The elderly, who may benefit most, often do not receive preventive services.1,2 Older American Indians and Alaska Natives (Al/ANs) typically have low income and educational levels, placing them at high risk for not receiving preventive services.3–5 Since the leading causes of death for older Al/ANs are heart disease, malignancies, injuries, and cerebrovascular disease, adherence to screening guidelines is of utmost importance. Yet, previous investigations of AIs were based on self-report, included only 1 or 2 preventive practices, and did not focus on older or urban populations.6 We therefore reviewed the medical records of 550 Al/AN elders seen in an urban primary care setting to evaluate the frequency recommended preventive procedures were received and to examine the factors associated with nonadherence.
Little is known about prevention among elderly or urban American Indian/Alaska Native (AI/AN) populations. We reviewed the medical records of 550 older urban AI/AN primary care patients to evaluate how frequently preventive measures were received. Adherence to guidelines was examined by a culturally appropriate (≥50 years) and standard age threshold (≥65 years), and by performance of preventive measures at any time (“ever”) and in the past year. Lifetime performance was inadequate for the many measures, including mammograms (56%), fecal occult blood testing (37%), audiometry (33%), visual acuity testing (50%), smoking cessation counseling (50%), and pneumococcal (22%) and influenza (49%) vaccinations. Performance of the measures was less frequent in the prior year, but did not differ by age threshold. Predictors of adherence included female gender, having insurance, and having more health problems and medications. Nonadherence infrequently resulted from patients' failure to comply with recommendations. We conclude that use of most preventive services among elderly urban AI/ANs is suboptimal and should be improved.
This study was conducted at the Seattle Indian Health Board which provides health care to King County AI/ANs. Patients represent approximately 250 tribes. Approximately 50% are unemployed; 80% have incomes below the “poverty line”; and 58% have no health insurance. A computerized information system was used to identify all AI/ANs who were at least aged 50 years seen between June 1994 and June 1995.
In AI/AN culture, “elder” status is not solely a function of chronological age and is conferred earlier than in the majority culture. For AI/ANs, impairments associated with aging may occur 20 years earlier than in the general population,7 and lower age eligibility criteria exist for some federal programs. Therefore, we used 2 definitions of “elderhood”: at least aged 50 years and at least aged 65 years. Preventive interventions relevant to Al/ANs were selected by 3 geriatricians and 2 general internists. Sixteen of the interventions were recommended by national organizations, and 2 were clinically based (incontinence, dementia screening).
An abstractor, supervised by a general internist, reviewed charts. Study forms were developed, pilot tested, modified, then reviewed for discrepancies between the abstractor and internist. Because training was an iterative process, interrater reliability tests were not applied. General information was collected on demographics, current smoking, alcohol/substance use, provider-documented medical and psychological problems, and current medications.
Data were recorded for mammograms, clinical breast examinations (CBEs), Pap smears, pelvic examinations, thyroid tests, fecal occult blood testing (FOBT), audiometry, ocular pressure testing, PPD placement, visual acuity testing, smoking cessation counseling, and pneumococcal and influenza vaccine administration. We determined whether cholesterol tests had been completed every 5 years, blood pressure was checked yearly for the last 5 years, and if falls, incontinence, and dementia had been assessed. For services performed off-site, we recorded whether a test had been ordered and performed, and, for FOBT, if cards had been given out and returned.
Group differences were examined with χ2 and t tests. Univariate models for mammography, CBEs, Pap smears, and FOBT excluded patients with breast, cervical, and colorectal malignancies. Multiple regressions determined predictors of adherence to preventive services with the total number of procedures as the dependent variable and adjusting for the lower number recommended for men (lifetime n = 13; annual n = 6) compared with women (lifetime n = 18; annual n = 9). Univariate logistic regression models examined compliance with individual annual screening interventions. For all regressions, independent variables included age, gender, marital status, employment, insurance type, current alcohol/substance use/smoking, number of medical and psychological problems, and current medications.
Charts were abstracted for all 550 AI/ANs (100%) who were at least aged 50 years and made at least 1 visit to the clinic. The patients' mean age was 60.6 years. Women comprised 61% of the 550 AI/ANs and 28% of all patients were married; 45% had a high school education or less; 43% were employed; and 49% received Medicare/Medicaid. A minority were alcohol users (26%), substance abusers (3%), or smokers (36%); 492 (89%) and 299 (54%) patients had medical and psychological problems, respectively.
Our guidelines are shown in Table 1. Table 2 summarizes the frequency screening was performed “ever” and during the past year. Except for vaccinations and FOBT, application of culturally appropriate age thresholds did not yield lower adherence rates. For both age groups, screening was less often performed during the prior year.
|Females aged 50–59 years||Mammography annually||USPTF, ACP, CTF, ACS, ACOG, AAFP|
|Females aged ≥60 years||Mammography every 2 years||USPTF, ACP, CTF, ACS, ACOG, AAFP|
|All females||Annual breast examination||USPTF, ACP, CTF, ACS, ACOG, AAFP|
|Sexually active females aged 19–70 years||Pap smear every 3 years after 2 normal||HEDIS, USPSTF, ACP, CTF, ACS, ACOG, AAFP|
|Ovarian/vaginal cancer||Pelvic examination annually||ACS, ACOG, AAFP|
|Colorectoral cancer||Fecal occult blood annually||USPSTF, ACP, ACS, ACOG|
|Aged ≥50 years|
|High cholesterol||Serum cholesterol every 5 years||HEDIS, USPSTF, ACP, AAFP, Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults|
|Aged 18–65 years|
|Hypertension||Blood pressure every visit; ≤every 2 years||USPSTF, ACP, CTF, Joint Committee on the Detection, Evaluation, and Treatment of High Blood Pressure, ACP, AAFP|
|All ages||if diastolic <85 and systolic <140;|
|Annual if diastolic 85–89 and systolic 140–159|
|Pneumococcal pneumonia||Pneumovax every 5 years||USPSTF, ACP, CDC, AAFP|
|Aged ≥65 years and high risk*|
|Influenza||Vaccinate annually||USPSTF, ACP, CTF, ACOG, AAFP, CDC Immunization Practices Committee|
|Ages ≥65 years and high risk*|
|Tuberculosis||PPD once||USPSTF, CDC, CTF, AAFP†|
|Smoking||Cessation counseling annually||USPSTF, ACP, CTF, ACOG, AAFP|
|Thyroid disease||Thyroid function tests, once then at clinical discretion||USPSTF, CTF, AAFP‡|
|Females aged ≥60 years|
|Hearing impairment||Hearing test at least once, then screening frequency at clinical discretion||AAFP, Institute of Medicine|
|Aged ≥65 years|
|Visual impairment||Vision test at least once, then screening frequency at clinical discretion||USPSTF, CTF, AAFP, American Academy of Ophthalmology|
|Aged ≥65 years|
|Glaucoma||At least once, then screening frequency at clinical discretion||USPSTF, CTF§, American Academy of Ophthalmology|
|Aged ≥65 years|
|Falls||Query about falls annually||USPSTF, AAFP∥|
|Dementia||Screen for dementia|
|Incontinence||Query about incontinence annually|
|Aged ≥50 y||Aged ≥65 y|
|Ever||Last Year||Ever||Last Year|
|Procedure||n = 550||n = 550||n = 157||n = 157|
|Mammography||188 (56)*||54 (33)†||60 (58)‡||25 (24)|
|Clinical breast examination||245 (73)*||101 (30)*||79 (76)‡||27 (26)|
|Pap smear||237 (71)*||106 (32)*||69 (66)‡||34 (33)|
|Pelvic examination||234 (70)*||79 (24)*||76 (73)‡||17 (16)|
|Thyroid function tests||171 (51)*||60 (18)*||58 (45)‡||10 (8)|
|Cholesterol determination||457 (83)||315 (57)||140 (89)||101 (64)|
|Blood pressure determination||539 (98)||517 (94)||157 (99)||151 (96)|
|Fecal occult blood||204 (37)||32 (6)||72 (46)||6 (4)|
|Audiometry||182 (33)||39 (7)||59 (37)||11 (7)|
|Visual acuity||275 (50)||13 (2)||90 (57)||0 (0)|
|Ocular pressure||238 (43)||3 (1)||81 (51)||0 (0)|
|Smoking cessation counseling§||98 (50)||37 (19)||23 (53)||9 (21)|
|Pneumococal vaccination||119 (22)||27 ( 5)||51 (32)||9 ( 6)|
|Influenza vaccination||270 (49)||163 (30)||114 (72)||60 (38)|
|PPD||117 (21)||11 (2)||34 (22)||0 ( 0)|
|Fall prevention||42 (8)||14 (3)||15 (9)||4 (3)|
|Dementia||16 (3)||4 (1)||5 (3)||2 (1)|
|Incontinence||24 (4)||11 (2)||9 (6)||2 (1)|
Female gender (P≤ .001), insurance status (Medicare, P≤ .05; Medicaid, P≤ .01), and more medical (P≤ .01) and psychological (P≤ .001) problems predicted lifetime screening for both age groups. Female gender (P≤ .001), insurance (Medicare, P≤ .01), current smoking (P≤ .05), more psychological problems (P≤ .001), and more medications (P≤ .001) predicted receipt of annual procedures. For individual annual measures among patients who were aged at least 50 years, the following were demonstrated: (1) male gender predicted counseling among smokers (P≤ .01); (2) number of medications (P≤ .001) and having insurance (Medicare, P≤ .01; private, P≤ .05; self-insured, P≤ .01) predicted influenza vaccine; (3) being single predicted fall prevention (P≤ .01); (4) drug use predicted dementia screening (P≤ .05); and (5) younger age(P≤ .01) and insurance coverage (Medicaid, P≤ .01; self-insured, P≤ .05) predicted pelvic examinations. For patients aged at least 65 years, alcohol use predicted FOBT (P≤ .01), and more psychological problems predicted pelvic examinations (P≤ .05).
Abnormal results were common: 8% and 0.5% of women had a mammogram reported as abnormal or consistent with cancer, respectively; 5% had an abnormal CBE; 1% had dysplasia/metaplasia on Pap smear; and 5% had an abnormal pelvic examination. Among all patients, 9% had a positive FOBT; 80% had abnormal hearing; 21% had increased ocular pressure; and 21% had a PPD of 10 mm or more. No ocular pressure and only 17% mammography appointments were missed; FOBT cards were not returned by 13% of patients.
This study, which is the first to focus on the preventive care received by older urban AI/ANs, found that the performance of recommended screening varied widely. For most measures, lifetime performance was suboptimal including those targeting women (mammograms, 56%), cardiovascular and malignant diseases (FOBT, 37%; smoking cessation counseling, 50%), infectious diseases (pneumococcal vaccination, 22%), vision and hearing (ocular pressure, 43%; audiometry, 33%), and common problems of older adults (falls, 8%; dementia 3%). Performance was lower in the past year for all measures but did not differ by age, suggesting providers were cognizant of the cultural definition of elders. By comparison, in the only comprehensive screening study among older adults, Pap smears had been received by 17%, mammography by 56%, and pneumococcal and influenza vaccinations by 57% and 32%, respectively.2
Implementation of preventive measures among AI/ANs is important since the life expectancy of Al/ANs is 8 years less than non-Indians8 and their 5-year cancer survival is among the worst of any U.S. racial/ethnic group. Moreover, incidence and mortality of breast cancer is rising among AI/ANs,9–11 and the high incidence of cervical cancer and poor survival9 has lead the Indian Health Service (IHS) to recommend annual Pap smears for all women. Alaska Native women also have the highest incidence and mortality rate of any racial/ethnic group for colorectal cancer,11 although rates vary by tribe and gender.10 Given these statistics, the lack of information on preventive care for older AI/ANs is surprising.
Directly comparing our results to previous investigations is difficult since none have focused on older urban AI/ANs, and substantial variation in screening rates has been observed. For example, among women across IHS service areas, from 35% to 78% have had mammography, 28% to 70% have had an annual CBE, and 26% to 69% have had a Pap smear in the past year.12 Others reported that during the past year, 45% of eligible AI/AN women had a CBE,13 but only approximately 10% had received a Pap smear.14,15 The only comprehensive study of AIs, conducted in a rural setting, found self-reported rates for blood pressure checks (>70%), cholesterol tests (41%), pneumonia vaccinations (31%), or ever having a Pap smear (88%), CBE (80%), or mammogram (76%) generally similar to the chart-documented rates reported here.16 Of interest, among older rural AIs, mammography has been associated with availability,13 age, knowledge of the procedure, belief that mammography detects breast cancer, and use of the native language16— factors we were unable to assess.
Finally, although cardiovascular and pulmonary diseases are among the leading causes of death for older Al/ANs,17,18 data on screening for cardiovascular risk factors or smoking cessation are limited. In this regard, elevated blood pressure (≤38%19) and cholesterol levels (≤32%19–21), as well as smoking (36%22) are common among AI/ANs. Urban AIs have the highest,20,23 and older adults the lowest, smoking rates.24 Additionally, mortality is considerably higher for pneumonia, influenza, and tuberculosis (approximately 9 times) among AI/AN elders compared to their non-Native counterparts.17 These striking figures, in conjunction with our findings of low rates of smoking cessation counseling, administration of vaccinations, and screening for tuberculosis, highlight the need for greater efforts to reduce cardiopulmonary morbidity and mortality among AIs.
This study is limited by the method used to document adherence. Chart reviews may be subject to biases of incomplete information and retrospective case ascertainment. They are also inadequate to fully examine patient-based factors related to adherence. Additionally, since these data were collected in an urban practice, our results may not generalize to reservation-based communities. This drawback is attenuated by the fact that approximately 60% of AI/ANs live significant portions of their lives in cities.
Increasing age, lack of income, education, or health insurance, and assumptions that tests are indicated only when problems exist, have been associated with inadequate screening.3,16,18,25,26 Although Native women most often failed mammography appointments in one study,27 when providers ordered studies, our patients typically complied. Tribal health care organizations should educate providers and patients about prevention, particularly since only a third of state public health agencies have cancer prevention/control programs targeting AI/ANs.28 Fortunately, screening can be improved when programs are designed for AI/ANs, input from communities, providers, and state health organizations is solicited,9,29 and lay health advisors and community health aides are trained.27,30 Lastly, our finding that many older, urban AI/ANs received inadequate preventive care underscores the need to develop culturally relevant educational materials.
Supported in part by grants 1P30AG/NE15292 from the National Institute of Aging and the National Institute of Nursing Research, 90-AM-0757 from the Administration on Aging to Drs. Buchwald and Manson, and P0143471 from the National Institute of Mental Health to Dr. Manson.
The authors would like to thank Drs. Jan Beals and Christina Mitchell for statistical advice. Special appreciation to Ralph Forquera and the Seattle Indian Health Board for their support in pursuing improved patient care through research.