Data were gathered in 34 hospitals—as 6 of the 40 hospitals chosen initially decided not to take part—, selected on the basis of their excellence when taking part in previous research projects and their interest in studies of this type. The criterion of geographical distribution was also taken into account, so that they were a representative sample of national territory.

Patients were included who were being treated in rheumatology surgeries that were not specialised in a single field, and those for whom recording information reliably would have been impossible were excluded (due to dementia, difficulty in understanding the questions, a lack of linguistic fluency, etc.) or patients with previous trauma that was the main cause of pain and that was not associated with their rheumatic disease.

Patients were selected according to the randomisation list sent to each hospital, based on the order in which they were seen in the surgery. This list was prepared in a centralised way and consisted of 50 randomly selected numbers from 1 to 200. If the corresponding patient fulfilled the selection criteria, they were asked to take part. If they refused to take part, they were replaced by the following patient on the random list. 30 patients per hospital were included (this number having been set for reasons of convenience, as a total sample of 1200 patients would make it possible to estimate prevalences and compare subgroups as well as making comparisons with data from the EPIDOR study). The distribution of new patients (PN) and those in follow-up (PR) included in each hospital was such that it approximately represents the ratio in a typical rheumatology surgery (where the ratio of new patients to patients in follow-up = 1:2).

This study was approved by the clinical research ethics committees and management bodies of all the participating hospitals, and it was undertaken according to the principles of the Helsinki Declaration and the International Guidelines for Ethical Review of Epidemiological Studies (Council for the International Organizations of Medical Sciences-Geneva, 1991). All of the subjects included gave their informed consent prior to taking part in the study.

The doctor completed a standardised questionnaire in each visit, and the patients completed a series of self-administered questionnaires. The second visit took place for those subjects who in their first visit had referred moderate to severe pain, or pain that was insufficiently controlled (pain ≥ 4 on the numerical pain scale), on condition that they had been prescribed a new treatment, modified their existing treatment or were waiting for their prescribed treatment to bring about changes within the short term. This second visit had to take place from one to 3 months after the first one, at the moment when it was estimated that pain should have been controlled or, at least when it was possible to definitively assess the response to the prescribed treatment.

Fieldwork took place from October 2012 to July 2013.

The variables included were:

• -

  Sociodemographic characteristics of the doctor and patient. Lifestyle habits of the latter.

• -

  Date symptoms commenced, reason for visiting (PN or PR for check-up), follow-up time in rheumatology.

• -

  Associated diseases (cardiovascular, respiratory, digestive, neurological, psychiatric, metabolic, kidney disease) and treatment.

• -

  Pain characteristics: duration, location (pain was considered to be generalised when it was located above and below the waist and on both sides of the body), the diagnosis of its cause, type and pattern, intensity and degree of interference–measured on the numerical visual scale ([NVS] used in its printed version, in which 0—no pain—up to 10—the worst imaginable pain—in a horizontal layout that was standardised in all of the hospitals)15 and a short questionnaire to evaluate the pain (BPI)16 and a neuropathic pain diagnostic questionnaire.8

• -

  Pain management: the treatment patients had been prescribed or had received during the month prior to their visit was described, together with the treatment prescribed by the rheumatologist in the basal visit. Information was recorded about the pharmacological analgesic treatment itself (simple analgesics, NSAID, COX-2, minor and major opioids, capsaicin and others), coadjuvants (differentiating between antidepressive drugs, anticonvulsants, muscle relaxants, local anaesthetics, SYSADOA, corticoids, DMD, biologicals, botulinum toxin and others) and interventionist techniques (including corticoid infiltrations, synoviorthesis, radiofrequency, ozone therapy, intra-articular hyaluronate and others). Information was also recorded about non-pharmacological treatment of any type (patient education, psychotherapy, relaxation, physical activity, rehabilitation, physiotherapy, hydrotherapy, masotherapy, chiropraxis, osteopathy, electrotherapy, TENS, acupuncture, homeopathy and others).

• -

  Therapeutic adherence to the treatment prescribed in the basal visit (evaluated by the rheumatologist in the follow-up visit).

• -

  Evaluation of clinical response: on the overall clinical impression of change scale,17 with a specific question about controlled pain, a question about whether the therapeutic response was sufficient to live a normal life, and a question about self-perceived controlled pain in comparison with previous pain.

• -

  Quality of life—using the SF1218 questionnaire, catastrophysing—using the Spanish version of the Pain Catastrophising Scale (PCS)19—, effect on mood—using the hospital anxiety and depression scale (HADS)20 and the Zung scale.21

Descriptive analysis was performed, calculating the absolute and relative frequency for the qualitative variables, and the average and standard deviation (SD) or median and interquartile range (IR) for the quantitative variables, depending on whether or not they were distributed normally. The differences between PN and PR were then analysed using the Student t-test for quantitative variables or non-parametric tests if normalcy could not be assumed, and the Chi–squared test corrected for Yates or Fisher continuity in 2:2 tables for qualitative variables.

P < .05 was considered statistically significant. The SPSS 21.0 program for Windows was used for statistical analysis.

Although at first the number of patients had been set at 30 per hospital for the total sample, when the number of participating hospitals was reduced, others were requested to increase the number of patients they included to compensate for the loss.

Recruitment took place in accordance with the protocol, including consecutive patients as specified by the randomisation table. A total of 1084 patients were recruited. Of the 1075 for whom the type of visit is known, 342 (32%) were PN and 733 (68%) were PR.

50% of the patients had visited the surgery at least twice during the previous year (IR 1–3) and had been in follow-up for at least 34 months (IR 3–94) (Fig. 1) for the distribution of the sample according to hospital.

Fig. 1.

Distribution of the sample by hospital.

Data corresponding to the sociodemographic and occupational variables are shown in Table 1. Regarding occupational disability, there was a significantly higher rate of recognised permanent disability among the patients who visited for a check-up (8.8% in PN; 17.3% in PR).

19% were habitual smokers—53.6% had never smoked—and 72.8% never or hardly ever consumed alcohol—frequent alcohol consumption was very rare (less than 1%)—. A high proportion of patients had a sedentary lifestyle—22.5% had not carried out any free time physical activity in the previous 6 months—. No statistically significant differences were found between PN and PR.

Fig. 2 shows patient comorbidities. Arterial hypertension (25.6%) and hypercholesterolaemia (19.4%) were the most frequent risk factors. No statistically significant differences were found between PN and PR in terms of comorbidity frequency except for digestive disease, which was more prevalent in the latter (15.5% vs. 22.0%, respectively).

Fig. 2.

Comorbidities.

64.7% had a high probability of developing anxiety and 22.0% had a high probability of depression (significantly higher in PN than in PR; 26.5% and 19.7%, respectively). 23% had clinically relevant catastrophisation (score  ≥ 30 on the PCS)22 and their health-related quality of life (according to their SF-12 score) was worse physically (average score 37.5 [SD: 10.6]) and mentally (44.9 [SD: 13.17]) than the reference population, without any significant difference here between PN and PR.

The prevalence and intensity of pain are shown in Table 2.

The locations of pain are shown in Fig. 3 (for all of the pains suffered by patients, regardless of their intensity) and Fig. 4 (for the dominant pain, defined as the one which the patient thought was the most intense or which h most affected their normal activity. Here it was only possible to identify one location).

Fig. 3.

Pain location.

Fig. 4.

Dominant pain location.

Nociceptive pain was the most common dominant pain (83.4% of all the patients in the sample)—somatic pain in 97.3%—, while neuropathic pain was far less common (12.0%) as was psychogenic pain (6.3%). The results for pain duration showed that the majority of the patients had suffered the pain for more than 3 months (80.5% of the total number of patients in the sample). Table 3 shows its form of presentation.

The lumbar area (68%) was the most common location for dominant spinal pain, followed by the hands (56%) in the upper limbs and the knees (45%) in the lower limbs, with similar distribution in PN and PR.

Diagnoses of the dominant pain causes are shown in Fig. 5.

Fig. 5.

Diagnosis associated with dominant pain.

Significant differences were found between PN and PR. In the first of these the most usual diagnoses in order of frequency were arthrosis, soft tissue rheumatism, vertebral pain and inflammatory arthropathies (rheumatoid arthritis, ankylosing spondylitis and psoriatic arthritis were the most frequent, at 44%, 19% and 17% of the total, respectively). In the PR inflammatory arthropathies, arthrosis and soft tissue rheumatism were the most frequent.

If we compare both groups, the cases of arthrosis, soft tissue rheumatism, vertebral pain and neuropathic pain were more common in PN as causes of the dominant pain. Cases of inflammatory arthropathy, on the other hand, were more common in the PR.

Respecting comorbidity treatment, 42.5% of the total number of patients in the sample had taken an IBP gastric protector and 24.5% had taken an antihypertensive drug. There was significantly more frequent use of IBP treatment as well as steroids in the PR group compared to the PN.

85.6% of the total number of patients in the sample had been prescribed a specific pharmacological and/or interventionist treatment for pain (PN 74.8%; PR 90.5%). The frequency according to treatment type is shown in Fig. 6. Regarding pharmacological analgesic treatment, 46.3% of all the patients in the sample had received traditional NSAIDs in the previous month (97.5% of which were oral) and 42.5% had received simple analgesics. Traditional as well as COXIB NSAIDs were used significantly more often by the PR than was the case for the PN, and this was also true for major opioids (48.4%, 12.8% and 5.6%, as opposed to 41.2%, 6.1% and 1.5%, respectively). In the sample as a whole 23.5% had received opioid treatment in the previous month (minor or major).

Fig. 6.

Patients with some type of treatment for pain.

DMD and biological therapy stand out among the coadjuvants used (17%), together with corticoids (13%) and muscle relaxants (12%). Significant differences were detected between PN and PR respecting the use of DMD, biological therapy and corticoids. Corticoids are used far more frequently in PN (95% vs. 16%). On the contrary, DMD and biological therapy are clearly used more in patients visiting during follow-up (24% vs. 2%).

Infiltrations were the most common interventionist technique (9.1% of the total, 10.9% in the PR and 5.3% in PN).

26.3% of patients in the sample had received non-pharmacological therapy in the previous month (significantly more PR than PN; 29.9% and 18.7%, respectively). The use of techniques involving physical mobilisation stands out as more frequent, with physiotherapy (11.1%), rehabilitation (10.5%) and physical exercise and sports activities (10.4%). Rehabilitation was used more in the PR, as was physical exercise. The former was used more often for mechanical disease—12.4% of arthrosis and 19.4% soft tissue rheumatism—than it was for inflammatory arthropathies, at 8.7%; this may explain, at least partially, the predominance of inflammatory disease in the PR).

Table 4 shows the treatment prescribed in the basal visit.

The data on the perception of controlled pain and satisfaction with treatment are shown in Table 5.

490 of the 515 patients who fulfilled the criteria visited again approximately 3 months after their first visit (with an average of 80 days between both visits and a standard deviation of 42). The average pain intensity during the previous week scored using the NVS was 4.5 (higher for PR than it was for PN; 4.7 vs. 4.1).

On the 11 point numerical scale (where 0 is “no pain” and 10 is “the worst imaginable pain”) in the short pain evaluation questionnaire, the median (Me) score for pain intensity was 6 for the maximum intensity in the previous 24 h, 3 for the minimum and 4 for average and current intensity. The difference between the PN and PR is significant for maximum intensity, as it is higher for the patients in follow-up (6 vs. 5), and also, although to a lesser degree, current pain intensity is also significantly higher. The degree to which pain is an interference, evaluated in the same questionnaire (on a scale where 0 means it “does not interfere” and 10 corresponds to “it interferes completely”) was higher for general activities, work and the ability to walk (Me = 4) than it was for mood, sleep and the capacity to enjoy life (Me = 3), and it was lowest in the capacity for social relations (Me = 2).

86.6% of the patients had adhered properly to the treatment they had been prescribed in their previous visit, according to the doctors. Their treatment had been modified for 35.9% of the patients since their basal visit (above all in terms of increasing dosage or frequency of administration), and this occurred more often in the PN than it did in the PR.

Table 6 shows data on perception that pain was controlled and degree of satisfaction with the treatment.