The WP2019 is a more stringent definition of chronic widespread pain than ACR1990 and is associated with an excess risk for premature death.
Keywords: Chronic widespread pain, Mortality, Cardiovascular disease
Abstract
Introduction:
Chronic widespread pain (CWP) has been suggested as a risk factor for mortality in cardiovascular diseases and malignancies. Different definition of CWP makes it difficult to compare previous studies.
Objectives:
The aim was to study whether mortality and certain causes of death were increased among people with CWP and whether the definition of CWP influenced outcome.
Methods:
This 25-year follow-up study included 2425 people from the general population, at baseline divided into 3 pain groups: CWP, chronic regional pain, and no chronic pain (NCP). Chronic widespread pain was defined according to the ACR1990 (CWPACR1990) and the more stringent WP2019 (CWPWP2019) criteria. Causes of death were derived from official national register. Mortality, adjusted for age, sex, socioeconomic status, and smoking habits were analyzed with Cox regression.
Results:
Overall mortality was not higher in people with CWPACR1990 (hazard ratio [HR] 1.08, P = 0.484) compared with NCP but significantly higher when using CWPWP2019 (HR 1.32, P = 0.033). People with CWPWP2019 had a higher mortality in diseases of the circulatory system (HR 1.32, P = 0.033) but not for neoplastic diseases. CWPACR1990 showed an increased mortality in malignancies of digestive organs. An increased mortality in influenza, pneumonia, acute kidney failure, and chronic kidney disease was observed for the CWPWP2019 definition.
Conclusion:
The more stringent WP2019 definition of CWP showed an excess risk for death, especially within diseases of the circulatory system. The results suggest that WP2019 defines a more vulnerable group in the population. Chronic widespread pain should be acknowledged in the clinic as a risk factor for increased mortality.
1. Introduction
Chronic pain is a common condition in the general population and is often associated with other diseases but could also be a disorder of its own with a multifactorial background. This is especially the case when chronic pain is widespread in the body. This view on chronic widespread pain has also been implemented in the latest 11th revision of the International Classification of Diseases (ICD-11).
Chronic widespread pain (CWP) has previously and most commonly been defined as included in the ACR1990 criteria for fibromyalgia. This definition requires pain for at least 3 months in the left side and right side of the body, above and below the waist, and in the axial skeleton. In addition, for fibromyalgia, pain must be present in at least 11 of 18 specified tender points on digital palpation.22 The ACR1990 definition of CWP has been criticized for being too including, and updated versions of the CWP definition have been proposed.18–20 The most recent of these definitions, the WP2019,18 requires that pain must be present for at least 3 months in 4 of 5 main pain regions: the upper and lower quadrants of the body and the spine. The 5 main pain regions is further divided into 3 pain sites, and pain must be present in at least 7 of these 15 pain sites. The requirement of 4 of 5 regions is also consistent with the definition of CWP in ICD-11.
The prevalence of CWP in the general population based on the ACR1990 criteria vary in the literature, with most measures in the range of 10% to 15%.11 A prevalence of 11.4% has previously been reported from the EPIPAIN cohort,4 the base for the study in this article. Pain for at least 3 months not fulfilling CWP criteria is usually referred to as chronic regional pain (CRP), with a prevalence of 23.9% in the EPIPAIN cohort.4 In terms of the WP2019 definition, the prevalence of CWP and CRP in the population is yet to be investigated.
Chronic widespread pain defined according to ACR1990 is more common in the female population4,11,16 and amongst elderly adults.4,11 Also, socioeconomic status4,11 and sleep disorders1,6 are associated to CWP. Factors associated to CWP defined according to WP2019 remain to be studied.
Previous studies have suggested that people with CWP have an excess mortality,5,9,10 because of not only cardiovascular diseases9,13,14,17 and malignant tumors9,12,13 but also pulmonary diseases,7,9 and chronic kidney disease.23 There are also diverging results, showing no excess mortality.2,3 Definitions of CWP, and if the results are controlled for different confounders, varies between studies, making direct comparisons difficult. It is also yet to be investigated how the more stringent WP2019 definition compares with the ACR1990 definition regarding excess mortality and certain causes of death.
The aim of this study was to determine whether overall mortality and known common causes of death (neoplasms and diseases of the circulatory system) were increased among people with CRP and CWP in a 25-year follow-up of a cohort of general population. In addition, to find out whether the definition of CWP, according to ACR1990 or WP2019, influenced these outcomes.
2. Materials and methods
2.1. Study population and study design
This longitudinal cohort study is based on baseline data from the EPIPAIN postal survey in 1995 and a 25-year follow-up of mortality data from the Swedish national registry. The EPIPAIN cohort is based on a representative sample of 3928 persons from the general population in southwest of Sweden. Based on the digital population register, a questionnaire was mailed to every 18th male and female aged 20 to 74 years in 2 municipalities (population 70 704). The 2425 persons who answered the postal survey in 1995 constitute the study cohort. Data from all participants was sent to the Swedish National Board of Health and Welfare and linked to the National Cause of Death Register.
2.2. Covariates
The EPIPAIN questionnaire had a focus on chronic pain with a pain mannequin for registration of pain experience in 18 predefined areas,4 but it also included questions regarding sociodemographic factors and lifestyle habits. Chronic pain (duration for at least 3 months during the last year) at baseline in 1995 was the main explanatory variable in this study. Based on the pain mannequin, pain was classified as CWP, CRP, or no experience of chronic pain (NCP). Both the ACR1990 and the WP2019 classifications were used, giving 2 sets of 3 pain groups: (1) CWPACR1990, CRPACR1990, and NCPACR1990 and (2) CWPWP2019, CRPWP2019, and NCPWP2019. There was no attempt to classify the CWP subgroup fibromyalgia because this would have required tender point count (ACR1990)22 or a more extensive questionnaire.18
Age, sex, smoking habits, and socioeconomic status were included as cofactors or confounders. Smoking habits at inclusion were divided into 3 groups: never smoking, previous smoking, and smoking.
Based on occupation at inclusion, socioeconomic status was classified according to the Swedish socioeconomic classification system15 and divided into 4 groups: intermediate/higher nonmanual employees, assistant nonmanual employees, manual workers, and other. The group other included self-employed, farmers, housewives, and students.
The follow-up time was from baseline, June 1, 1995, to end of the study (May 31, 2020) or until the participant died.
2.3. Main outcome variable
Main outcome variable was cause of death, derived from the National Cause of Death Register at the Swedish National Board of Health and Welfare, classified according to International Classification of Diseases (ICD-9 and ICD-10). This register gives a full coverage of all deaths in Sweden, including information on cause and date of death. A translation table from the National Board of Health and Welfare was used to convert the ICD-9 codes into ICD-10 codes. Eighteen of the deaths were not coded in the register or had incorrect codes and were set as “unknown.”
Causes of death were divided into primary and contributing causes of death. Causes of death were then divided into chapters and blocks according to ICD-10. Results are presented separately for primary causes of death and for total causes of death (including both primary and contributing causes). The focus in this study was on neoplasms and diseases of the circulatory system, but all causes of death in the study population were included in the analyses.
2.4. Statistical methods and analyses
Chi-square tests were used to test for differences in proportions between the groups. Cox regression analyses were used to analyze the primary outcomes, such as overall deaths and different causes of deaths, and their associations to pain group belonging (defined by ACR1990 or WP2019) controlled for age, sex, smoking habits, and socioeconomic status. Due to the small number of primary causes of death in some of the ICD-10 chapters and most of the blocks (subgroups within chapters), and because a contributing cause could be important, primary and contributing cause of death were combined in a secondary set of analyses. A threshold value of more than 50 study participants in an outcome was set to allow for 10 participants for each cofactor in the regression analyses. The time of survival was measured from the start of the study to the study participant deceased or until the study ended, giving a span from 0 to 25 years. The associations to the studied possible predictors are presented as hazard ratio (HR). IBM SPSS statistics version 26 was used to analyze the data. All tests were 2 tailed and conducted at the 0.05 significance level.
2.5. Ethics
This study was approved by the Regional Ethics Review Board Lund, Sweden (Dnr 2016/132). This provided the possibility to collect the information of deceased and the causes of death of the study participants from the National Board of Health and Welfare. All the data of the study participants were encrypted so that the data could not be traced back to the individual. The participants had at baseline given their written informed consent to participate in the study.
3. Results
3.1. Descriptive data
At baseline there were a total of 2425 participants, 1132 (46.7%) male and 1293 (53.3%) female, with a mean (SD) age of 46.5 (15.4) years. The prevalence of CWP was 12.5% (n = 303) when defined according to the ACR1990 and 8.2% (n = 200) when defined according to WP2019. Corresponding prevalence of CRP was 24.2% (n = 599) and 28.3% (n = 686), respectively. The distribution of studied cofactors in relation to pain groups defined according to ACR1990 and WP2019 is presented in Table 1.
Table 1.
Demographic factors at baseline 1995 in relation to pain groups defined according to ACR1990 or WP2019.
| NCP | CRP | CWP | Total | |
|---|---|---|---|---|
| ACR1990 | % (n) | % (n) | % (n) | % (n) |
| Sex | ||||
| Male | 49.7 (729) | 46.3 (272) | 31.4 (95) | 46.5 (1096) |
| Female | 50.3 (737) | 53.7 (316) | 68.6 (208) | 53.5 (1261) |
| 100 (1466) | 100 (588) | 100 (303) | 100 (2357) | |
| Age | ||||
| 20–33 | 30.7 (450) | 19.9 (117) | 9.9 (30) | 25.3 (597) |
| 34–46 | 24.9 (365) | 23.8 (140) | 19.5 (59) | 23.9 (564) |
| 47–58 | 22.8 (334) | 27.6 (162) | 32.7 (99) | 25.2 (595) |
| 59–74 | 21.6 (317) | 28.7 (169) | 38.0 (115) | 25.5 (601) |
| 100 (1466) | 100 (588) | 100 (303) | 100 (2357) | |
| Socioeconomic status | ||||
| A | 28.2 (413) | 21.9 (129) | 11.2 (34) | 24.4 (576) |
| B | 13.8 (202) | 13.3 (78) | 15.2 (46) | 13.8 (326) |
| C | 43.2 (634) | 52.9 (311) | 61.4 (186) | 48.0 (1131) |
| D | 14.8 (217) | 11.9 (70) | 12.2 (37) | 13.7 (324) |
| 100 (1466) | 100 (588) | 100 (303) | 100 (2357) | |
| Smoking habits | ||||
| Never | 54.8 (797) | 45.3 (264) | 48.2 (145) | 51.6 (1206) |
| Previous | 24.3 (353) | 30.2 (176) | 27.9 (84) | 26.2 (613) |
| Smoker | 21.0 (305) | 24.5 (143) | 23.9 (72) | 22.2 (520) |
| 100 (1455) | 100 (583) | 100 (301) | 100 (2339) | |
| WP2019 | ||||
| Sex | ||||
| Male | 49.7 (729) | 45.5 (312) | 26.0 (52) | 46.5 (1093) |
| Female | 50.3 (737) | 54.5 (374) | 74.0 (148) | 53.5 (1259) |
| 100 (1466) | 100 (686) | 100 (200) | 100 (2352) | |
| Age | ||||
| 20–33 | 30.7 (450) | 18.7 (128) | 9.0 (18) | 25.3 (596) |
| 34–46 | 24.9 (365) | 23.6 (162) | 18.5 (37) | 24.0 (564) |
| 47–58 | 22.8 (334) | 28.7 (197) | 31.5 (63) | 25.3 (594) |
| 59–74 | 21.6 (317) | 29.0 (199) | 41.0 (82) | 25.4 (598) |
| 100 (1466) | 100 (686) | 100 (200) | 100 (2352) | |
| Socioeconomic status | ||||
| A | 28.2 (413) | 21.1 (145) | 8.5 (17) | 24.4 (575) |
| B | 13.8 (202) | 13.4 (92) | 15.5 (31) | 13.8 (325) |
| C | 43.2 (634) | 53.6 (368) | 63.5 (127) | 48.0 (1129) |
| D | 14.8 (217) | 11.8 (81) | 12.5 (25) | 13.7 (323) |
| 100 (1466) | 100 (686) | 100 (200) | 100 (2352) | |
| Smoking habits | ||||
| Never | 54.8 (797) | 46.2 (314) | 45.5 (91) | 51.5 (1202) |
| Previous | 24.3 (353) | 29.5 (200) | 29.5 (59) | 26.2 (612) |
| Smoker | 21.0 (305) | 24.3 (165) | 25.0 (50) | 22.3 (520) |
| 100 (1455) | 100 (679) | 100 (200) | 100 (2334) |
A = intermediate/higher nonmanual employees, B = assistant nonmanual employees, C = manual workers, D = other (self-employed, farmers, housewives, and students).
CRP, chronic regional pain; CWP, chronic widespread pain; NCP, no chronic pain.
3.2. Deaths
A total of 608 (25.1%) out of 2425 study participants had died at the end of the study, 337 of the males and 271 of the females (29.8% vs 21.0%; P < 0.001), with a mean age of 77.6 years (males 75.9 vs females 79.7; P < 0.001).
Within the pain groups according to the ACR1990 definition, the unadjusted mortality was 21.4% in NCPACR1990, 28.8% in CRPACR1990, and 35.5% in CWPACR1990 (P < 0.001). With pain groups defined according to WP2019, the unadjusted mortality was 21.4% in NCPWP2019, 27.8% in CRPWP2019, and 39.5% in CWPWP2019 (P < 0.001). Cox regression analysis with death as outcome in relation to pain group according to the ACR1990 criteria showed no significant associations for CRPACR1990 (HR 1.05, P = 0.600) or CWPACR1990 (HR 1.08, P = 0.484) in relation to NCPACR1990, adjusted for age, sex, socioeconomic status, and smoking habits (Table 2 and Fig. 1). When pain groups were classified according to WP2019, there was a significant association for CWPWP2019 (HR 1.32, P = 0.033) but not for CRPWP2019 (HR 0.98, P = 0.860), adjusted for age, sex, socioeconomic status, and smoking habits (Table 2). Figure 1 displays the result visually as Kaplan–Meier curves for the different pain groups.
Table 2.
Cox regression analysis with hazard ratio for death during the 25-year follow-up time in relation to the 2 studied definitions of chronic widespread pain/pain groups.
| Pain group ACR1990 | Pain group WP2019 | |||||
|---|---|---|---|---|---|---|
| Deceased Yes = 577, No = 1762 Missing data* = 86 |
Deceased Yes = 574, No = 1760 Missing data* = 91 |
|||||
| Data from inclusion | n | HR (95% CI) | P | n | HR (95% CI) | P |
| Sex | ||||||
| Male | 1087 | 1 | 1084 | 1 | ||
| Female | 1252 | 0.58 (0.49–0.69) | 0.000 | 1250 | 0.57 (0.48–0.68) | 0.000 |
| Age (y) | 2339 | 1.13 (1.12–1.14) | 0.000 | 2334 | 1.13 (1.12–1.14) | 0.000 |
| Pain group | ||||||
| NCP | 1455 | 1 | 1455 | 1 | ||
| CRP | 583 | 1.05 (0.87–1.28) | 0.600 | 679 | 0.98 (0.82–1.18) | 0.860 |
| CWP | 301 | 1.08 (0.87–1.36) | 0.484 | 200 | 1.32 (1.02–1.70) | 0.033 |
| Socioeconomic status | ||||||
| A | 575 | 1 | 574 | 1 | ||
| B | 326 | 1.17 (0.88–1.56) | 0.276 | 325 | 1.16 (0.87–1.55) | 0.302 |
| C | 1121 | 1.23 (0.99–1.54) | 0.066 | 1119 | 1.22 (0.98–1.53) | 0.078 |
| D | 317 | 1.29 (0.97–1.72) | 0.083 | 316 | 1.27 (0.95–1.70) | 0.103 |
| Smoking habits | ||||||
| Never | 1206 | 1 | 1202 | 1 | ||
| Previous | 613 | 1.13 (0.92–1.38) | 0.248 | 612 | 1.13 (0.92–1.38) | 0.253 |
| Smoker | 520 | 2.10 (1.70–2.58) | 0.000 | 520 | 2.10 (1.70–2.59) | 0.000 |
Cases with missing data, thus excluded from the analysis.
A = intermediate/higher nonmanual employees, B = assistant nonmanual employees, C = manual workers, D = other (self-employed, farmers, housewives, and students).
95% CI, 95% confidence interval; CRP, chronic regional pain; CWP, chronic widespread pain; HR, hazard ratio; NCP, no chronic pain.
Figure 1.
Cumulative survival reported as a percentage of the total number of study participants in the analysis, 2339 for ACR1990, and 2334 for WP2019. Time in years is the follow-up time. Adjusted for age, sex, smoking habits, and socioeconomic status.
3.3. Primary and total cause of death
Cause of death for all 608 deceased divided into chapters according to the ICD-10 codes is presented in Table 3. The most common causes were diseases of the circulatory system, neoplasms, and diseases of the respiratory system. Also, regarding total causes of death, including both primary and contributing causes of death, the major causes of death were within diseases of the circulatory system, neoplasms, and diseases of the respiratory system (Table 3).
Table 3.
All causes of death according to international classification of diseases-10 A00-Z99 codes.
| Causes of death, chapters according to ICD-10 codes A00-Z99 | Primary cause of death | Total cause of death | ||
|---|---|---|---|---|
| n | % | N | % | |
| A00-B99 certain infectious and parasitic diseases | 12 | 0.5 | 44 | 1.8 |
| C00-D48 neoplasms | 204 | 8.4 | 231 | 9.5 |
| D50-D89 diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism | 2 | 0.1 | 13 | 0.5 |
| E00-E90 endocrine, nutritional and metabolic diseases | 16 | 0.7 | 71 | 2.9 |
| F00-F99 mental and behavioural disorders | 25 | 1.0 | 87 | 3.6 |
| G00-G99 diseases of the nervous system | 21 | 0.9 | 54 | 2.2 |
| H00-H59 diseases of the eye and adnexa | — | — | 2 | 0.1 |
| I00-I99 diseases of the circulatory system | 207 | 8.5 | 349 | 14.4 |
| J00-J99 diseases of the respiratory system | 38 | 1.6 | 136 | 5.6 |
| K00-K93 diseases of the digestive system | 16 | 0.7 | 43 | 1.8 |
| L00-L99 diseases of the skin and subcutaneous tissue | 2 | 0.1 | 12 | 0.5 |
| M00-M99 diseases of the musculoskeletal system and connective tissue | 5 | 0.2 | 20 | 0.8 |
| N00-N99 diseases of the genitourinary system | 8 | 0.3 | 67 | 2.8 |
| Q00-Q99 congenital malformations, deformations,s and chromosomal abnormalities | 2 | 0.1 | 2 | 0.1 |
| R00-R99 symptoms, signs, and abnormal clinical and laboratory findings, not elsewhere classified | 14 | 0.6 | 158 | 6.5 |
| S00-T98 injury, poisoning and certain other consequences of external causes | 3 | 0.1 | 44 | 1.8 |
| V01-Y98 external causes of morbidity and mortality | 17 | 0.7 | 58 | 2.4 |
| Z00-Z99 factors influencing health status and contact with health services | — | — | 21 | 0.9 |
| Unknown. Study participants without ICD-code, or with a nontranslatable ICD-9 code into ICD-10 | 16 | 0.7 | 20 | 0.8 |
One column includes only the primary cause of death, and the other column includes both primary and cause of death as the total cause of death. The total study population was 2425 people.
ICD-10, international classification of diseases-10.
3.4. Primary cause of death in relation to pain group
3.4.1. Neoplasms
Cox regression analysis with all neoplasms (ICD-10 codes C00-D48) as the primary cause of death as outcome controlled for age, sex, smoking habits, and socioeconomic status, did not show any significant associations for the pain groups CRPACR1990 or CWPACR1990 in relation to NCPACR1990. Age, sex, and smoking showed significant associations (Table 4). Also, when the pain groups were classified according to WP2019, no significant associations were seen for CRPWP2019 or CWPWP2019, but age, sex, and smoking habits showed significant associations (Table 4). Figure 2 displays the result visually as Kaplan–Meier curves for the different pain groups.
Table 4.
Cox regression analysis with neoplasms as the primary cause of death according to international classification of diseases-10 codes over a 25-year follow-up time.
| Data from inclusion | Pain group according to ACR1990 | Pain group according to WP2019 | ||||
|---|---|---|---|---|---|---|
| C00-D48 as the primary cause of death Yes = 198, No = 2141 Missing data* = 86 Censured data† = 0 |
C00-D48 as the primary cause of death Yes = 198, No = 2136 Missing data* = 91 Censured data† = 0 |
|||||
| n | HR (95% CI) | P | N | HR (95% CI) | P | |
| Sex | ||||||
| Male | 1087 | 1 | 1084 | 1 | ||
| Female | 1252 | 0.64 (0.48–0.86) | 0.003 | 1250 | 0.64 (0.47–0.86) | 0.003 |
| Age (y) | 2339 | 1.10 (1.09–1.12) | 0.000 | 2334 | 1.10 (1.09–1.12) | 0.000 |
| Pain group | ||||||
| NCP | 1455 | 1 | 1455 | 1 | ||
| CRP | 583 | 0.98 (0.70–1.36) | 0.885 | 679 | 0.99 (0.72–1.35) | 0.928 |
| CWP | 301 | 1.21 (0.83–1.76) | 0.327 | 200 | 1.35 (0.87–2.09) | 0.176 |
| Socioeconomic status | ||||||
| A | 575 | 1 | 574 | 1 | ||
| B | 326 | 0.83 (0.51–1.35) | 0.453 | 325 | 0.83 (0.51–1.34) | 0.438 |
| C | 1121 | 1.02 (0.72–1.44) | 0.927 | 1119 | 1.01 (0.71–1.43) | 0.963 |
| D | 317 | 0.70 (0.41–1.20) | 0.190 | 316 | 0.69 (0.40–1.18) | 0.170 |
| Smoker | ||||||
| Never | 1206 | 1 | 1202 | 1 | ||
| Previous | 613 | 1.26 (0.89–1.79) | 0.188 | 612 | 1.26 (0.89–1.78) | 0.188 |
| Smoker | 520 | 2.21 (1.56–3.14) | 0.000 | 520 | 2.20 (1.55–3.12) | 0.000 |
Cases with missing data, thus excluded from the analysis.
Deceased because of other cause of death before the first participant deceased of the current cause of death, thus excluded from the analysis.
A = intermediate/higher nonmanual employees, B = assistant non-manual employees, C = manual workers, D = other (self-employed, farmers, housewives, and students).
95% CI, 95% confidence interval; CRP, chronic regional pain; CWP, chronic widespread pain; HR, hazard ratio; NCP, no chronic pain.
Figure 2.
Survival function with neoplasms as the primary cause of death. Cumulative survival reported as a percentage of the total number of study participants in the analysis, 2339 for ACR1990, and 2334 for WP2019. Time in years is the follow-up time. Adjusted for age, sex, smoking habits, and socioeconomic status. Pain group belonging at baseline according to ACR1990 in the upper diagram and WP2019 in the lower.
3.4.2. Diseases of the circulatory system
Cox regression analysis with diseases of the circulatory system (ICD-10 codes I00-I99) as the primary cause of death as outcome, adjusted for age, sex, smoking habits, and socioeconomic status did not show any significant associations for the pain groups CRPACR1990 (HR 1.35, P = 0.072) or CWPACR1990 (HR 1.35, P = 0.128) in relation to NCPACR1990. Age, sex, socioeconomic status, and smoking showed significant HR. When the pain groups were classified according to WP2019, there was a significant association for the pain group CWPWP2019 (HR 1.70, P = 0.016) but not for CRPWP2019 (HR 1.21, P = 0.239), adjusted for age, sex, socioeconomic status, and smoking (Table 5). Figure 3 displays the results visually as Kaplan–Meier curves for the different pain groups.
Table 5.
Cox regression analysis with diseases of the circulatory system as the primary cause of death according to international classification of diseases-10 codes over a 25-year follow-up time.
| Data from inclusion | Pain group according to ACR1990 | Pain group according to WP2019 | ||||
|---|---|---|---|---|---|---|
| I00-I99 as the primary cause of death Yes = 191, No = 2146 Missing data* = 86 Censured data† = 2 |
I00-I99 as the primary cause of death Yes = 188, No = 2144 Missing data* = 91 Censured data† = 2 |
|||||
| n | HR (95% CI) | P | n | HR (95% CI) | P | |
| Sex | ||||||
| Male | 1087 | 1 | 1084 | 1 | ||
| Female | 1252 | 0.48 (0.35–0.65) | 0.000 | 1250 | 0.46 (0.33–0.63) | 0.000 |
| Age (y) | 2337 | 1.17 (1.15–1.19) | 0.000 | 2332 | 1.17 (1.15–1.19) | 0.000 |
| Pain group | ||||||
| NCP | 1455 | 1 | 1455 | 1 | ||
| CRP | 583 | 1.35 (0.97–1.88) | 0.072 | 679 | 1.21 (0.88–1.67) | 0.239 |
| CWP | 301 | 1.35 (0.92–1.97) | 0.128 | 200 | 1.70 (1.10–2.61) | 0.016 |
| Socioeconomic status | ||||||
| A | 575 | 1 | 574 | 1 | ||
| B | 326 | 1.09 (0.62–1.90) | 0.770 | 325 | 1.10 (0.63–1.93) | 0.743 |
| C | 1121 | 1.43 (0.94–2.16) | 0.094 | 1119 | 1.43 (0.94–2.19) | 0.096 |
| D | 317 | 1.98 (1.22–3.23) | 0.006 | 316 | 2.00 (1.22–3.28) | 0.006 |
| Smoker | ||||||
| Never | 1206 | 1 | 1202 | 1 | ||
| Previous | 613 | 1.24 (0.88–1.75) | 0.227 | 612 | 1.25 (0.88–1.76) | 0.216 |
| Smoker | 520 | 1.96 (1.33–2.89) | 0.001 | 520 | 1.99 (1.35–2.94) | 0.001 |
Cases with missing data, thus excluded from the analysis.
Deceased because of other cause of death before the first participant deceased of the current cause of death, thus excluded from the analysis.
A = intermediate/higher non-manual employees, B = assistant nonmanual employees, C = manual workers, D = other (self-employed, farmers, housewives, and students).
95% CI, 95% confidence interval; CRP, chronic regional pain; CWP, chronic widespread pain; HR, hazard ratio; NCP, no chronic pain.
Figure 3.
Survival function with diseases of the circulatory system as the primary cause of death. Cumulative survival reported as a percentage of the total number of study participants in the analysis, 2337 for ACR1990, and 2332 for WP2019. Time in years is the follow-up time. Adjusted for age, sex, smoking habits, and socioeconomic status. Pain group belonging at baseline according to ACR1990 in the upper diagram and WP2019 in the lower.
3.5. Total cause of death
Due to the small number of primary causes of death in some of the ICD-10 chapters and most of the blocks (subgroups within chapters), and because a contributing cause could be important, primary and contributing cause of death were combined into “total cause of death” in a secondary set of analyses. All analyses were performed in the same manner as above and presented in Table 6. CWPACR1990 was significantly associated to “malignant neoplasms of digestive organs” (HR 1.93, P = 0.034), “other forms of heart disease” (HR 1.47, P = 0.040), and “diseases of arteries, arterioles and capillaries” (HR 2.19, P = 0.028). CWPWP2019 was significantly associated to “diseases of the circulatory system” (HR 1.50, P = 0.017), “diseases of the genitourinary system” (HR 2.28, P = 0.024), “other forms of heart disease” (HR 2.21, P < 0.001), “diseases of arteries, arterioles, and capillaries” (HR 2.62, P = 0.016), “influenza and pneumonia” (HR 2.12, P = 0.037), and “acute kidney failure and chronic kidney disease” (HR 2.16, P = 0.044).
Table 6.
Hazard ratio (95% confidence interval) and P value for the pain groups in relation to total cause of death as international classification of disease-codes adjusted for age, sex, smoking habits and socioeconomic status.
| CRPACR1990 | CWPACR1990 | CRPWP2019 | CWPWP2019 | |
|---|---|---|---|---|
| Neoplasm C00-D48 | HR 1.02 P = 0.913 |
HR 1.14 P = 0.472 |
HR 0.99 P = 0.970 |
HR 1.28 P = 0.249 |
| Diseases of the circulatory system I00-I99 | HR 1.17 P = 0.215 |
HR 1.15 P = 0.370 |
HR 1.05 P = 0.696 |
HR 1.50 P = 0.017 |
| Diseases of the respiratory system J00-J99 | HR 0.90 P = 0.637 |
HR 1.05 P = 0.830 |
HR 0.81 P = 0.316 |
HR 1.52 P = 0.116 |
| Diseases of the genitourinary system N00-N99 | HR 1.22 P = 0.494 |
HR 1.57 P = 0.173 |
HR 1.10 P = 0.751 |
HR 2.28 P = 0.024 |
| Malignant neoplasms of digestive organs C15-C26 | HR 1.01 P = 0.971 |
HR 1.93 P = 0.034 |
HR 1.21 P = 0.501 |
HR 1,88 P = 0.094 |
| Other forms of heart disease I30-I52 | HR 1.35 P = 0.067 |
HR 1.47 P = 0.040 |
HR 1.18 P = 0.297 |
HR 2.21 P = 0.000 |
| Diseases of arteries, arterioles and capillaries I70-I99 | HR 1.71 P = 0.106 |
HR 2.19 P = 0.028 |
HR 1.54 P = 0.181 |
HR 2.62 P = 0.016 |
| Influenza and pneumonia J09-J18 | HR 0.63 P = 0.239 |
HR 1.47 P = 0.248 |
HR 0.62 P = 0.190 |
HR 2.12 P = 0.037 |
| Acute kidney failure and chronic kidney disease N17-N19 | HR 0.93 P = 0.834 |
HR 1.59 P = 0.181 |
HR 0.88 P = 0.689 |
HR 2.16 P = 0.044 |
Minimum n for cause of death >50 people. Only pain group and total cause of death are displayed in the table as HR and P-value.
CRP, chronic regional pain; CWP, chronic widespread pain; HR, hazard ratio.
4. Discussion
4.1. Main findings
The main finding in this 25-year follow-up of a general population cohort was that CWP, as defined by the more stringent WP2019, was a risk factor for excess mortality, adjusted for age, sex, smoking habits, and socioeconomic status. An increased risk was not seen when having CWP was defined using the less stringent ACR1990 definition. The most common causes of death were similar to those in the general population, being diseases of the circulatory system and neoplasms. When subgrouping causes of deaths according to ICD-10 chapters, the higher risk for death in those with CWP as defined by WP2019 was seen for diseases of the circulatory system but not for neoplasms overall.
4.2. Overall mortality risk
No increased overall mortality was shown for the pain groups CRPACR1990 and CWPACR1990 when compared with the NCP ACR1990 group. This is in line with a previous systematic review2 and a study on the mortality risk in fibromyalgia according to the ACR1990 definition.21 On the other hand, the result stands in contrast to previous studies which have reported an association between CWP and excess mortality.5,9,10,17 When CWP, in this study, was defined according to the more stringent WP2019 definition, participants with CWPWP2019 showed an excess mortality, being more in line with these previous studies reporting CWP as a risk factor. The result in this study could be because the definition of CWP according to WP2019 defines a more vulnerable population than the ACR1990 definition. This stronger association with mortality is seen throughout this study, controlled for sex, age, and smoking.
4.3. Risk of death from different causes
As expected, the most common cause of death was diseases of the circulatory system. This matches the statistics of most common cause of death according to the World Health Organization 2019, Eurostat 2016, and the Swedish National Board of Health and Welfare 2019. Furthermore, neoplasms and diseases of the respiratory system were amongst the most common causes of death. This also indicates that the study population is representative of a normal population in Sweden regarding causes of death.
There was no statistically significant excess mortality in neoplasms overall as a primary or total (including neoplasm as contributing cause) cause of death for participants with CWP regardless of the definition (ACR1990 or WP2019). However, the WP2019 definition gave a higher but nonsignificant HR than the ACR1990 definition. An excess mortality because of neoplasms in people with CWP has been reported in some previous studies,9,10,12,13 but there are also reports where no such excess risk has been found.2,3 A potential risk for excess mortality in neoplasms with a more stringent CWP definition could be of interest to evaluate in further larger studies.
Chronic widespread pain has previously been suggested to be associated to excess mortality in diseases in the circulatory system.10,13,14,17 In this study, there was a significant association between CWPWP2019 and excess death from a primary cause within the circulatory system. The associations were also nearly significant for CRPACR1990 and CWPACR1990 and became significant for those 2 groups when analyzed together. The finding that there is an association of excess mortality even with less widespread pain is graphically seen in Figure 3 where the Kaplan–Meier curves overlap for CRP and CWP when defined according to ACR1990 but get separated with the more stringent WP2019 definition.
Previously, CWP correlated with an increased risk of hospitalization because of infectious diseases where ICD-10 J00-J22 are included.8 Thus, the link between CWP and infectious diseases could be a possible cause of excess mortality in diseases of the respiratory system. In this study, this is illustrated by a significant two-fold higher HR for CWPWP2019 in relation to total causes of deaths due to influenza and pneumonia.
“Diseases of the genitourinary system” and its subgroup “acute kidney failure and chronic kidney disease” was found significant in the CWPWP2019 group. These findings are new and need to be further investigated, but chronic kidney disease has previously been reported to be associated to CWP.23
4.4. The definition of chronic widespread pain
This study shows that using different definitions of CWP may have an in influence on long-term follow-up results on mortality and causes of death.
The authors behind the ACR1990 and WP2019 definitions have stated that WP2019 is more specific for identifying individuals with authentic CWP. They argue that ARC1990 should be replaced with WP2019 for classification of CWP.18 There is no universal definition of CWP used in previous studies, making it difficult to compare the results, but the new ICD-11 definition is in line with the WP2019 definition. Based on the results of this study, the WP2019 definition seems to define a CWP group that has an increased correlation to excessive mortality.
4.5. Strengths and weaknesses
The major strength of the study is the well-established cohort that have been followed for 25 years with mortality data from the official Swedish death register, including all deaths in Sweden. Another strength is the possibility of applying the different definitions of CWP in this cohort because of the detailed pain report at baseline. It is a weakness that, although following a relatively large population sample, some causes of deaths are too rare to study without the risk of having too low power to rule out an increased risk. It is also a weakness that the study does not take pain developed over time into account, although still interestingly shows that a baseline report of CWP have an impact on mortality over 25 years. The CWP concept, as applied in this study, does not rule out remaining pain from an inflammatory rheumatic disease. The contribution to CWP from such disorders is expected to be small because of a low prevalence in the population in relation to that of CWP. Because of restrictions in data that could be added to the national register, it was not possible to control for all possible confounders. This is also the case for possible comorbidities such as affective disorders.
5. Conclusions
The more stringent WP2019 definition of CWP showed an excess risk for death, especially within diseases of the circulatory system, for those having CWP at baseline in a 25-year follow-up of 2425 people from the general population. The results suggest that WP2019 defines a more vulnerable population. There is a need to evaluate this further in larger studies. Based on our results, we suggest using the WP2019 definition instead of the ACR1990 definition in future studies on CWP and mortality risk. It is also still to be studied how the new definition of CWP in ICD-11 will associate with mortality. The current findings suggest that in the clinic, CWP should be acknowledged as a risk factor for increased mortality besides known risk factors.
Disclosures
The authors have no conflict of interest to declare.
Acknowledgements
The study was financially supported by a nonrestricted grant from the Swedish Rheumatism Association.
Data cannot be made freely available as they are subject to secrecy in accordance with the Swedish Public Access to Information and Secrecy Act. Request to make data available to reproduce the findings in the study should be made to the corresponding author Stefan Bergman. This will then be subject to a review of secrecy.
Footnotes
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
References
- [1].Aili K, Andersson M, Bremander A, Haglund E, Larsson I, Bergman S. Sleep problems and fatigue as predictors for the onset of chronic widespread pain over a 5- and 18-year perspective. BMC Musculoskelet Disord 2018;19:390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Asberg AN, Heuch I, Hagen K. The mortality associated with chronic widespread musculoskeletal complaints: a systematic review of the literature. Musculoskeletal Care 2017;15:104–13. [DOI] [PubMed] [Google Scholar]
- [3].Asberg AN, Stovner LJ, Zwart JA, Winsvold BS, Heuch I, Hagen K. Chronic musculoskeletal complaints as a predictor of mortality-The HUNT study. PAIN 2016;157:1443–7. [DOI] [PubMed] [Google Scholar]
- [4].Bergman S, Herrstrom P, Hogstrom K, Petersson IF, Svensson B, Jacobsson LT. Chronic musculoskeletal pain, prevalence rates, and sociodemographic associations in a Swedish population study. J Rheumatol 2001;28:1369–77. [PubMed] [Google Scholar]
- [5].Da Silva JAP, Geenen R, Jacobs JWG. Chronic widespread pain and increased mortality: biopsychosocial interconnections. Ann Rheum Dis 2018;77:790–2. [DOI] [PubMed] [Google Scholar]
- [6].Gupta A, Silman AJ, Ray D, Morriss R, Dickens C, MacFarlane GJ, Chiu YH, Nicholl B, McBeth J. The role of psychosocial factors in predicting the onset of chronic widespread pain: results from a prospective population-based study. Rheumatology (Oxford) 2007;46:666–71. [DOI] [PubMed] [Google Scholar]
- [7].Holmberg T, Davidsen M, Thygesen LC, Krøll MJ, Tolstrup JS. Mortality among persons experiencing musculoskeletal pain: a prospective study among Danish men and women. BMC Musculoskelet Disord 2020;21:666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Lindgren H, Bergman S. Chronic musculoskeletal pain predicted hospitalisation due to serious medical conditions in a 10 year follow up study. BMC Musculoskelet Disord 2010;11:127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Macfarlane GJ, Barnish MS, Jones GT. Persons with chronic widespread pain experience excess mortality: longitudinal results from UK Biobank and meta-analysis. Ann Rheum Dis 2017;76:1815–22. [DOI] [PubMed] [Google Scholar]
- [10].Macfarlane GJ, McBeth J, Silman AJ. Widespread body pain and mortality: prospective population based study. BMJ 2001;323:662–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Mansfield KE, Sim J, Jordan JL, Jordan KP. A systematic review and meta-analysis of the prevalence of chronic widespread pain in the general population. PAIN 2016;157:55–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].McBeth J, Silman AJ, Macfarlane GJ. Association of widespread body pain with an increased risk of cancer and reduced cancer survival: a prospective, population-based study. Arthritis Rheum 2003;48:1686–92. [DOI] [PubMed] [Google Scholar]
- [13].McBeth J, Symmons DP, Silman AJ, Allison T, Webb R, Brammah T, Macfarlane GJ. Musculoskeletal pain is associated with a long-term increased risk of cancer and cardiovascular-related mortality. Rheumatology (Oxford) 2009;48:74–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Ronnegard AS, Nowak C, Ang B, Arnlov J. The association between short-term, chronic localized and chronic widespread pain and risk for cardiovascular disease in the UK Biobank. Eur J Prev Cardiol 2022;29:1994–2002. [DOI] [PubMed] [Google Scholar]
- [15].SCB. Swedish socioeconomic classification (in Swedish). Stockholm, Sweden: Statistics Sweden, 1984. [Google Scholar]
- [16].Tan AC, Jaaniste T, Champion D. Chronic widespread pain and fibromyalgia syndrome: life-course risk markers in young people. Pain Res Manag 2019;2019:6584753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Tesarz J, Eich W, Baumeister D, Kohlmann T, D'Agostino R, Schuster AK. Widespread pain is a risk factor for cardiovascular mortality: results from the Framingham Heart Study. Eur Heart J 2019;40:1609–17. [DOI] [PubMed] [Google Scholar]
- [18].Wolfe F, Butler SH, Fitzcharles M, Hauser W, Katz RL, Mease PJ, Rasker JJ, Russell AS, Russell IJ, Walitt B. Revised chronic widespread pain criteria: development from and integration with fibromyalgia criteria. Scand J Pain 2019;20:77–86. [DOI] [PubMed] [Google Scholar]
- [19].Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Hauser W, Katz RL, Mease PJ, Russell AS, Russell IJ, Walitt B. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthritis Rheum 2016;46:319–29. [DOI] [PubMed] [Google Scholar]
- [20].Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Katz RS, Mease P, Russell AS, Russell IJ, Winfield JB, Yunus MB. The American College of Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res (Hoboken) 2010;62:600–10. [DOI] [PubMed] [Google Scholar]
- [21].Wolfe F, Hassett AL, Walitt B, Michaud K. Mortality in fibromyalgia: a study of 8,186 patients over thirty-five years. Arthritis Care Res (Hoboken) 2011;63:94–101. [DOI] [PubMed] [Google Scholar]
- [22]. Wolfe F Smythe HA Yunus MB Bennett RM Bombardier C Goldenberg DL Tugwell P Campbell SM AbelesM Clark P Fam AG Farber SJ Fiechtner JJ Franklin CM Gatter RA Hamaty D Lessard J Lichtbroun AS Masi AT McCain GA Reynolds WJ Romano TJ Russell IJ Sheon RP.. The American college of rheumatology 1990 criteria for the classification of fibromyalgia. Report of the multicenter criteria committee. Arthritis Rheum 1990;33:160–72. [DOI] [PubMed] [Google Scholar]
- [23].Yang XH, Zhang BL, Gu YH, Zhan XL, Guo LL, Jin HM. Association of sleep disorders, chronic pain, and fatigue with survival in patients with chronic kidney disease: a meta-analysis of clinical trials. Sleep Med 2018;51:59–65. [DOI] [PubMed] [Google Scholar]