The use of the Spanish Healthcare Services: a comparison between people with and without pain

Abstract

Background

Chronic pain is a prevalent condition, requiring regular access to healthcare services. Non-chronic continuous pain may represent an early stage preceding chronic pain. We compared healthcare use due to any reason among people without pain, chronic and non-chronic continuous pain in Spain, and factors related.

Methods

A cross-sectional survey in the general Spanish adult population (n = 7058) was conducted. Participants were classified as no pain, non-chronic continuous pain, or chronic pain. Data included sociodemographic information, health-related quality of life, anxiety, depression, social support, labour consequences, and healthcare use due to any reason in the past four weeks. Logistic regression models were used to identify factors associated with healthcare utilization in the 3 groups.

Results

Prevalence of chronic pain was 25.9%, and non-chronic continuous pain, 7.7%. Both groups used healthcare more than the no pain group (≈ 59% vs. 28%, p < 0.001). Better quality of life was consistently associated with a lower likelihood of healthcare visits across groups, while sick leave increased its likelihood, especially among chronic pain subjects (OR > 2.5). Female sex and social support were associated with healthcare use in the no-pain group, whereas increasing age was associated with fewer visits in the non-chronic continuous pain group.

Conclusion

Chronic and non-chronic continuous pain are associated to increased healthcare utilization due to any reason. Primary care planning should prioritize integrating bio-psychosocial interventions and decoupling routine clinical care from mandatory bureaucratic disability evaluations to alleviate system overloading.

Background

The International Association for the Study of Pain (IASP) defines pain as “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage.” Pain may become chronic, being considered chronic pain (CP) when it recurs or persist longer than 3 months (ICD-11) [1]. However, a preceding stage may be considered, in which the pain is not yet chronic (less than 3 months in duration) but has also not resolved within the expected timeframe to be considered acute. In this article, as well as in previous work by our group [2], we refer to this type of pain as “non-chronic continuous pain” (NCCP). Studies show that this pain may progress to CP [3], representing an early warning sign that should be considered, in order to prevent its evolution to chronicity [4]. Thus, the knowledge of this type of pain is critical for secondary prevention to stop the transition into long-term persistence.

CP is a complex disease that includes biological, psychological, and social dimensions [5]. In Spain, CP affects an estimated 25.9% of the adult population [2] and, as in several studies carried out in Europe, their prevalence is higher in women, older adults, and individuals with lower socioeconomic status [6], [7]. Other factors include unhealthy lifestyle behaviour [8], comorbid mental conditions (such as anxiety and depression), lower educational attainment, unemployment, and lack of social support [9], [10]. Low social support is a critical modulator of pain perception and coping, although not always adequately addressed [11]. Furthermore, CP significantly diminishes quality of life and restrics physical, emotional, and social engagement, contributing to reduced productivity and increased absenteeism [2], [12].

Given the nature of this problem, we understand that NCCP and CP require attention from public health systems [13]. Individuals with CP often require regular healthcare access, resulting in distinct utilization patterns that reflect their increased needs and leads to barriers to care ]14]. Previous studies highlight the burden of CP on healthcare systems, noting high use of primary care, emergency services, and specialist consultations [15] which leads to an increase in the individual, social, and health cost. In spite of this, direct comparisons between the general population and people with CP remain unclear. Additionally, the use of healthcare services in subject with NCCP could be important. However, to the best of our knowledge, this issue is not well known.

In view of the above, we aimed to compare the use of healthcare resources due to any cause among people with no-pain, NCCP and CP, as well as their associated factors in each of these subgroups.

Methods

Study design

The study, which is part of the “Pain Barometer” project, is an observational cross-sectional study carried on a representative sample of the general Spanish population (18–85 years) to determine the prevalence and characteristics of CP in the Spanish population [2].

Sampling method

The study population was obtained through volunteer subjects from a census property of the company CINT (platform for gathering digital information oriented exclusively to purposes related to market research, providing access to panels of participants that comply with the requirements of the European Society for Opinion and Marketing Research, ESOMAR and the International Organization for Standardization, ISO). Access to the subjects who participated in the survey was carried out through Access’ own platform, guaranteeing anonymity and without having access to the personal information of the individuals. Data collection was conducted by More Than Research (Madrid, Spain), an independent demoscopic company [16]. Data were provided to the authors of this manuscript fully anonymised. This study was performed in line with the principles of the Declaration of Helsinki.

A sample size was calculated for the main objective of the project: to estimate a prevalence of CP of 28.4% (estimated from preliminary results from a pilot study in the target population), with a 95% confidence level and accuracy of 13%. Quotas by strata were established according to sex, age, and Autonomous Community (specific territorial boundaries in Spain) proportionally to the general population distribution to guarantee a representative sample.

The information was collected using computer-assisted web interviewing (CAWI) for individuals from 18 to 75 years old, and computer-assisted telephone interviewing (CATI) for individuals from 76 to 85 years of age. A total of 7,058 interviews were obtained (6,394 CAWI and 664 CATI) for the original study.

Measures and variables

Data were collected between June 13th and June 20th, 2022 through a structured questionnaire. The questionnaire was specifically developed for the Pain Barometer study, and its Spanish version has been previously described elsewhere [17] (see Annex 2 within that document). As reported in our previous study [2], two screening questions were used to identify the people with NCCP and CP following the criteria of the International Association for the Study of Pain (IASP). An individual was considered to have NCCP if they answered “yes” to the question: “Have you suffered from pain at least 4 or 5 days a week during the last month?” They were considered to have CP if, once they had answered “yes” to the aforementioned question, they also answered “yes” to the question: have you had it for at least 3 months?”

The data collected included sociodemographic data (age, sex, educational level and employment status), the presence of anxiety and depression, health related quality of life (HRQL), social support, and labour consequences (sick leave in the last year).

Regarding the healthcare resource utilization, data on the frequency of visits to healthcare facilities over the past 4 weeks due to any cause (not restricted to pain-related causes) were collected. This includes how often subjects attended primary, specialized, hospital, and emergency care. By not restricting the analysis to pain-related healthcare visits, we will be able to assess the overall use of healthcare services and compare it between groups. Hereafter, when reporting healthcare resource use, the reader can assume that it refers to visits due to any cause.

Anxiety and depression were evaluated using the Spanish adaptation of the Hospital Anxiety and Depression Scale [18]. This questionnaire consists of 14 items divided into two subscales of 7 items for anxiety and seven items for depression. Responses are collected on a 4-point Likert-type scale ranging from 0 to 3. The range of scores for each subscale is 0–21, with higher scores corresponding to higher levels of anxiety and depression. For both subscales, the authors suggested that scores above 11 would indicate a ‘case’ and between 8–11 would be considered a ‘doubtful case’ of anxiety and depression [19]. The internal consistency for the chronic pain patients in the Spanish population has yielded satisfactory results, with Cronbach alphas of 0.74 for the anxiety dimension and 0.80 for the depression one [20].

Health-related quality of life was evaluated with the Spanish version of the SF-12v.2 health survey [21]. This scale is comprised of 12 items and eight dimensions: physical functioning (PF), role-physical (RP), bodily pain (BP), general health perception (GH), vitality (V), social functioning (SF), role-emotional (RE), and mental health (MH). Scoring of the dimensions range from 0 to 100, with the highest scores indicating a better quality of life. Moreover, this instrument has two global dimensions: the physical component summary (PCS), which assesses global physical health, and the mental component summary (MCS), which evaluates global mental health. The scores of these global components are studentized with a median = 50. For this study, only the global dimensions of PSC and MSC were included.

Perceived social support was assessed using the Duke Perceived Social Support Questionnaire (DUKE). This questionnaire is composed of two subscales: confidential social support (ability to communicate with others) and affective social support (received affection and empathy). Scores range from 11 to 55, with higher scores indicating greater perceived social support. Scores below 32 are considered to indicate low perceived social support [22].

Statistical analysis

A descriptive analysis was carried out for the three groups of pain (CP, NCCP and without pain) differentiating between people who had/had not visited healthcare in the last four weeks. To determine the differences between subjects who had/had not visited healthcare in the last 4 weeks in each pain group, non-parametric Mann-Whitney U tests were used, given the nature of the quantitative variables compared. Chi-square tests were used for qualitative variables, with the Likelihood ratio when more than 20% of the expected values were lower than 5. The normality of the distribution of the variables was checked using the Kolmogorov-Smirnov test.

Focusing on the people who used healthcare services, we described the resources used (primary, specialized, hospital or emergency) and compared them among the 3 groups of pain (no pain, non-chronic continuous pain, and chronic pain) with Chi-square test and post-hoc comparisons between pairwise groups, with Chi-squared tests as well.

Three binary logistic regression models were fitted to find the factors related to the use of health resources due to any cause in: (A) population with no pain; (B) people with NCCP and (C) people with CP. The dependent variable was the healthcare visit in the last 4 weeks due to any reason (Yes/No) in all models. All the sociodemographic and health related covariates were tested for potential inclusion in the models. A stepwise selection method was used, based on the Wald tests. As this method might imply risk of overfitting, unstable coefficient estimates or inflated Type I error, we also present adjusted estimates considering all covariates regardless of statistical significance, for direct comparison. The global goodness of fit was checked with the Hosmer-Lemeshow test.

All the analyses were conducted using the IBM SPSS v.29 (IBM Corp, Chicago, IL, USA). In all cases, a significance level of α = 0.05 was established.

Results

Characteristics, frequency of healthcare visits due to any reason and comparison between subjects who had/had not visited healthcare in the last 4 weeks in the groups with CP, NCCP, and no pain

A total of N = 7058 subjects took part in the survey, of which N = 1825 (25.9%) reported CP and 543 (7.69%) had NCCP. Their characteristics can be seen in Table 1.

Table 1.

Characteristics and comparison between subjects who had/had not visited healthcare due to any reason in the last 4 weeks in the groups with chronic pain (CP), non-chronic continuous pain (NCCP), and no pain

Variable	Category	No pain (N = 4.690) Healthcare visit in the last 4 weeks					Non chronic continuous pain (N = 543) Healthcare visit in the last 4 weeks					Chronic pain (N = 1.825) Healthcare visit in the last 4 weeks
		No		Yes		p-value	No		Yes		p-value	No		Yes		p-value
		N	%	N	%		%	N	%	N		N	%	N	%
Age group	18–34 35–54 55–75 76–85	761 1355 941 317	22.6 40.2 27.9 9.4	357 444 368 147	27.1 33.7 28 11.2	< 0.001 a	63 85 60 11	28.8 38.8 27.4 5	115 107 80 22	35.5 33 24.7 6.8	0.252a	106 281 279 76	14.3 37.9 37.6 10.2	194 438 360 91	17.9 40.4 33.2 8.4	0.040 a
Age	Mean (SD) Median (RI)	49.16 (16.32) 48 (25)		49.03 (17.8) 48 (30)		0.629b	45.89 (15.94) 45 (24)		44.91 (17.33) 42.5 (28)		0.367b	52.78 (15.38) 53 (22)		50.65 (15.68) 51 (22)		0.004 b
Sex	Male Female	1899 1475	56.3 43.7	653 663	49.6 50.4	< 0.001 a	96 123	43.8 56.2	138 186	42.6 57.4	0.774a	293 449	39.5 60.5	460 623	42.5 57.5	0.203a
Educational level	No studies Primary education Secondary education Technical/Vocational education University education Unspecified	33 212 670 786 1657 16	1 6.3 19.9 23.3 49.1 0.5	11 81 256 299 668 1	0.8 6.2 19.5 22.7 50.8 0.1	0.403a	1 9 35 58 116 0	0.5 4.1 16 26.5 53 0	2 19 71 77 154 1	0.6 5.9 21.9 23.8 47.5 0.3	0.362c	9 63 180 218 270 2	1.2 8.5 24.3 29.4 36.4 0.3	10 71 221 317 460 4	0.9 6.6 20.4 29.3 42.5 0.4	0.082a
Employment status	Currently working Unemployed Retired Permanent absolute disability Studying Unpaid housework	2085 298 697 25 133 136	61.8 8.8 20.7 0.7 3.9 4	687 121 325 32 86 65	52.2 9.2 24.7 2.4 6.5 4.9	< 0.001 a	145 23 28 2 13 8	66.2 10.5 12.8 0.9 5.9 3.7	190 30 58 4 25 17	58.6 9.3 17.9 1.2 7.7 5.2	0.414a	370 68 193 38 23 50	49.9 9.2 26 5.1 3.1 6.7	551 130 236 69 29 68	50.9 12 21.8 6.4 2.7 6.3	0.141a
Sick leave in the last year (all causes).	No Yes	1656 429	79.4 20.6	470 217	68.4 31.6	< 0.001 a	43 102	29.7 70.3	87 103	45.8 54.2	0.003 a	114 256	30.8 69.2	314 237	57 43	< 0.001 a
Health status
Depression (HADS-D)	Mean (SD) Median (RI)	3.91 (3.62) 3 (5)		4.77 (3.87) 4 (6)		< 0.001 b	5.5 (3.93) 5 (6)		5.72 (4) 6 (6)		0.554b	6.54 (4.34) 6 (6)		7.67 (4.48) 7 (7)		< 0.001 b
	No case Doubtful case Case	2793 376 205	82.8 11.1 6.1	996 201 119	75.7 15.3 9	< 0.001 a	149 41 29	68 18.7 13.2	220 56 48	67.9 17.3 14.8	0.827a	461 151 130	62.1 20.4 17.5	544 264 275	50.2 24.425.4	< 0.001 a
Anxiety (HADS-A)	Mean (SD) Median (RI)	4.52 (3.66) 4 (5)		5.57 (4.07) 5 (6)		< 0.001 b	6.35 (3.72) 6 (5)		6.99 (4.26) 7 (6)		0.116b	7.16 (4.25) 7 (6)		8.62 (4.55) 8 (7)		< 0.001 b
	No case Doubtful case Case	2666 473 235	79 14 7	918 229 169	69.8 17.4 12.8	< 0.001 a	136 52 31	62.1 23.7 14.2	181 72 71	55.9 22.2 21.9	0.075a	405 180 157	54.6 24.3 21.2	452 284 347	41.7 26.2 32	< 0.001 a
HRQL (SF12v.2). Physical component summary (PCS).	Mean (SD) Median (RI)	51.7 (6.85) 53.17 (7.64)		48.79 (8) 50.26 (10.62)		< 0.001 b	46.49 (8.13) 46.82 (10.93)		43.99 (8.88) 45.06 (11.66)		0.002 b	38.93 (10.81) 39.69 (16.08)		37.32 (10.24) 37.78 (15.33)		0.001 b
HRQL (SF12v.2). Mental component summary (MCS).	Mean (SD) Median (RI)	49.46 (10) 51.5 (14.35)		46.62 (11.23) 47.76 (16.36)		< 0.001 b	44.59 (10.06) 43.78 (15.54)		44.57 (11.41) 44.32 (16.49)		0.993b	44.39 (11.58) 44.07 (16.64)		41.09 (11.47) 40.26 (15.94)		< 0.001 b
Social support (DUKE)	Mean (SD) Median (RI)	41.75 (10.18) 43 (16)		41.08 (10.2) 43 (16)		0.028 b	38.5 (10.68) 40 (15)		39.64 (10.44) 41 (15)		0.254b	37.92 (11.42) 39 (16)		36.57 (11.61) 37 (17)		0.010 b
	Low perceived social support Normal perceived social support	497 2877	14.7 85.3	232 1084	17.6 82.4	0.014 a	57 162	26 74	64 260	19.8 80.2	0.085a	190 552	25.6 74.4	339 744	31.3 68.7	0.008 a

SD: Standard deviation; IR: Interquartile range; HRQL: Health related quality if life

^aChi-square.^bU Mann-Whitney.^cLikelihood ratio

Bold p-values highlight results statistically significant

Regarding the frequency of healthcare visits, we highlight that the percentage of healthcare visits in the last 4 weeks in both CP and NCCP groups was above 59%, significantly higher than the group without pain (28.1%, p < 0.001). On the other hand, CP patients visited primary care and specialized care more frequently than NCCP patients and people without pain (Table 2). The health resources were more used by younger people in the group with CP (p = 0.004) and by women in the group without pain (p < 0.001). In subjects without pain and in CP patients, people who had used healthcare services in the last 4 weeks presented higher scores of depression and anxiety scales and more cases of these pathologies, lower levels of physical and mental HRQL, and lower perceived social support (p < 0.05 in all cases) (Table 1).

Table 2.

Frequency of healthcare visits due to any cause by groups

Variable	Category	No pain (A) (N = 4.690)		Non chronic continuous pain (B) (N = 543)		Chronic pain (C) (N = 1.825)		p-value	p-value A vs. B	p-value A vs. C	p-value B vs. C
		N	%	N	%	N	%
Visit to healthcare due to any reason in the last 4 weeks	Yes No	1.316 3.374	28.1 71.9	324 219	59.7 40.3	1.083 742	59.3 40.7	< 0.001 a	< 0.001 a	< 0.001 a	0.892a
Frequency of primary care visits (only those who answered yes to the first question in this table)	Never 1 time 2 times 3 times or more	261 791 176 88	19.8 60.1 13.4 6.7	58 172 63 31	17.9 53.1 19.4 9.6	128 565 247 143	11.8 52.2 22.8 13.2	< 0.001 a	0.007 a	< 0.001 a	0.012 a
Frequency of specialized care visits (only those who answered yes to the first question in this table)	Never 1 time 2 times 3 times or more	615 524 115 62	46.7 39.8 8.7 4.7	148 117 32 27	45.7 36.1 9.9 8.3	309 464 169 141	28.5 42.8 15.6 13	< 0.001 a	0.054a	< 0.001 a	< 0.001 a
Frequency of hospital care visits (only those who answered yes to the first question in this table)	Never 1 time 2 times 3 times or more	1.119 140 26 31	85 10.6 2 2.4	253 43 13 15	78.1 13.3 4 4.6	848 139 33 63	78.3 12.8 3 5.8	< 0.001 a	0.006 a	< 0.001 a	0.709a
Frequency of emergency visits (only those who answered yes to the first question in this table)	Never 1 time 2 times 3 times or more	934 313 41 28	71 23.8 3.1 2.1	156 129 22 17	48.1 39.8 6.8 5.2	657 279 70 77	60.7 25.8 6.5 7.1	< 0.001 a	< 0.001 a	< 0.001 a	< 0.001 a

^aChi-square

Bold p-values highlight statistically significant results

Factors related to healthcare visits due to any reason by groups

In the group without pain (Model A), the factors related to healthcare visit included female sex (OR = 1.294), better perceived social support (OR = 1.021) and sick leave in the last year (OR = 1.45). On the other hand, people with better quality of life (both physical and mental) were less likely to visit healthcare (Table 3).

Table 3.

Factors related to healthcare visit due to any reason by groups

Models for healthcare visit due to any reason.	Model A No pain (N = 2772)			Model B Non-chronic continuous pain (N = 335)			Model C Chronic pain (N = 921)
Variable	OR	95%CI(OR)	p	OR	95%CI(OR)	p	OR	95%CI(OR)	p
Intercept	39.293 131.547		< 0.001 < 0.001	16.872 13.09		0.001 0.155	11.246 1.98*10^10		< 0.001 > 0.999
Sex Male* Female	1.294 1.264	(1.079;1.551) (1.052;1.519)	0.005 0.012
Age (years)				0.973 0.974	(0.954;0.991) (0.955;0.994)	0.004 0.009
PCS (Physical quality of life. The higher, the better)	0.930 0.923	(0.917;0.942) (0.910;0.936)	< 0.001 < 0.001	0.965 0.966	(0.938;0.994) (0.934;0.998)	0.018 0.048	0.974 0.970	(0.960;0.989) (0.953;0.988)	0.001 0.001
MCS (Mental quality of life. The higher, the better)	0.956 0.952	(0.947;0.966) (0.938;0.965)	< 0.001 < 0.001				0.968 0.969	(0.955;0.981) (0.949;0.991)	< 0.001 0.005
Social support (DUKE score. The higher, the better)	1.021 1.015	(1.011;1.032) (1.004;1.026)	< 0.001 0.006
Sick leave in the last year No* Yes	1.450 1.416	(1.184;1.776) (1.154;1.737)	< 0.001 < 0.001	1.723 1.716	(1.070;2.773) (1.055;2.792)	0.025 0.030	2.539 2.513	(1.907;3.380) (1.876;3.368)	< 0.001 < 0.001
Dependent variable: Visit to healthcare due to any reason in the last 4 weeks (Yes vs. No*). *Reference category. Italic values: Adjusted values when including all covariates in the model regardless of statistical significance.	Hosmer-Lemeshow: Chi-square = 14.587; df = 8; p = 0.068. Hosmer-Lemeshow: Chi-square = 7.026; df = 8; p = 0.534.			Hosmer-Lemeshow: Chi-square = 5.399; df = 8; p = 0.714. Hosmer-Lemeshow: Chi-square = 5.0807; df = 8; p = 0.669.			Hosmer-Lemeshow: Chi-square = 7.350; df = 8; p = 0.499. Hosmer-Lemeshow: Chi-square = 11.049; df = 8; p = 0.199.

In the NCCP group (Model B), physical quality of life (OR = 0.965) and sick leave (OR = 1.723) showed an association like the observed in people without pain. In addition, people were less likely to visit healthcare as age increased (OR = 0.973) (Table 3).

Finally, in the CP group (Model C), physical and mental HRQL were found to be related to less likelihood of healthcare visits again (ORs around 0.97), and sick leave increased the risk (more than 2.5 times), with a higher OR compared to the other groups (Table 3).

The three models showed a good global fit. In addition, when comparing the results with the estimations in the saturated models, we found virtually identical results, minimizing the risks expressed in the methods section.

Discussion

In this paper, we compared the healthcare visits due to any reason among people with no-pain, NCCP and CP, as well as their associated factors. Among the main results, we highlight the differences related to the healthcare consultations depending on the group. These results will be discussed below.

Participants with NCCP and CP exhibited a significantly higher frequency of healthcare utilization in the last 4 weeks compared to those without pain. The most significant disparity was observed in visits to specialists and/or emergency departments. This trend is consistent with global data. A survey carried out in Germany, France, Italy, Spain and the UK showed that 93% of people with CP had consulted a healthcare professional in the last month and 25% of those with severe pain had been attended in an emergency room [23]. Supporting this, a recent population-based study from Denmark indicates that individuals reporting high-intensity pain present, on average, twice as many primary care consultations and significantly higher hospitalization rates compared to those without pain [24].

This high use might be related to several factors. Specialist physicians provide more in-depth care for complex conditions like CP, and as waiting times for specialists increase [25], patients may concurrently seek more immediate attention in emergency care. Furthermore, recent evidence suggests that healthcare utilization pattern among pain patients are closely linked to treatment efficacy and safety profile. Specifically, North American studies have demonstrated that the association between pain treatment regimens and subsequent healthcare use reflects both medication side effects and ongoing, poorly controlled pain [26]. Additionally, CP patients undergoing treatment may also experience acute pain flares, which often coincide with new alarming symptoms and concerns about severe pain-related impairment, frequently tracking alongside the utilization of health resources for immediate relief or treatment readjustments [27–29].

The characteristics of patients using health resources were different between groups. Younger people in the general population and in the CP groups were more likely to have visited healthcare during the four weeks of the assessment period. This finding contrasts the literature, which indicates that older people usually utilise healthcare services more frequently in both populations [30]. This shift can be explained by distinct clinical and life-stage drivers. Our descriptive analysis (data not shown) reveals that the younger group show more cases of anxiety and depression, regardless of pain status. This is consistent with global trends marking mental health as a dominant factor of care-seeking in youth [31]. Furthermore, healthcare utilization in working-age cohorts frequently co-occurs with conditions that threaten occupational functioning, such as migraines and lumbar or cervical pain [32, 33], which were the most prevalent diagnoses in our sample of young adults (data not shown). We hypothesize that these clinical factors may be exacerbated by broader systemic determinants. In Spain, high youth unemployment (27.2% in 2021, coinciding with our data collection period [34] and labour market instability could act as chronic stressors. Although not directly measured in our study, this socioeconomic distress might manifest as a dual driver for healthcare use: first, by the somatization of psychological distress, and second, by increasing the bureaucratic need for medical evaluations to justify sick leaves in highly precarious work environments. Conversely, older adults in our sample might face different access barriers or normalise pain as a part of aging [35], leading to more proactive, acute care-seeking behaviours in the younger cohort [36]. These findings have direct implications for primary care and preventive medicine. First, prevention strategies must shift from purely biomechanical approaches to integrated bio-psychosocial interventions within workplaces, aiming to mitigate the impact of occupational stress on pain chronification.

Regarding the factors associated with the healthcare visit, our regression analysis results suggest that sociodemographic factors have only a minor impact on healthcare utilization, as only being female was associated with more healthcare visits in people without pain. This finding calls into question the preconceived idea that women with pain use healthcare services or seek medical attention more frequently than men. To explain this, we hypothesize that women categorised under “no chronic pain” may actually be experiencing diagnosis inequities such as underdiagnosis, delayed recognition of symptoms, and the greater burden of chronicity, as stated in The Spanish White Paper on Health and Gender [37]. However, when pain is present, sex differences observed in the no-pain group disappear, possibly because the experience of chronic or continuous pain becomes a levelling factor that leads both men and women to seek medical attention according to need rather than sex condition. The Spanish National Health Survey shows a high proportion of both men and women with CP who use the healthcare system, but provides no evidence of substantial differences between them. Similarly, Dueñas et al. [38] found no significant differences by sex in the number of healthcare visits during the previous year due to CP, nor major differences in the proportions of visits to primary care professionals or to specialised care, which reinforces our results.

Other factors, such as quality of life, remain consistently associated across all groups. This aligns with other studies comparing groups with and without CP, where healthcare consultations were primarily tied to mental health issues and other existing health problems (comorbidities) [39]. In this vein, we found that having a worse mental and physical health, and being on sick leave were related to healthcare visits in both groups. This is not surprising, since individuals on sick leaves receiving temporary disability benefits must adhere to a frequent and rigorous schedule of medical consultations. Spanish laws force ongoing medical monitoring to assess its evolution, probable duration, and eventual resolution/recovery. Failure to attend these mandatory reviews may result in the suspension or loss of the associated financial/salary benefit [40].

We also observed that higher levels of social support were positively associated with healthcare visits, but only in the group without pain, while other studies find the opposite relationship [41]. A possible explanation is that individuals with stronger social networks may be more encouraged to seek medical attention, or may find it easier to access healthcare through practical support from relatives or peers [42]. It is also possible that the protective role of social support becomes more evident in populations with chronic conditions, whereas in healthier individuals, it may have a different effect. This unexpected result highlights the complexity of the relationship between social support and healthcare utilization, suggesting the need for further research.

Limitations

Some limitations need to be stated. First, we have a design limitation due to the cross-sectional nature of the study. In addition, the panel recruitment method might lead to a potential selection bias. Nonetheless, the demoscopic company responsible for participant recruitment has extensive experience in this type of study and implements procedures designed to minimise this risk. Furthermore, the distribution of the sample with respect to key variables such as sex and age does not suggest the presence of substantial sampling bias. However, given the nature of the recruitment process, this potential source of bias should not be overlooked.

Another limitation is derived from the use of self-reported data. This is a usual limitation in health studies that has been analysed previously [43–45]. Self-reported health information has shown moderate to high reliability for many chronic conditions when compared with medical records or clinical assessments, and it has been widely used in epidemiological research. Specifically, a revision addressing accuracy of self-reported utilisation of health care services [46] concludes that the accuracy depends, among other factors, on recall time frame. In our study, the use of a 4-week recall period is likely to have minimised this risk. Also, participants were asked to report healthcare visits for any reason rather than visits related to a specific condition. This is likely to facilitate recall and may contribute to the reliability of the information provided.

A potential methodological limitation is the use of a stepwise selection method based on the Wald tests. As stated in the methods section, it might imply risk of overfitting, unstable coefficient estimates or inflated Type I error. We tried to minimise this risk presenting adjusted estimates considering all covariates regardless of statistical significance, for direct comparison. Nonetheless, a theory-driven covariate strategy might be preferable in a future analysis regarding this issue.

Particular consideration should be given to the potential limitation arising from the definition of non-chronic continuous pain. While the definition of chronic pain is clearly stated, the definition of non-chronic continuous pain is not standard and only partially aligned with subacute pain definitions. This might lead to potential misclassification and limited external validity. In spite of this, we decided to use this definition as we were particularly interested in non-chronic pain that manifests at least 4–5 days per week. This led us to the definition of non-chronic continuous pain, which is coherent with previous studies carried out in Spain. As strengths, we highlight the large sample size and the use of validated scales, thereby making the information obtained more valid and reliable, despite being self-reported.

Conclusion

To sum up, this study shows that both chronic pain and non-chronic continuous pain populations exhibit a greater frequency of healthcare resources utilization than individuals without pain, although not necessarily due to pain itself. The factors associated with healthcare visits due to any reason varied across groups, and include sex, age, quality of life, and social support. Notably, within the group without pain, being female was associated with higher healthcare utilization, which highlights the need to monitor potential clinical gender bias, such as underdiagnosis or the delayed recognition of underlying conditions. In addition, findings reveal that care-seeking in younger people could be linked to a high prevalence of mental health comorbidities and sick leave emerged as a factor related in all groups, with the strongest association among people with chronic pain. These results have direct implications for primary care planning and strategic resource allocation. The Spanish healthcare system must integrate bio-psychosocial interventions within youth and occupational care, while prioritizing the mental health sphere and specialized support within primary care settings. Moreover, optimizing resource utilization requires decoupling routine clinical care from mandatory administrative/bureaucratic medical evaluations for temporary disability, thereby alleviating primary care overloading and ensuring that clinical consultations are allocated based on therapeutic need rather than legal requirements.

Abbreviations

BP

Bodily pain

CATI

Computer-assisted telephone interviewing

CAWI

Computer-assisted web interviewing

CP

Chronic pain

ESOMAR

European Society for Opinion and Marketing Research

GH

General health perception

HRQL

Health-related quality of life

IASP

International Association for the Study of Pain

ICD-11

International Classification of Diseases, 11th Revision

ISO

International Organization for Standardization

MCS

Mental component summary

MH

Mental health

NCCP

Non-chronic continuous pain

OR

Odds ratio

PCS

Physical component summary

PF

Physical functioning

RE

Role-emotional

RP

Role-physical

SF

Social functioning

SF-12v.2

12-Item Short Form Survey, Version 2

V

Vitality

Author contributions

IF and AS conceived the idea of the article. The fieldwork was performed by AS, HdS, MD, SR, AE and IF. MD and AS performed the statistical analysis. AS and HdS were mainly responsible for writing the manuscript. IF, MD, AE and SR were involved in revising the content. All the authors approved the final version of the manuscript.

Funding

Funding for open access publishing: Universidad de Cádiz/CBUA. This work was supported by a collaboration between the Foundation for the management of Biomedical Research of Cádiz (FCÁDIZ), and the Grünenthal Foundation. The Grünenthal Foundation is a private non-profitable organisation that promotes the dissemination of scientific knowledge and supports research.

Data availability

The data supporting the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions, as they contain information that could compromise the privacy of research participants.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and complies with the Spanish Biomedical Research Act (Law 14/2007) [47] and Organic Law 3/2018 on Data Protection (LOPDGDD) [48]. In accordance with the 17th Additional Provision of Law 3/2018 [49], the use of pseudonymized data for health research is permitted without additional explicit consent when a strict functional separation between the data controllers and the research team is guaranteed.

Consequently, formal approval by a Research Ethics Committee (REC) was deemed unnecessary as the project falls outside the scope of Royal Decree 1090/2015 [50]. The study is strictly observational, involves no clinical intervention, and the research team only had access to a dataset that was effectively anonymized at the source by the data collection agency.

All participants provided explicit informed consent at the beginning of the questionnaire (via CATI or CAWI platforms). While the data collection agency managed personal identifiers for recruitment purposes, these were strictly decoupled from the responses. The final dataset delivered to the researchers was stripped of all personal identifiers, ensuring the confidentiality and privacy of participants in full compliance with current ethical and legal standards.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.