The prevalence of gout is on the rise worldwide, especially among newly industrialized populations. We evaluated the prevalence of gout in the recently established Hmong of Minneapolis/St. Paul (MSP) compared with that in non-Hmong populations.
The prevalence of self-reported gout in the Hmong population was estimated from 2 cross-sectional community surveys and compared with national data extrapolated from the Third National Health and Nutrition Examination Survey. The prevalence of physician-diagnosed gout in Hmong and non-Hmong MSP residents was separately estimated from the diagnosis codes of 11 MSP primary care clinics.
The prevalence of self-reported gout among MSP Hmong was 2-fold higher than in the general US population (6.5% versus 2.9%; P < 0.001). Although women of both groups reported gout at a rate of 1.9%, Hmong men were significantly more likely than their non-Hmong counterparts to report gout (11.5% versus 4.1%; P < 0.001). Similar results were observed when investigating physician-diagnosed gout in MSP (2.8% Hmong versus 1.5% non-Hmong; P < 0.001). No difference was observed between the women of the 2 groups (0.8% versus 0.7%; P = 0.833), whereas Hmong men were more than twice as likely to be diagnosed with gout compared with their non-Hmong counterparts (6.1% versus 2.5%; P < 0.001). Among Hmong men, advancing age was associated with a considerably higher likelihood of being diagnosed with gout.
A significant association is observed between Hmong ethnicity and gout, both self-reported and physician diagnosed. This unique population may provide an opportunity to further our understanding of the pathophysiology of gout.
Gout is a common rheumatic disease caused by elevated serum concentrations of uric acid, and affects more than 5 million men and women in the US (1). The risk of developing gout is dependent on a number of factors, including sex, age, ethnicity, and environment, although the interplay of these factors has not been well defined. Moreover, genetic factors play a significant role in the pathogenesis of gout, with both serum uric acid concentrations and renal uric acid clearance having been shown to have significant heritability (2, 3). Despite the difficulties in estimation of gout prevalence due to heterogeneous methodology, gout among different Western countries has progressively increased in the recent decade (4).
Over the past 90 years, serum uric acid concentrations have shown a significant increase from 3.5 mg/dl to 6–6.5 mg/dl in the general US male population (5, 6). In addition to the development of gouty arthritis, hyperuricemia has been associated with several conditions, such as the metabolic syndrome, hypertension, and chronic kidney disease (7, 8), although the relative importance of these associations has not been fully established (8–10).
An increased prevalence of gout is noted among indigenous, native populations concurrently with the adoption of a more Western lifestyle (11–13). A group of newly arrived Americans, the Hmong, initiated immigration following the conclusion of the Vietnam conflict in 1975. This ethnically distinct population of Laotian refugees has been observed to be uniquely susceptible to uric acid kidney calculi (14). As such, we sought to define the prevalence of gout among the Hmong of Minneapolis/St. Paul (MSP), Minnesota, and determine their status as a novel gouty population.
MATERIALS AND METHODS
To determine the prevalence of gout in the Hmong population living in MSP, in an Institutional Review Board–approved protocol, 2 cross-sectional community surveys were conducted and the billing data of the MSP HealthEast Care System were reviewed from 2001 through 2004. Data were then compared with non-Hmong populations to determine if there were significant differences in the prevalence of gout between Hmong and non-Hmong populations.
Community survey data collection.
The surveys were developed collaboratively by the Hmong HealthCare Professionals Coalition (HHCPC) and the research team. The survey was translated into Hmong and then back translated into English to ensure the accuracy of the translations. Both of the surveys included questions regarding age, sex, personal and family history of kidney stones, and gout. All of the volunteer surveyors attended a training session and were fluent in the Hmong language. Since some Hmong individuals are bilingual and prefer to converse in English, the surveys were completed verbally with the surveyor obtaining consent in Hmong and then conducting the survey in the language with which the participant was most comfortable. The first survey was conducted at the Hmong Sports Festival (July 4–5, 2003), where an estimated 40,000 Hmong were expected to attend. Surveyors were stationed at an HHCPC booth that was distributing educational materials on kidney stones. The attendees visiting the booth were invited to participate in a survey of gout in the Hmong people. Self-reported gout was defined as a positive response to the question, “Have you had gout?” The second survey was conducted at the Hmong New Year celebration (November 22, 2003) in St. Paul, Minnesota, where an estimated 30,000 Hmong were expected to attend. Volunteer surveyors were stationed throughout the event. Surveyors approached every fifth person who passed by and invited that individual to participate in a survey of gout in the Hmong people. Self-reported gout was defined as a positive response to the question, “Do you have gout?” The prevalence of gout in the US population was extrapolated from data collected via the Third National Health and Nutrition Examination Survey (NHANES-III; 1988–1994) (15). This nationwide survey was designed to obtain nationally representative information on the health and nutritional status of the US population through interviews and physical examinations. NHANES-III was conducted on a nationwide probability sample of approximately 33,994 persons ages >2 months. The NHANES-III question defining self-reported gout read, “Has a doctor ever told you that you have gout?”
Clinic data collection.
Billing data from the 11 primary care clinics belonging to the MSP HealthEast Care System were reviewed for the period extending from September 2001 through April 2004. The patients were divided into Hmong and non-Hmong groups based on surname. The surnames used to classify the Hmong group are listed in Table 1 (16). Patients with surnames not in Table 1 were classified as non-Hmong. The MSP HealthEast Care System billing system tracked up to 4 diagnoses for each patient visit to a clinic. If a patient had a diagnosis code of gout (International Classification of Diseases, Ninth Revision, code 274.0–274.9) during any of their visits they were classified as having gout. HealthEast Care System data, including age, sex, and gout diagnosis, were compared for Hmong and non-Hmong groups. These data were compared using Pearson's chi-square test to determine if the prevalence of gout was different for Hmong patients than for non-Hmong patients.
Table 1. Hmong surnames
A total of 325 people agreed to be interviewed at the Hmong Sports Festival. A response rate denominator is not available because the very small fraction of the event attendees visiting the HHCPC booth was not measured. Of the 404 individuals who were approached at the Hmong New Year celebration, 85 refused, resulting in a response rate of 79.0%. A combined 644 separate individuals participated at one of the events. Sixteen participants produced incomplete surveys (missing age, sex, or gout information) and 9 surveys were completed by minors. These cases were excluded, leaving 619 surveys for analysis. There were 19,618 adults ages ≥18 years who completed the NHANES-III survey. Three participants did not know if they had been told by a doctor that they had gout. The unsure individuals were excluded, leaving 19,615 surveys to analyze and compare with the Hmong survey results.
Among the Hmong survey respondents, 52.3% were women, compared with 53.1% of the NHANES-III respondents (P = 0.69). The mean ± SD age of the Hmong survey respondents was 43.6 ± 14.1 years compared with 48.2 ± 20.5 years for the NHANES-III respondents (P < 0.001). Whereas 6.5% of the Hmong survey respondents reported they had gout, only 2.9% of the NHANES-III respondents reported they had gout (P < 0.001). Although women of both groups reported gout at a rate of 1.9%, the Hmong men reported having gout significantly more frequently than their non-Hmong counterparts (11.5% versus 4.1%; P < 0.001). In particular, Hmong men in the age groups 40–49, 50–59, and 60–69 years reported gout considerably more often than the NHANES-III men in the same age groups (P < 0.001 for all 3 age groups) (Figure 1). There was no statistical difference observed between Hmong and non-Hmong women in any age group (Figure 1). No significant differences were observed between the sex-specific age groups of the 2 events.
Clinic billing data.
A total of 95,302 patients ages ≥18 years were seen at the MSP HealthEast Care System during the study period. Surnames common to both Hmong and other Asian ethnic groups (Lee and Chang) were excluded (641 total). Six additional patients were excluded because of missing sex information. A total of 94,655 patients remained for the analysis, including 1,294 identified as Hmong.
Among the Hmong HealthEast Care System patients, 61.8% were women, compared with 58.1% of the non-Hmong HealthEast Care System patients (P = 0.01). The mean ± SD age of the Hmong patients was 34.8 ± 15.8 years and the mean ± SD age of the non-Hmong patients was 43.2 ± 19.1 years (P < 0.001). Overall, there was a significant association between patient ethnicity and gout diagnosis. Whereas 2.8% of the Hmong patients had a gout diagnosis, only 1.5% of the non-Hmong were diagnosed with gout (P < 0.001). No significant difference was observed between the women of the 2 groups (0.8% versus 0.7%; P = 0.833). The Hmong men, however, were more than twice as likely to be diagnosed compared with their non-Hmong counterparts (6.1% versus 2.5%; P < 0.001). In particular, Hmong male patients in the age groups <30, 30–39, 40–49, 50–59, and ≥70 years carried a diagnosis of gout significantly more often than the non-Hmong male patients of the same age group (P = 0.001 for the age groups 40–49 and ≥70 years and P < 0.001 for all others) (Figure 2). The Hmong female patients in the age groups 50–59 and 60–69 years had a gout diagnosis significantly more often than the non-Hmong female patients in the same age groups (P = 0.04 and 0.001, respectively) (Figure 2).
In this study, we investigated the prevalence of gout in the MSP Hmong population by 2 methods. Self-reported gout was surveyed in Hmong individuals attending cultural events on 2 occasions, generating consistent results. Physician-reported gout was estimated from the billing data of a large primary care clinic system. Both self-reported and physician-reported gout in the Hmong population was found to be significantly more prevalent than in the non-Hmong comparison groups.
Hmong refer to gout as mob ko taw vwm (crazy foot pain), a term not far removed from the classic term podagra. Although to our knowledge no published data are available, Hmong elders relate that gout was an affliction in Laos, but that it is more prominent in the US. The Hmong have undergone a massive and abrupt geographic and cultural translocation. Modernization of other underdeveloped societies, most notably New Zealand Maoris and Taiwanese aborigines, has been associated with the emergence of widespread gout (11–13, 17–19). The prevalence of gout in these groups far exceeds that of Western industrialized populations, yet gout was not recognized until their adoption of Western habits. A comparison of gout prevalence in these cultures and the US is shown in Table 2. The higher prevalence of gout in some of these populations is likely related to dietary influences and the presence of other comorbid conditions. The Maori are exposed to all major risk factors, including high purine intake, alcohol consumption, obesity, glucose intolerance, and hypertension (17, 19, 20). Similarly, alcohol, central obesity, diabetes mellitus, and hypertension have been identified as significant risk factors for hyperuricemia and gout in Taiwanese aborigines (12, 13, 18, 21–23). Conversely, the prevalence of gout in a Vietnamese urban community was reported as being much lower, at 0.14% (24). However, it is noteworthy to mention that the genetic backgrounds of the Vietnamese and Laotian Hmong populations are distinctly different.
Knowledge of risk factors and biomarkers in the Hmong is limited. Immigration has provided an improved supply of meat, fish, and poultry (25, 26), and a dramatic increase in the intake of these protein sources has been noted. Alcohol is consumed by men on social occasions, but women usually do not drink (25, 26). Patterns consistent with a high prevalence of obesity and hypertension have been observed among American-born Hmong youth (27, 28). While recognized in the Hmong community as a “new” condition, diabetes mellitus is prevalent in adult Hmong (29). Although societal modernization is consistent with the presentation of other gouty populations, the Hmong experience was uniquely accelerated. Whereas other populations adopted Western habits in response to improved social and economic conditions in their native “homeland,” the Hmong immigrated en masse to the US as a sequela of the Vietnam conflict (16). The recency of the resettlement (the final group of refugees arrived in the second half of 2004) places this emerging, novel gouty population at the forefront of any discussion of pathophysiology and treatment.
Genetic factors may also contribute significantly to the overall disease burden of gout in the Hmong population, although the relative contribution of genetics to the pathogenesis of gout in this ethnic group is difficult to estimate at this time. The Hmong population may be an ideal candidate for this genetic exploration because the Hmong have a high cultural identity and a low marriage rate outside of their ethnic group. This cultural habit has been maintained from first-generation Hmong refugees. Furthermore, the young median age of the Hmong compared with the total US population (30) may allow exploration of the contribution of genetic factors as independent confounders in the pathogenesis of primary gout (31).
The higher prevalence of gout in the Hmong was particularly observed among men. This finding may result from a differential exposure of Hmong men to environmental factors (such as alcohol or dietary protein intake). Alternatively, genetic polymorphisms unique to the Hmong could influence serum uric acid concentrations with pronounced sex-specific effects (32). Interestingly, postmenopausal Hmong women exhibited a greater prevalence of gout than premenopausal women (Figure 2). This is likely a reflection of the reduced exposure to estrogen, which enhances renal clearance of uric acid (33). Another potential factor that could lead to a greater prevalence of gout in this subgroup is the use of diuretics, which are known to lower uric acid clearance and consequently increase serum uric acid levels.
The cross-sectional community surveys of the present study were subject to several limitations. First, the use of convenience samples limits the generalizability of the present data to the larger Hmong population. Second, the surveyors' interactions with the sample population were not parallel between the 2 events. In one case, all of the attendees who had approached a health education booth were solicited, whereas in the other, interviewers using a fixed selection procedure actively approached attendees. Third, to our knowledge, the validity of self-reported gout has never been examined against medical records. Relative to physician-reported medical history, self-reported use of gout medication has shown low sensitivity (68%) and high specificity (98%), suggesting that data reported in the current study may be an underestimation (34). Finally, the comparison of the present data with the general US population is open to critique due to differences between the NHANES-III survey and our self-report evaluation tools. In addition to wording differences, evaluation technique may have played a role in the results. Bergmann et al have observed that personal interviews were nearly twice as likely to detect a history of gout when compared with self-administered surveys, suggesting that observed differences in the present study may be the result of technique (35).
Our investigation of clinic data was also not without limitations. One obvious limitation is the validity of gouty diagnosis using medical record review, which may overestimate the incidence and prevalence of gout (36, 37). Additionally, we did not extract information on risk factors associated with gout (such as hypertension, type 2 diabetes mellitus, or obesity) or on medications used for the treatment of gout (allopurinol or colchicines). The use of billing records to identify the prevalence of gout is also complicated by a lack of health insurance coverage among immigrants. It is estimated that the foreign-born population is nearly 3 times more likely than native-born Americans to be without health insurance (38). As such, the Hmong could conceivably be less likely to seek care while being at a relatively higher risk of gout. The use of surname in defining ethnicity can also be tenuous in a multicultural society such as that of the US. Fortunately, most Hmong clan names are not common in non-Hmong populations. In addition, most Hmong people living in St. Paul, Minnesota, are first- or second-generation immigrants, and therefore it is assumed that surnames have not changed greatly due to marriage.
Current data strongly suggest that the Hmong American population of MSP has a unique predisposition to gout. Possible long-term consequences of this susceptibility include an increased risk for chronic kidney disease and cardiovascular disease (5, 6, 39). Although the adoption of a westernized lifestyle may have played a key role in the high incidence of gout in this population, further investigation is important to elucidate the relative roles of inherent and acquired risk factors. Increased understanding of gout pathophysiology and its potential genetic basis may result in improved care of both the Hmong and the general population at large.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Portis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Portis, Laliberte, Tatman, Moua, Culhane-Pera, Sakhaee.
Acquisition of data. Portis, Moua, Culhane-Pera.
Analysis and interpretation of data. Portis, Laliberte, Tatman, Moua, Culhane-Pera, Maalouf, Sakhaee.
We wish to recognize the critical assistance in Hmong linguistic and cultural translation and survey performance by the HHCPC, Mao Heu Thao, as well as Phua Xiong, MD, Peter Yang, MD, Bobby K. Yang, MD, Foung Lo, MD, and Kang Xiaaj, MD. We also wish to acknowledge the assistance of Ms Hadley Armstrong in the editorial review of this manuscript.