A critical review of the literature

Copyright © 2010 American Dental Association. All rights reserved. Reprinted by permission. on Sptem er 1, 2010 jadada.org D ow nladed from lence of asthma has been increasing since the 1980s across all age, sex and racial groups. Asthma appears to be, after dental caries, the most prevalent chronic childhood disease in the United States. In the 1980s, asthma was associated with 500,000 hospitalizations and 6.5 million office visits to physicians annually. Based on national survey results published in the late 2000s, 7 percent of adults and 9 percent of children in the United States have asthma, necessitating 10.6 million office visits to physicians. In 2006, 444,000 hospital discharges involved patients whose first-listed diagnosis was asthma. Caries remains an important health problem across all age groups in the United States, but it is documented better in children than in adults. Mixed evaluations have resulted from looking at trends in terms of considering a glass half full or half empty. Namely, dental caries prevalence for many children has decreased in the past few decades, but although the proportion of cariesfree children has increased, the reduction in caries burden has not been shared equally among children. If the caries experience in primary and permanent teeth is considered jointly, the proportions of caries-free children continue to decrease in adolescence: data from the Third National Health and Nutrition Examination Survey, conducted from 1999 through 2004, indicated that 59 percent of 12to 19-year-olds have had dental caries in their permanent teeth and 23 percent have untreated decay. Our rationale for undertaking this review of the literature is that people with asthma may become more susceptible to caries directly (through biological mechanisms), indirectly (through pharmacological mechanisms) or both. The 2000 report of the U.S. surgeon general titled Oral Health in America indicated that asthma and caries, together with learning difficulties and social problems, are correlated closely enough with social disadvantage to be designated sentinel diseases. Questions that remain to be addressed are whether a link exists between asthma and caries and if so, what its nature is. Is the association between increased severity and frequency of asthma conditions and increased experience of caries a direct relationship? Is this relationship pharmacological or biophysiological in nature? Are asthma and caries separate sequelae of poor access to different health care services? Or are all of these the case? We present a semistructured review of the scientific and professional literature in which we attempted to ascertain the strength of the evidence supporting an association between asthma and caries. Because many reports pertain to children, we will emphasize, but not limit the review to, younger age groups. MATERIALS AND METHODS We conducted a general literature review with substantial structured review components, rather than a definitive, systematic review. Sources. In March 2010, two oral epidemiologists (G.M. and O.L.) searched Medline for articles published from 1976 through 2010 by using the Ovid Web Gateway. The search strategy included the National Library of Medicine Medical Subject Headings (MeSH) terms (“asthma” and [“dental caries” or “dental caries susceptibility”]). They limited the search to studies involving human participants and published in English. We designed this search strategy to ensure high sensitivity initially, rather than high specificity. Although we did not contact editors or authors (with one exception), we undertook a hand-search review of the list of references in every article we identified. Study selection. Two oral epidemiologists (G.M. and J.D.S.) reviewed the list of titles and abstracts for articles generated by the search engine to identify those that appeared to be research reports addressing the structured review question. They explicitly eliminated conference proceedings and abstracts, editorials, opinion pieces and unpublished studies. Owing to the general nature of the question (association) and the limited number of research studies available, 1062 JADA 141(9) http://jada.ada.org September 2010 C O V E R S T O R Y ABBREVIATION KEY. BOP: Bleeding on probing. dfs: Decayed or filled surfaces (primary teeth). DFS: Decayed or filled surfaces (permanent teeth). defs: Decayed, extracted or indicated for extraction, or filled surfaces (primary teeth). deft: Decayed, extracted or indicated for extraction, or filled (primary) teeth. dft: Decayed or filled (primary) teeth. dmfs: Decayed, missing or filled surfaces (primary teeth). DMFS: Decayed, missing or filled surfaces (permanent teeth). dmft: Decayed, missing or filled (primary) teeth. DMFT: Decayed, missing or filled (permanent) teeth. DS: Decayed surfaces (permanent teeth). ENO: Exhaled nitric oxide. FS: Filled surfaces (permanent teeth). MeSH: Medical Subject Headings. NA: Not applicable. NHANES III: Third National Health and Nutrition Examination Survey. SES: Socioeconomic status. SGH: Salivary gland hypofunction. WHO: World Health Organization. Copyright © 2010 American Dental Association. All rights reserved. Reprinted by permission. on Sptem er 1, 2010 jadada.org D ow nladed from they included most evidence levels in the review— that is, cross-sectional, case-control and cohort studies, as well as clinical trials. Study selection criteria dictated that an article describe a clinical, epidemiologic, microbial or survey investigation of dental caries in relation to asthma-related disease presentations and that it present a clinical measurement of caries (such as decayed, missing or filled [permanent] teeth [DMFT]; decayed, extracted or indicated for extraction, or filled surfaces [defs] of primary teeth; and so on) as the outcome variable. We read the articles that appeared to qualify for inclusion. Evaluation. We abstracted the articles, evaluated the quality of the measures used in the studies and reviewed the studies’ designs and analytic methods. We also sought to identify variables that indexed socioeconomic status, social class or any social position measure; we noted participant recruitment, measures of caries and asthma, and statistical analysis approaches. We discussed and resolved differences in interpretation; our review was not masked. For studies in which investigators had performed multiple bivariate tests and reported exact probabilities, we used Holm’s sequential strategy with α = .05 as the level of significance to reduce the familywise error rate. For example, for studies in which three tests were performed, we ranked the computed P values in increasing order, comparing the smallest value (αH1) with .05/3, the next smallest significance level (αH2) with .05/2 and the last (that is, the largest) (αH3) P value with .05/1, or αH = .05. We deemed comparisons statistically significant if P < αH.