Portraying Islam and Muslims in MEDLINE: A content analysis Social Science & Medicine 65 (2007) 2425–2439 Portraying Islam and Muslims in MEDLINE: A content analysis Lance Daniel Laird a,� , Justine de Marrais b , Linda L. Barnes a a Boston University School of Medicine, Boston, MA, USA b Boston Medical Center, MA, USA Available online 4 September 2007 Abstract The growing number and diversity of Muslims in the United States and Western Europe challenge clinicians and researchers to understand this population’s perspectives and experiences regarding health and biomedicine. For information about Muslim patient populations, clinicians and researchers routinely consult medical literature. To examine how this literature portrays Muslims, we conducted an ethnographic content analysis of 2342 OVID MEDLINE-indexed abstracts from 1966 through August 2005, derived from a Boolean search for ‘‘islam or muslim or muslims.’’ Manifest (explicitly stated) themes included Muslim religious practices, Islamic law and ethics, history of Islamic medicine, public health, social medicine, and cultural competence. Latent (underlying) themes implied that being an observant Muslim poses health risks; Muslims are negatively affected by tradition, and should adopt modernity; and that ‘‘Islam’’ is a problem for biomedical healthcare delivery. A countervailing latent theme implies that being Muslim may promote good health. We discuss ambiguities in uses of the term ‘‘Muslim;’’ implications of Muslim practices for health management and healthcare delivery; and ways in which MEDLINE-indexed literature intersects with orientalist and colonialist discourse about religious Others. Such intersections highlight connections with potential structural inequalities in healthcare delivery to Muslim patients. r 2007 Elsevier Ltd. All rights reserved. Keywords: Ethnographic content analysis; Islam; Muslims; Religion; MEDLINE Introduction Popular media and political rhetoric devote grow- ing attention to Muslims and Islam, while Muslim populations in the US and Western Europe continue to rise. The hostility and suspicion facing Muslims following September 11, 2001, and Western military interventions in Muslim-majority countries, challenge clinicians, researchers, and policymakers to better understand diverse Muslim perspectives, experiences, and ways of practicing Islam relative to health and biomedicine. Researchers seeking such information regularly consult medical literature. Yet this literature is not produced in a vacuum. Media reports, political rhetoric, and legislative action regularly promote negative stereotypes that characterize Muslims as an out-group (Runnymede Trust et al., 2004). Such portraits both reflect and filter into Western policy interests in the Middle East (Gerges, 2003; McAlister, 2005) and affect Muslims in Western societies (Shaheen, 2003). ARTICLE IN PRESS www.elsevier.com/locate/socscimed 0277-9536/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2007.07.029 � Corresponding author. Tel.: +1 617 414 3660; fax: +1 617 414 3679. E-mail addresses: lance.laird@bmc.org (L.D. Laird), justine.demarrais@bmc.org (J. de Marrais), linda.barnes@bmc.org (L.L. Barnes). www.elsevier.com/locate/socscimed dx.doi.org/10.1016/j.socscimed.2007.07.029 mailto:lance.laird@bmc.org mailto:justine.demarrais@bmc.org mailto:linda.barnes@bmc.org US polls indicate that from one-quarter (Council on American-Islamic Relations, 2004, 2006a) to one-half (Dean & Fears, 2006) of Americans hold negative views of Islam and Muslims. Islamophobia and anti-Muslim hate crimes have increased sig- nificantly in the US (Council on American-Islamic Relations, 2006b) and in Europe (Allen & Nielsen, 2002; Runnymede Trust et al., 2004). Islamophobic discourse portrays Islam as monolithic and threa- tening, Muslims as using their religion to gain advantages, and Muslim cultures as significantly different from other cultures. Anti-immigrant atti- tudes frequently accompany such messages (Com- mission on British Muslims and Islamophobia, 1997), building on a history of orientalist represen- tations of Muslim societies as deficient or deviant versions of a modern, rational, Europe (Said, 1978). Medical literature includes studies involving ‘‘Islam,’’ a religious tradition, and ‘‘Muslims,’’ the people who follow this tradition. We hypothesized that a systematic content analysis of MEDLINE- indexed abstracts might illuminate meanings as- signed to both terms in health-related contexts and in scientific writing. To that end, we examined such abstracts from over a 40-year period, looking for recurring themes or ‘‘frames.’’ We asked, which themes are manifest (explicit), and which, latent (underlying, and perhaps unintended)? We won- dered whether trends in Western media and popular culture about Muslims might enter biomedical discourse. If so, how, and what might be their implications for healthcare settings? Medical anthropologists argue that the ‘‘biome- dical gaze’’ (Foucault, 1973; Lock & Gordon, 1988) pathologizes and Otherizes its objects. Yet does this tendency adequately explain possible negative re- presentations of Muslims and Islam? One study of ‘‘Chinese people’’ in North American social work literature, for instance, suggests that professional service providers often replicate cultural biases that ‘‘otherize’’ and ‘‘essentialize’’ ethnic minorities (Tsang, 2001). When biomedical pathologizing tendencies intersect with discourses about margin- alized minorities, they can reproduce cultural prejudices and reinforce disparities in health out- comes and healthcare access. We propose, therefore, not to demonstrate that Muslims are treated uniquely in medical literature, but to explicate how representations of Muslims and Islam in medical literature may intersect with broader cultural discourses within which readers of these texts are situated. Methods Ethnographic content analysis of texts (Altheide, 1996) derives from a combination of literary content analysis (Downe-Wamboldt, 1992) and ethno- graphic interview analysis (Weston et al., 2001). For example, ethnographic content analyses of cancer-related stories in mainstream print media have shown how dominant discourses, disparity of coverage, and the lack of ethnically specific infor- mation shape public awareness and influence healthcare delivery (Clarke, 2004; Clarke & Everest, 2006; Hoffman-Goetz & Friedman, 2005). Ethnographic content analysis aims to define and describe patterns within a set of documents. Through reiterated comparison and contrasting of data, researchers identify themes that shape the discourse related to their inquiry. Subsequent analysis differentiates between manifest and latent themes (Altheide, 2002), and between explicit and implicit meanings. One cannot infer the intentions of the authors, which in this case we presumed to be beneficent and compassionate. When citing exam- ples, therefore, we comment only on subtle, textured meanings within the abstracts before us. While implicit meanings may be unintentional, they invariably interact with textual and cultural con- texts that shape readers’ attitudes in ways that may generate or reinforce bias. This is potentially significant, because unexamined provider biases are known to affect access and treatment (Burgess, Fu, & van Ryn, 2004; Schulman et al., 1999). Sample selection A Boolean search in OVID MEDLINE (1966- present), on August 26, 2005, using the keywords, ‘‘islam or muslim or muslims’’ produced 2342 abstracts, after excluding duplicates and non- pertinent Boolean hits (e.g., ‘‘Islam’’ as an author’s surname). Abstracts are more available than articles to most clinicians, and limn the broad, initial portrait we analyze here, so we limited the study to these results. (Following completion of our analysis, we discovered that inclusion of the alternate spelling ‘‘Moslem’’ would have yielded another 89 results.) Although most MEDLINE users filter search results, we could not predict individual search para- meters, and so included all results to avoid selection bias (del Pozo & Fins, 2005). We did not limit by article type, assuming that full representation of any ARTICLE IN PRESS L.D. Laird et al. / Social Science & Medicine 65 (2007) 2425–24392426 https://isiarticles.com/article/1240