. Author manuscript; available in PMC: 2010 Jun 28.

Published in final edited form as: Epidemiol Rev. 2009 Jun 16;31:113–129. doi: 10.1093/ajerev/mxp004

Table 1.

Literature on Cardiovascular Health Disparities Among Native Hawaiians and Other Pacific Islanders, January 1998–December 2008

Study Authors and Year (Ref. No.)	Sample and No. of Subjects	% of Total Sample Who Were NHOPI	Age, years	Study Design	Main Goal and Outcomes/Findings	Limitations
Novotny et al., 1998 (16)	66 Native Hawaiian women out of a total multiethnic sample of 421	16	25–35	Observational, cross-sectional, clinical research sample	Examine anthropometric variations between multiethnic women and their relation to blood pressure and cholesterol.	Sample bias; relatively small study.
Novotny et al., 1998 (16)		16	25–35	Observational, cross-sectional, clinical research sample	Native Hawaiian women had the highest measures of adiposity. Cholesterol was not associated with adiposity. Blood pressure was associated with adiposity but did not vary by ethnicity.	Sample bias; relatively small study.
Verderber et al., 1999 (20)	23 Pacific Islanders out of a total multiethnic sample of 60	38	40–85	Observational, cross-sectional, hospital surgery sample	Describe risk factors for cardiac disease and post-CABG outcomes and complications in multiethnic patients.	Sample bias (single hospital, relatively small sample). No adjustment for physician as covariate.
Verderber et al., 1999 (20)		38	40–85	Observational, cross-sectional, hospital surgery sample	No ethnic differences in postoperative complications in first 20 hours. At 48 hours post-CABG, Pacific Islanders required significantly more care than Japanese (P < 0.01).
Taira et al., 2001 (21)	361 Native Hawaiians out of a total multiethnic sample of 2,962	11	≥18	Observational, cross-sectional, administrative database sample	Examine differences in revascularization and mortality rates following acute coronary syndrome in a multiethnic patient population.	Cross-sectional; claims data from single health insurer.
Taira et al., 2001 (21)		11	≥18		Ethnic differences in types of procedures received were significant only for men. Native Hawaiian men were less likely to receive percutaneous coronary intervention (adjusted OR = 0.51, 95% CI: 0.34, 0.75) and more likely to receive CABG (adjusted OR = 1.8, 95% CI: 1.2, 2.7) than whites. No significant ethnic differences in women.	Cross-sectional; claims data from single health insurer.
Grandinetti et al., 2002 (18)	572 Native Hawaiians; nonpregnant	100	≥30	Observational, retrospective cohort, population-based sample	Examine degree of Hawaiian ancestry and blood pressure and relation to other covariates.	Percentage of Hawaiian ancestry was self-reported; cross-sectional design; sample population only from rural location. Assessment of adequate blood pressure control with medications not described.
Grandinetti et al., 2002 (18)	572 Native Hawaiians; nonpregnant	100	≥30		Adjusted prevalence of hypertension increased with degree of Hawaiian ancestry, except for 100% Hawaiian ancestry. Hawaiian ancestry was significantly associated with systolic and diastolic blood pressure (P < 0.0001). After adjustment for all significant covariates, only diastolic blood pressure was associated with Hawaiian ancestry (P = 0.008).
Sundaram et al., 2005 (17)	510 NHOPI women out of a total multiethnic sample of 120,035	<1	≥18	Observational, cross-sectional telephone survey, 2001 BRFSS	Determine prevalence of cardiovascular disease risk factors among multiethnic women in 2001 BRFSS.	All data were self-reported. Cross- sectional; relatively small sample compared with other ethnic groups.
Sundaram et al., 2005 (17)		<1	≥18	Observational, cross-sectional telephone survey, 2001 BRFSS	NHOPI women had second highest prevalence of hypertension at 33.7% (age-adjusted) and highest prevalence of high cholesterol (23.9%).
Yeo et al., 2005 (45)	377 Pacific Islanders out of a total multiethnic sample of 2,598	14	≥18	Observational, case-control, hospital administrative database sample	Examine ethnic differences in percutaneous coronary intervention success and/or complications in hospitalized patients.	Sample bias (recruited from a single hospital). Cross-sectional data; no long-term outcomes available.
Yeo et al., 2005 (45)		14	≥18		Despite higher rates of diabetes, hypertension, obesity, and renal failure in NHOPI, there was no difference in percutaneous coronary intervention complications compared with whites or Japanese.
Chiem et al., 2006 (46)	228 Chamorros	100	≥18	Observational, cross- sectional telephone survey, community administrative database sample	Describe cardiovascular disease risk factors in Chamorro community to aid in developing programs.	Sample bias (recruited from community database). No adjustment for covariates. Cross-sectional data. Self-reported health risk factors.
Chiem et al., 2006 (46)	228 Chamorros	100	≥18		Crude frequencies of hypertension, diabetes, hyperlipidemia, and physical inactivity were higher in Chamorros than in US whites. Chamorro women were more likely to have hypertension and diabetes. Men were more likely to have elevated body mass index and cholesterol.
Ezeamama et al., 2006 (22)	1,289 Samoans (American and Western Samoan)	100	25–58	Observational, prospective cohort, population-based sample	Investigate cross-sectional and prospective associations between SES and cardiovascular disease risk factors and predict the probability of risk factors by SES level between American Samoa and Samoa.	Relatively large sample of Samoan participants; subjects were younger and thus there were fewer cases of cardiovascular disease risk factors. Men lost to follow-up in both locations were more likely to be employed in the wage-labor market and may have introduced attrition bias.
Ezeamama et al., 2006 (22)	1,289 Samoans (American and Western Samoan)	100	25–58	Observational, prospective cohort, population-based sample	High SES was associated with increased odds of risk factors in less developed Western Samoa and decreased odds of risk factors in more developed American Samoa. Inverse association between SES and risk factors in Western Samoa vs. American Samoa is attributable to heterogeneity across the Samoan Islands in specific exposures to economic development and natural history of individual risk factors.
Grandinetti et al., 2006 (19)	185 Native Hawaiians out of a total multiethnic sample of 588; nonpregnant	31	≥30	Observational, retrospective cohort; randomly selected nested study of population- based sample	Examine angiotensin-converting enzyme gene polymorphisms and increased Q-Tc (associated with heart disease) between ethnic groups.	Cross-sectional prevalence Cases may lead to bias.
Grandinetti et al., 2006 (19)		31	≥30		Increased Q-Tc was highest among persons with the ACE II genotype. ACE II genotype was higher in Native Hawaiians (50.3%) than in whites (21%). After adjustment, prevalence of increased Q- Tc was significantly associated with ACE II genotype independently of ethnicity (P < 0.01).	Cross-sectional prevalence Cases may lead to bias.
Kretzer et al., 2007 (40)	15 Native Hawaiians out of a total multiethnic sample of 23	65	≥30	Quasi-experimental (no control group), pre-post intervention, community sample	Evaluate whether a class on self-identity via ho‘oponopono would improve high blood pressure.	No control group, small sample size, and potentially biased sample. No adjustment for medication changes.
Kretzer et al., 2007 (40)	15 Native Hawaiians out of a total multiethnic sample of 23	65	≥30		Ho‘oponopono intervention (4-hour group class) reduced mean systolic and diastolic blood pressure, which was sustained for 2 months after intervention. Results for the 15 Native Hawaiians were aggregated with those for the remaining 8 other non-Hawaiian participants.
Taira et al., 2007 (47)	3,746 Native Hawaiians out of a total multiethnic sample of 28,395	13	≥18	Observational, cross- sectional; health insurance administrative database sample	Examine factors associated with antihypertensive medication adherence within a multiethnic patient population using administrative claims data.	Cross-sectional claims data from a single health insurance plan. Patient adherence was measured by filling of prescriptions, not at patient level. Lack of information on impact of comorbid conditions on medication adherence.
Taira et al., 2007 (47)		13	≥18		Overall adherence in all ethnic groups was less than 65%. After adjustment, Native Hawaiians were less likely to adhere than whites (OR = 0.84, 95% CI: 0.78, 0.91), and this was consistent across therapeutic classes. Other patient factors associated with lower adherence: younger age, higher morbidity, and history of heart disease. Seeing a physician of the same ethnicity did not improve adherence.
Kaholokula et al., 2008 (39)	36 NHOPI and family caregivers	100	≥18	Focus group, community sample	Identify health beliefs, attitudes, practices, and social and family relations important in heart failure treatment among NHOPI.	Small sample size, and only 30% were heart failure patients. Qualitative study design with potentially subjective responses.
Kaholokula et al., 2008 (39)	36 NHOPI and family caregivers	100	≥18	Focus group, community sample	Native Hawaiians with heart failure reported coping experiences of denial of illness, hopelessness, and despair and relied on spiritual/religious beliefs for support. Samoans preferred being treated by physicians, while Native Hawaiians preferred traditional healers. Barriers to heart failure care include poor knowledge, lack of trust of the physician, a poor patient-physician relationship, finances, dietary changes, and increased demands on time.

Abbreviations: ACE II, angiotensin-converting enzyme insertion/insertion; BRFSS, Behavioral Risk Factor Surveillance System; CABG, coronary artery bypass graft; CI, confidence interval; NHOPI, Native Hawaiians and other Pacific Islanders; OR, odds ratio; Q-Tc, corrected Q–T interval; SES, socioeconomic status.