Epidemiology of oral cancer in Connecticut, 1935 to 1985

Trends in the incidence of lip, oral, and pharyngeal cancer: Connecticut, 1935-94

Using Connecticut Tumor Registry data we explored trends in age-adjusted (AARs) and age-specific (ASRs) incidence rates for lip, oral, and pharyngeal cancer over the 60-year period 1935-94. Particular attention was given to findings from the most recent series of 5-year periods that have not been previously analyzed. There was a long-term decline in lip cancer AARs by period, and ASRs generally fell over time and with successive birth cohorts. This notable decrease in rates continued through 1990-94 and with the more recent cohorts, particularly among males. Oral cancer AARs for males peaked in the early 1960s and the late 1970s while rates for pharyngeal cancer increased into the late 1970s. For each of these sites, rates began to decline in 1980-84 and have continued to fall into the first half of the 1990s. Among females AARs for oral and pharyngeal cancer increased more than threefold between 1935-39 and 1980-84; however, encouragingly, and in sharp contrast to the earlier trend, rates fell during the most recent 10-year period. ASRs for oral and pharyngeal cancer were increasing by the birth cohort of 1900 and tended to increase through the cohort of 1920 for males and the cohorts of 1920-30 for females. ASRs for subsequent cohorts have remained relatively unchanged or decreased modestly. It is doubtful that the observed trends in lip, oral and pharyngeal cancer incidence are primarily artifactual, but more likely represent secular changes in exposure to environmental risk factors.

Trends in site-specific mortality from oral and pharyngeal cancer among Japanese males, 1950-94

In order to obtain some etiological clues for oral and pharyngeal cancer in Japanese males, we examined mortality trends from cancer of the lip, buccal cavity and pharynx by site with other descriptive epidemiological features (1950-94). The data were obtained from Vital Statistics (Ministry of Health and Welfare, Japan). The 45-year trend in mortality was evaluated by computing crude and age-adjusted mortality rates, age-specific mortality rates, with birth cohort analysis. The crude mortality rate from cancer of the lip, buccal cavity and pharynx has increased by 3.7-fold, and age adjusted mortality rate by 1.7-fold in the 45-year period, with different mortality trends by site. In general, age-specific mortality has also been increasing with advancing age. The birth cohort analysis showed that the higher the mortality rate, the younger the birth cohort, but with different effects of birth cohort by site. Mortality from cancer of the lip, buccal cavity and pharynx has consistently been increasing in Japanese males from 1950 to 1994, with similar upward mortality trends for some specific sites.

How you look determines what you find: severity of illness and variation in blood transfusion for hip fracture

PURPOSE

Utilization report cards are commonly used to assess hospitals. However, in practice, they rarely account for differences in patient populations among hospitals. Our study questions were: (1) How does transfusion utilization for hip fracture patients vary among hospitals? (2) What patient characteristics are associated with transfusion and how do those characteristics vary among hospitals? (3) Is the apparent pattern of variation of utilization among hospitals altered by controlling for these patient characteristics?

SUBJECTS AND METHODS

We included consecutive hip fracture patients aged 60 years or older who underwent surgical repair between 1982 and 1993 in 19 hospitals from four states, excluding those who refused blood transfusion, had multiple trauma, metastatic cancer, multiple myeloma, an above the knee amputation, or were paraplegic or quadriplegic. The outcome of interest was postoperative blood transfusion. "Trigger hemoglobin" was the lowest hemoglobin recorded before transfusion or recorded at any time during the week before or after surgery for patients who were not transfused.

RESULTS

There was considerable variation in transfusion among hospitals postoperatively (range 31.2% to 54.0%, P = 0.001). Trigger hemoglobin also varied considerably among hospitals. In unadjusted analyses, four of nine teaching and two of nine nonteaching hospitals had postoperative transfusion rates significantly higher than the reference (teaching) hospital, while one nonteaching hospital had a lower rate. In an analysis controlling for trigger hemoglobin and multiple clinical variables, one of nine teaching and four of nine nonteaching hospitals had rates higher than the reference hospital, while four teaching hospitals and one nonteaching hospital had lower rates.

CONCLUSIONS

The apparent pattern of variation of transfusion among hospitals varies according to how one adjusts for relevant patient characteristics. Utilization report cards that fail to adjust for these characteristics may be misleading.

Trends and differentials in mortality from cancers of the oral cavity and pharynx in the United States, 1973-1987

BACKGROUND

This analysis consisted of an examination of trends and differentials in mortality from cancers of the oral cavity and pharynx in the United States for a recent 15-year period.

METHODS

The authors have used national cause-of-death data for the United States and intercensal population estimates to examine mortality from oral and pharyngeal cancers between 1973 and 1987 and to study differentials according to gender, race, and region of residence.

RESULTS

The overall mortality rate from these cancers decreased by 19% during the 15-year period, with most of the decline occurring after 1979. Mortality was much higher for men than for women and for blacks than for whites throughout the interval. Despite the overall decline, mortality rates increased among blacks, especially among black men. Mortality was highest in the South Atlantic, New England, and Mid-Atlantic states and lowest in the Mountain states.

CONCLUSIONS

The disparity between male and female mortality from oral and pharyngeal cancer stems mainly from differences in the likelihood of developing these cancers, whereas the differences between blacks and whites appears to arise more from differences in survival than in incidence. Different age patterns of mortality for blacks and whites exist, in which mortality among whites, but not among blacks, rises continuously with age. An unexplained finding was that mortality rates were reported to have fallen in recent years, whereas incidence and survival rates have reportedly remained almost unchanged. This apparent inconsistency may have resulted from declines in the incidence of oral and pharyngeal cancers that have been masked by improved detection.

Head and neck carcinomas in patients under the age of forty years

From 1973 to 1988, 1,104 patients with head and neck carcinomas (excluding thyroid carcinoma) were admitted to the Yokohama City University Hospital, and among them, 48 (4.3%) were under the age of 40 years. We made a retrospective survey of these young patients. There were 36 males and 12 females. An increasing ratio of male to female was seen with increasing age. The primary site was most frequently the nasopharynx (29 cases), followed by the maxillary sinus (9 cases), and the tongue (4 cases). Squamous cell carcinoma in these young patients characteristically showed a low grade of differentiation.

Toward assessing trends in oral mucosal lesions: lessons learned from oral cancer

Trend data on oral mucosal lesions are limited. The literature is comprised of a large number of studies documenting the prevalence of various lesions in different populations. Trend data on cancers of the oral cavity and pharynx are the exception. An overview of the international epidemiologic studies of these cancers is provided. General conclusions across studies and countries for trend data of these cancers cannot easily be drawn, since variations exist in data collection methods, risk factors, and other characteristics of the study populations. This paper also reviewed the available prevalence data for oral mucosal lesions in general as well as data on leukoplakia, recurrent herpes labialis, and recurrent aphthous ulcers. A few studies of the incidence of leukoplakia are described. In order to explore the potential for future trend data on oral mucosal lesions, this paper discusses issues related to the feasibility and structure of monitoring these lesions. Criteria that could be used to identify which conditions should be monitored in the future were discussed, such as, those that present a public health problem, those that are sentinel events for serious conditions, and infectious lesions. Once lesions are identified, it is clear that international consensus for diagnostic criteria of these lesions is needed in order to permit comparisons to be made among studies. Finally, strategies for monitoring these conditions and factors that may affect future trends are reviewed.

Lip cancer. Incidence trends in Connecticut, 1935-1985

Suspicions have recently arisen that cancer of the lip may exert an undue influence on overall oral cancer statistics and, therefore, possibly distort the true image of intraoral cancer. The authors investigated this question through epidemiologic analysis. A total of 2291 cases of lip cancer accessioned by the Connecticut Tumor Registry (CTR) from 1935 to 1985 (23.6% of all oral cancer) were analyzed. Occurrence trends for males and females had different patterns: for men, the age-adjusted incidence rates showed a fivefold decrease during the 51-year study; for women, the rates were relatively low and constant during the same period. Analysis for age-specific rates revealed that the older the age group, the higher the incidence rates for both sexes. Squamous cell carcinoma accounted for at least 87.4% of all lip cancers (96.2% if nonspecified epithelial neoplasms are assumed to be squamous cell carcinoma). The vermilion border of lower lip was the most common site. Moderately differentiated tumors were most common (48.5%), closely followed by well-differentiated tumors (44.2%). Analysis by county showed that the crude incidence rates for males in New London and Windham counties exceeded the average Connecticut statewide rates. The authors concluded that the epidemiology of Connecticut lip cancer differs significantly from that of intraoral squamous cell carcinoma in the same population studied within the same period of time. Epidemiologic studies involving "oral cancer" should direct attention to anatomic subsite to consider differences in disease trends according to specific location.

Changing trends in oral cancer in the United States, 1935 to 1985: a Connecticut study

An extensive epidemiologic study was undertaken in an attempt to analyze patterns of oral cancer distribution and demography in Connecticut between 1935 and 1985. Sources of data for the investigation included both the Connecticut Tumor Registry (CTR) and the University of Connecticut Oral Pathology Biopsy Service. During the 51-year study period, 9,708 cases of primary oral cancer were reported to the CTR. Male age-adjusted incidence rates for overall oral cancer remained stable between 1935 and 1964 (14.5 to 14.8 per 100,000), with a gradual decline to 10.9 per 100,000 in the early 1980s. In contrast, age-adjusted rates for females advanced approximately threefold, from 1.4 per 100,000 in the 1930s to 4.1 per 100,000 in the early 1980s. There was a decrease in age-specific rates of oral cancer in males aged 70 and older; in contrast, age-specific incidence rates in females increased steadily over the same period. It was also found that female birth cohorts born in 1900 and later exhibited higher oral cancer incidence rates than those of previous cohorts. Between the 1960s and the present, male patients 30 to 39 years of age exhibited a nearly fourfold increase in oral cancer incidence; this was not observed among similarly aged females. Connecticut counties with highest oral cancer incidence rates in both sexes were the more densely populated Hartford and New Haven counties. In general, the picture of oral cancer, as revealed through analysis of cases accessioned by the University biopsy service between 1975 and 1986, exhibited similar trends to those disclosed by analysis of CTR data.(ABSTRACT TRUNCATED AT 250 WORDS)

Intraoral squamous cell carcinoma. Epidemiologic patterns in Connecticut from 1935 to 1985

There were 6181 cases of invasive intraoral squamous cell carcinoma accessioned by the Connecticut State Tumor Registry from 1935 to 1985. Cases were analyzed for age, sex, lesion site, and histologic differentiation. Crude, age-specific, and age-adjusted incidence rates plus birth cohort analyses were also calculated. It was found that incidence rates for both men and women increased over the 51-year period of study. For men, age-adjusted incidence rates (1970 United States standard) increased from 4.9/100,000 in 1935 to 1939 to 8.5/100,000 in 1980 to 1985; for women, rates increased from 0.5/100,000 to 3.3/100,000 for the same period. The male-to-female ratio for intraoral squamous cell carcinoma declined dramatically from 9.8 to 2.6 during the 51-year study period primarily because of the steep rate of increased incidence in women relative to that seen in men. The peak age of intraoral squamous cell carcinoma was the seventh decade. Age-specific analysis showed that the older the age group, the higher the incidence for both sexes. During recent years, there was evidence of slightly increased incidence in men younger than 40. The tongue was the most common site for intraoral squamous cell carcinoma, followed closely by the floor of the mouth. Moderately differentiated tumors were most common (54.3% of the total), followed by both well-differentiated cases (29.1%) and those that were poorly differentiated (16.6%).

Oral cancer: a survey of 566 cases from the University of Connecticut Oral Pathology Biopsy Service, 1975-1986

A survey of the University of Connecticut Oral Pathology Biopsy Service was undertaken to analyze cases of oral cancer accessioned during the 12-year period, 1975 through 1986 inclusive. Of 33,429 total specimens accessioned, there were 546 malignant oral neoplasms diagnosed and reported. Sixty-five (11.5%) originated from out of state. Invasive intraoral squamous cell carcinoma was the predominant tumor (69.7% of total), whereas lip cancer constituted only 2.8% of all malignancies. Minor salivary gland adenocarcinomas accounted for 11% of total malignancies whereas verrucous carcinoma, carcinoma in situ, and miscellaneous other forms of oral cancer accounted for the remainder (4.6%, 5.3%, and 6.6%, respectively). Cases of invasive squamous cell carcinoma were further analyzed by year, sex distribution, location subsite, age at diagnosis, and histologic grade. With the exception of histologic grading, we found that the characterization of cases of squamous cell carcinoma within the biopsy service tended to parallel results from a separate but related statewide analysis of both oral cancer and intraoral squamous cell carcinoma from Connecticut over a much longer time span. We concluded that the picture of oral cancer as characterized by cases within the University of Connecticut Oral Pathology Biopsy Service is generally reflective of the disease on a statewide level.