Head and Neck Cancer Mortality,Socioeconomic Status,and Distribution of Otolaryngologists

Abstract

Objectives:

Assess the relationship between distribution of otolaryngologists and head and neck cancer mortality rates, taking into account measures of socioeconomic status.

Methods:

We used the number of deaths from head and neck cancer in each U.S. county during 2002-2009 reported by National Center for Health Statistics. Due to low numbers in many counties, we aggregated counties into the 949 Health Service Areas (HSAs), which reflect hospital referral patterns. We used the Area Resource File to obtain numbers of otolaryngologists, general practice physicians, and dentists, as well as urban/rural designation, poverty rate, educational attainment, median household income, and racial minority proportion for each HSA. The impact of these covariates was evaluated on mortality rates adjusted for age and sex distribution.

Results:

The crude mortality rate was 4.1/100,000. There were no otolaryngologists in 274 HSAs (median 2, range 0-351). There were no significant relationships between number of otolaryngologists, general practitioners, or dentists, as continuous variables, and mortality rate. However, HSAs with no otolaryngologists had a higher mortality rate (by 0.32/100,000) than those with one or more otolaryngologist (p=.005). Mortality rates were significantly higher in HSAs with lower median income and greater poverty rate, non-white population, and proportion of adults without high school education, and also in entirely rural HSAs (P < .0001). On multivariate analysis, adverse socioeconomic factors and rural status were independently associated with higher mortality (p < .0001), but presence of an otolaryngologist was not (P = .23).

Conclusions:

Socioeconomic factors are stronger independent predictors of head and neck cancer mortality than otolaryngologist distribution.