Complementary Medicine Use in Patients with Low Back or Neck Pain: Results of the 2017 National Health Interview Survey

Abstract

Objective

Complementary and integrative medicine (CIM), including mind–body medicine (MBM), is a treatment option that has been proved to alleviate symptoms in patients with low back and neck pain.

Methods

Data from the National Center for Health Statistics’ National Health Interview Survey 2017 were used to examine prevalence and predictors of MBM use and consultation of CIM practitioners among patients with low back and neck pain in the American adult population.

Results

A weighted total of 81,671,436 participants (33.1%) reported low back and neck pain. Of those, more participants used mantra meditation, mindfulness meditation, spiritual meditation, guided imagery, and progressive relaxation than did participants without low back and neck pain (all P < 0.001). Spiritual meditation was the most popular (12.6%), followed by mindfulness meditation (6.9%) and progressive relaxation (5.4%). Participants with low back and neck pain consulted practitioners of CIM more often than did those without low back and neck pain (P < 0.001). MBM use was predicted by being female, having a higher educational level, being employed, and living in the Western United States. Consultation of practitioners of CIM was predicted by younger age, being female, not being married or living with a partner, having a higher educational level, being employed, living in the Western United States, and being uninsured.

Conclusions

MBM is a popular treatment option for people with low back and neck pain. Practitioners of CIM are also being sought out by people with such complaints. Further research in this direction is relevant.

Low Back Pain, Neck Pain, Mind–Body Medicine, National Health Interview Survey

Introduction

In 2017, back pain was the leading cause of nonfatal health loss among all age groups and both sexes. Neck pain ranked ninth [1,2]. Conventional treatment methods do not always alleviate pain symptoms. Also, the use of painkillers can have strong side effects and may be addictive. Therefore, patients seek alternative pain relief or pain relief that is complementary to conventional medical treatments [3,4]. For the treatment of subacute and chronic nonspecific low back pain, complementary and integrative medicine (CIM), including active mind–body medicine (MBM) modalities like yoga or pilates, are more effective in terms of pain reduction and improved functionality than are general medical care and passive therapy measures [5–7]. With regard to neck pain, yoga and tai chi, for example, show promising signs of symptom relief [8,9]. An analysis of the 2012 U.S. National Health Interview Survey (NHIS) on low back pain revealed that MBM is becoming an important component of treatment for people with low back pain [10]. Ghildayal et al. [10] showed that 14.9% of the population with low back pain had used MBM in the prior 12 months and that most people used MBM specifically to treat their back pain. With regard to neck pain, as well, patients seek pain relief apart from conventional medical treatments [11]. A study on patients with chronic pain found that patients who used CIM were younger, more likely to be white, had a higher educational level, and were more dissatisfied with pain care [12].

The present exploratory study uses the NHIS 2017 to estimate the prevalence and predictors of MBM use and CIM consultation in patients with neck or low back pain in the American adult population.

Methods

Data Source

Data from the 2017 NHIS public-use data file were used [13]. The NHIS is a nationally representative survey by the National Center for Health Statistics (NCHS) to monitor the health of the U.S. population and has been the primary source of information about the nation’s health since 1957. In 2017, it covered a total of 32,617 households, with 26,742 adults providing data, resulting in a response rate of 66.5%. Participants did not receive any incentives or compensation for participating in the NHIS.

Outcome Measures

Assessment of MBM Use

The following MBM modalities were assessed by the NHIS 2017 with the question, “During the past 12 months, did you use …?” (yes/no/refused/don’t know): 1) mantra meditation, including Transcendental Meditation^®, relaxation response, and clinically standardized meditation; 2) mindfulness meditation, including Vipassana, Zen Buddhist meditation, mindfulness-based stress reduction, and mindfulness-based cognitive therapy; 3) spiritual meditation, including centering prayer and contemplative meditation; 4) guided imagery; 5) progressive relaxation; 6) yoga; 7) tai chi; and 8) qi gong. Furthermore, general MBM use by patients with low back or neck pain, as operationalized by a dichotomous yes/no indicator of any MBM use within the prior 12 months, was assessed.

Consultation of Practitioners of CIM

Consultation of practitioners of CIM was assessed with the question, “During the past 12 months, did you see …?” (yes/no/refused/don’t know): 1) a chiropractor and/or 2) a practitioner of naturopathy, and/or 3) a practitioner of traditional medicine, such as a shaman, curandero, yerbero, sobador, or Native American healer, and/or 4) a practitioner of homeopathic treatment. Furthermore, general consultation of practitioners of CIM, as operationalized by a dichotomous yes/no indicator of any consultation with any CIM practitioner within the prior 12 months, was evaluated.

Assessment of Low Back and Neck Pain

Participants with low back or neck pain were identified with the following question: “During the past 3 months, did you have neck pain (low back pain)?” (yes/no/refused/don’t know). Participants were asked to report pain that had lasted a whole day or more instead of reporting fleeting or minor aches or pains. Low back and neck pain were combined into one variable, which was coded as 1 for “had back and/or neck pain” or 0 for “had no back and/or neck pain/refused/don’t know.”

Predictors

Predictors that were included on the basis of previous literature about MBM use in patients with pain and on the basis of their availability within the 2017 NHIS were 1) sociodemographic factors and 2) health care access [12,14,15]. Sociodemographic factors included age (18–29 years, 30–39 years, 40–49 years, 50–64 years, and ≥65 years), sex (female vs male), race/ethnicity (non-Hispanic white, Hispanic, African American, Asian, and other), U.S. region (West, Northeast, Midwest, and South), marital status (married/living with a partner vs not married/living with a partner), educational level (less than college vs some college or more), and employment status (employed vs unemployed). Furthermore, insurance status (insured vs uninsured) was included as a potential predictor.

Statistical Analysis

First, sociodemographic differences between participants with and without low back or neck pain were analyzed with chi-squared tests. Second, the 12-month prevalence of MBM use was compared between participants with and without neck or low back pain with chi-squared tests. Third, sociodemographic characteristics between patients with low back or neck pain using or not using MBM and consulting or not consulting practitioners of CIM were compared with chi-squared tests. Fourth, predictors of MBM use and consultation of practitioners of CIM were analyzed with multiple logistic regression analysis. Only those potential predictors associated with the respective dependent variable with P < 0.05 on univariate analysis (chi-squared test) were included in the regression analyses. Population-based estimates were calculated with weights calibrated to the 2010 Census–based population. Relative weights were used for regression analyses and chi-squared tests because absolute weights would inflate the sample size in inferential tests. Adjusted odds ratios with 95% confidence intervals were calculated. All analyses were performed in the Statistical Package for Social Sciences software (IBM SPSS Statistics for Windows, release 27.0; IBM Corporation, Armonk, NY). A P value <0.05 was considered significant.

Results

Sample Characteristics

Table 1 presents the characteristics of participants with and without low back or neck pain. A weighted total of 81,671,436 participants (33.1%) reported having experienced low back or neck pain within the prior 3 months. Differences between participants with and without low back or neck pain were found for sex, relationship status, education, employment, and health insurance status but not for age, race/ethnicity, or region. Separate analyses for neck and low back pain can be found in the Supplementary Data.

Table 1.

Sociodemographic characteristics of the study sample (weighted N = 246,657,271)

Characteristics	Individuals With Low Back or Neck Pain (Weighted n = 81,671,436 [33.1%])	Individuals Without Low Back or Neck Pain (Weighted n = 164,985,835 [66.9%])	P
Age			0.30
18–29 y	13,123,194 (16.1)	27,369,718 (16.6)
30–39 y	12,964,922 (15.9)	25,063,783 (15.2)
40–49 y	12,260,384 (15.0)	23,987,524 (14.5)
50–64 y	21,198,077 (15.0)	44,038,394 (26.7)
65+ y	22,124,859 (27.1)	44,526,416 (27.0)
Sex			<0.001
Male	35,671,938 (43.7)	83,318,150 (50.5)
Female	45,999,498 (56.3)	81,667,685 (49.5)
Race/ethnicity			0.29
Non-Hispanic white	57,115,412 (69.9)	116,762,323 (70.8)
Hispanic	9,903,730 (12.1)	19,569,277 (11.9)
African American	8,916,124 (10.9)	18,110,477 (11.0)
Asian	4,503,089 (5.5)	8,417,057 (5.1)
Other	1,233,081 (1.5)	2,126,701 (1.3)
Relationship status			<0.01
Married/living with a partner	50,295,804 (61.6)	98,183,928 (59.5)
Not married/living with a partner	31,275,744 (38.3)	66,524,870 (40.3)
Education			<0.001
Less than college	31,596,061 (38.7)	56,987,236 (34.5)
Some college or more	49,692,765 (60.8)	107,376,015 (65.1)
Employment			<0.001
Unemployed	36,282,804 (44.4)	55,986,702 (33.9)
Employed	45,376,150 (55.6)	108,934,946 (66)
Region			0.68
West	19,668,777 (24.1)	38,708,695 (23.5)
Northeast	14,739,017 (18.0)	30,430,043 (18.4)
Midwest	17,804,567 (21.8)	35,988,202 (21.8)
South	29,459,075 (36.1)	59,858,895 (36.3)
Health insurance			<0.001
Insured	73,863,072 (90.4)	146,913,884 (89.0)
Uninsured	7,808,364 (9.6)	18,071,951 (11.0)

Values are given as n (%). P values were derived from chi-squared tests using relative weights.

Data source: NCHS NHIS 2017.

Prevalence of MBM Use

Figure 1 shows the 12-month prevalence of MBM use. A weighted total of 21,952,829 participants (26.9%) with low back or neck pain reported having used MBM within the prior 12 months, compared with a weighted total of 36,154,134 participants (21.9%) without low back or neck pain (P < 0.001). Participants with low back or neck pain were significantly more likely (all P < 0.001) to use spiritual meditation (weighted n = 10,288,572 [12.6%]), followed by mindfulness meditation (weighted n = 5,658,980 [6.9%]), progressive relaxation (weighted n = 4,369,655 [5.4%]), mantra meditation (weighted n = 4,185,719 [5.1%]), and guided imagery (weighted n = 3,155,400 [3.9%]) than were participants without low back or neck pain. No significant differences between the groups were found for practicing yoga (P = 0.12), tai chi (P = 0.05), or qi gong (P = 0.37).

Figure 1.

Twelve-month prevalence of MBM use in participants with or without low back or neck pain (LBNP). Data source: NCHS NHIS 2017. *P < 0.001.

Consultations of Practitioners of CIM

Figure 2 shows the 12-month prevalence of consultations with practitioners of CIM. A weighted total of 16,320,531 participants (20.0%) with low back or neck pain reported having consulted a practitioner of CIM within the prior 12 months, compared with a weighted total of 14,199,190 participants (8.6%) without low back or neck pain (P < 0.001). Participants with low back or neck pain were significantly more likely (all P < 0.001) to consult a chiropractor (weighted n = 14,132,961 [17.3%]), followed by a practitioner of naturopathy (weighted n = 1,798,555 [2.2%]), a practitioner of homeopathy (weighted n = 1,630,932 [2.0%]), and a practitioner of traditional medicine (weighted n = 616,352 [0.8%]).

Figure 2.

Twelve-month prevalence of consultations with practitioners of CIM in participants with or without low back or neck pain (LBNP). Data source: NCHS NHIS 2017.

Differences and Predictors of MBM Use in Participants with Low Back or Neck Pain

MBM users and MBM non-users with low back or neck pain differed with regard to sex, education, employment status, region, and health insurance status (Table 2). Regression analyses showed that MBM use was predicted by being female, having a higher educational level, being employed, and living in the Western United States (Table 3).

Table 2.

Comparison of characteristics among patients with low back or neck pain using or not using MBM and consulting or not consulting a practitioner of CIM

Characteristics	Using MBM (Weighted n = 21,952,829 [26.9%])	Not Using MBM (Weighted n = 59,718,607 [73.1%])	P	Consulting a Practitioner of CIM (Weighted n = 16,320,531 [20.0%])	Not Consulting a Practitioner of CIM (Weighted n = 65,350,905 [80.0%])	P
Age			0.37			0.01
18–29 y	3,730,160 (17.0)	9,393,034 (15.7)		2,953,839 (18.1)	10,169,355 (15.6)
30–39 y	3,364,138 (15.3)	9,600,784 (16.1)		2,802,339 (17.2)	10,162,583 (15.6)
40–49 y	3,459,049 (15.8)	8,801,335 (14.7)		2,430,520 (14.9)	9,829,864 (15.0)
50–64 y	5,568,294 (25.4)	15,629,783 (26.2)		3,847,652 (23.6)	17,350,425 (26.5)
65+ y	5,831,188 (26.6)	16,293,671 (27.3)		4,286,181 (26.3)	17,838,678 (27.3)
Sex			<0.001			0.01
Male	7,239,642 (33.0)	28,432,296 (47.6)		6,692,297 (41.0)	28,979,641 (44.3)
Female	14,713,187 (67.0)	31,286,311 (52.4)		9,628,234 (59.0)	36,371,264 (55.7)
Race/ethnicity			0.89			0.28
Non-Hispanic white	15,428,779 (70.3)	41,686,633 (69.8)		11,376,519 (69.7)	45,738,893 (70.0)
Hispanic	2,711,303 (12.4)	7,192,427 (12.0)		2,000,012 (12.3)	7,903,718 (12.1)
African American	2,368,975 (10.8)	6,547,149 (11.0)		1,696,717 (10.4)	7,219,407 (11.0)
Asian	1,129,898 (5.1)	3,373,191 (5.6)		1,047,313 (6.4)	3,455,776 (5.3)
Other	313,874 (1.4)	919,207 (1.5)		199,970 (1.2)	1,033,111 (1.6)
Relationship status			0.12			<0.001
Married/living with a partner	13,232,866 (60.3)	37,062,938 (62.1)		10,643,009 (65.2)	39,652,795 (60.7)
Not married/living with a partner	8,703,464 (39.6)	22,572,280 (37.9)		5,639,530 (34.6)	25,636,214 (39.2)
Education			<0.001			<0.001
Less than college	5,101,335 (23.2)	26,494,726 (44.4)		4,571,517 (28.0)	27,024,544 (41.4)
Some college or more	16,788,003 (76.5)	32,904,762 (55.1)		11,718,137 (71.8)	37,974,628 (58.1)
Employment			<0.001			<0.001
Unemployed	8,559,371 (39.0)	27,723,433 (46.4)		5,780,249 (35.4)	30,502,555 (46.7)
Employed	13,393,458 (61.0)	31,982,692 (53.6)		10,540,282 (64.6)	34,835,868 (53.3)
Region			<0.001			<0.001
West	6,217,500 (28.3)	13,451,277 (22.5)		4,612,915 (28.3)	15,055,862 (23.0)
Northeast	4,156,316 (18.9)	10,582,701 (17.7)		2,855,868 (17.5)	11,883,149 (18.2)
Midwest	4,901,786 (22.3)	12,902,781 (21.6)		4,090,171 (25.1)	13,714,396 (21.0)
South	6,677,227 (30.4)	22,781,848 (38.1)		4,761,577 (29.2)	24,697,498 (37.8)
Health insurance			<0.01			<0.001
Insured	20,160,983 (91.8)	53,702,089 (89.9)		15,128,635 (92.7)	58,734,437 (89.9)
Uninsured	1,791,846 (8.2)	6,016,518 (10.1)		1,191,896 (7.3)	6,616,468 (10.1)

Values are given as n (%).

Data source: NCHS NHIS 2017.

Table 3

Predictors of MBM use in individuals with low back or neck pain, 2017 NHIS

Predictors	Adjusted OR (95% CI)	P
Sex
Male	1 (Reference)
Female	1.93 (1.75–2.14)	<0.001
Education
Less than college	1 (Reference)
Some college or more	2.50 (2.24–2.79)	<0.001
Employment
Unemployed	1 (Reference)
Employed	1.25 (1.13–1.38)	<0.001
Region		<0.001
West	1 (Reference)
Northeast	0.83 (0.72–0.96)	0.01
Midwest	0.82 (0.72–0.95)	0.01
South	0.65 (0.57–0.74)	<0.001
Health insurance
Insured	1 (Reference)
Uninsured	1.03 (0.86–1.22)	0.75

Note. n = 9,457 (unweighted); Method = Enter; OR = odds ratio; CI = confidence interval; R² Cox and Snell = 0.062; R² Nagelkerke = 0.090; Modell χ²(7) = 562.146, P < 0.001.

Data source: NCHS NHIS 2017.

Differences and Predictors of Consultations of Practitioners of CIM in Participants with Low Back or Neck Pain

Participants with low back or neck pain who consulted a practitioner of CIM differed in age, sex, relationship status, education, employment, region, and health insurance status from those who did not consult a practitioner of CIM (Table 2). Regression analyses showed that consultations of practitioners of CIM were predicted by younger age, being female, not being married or living with a partner, having a higher educational level, being employed, living in the Western United States (Colorado, Montana, Utah, Wyoming, California, Hawaii, Nevada, Alaska, Idaho, Oregon, Washington), and being uninsured (Table 4).

Table 4.

Predictors of consultations of practitioners of CIM in individuals with low back or neck pain, 2017 NHIS

Predictors	Adjusted OR (95% CI)	P
Age		<0.01
18–29 y	1 (Reference)
30–39 y	0.97 (0.81–1.16)	0.71
40–49 y	0.85 (0.70–1.02)	0.09
50–64 y	0.77 (0.65–0.90)	<0.01
65+ y	0.83 (0.70–0.97)	0.02
Sex
Male	1 (Reference)
Female	1.21 (1.08–1.35)	0.001
Relationship status
Married/living with a partner	1 (Reference)
Not married/living with a partner	1.14 (1.02–1-27)	0.02
Education
Less than college	1 (Reference)
Some college or more	1.64 (1.45–1.84)	<0.001
Employment
Unemployed	1 (Reference)
Employed	1.47 (1.32–1.65)	<0.001
Region		<0.001
West	1 (Reference)
Northeast	0.78 (0.66–0.91)	<0.01
Midwest	1.00 (0.86–1.16)	0.97
South	0.67 (0.58–0.77)	<0.001
Health insurance
Insured	1 (Reference)
Uninsured	1.31 (1.07–1.61)	<0.01

n = 9,457 (unweighted); Method = Enter; OR = odds ratio; CI = confidence interval; R² Cox and Snell = 0.027; R² Nagelkerke = 0.043; Modell χ²(12) = 50.773, P < 0.001.

Data source: NCHS NHIS 2017.

Discussion

About one third of the adult U.S. population suffers from low back or neck pain. Analyses of the NHIS 2017 data show that more than a quarter of these pain patients had used MBM in the prior 12 months, which is significantly more than among patients without low back or neck pain. Our findings show that, in particular, employed females with higher educational levels who live in the Western United States are more likely to use MBM for treating low back and/or neck pain. This is consistent with previous research showing that MBM use is generally associated with these factors. Educated women are more likely than other groups of people to use MBM practices such as tai chi, qi gong, and yoga [16,17]. The most used MBM therapies were yoga and spiritual meditation, followed by mindfulness meditation and progressive relaxation. Yoga, however, was practiced just as frequently by patients without low back or neck pain.

The fact that educated women in particular request such procedures is not surprising. Although yoga used to be practiced primarily by men [18], the proportion of men practicing yoga is vanishingly small in modern times. Interestingly, not only are female patients more likely to use CIM, it is also the case that female physicians are more likely than their male counterparts to prescribe CIM [19]. As a great number of men suffer from low back or neck pain, an increased use of yoga among this population would be helpful.

Interestingly, significant group differences between participants with and without low back or neck pain were found only with regard to the use of passive MBM modalities, like meditation, but not with regard to the active modalities, like yoga, tai chi, and qi gong. This is interesting, as medical guidelines like the American College of Physicians Clinical Practice Guideline especially recommend active modalities like yoga for treating low back pain [20,21]. Since 2017, tai chi and mindfulness-based stress reduction also have been included in the guideline. Even though yoga was the most used MBM modality, participants used yoga irrespective of whether they had low back or neck pain. Patients with low back or neck pain should be made even more aware of the benefits of active MBM procedures for their pain.

Patients with low back or neck pain consult a practitioner of CIM more often than do patients without low back or neck pain. In particular, young, employed females with higher educational levels who live in the Western United States, are not married or living with a partner, and do not have health insurance turn to a practitioner of CIM for treating their low back or neck pain. The fact that CIM is used especially in the West might be related to the fact that the density of CIM practitioners is higher there than in most other areas of the United States [27]. Also, even though the rate of CIM use is lower in the geographic South, the absolute number of Americans using CIM is highest in the South, which probably has to do with the fact that more people live in the South.

Compared with previous findings, the number of participants using MBM to treat their low back or neck pain is steadily increasing. In 2012, 14.9% of participants with low back pain stated that they used MBM. In the evaluation of the 2017 data set, the figure is already 26.9%. However, these figures are not fully comparable, as the 2012 evaluation of Ghildayal et al. [10] evaluated only low back pain and did not include patients with neck pain. In our studied population, only 20% of the population with low back or neck pain reported visiting a practitioner of CIM. This contrasts with findings from Goode et al. [11], who conducted a cross-sectional telephone survey of a representative sample of households in North Carolina in 2006, in which 41% percent reported visiting one or more complementary care providers in the prior 12 months, including chiropractors, acupuncturists, and massage therapists. Here again, only people with chronic neck pain and no chronic lower back pain were interviewed. Accordingly, these two findings are only conditionally comparable. Although the numbers in the present study are lower than those in previous studies, they are nevertheless high.

Americans seek help for their pain symptoms from MBM and practitioners of CIM. This use of CIM is also a major economic factor: Americans spent $30.2 billion out-of-pocket on complementary health approaches like MBM in 2012 [22]. Interestingly, CIM users with chronic back pain have health care expenditures that are $734 lower than those of nonusers [23].

In line with the literature, we found that, in particular, females with higher educational levels and living in the Western United States used MBM and consulted practitioners of CIM [24]. Consideration should be given to how other population groups could be specifically addressed. Against the background of the opioid crisis prevailing in the United States [25,26], it is important to draw the attention of as broad a section of the population as possible to alternatives to pharmacological pain therapy.

There are several study limitations. The information collected under the NHIS is based solely on participants’ self-reported data, so a recall bias cannot be ruled out. Furthermore, the back or neck pain recorded is not a medical diagnosis, but rather self-reported pain. It could indicate a person suffering from clinically diagnosed chronic pain or a person who had only one day of pain in the prior 3 months. It would be valuable to look at data on clinically diagnosed pain patients. Because of the limited information within the NHIS data set, it would not be possible to analyze the duration of pain and the details of MBM use. Unfortunately, no other CIM procedures were surveyed beyond those described here, so we cannot make any statements about other procedures, such as acupuncture and massage. Furthermore, because of the cross-sectional design of the NHIS, it is not possible to draw causal conclusions. Also, the 2017 NHIS did not capture CIM use for the treatment of low back or neck pain. Therefore, although there is a correlation between these complaints and CIM use, it cannot be concluded with certainty that individuals used CIM to treat low back or neck pain. Because of the different time windows in the questions (low back or neck pain was surveyed for a time window of 3 months, and MBM use was surveyed for a time window of 12 months), it is also possible that the use of MBM took place before the appearance of low back or neck pain.

Conclusion

A large proportion of the American population suffers from low back or neck pain. U.S. adults with low back or neck pain were significantly more likely to use MBM and more likely to consult a CIM practitioner than were adults with no low back or neck pain. Hence, the use of MBM is an important and growing part of health care for people with low back or neck pain and should be investigated further.

Supplementary Data

Supplementary Data may be found online at http://painmedicine.oxfordjournals.org.

Funding sources: No specific funding was received.

Conflicts of interest: The authors have nothing to disclose.

Disclaimer: Analyses, interpretations, and conclusions are the responsibility of the author (recipient of the data file) and not of the National Center for Health Statistics, which is responsible only for the initial data.