Complementary and Alternative Medicine Use Among a Multiethnic ...

THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE Volume 10, Number 6, 2004, pp. 1061–1066 © Mary Ann Liebert, Inc.

Complementary and Alternative Medicine Use Among a Multiethnic Sample of Older Adults with Diabetes NANCY E. SCHOENBERG, Ph.D.,1 ELEANOR PALO STOLLER, Ph.D.,2 CARY S. KART, Ph.D.,3 ADAM PERZYNSKI, M.A.,2 and ELIZABETH E. CHAPLESKI, Ph.D., M.S.W.4

ABSTRACT Objective: This study describes complementary and alternative medicine (CAM) use for diabetes self-management among a multiethnic sample with an aim of better understanding lay perspectives on CAM’s utility and determining whether CAM practices undermine conventional diabetes self-management. Design: During in-depth interviews with 80 older adults, data were collected on sociodemographics, the full range of self-management practices, and attitudes toward CAM. Analysis included descriptive measures of association and line-by-line coding. Setting/location: Trained interviewers recruited respondents from four health or social service sites. Sites were selected because they contained a large clientele of the targeted ethnic group and had been involved successfully in previous research studies. Subjects: Twenty (20) adults age 50 and older from each of the groups most adversely affected by diabetes (African Americans, Hispanics, Native Americans, rural whites) participated in the study. Outcome measures: Self-management strategies (included CAM) were assessed through a semistructured interview guide. Structured instruments obtained data on sociodemographics and health history. The 15-item Summary of Diabetes Self-Care Activities Questionnaire (SDSCA) assessed the frequency of completing recommended self-management activities. Results: One in four elders reported using CAM, with respondents’ cultural background associated with the CAM modality. We found no relationship between standard biomedical regimens and CAM use, supporting respondents’ suggestion that CAM supplements rather than substitutes for biomedical self-management. Respondents suggested that use of CAM was limited by CAM’s inaccessibility, and, underlying all, the dominance of conventional biomedical therapies that undermines belief in CAM’s effectiveness. Conclusion: Older adults with diabetes use a flexible configuration of diabetes–self-management techniques, including culturally specific CAM modalities. CAM use, however, occupies a limited role in diabetes self-management, largely because of the predominance of conventional biomedical regimens.

INTRODUCTION

A

ffecting approximately 17 million persons, or 6.2% of the U.S. population, type 2 diabetes is the leading cause of nontraumatic amputations, new cases of blindness and

end-stage renal disease and increases the risk for stroke, high blood pressure, and numerous other complications. In 1999, approximately 450,000 deaths, or nearly one fifth of all deaths, were attributed to diabetes (Centers for Disease Control, 2002). With 2200 and 800,000 new cases of daily and

1Departments

of Behavioral Science, Anthropology and Internal Medicine, University of Kentucky, Lexington KY. Western Reserve University, Cleveland, OH. 3Scripps Gerontology Center, Miami University, Oxford, OH. 4Institute of Gerontology, Wayne State University, Detroit, MI. 2Case

1061

1062 annually, respectively, type 2 diabetes is expected to increase by 165% by 2050 (Boyle et al., 2001). While this escalating rate is cause for considerable concern nationally, older adults and certain ethnic groups bear a disproportionate burden from diabetes. Approximately 7 million adults 65 and older years, or 20.1% of this age group, have diabetes (Pearce et al., 1997) and Native Americans, Mexican Americans, and African Americans have age/gender-standardized rates of diabetes 2.5, 2.0, and 1.7 times higher than non-Hispanic whites (Harris, 1998). These elevated rates raise numerous concerns, among them that these groups often lack the requisite resources to fully engage in self-management, the cornerstone of glycemic control (Mensing et al., 2002). To account for suboptimal adherence to conventional self-management, some researchers have suggested that complementary and alternative medicine (CAM), defined as “a group of diverse medical and health care systems, practices, and products that are not presently considered to be part of conventional medicine . . . ” (National Center for Complementary and Alternative Medicine, 2004), undermines diabetes regimens (Mensing et al., 2002). Although recent literature suggests that CAM is more often used as a complement than an alternative to conventional diabetes treatment (Egede et al., 2002; Yeh et al., 2002), further exploration of the interface between CAM and conventional diabetes management is warranted because we lack sufficient insights on CAM attitudes and practices among those most affected by diabetes, older adults, and traditionally underserved populations (McMahan and Lutz, 2004). Existing research reports that one half to two thirds of older adults use CAM strategies, with higher use patterns among women, older adults, those with higher educational levels, and more comorbidities (Barnes et al., 2004; McMahan and Lutz, 2004). The few reports of CAM prevalence among persons with diabetes indicate a greater likelihood of CAM use, although less often for diabetes and more directed at musculoskeletal problems (Egede et al., 2002). Debate continues on the prevalence of CAM use among traditionally underserved populations. While important factors including cultural traditions, historical exclusion from conventional medical care, and the lower costs of some CAM approaches might lead to greater CAM use among these groups (Arcury et al., 2003; Schoenberg and Drew, 2002), some research documents more limited CAM use, particularly to manage chronic conditions (Egede et al., 2002; Hunt et al., 2000). To better understand the prevalence of CAM use and to explore whether such CAM practices undermine conventional recommendations for diabetes management, this article focuses on three questions: (1) Do African-American, Hispanic, Native-American, and rural White adults employ CAM practices in their diabetes self-management repertoires?; (2) If so, what types of CAM practices are employed and for what purpose?; and (3) Does the use of such strate-

SCHOENBERG ET AL. gies undermine biomedically standard approaches to selfmanagement?

MATERIALS AND METHODS For each of the four groups—African Americans, Hispanics (all of Mexican descent and the majority of whom were native-born), Native Americans (Great Lakes Indians), and rural whites—interviews were conducted with 20 respondents 50 years of age or older identified as having diabetes. Rural whites were included because they face many of the same risk factors for poor diabetes outcomes, including compromised access to health services, low income, and education level (Harris, 1998). Participants were recruited from four sites (three health clinics and one senior center) predominantly attended by members of the targeted group. While recruitment at conventional health clinics may have the unintended consequence of enrolling participants with a proclivity toward convention medicine, we found this venue essential to insure adequate numbers of participants with diabetes. In addition, nearly all older adults use a conventional provider, even those who use CAM, thus minimizing concerns about excluding CAM users (Cherniack et al., 2001; McMahan and Lutz, 2004). Staff at each site identified individuals with diabetes either by personal knowledge or chart review, invited eligible individuals to participated, and, if acceptable, followed up and conducted the interview. During that interview, all participants provided informed consent and questionnaires were administered. Participants received a $15 honorarium. All protocols were approved by the Institutional Review Boards. Interviews generally took between 45 and 90 minutes and were in two parts; semistructured questions on sociodemographics, health history, and explanatory models of diabetes and a 15-item Summary of Diabetes Self-Care Activities Questionnaire (SDSCA) to assess the frequency of completing different regimen activities during the preceding seven days (Toobert et al., 2000). Because this study focused on social and behavioral factors associated with CAM use and diabetes regimen, no clinical data were gathered. For the quantitative analyses, CAM variables (type, frequency of mention, and use) were combined with the sociodemographic, health history, and self-management variables and entered into an SPSS (SPSS Inc., Chicago, IL) data file to ascertain the distribution of CAM use and other relevant variables, which are operationalized in Tables 1 and 2. To obtain both qualitative and quantitative insights, we used a modest sample size and undertook purposive sampling; thus, our quantitative analyses consist of nonparametric statistics, including the Kruskal-Wallis one-way analysis of variance and the Spearman rank correlation coefficient. For the qualitative analyses, upon completion of transcription of the tape-recorded interviews, the research

1063

CAM USE IN DIABETES TABLE 1. SELECTED CHARACTERISTICS

Age (Mean, in years) (SD) Gender Female Male Marital status Married Divorced Widowed Other Educationa (years completed) Employment status Retired Employed Disabled Other Health statusb,c Mean (SD)

OF

STUDY PARTICIPANTS (n
80)

African American (n
20)

Hispanics (n
20)

Native American (n
20)

Whites (n
20)

Total (n
80)

69.30 (SD
12.20)

69.60 (SD
8.98)

66.15 (SD
5.88)

65.50 (SD
8.41)

67.64 (SD
9.16)

70.0% 30.0%

68.4% 31.6%

60.0% 40.0%

64.7% 35.3%

65.8% 34.2%

50.0% 15.0% 25.0% 10.0% 11.65 (SD
3.65)

45.0% 5.0% 45.0% 5.0% 5.50 (SD
4.96)

40.0% 20.0% 35.0% 5.0% 12.10 (SD
2.49)

40.0% 10.0% 50.0% — 7.80 (SD
3.12)

43.8% 12.5% 38.8% 5.0% 9.26 (SD
4.53)

55.0% 15.0% 25.0% 5.0% 3.10 (0.97)

55.5% 10.0% 20.0% 15.0% 3.35 (0.93)

63.2% 26.3% 5.3% 5.3% 2.95 (1.23)

35.0% 10.0% 35.0% 20.0% 3.85 (0.99)

51.9% 15.2% 21.5% 11.4% 3.31 (1.07)

Test, p  0.000. Test, p  0.041. cSelf-reported, on scale from 1–5, where 1
excellent, 2
very good, 3
good, 4
fair, and 5
poor. SD, standard deviation. aKruskal-Wallis

bKruskal-Wallis

team initiated line-by-line coding to identify and describe CAM practices. We highlighted chunks of text that contained references to any management strategy outside of conventional biomedical diabetes management, attaching codes to these text segments. Then, researchers compiled the codes into codebooks, which we merged after numerous refinement and elaboration. The final codebook consisted of a set of quantifiable variables with corresponding categories. TABLE 2. FREQUENCY (PERCENT)

OF

RESULTS Quantitative data Sample. Table 1 summarizes the sample’s characteristics. A majority of the sample was female (65.8%), mean age was 68 years, and mean level of education was 9.3 years. Most respondents were either married or widowed and, aside CAM STRATEGIES

BY

ETHNIC GROUP

Ethnic group African American CAM Strategy Dietary remedies Teas Herbal remedies Spiritual interventions Other Any of the above CAMc n

1 1 2 1 1 4 20

(5%)b (5%) (10%) (5%) (5%) (20%)

Hispanic 3 2 8 1 1 10 20

(15%) (10%) (40%) (5%) (5%) (50%)

Native american 2 1 1 1 2 3 20

(10%) (5%) (5%) (5%) (10%) (15%)

Rural white 1 0 1 1 0 3 20

(5%) (0%) (5%) (5%) (0%) (15%)

Totala 7 4 12 4 4 20 80

(9%) (5%) (15%) (5%) (5%) (25%)

aThis column indicates the number and percentage of the total sample that indicate use of a particular CAM strategy. Thus, 9% of the total sample, or 7 participants, indicated that they selected a specific food to reduce their blood sugar. bThese figures indicate the number and the percentage of individuals from a particular group who indicated using the specific CAM strategy listed for control of diabetes. Thus, for dietary remedies one African American elder indicated a particular food that she took to control blood sugar. cThis row represents the number and the percentage of individuals from a particular group who indicated their use of any of the CAM strategies. Because the same person may list multiple CAM strategies, this row’s total may be less than the sum total of CAM strategies. Indeed, this is the case for all groups except rural whites. CAM, complementary and alternative medicine.

1064 from the white group, the majority was retired. Statistically significant differences emerged in education and health status. CAM use. One in four respondents mentioned using some type of CAM procedure for managing their diabetes (Table 2). CAM use overall was higher among Hispanic respondents (50%) than African American (20%), Native American (15%), or rural whites (15%) respondents. Many of the respondents indicating use of CAM for diabetes used only one modality, including all rural whites and two thirds of Hispanics. Among African-American and Native American respondents, one half and two thirds, respectively, of those who reported CAM use selected multiple modalities. No differences in these multiple use patterns were distinguishable for education, gender, age, or other characteristics. Self-management activities and CAM. To explore the impact of reported CAM use on self-management of diabetes, we examined bivariate associations between CAM use and each element of conventional self-management. In data not shown, neither the individual CAM strategies reported in Table 2 nor the composite indicator (i.e., uses any CAM strategy) demonstrated any meaningful association with any self-management strategies. These results suggest that CAM use is independent of and does not impact upon elderly patients with diabetes use of conventional self-management approaches.

Qualitative data Respondents described a wide array (although not a widespread use) of CAM approaches for diabetes including spiritual practices, nutritional supplements, herbal or plant treatments, and dietary and exercise approaches that fall outside conventional recommendations. In addition, narrative analyses drew out the following themes: (1) respondents use CAM approaches when emphasizing a holistic, lifestyle approach to diabetes self-management; (2) CAM use reflects respondent’s background; and (3) respondents note several reasons for a limited use of CAM approaches, including lack of availability of the modality, underlain by the influence of biomedical practice that strongly promotes the efficacy of conventional management. CAM is a useful part of holistic diabetes self-management. Nearly all of those respondents reporting use of CAM and approximately one quarter of those who did not report CAM use suggested that unconventional self-management strategies fit a general prescription for a healthy lifestyle. For example, in addition to explaining diabetes-based prohibitions regarding sugars, starches, and fat, some respondents warned against consuming specific foods while promoting other items. While many emphasized the importance of multiple daily servings of vegetables, others specified

SCHOENBERG ET AL. fruits, particularly bananas, as important to maintaining stable blood sugar. A few respondents mentioned the efficacy of herbal teas, particularly green tea, and nutritional supplements for keeping their diabetes under control. In addition, many respondents concur with the conventional notion that stress elevates blood sugar, but took this general lifestyle issue one step further by integrating strategies to control their stress and blood sugar, including prayer, meditation, and yoga. CAM use appears associated with ethnic/residential background. Approximately half of those respondents who describe using CAM modalities tied the approaches to their background or cultural practices, particularly in the realm of food or beverage choices that fall slightly outside of conventional dietary management strategies. For example, in discussing the importance of a healthful diet, one quarter of the Native American sample explained that their high prevalence of diabetes resulted from the intrusion of the “white man’s diet” and the loss of native foods, a situation many try to rectify by incorporating more traditional food such as buffalo. Because many Native Americans in our study viewed alcohol as a contributing factor in diabetes onset, moderating or eliminating alcohol intake was viewed as a way to prevent diabetes onset and to control blood sugar once diagnosed. Consistent with previous research, Hispanics reported a broad range of remedies, including teas brewed from cat’s claw (uña de gato) plant or corn silk. Others mentioned ingesting aloe or eating cooked cactus (nopales), making tea from sheep stool, and taking a spoonful of one’s own urine mixed with olive oil to manage diabetes. Religious or spiritual CAM modalities were also grounded in cultural background. Several Native American respondents recommended involvement in sweat lodges, native healing ceremonies, and dance rituals. Several rural whites reported appealing directly to God for intervention, whereas Hispanics mentioned prayer during Mass. Several reasons underlie limited use of CAM. Despite this variety of CAM approaches, respondents’ narratives also reveal several reasons for not using CAM strategies. First, among those rejecting CAM, 1 in 10 individuals questioned the efficacy of alternative approaches, particularly in light of their costliness. A 55-year-old rural white man said: I’ve never taken anything other than what the doctors here have given me. Uh, well, I’ll take that back. I have tried a couple of vitamins, seems like somebody told me to take some potassium one time and some other kind of vitamin, some kind of acid, well, I forgot the name of it. But it didn’t seem to make any difference. It seems like it was just a ten dollar waste. Approximately one quarter of the sample, nearly all of whom are Hispanic, described being unfamiliar with where to pur-

1065

CAM USE IN DIABETES chase CAM products. For example, a 77-year-old Hispanic woman from Ohio indicates that she doesn’t use alternative dietary approaches “because up here it’s hard to find nopales [cactus] and all this stuff . . . in California or Arizona and all the places, you can find it.” Underlying these explanations is a view expressed by more than half of the sample that the dominating role played by the biomedical sphere in diabetes management has eclipsed CAM use. This is exemplified in statements like “I don’t believe in that kind of stuff” and “(The) only approaches I use are the ones that the doctors gave me and the nurse. Those are the only ones that I use.” A 72-year-old African American woman with diabetes for 18 years suggests, “(Do) what the doctor tell you to do and take your medication. Your answer is in all your pills, do whatever he says. That’s controlling it.”

DISCUSSION This study demonstrates several findings that advance knowledge about CAM and diabetes self-management among four groups at elevated risk from the disease. Specifically, older adults use a flexible configuration of diabetes self-management that integrates a predominantly biomedical regimen with a more holistic and culturally shaped regimen; that a modest but meaningful number of these elders use CAM modalities that complement rather than substitute for conventional approaches; and that evaluations of CAM’s utility is assessed using criteria similar to that for conventional therapies—cost, availability, and the high degree of efficacy of conventional therapies perceived and promoted by HCPs. Consistent with other recent studies, we found CAM use occupies a distinctive, although modest, component of overall diabetes self-management (Egede et al., 2002; Hunt et al., 2000; Yeh et al., 2002). Specifically, one in four respondents mentioned using CAM for diabetes control, with Hispanics indicating the use of CAM therapies more often than any other group. Our results corroborate existing reports on CAM use among Hispanic elders, much of which indicates use of herbal medicines and other practices considered outside of conventional regimens (Hunt et al., 2000; Loera et al., 2001; Mercado-Martinez and Ramos-Herrera, 2002; Rivera et al., 2002). Speculations about the origins of these distinct CAM use patterns have focused on the effects of acculturation or assimilation, language, and socioeconomic status (Markides and Black, 1996). While some of these same factors may be said to apply to Native Americans, research conducted among a sample of urban Great Lakes Indians similar to our study participants suggests a greater level of acculturation from their rural and reservation counterparts in their use of traditional and conventional health services (Chapleski et al., 1997; Jackson and Chapleski, 2000). In addition, our results corroborate other reports that

CAM practices may supplement but do not undermine conventional management efforts (Egede et al., 2002; Hunt et al., 2000; Yeh et al., 2002). By comparing conventional self-management activities with CAM use patterns, we have found unwarranted the often stated concern that patients’ preference for alternative modalities undermines their conventional self-management activities and, by extension, their glycemic control. Those participants using CAM often describe it as one of several simultaneous modalities that fits into a total package of self-management approaches. Finally, our semistructured interviews allowed us to explore the ways in which respondents evaluate the utility of CAM approaches. Results suggest that people use the same criteria for judging the effectiveness of CAM treatments that they use for conventional treatments: the accessibility of the treatment (both cost and availability) and their assessments of the modality’s effectiveness (Sirois and Gick, 2002). In environments that may not offer extensive CAM alternatives and instead strongly promote the efficacy of pharmaceutical agents, diabetes management is heavily toward conventional therapies. Moreover, because participants generally regard their HCPs as the key managers of their diabetes care, the message that conventional medical care has developed extremely effective and relatively easily administered pharmaceutical agents to control diabetes, appears to have become internalized.

LIMITATIONS AND CONCLUSIONS The exploratory research design in this study, particularly our use of nonprobability samples, limits the generalizability of our results. Because we selected the field sites to ensure the availability of disproportionately affected population groups, there is a possibility of confounding geographical and ethnic differences. The diversity of fieldsites is, in our view, appropriate and desirable because our intention was to descriptively gauge lay perspectives from the four groups to whom diabetes represents a great burden rather than to estimate the distribution of CAM perspectives across groups. Finally, while it is beyond the scope of this study to evaluate whether the use of CAM influenced clinical outcomes, determining such an impact holds considerable significance. Despite these limitations, study findings diminish concerns about CAM undermining conventional diabetes selfmanagement approaches and instead underscore the powerful role that conventional HCPs play in the management of their patients’ diabetes care. With this in mind, HCPs are advised to acquaint themselves with more commonly used CAM approaches, nonjudgmentally discuss CAM modalities, and in fact, encourage the use of particular CAM approaches that enhance conventional therapies and that derive from culturally based practices.

1066

SCHOENBERG ET AL.

ACKNOWLEDGMENTS This research has been support by a grant from the National Institutes of Health/National Institute on Aging (#AG17347) made to Cary S. Kart. We gratefully acknowledge the support of The Ohio Long-Term Care Research Project, Scripps Gerontology Center, Miami University, and the other fieldsites where this project took place.

REFERENCES Arcury T, Quandt A, Bell R, Mellen B, Smith S, Skelly A, Wetmore L. Gender-ethnic differences in the use of CAM for treating diabetes among rural older adults. Diabetes 2003;52:410. Barnes PM, Powell-Griner E, McFann K, Nahin RL. [no. 343] Advance data from vital and health statistics. Complementary and Alternative Medicine Use Among Adults: United States, 2002. Hyattsville, MD: National Center for Health Statistics, 2004:343. Boyle JP, Honeycutt AA, Venkat Narayan KM, Hoerger TJ, Geiss LS, Checn H, Thompson TJ. Projection of diabetes burden through 2050. Diabetes Care 2001;24:1936–1940. Centers for Disease Control. Online document at: www.cdc.gov/ diabetes/pubs/estimates.htm 2002. Accessed September 10, 2003. Chapleski E, Gelfand D, Pugh K. Great Lakes American Indian elders and service utilization: Does residence matter? J Appl Gerontol 1997;16:333–354. Cherniack EP, Senzel RS, Pan CX. Correlates of use of alternative medicine by the elderly in an urban population. J Altern Complement Med 2001;7:277–280. Egede LE, Ye X, Zheng D, Silverstein MD. The prevalence and pattern of complementary and alternative medicine use in individuals with diabetes. Diabetes Care 2002;25:324–329. Harris MI. Diabetes in America: Epidemiology and scope of the problem. Diabetes Care 1998;21:11–14. Hunt LM, Arar NH, Akan LL. Herbs, prayer, and insulin. J Family Pract 2000;49:216–223. Jackson DD, Chapleski EE. Not traditional, not assimilated: Elderly American Indians and the notion of cohort. J Cross Cultural Gerontol 2000;5:229–259. Loera JA, Black SA, Markides KS, Espino DV, Goodwin JS The use of herbal medicine by older Mexican Americans. J Gerontol 2001;56:714–719.

Markides KS, Black SA Aging and health behaviors in Mexican Americans. Family Commun Health 1996;19:11–19. McMahan S, Lutz R. Alternative therapy use among the young old (65 to 74): An evaluation of the MIDUS database. J Appl Gerontol 2004;23:91–103. Mensing C, Boucher J, Cypress M, Weinger K, Mulcahy K, Barta P, Hosey G, Kopher W, Lasichak A, Lamb B, Mangan M, Norman J, Tanja J, Yauk L, Wisdom K, Adams C. National standards for diabetes self-management education. Diabetes Care 2002;25:S140–147. Mercado-Martinez FJ, Ramos-Herrera IM. Diabetes: The layperson’s theories of causality. Qual Health Res 2002;12:792–806. National Center for Complementary and Alternative Medicine. Online document at: http://nccam.nih.gov/ Accessed August 2004. Pearce D, Griffin T, Kelly J, Mikkelsen L. An overview of the population in Europe and North America. Population Trends 1997;89:24–36. Rivera J, Ortiz M, Lawson M, Verma K. Evaluation of the use of complementary and alternative medicine in the largest United States–Mexico border city. Pharmacotherapy 2002;22:256–264. Schoenberg NE, Drew EM. Articulating silence: Experiential certitude and biomedical controversies over hypertension symptomatology. Med Anthropol Q 2002;16:458–475. Sirois FM, Gick ML. An investigation of the health benefits and motivations of complementary medicine clients. Social Science Med 2002;55:1025–1037. Toobert DT, Hampson SE, Glasgow RE. The summary of diabetes self-care activities measure: Results from 7 studies and a revised scale. Diabetes Care 2000;23:943–950. Yeh GY, Eisenberg DM, Davis RB, Phillips RS. Use of complementary and alternative medicine among persons with diabetes mellitus: Results of a National Sample. Am J Public Health 2002;92:1648–1652.

Address reprint requests to: Nancy E. Schoenberg, Ph.D. Departments of Behavioral Science, Anthropology, and Internal Medicine University of Kentucky 125 College of Medicine Office Building Lexington, KY 40536-0086 E-mail: [email protected]