SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

ABSTRACT: Context: Regular screenings are important for reducing cancer morbidity and mortality. There are several barriers to receiving timely cancer screening, including overweight/obesity. No study has examined the relationship between overweight/obesity and cancer screening among American Indian/Alaska Natives (AI/ANs). Purpose: To describe the prevalence of fecal occult blood testing (FOBT) and prostate-specific antigen (PSA) testing among AI/AN men within the past year by age and rurality, and determine if body mass index (BMI) is associated with screening. Methods: A national cross-sectional survey was administered face-to-face to 2,447 AI/AN men at least 55 years of age in 2004-2005. Participants were asked when they last had FOBT and PSA testing. BMI was derived from self-reported height and weight, and rurality of residence was defined by rural-urban commuting area codes. We assessed the association of cancer screening and BMI with logistic regression models, adjusting for demographic and health factors. Findings: Prevalence of up-to-date FOBT and PSA testing were 23% and 40%, respectively. Older men were more likely than younger men to have FOBT and PSA testing. BMI was not associated with receipt of FOBT or PSA testing. Conclusions: This is the first study to examine obesity and health care in AI/ANs. As in other populations, FOBT and PSA testing were suboptimal. Screening was not associated with BMI. Studies of AI/AN men are needed to understand the barriers to receiving timely screenings for prostate and colorectal cancer.

Colorectal cancer is the third most common cancer among US men, with an estimated 49,000 new cases and 28,000 deaths in 2006. Prostate cancer is the most common cancer among males with 234,000 cases in 2006, accounting for 33% of all cancers in men and 27,000 deaths in 2006.1 Among American Indian/Alaska Native (AI/AN) men, the case mortalities for colorectal cancer and prostate cancer are lower than that of the US general population and are declining.1,2 However, the case mortalities for colorectal and prostate cancers among AI men in the Northern Plains exceed those for US men of all racial/ethnic populations by 59% and 48%, respectively.

Recent studies have linked obesity with cancer, including colorectal and prostate cancer.3–5 In the United States, 67% of all adult men and 77% of AI adult men are overweight/obese.6 Obesity is associated with decreased access to care and lower rates of some cancer screening behaviors.7,8 The influence of obesity on receipt of colorectal cancer screening tests, however, is inconsistent but appears to vary with sex and test type.9 In 2 available studies on prostate specific antigen (PSA) testing, men with a higher body mass index (BMI) were more likely than their lighter counterparts to have had recent PSA testing.10,11

Rurality is another barrier to health care access and cancer screening.12,13 Rural residents are more likely to experience poor health and chronic conditions than their urban counterparts.14 Rural residents also visit health care providers less frequently and receive fewer preventive services, including cancer screenings.12

To our knowledge, no study has examined the relationship between BMI and cancer screening among AI/ANs. Such a study is needed to assess whether overweight/obesity act as barriers to or promote preventive cancer care in Native populations, especially elders. We analyzed data from a nationwide survey of 2,447 elderly AI/ANs to (1) describe the prevalence of fecal occult blood testing (FOBT) and PSA testing within the past year by age and rurality, and (2) determine whether BMI is associated with receipt of FOBT and PSA testing.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

This study used data from a cross-sectional national survey conducted in 2004-2005 by the National Resource Center on Native American Aging. A letter soliciting participation in the study was mailed to all 243 regional administrators of the Title VI Native American Nutrition program. Each administrator receives federal funds to provide meals to AI/AN elders within their tribal service area population. A total of 94 tribal service areas volunteered to participate in the study. Resource Center staff procured tribal approval from each participating site. Trained interviewers, selected from the local area, contacted the elders to solicit their study participation and set up the interview date, time, and location.

FOBT was assessed through the question, “How long has it been since you had your last blood stool test using a home kit?” Receipt of a PSA test was assessed through the question, “How long has it been since you had your last PSA, prostate-specific antigen test, a blood test used for men for prostate cancer?” BMI was derived using the respondents' self-reported current height and weight. Covariates, derived from the literature as potential confounders, included age, marital status, education, limitations in activities of daily living, smoking status, health insurance coverage, and rurality of residence assessed by rural-urban commuting area codes.15 Health insurance status was defined by a 3-category variable indicating insurance coverage, Indian Health Service (IHS) coverage only, or no coverage at all. We distinguished IHS coverage from the others because it provides many AI/AN elders with access to health care services on reservations, but is not health insurance coverage.16 The remaining insurance categories were combined since health insurance coverage in general is linked with increased access to primary care,17 including cancer screenings.18

We analyzed data from men who had complete records for all model covariates and claimed not to have ever been diagnosed with colorectal or prostate cancer. After excluding all incomplete records, data were available on 2,447 elders for the FOBT analysis and 2,339 for the PSA analysis. We compared the frequency distributions for all independent variables by screening status, and assessed the association of cancer testing and BMI by creating 2 logistic regression models. The first modeled the log-odds of FOBT receipt within the past year as a function of BMI category and the covariates. The second modeled the log-odds of PSA test receipt within the past year as a function of BMI category and the covariates.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Overall, 23% of elderly AI/AN men had a FOBT within the past year. Having a FOBT was associated with higher age, being married, being unemployed during the past year, having a regular health care provider, having health insurance, having higher numbers of limitations in activities of daily living, and being a non-smoker (Table 1). Overall, 40% had a PSA test in the past year. Having a PSA test was associated with older age, being married, higher education level, having health insurance, having higher numbers of limitations in daily living, and non-smoking status. In addition, men living in isolated rural communities were less likely than those living in larger communities to have obtained a recent PSA test.

Table 1.  Sociodemographic and Health Characteristics by Cancer Screening Status, AI/AN Males Aged 55 and Older (2004-2005)
CharacteristicsFOBT in Past YearP-valuePSA Test in Past YearP-value
Yes n = 573No n = 1,874Yes n = 935No n = 1,404
  1. *Health insurance includes Medicare, Medicaid, private insurance, Veterans Administration, tribal insurance and/or other insurance type. NS, not significant.

  2. **P < .01.

  3. ***P < .001.

Sociodemographic%% %% 
 Age, years  ***  ***
  55-643442 3645 
  65-744039 4237 
  ≥752719 2218 
 Married6453***6350***
 ≥High school7672NS7771***
 Employed in past year2332***3031NS
 Residence  NS  ***
  Urban2020 1921 
  Large rural2017 2016 
  Small rural1817 2015 
  Isolated rural4346 4148 
 Regular health provider8372***8169***
 Health insurance  **  ***
  Yes*8177 8273 
  Indian Health Service1217 1219 
  No 7 7  6 8 
Health
 ADL limitations  ***  **
  06875 7076 
  1-32219 2218 
  ≥410 6  8 6 
 Cigarette smoker2331***2433***
 BMI category  NS  NS
  Underweight 2 2  1 2 
  Healthy2020 1920 
  Overweight3939 3640 
  Obese I2523 2622 
  Obese II 910 11 9 
  Obese III 6 7  6 7 

We examined the prevalence of FOBT and PSA testing in the past year by rurality of residence and age (<65 vs 65+). For FOBTs, prevalence in urban areas did not differ by age. However, for rural regions, men aged 65 and older showed a higher prevalence of recent FOBTs than younger men. For PSA tests, men aged 65 and older had higher percentages than younger men for all geographic classifications.

We calculated the adjusted screening odds ratios for FOBT and PSA tests in the past year by BMI category (Table 2). The category, underweight (ie, BMI score < 18.5), was used as the reference group in the logistic regression models. BMI was not associated with either FOBT or PSA testing, with adjustment for covariates.

Table 2.  FOBT and PSA Test Receipt Odds Ratios (Adjusted) by BMI Category, AI/AN Males Aged 55 and Older (2004-2005)
BMI CategoryFOBTPSA Test
OR95% CIOR95% CI
  1. OR, odds ratio; CI, confidence interval.

  2. Reference group: underweight (ie, BMI score < 18.5).

Underweight 1 1
Healthy1.0(0.5, 2.0)1.5(0.7, 3.1)
Overweight1.0(0.5, 2.0)1.3(0.6, 2.6)
Obese I1.0(0.5, 2.2)1.7(0.8, 3.5)
Obese II0.9(0.4, 1.9)1.6(0.8, 3.3)
Obese III0.8(0.3, 1.8)1.2(0.6, 2.6)

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

The American Cancer Society recommends that all men aged 50 years and older obtain FOBT and PSA testing annually,1 although there is no consensus regarding the recommendation for PSA.19 We found less than half of older AI/AN men had received these tests within the past year, and these rates were lower than for similarly-aged US men.20 Receipt of FOBT and PSA were both associated with older age, being married, having health insurance, and being a non-smoker. In addition, receipt of FOBT was linked with having a regular health provider and more limitations in activities of daily living. Having a recent PSA test was also related to living in larger rural towns. PSA testing requires a visit to a clinic or health care provider and rural residents have decreased access to primary health care services.21

In this analysis, BMI category was not associated with either FOBT or PSA testing among older AI/AN men. In contrast, 2 recent studies reported overweight/obese men were more likely than healthy-weight men to have been recently screened for prostate cancer.10,11 However, men in those studies were younger and primarily white or African-American. The authors hypothesized overweight/obese men experienced urogenital symptoms due to excess bodyweight and were thus screened at higher rates.10 Other studies have suggested higher intra-abdominal pressure chronically stresses the pelvic floor22 and BMI is positively associated with lower urinary tract health conditions (eg, incontinence).23 Also, overweight/obese people may be more likely to get screened because of more frequent health care provider use due to comorbid conditions associated with their weight. It is plausible that, given the very high prevalence of obesity among AI/ANs,6,24 being overweight has become the norm. As a result, discrimination by body weight, especially in health care settings, may be diminished and heavier persons would be no less likely than healthy weight individuals to be screened for cancer.

Our findings are cause for concern given that the prognosis of prostate and colorectal cancer is greatly improved when these malignancies are detected early. About 20% of our sample was covered only by the IHS, which is the main source of health care for many AI/ANs, especially those living on or near reservations.25 The IHS provides only primary care which includes FOB/PSA testing; however, colonoscopies are generally unavailable. Moreover, the funding appropriated to the IHS can address only about 65% of the health care needs of its service population.25

This study has limitations. The sample was not selected randomly, and may therefore not accurately represent the whole US Native population. Other unknown selection biases could have been introduced at the tribal level. Although the survey was advertised to tribes across the country, it is likely that not all tribes were aware of the survey, wished to participate, or had the requisite human resources, information, or training services. Even so, this is the largest survey of AI/AN elders ever conducted.

In conclusion, our findings indicated problem areas among older AI/AN men with obesity and low cancer testing rates, but the 2 were unrelated in this sample. Future studies are needed to assess barriers to receiving timely prostate and colorectal cancer screenings. Study findings suggest that absence of a usual health provider, and lacking health insurance should be more closely examined as tangible obstacles that hinder AI/AN men from receiving preventive services.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References