Abstract
Objective. Back pain (BP) is frequent in the community; its prevalence in México is 6%. Our objective was to determine the prevalence of BP in Mexican communities and determine its most important characteristics.
Methods. A cross-sectional study of individuals aged > 18 years was conducted in Mexico City and in urban communities in the state of Nuevo León. Sampling in Mexico City was based on community census and in Nuevo León, on stratified, balanced, and random sampling. Procedures included a door-to-door survey, using the Community Oriented Program for the Control of Rheumatic Diseases, to identify individuals with BP > 1 on a visual analog scale in the last 7 days. General practitioners/rheumatology fellows confirmed and characterized BP symptoms.
Results. In all, 8159 individuals (mean age 43.7 yrs, two-thirds female) were surveyed and 1219 had BP. The prevalence of nontraumatic BP in the last 7 days was 8.0% (95% CI 7.5–8.7). The mean age of these individuals was 42.7 years, and 61.9% were female. Thirty-seven percent had inflammatory BP [prevalence of 3.0% (95% CI 2.7–3.4)]. Compared with the state of Nuevo Léon, the characteristics and consequences of BP in Mexico City were more severe. In logistic regression analysis, living in Mexico City, having a paid job, any kind of musculoskeletal pain, high pain intensity, and obesity among other variables were associated with BP.
Conclusion. The prevalence of nontraumatic BP in the last 7 days in urban communities in México is 8.0%. However, clinical features and consequences differed among the communities studied, suggesting a role for local factors in BP.
Back pain (BP) is one of the most important musculoskeletal (MSK) disorders affecting humans. The prevalence of BP in the general population ranges from 22% to 33%1, but diverse studies indicate that 11% to 84% of the population suffers from BP at some time in their lives2. Moreover, the prevalence of BP increases with age, but in most cases BP occurs between 45 and 49 years of age3. The consequences of BP are linked to transitory or permanent disability4, job loss, and significant costs5.
The prevalence of BP and other MSK pain disorders in México is only partially known. In 2002, Cardiel and Rojas-Serrano6 found a prevalence of 23% for MSK pain and 6.3% for BP among 2500 individuals from an urban community in México City. In their study, Cardiel and Rojas-Serrano relied on an adaptation of the Community-Oriented Program for the Control of Rheumatic Diseases (COPCORD) stage 1 questionnaire for the Mexican population as a screening tool. Other COPCORD studies, particularly those in developing nations, report figures that ranged from 4.3% to 43.8%7,8,9,10,11,12,13,14. In addition to such findings, some studies have also attempted to determine the cause of BP, but results vary widely1,2,3. Overall, only 15% to 20% appear related to identifiable causes15.
In our study, we determined prevalence of BP in the community using the Mexican adaptation of COPCORD as a screening tool. We also tried to determine some of the most important characteristics of BP and identify factors related to its presence in the community.
MATERIALS AND METHODS
We performed a cross-sectional study of individuals ≥ 18 years of age residing for ≥ 6 months in México City and in urban communities in the state of Nuevo León. The protocol of our study was approved by the Ethics Committee of each center. All participants in the study were informed about procedures of the protocol, and after agreeing to participate, they signed an informed consent.
Sampling and settings
Sample size calculation was based on a prevalence of MSK pain of 50%, a level of uncertainty of 3%, a confidence level of 95%, and a statistical power of 80% to discriminate up to 5% differences in prevalence16. In México City, sampling was based on a community census performed by the American British Cowdray Medical Center primary health clinic as part of a health program for vulnerable populations. In the state of Nuevo León, sampling was based on mixed and complex strategies in which an updated census of the country17 was used to generate a stratified (by region), balanced, and random sample of subjects ≥ 18 years of age, representative of all health states; in each region of the state, a second random assignment was done to select one or more municipalities (according to proportional sample size); in each municipality one or more basic geostatistical areas13,14 were selected to carry out interviews in all homes. The study was carried out in Corpus Christi, Santa Lucía, and Cuajimalpa III districts in Cuajimalpa, México City, and in urban communities of the state of Nuevo León. México City — the largest city in the country — has a population of 8,720,916 inhabitants (8.4% of Mexico’s population); 99.7% of that population is urban and their mean education level is 10.2 years. The state of Nuevo León — divided into 7 regions, located in northeast México, on the southern border of the USA — has a population of 4,199,292 inhabitants.
Procedures
During the first phase of the study, 6 interviewers were trained on the protocol, and procedures were standardized. Individuals age ≥ 18 years who lived at an address provided in the last 6 months were invited to participate in the study. The COPCORD questionnaire was administered at home, in a door-to-door survey. If needed, interviewers returned up to 5 times to the same home to administer the questionnaire. Surveys were cross-checked by the different interviewers during screening and later by the coordinators. Individuals with BP — including pain associated with trauma — scoring > 1 on a visual analog scale (VAS) ranging from 0 (no pain) to 10 (very painful), occurring in the last 7 days or at any time in the past, were examined by general practitioners and rheumatology fellows to confirm the information and characterize BP symptoms. The interval between the door-to-door survey and physical examination was 7 days. The definition of BP for the calculation of prevalence and all other analysis was nontraumatic BP scoring > 1 on a VAS (0 to 10) occurring in the last 7 days as confirmed by a study-trained physician.
Statistical analysis
We performed an exploratory analysis of the variables included in the theoretical model. Univariate analysis of each variable was done using one-way and 2-way analysis of variance for the continuous variables, and chi-squared test for ordinal, nominal, or categorical variables. Adjusted OR were estimated through logistic regression. Statistical analysis was performed using Stata SE version 9.0 for Windows18.
RESULTS
Surveys were carried out between August 2008 and August 2009. In total, 8159 individuals from all the communities included in the study participated in the COPCORD survey. Their mean age was 43.7 years and nearly two-thirds of the participants were female (Table 1). Most individuals were married and had a paid job.
A total of 4201 (51.4%) individuals reported having MSK pain over the last 7 days, with a median intensity of 2 (Table 1). Most individuals did not relate their MSK pain to trauma and most had never experienced physical limitation as a consequence of it; past physical and current limitations were reported by 12.8% and only 5.8%, respectively. Around 30% had problems in adjusting to discomfort and 8.7% had physical disability according to the Health Assessment Questionnaire-Disability Index score. Regarding treatment, 40.7% received some form of treatment, which mostly (84%) consisted of nonsteroidal antiinflammatory drugs or simple analgesics. Most individuals with MSK pain (74.6%) took medications by themselves or by recommendation of non-health professionals. On the other hand, 5034 (61.7%) individuals reported some comorbidity.
In the door-to-door survey, 1219 individuals (14.9%, 95% CI 14.1–15.7) reported BP. However, the definite prevalence of nontraumatic BP in the last 7 days and/or in the past was 13.9% (95% CI 13.2–14.7) according to physician-confirmed diagnosis in 1141 out of 1219 individuals. Mean age was 42 years and nearly 60% of the cases were female (Table 2). Besides BP in the last 7 days, most individuals (69.1%) reported having BP in the past (Table 2). Less than 25% of all individuals with pain related its onset to trauma.
Six hundred-sixty individuals, 303 from México City and 357 from the state of Nuevo León, fulfilled the definition of nontraumatic BP scoring > 1 on a VAS in the last 7 days, with a prevalence of 8.0% (95% CI 7.5–8.7). The mean pain intensity on VAS was 6.8. More than 50% of individuals with BP received medication. However, less than 10% of BP cases had any current physical limitation. Interestingly, 37.8% of individuals with BP had symptoms compatible with inflammatory BP (IBP), for which prevalence in these communities was 3.0% (95% CI 2.7–3.4).
Comparison of demographic and clinical variables related to BP between México City and Nuevo León communities yielded significant differences in several variables (Table 2). The prevalence of nontraumatic BP in the last 7 days was 3.7% (95% CI 3.3–4.1) in México City and 4.3% (95% CI 3.9–4.8) in Nuevo León (p < 0.1). The prevalence of BP not related to trauma versus BP related to trauma were 9.0% (95% CI 8.1–9.9) and 2.9% (95% CI 2.4–3.4) in México City; and 12.5% (95% CI 11.5–13.6) and 3.4% (95% CI 2.9–4.0) in Nuevo León (p < 0.1 vs México City in both comparisons). Similarly, the prevalence of BP at any time in the past and IBP were 7.4% (95% CI 6.6–8.2) and 4.6% (95% CI 3.9–5.2) in México City and 11.8% (95% CI 10.9–12.9) and 1.5% (95% CI 1.1–1.9) in Nuevo León (p < 0.1 vs México City in both comparisons).
Overall, BP was higher as measured by pain VAS and appeared to have more consequences among México City individuals than those from Nuevo León (Table 3). The prevalence of comorbidities was generally higher in México City (Table 3). Rank order was also different; for example, gastritis, peripheral vascular disease, and blood hypertension were ranked in first to third place in México City, whereas in Nuevo León, these ranks were occupied by blood hypertension, gastritis, and diabetes mellitus.
Individuals with nontraumatic BP scoring > 1 on VAS in the last 7 days differed from all other individuals — regardless of their clinical status — in some variables at a significant level (Table 4). As expected, such differences reflected the presence of pain and associated factors.
In logistic regression analysis, the variables associated with presence of BP were living in México City and having a paid job; while the variables associated with MSK pain in the last 7 days and any time in the past were high pain intensity and treatment, including complementary/alternative medicine (Table 5). Obesity and other comorbidities were also associated with BP.
DISCUSSION
The prevalence of nontraumatic BP in the urban communities of 2 major cities in México was 8.0%, which is in the range of most studies in this area. Interestingly, however, such prevalence is higher than that in other Latin American countries with the COPCORD stage 1 questionnaire, and in the wide range of reports from the Middle East and Asia with the same methodology. Specifically, the prevalence of BP in reports from Latin America was 6.3% in México City6, 7.0% in Lima, Peru19, and 5.7% in Santa Clara, Cuba7. Lower figures have also been found in Bangladesh (9.2%)12, China (8.0%)13, Vietnam (11%)11, and Pune, India (5.5%)20. Prevalence of BP in Iran was 15.4%21 and in Thailand 22.7%9. Based on such data, we may conclude that although all these studies relied on COPCORD as a screening tool, variations in the prevalence of BP exist.
Demography of the populations in such studies shows little variation, and methodology appears to be the same. It is possible that local additional factors influenced our results, particularly the mean age of the population.
When we compared the prevalence and the characteristics of BP in the 2 communities in our study, we found significant differences. In México City communities, the prevalence of BP was in general lower than in the state of Nuevo León. Only IBP was more frequently found in México City. It is possible that differences in sampling and perhaps response rate could explain some differences; by looking at the communities included in this study, it seems that those in México City were more subject to chronic social stress than in the state of Nuevo León. Interestingly, most indicators of severity, including the consequences of BP, were significantly higher in México City. There were also significant differences between the 2 cities in regard to the frequency and rank order of comorbidity. These differences could also be explained by demographic features.
There were differences between individuals with and without BP, which in logistic regression analysis appeared to be positively associated with the presence of BP. These included pain intensity, markers of disease severity, and having a paid job. Interestingly, 16% to 25% of the compensation paid because of disease in the United States was for BP, and 52% of all BP cases were work-related23.
More men than women had BP (15.4% vs 13%, respectively), which contrasts with most COPCORD studies. In Thailand, 9.9% to 17.3% were women and 4% to 11% men9; in China, 4.7% were male and 7% female13; and in India, 7.3% were male and 15.4% female10. The relationship between BP and work includes farming activities and activities within the informal employment sector. Similar results were found in our study (data not shown).
One-third of individuals with BP in our study had IBP, with an overall prevalence in the community of 3.0%. The identification of IBP was based on identification of specific symptoms, including age at onset and duration of BP, as well as pain and stiffness triggered by bed rest that improves with movement. Although no other COPCORD study approached IBP, it is known that around 5% of individuals with BP have IBP24.
One of the limitations of our study was the over-representation of women and a lower response rate in México City, which could account for some of the differences between the 2 cities. Another limitation was the difference between cities in the methodology for case ascertainment, particularly for individuals with IBP. Lastly, we have to consider that there could be some representative limitations in our study, although sampling was very carefully done.
Nevertheless, we may conclude from this study that BP is a common MSK disorder in the Mexican population, where 8.0% of individuals included in this study complained of nontraumatic BP in the last 7 days. BP predominantly affects middle-aged individuals, who, interestingly, rarely have limited function or inability to work as a consequence of BP. Finally, we found a relatively low prevalence of IBP, but importantly, this finding may indicate the need for study of spondyloarthropathies. Differences between communities suggest a role for local factors in the prevalence and consequences of BP.
Acknowledgments
For their help in the study, we are deeply grateful to Gerlinde Schlang, Nadia Saavedra-Cervantes, Diana Anaya, Everardo Alvarez, Bernardo Julian, Ivonne Arena, Adalberto Loyola, Veronica Linares, Nidya Cordero-Maldonado, Denisse Alvarez, Nancy García, Guillermo Perez, and Sara Marín. We also thank all individuals from Santa Lucía, Cuajimalpa III, and Corpus Christi community who participated in the study.
Footnotes
-
Supported by a research grant from the National Council of Science and Technology (CONACYT) Project Salud 2007-C01-69765 and Salud 2007-C01-69439.