Back Pain Prevalence and Visit Rates: Estimates From U.S. National Surveys, 2002

Article (PDF Available)inSpine 31(23):2724-7 · December 2006with937 Reads
DOI: 10.1097/01.brs.0000244618.06877.cd · Source: PubMed
Abstract
Review and analysis of data from two U.S. national surveys in 2002. To examine the prevalence of back pain and physician visits for back pain in the United States. National data on the prevalence of back pain become available only intermittently. We summarized published data from the 2002 National Health Interview Survey (NHIS) on the prevalence of back pain and compared it with earlier surveys. We also analyzed the 2002 National Ambulatory Medical Care Survey (NAMCS) to determine physician visit rates for back pain. In the 2002 NHIS, there were 31,044 adult respondents. Low back pain lasting at least a whole day in the past 3 months was reported by 26.4% of respondents, and neck pain was reported by 13.8%. Among racial groups, American Indians and Alaska Natives had the highest prevalence of low back pain, and Asian Americans had the lowest. Prevalence generally declined with greater levels of education and increasing income. Prevalence estimates were consistent with those from previous surveys, although methodologic differences limited comparisons. NAMCS data suggested that the proportion of all physician visits attributable to low back pain (2.3% in 2002) has changed little since the early 1990s. About one fourth of U.S. adults report low back pain in the past 3 months; the proportion of physician visits attributed to back pain has changed little in the past decade.
SPINE Volume 31, Number 23, pp 2724–2727
©2006, Lippincott Williams & Wilkins, Inc.
Back Pain Prevalence and Visit Rates
Estimates From U.S. National Surveys, 2002
Richard A. Deyo, MD, MPH,*†‡§ Sohail K. Mirza, MD, MPH,†§ and Brook I. Martin, MPH*§
Study Design. Review and analysis of data from two
U.S. national surveys in 2002.
Objectives. To examine the prevalence of back pain
and physician visits for back pain in the United States.
Summary of Background Data. National data on the prev-
alence of back pain become available only intermittently.
Methods. We summarized published data from the
2002 National Health Interview Survey (NHIS) on the
prevalence of back pain and compared it with earlier sur-
veys. We also analyzed the 2002 National Ambulatory
Medical Care Survey (NAMCS) to determine physician
visit rates for back pain.
Results. In the 2002 NHIS, there were 31,044 adult
respondents. Low back pain lasting at least a whole day in
the past 3 months was reported by 26.4% of respondents,
and neck pain was reported by 13.8%. Among racial
groups, American Indians and Alaska Natives had the
highest prevalence of low back pain, and Asian Ameri-
cans had the lowest. Prevalence generally declined with
greater levels of education and increasing income. Prev-
alence estimates were consistent with those from previ-
ous surveys, although methodologic differences limited
comparisons. NAMCS data suggested that the proportion
of all physician visits attributable to low back pain (2.3%
in 2002) has changed little since the early 1990s.
Conclusions. About one fourth of U.S. adults report
low back pain in the past 3 months; the proportion of
physician visits attributed to back pain has changed little
in the past decade.
Key words: prevalence, back pain, neck pain, trends,
physician visits. Spine 2006;31:2724 –2727
Back pain is extremely common, but there are few recent
national data for the United States. Rising rates of imag-
ing, spine surgery, spinal injections, and use of comple-
mentary and alternative providers for back pain have
created an impression of increasing prevalence of back
pain.
1–4
The best population-based data on symptoms
come from well-designed national surveys, but such data
become only intermittently available. The recent release
of data from the 2002 National Health Interview Survey
(NHIS) and National Ambulatory Medical Care Survey
(NAMCS) provided an opportunity to examine recent
estimates of back pain prevalence in the United States,
their variation with demographic features, and back pain
visit rates to physician offices.
Methods
Most of the data presented here are derived from two national
surveys conducted by the National Center for Health Statistics,
a branch of the Centers for Disease Control and Prevention.
The first is the NHIS (2002). The second is the NAMCS (2002).
These surveys provide national estimates for a broad range of
health measures among the civilian, noninstitutionalized adult
population. We used them to examine the prevalence of back
pain, its demographic correlates, and the proportion of physi-
cian visits attributable to low back pain.
A recent summary of data from the NHIS provided an oppor-
tunity to examine the prevalence of back pain. The data in Tables
1 and 2 are summarized from the report.
5
We have not obtained
the data to perform additional analyses. The 2002 NHIS survey
sampled 36,161 households, from which there was an 88.1%
response rate. Among these households were 36,787 adults who
were eligible for the adult questionnaire, and data were actually
collected from 31,044 adults. The final response rate of 74.4% of
adults reflected a combination of nonresponse rates from family
units and the nonresponse rates from adults within responding
units. The adults who were sampled were assigned weights reflect-
ing the probability of being sampled, which allowed estimates of
national percentages. The data presented here were age-adjusted
to the 2000 U.S. population by the National Center for Health
Statistics. The age groups used for age adjustment were 18 to 44
years, 45 to 64 years, 65 to 74 years, and 75 years or greater. Age
adjustment was used to permit comparison among demographic
subgroups that may have different age structures.
5
We compared these results with previous surveys of back
pain prevalence, including a 1985 national survey conducted
by Louis Harris & Associates.
6,7
Adults over age 18 were sam-
pled in every state of the United States in proportion to the state
population. Telephone numbers were randomly selected from
the Harris library of telephone directories, and the final two
digits were altered with random numbers. Weighting was ap-
plied to education, age, sex, and race to bring them into line
with actual proportions of the population. We also reviewed
data from a citywide population survey conducted in Dayton,
OH in 1973.
8
In this survey, families were chosen by a propor-
tionate random sampling scheme that blocked on demographic
characteristics to produce a sample with a distribution similar
to that found in then-current census data. Finally, we used data
from a previous U.S. government survey, the second National
Health and Nutrition Examination Survey (NHANES II).
9
This
involved 10,404 adults age 25 or older who completed both an
interview and a physical examination. Again, weights were in-
corporated to adjust for probability of selection, nonresponse,
age, sex, and race to make the sample nationally representative.
Unfortunately, the definitions of back pain used in these sur-
veys varied somewhat. In the 2002 NHIS, respondents were asked
From the Departments of *Medicine, †Orthopaedic Surgery, ‡Health
Services, and the §Center for Cost and Outcomes Research, University
of Washington, Seattle, WA.
Acknowledgment date: August 15, 2005. First revision date: Novem-
ber 8, 2005. Acceptance date: January 24, 2006.
Supported by Grant No. P60 AR 48093 from the National Institute for
Arthritis, Musculoskeletal and Skin Diseases.
The manuscript submitted does not contain information about medical
device(s)/drug(s).
Federal funds were received in support of this work. No benefits in any
form have been or will be received from a commercial party related
directly or indirectly to the subject of this manuscript.
Address correspondence and reprint requests to Richard A. Deyo, MD,
MPH, Center for Cost and Outcomes Research, Box 359736, Harbor-
view Medical Center, 325 Ninth Avenue, Seattle, WA 98104; E-mail:
deyo@u.washington.edu
2724
“During the past 3 months, did you have low back pain?” Re-
spondents were instructed to report pain that had lasted a whole
day or more and conversely, not to report fleeting or minor aches
or pains. In contrast, the Louis Harris survey asked about any
back pain in the past year, and the Dayton survey asked about
“frequent low back pain in the past year.” We assumed that back
pain lasting for a day and occurring in the past 3 months would be
less common than any back pain occurring in the past year, but
more common than “frequent back pain.” We therefore expected
the 2002 NHIS results to fall between the estimates of 1 year
prevalence and the prevalence of “frequent” back pain in the past
year. The NHANES survey asked “Have you ever had pain in
your back on most days for at least 2 weeks?” We assumed this
figure would be lower than the 2002 NHIS figure, which asked
only about back pain for a day, even though the NHANES ques-
tion applied to lifetime prevalence.
The NAMCS is a study of visits to medical office practices.
Each year, 3,000 physicians are randomly selected to provide
data on approximately 30 patient visits over a 1-week period.
The sample includes office-based physicians who are not feder-
ally employed, identified by the American Medical Association
and the American Osteopathic Association. Physicians in anes-
thesiology, pathology, and radiology are excluded, and only
office visits are captured (not hospital, nursing home, or house
calls). All 50 states and the District of Columbia are included.
For this report, we obtained NAMCS data for the years
1996 to 2002 to assess trends in office visits for back pain. For
this purpose, we used International Classification of Diseases
Version 9, Clinical Modification (ICD-9-CM) diagnosis codes
that identified the problem as being in the lumbar spine. In a
small number of cases, the location in the spine was not clear
from ICD codes, but we included patients whose chief com-
plaint indicated a low back problem. The codes were based
on a previously described algorithm,
10
which was recently up
-
dated.
11
We excluded patients with pregnancy, malignancies,
injury resulting from motor vehicle accidents, or fractures.
Results
In the 2002 NHIS, back pain was the most frequent type
of pain reported by respondents (Table 1). There were
26.4% of respondents who reported low back pain
within the past 3 months, while smaller percentages re-
ported severe headache, neck pain, or face or jaw pain.
This prevalence of back pain corresponds to an estimate
of over 54 million American adults who have experi-
enced low back pain in the past 3 months.
Back pain was more common among adults over the age
of 45 than among younger adults (Table 2). However, the
prevalence of back pain fell slightly among the oldest
adults, as noted in previous surveys.
9
Women were some
-
what more likely than men to report back pain. Among
racial groups, American Indians and Alaska Natives had
the highest prevalence of back pain, while Asian Americans
had the lowest prevalence. White, black, and Hispanic re-
spondents fell between these extremes.
The prevalence of back pain generally fell with greater
levels of education. Thus, respondents with less than a
high school diploma had a prevalence of almost 32%
versus approximately 22% among those with a bache-
lor’s degree or higher. Similarly, the prevalence of back
pain fell consistently with increasing levels of family in-
come. Regarding marital status, the prevalence of back
pain was highest among widowed adults and lowest
among those who had never married.
These prevalence estimates from 2002 seemed consis-
tent with those from earlier surveys (Table 3).
12
As ex
-
pected, the 2002 prevalence estimates were substantially
lower than those reported by the Harris survey for “any
back pain in the past year.” On the other hand, they were
somewhat higher than estimates of “frequent low back
pain” in the past year, or the prevalence of back pain
Table 1. Prevalence of Various Pain Syndromes in the
Past 3 Months Reported by U.S. Residents, 2002
Age-Adjusted % (SE)
Low back pain 26.4 (0.32)
Migraine or severe headache 15.0 (0.25)
Neck pain 13.8 (0.25)
Face or jaw pain 4.6 (0.15)
Data are from National Health Interview Survey.
5
For each type of pain, respon
-
dents were instructed to report pain that occurred during the past 3 months that
had lasted a whole day or more. They were instructed not to report fleeting or
minor aches and pains. Individuals may be represented in more than one row.
Table 2. Prevalence of Back Pain by Respondent
Characteristics, 2002
Selected Characteristic
Age-Adjusted % With
Back Pain (SE)
Respondent age
18–44 yr 23.7 (0.43)
45–64 yr 29.8 (0.58)
65–74 yr 28.8 (0.97)
75 yr or more 28.7 (1.03)
Sex
Male 24.3 (0.45)
Female 28.3 (0.43)
Race/ethnicity
White (non-Hispanic) 27.4 (0.39)
Black 23.9 (0.85)
Hispanic 24.3 (0.81)
American Indian/Alaska Native 35.0 (4.46)
Asian 19.0 (1.59)
Native Hawaiian/Pacific Islander 26.8 (10.76)*
2 or more races
Black and white 20.0 (5.25)
American Indian/Alaska Native and white 45.5 (4.24)
Education
High school diploma 31.8 (0.81)
High school diploma or GED 28.6 (0.59)
Some college 28.9 (0.64)
Bachelor’s degree or higher 22.2 (0.59)
Family income
$20,000 31.8 (0.79)
$20,000–$34,999 29.9 (3.21)
$35,000–$54,999 26.4 (0.83)
$55,000–$74,999 25.0 (0.77)
$75,000 or more 23.3 (1.81)
Marital status
Married 26.0 (0.41)
Widowed 35.8 (3.21)
Divorced or separated 31.0 (0.83)
Never married 22.7 (0.77)
Living with a partner 34.3 (1.81)
Data are from National Health Interview Survey.
5
*Large standard error, reflecting small sample. Caution suggested by the
surveyors as data do not meet intended standard of precision.
2725Back Pain Prevalence
Deyo et al
lasting for at least 2 weeks. The 4 tabulated surveys to-
gether span some 30 years. However, because of differ-
ences in the way questions were asked and differences in
sampling methods, we cannot determine if there is a
trend in prevalence over time.
As estimated from the NAMCS, office visits for low back
pain remained a fairly steady fraction of all office visits
between 1996 and 2002 (Figure 1). In 2002, lower back
pain accounted for approximately 2.3% of all office visits,
the same fraction observed in 1996. Although the estimated
total number of office visits for all reasons increased from
about 734 million in 1996 to 890 million in 2002, the
increase seemed to be true for a wide range of conditions.
The estimate of 2.3% of visits is similar to the 2.8%
reported from 1989–1990 NAMCS data.
13
The higher
percentages reported in 1989 –1990 may have been be-
cause cases were included as lumbar in that report even if
location in the spine was ambiguous from ICD-9 codes.
Discussion
The NHIS provides updates of back pain prevalence in
the United States. The survey questions are not compa-
rable with those used in previous national surveys, so we
cannot determine if there is a trend in prevalence, al-
though the recent values seem consistent with earlier sur-
veys. We also cannot determine any possible changes in
pain severity, diagnoses, or work absenteeism. Another
review of trend data, including international estimates,
suggested that the prevalence of low back pain has not
changed significantly over the past 30 years.
14
Other estimates include a value of 41% of American
adults between the ages of 26 and 44 years having had
back pain in the past 6 months.
14
Several European stud
-
ies have examined the prevalence of any back pain in the
last month, producing estimates that center around
40%.
14
The estimate derived from the 2002 NHIS for
3-month prevalence is somewhat lower, but this may be
related to the instruction to ignore trivial back pain in
this Health Interview Survey.
These data confirm earlier findings suggesting that the
prevalence of back pain is highest among patients with low
levels of education and income.
8,9
It remains unclear
whether this is a consequence of more ergonomically de-
manding jobs, greater life stress, or other factors. The data
also identify substantial racial differences in the prevalence
of back pain, which may warrant further investigation. In
comparison with other body sites, back pain remains the
most common pain problem reported by U.S. adults.
The proportion of all physician visits related to low back
pain has remained relatively stable. In contrast, there is
evidence of increasing use of back surgery,
1
complementary
and alternative medicine,
2
and injection procedures.
3
One
hypothesis would be that the prevalence of back pain is
relatively stable, but therapy is becoming more intensive,
with greater “per patient” use of these treatments.
Table 3. Estimates of U.S. Prevalence of Back Pain in Various Surveys
Any Back Pain in Past Year
(%) (Louis Harris Survey
Group, 1985; n 1,254)
At Least 1 Day of Back Pain
in Past 3 Months (%)
(NHIS, 2002; n 31,044)
“Frequent” LBP in Past 12
Months (%) (Dayton, Ohio,
1973; n 2,782)
Lifetime Prevalence of LBP
Lasting at Least 2 Weeks
(NHANES II, 1976–1980)
All adults 56 26.4 18 13.8
Male 53 24.3 15 14.2
Female 57 28.3 20 13.4
Black 46 23.9 19 11.4
White 59 27.4 19 14.2
Over the age of 65 49 28.8 18 16
Figure 1. Percent of office visits
for low back pain as primary di-
agnosis among adults 18 years or
older (NAMCS, 1996–2002).
2726 Spine
Volume 31
Number 23
2006
Based on this survey and earlier surveys, we can roughly
generalize that about half of adults have low back pain
during any given year and about two thirds have low back
pain at some time in their lives.
8
About one fourth have had
at least a day of back pain during the past 3 months, and
about 15% of adults report “frequent” back pain or pain
lasting greater than 2 weeks annually.
9
Key Points
Low back pain remains the most common type
of pain reported by U.S. adults; 26.4% report back
pain lasting at least a day in the past 3 months.
Back pain prevalence falls with increasing levels
of education and income.
The proportion of all physician visits attributed
to back pain has remained fairly steady since 1990.
The prevalence of back pain estimated from
2002 national data is consistent with estimates
from earlier decades, but differences in question-
naire wording make it impossible to determine if
there are significant trends.
References
1. Gray DT, Deyo RA, Kreuter W, et al. Population-based trends in volumes
and rates of ambulatory lumbar spine surgery. Spine 2006;31:1957– 63.
2. Kessler RC, Davis RB, Foster DF, et al. Long-term trends in the use of
complementary and alternative medical therapies in the United States. Ann
Intern Med 2001;135:262– 8.
3. Carrino JA, Morrison WB, Parker L, et al. Spinal injection procedures: vol-
ume, provider distribution, and reimbursement in the U.S. Medicare popu-
lation from 1993 to 1999. Radiology 2002;225:723–9.
4. Abelson R. An MRI machine for every doctor? Someone has to pay. New
York Times, March 13, 2004, A1.
5. U.S. Department of Health and Human Services. Summary Health Statistics
for U.S. Adults: National Health Interview Survey, 2002. Vital and Health
Statistics Series 10, No. 222 [DHHS Publication No. (PHS) 2004-1550].
Hyattsville, MD, 2004. Available at www.cdc.gov/nchs/data/series/sr_10/
sr10_222.pdf, accessed August 20, 2004.
6. Taylor H, Curran NM. The Nuprin Report. New York: Louis Harris, 1985.
7. Sternbach RA. Survey of pain in the United States: the Nuprin Pain Report.
Clin J Pain 1986;2:49–53.
8. Reisbord LS, Greenland S. Factors associated with self-reported back-
pain prevalence: a population-based study. J Chron Dis 1985;38:691–
702.
9. Deyo RA, Tsui-Wu YJ. Descriptive epidemiology of low-back pain and its
related medical care in the United States. Spine 1987;12:264 8.
10. Cherkin DC, Deyo RA, Volinn E, et al. Use of the International Classification
of Diseases (ICD-9-CM) to identify hospitalizations for mechanical low back
problems in administrative databases. Spine 1992;17:817–25.
11. Deyo RA, Gray DT, Kreuter W, et al. United States trends in lumbar fusion
surgery for degenerative conditions. Spine 2005;30:1441–5.
12. Lawrence RC, Helmick CG, Arnett SC, et al. Estimates of the prevalence of
arthritis and selected musculoskeletal disorders in the United States. Arthritis
Rheum 1998;41:778 –99.
13. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low back pain:
frequency, clinical evaluation, and treatment patterns from a U.S. national
survey. Spine 1995;20:11–9.
14. Nachemson A. Epidemiology and the economics of low back pain. In:
Herkowitz HN, et al, eds. The Lumbar Spine, 3rd ed. Philadelphia: Lippin-
cott, Williams & Wilkins, 2004:3–10.
2727Back Pain Prevalence
Deyo et al
    • "Prevalence rates were reported highest among Native Americans and lowest among Asian Americans. Increasing income and higher education levels had moderating effects on reported back pain [6]. Throughout the adult life cycle, prevalence rates increase until the 60–65 year age group and then again decline, with the peak incidence of LBP in the third decade of life [7]. "
    [Show abstract] [Hide abstract] ABSTRACT: Low back pain (LBP) is a heterogeneous condition with high prevalence, high morbidity, and large economic burden. According to the 2010 Global Burden of Disease Study, low back pain ranked highest among 291 studied disorders in terms of years lived with disability (YLDs), with a global point prevalence estimated to be 9.4 % (95 % CI 9.0–9.8). In 2005, direct expenditures for spine problems in the United States were estimated at $85.9 billion, which represented 9 % of the total national healthcare expenditures, similar to costs associated with arthritis, cancer, and diabetes, and only exceeded significantly by those for heart disease and stroke. In the United States between 2004 and 2008, it is estimated that over two million episodes of back pain resulting in presentation for emergency care occurred, yielding an incidence rate of 1.39/1,000 person-years. In workers 40–65 years of age, back pain costs employers an estimated $7.4 billion/year in lost productive time. Commonly, disorders of the neck and back are self-limiting conditions, which require only judicial use of imaging and rarely more invasive treatments. Many national and international groups have produced high-quality, evidence-based recommendations to aid in the diagnosis and treatment of low back pain (Table 1).
    Chapter · Jan 2017 · PLoS ONE
    • "Low back pain (LBP) and neck pain (NP) are frequently reported in the general populations [1, 2]. In spite of the self-limiting nature of most spinal pain conditions and small effects of commonly applied treatment interven- tions [3][4][5][6], many spinal pain sufferers will consult professional help for their problems with frequent visits to primary care clinics [7, 8]. In a recent article exploring the natural course of acute neck and low back pain in the general population we found a rapid decrease of pain within the first month [6]. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Low back and neck pain are commonly reported in the general population and represent frequent causes for health care consultations. The main aim of this study was to describe the determinants of health care contact during a 1-year period in a general population with recent onset spinal pain. Methods From 9056 participants in a general health survey in Norway we identified 219 persons reporting a recent onset (<1 month) of low back or neck pain. Questionnaires were given at 1 (baseline), 2, 3, 6 and 12 months after pain debut. The main outcome was self-reported health care contact due to spinal pain. Associations between health care contact and pain-related factors, other somatic and mental health factors, pain-related work limitations, physical activity and sociodemographic factors were explored. Results Conventional health care was sought by 93 persons (43 %) at least once throughout the year following the onset of pain. 18 persons (8 %) sought alternative health care only and 108 persons (49 %) sought no kind of health care. Baseline reports of coexisting low back and neck pain of equal intensity, poor self-reported health, symptoms of anxiety or depression, obesity and smoking were all associated with an increased tendency to seek conventional health care. Pain intensity and pain-related work limitations at each occasion were strongly associated with concurrent health care contact throughout the year. Higher education was associated with a reduced tendency to contact health care and no association was found for physical activity. Conclusion The main finding in this study was that people from the general population who seek health-care for a new incident of neck or low back pain report more symptoms of mental distress, poorer self-reported health and more intense pain with stronger work limitations compared to those who do not. The findings suggest that identification of complementary symptoms is highly relevant in the examination of spinal pain patients, even for those with recent onset of symptoms.
    Full-text · Article · Dec 2016
    • "Low back pain (LBP) is the most common occupational disorder in North America [1, 2], a major cause of work absenteeism [3, 4] and a leading cause of disability worldwide [5]. The 2010 Global Burden of Disease Study revealed that the LBP disability-adjusted life years increased from 58.2 million in 1990 to 83.0 million in 2010[6], although the majority of LBP patients experience non-specific symptoms that cannot be attributed to a serious disease [7]. "
    [Show abstract] [Hide abstract] ABSTRACT: Purpose To estimate the clinical effectiveness and to systematically review the literature of full economic evaluation of chiropractic care compared to other commonly used care approaches among adult patients with non-specific LBP. Study Design Systematic reviews of interventions and economic evaluations. Methods A comprehensive search strategy was conducted to identify 1) pragmatic randomized controlled trials (RCTs) and/or 2) full economic evaluations of chiropractic care for low back pain compared to standard care delivered by other healthcare providers. Studies published between 1990 and 4th June 2015 were considered. Primary outcomes included pain, functional status and global improvement. Study selection, critical quality appraisal and data extraction were conducted by two independent reviewers. Data from RCTs with low risk of bias were included in a meta-analysis to determine effect estimates. Cost estimates of full economic evaluations were converted to 2015 USD and results summarized using Slavin’s qualitative best-evidence synthesis. Results Six RCTs and three full economic evaluations were scientifically admissible. Five RCTs with low risk of bias compared chiropractic care to exercise therapy (n = 1), physical therapy (n = 3) and medical care (n = 1). Overall, we found similar effects for chiropractic care and the other types of care and no reports of serious adverse events. Three low to high quality full economic evaluations studies (one cost-effectiveness, one cost-minimization and one cost-benefit) compared chiropractic to medical care. Given the divergent conclusions (favours chiropractic, favours medical care, equivalent options), mixed-evidence was found for economic evaluations of chiropractic care compared to medical care. Conclusion Moderate evidence suggests that chiropractic care for LBP appears to be equally effective as physical therapy. Limited evidence suggests the same conclusion when chiropractic care is compared to exercise therapy and medical care although no firm conclusion can be reached at this time. No serious adverse events were reported for any type of care. Our review was also unable to clarify whether chiropractic or medical care is more cost-effective. Given the limited available evidence, the decision to seek or to refer patients for chiropractic care should be based on patient preference and values. Future studies are likely to have an important impact on our estimates as these were based on only a few admissible studies.
    Full-text · Article · Aug 2016
Show more

We use cookies to give you the best possible experience on ResearchGate. Read our cookies policy to learn more.