We examined the National Health Interview Survey (NHIS) 2012 to explore how US adult consumers of CAM differ by gender in terms of their sociodemographic characteristics, current health conditions, and perceived benefits of CAM. All individuals who completed the adults core interviews (
The use of complementary and alternative medicine (CAM) amongst adults is substantial in both the United States (USA) [
A 2010 review of surveys investigating CAM use among community-based adults indicated an association between CAM use and gender, with women more likely than men to use CAM [
Gender differences in social determinants of health and illness, as well as health care decision-making, have been explored by various researchers over the past two decades [
To date, few have examined in depth the factors that differentiate male and female consumers of CAM. In response to this significant research gap, this paper reports the first focused analysis of gender differences in CAM use amongst US adults. Using data from the 2012 NHIS, this study specifically aimed to explore how US adult consumers of CAM differ by gender in terms of their sociodemographic characteristics, current health conditions, and perceived benefits of CAM.
This study is a secondary analysis of 2012 National Health Interview Survey data. NHIS is a cross-sectional household interview survey conducted periodically by Centers for Disease Control and Prevention’s (CDC) National Center for Health Statistics (NCHS). The target population for the NHIS is the civilian noninstitutionalized population of the United States. The core questionnaires provide information on demographics, health status, health behaviors, and health care access and utilization. Supplemental questions on CAM use (i.e., Adult CAM Supplement) are collected every five years on randomly selected members of a household, including one adult (18 years or older) and one child (0–17 years old). The total household response rate for 2012 was 77.6%. The interviewed sample consisted of 42,366 households, which yielded 108,131 persons in 43,345 families, including 34,525 persons being 18 years of age and older. The conditional response rate for the sample adult component was 79.7% (i.e., the number of completed sample adult interviews [
In the 2012 NHIS, the Adult CAM Supplement collected information from sample adults regarding their use of 18 nonconventional health care practices, including acupuncture, Ayurveda, biofeedback, chelation therapy, chiropractic or osteopathic manipulation, craniosacral therapy, energy healing therapy, hypnosis, massage, naturopathy, traditional healing, movement therapy (Pilates/Trager psychophysical integration/Feldenkrais), herbal and nonvitamin supplementation, vitamin and mineral supplementation, homeopathy, special diets, yoga/tai chi/qi gong, and relaxation techniques (meditation/guided imagery/progressive relaxation). While some questions were asked of each health care practice, other questions were asked only for the top three modalities deemed by the respondent to be the most important to their health [
Based on the study objectives, CAM use, major sociodemographic variables, and perceived benefits of using CAM, as well as the most popular health conditions for which CAM was used, were selected as the main variables of the study, each of which are defined below.
reasons for using CAM for the first top therapy (i.e., general wellness or general disease prevention, improving energy, improving immune function, improving athletic or sports performance, and improving memory or concentration); motivations for using CAM for the first top therapy (i.e., eating healthier, eating more organic foods, cutting back on or stop drinking alcohol, and doing exercise more regularly); outcomes of CAM use for the first top therapy (i.e., sense of control over one’s health, reduced stress level or relaxation, better sleep, feeling better emotionally, coping with health problems easier, improved overall health and feeling better, improved relationships with others, and improved attendance at job or school).
Two questions regarding the helpfulness of CAM use were also included under perceived benefits. The two questions asked how much the first top therapy helped with the most important reasons for CAM use and with the specific health problems, respectively. Responses to these two questions were a great deal, some, only a little, or not at all.
Analyses were performed using Stata 9.0 (Stata Statistical Software: Release 9. College Station, TX: StataCorp LP). All analyses used the NHIS Sample Adult Weight—Final Annual (WTFA_SA) including design, ratio, nonresponse, and poststratification adjustments for sample adults [
In NHIS 2012, 34,325 adults were included in the Adult CAM Supplement subset. Of these, 29.6% (
Weighted sociodemographic characteristics of non-CAM users and CAM users and male and female CAM users.
Non-CAM user
( |
CAM user
( |
|
CAM users |
|
||
---|---|---|---|---|---|---|
Male ( |
Female ( |
|||||
Age, mean (95% CI) | 46.5 (46.2, 46.8) | 47.0 (46.5, 47.7) | 0.070 | 47.3 (46.7, 47.9) | 46.7 (46.2, 47.2) | 0.174 |
Sex | <0.001 | |||||
Male | 51.2 | 40.9 | ||||
Female | 48.8 | 59.1 | ||||
Region of USA ( |
<0.001 | 0.22 | ||||
Northeast | 18.7 | 17.0 | 15.9 | 17.7 | ||
Midwest | 21.5 | 25.6 | 25.3 | 25.8 | ||
South | 39.7 | 28.8 | 29.3 | 28.5 | ||
West | 20.1 | 28.6 | 29.6 | 28.0 | ||
Race/ethnicity ( |
<0.001 | 0.93 | ||||
Hispanic | 17.2 | 9.5 | 9.5 | 9.4 | ||
Non-Hispanic White | 62.8 | 77.4 | 77.4 | 77.4 | ||
Non-Hispanic Black | 14.0 | 6.8 | 6.9 | 6.8 | ||
Non-Hispanic Asian | 5.2 | 5.5 | 5.6 | 5.5 | ||
Non-Hispanic Other | 0.8 | 0.8 | 0.7 | 0.8 | ||
Education | <0.001 | 0.001 | ||||
Less than high school | 20.6 | 8.3 | 9.8 | 7.3 | ||
High school graduate and some degree | 56.3 | 50.9 | 50.1 | 51.5 | ||
Bachelor degree | 15.7 | 24.7 | 23.2 | 25.7 | ||
Master degree and higher | 7.4 | 16.1 | 16.9 | 15.5 | ||
Personal earning in the past year (US$) | <0.001 | <0.001 | ||||
<$10,000 | 8.1 | 8.2 | 6.7 | 9.2 | ||
$10,000–$19,999 | 9.3 | 8.1 | 6.6 | 9.2 | ||
$20,000–$34,999 | 11.8 | 12.6 | 11.9 | 13.1 | ||
$35,000–$54,999 | 11.0 | 14.0 | 14.9 | 13.3 | ||
$55,000–$74,999 | 5.3 | 8.9 | 10.4 | 7.9 | ||
$75,000+ | 6.7 | 12.1 | 18.6 | 7.6 | ||
Refused to report or do not know | 11.4 | 9.0 | 9.1 | 8.9 | ||
Did not work in the past year | 36.4 | 27.1 | 21.7 | 30.9 | ||
Marital status/relationship | <0.001 | <0.001 | ||||
Married or living with a partner | 58.6 | 64.4 | 68.3 | 61.6 | ||
Divorced or separated | 11.2 | 11.5 | 9.6 | 12.8 | ||
Widowed | 6.5 | 4.9 | 1.9 | 6.9 | ||
Never married | 23.6 | 19.3 | 20.1 | 18.8 | ||
Family spending on medical care | <0.001 | 0.12 | ||||
0 | 12.9 | 7.0 | 7.6 | 6.6 | ||
$1–499 | 35.9 | 29.2 | 28.0 | 30.0 | ||
$500–1999 | 30.1 | 34.3 | 34.5 | 34.2 | ||
$2000–2999 | 9.7 | 11.8 | 12.9 | 11.1 | ||
$3000–4999 | 5.6 | 9.1 | 8.7 | 9.4 | ||
$5000+ | 5.8 | 8.6 | 8.3 | 8.8 |
Of the CAM users, nearly 60% were women. In addition, female CAM users were significantly more likely than male CAM users to use more than one CAM modality (male:
Using the total sample, logistic regression revealed that women were about three times more likely than men to use CAM (OR = 2.8; 95% CI: 2.5, 3.0), while race/ethnicity, education, personal earnings in the past year, family medical expenses, and marital status were all associated with CAM use (Table
Weighted logistic regression models of CAM use*.
Overall model | Men only | Women only | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
OR† | 95% CI‡ |
|
OR | 95% CI |
|
OR | 95% CI |
| ||||
Female1 | 2.75 | 2.50 | 3.03 | 0.00 | ||||||||
Region of USA | ||||||||||||
Midwest2 | 1.43 | 1.24 | 1.65 | 0.00 | 1.19 | 0.94 | 1.50 | 0.16 | 1.69 | 1.39 | 2.05 | 0.00 |
South | 0.70 | 0.60 | 0.82 | 0.00 | 0.87 | 0.70 | 1.09 | 0.24 | 0.63 | 0.52 | 0.78 | 0.00 |
West | 2.87 | 2.46 | 3.35 | 0.00 | 1.78 | 1.41 | 2.26 | 0.00 | 3.21 | 2.61 | 3.93 | 0.00 |
Race/ethnicity | ||||||||||||
Non-Hispanic White3 | 3.01 | 2.70 | 3.36 | 0.00 | 5.93 | 4.45 | 7.91 | 0.00 | 2.11 | 1.81 | 2.46 | 0.00 |
Non-Hispanic Black | 0.70 | 0.58 | 0.84 | 0.00 | 3.03 | 2.11 | 4.33 | 0.00 | 0.43 | 0.34 | 0.54 | 0.00 |
Non-Hispanic Asian | 0.14 | 0.11 | 0.18 | 0.00 | 0.44 | 0.25 | 0.76 | 0.00 | 0.88 | 0.65 | 1.19 | 0.40 |
Non-Hispanic Other | 1.34 | 0.81 | 2.23 | 0.26 | 3.47 | 1.67 | 7.21 | 0.00 | 1.07 | 0.53 | 2.18 | 0.85 |
Education | ||||||||||||
High school graduate and some degree4 | 1.15 | 1.03 | 1.30 | 0.02 | 2.60 | 1.93 | 3.50 | 0.00 | 0.89 | 0.76 | 1.04 | 0.13 |
Bachelor degree | 1.93 | 1.65 | 2.26 | 0.00 | 4.00 | 2.88 | 5.55 | 0.00 | 1.46 | 1.19 | 1.81 | 0.00 |
Master degree and higher | 5.33 | 4.49 | 6.34 | 0.00 | 5.58 | 3.93 | 7.91 | 0.00 | 5.91 | 4.64 | 7.53 | 0.00 |
Personal earning in the past year (US$) | ||||||||||||
$10,000–$19,9995 | 1.07 | 0.85 | 1.36 | 0.56 | 0.91 | 0.61 | 1.36 | 0.65 | 1.00 | 0.75 | 1.34 | 0.99 |
$20,000–$34,999 | 1.39 | 1.12 | 1.72 | 0.00 | 1.13 | 0.80 | 1.59 | 0.49 | 1.19 | 0.90 | 1.57 | 0.22 |
$35,000–$54,999 | 0.76 | 0.61 | 0.94 | 0.01 | 0.50 | 0.35 | 0.71 | 0.00 | 1.19 | 0.89 | 1.58 | 0.24 |
$55,000–$74,999 | 1.66 | 1.29 | 2.14 | 0.00 | 1.02 | 0.70 | 1.48 | 0.92 | 1.52 | 1.06 | 2.18 | 0.02 |
$75,000+ | 1.51 | 1.19 | 1.93 | 0.00 | 1.03 | 0.73 | 1.45 | 0.88 | 1.23 | 0.83 | 1.80 | 0.30 |
Refused to report or do not know | 2.30 | 1.87 | 2.82 | 0.00 | 0.68 | 0.47 | 0.97 | 0.04 | 3.71 | 2.81 | 4.89 | 0.00 |
Did not work in the past year | 1.10 | 0.93 | 1.31 | 0.27 | 0.79 | 0.58 | 1.08 | 0.14 | 0.89 | 0.71 | 1.12 | 0.33 |
Marital status/relationship | ||||||||||||
Divorced or separated6 | 1.57 | 1.39 | 1.76 | 0.00 | 1.11 | 0.91 | 1.36 | 0.30 | 2.06 | 1.75 | 2.41 | 0.00 |
Widowed | 2.06 | 1.81 | 2.34 | 0.00 | 3.06 | 2.41 | 3.89 | 0.00 | 2.85 | 2.39 | 3.39 | 0.00 |
Never married | 0.99 | 0.85 | 1.14 | 0.84 | 0.95 | 0.78 | 1.16 | 0.61 | 0.97 | 0.79 | 1.19 | 0.76 |
Family spending on medical care | ||||||||||||
$1–499 | 0.71 | 0.62 | 0.81 | 0.00 | 1.50 | 1.17 | 1.93 | 0.00 | 0.50 | 0.42 | 0.60 | 0.00 |
$500–1999 | 0.61 | 0.52 | 0.71 | 0.00 | 0.98 | 0.75 | 1.28 | 0.90 | 0.68 | 0.56 | 0.84 | 0.00 |
$2000–2999 | 1.84 | 1.55 | 2.18 | 0.00 | 0.83 | 0.59 | 1.18 | 0.30 | 3.98 | 3.12 | 5.08 | 0.00 |
$3000–4999 | 1.07 | 0.84 | 1.37 | 0.58 | 1.79 | 1.24 | 2.60 | 0.00 | 0.96 | 0.71 | 1.31 | 0.81 |
$5000+7 | 0.92 | 0.73 | 1.17 | 0.51 | 1.29 | 0.90 | 1.86 | 0.17 | 1.01 | 0.73 | 1.38 | 0.97 |
1Reference = male; 2reference = Northeast; 3reference = Hispanics; 4reference = less than high school; 5reference = less than $10,000; 6reference = married or living together; 7reference = 0.
Higher education was associated with CAM use among both men and women, with the exception of an insignificant difference when comparing high school to less than high school among women. When compared to no family medical expenses, higher expenses in women were not associated with a higher likelihood of CAM use, but expenses between $3000 and $4999 were associated with more CAM use in men (OR = 1.79; 95% CI: 1.24, 2.60). Both widowed men (OR = 3.06; 95% CI: 2.41, 3.89) and women (OR = 2.85; 95% CI: 2.39, 3.39), as well as divorced or separated women (OR = 2.06; 95% CI: 1.75, 2.41), were more likely to use CAM compared to men/women who were married and living together.
Figure
Top 10 health problems that men and women used CAM for.
The perceived benefits of CAM use, as reported by the men and women in our analyses, are presented in Table
Comparison of the reasons, motivations, and outcomes for using the first top CAM therapy by gender (
Perceived benefits | Male ( |
Female ( |
|
---|---|---|---|
Reasons | |||
For general wellness or general disease prevention | 62.2 | 66.8 | <0.001 |
To improve energy | 28.4 | 38.5 | <0.001 |
To improve immune function | 23.2 | 28.4 | <0.001 |
To improve athletic or sports performance | 23.7 | 18.3 | <0.001 |
To improve memory or concentration | 16.0 | 18.2 | 0.024 |
Motivations | |||
To eat healthier | 20.2 | 27.3 | <0.001 |
To eat more organic foods | 11.9 | 14.2 | 0.009 |
To cut back on or stop drinking alcohol ( |
7.0 | 7.1 | 0.873 |
To cut back on or stop smoking cigarettes ( |
13.3 | 15.1 | 0.427 |
To exercise more regularly | 22.8 | 30.0 | <0.001 |
Outcome | <0.001 | ||
Gave a sense of control over one’s health | 36.8 | 43.8 | <0.001 |
Reduced stress level or relaxation | 41.9 | 54.5 | <0.001 |
Better sleep | 36.1 | 43.4 | <0.001 |
Feeling better emotionally | 33.6 | 45.2 | <0.001 |
Made it easier to cope with health problems | 32.5 | 37.5 | <0.001 |
Improved overall health and feeling better | 67.3 | 71.1 | 0.001 |
Improved relationship with others | 20.1 | 22.6 | 0.024 |
Improved attendance at job or school ( |
16.2 | 16.8 | 0.562 |
How much the first therapy helped with the most important reasons for CAM use ( |
<0.001 | ||
A great deal | 37.2 | 45.2 | |
Some | 44.0 | 41.8 | |
Only a little | 15.3 | 10.5 | |
Not at all | 3.6 | 2.5 | |
Used first of top three therapies for specific health problems | 43.0 | 42.9 | 0.924 |
How much the first therapy helped with specific health problems ( |
<0.001 | ||
A great deal | 47.5 | 56.0 | |
Some | 36.5 | 31.6 | |
Only a little | 11.7 | 9.1 | |
Not at all | 4.3 | 3.4 |
2Sample adults 18+ who have used first of top three modalities and who currently smoke every day or some days.
3Sample adults 18+ who have used first of top three modalities and who worked or attended school in the past year.
4Sample adults 18+ who have used first of top three modalities and two or more reasons for seeing a practitioner/using modality chosen.
This study has interrogated data from the Adult CAM Supplement of the 2012 NHIS in order to further our understanding of gender differences with respect to CAM use. Specifically, the study has gained new insights into how male and female CAM users differ in their sociodemographic characteristics, reasons, and motivations for CAM use and the health conditions for which CAM is used, shedding new light on the profile of CAM users in the USA.
Our findings indicate that, compared to non-CAM users, users of CAM are more likely to be women, to have higher education and earnings, to be divorced or widowed, and not to reside in the South of the USA. Women, representing the majority (60%) of CAM consumers, were three times more likely to use CAM than men. This supports previous NHIS reports of CAM use in 2002 and 2007 [
In addition to the above-mentioned factors differentiating CAM users from nonusers, several elements clearly distinguished male consumers from female consumers. Firstly, personal income level differs between male and female CAM users, although it is not different between CAM and non-CAM users. Male CAM users, for instance, were more likely to earn a higher income (US$35,000 or more per annum) than female users. This observation is in line with the findings from a recent Norwegian study [
Secondly, marital status appears to have a different impact on male and female CAM users. While previous studies suggest that mutual support of married individuals may promote greater CAM use among married ones than among divorced counterparts [
Thirdly, our analyses revealed that male and female CAM users may differ by race/ethnicity. Our findings showed that all ethnicities except non-Hispanic Asian in men and only non-Hispanic White in women were significantly more likely to use CAM. On a broader level this corresponds to previous NHIS findings that have reported Asian adults as being generally less likely to use CAM compared to White, native, or American-Indian adults [
Of all the reasons, motivations, and outcomes surveyed, a significant differential between women and men was identified. Relative to men, women were motivated to use CAM for a number of reasons (e.g., general wellness or general disease prevention, improving energy, and improving immune function), all of which related to a need to improve one’s health and well-being. To some extent, this finding is not surprising as women are generally more likely than men to utilize preventative health care services [
Male CAM users in our study, on the other hand, were more likely to use CAM to improve athletic or sports performance. This finding is consistent with an earlier survey in New York that more male than female reported the use of CAM to enhance performance [
Another finding from our analyses was that women who used CAM were more likely to report positive outcomes and greater benefit from their CAM use compared to male CAM users. Putting the potential social desirability bias aside, one possible explanation for this finding may be that women are more responsive to the effects of CAM on mind-body well-being, although this speculation has yet to be substantiated by future studies examining both self-reported clinical outcomes and some objective biomarkers. Another possible explanation for this finding could be that the health care needs of women, such as the desire for autonomy in health care decisions [
The findings from our study indicate no substantial differences in the top health conditions that male and female CAM users report. In line with previous NHIS reports [
There are several limitations to our study. First of all, as all of questions in NHIS are self-reported and most of CAM questions were asked regarding the experience in the past 12 months, our study is subject to recall bias and social desirability bias. Secondly, because of the nature of the cross-sectional study design, our findings should be interpreted with caution and we cannot draw conclusions about possible causal pathways between two explored variables in our study. These limitations should be balanced against the strengths of the study, including the large sample size and representativeness of the US population.
This is the first known study in the USA that has sought to understand how male CAM users differ from female CAM users with respect to sociodemographic characteristics and perceived benefits of CAM use. Our paper provides foundation information regarding gender difference of CAM use and provides a platform for further in-depth examination of how and why males and females differ in their reasons for CAM use. Furthermore, our findings demonstrate that it is important that those in clinical practice engage and enquire with their male and female patients regarding possible CAM use in order to help provide safe, effective, coordinated, and equitable health care.
The authors declare that there is no conflict of interests regarding the publication of this paper.