Abstract
Objectives
The aim of this study was to determine if changes in annual temperature influence the prevalence of frequent otitis media (FOM) and respiratory allergy in children.
Methods
Annual prevalence data for FOM (defined as 3 or more ear infections per year), respiratory allergy, and seizures (nonrespiratory, control condition) in children were extracted from the National Health Interview Survey for 1998 to 2006. Average US annual temperatures for the same period were recorded from the Environmental Protection Agency. Complex samples logistic regression analyses were performed to identify possible correlations between annual temperature and each of the 3 disease conditions, controlling for age and sex.
Results
A total of 113 067 children were studied (mean age, 8.6 years; 51.1% girls). Overall prevalences (±95% confidence interval) were 6.3% ± 0.2%, 11.8% ± 0.2%, and 0.7% ± 0.1% for FOM, respiratory allergy, and seizures (nonrespiratory, control condition), respectively. Average annual temperatures ranged from 53.64°F to 55.09°F. Regression analysis found that annual temperature did not influence the prevalence of FOM ( P = .681); male sex and younger age were associated with a higher prevalence of FOM ( P = .025 and P < .001, respectively). Similarly, annual temperature did not influence prevalence of respiratory allergy ( P = .883); male sex and increasing age were associated with a higher prevalence of respiratory allergy (both P < .001). Annual temperature and sex did not influence seizure prevalence; however, increasing age was negatively associated.
Conclusions
Changes in average annual temperature do not appear to influence the prevalence of otitis media or respiratory allergy. This negative finding suggests that although global warming continues to affect our environment, childhood otolaryngologic disease prevalence may not be directly influenced.
1
Introduction
Because worldwide ambient temperatures continue to increase because of global warming, there is growing concern regarding the possible detrimental health effects of climate change . Health concerns have centered on discussions about increased incidence of infectious disease, vector-borne and diarrheal illness, and allergy, as well as the availability of food sources. Few studies have investigated the possible correlation of average annual temperatures and otolaryngologic disease prevalence, and those that have were conducted in the adult population .
Among respiratory diseases that may be impacted by global warming, otitis media deserves special consideration because of its tremendous disease burden . Although several risk factors for otitis media have been identified, the effect of environmental temperature on the prevalence of respiratory diseases, including otitis media, has not been described. The current study aims to discover a possible correlation between average annual temperature and the prevalence of respiratory allergy and frequent otitis media (FOM). To do so, it uses the National Health Interview Survey (NHIS), considered a gold standard for population-based disease prevalence in the United States . The NHIS is a multipurpose household-based health survey conducted by the National Center for Health Statistics and is the principal source of information on the health of the civilian, noninstitutionalized household population of the United States.
2
Methods
The study population consisted of the pediatric survey population (age, 0–18 years) extracted from the NHIS as aggregated in the Integrated Health Interview Series. Because this is a completely de-identified data set in the public domain, institutional review board approval was not required. In addition to standard demographic data, annual prevalence data for FOM (defined as 3 or more ear infections per year), respiratory allergy, and seizures (a nonrespiratory, control condition) in children were extracted from the NHIS for 1998 to 2006. The variable frequent ear infection was based solely on parental response to the specific question asked for each child: “During the past 12 months, has your child had three or more ear infections?” Other questions asked were as follows: “During the past 12 months, has your child had any type of respiratory allergy?” and “During the past 12 months, has your child had seizures?” These variables therefore represent a 12-month prevalence for each of these disorders. Average US annual temperatures for the same period was recorded from the Environmental Protection Agency. Average temperature data are gathered both from land-based weather stations and satellite measurements. Data were imported into SPSS 17.0 (SPSS, Inc, Chicago, IL) and cross-checked for accuracy with the original data sets.
Standard demographic statistics were computed for the study population. Logistic regression analyses were performed with each of the 3 disease condition variables (dependent variables) and annual temperatures. Age and sex were the independent predictor variables. Because the NHIS is a complex sample survey, appropriate methods for statistical analyses with complex samples were applied. Statistical significance was set at P < .05. Graphical analysis was also conducted to confirm statistical relationships visually.
2
Methods
The study population consisted of the pediatric survey population (age, 0–18 years) extracted from the NHIS as aggregated in the Integrated Health Interview Series. Because this is a completely de-identified data set in the public domain, institutional review board approval was not required. In addition to standard demographic data, annual prevalence data for FOM (defined as 3 or more ear infections per year), respiratory allergy, and seizures (a nonrespiratory, control condition) in children were extracted from the NHIS for 1998 to 2006. The variable frequent ear infection was based solely on parental response to the specific question asked for each child: “During the past 12 months, has your child had three or more ear infections?” Other questions asked were as follows: “During the past 12 months, has your child had any type of respiratory allergy?” and “During the past 12 months, has your child had seizures?” These variables therefore represent a 12-month prevalence for each of these disorders. Average US annual temperatures for the same period was recorded from the Environmental Protection Agency. Average temperature data are gathered both from land-based weather stations and satellite measurements. Data were imported into SPSS 17.0 (SPSS, Inc, Chicago, IL) and cross-checked for accuracy with the original data sets.
Standard demographic statistics were computed for the study population. Logistic regression analyses were performed with each of the 3 disease condition variables (dependent variables) and annual temperatures. Age and sex were the independent predictor variables. Because the NHIS is a complex sample survey, appropriate methods for statistical analyses with complex samples were applied. Statistical significance was set at P < .05. Graphical analysis was also conducted to confirm statistical relationships visually.
3
Results
Overall, 113 067 children were studied (mean age, 8.6 years; 51.1% girls). The overall prevalences (±95% confidence interval) were 6.3% ± 0.2%, 11.8% ± 0.2%, and 0.7% ± 0.1% for FOM, respiratory allergy, and seizures, respectively. This corresponds to US annual prevalence estimates of 4.07 million, 7.63 million, and 0.45 million children, respectively. The average annual temperatures in the United States ranged from 53.64°F to 55.09°F during the study period ( Table 1 ).
| Year | Mean annual temperature (°F) | n |
|---|---|---|
| 1998 | 54.94 | 98 785 |
| 1999 | 54.53 | 97 059 |
| 2000 | 53.84 | 100 618 |
| 2001 | 54.23 | 100 760 |
| 2002 | 53.66 | 93 386 |
| 2003 | 53.88 | 92 148 |
| 2004 | 53.64 | 94 460 |
| 2005 | 54.08 | 98 649 |
| 2006 | 55.09 | 75 716 |
The results of the statistical analysis are presented in Table 2 . Logistic regression analyses found that annual temperature did not influence the prevalence of FOM ( P = .681), whereas male sex and younger age were associated with a higher prevalence of FOM ( P = .025 and P < .001, respectively). Similarly, annual temperature did not influence prevalence of respiratory allergy ( P = .883); male sex and increasing age were associated with a higher prevalence of respiratory allergy (both P < .001). Annual temperature and sex did not have a statistically significant influence on seizure prevalence ( P = .949 and .522, respectively), whereas increasing age was associated with a lower seizure prevalence ( P = .004). Relationships among the disease conditions and annual temperature are graphically presented in Figs. 1 to 3 .
| Disease condition | Odds ratio for disease | 95% confidence interval | P | |
|---|---|---|---|---|
| FOM | ||||
| Intercept | 0.080 | 0.003 | 2.289 | .145 |
| Age | 0.880 | 0.875 | 0.886 | <.001 |
| Male sex | 1.068 | 1.008 | 1.131 | .025 |
| Annual temperature | 1.013 | 0.952 | 1.078 | .681 |
| Respiratory allergy | ||||
| Intercept | 0.068 | 0.007 | 0.713 | .031 |
| Age | 1.044 | 1.040 | 1.048 | <.001 |
| Male sex | 1.216 | 1.161 | 1.273 | <.001 |
| Annual temperature | 1.003 | 0.961 | 1.048 | .883 |
| Seizures | ||||
| Intercept | 0.006 | 0.001 | 277.0 | .352 |
| Age | 0.974 | 0.956 | 0.992 | .004 |
| Male sex | 1.063 | 0.882 | 1.280 | .522 |
| Annual temperature | 1.006 | 0.825 | 1.228 | .949 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
