broader Filipino culture, the first author drew on experiences, observations, and conversations that occurred during 11 visits to the Philippines (over the past 13 years). To develop a cultural understanding of the cause, treatment, and prevention of CL±P, Kleinman's explanatory models of illness theory were employed to elucidate participants’ explanations for CL±P. These data were gathered as part of a larger study exploring Filipinos’ beliefs regarding causes of CL±P and their consumption of prenatal vitamins (Daack-Hirsch, 2007). Institutional Review Boards at the University of Iowa in the United States and the H.O.P.E Foundation Inc. in the Philippines approved this project. Participants provided written informed consent.

Setting

We conducted the study in the province of Negros Occidental, Philippines. We collected data in four cities: Bacolod, Silay, Kabankalan, and Himamylan. Bacolod is the capital of Negros Occidental and is classified as a highly urbanized city because its population is greater than 200,000. The other three cities contain both rural and urban communities.

Participant Ascertainment

We used a stratified, purposeful recruitment strategy: a qualitative sampling method designed to maximize the heterogeneity of participants on the basis of characteristics that might differentiate the explanatory models. This type of sampling method is not statistically representative but rather is designed to generate a range of individual responses that are informationally representative of the larger community (Coyne, 1997; Sandelowski, 2000). Three groups of people were invited to participate in this study: individuals who either themselves have a CL±P or have a child with CL±P (cleft group), individuals who neither have a CL±P nor have children with CL±P (noncleft group), and local health care workers. We focused on working class Filipinos because they experience a greater risk for orofacial clefting compared with Filipinos from a higher socioeconomic class (Chung et al., 1987; Cembrano et al., 1995; Murray et al., 1997), yet the working class does not have the same access to treatment for cleft and often relies on charitable medical missions to surgically correct orofacial cleft (Phillips, 1999; Jones, 2005). They typically do not have access to secondary services related to care of people with orofacial clefting, such as speech therapy, dentistry, and psychological services. We included local health care workers in the study because, throughout the Philippines, they are often the primary connection between working class people and the health care delivery system (Ron, 1999).

For the purposes of this study, the term working class was defined as people employed in jobs with low wages and little or no benefits or who are unemployed (Elmer Social Science Dictionary, 2003). Such occupations could include, but were not limited to, manual labor, service and repair work, agriculture, street vendors, and public transportation work. For the purposes of this study, the term health care worker was defined as any person who engages in health and health-related work and is employed at a city health center, rural health unit, and/or a barangay health station. These health care workers have trained in programs at accredited governmental or nongovernmental organizations and are paid for, or voluntarily render, primary health care services (Magna Carta of Public Health Workers, 1992).

These three groups of people—cleft, noncleft, and health care workers—were targeted for enrollment because, hypothetically, their explanations (explanatory models) of CL±P might vary because of differences in their experience with people who have CL±P and their access to health care. Men and women between the ages of 18 and 40 years who could give written, voluntary, and informed consent for themselves were eligible to participate in the study. Participant ascertainment ceased once reported explanations for cause, prevention, and treatment of CL±P became redundant within each of the three groups of interviewees (Sandelowski, 1995).

H.O.P.E. Foundation staff identified and referred the cleft group participants to the research team. The H.O.P.E. Foundation Inc. (a nongovernmental organization) maintains a database of all people who have received CL±P surgery through medical missions in Negros.

Participants in the noncleft group and the local health care worker group were drawn from the same cities as the cleft group participants. Noncleft participants were either referred to us by health workers from various health centers or by the research team during “consultation day” at their local barangay (village or neighborhood) health offices. Consultation day is a weekly service in which a government health team holds a clinic at a local health office to provide free checkups and treatment. Local health centers serve as a natural gathering place for people living in the barangay. Once consultations were completed, we invited people to stay and be interviewed for our study. All participants were interviewed in the participant's home, local health office, or the H.O.P.E. Foundation Inc. office.

Interview Guide

The first author developed an interview guide based on explanatory models of illness theory (Appendix A). An explanatory model is typically constructed by asking people questions about their beliefs regarding five elements: (1) etiology, (2) onset of symptoms, (3) pathophysiology, (4) course of illness (sick role and severity of the disorder), and (5) treatment (Kleinman, 1978; Kleinman et al., 1978; Cohen et al., 1994). Because we used explanatory models of illness theory to explore a birth defect, questions were modified to fit the nonepisodic nature of birth defects. The first author developed guided, open-ended questions with probes to elicit individual explanations for CL±P. Questions were based on four elements, which represent the theoretical framework of explanatory models: (1) label (what people name CL±P), (2) etiology, (3) how and when CL±P forms, and (4) treatment and prevention. We also collected basic demographic information including address, date of birth, education, occupation, monthly income, civil status, and religion.

Data Collection

Ilonggo is the Filipino dialect spoken on Negros Occidental; however, the first author (S.D.-H.) does not speak Ilonggo, so collecting data required an oral back-translation process. The oral back-translation process involved three people: the first author and the two Filipino research nurses. Back translations started by having the first author ask a question from the questionnaire, in English. Then, one of the Filipino nurses asked the same question out loud in Ilonggo. Immediately, the second Filipino nurse verbally translated the question back to English, to allow the first author's use of English words to be modified, reflecting a more typical Filipino-English vernacular. This made translating English into Ilonggo easier for the Filipino nurses during the interviews and allowed the team to clarify the meaning of questions asked in English. Also, many participants could understand English, so it was important that questions were asked using a Filipino-English vernacular. Before changing an English word, the team discussed whether use of the different word altered the original meaning of the question. Interviews were conducted in English, Ilonggo, or a combination of the two languages, depending on the participant's preference. The first author conducted interviews in English; the second author (H.G.) conducted interviews in Ilonggo; both authors conducted interviews when both languages were used. In these interviews, S.D.-H. was the primary interviewer and H.G. acted as an interpreter. All interviews were audiotaped.

Interviews began by showing all participants an identical photograph of a Filipino child with a cleft lip and palate. Participants were asked to identify or label the cleft using Filipino words and descriptions and to name the cleft, with the thought that people may have different labels for cleft depending on what they believed caused it (Nichter and Nichter, 1996). The interview continued by asking all participants open-ended questions pertaining to each of the explanatory model elements described above. Only participants in the cleft group were asked what they specifically thought caused CL±P in themselves or their child in addition to what they thought caused CL±P in general and what had been done specifically for them or their child as treatment for clefting (Appendix A).

Data Analysis

The audiotapes were transcribed verbatim. To verify the interpretation that took place during the face-to-face interview, H.G. translated Ilonggo to English after transcription, and the recorded interpretation and written translation were compared. Transcriptions were also checked for accuracy by simultaneously listening to the recording and reading the transcript. Data were stored and analyzed using NVivo 2 software. Using content analysis, each interview and set of field notes were coded for broad topical categories and more descriptive subcategories according to the four modified elements representing explanatory models described above (Crabtree and Miller, 1999; Sandelowski, 2000). Next, to identify differences and similarities in attitudes and beliefs about the cause, prevention, and treatment of clefting within and across groups, data matrixes were constructed. NVivo 2 software was used to create matrixes. A matrix was constructed for each element–-label, etiology, treatment, and prevention—by listing all participant response categories and subcategories by participant group (cleft, noncleft, and health worker). Categories were summarized in narrative and tabulated for presentation (Angrosino, 2007; Gibbs, 2007).

Results

Participant Characteristics

We interviewed 80 people in 69 interviews. Eleven interviews with health care workers were conducted: three in groups and eight as individuals (n = 22). One group interview consisted of nine health care workers; the other two group interviews were with three and two health care workers. Although we did not intend to interview groups of people, it so happened that at three of the health centers, when we began to interview one of the health care workers, others naturally joined us, thus forming a group. Rather than discourage the group discussion and lose the opportunity to gather information, the health care workers were interviewed together, as they preferred. This type of group interview is called a natural group and is an interview strategy commonly used in ethnography (Coriel, 1995). The interviews with people in the cleft group (n = 27) and noncleft group (n = 31) were conducted as individual interviews, for a total of 58 interviews.

Table 1

Participant Demographic Characteristics Stratified by Interview Group

Demographics	Health Care Worker (n = 22)	Noncleft Group (n = 31)	Cleft Group (n = 27)	Total (n = 80)
Gender, n (%)
Female	22 (100)	23 (74)	20 (74)	65 (81)
Male	0	8 (26)	7 (26)	15 (19)
Age
Range	21 to 53	18 to 50	22 to 45	18 to 53
Mean (SD)	38(9.11)	30 (7.97)	34 (6.29)	33 (8.45)
Median	38	27	35	34
Education, n (%)
Elementary	0	4 (13)	7 (26)	11 (14)
Some high school	4 (18)	5 (16)	3 (11)	12 (15)
High school graduate	4 (18)	10 (32)	8 (30)	22 (28)
Vocational	0	2 (6)	4 (15)	6 (8)
Some college	6 (27)	4 (13)	2 (7)	12 (15)
College graduate	8 (36)	6 (19)	3 (11)	17 (21)
Income, n (%)
<6000 P^*	17 (77)	28 (90)	21 (78)	66 (83)
6000 to 10,000 P	2 (9)	1 (3)	4 (15)	7 (9)
11,000 to 20,000 P	3 (14)	2 (6)	2 (7)	7 (9)
Occupation, n (%)
Community health worker	15 (68)	0	0	15 (19)
Certified midwife	6 (26)	0	0	6 (8)
Nurse	1 (5)	0	0	1 (1)
Service	0	6 (19)	10 (37)	16 (20)
Domestic	0	16 (52)	10 (37)	26 (33)
Agriculture	0	2 (6)	3 (11)	5 (6)
Labor	0	0	1 (4)	1 (1)
Professional	0	4 (13)	1 (4)	5 (6)
None	0	3 (10)	2 (7)	5 (6)
Civil status, n (%)
Married	16 (73)	21 (68)	22 (81)	59 (74)
Single	3 (14)	5 (16)	2 (7)	10 (13)
Separated	1 (5)	0	0	1 (1)
Common law	0	5 (16)	3 (11)	8 (10)
Widow	1 (5)	0	0	1 (1)
No data	1 (5)	0	0	1 (1)
Religion, n (%)
Catholic	14 (64)	30 (97)	23 (85)	67 (84)
Baptist	2 (9)	0	1 (4)	3 (4)
Born-again Christian	2 (9)	1 (3)	1 (4)	4 (5)
Mormon	1 (5)	0	1 (4)	2 (3)
Protestant	1 (5)	0	0	1 (1)
Islam	1 (5)	0	0	1 (1)
Jehovah Witness	1 (5)	0	0	1 (1)
Agilipay	0	0	1 (4)	1 (1)

P=peso; In April 2005 US $15P50.

Table 1 provides a summary of the participants’ characteristics, stratified by participant group type (cleft, noncleft, and health care worker). Most (84%) were married or in common-law relationships. About 29% of participants had less than a high school education, 27% were high school graduates, and 44% had more than a high school education. The majority (93%) held jobs in service, labor, or agriculture, with 83% reporting monthly family incomes of less than 6000 pesos (∼$125.00 USD). Eighty-four percent reported their religious affiliation as Catholic. Participants ranged in age from 18 to 53 years (mean age, 33 years).

Participant Responses Based on Explanatory Models Elements

The results described below are categorized according to the four elements representing the explanatory models of illness framework: (1) label (the name people give to CL±P), (2) etiology, (3) how and when CL±P forms, and (4) treatment and prevention. Whether the health care workers were interviewed as part of a group or as individuals, their answers did not generate different categories relating to the cause, prevention, and treatment of CL±P compared with participant responses from the cleft and noncleft groups; therefore, the data compiled from all interviews with health care workers are presented together.

Table 2

Summary of Causal Categories Stratified by Interview Group^*

Cause of Clef ting Characteristics, n (%)	Cleft (n =27)	Noncleft (n = 31)	Health Care Worker (n = 11)^†	Total Responses
Inborn	2 (7)	1 (3)	0	3
God's will	4 (15)	2 (6)	0	6
Accident as an adult	0	1 (3)	0	1
Do not know	0	1 (3)	0	1
Cravings	8 (30)	4 (13)	0	12
Environmental factors	6 (22)	13 (42)	10 (91)	29
Force to the fetal face	11 (41)	13 (42)	5 (45)	29
Inheritance	10 (37)	16 (52)	7 (64)	33

Participants could have multiple responses.

†

Based on 11 interviews, three group interviews, and eight individual interviews.

Label

Universally, CL±P was termed sungi among Ilonggo speakers on Negros Occidental (n = 75), regardless of whether the participant was categorized as cleft, noncleft, or health worker. Only four people offered an alternative name for CL±P, in another dialect. In Cebuano and Tagalog dialects, CL±P is called bungi. We did not ask people to specifically label the cleft palate because most people did not notice that the child in the picture had a cleft palate. However, three people did label the cleft palate as punga, the Ilonggo word specific for cleft palate. One person labeled the cleft lip as sambil, also an Ilonggo word, meaning swelling (in this case, swelling of the lips).

Etiology: How and When Cleft Forms

Data for etiology and how and when cleft forms are presented together because participants’ explanations of how and when cleft forms were part of their explanations about the cause of cleft. Participants’ responses to the question “What causes clefting?” were put into eight broad categories (Table 2). The four most common causes given were inheritance, force to the fetal face, environmental factors, and cravings. Each of the more common explanations of cause of CL±P is described below in detail.

Inheritance was the most frequently given explanation for cause of CL±P. Ten people from the cleft group, 16 from the noncleft group, and seven responses from the health care worker group believed inheritance is a possible cause of CL±P. When asked how inheritance causes CL±P, participants talked about other relatives who had an orofacial cleft as a source of inheritance. They also described inheritance as being in the blood: “It came through the blood” or “It was present in a distant relative” and showed up now after a few generations of no one having an orofacial cleft. A vertical pattern of inheritance was cited, but when sibling pairs were affected, it was not always recognized as inheritance (see Appendix B, Box 1, for an example of a participant response).

The second most frequent explanation given for the cause of CL±P was force to the fetal face (n = 11 cleft group; n = 13 noncleft group; n = 5 health care worker group). When asked how and when CL±P formed due to force, participants explained that CL±P resulted from fetal positioning and force or pressure applied to the pregnant abdomen. Participants described the fetus in a fetal position, with her thumbs or fingers in or near the mouth, doing what is often referred to as thumb sucking. A sudden movement by the mother, such as slipping, falling, or being startled, or a constant pressure to the mother's abdomen caused the fetus to tear the face open. In the case of cleft of the lip only, participants described the thumbs or fingers setting outside the mouth and pushing into the oral cavity causing a cleft lip and sometimes all the way through tearing the palate too. Alternatively, they described the fetus sucking her thumbs or fingers and pulling them through the palate and lip. In the case of a bilateral cleft lip, they believed two digits in or near the mouth caused the cleft, while a unilateral cleft was caused by a single digit. Participants believed the face formed late in pregnancy–-after the fifth month–-with the face forming at 7 months being the most common (n = 8) response; they also believed that force or pressure could cause CL±P only after the face had formed (see Appendix B, Box 2, for examples of participants’ responses).

Slipping and falling were the most common types of external force reported (n = 21). Similarly, attempted abortions (n = 10) and pressure on the mother's abdomen (n = 4) were also described as putting force on the fetal face, causing it to tear open. While participants from both the cleft and noncleft group claimed an attempted abortion could cause cleft, no one from the cleft group stated specifically that an attempted abortion caused their child's cleft. Attempted abortions were described as those that use a catheter, abdominal massage, or medications. Participants described damage to the face from abortion methods, either those that directly contacted the fetal face (e.g., via a catheter) or those that they perceived shook the uterus (e.g., abortion induced by medication, such as ipecac, which causes cramping and vomiting). Folk medicines used to induce abortions are sold at the open market and induce severe abdominal cramping and vomiting. One woman told us that abortions could be induced by a vigorous massage performed by a traditional midwife (hilot). It is notable that health care workers did not claim attempted abortion can cause CL±P.

A number of the causes of cleft reported by participants were categorized by the research team as environmental factors because they were neither genetic factors nor related to pressure to the fetal face, fate, or cravings. Chickenpox, birth control pills, medications, smoking, vitamin use (too much or too little), stress, poor nutrition, alcohol use, infection, and pollution were all categorized as environmental exposures. This category was created based on biomedical explanations for cleft and may not be recognized by Filipinos. The 29 responses in the environmental category were collected from 20 participants/interviews. At least one environmental factor was reported as a cause of cleft by six participants from the cleft group, 13 participants from the noncleft group, and in 10 interviews from the health care worker group (Table 2). Participants did not elaborate on how or when CL±P happens when they named an environmental factor as a cause.

Cravings were given as the cause of CL±P by eight participants from the cleft group and four from the noncleft group. Health care workers did not claim cravings can cause CL±P. The period of prenatal craving was described by participants as the first 3 months of pregnancy. Participants believed that during the period of craving, if a mother or father saw and thought about an object and/or person, then the fetus is vulnerable to taking on characteristics of that person or thing. Participants said that after this initial 3-month period, cravings will not lead to CL±P. Participants in this study explained craving in two ways. First, a craving could be observing a person with a CL±P and thinking about them during the period of craving. Second, acquiring CL±P was explained as retribution for staring at, teasing, or reprimanding someone with a CL±P during the period of craving. Participants also described craving as “fondness for looking at a cleft person,” unable to get the image of a person who has CL±P off their mind, or the urge to tease a person who has CL±P (see Appendix B, Box 3, for examples of participants’ responses).

Beliefs about one cause of CL±P did not necessarily exclude others as possible causes of CL±P. Twelve participants (44%) in the cleft group gave discordant explanations for what caused CL±P in general and what specifically caused their child's cleft or their cleft. In the cleft group, the three most common explanations for what causes CL±P in general were force to the fetal face (n = 11), inheritance (n = 10), and cravings (n = 8). These three explanations were reported with about the same frequency. However, the frequency of responses to the follow-up question, “What caused your child's or your cleft?” differed. When asked what caused their child's cleft, inheritance was the most commonly offered explanation (n = 15), followed by force to the fetal face (n = 7), cravings (n = 5), and God's will (n = 4; Figure 1).

Prevention and Treatment

Participants identified 15 prevention methods for CL±P. The most frequently reported methods were family planning (n = 25), taking care not to fall or put pressure on the pregnant abdomen (n = 12), and taking vitamins during pregnancy (n = 8). Participants described family planning as the use of some method to prevent the birth of subsequent children.

Figure 1.

Discordant explanations for cause of CL±P among the cleft group participants.

Prevention methods that we considered were safeguards against environmental factors were condensed into one category: environmental prevention. These methods included alcohol and tobacco avoidance, exercise, a healthy diet, vitamins, regular checkups, and use of prescribed medicines only. (The first author created the broad category of environmental prevention, but it is notable that this category might not be recognized by Filipinos.) The participants talked about engaging in these prevention strategies during pregnancy but not before; on average, women in our study began their prenatal care about 3½ months after they missed their period (range, 1 to 9 months).

Condensing environmental prevention strategies into a single category allowed us to draw general conclusions about participants’ responses; these data are summarized in Table 3. The most common prevention strategy was family planning, followed by environmental prevention, then avoiding falls. Within interview groups, however, this pattern did not hold true. For example, although family planning was the most frequent prevention method cited by participants in both the cleft and noncleft groups, it was not mentioned by the health care workers. In the cleft and noncleft groups, participants who thought family planning was a way to prevent CL±P said that to prevent CL±P, they must prevent the birth of more children and made no distinction between a single occurrence of CL±P and recurrence. Furthermore, the belief in the preventative value of family planning was not limited to those who believed CL±P was inherited (see Appendix B, Box 4, for examples of participants’ responses).

Table 3

Summary of Prevention Categories Stratified by Interview Group^*

Characteristic, n (%)	Prevention Category
Characteristic, n (%)	Cleft (n = 27)	Noncleft (n = 31)	Health Care Worker (n = 11)^†	Total
More research on cause is needed first	1 (4)	0	0	1
Be careful in general	1 (4)	2 (6)	0	3
Do not overwork during pregnancy	0	3 (10)	1 (9)	4
There is nothing you can do to prevent cleft	1 (4)	3 (10)	0	4
I do not know	3 (11)	1 (3)	1 (9)	5
Do not mock, reprimand, or dwell on the image of a cleft person	3 (11)	2 (6)	0	5
Do not slip or fall; avoid pressure to the abdomen during pregnancy	4 (15)	7 (26)	1 (9)	12
Environmental prevention	3 (11)	11 (35)	3 (27)	17
Family planning	16 (59)	9 (29)	0	25

Participants could have multiple responses.

†

Based on 11 interviews, three group interviews and eight individual interviews.

Explanations for cause and prevention of CL±P were not necessarily congruent; 21 of 69 interviews contained ideas about prevention that did not necessarily agree with the cause the participant had reported. For example, three participants said that to prevent clefting, women should eat healthy diets during pregnancy, but these three participants gave different causes for CL±P: (1) attempted abortion, (2) God's will, (3) slipping and falling, and (4) inheritance. In addition, attempted abortion, God's will, and inheritance were given as causes of CL±P by participants who reported CL±P could be prevented if women avoided slipping and falling. Attempted abortion, God's will, slipping and falling, inheritance, pressure on the mother's abdomen, and teasing were all causes given by participants who reported family planning could prevent CL±P.

When asked what treatments are available for children with CL±P, all participants responded “surgery.” In addition to surgery, two mothers described taking extra time to feed their cleft children, as compared with their other children, as a treatment. No other special or extraordinary treatments for children with CL±P were described.

Discussion

The primary purpose of this study was to describe beliefs about the cause, prevention, and treatment of cleft lip with or without cleft palate (CL±P) among working class people and health care workers in the Philippines before developing appropriate educational and prevention programs. Using Kleinman's explanatory models of illness as a theoretical framework, we elicited a number of ethnomedical explanations for the cause and prevention of CL±P. These findings provide insights for clinicians and researchers about alternative explanations of orofacial clefting that may be offered by Filipino patients and their families.

Clinical Implications

Findings from this study are directly applicable to the care of Filipino children and families with orofacial clefting in many settings, including in the United States. The U.S. census taken in 2000 reported that 1.8 million people identified themselves as Filipino (United States Census Bureau, 2007). In 2000, roughly 1.4 million Filipinos living in the United States were born in the Philippines (United States Census Bureau, 2007). Furthermore, health care providers from all over the world participate in medical missions in the Philippines treating people with cleft lip and/or cleft palate.

When clinicians and patients explain the cause of an illness in very different ways, patients are more prone to noncompliance, dissatisfaction with their care, and missed diagnosis, and they also tend to receive inappropriate or poor care (Kleinman et al., 1978; Tripp-Reimer et al., 2001; Thackeray and Neiger, 2003; Andrulis and Brach, 2007; Haidet et al., 2008). We suggest that one way to negotiate divergent explanations for CL±P, or patient problems related to CL±P, is to explore their beliefs by asking the Kleinman questions (Kleinman, 1978; Andrulis and Brach, 2007): What do you think caused your illness or problem? How does the illness or problem happen? What kind of treatment do you think you should receive? What are the most important results you hope you receive from this treatment?

While one explanation for CL±P might be acceptable in one situation, another explanation might be relevant in the context of a different situation. Interestingly, we found that about half of the participants in the cleft group gave discordant explanations for what caused CL±P in general and what specifically caused their child's cleft or their cleft. We also found incongruence between explanations for cause of cleft and how to prevent cleft. Based on these findings, we suggest adding the following to the Kleinman questions: What do you think caused the illness or problem in your child (or yourself) specifically? How do you think the illness or problem can be prevented?

While the goal of elucidating the patient's explanations for CL±P is to help the patient make sense of his or her experience, it can be very time-consuming for health care providers. However, even in the research setting, where we had more time and were using interpreters, it took only between 20 and 30 minutes per interview to obtain this information. In the clinical setting, this may be time well spent as the conversation may lead to better rapport with new families and may minimize time spent later on issues related to misunderstandings such as noncompliance, dissatisfaction with care, missed diagnosis, and inappropriate or poor care.

A clinician should also consider his or her own explanation for CL±P and compare it to the patient's explanation. By listing problems from both the patient's and clinician's points of view, and noting any conflicting views, clinicians can identify hidden discrepancies. With this information, clinicians can find ways to carefully and simply offer biomedical explanations for the illness or problem, explaining discrepancies and pointing out when patient and clinician views are the same. When possible, clinicians should frame the discussion using the patient's terms, thereby negotiating a common view of the illness or problem and an agreed upon plan for treatment (Kleinman, 1978; Andrulis and Brach, 2007).

For example, inheritance was a common causal explanation of CL±P for Filipinos in this study. Participants recognized that a family history of CL±P is a risk factor for having a child who has CL±P. A positive family history is also a major risk factor in biomedical explanations for CL±P. When health care providers and public health educators discuss concepts of genetic risk factors with Filipinos, they might anchor their explanations in shared concepts of inheritance. The clinician could initiate this conversation by using the patient's own pedigree.

Like inheritance, force to the fetal face caused by falls and pressure to the pregnant abdomen was a pervasive explanation for CL±P. However, unlike inheritance, fall/pressure as an explanation for CL±P is not supported by biomedical studies. Falls and pressure to the abdomen pose another challenge in that people who believe this is a cause generally did not believe this could cause CL±P until later in the pregnancy. In this explanation, the fetus is at risk toward the end of pregnancy, a belief that is not congruent with the biomedical understanding that CL±P forms in the early stages of pregnancy (Prescott and Malcolm, 2002; Murray and Schutte, 2004). By identifying this disparity between the clinician's and patient's explanations of CL±P, we gain insight into why a woman in the early stages of pregnancy may be reluctant to take on prevention strategies, such as vitamins and reductions of environmental exposures.

In this study, another common causal explanation for CL±P given by the participants was what Filipinos referred to as “cravings.” Filipinos described craving as seeing, teasing, and/or reprimanding a person with CL±P. Central to these acts as a cause for clefting is their timing during pregnancy. Filipinos explained that craving caused cleft only when it occurred during what they called the “period of craving” (the first trimester of pregnancy). This concept of the period of craving offers common ground on which to frame educational materials on the causes of cleft because, from a biomedical point of view, this period is also the time when the fetus is susceptible to development of cleft (Prescott and Malcolm, 2002). Although their ideas about the causes of CL±P differ, both groups agree that the first trimester of pregnancy is a susceptible time for the fetus. This common ground can be used as a starting point from which to correct misconceptions about what the fetus is actually susceptible to. For example, in patient educational materials used in the Philippines, using the idea of a period of craving to frame the importance of tobacco and alcohol avoidance, vitamins, and good nutrition during the first trimester might increase the understanding that these factors play a role in the cause and prevention of CL±P. Thus, they may become more accepted as viable explanations for CL±P and incorporated into prevention methods.

Research Implications

To the best of our knowledge, this is the first study to apply Kleinman's explanatory models of illness as a framework to explore beliefs about the cause, prevention, and treatment of a birth defect. We found that by modifying the types of questions posed in Kleinman's explanatory models, we could elicit explanatory models regarding orofacial clefting. This may be an appropriate framework for use in other studies on ethnomedical explanations for birth defects. Our findings support Kleinman's theory that people construct their explanations based on their own experience and that explanations are situation dependent. In this study, participants operationalized these theories by providing contextual details about the situation they or others were in to cause the baby to have cleft. We also found discordant explanations for what people thought caused cleft in general versus what they thought caused cleft in their child, and we found that some participants gave discordant explanations about cause and prevention, suggesting that particular explanations about cause and prevention are not mutually exclusive but situation dependent.

To develop patient-centered treatment plans, there is a need for patients and health care providers to develop an understanding of each other's perspectives about the illness. However, assessing the level of concordance between provider-patient explanations for illness is time-consuming. Quantitative instruments may aid health care providers to more efficiently evaluate provider-patient differences. Haidet et al. (2008) developed the CONNECT instrument for measuring the degree to which patients’ and providers’ perspectives differ. Their long-term goal is to develop an instrument that has applicability across a variety of medical conditions and contexts. Findings from our study could inform modifications to the CONNECT instrument for use with families of children with craniofacial birth defects. The CONNECT instrument does not elicit patients’ specific explanations for cause, treatment, and prevention of an illness. Full elucidation of a patient's perspectives about their illness experience may require an in-depth discussion.

Limitations

A potential limitation to this study is that using translation/interpretive methods to convert research findings from one language to another imposes bias on data sets. This limitation was minimized by the cross-checking procedures (Johnson, 1993) and an experienced research team. Cross-checking the data involved note taking, debriefing, and double translation. After each interview, the research team debriefed by reviewing notes and discussing whether they agreed on the concepts presented and whether these concepts were consistent with what was already heard or known. The discussion provided an opportunity to ask the participant to clarify content at the time of the interview. A double translation process was also used to cross-check the data. The interviews conducted by the lead author involved one of the Filipino nurses providing real-time English and Ilonggo translation during the interviews. To verify conceptual accuracy of the translation process, Ilonggo was translated to English a second time during transcription.

Finally, the use of a multicultural research team minimized the bias due to translation/interpretive methods. Ideally, working with translators/interpreters who are from the culture and have knowledge in the research process and goals of a particular project promotes collaboration and produces a mutual understanding of the data (Temple and Young, 2004; Larkin et al., 2007).

Conclusion

This unique study describes Filipinos’ explanations of cause, prevention, and treatment (explanatory models) for CL±P. We found that by modifying the types of questions posed in Kleinman's explanatory models, we could effectively elicit explanatory models regarding a birth defect. We also propose that by modifying the types of questions posed in Kleinman's explanatory models, health care providers can ask patients the right kind of questions about the cause and prevention of illness; this approach can be used to promote a more accurate representation of the patient's illness experience. By eliciting patients’ explanatory models and comparing them with the biomedical models, clinicians can negotiate between the divergent explanations to improve health outcomes. Patients’ individual explanatory models may explain their behaviors in terms of how they seek health care information, whom they seek information from, and how they choose between prevention and treatment alternatives. By understanding the patient's explanatory models, the clinician can anticipate which patients may or may not adhere to treatment plans. Moreover, the findings from this study can guide the design of health campaigns in the Philippines regarding orofacial clefting, thus making informational brochures and programs more effective.

Footnotes

Acknowledgments

We gratefully acknowledge the participants of this study for sharing their time and knowledge with us. We thank Ma. Buena Nepomuceno, May Ann Estosa, and Gino Gamboa for being superb research team members. Organizational and technical support in the Philippines was provided by the board and staff of the H.O.P.E. Foundation Inc. We have greatly benefited from the dedicated mentoring of Drs. Toni Tripp-Reimer, Kathy Clark, Jeff Murray, Bernard Sorofman, and Janet Williams. This research was supported by National Institute of Health (NIH) grant F31 NR008655, International Society of Nurses in Genetics (ISONG) research award, and Sigma Theta Tau Gamma Chapter Research Award.

Appendix A: Interview Guide

Begin by showing a picture of a child with cleft and asking the participant to name the birth defect (cleft) using his or her own words.

Appendix B: Examples of Participant Responses

Box 4

Prevention

Box 4

Prevention

References

Anderson

. The Aims and Achievements of Medical Anthropology: Magic, Science, and Health. Orlando: Harcourt Brace College Publishers; 1996.

Andrulis

D.P.

, Brach

. Integrating literacy, culture and language to improve health care quality in diverse populations. Am J Health Behav. 2007; 31(suppl 1): S122-S133.

Angrosino

. Doing ethnographic and observational research (book 3 of the SAGE Qualitative Research Kit). London: Sage; 2007.

Cabassa

L.J.

, Hansen

M.C.

, Palinkas

L.A.

, Ell

. Azucar y nervios: explanatory models and treatment experiences of Hispanics with diabetes and depression. Soc Sci Med. 2008; 66: 2413-2424.

Casiday

R.E.

, Hungin

A.P.S.

, Cornford

C.S.

, de Wit

N.J.

, Blell

. GPs' explanatory models for irritable bowel syndrome: a mismatch with patient models? Fam Pract. 2009a; 26: 34-39.

Casiday

R.E.

, Hungin

A.P.S.

, Cornford

C.S.

, de Wit

N.J.

, Blell

. Patients' explanatory models for irritable bowel syndrome: symptoms and treatment more important than explaining aetiology. Fam Pract. 2009b; 26: 40-47.

Cembrano

J.R.J.

, de Vera

J.S.

, Joaquino

J.B.

, Ng

E.F.

, Tongson

T.L.

, Manalo

P.D.

, Fernandez

G.C.

, Encarnación

R.C.

. Familial risk of recurrence of clefts of the lip and palate. PJSS. 1995; 50: 37-40.

Cheng

L.R.

. Asian-American cultural perspectives on birth defects: focus on cleft palate. Cleft Palate J. 1990; 27: 294-300.

Chung

C.S.

, Mi

M.P.

, Beechert

A.M.

. Genetic epidemiology of cleft lip with or without cleft palate in the population of Hawaii. Gen Epi. 1987; 4: 415-423.

10.

Cohen

M.Z.

, Tripp-Reimer

, Smith

, Sorofman

, Lively

. Explanatory models of diabetes: patient practitioner variation. Soc Sci Med. 1994; 38: 59-66.

11.

Cooper

M.E.

, Ratay

J.S.

, Marazita

M.L.

. Asian oral-facial cleft prevalence. Cleft Palate Craniofac J. 2006; 43: 580-589.

12.

Coriel

. Group interview methods in community health research. Med Anthropol. 1995; 16: 195-210.

13.

Coyne

I.T.

. Sampling in qualitative research: purposeful and theoretical sampling: merging or clear boundaries? J Adv Nurs. 1997; 26: 623-630.

14.

Crabtree

, Miller

. Using codes and code manuals: a template organizing style of interpretation. In: Crabtree

, Miller

, eds. Doing Qualitative Research. 2nd ed. Thousand Oaks, CA: Sage; 1999: 163-177.

15.

Crocker

E.C.

, Crocker

. Some implications of superstitions and fold beliefs for counseling parents of children with cleft lip and cleft palate. Cleft Palate J. 1970; 7: 124-128.

16.

Daack-Hirsch

S.E.

. Context for Filipino Community Based Orofacial Cleft Prevention. Iowa City, IA: University of Iowa; 2007, Dissertation.

17.

dela Cruz

F.A.

, Galang

C.B.

. The illness beliefs, perceptions, and practices of Filipino Americans with hypertension. J Am Acad Nurse Pract. 2008; 20: 118-127.

18.

Elmer Social Science Dictionary. Available at: www.elissetche.org/dico/R.htm. Accessed June 6, 2006.

19.

Galanti

. Culture and medicine. Filipino attitudes toward pain medication: a lesson in cross-cultural care. West J Med. 2000; 173: 278-279.

20.

Gibbs

G.R.

. Analyzing Qualitative Data (Book 6 of the SAGE Qualitative Research Kit). London: Sage; 2007.

21.

Global Strategies to Reduce the Health-Care Burden of Craniofacial Anomalies: Report of the WHO Meetings on International Collaborative Research on Craniofacial Anomalies. Geneva, Switzerland, November 2000; Park City, Utah, May 2001. World Health Organization, 2002.

22.

Guthrie

H.A.

, Guthrie

G.M.

, Fernandez

, Barba

C.V.

. Nutritional intervention: a key to child survival. Asia Pac J Public Health. 1990; 4: 53-58.

23.

Haidet

, O'Malley

K.J.

, Sharf

B.F.

, Gladney

A.P.

, Greisinger

A.J.

, Street

R.L.

. Characterizing explanatory models of illness in healthcare: development and validation of the CONNECT instrument. Patient Educ Couns. 2008; 73: 232-239.

24.

Horton

, Barker

J.C.

. Rural Latino immigrant caregivers' conceptions of their children's oral disease. J Public Health Dent. 2008; 68: 22-29.

25.

Johnson

R.B.

. Examining the validity of qualitative research. Education. 1993; 118: 282-292.

26.

Jones

L.D.

. Operation Smile—changing lives one smile at a time. Arch Pediatr Adolesc Med. 2005; 159: 211.

27.

Kleinman

. Clinical relevance of anthropological and cross-cultural research: Concepts and strategies. Am J Psychiatry. 1978; 135: 427-431.

28.

Kleinman

, Eisenberg

, Good

. Culture, illness and care: clinical lessons from anthropologic and cross-cultural research. Ann Intern Med. 1978; 88: 251-258.

29.

Larkin

P.J.

, Dierckx de Casterle

, Schotsmans

. Multilingual translation issues in qualitative research: reflections on a metaphorical process. Qual Health Res. 2007; 17: 468-476.

30.

Leslie

L.K.

, Plemmons

, Monn

A.R.

, Palinkas

L.A.

. Investigating ADHD treatment trajectories: listening to families' stories about medication use. J Dev Behav Pediatr. 2007; 28: 179-188.

31.

Magna Carta of Public Health Workers. Republic Act No.7305. Date enacted: March 26, 1992.

32.

McKenzie

J.L.

, Chrisman

N.J.

. Healing herbs, gods, and magic. Nurs Outlook. 1977; 25: 326-329.

33.

Miranda

B.F.

, McBride

M.R.

, Spangler

. Filipino-Americans. In: Purnell

L.D.

, Paulanka

B.J.

, eds. Transcultural Health Care: A Culturally Competent Approach. Philadelphia: F.A. Davis; 1998: 245-272.

34.

Munger

, Sauberlich

, Corcoran

, Nepomuceno

, Daack-Hirsch

, Solon

F.S.

. Maternal vitamin B-6 and folate status and risk of oral facial birth defects in the Philippines. Birth Defects Res A Clin Mol Teratol. 2004; 70: 464-471.

35.

Murray

J.C.

. Gene/environment causes of cleft lip and/or palate. Clin Genet. 2002; 61: 248-256.

36.

Murray

J.C.

, Daack-Hirsch

, Buetow

, Munger

, Espina

, Paglinawan

, Villanueva

, Rary

, Magee

. Clinical and epidemiologic studies of cleft lip and palate in the Philippines. Cleft Palate Craniofac J. 1997; 34: 7-10.

37.

Murray

J.C.

, Schutte

B.C.

. Cleft palate: players, pathways, and pursuits. J Clin Invest. 2004; 113: 1676-1678.

38.

Nichter

, Nichter

. Acute Respiratory Illness: Popular Health Culture and Mother's Knowledge in the Philippines. Vol. 3. Dordrecht, the Netherlands: Kluwer Academic; 1996.

39.

Phillips

D.F.

. Operation Smile volunteers travel far to transform lives. JAMA. 1999; 281: 597-598.

40.

Prescott

N.J.

, Malcolm

. Folate and the face: evaluating the evidence for the influence of folate genes on craniofacial development. Cleft Palate Craniofac J. 2002; 39: 327-331.

41.

Riley

B.M.

, Schultz

R.E.

, Cooper

M.E.

, Goldstein-McHenry

, Daack-Hirsch

, Lee

K.T.

, Dragan

, Vieira

A.R.

, Lidral

A.C.

, Marazita

M.L.

. A genome-wide linkage scan for cleft lip and cleft palate identifies a novel locus on 8p11–23. Am J Med Genet A. 2007; 143: 846-852.

42.

Ron

. NGOs in community health insurance schemes: examples from Guatemala and the Philippines. Soc Sci Med. 1999; 48: 939-950.

43.

Sandelowski

. Sample size in qualitative research. Res Nurs Health. 1995; 18: 179-183.

44.

Sandelowski

. Combining qualitative and quantitative sampling, data collection and analysis techniques in mixed-method studies. Res Nurs Health. 2000; 23: 246-255.

45.

Schultz

R.E.

, Cooper

M.E.

, Daack-Hirsch

, Shi

, Nepomucena

, Graf

K.A.

, O'Brien

E.K.

, O'Brien

S.E.

, Marazita

M.L.

, Murray

J.C.

. Targeted scan of fifteen regions for nonsyndromic cleft lip and palate in Filipino families. Am J Med Genet A. 2004; 125: 17-22.

46.

Temple

, Young

. Qualitative research and translation dilemmas. Qual Res. 2004; 4: 161-178.

47.

Thackeray

, Neiger

B.I.

. Use of social marketing to develop culturally innovative diabetes intervention. Diabetes Spectrum. 2003; 16: 15-20.

48.

Tripp-Reimer

, Choi

, Skemp Kelly

, Enslein

. Cultural barriers to care: inverting the problem. Diabetes Spectrum. 2001; 14: 13-22.

49.

United States Census Bureau. United States census 2000. Available at: http://www.census.gov/. Accessed April 5, 2007.

50.

Wilcox

C.E.

, Washburn

, Patel

. Seeking help for attention deficit hyperactivity disorder in developing countries: a study of parental explanatory models in Goa, India. Soc Sci Med. 2007; 64: 1600-1610.

51.

Williams

P.D.

. Disease concepts and treatment in the Philippines: reviews of the literature. Maternal Child Nurs J. 1978; 8: 95-102.

52.

Williams

P.D.

, Williams

A.R.

, Lopez

, Tayko

N.P.

. Mothers' developmental expectations for young children in the Philippines. Int J Nurs Stud. 2000; 37: 291-301.

53.

Yamada

, Caballero

, Matsunaga

, Agustin

, Magana

. Attitudes regarding tuberculosis in immigrants from the Philippines to the United States. Fam Med. 1999; 31: 477-482.

54.

Zucchero

T.M.

, Cooper

M.E.

, Maher

B.S.

, Daack-Hirsch

, Nepomuceno

, Ribeiro

, Caprau

, Christensen

, Suzuki

, Machida

. Interferon regulatory factor 6 (IRF6) gene variants and the risk of isolated cleft lip or palate. N Engl J Med. 2004; 351: 769-780.