Chronic pain (consistent pain for more than 3 months) has a direct impact on quality of life, days off work, and healthcare costs.1 In the United Kingdom, back pain accounted for £1,632 million in direct expenditure and £10,668 million in indirect costs.2 Furthermore, the total cost estimated for the 22 million patients with migraine in the US was 14.4 billion US dollars3 and 27 billion euros for the 41 million patients in Europe.4 Although the most frequent localization is in the low back,5 neck and shoulder pain are also common forms of chronic pain.1,6

A common finding of these studies was that the prevalence of pain was higher in women than in men.6–15 Bingefors and Isacson reported a prevalence of back pain of 24.3% in women and 20.9% in men, and a prevalence of headache of 17.6% in women and 6.7% in men.9In Spain, only a few studies have investigated the prevalence of pain in adults.7,10–15 Indeed, only two studies have analyzed the prevalence of chronic pain.12,15 The first reported a prevalence of back pain (including the neck and low back) of 21.5%.12 The second studied chronic pain in the upper back and neck and found a prevalence of 14.7%.15 Again, the Spanish studies found a higher prevalence of pain in women.7,10–15

The objectives of this study were 1) to estimate and compare the prevalence of chronic headache (CH), chronic neck pain (CNP) and chronic low back pain (CLBP) by sociodemographic variables in women and men living in the autonomous region of Madrid, and 2) to compare differences between women with and without chronic head, neck or low back pain according to self-perceived health status, lifestyle habits, and use of health services or alternative medicine, and to identify which factors are associated with each of the pain locations in women.

The prevalence of CH, CNP and CLBP in women (7.3%, 8.4%, 14.1% respectively) was higher (P <0.001) than in men (2.2%, 3.2%, 7.8% respectively). In all age groups and all sociodemographic variables, the prevalence in women was at least twice as high as that in men (Table 1).

The prevalence of CH among women was highest in the group aged 45-64 years (9.1%); among men, the prevalence was higher in the groups aged 25-44 and 45-64 years (2.6% and 2.3%, respectively). In both sexes, the prevalence of CNP and CLBP was highest among the group aged ≥65 years. In all age groups, the highest prevalence of chronic pain corresponded to CLBP, except in women in the group aged 16-24 years, in whom the prevalence of CH was higher. The least prevalent type of pain was CNP among persons aged 16-44 years of both sexes, and was CH among those aged 45 years or more. Participants who reported their marital status as ‘married’ showed the highest prevalence of CH whereas those who reported being divorced/separated/widowed showed the highest prevalence of CNP and CLBP. Women and men with no studies or only primary studies had more CNP and CLBP. Finally, participants born in Spain reported more CNP, whereas women born outside Spain had more CLBP (Table 1).

Women with chronic pain of any localization reported poorer self-perceived health than those without pain, showed higher psychological distress, had a greater number of chronic comorbid conditions and used medication and health services more frequently than women with no pain (Tables 2 and 3). Obesity and mobility limitations were associated with the presence of CNP and CLBP. Finally, the prevalence of emergency visits was 10% higher among women with chronic pain than among non-sufferers.

Our results indicate that CH, CNP and CLBP are a major public health problem in the autonomous region of Madrid. Overall, the most prevalent type of pain was CLBP. Chronic pain was more than twice as common in women than in men. In women, the use of analgesics and drugs for inflammation and having ≥3 chronic diseases was significantly associated with chronic pain. Finally, we found that CH, CNP and CLBP have their own independently associated factors.

The prevalence of CH, CNP and CLBP found in our study was slightly different to that reported by other authors.1,6,12,15,21,22 Previous studies found higher prevalences than the current results. In the Netherlands, Picavet and Schouten found a prevalence of 14.3% of CNP (but without showing the data separately by gender).6 In Spain, previous studies reported prevalences of 21.5%12 and 14.7%15 for both CNP and CLBP together. There are also some studies reporting lower prevalence rates. In North Carolina (US), the prevalence of CNP was 2.2%,21 and that of CLBP was 12.2% in women and 8% in men.22 Differences among studies may be related to different definitions of chronic pain, the recall period used, the study population, data collection, and other factors.

In agreement with previous data, we also found that CLBP was the most common form of pain.5 Our results also agree with previous studies in which women reported a higher prevalence of chronic pain.6–11,13,14

The higher prevalence of chronic pain in women could be explained by several biological differences between women and men. Women have less efficient pain habituation and a greater susceptibility to develop temporal summation of chemically-evoked pain,23 and mechanically-evoked pain,24 as well as less efficient diffuse noxious inhibitory mechanisms25 than men. Other mechanisms are related to estrogen fluctuations and certain genetic relationships in women.26

In addition to biological differences, other possible explanations may include the following: 1) in Spain, women experience higher psychological distress than men, and psychological distress is associated with pain;272) women exhibit a greater number of concomitant chronic diseases,28 which are associated with an increased likelihood of experiencing chronic pain and psychological distress; and 3) unconscious healthcare provider bias may affect healthcare delivery with women being diagnosed less or being treated less effectively or aggressively than men.

With regard to the first point, a recent study conducted in Spain found that women presenting with invalidating musculoskeletal pain had a higher probability (OR = 1.38) of suffering psychological distress than men.27

In a similar population-based study using data from the 2005 Madrid Regional Health Survey, Esteban-Peña et al28 showed that all the chronic diseases included in our study were more prevalent in women than in men: osteoarthritis/arthritis/rheumatism (25.4% vs. 8.9%), allergy (13.5% vs. 10.8%), osteoporosis (8.5% vs. 1%), varicose veins (13.4% vs. 3.4%), and depression (12.4% vs. 5.3%).

Regarding the third point, previous studies have shown that patient and physician gender may impact the process of medical care.29 Physical examinations are less frequently performed in women, which may delay diagnostic tests, thereby increasing the probability of pain becoming chronic in women.30 Nevertheless, it is generally assumed that women have more severe chronic pain and depression than men.30

We found that women with CH, CNP and CLBP consumed more analgesics and drugs for inflammation, especially those with CH. Radat and Lanteri-Minet reported that acute headache medications containing psychoactive components, e.g., barbiturates or opiates, were associated with an increased risk of medication-overuse headache.31 This finding may serve as a possible explanation for the current results, indicating that higher drug consumption may perpetuate chronic pain. Medication overuse in individuals with headache may be related to patients’ dependency on some drugs or to the inefficacy of some drugs, particularly antiinflammatory drugs, in providing pain relief.

As well as the higher consumption of analgesics and drugs for inflammation, we also found an association between CLBP and the use of tranquilizers. In a previous study in Spain, we found similar results: the use of tranquilizers was 1.65 times higher among persons with invalidating musculoskeletal pain. A previous study showed that low back pain was strongly associated with osteoarthritis.14 This finding is expected, since facet joint dysfunction, e.g., due to osteoarthritis, is generally considered as a nociceptive source in low back pain.32 Sleeping pills are associated with CH, which agrees with the results observed by Strine et al,33 who found that headache sufferers were more likely to report insomnia, excessive sleepiness, and depressive or anxiety symptoms.

We found that worse self-perceived health and more visits to the emergency center were associated with CH. A recent review investigating determinant and risk factors for neck pain in the general population supports the association between poor psychological health status and neck pain.34

Finally, the search for alternative medicine for pain relief was also associated with CNP and CLBP. This finding may be related to poor recovery with pharmacological treatment.35 Importantly, musculoskeletal pain is a multidimensional problem in which structure and function interact with personal, environmental and social factors.36 We suggest that, as conventional drug treatment covers one part of the puzzle of chronic pain, patients will usually seek additional help from alternative medicine providers.

The strengths of this study include a large sample size, random selection of the study population and the possibility of analyzing a large number of variables that can often not be obtained from medical records. Information was collected individually, and always from the perspective of citizens, which complements the information that can be obtained from other study sources, e.g., health services, hospital data, etc. However, this study has a number of limitations. Firstly, the questions included in the survey to classify patients as having CH, CNP and CLP included self-reported answers and have not been validated. Nevertheless, this method is commonly used in epidemiological studies.7,13,14,27,28 A strength of the diagnosis is that pain should be diagnosed by a physician. Secondly, information obtained from surveys may be subject to recall errors or to participants’ tendency to provide socially desirable responses. Thirdly, the Madrid Regional Health Survey includes non-institutionalized individuals; therefore, our results cannot be generalized to the institutionalized Spanish population, which may lead to underestimation of the prevalence of pain. Fourthly, the survey did not collect information on pain characteristics, i.e., duration, severity, cause, and working conditions, workload, environment, type of work day, i.e. part-time or full-time, all of which may affect the presence of CH, CNP and CLBP. Finally, since this population-based study has a cross-sectional design, a cause and effect relationship cannot be established. Nevertheless, Picavet and Hazes demonstrated that health surveys are a valuable source of information on musculoskeletal pain problems, which is not available from most other sources of information.37

This population-based study indicates that CH, CNP, and CLBP are a public health problem in the autonomous region of Madrid, particularly among women. CLBP was the most prevalent pain. Women had a higher overall prevalence than men for chronic pain in all locations. In women, chronic pain was associated with the use of analgesics and drugs for inflammation and with a greater number of comorbid diseases. Clinicians should be aware of these associated conditions to improve the management of women with CH, CNP, and CLBP.

All authors participated in the study design, data analysis, data interpretation, and drafting and approval of the manuscript submitted for publication.

None.

None.