Abstract
Objectives. Number of pain sites (NPS) is a potentially important marker of health-related quality of life (HRQoL) but remains unexplored in older people. This cross-sectional study investigated whether, in older people including the oldest old, NPS was independently associated with poorer mental and physical HRQoL and if the association was moderated by age.
Methods. A postal questionnaire sent to a population sample of adults aged ≥50 years in North Staffordshire, UK, included the 12-item Short Form Health Survey (SF-12) mental component summary (MCS) and physical component summary (PCS), a blank body pain manikin, socio-demographic, health behaviour and morbidity questions. Participants shaded sites of pain lasting ≥1 day in the past 4 weeks on the manikin. OA consultation data were obtained for participants consenting to medical records review.
Results. A total of 13 986 individuals (adjusted response 70.6%) completed a questionnaire, of which 12 408 provided complete pain data. The median NPS reported was 4 [interquartile range (IQR) 0–8]. General linear models showed that an increasing NPS was significantly associated with poorer MCS (β = −0.43, 95% CI −0.46, −0.40) and PCS (β = −0.87, 95% CI −0.90, −0.84). Adjustment for covariates attenuated the associations but they remained significant (MCS: β = −0.28, 95% CI −0.31, −0.24; PCS: β = −0.63, 95% CI −0.66, −0.59). The association between NPS and MCS or PCS was moderated by age, but the strongest associations were not in the oldest old.
Conclusion. NPS appears to be a potentially modifiable target for improving physical and mental HRQoL in older people. Future analyses should investigate the influence of NPS on HRQoL over time in older people.
Keywords: aged, cross-sectional survey, health-related quality of life, mental health, multisite pain, pain sites, physical health
Introduction
Musculoskeletal pain commonly occurs at multiple body sites in community-dwelling older adults, with 21–43% of persons aged ≥65 years reporting pain at two or more sites [1–3], with the variation possibly dependent on the number of pain sites (NPS) and chronicity of pain measured. The prevalence of multiple pain sites appears relatively stable over time [4] and similar across age groups [5], with studies of older people showing only a slight decline in the prevalence of multiple site pain after about age 75 years [2, 3, 6].
NPS has been shown to have an almost linear relationship with poor health outcomes in a population aged 24–76 years, with a greater NPS associated with reductions in overall health, sleep quality, psychological health [5], functional ability [7] and work disability [8]. In older populations there is evidence of a dose–response relationship between the extent of pain (none, single site, multiple sites and/or widespread) and some health outcomes related to older age: poorer lower extremity function [2], risk of falls [6], risk of disability [1] and sleep difficulties [9] in those aged from 65, 70, 65 and 64 years, respectively. Furthermore, the prevalence of pain that interferes with daily life continues to increase with age, from 32% in women aged 50–59 years to 50% in those aged ≥80 [10].
Health-related quality of life (HRQoL) is a concept that represents an individual’s perceived health status and overall physical and mental well-being that is not specific to any disease [11]. In a national debate in the UK on measures of well-being, overall health and individual well-being were two of the domains found to be important to individuals [12]. Although persons with more extensive pain, such as those with widespread pain or FM, report poorer HRQoL than those with no widespread pain [13, 14], to our knowledge no previous studies have investigated the relationship between NPS and HRQoL in individuals >75 years of age (the oldest old). NPS represents not only a simple and useful gauge of how much pain a person has [15], but also potentially a focus for intervention strategies in which physical and mental HRQoL are key disease-independent outcomes in the oldest old. The aim of this study was to test the hypotheses that in community-dwelling older people, an increasing NPS is associated with reduced HRQoL and that the relationship is moderated by age, with the greatest impact in the oldest old.
Methods
The North Staffordshire Osteoarthritis Project (NorStOP) included a large population-based survey of musculoskeletal pain in adults aged ≥50 years from North Staffordshire, UK, using a postal questionnaire. Details of NorStOP survey methods have been published previously [10, 16, 17]. Briefly, questionnaires were mailed with a letter from the general practice and a study information leaflet. Reminders were sent to non-responders 2 and 4 weeks after the initial questionnaire. Consent to use the data collected in the postal questionnaires was implied through return of the questionnaires to the research centre [18]. The questionnaire included a consent form on which participants could additionally provide written permission for their medical records to be reviewed. Approval for the study was granted by the North Staffordshire Research Ethics Committee (reference numbers 1351 and 1430).
Study population
The sampling frame for NorStOP was all patients aged ≥50 years registered with six general practices (n = 20 293) who were part of Primary Care Research West Midlands North (http://www.crncc.nihr.ac.uk/about_us/ccrn/wmids-north/corporate/pcrn_westmids_north). In the UK, general practice registers provide convenient sampling frames for population surveys, with ∼98% of the British population registered with a general practitioner (GP) [19]. Prior to mailing, 79 people were excluded by their GPs, e.g. due to severe psychiatric or terminal illness, resulting in 20 214 questionnaires being mailed. During mailing, 396 people were excluded (143 deaths or departures from the practices, 53 people with cognitive problems and 200 questionnaires returned as addressee unknown), giving an eligible study population of 19 818.
Study questionnaire
Primary outcome measures
Mental HRQoL and physical HRQoL were measured using the mental and physical component summary (MCS and PCS) scales of the Medical Outcomes Study 12-item Short Form Health Survey (SF-12) [20]. The SF-12 is internationally validated [21], with evidence for acceptable reliability [22, 23] and validity [22–24] in older people, although evidence for its internal construct validity varies [22, 23]. MCS and PCS scores, standardized to US general population scores [mean 50 (s.d. 10)], range from 0 to 100, with lower scores indicating worse HRQoL [20].
Primary exposure measure
NPS was measured by asking if, in the past 4 weeks, participants had experienced pain lasting for ≥1 day in any part of their body [16]. Those answering yes were asked to shade the site(s) of their pain(s) on a blank body manikin. Completed manikins were scored using a transparent template that divided the manikin into 44 mutually exclusive pain sites. NPS was then summed to give a total score ranging from 0 to 44. These data collection and scoring methods have been routinely used to measure pain location and distribution in both clinical and research settings [10, 13, 16–18, 25–33] and have been shown to have adequate test–retest and high inter- and intrarater reliability for measuring pain distribution [31] and provide a similar prevalence of pain to written questions [29].
Potential confounders of the relationship between pain and HRQoL
The following self-reported data on factors potentially confounding the relationship between pain and HRQoL were collected. The individual social factors were employment status, marital status and socio-economic status [obtained by classifying current/most recent occupation according to the Standard Occupational Classification 2000 [34], from which the National Statistics Socio-economic Classification (NS-SEC) [35] was derived].
The health behaviours measured were self-reported BMI (calculated from weight in kilograms/height in metres squared), smoking status and frequency of alcohol consumption.
Morbidities commonly associated with older age were assessed by asking if participants suffered from chest problems, heart problems, deafness, problems with eyesight (excluding the need for glasses), elevated blood pressure and diabetes.
OA, which may be associated with HRQoL, was measured by electronic recording of OA (as a Read code) by a GP in a consultation. Read codes are a hierarchy of morbidity, symptom and process codes used to label consultations in UK general practice [36] and map to disease codes in the International Classification of Diseases 10. Read codes starting with N05 were used to identify the diagnosis of OA. In responders who had consented to use of their medical records, consultation records for OA were identified for the 2 years prior to baseline.
Statistical analysis
The analysis included participants who provided complete pain data, defined as either yes to pain in the past 4 weeks and shading on the manikin or no to pain in the past 4 weeks question and no shading on the manikin. Participant characteristics are presented according to NPS, for which those reporting one or more pain sites were categorized into four groups with approximately equal numbers of respondents (1–3, 4–6, 7–11 and 12–44 pain sites) [10]. Chi-square and one-way analysis of variance (ANOVA) tests examined the strength of the associations between NPS and all other measures. For analysis of the association between MCS or PCS mean scores and NPS, participants reporting ≥30 pain sites were grouped together (30–44), since there were few participants with values in this range (n = 193); a one-way ANOVA was used to test this association according to age group and it was illustrated using a lowess scatterplot.
The associations between MCS or PCS scores and NPS (0–44) were analysed using general linear models. Results are presented as β coefficients with 95% CIs. The adjusted R2 values were used to describe the percentage of variability that was explained by each model. Standard residual diagnostics were applied to assess model fit (see supplementary data, available at Rheumatology Online). The analyses were conducted as follows: (i) The linear regression models were cumulatively adjusted for (a) age group and sex, (b) BMI, alcohol, smoking, employment status, marital status and individual socio-economic status, (c) morbidities and (d) consultation for OA. (ii) An interaction term between age group and NPS (age group × NPS), i.e. categorical variable × continuous variable, was added to the model to test moderation by age group in the fully adjusted model. A significant interaction between age group and NPS would indicate that the effect of NPS on HRQoL was different in different age groups. (iii) In the case of a significant interaction, separate fully adjusted models (with no interaction term) of the association between HRQoL and NPS were derived for each age group to examine any trend in the strength of association. Data were analysed with PASW Statistics version 18 (SPSS, Chicago, IL, USA). Multiple imputation was applied to assess the impact of missing data on the results (see supplementary data, available at Rheumatology Online).
Results
From the eligible study population of 19 818, a total of 13 986 people completed and returned questionnaires, giving an adjusted response of 70.6%. Of those, 12 408 participants provided complete pain data (88.7%). A total of 1578 participants did not provide complete pain data (275 answered yes to pain in the past 4 weeks but did not shade on the manikin; 77 answered no to pain in the past 4 weeks but shaded pain on the manikin; 1226 did not answer the question about pain in the past 4 weeks).
A total of 8890 (71.6%) participants reported one or more pain sites out of a possible 44; 669 (5.4%) had a single site of pain, 8221 (66.3%) participants reported pain at two or more sites and 6408 (51.6%) reported pain at four or more sites. The distribution of NPS in the study population showed a similar pattern for each age group (Fig. 1).
Fig. 1.
Distribution of number of pain sites in NorStOP participants according to age group
NorStOP: North Staffordshire Osteoarthritis Project.
Female participants were more likely to report a higher NPS than males, but there was no relationship with age (Table 1). Most health and socio-economic circumstances were significantly associated with increasing NPS: MCS and PCS scores decreased (worsening mental and physical HRQoL) and BMI, the likelihood of being a current/previous smoker, reporting a morbidity, not working due to ill health or being a routine/manual worker increased.
Table 1.
Characteristics of the NorStOP study participants according to number of pain sites
| Total | Number of pain sites |
P-value | |||||
|---|---|---|---|---|---|---|---|
| 0 | 1–3 | 4–6 | 7–11 | 12–44 | |||
| Overall, n (%) | 3518 (28.4) | 2482 (20.0) | 2322 (18.7) | 2022 (16.3) | 2064 (16.6) | ||
| Age, n (%), years | |||||||
| 50–59 | 4071 (32.8) | 1166 (33.1) | 800 (32.2) | 754 (32.5) | 693 (34.3) | 658 (31.9) | 0.116 |
| 60–69 | 3820 (30.8) | 1019 (29.0) | 792 (31.9) | 720 (31.0) | 617 (30.5) | 672 (32.6) | |
| 70–79 | 3061 (24.7) | 912 (25.9) | 592 (23.9) | 591 (25.5) | 489 (24.2) | 477 (23.1) | |
| ≥80 | 1456 (11.7) | 421 (12.0) | 298 (12.0) | 257 (11.1) | 223 (11.0) | 257 (12.5) | |
| Sex, n (%) | |||||||
| Female | 6910 (55.7) | 1863 (53.0) | 1325 (53.4) | 1256 (54.1) | 1182 (58.5) | 1284 (62.2) | <0.001 |
| Male | 5498 (44.3) | 1655 (47.0) | 1157 (46.6) | 1066 (45.9) | 840 (41.5) | 780 (37.8) | |
| SF-12, mean (s.d.) | |||||||
| MCS | 49.02 (11.18) | 52.13 (9.36) | 50.59 (10.49) | 49.23 (11.09) | 46.96 (11.62) | 43.53 (12.12) | <0.001 |
| PCS | 41.00 (12.56) | 48.95 (9.77) | 43.42 (11.43) | 40.05 (11.41) | 36.51 (11.22) | 29.73 (9.85) | <0.001 |
| BMI, mean (s.d.) | 26.56 (4.66) | 25.65 (4.05) | 26.33 (4.34) | 26.50 (4.42) | 27.31 (5.02) | 27.68 (5.50) | <0.001 |
| Alcohol, n (%) | |||||||
| < once per week | 5619 (45.9) | 1504 (43.2) | 1009 (41.2) | 1016 (44.3) | 931 (46.4) | 1159 (57.1) | <0.001 |
| ≥ once per week | 6635 (54.1) | 1974 (56.8) | 1438 (58.8) | 1277 (55.7) | 1074 (53.6) | 872 (42.9) | |
| Smoking, n (%) | |||||||
| Never | 5147 (41.9) | 1560 (44.8) | 1102 (44.9) | 921 (40.1) | 794 (39.6) | 770 (37.7) | <0.001 |
| Previous | 5200 (42.3) | 1385 (39.7) | 984 (40.1) | 1016 (44.2) | 895 (44.6) | 920 (45.0) | |
| Current | 1939 (15.8) | 541 (15.5) | 367 (15.0) | 361 (15.7) | 316 (15.8) | 354 (17.3) | |
| Employment status, n (%) | |||||||
| Employed | 3257 (27.1) | 1052 (30.9) | 734 (30.5) | 653 (28.9) | 497 (25.3) | 321 (16.1) | 0.001 |
| Not working due to ill health | 951 (7.9) | 90 (2.6) | 101 (4.2) | 155 (6.9) | 207 (10.6) | 398 (20.0) | |
| Retired | 6747 (56.1) | 1935 (56.8) | 1352 (56.3) | 1263 (55.9) | 1072 (54.7) | 1125 (56.5) | |
| Unemployed/seeking work | 126 (1.0) | 32 (0.9) | 27 (1.1) | 29 (1.3) | 18 (0.9) | 20 (1.0) | |
| Housewife | 663 (5.5) | 202 (5.9) | 134 (5.6) | 106 (4.7) | 124 (6.3) | 97 (4.9) | |
| Other | 274 (2.3) | 94 (2.8) | 55 (2.3) | 52 (2.3) | 43 (2.2) | 30 (1.5) | |
| Marital status, n (%) | |||||||
| Married/cohabiting | 8300 (67.7) | 2337 (67.4) | 1695 (69.1) | 1581 (68.8) | 1361 (68.1) | 1326 (64.9) | 0.026 |
| Separated, divorced, widowed, single | 3962 (32.3) | 1132 (32.6) | 758 (30.9) | 717 (31.2) | 638 (31.9) | 717 (35.1) | |
| Socio-economic status, n (%) | |||||||
| Managerial/professional | 2023 (17.5) | 652 (19.8) | 414 (17.9) | 372 (17.1) | 311 (16.4) | 274 (14.4) | 0.001 |
| Intermediate | 2077 (17.9) | 597 (18.2) | 420 (18.1) | 393 (18.1) | 334 (17.7) | 333 (17.5) | |
| Routine/manual | 7317 (63.2) | 1998 (60.7) | 1452 (62.7) | 1374 (63.3) | 1225 (64.7) | 1268 (66.5) | |
| Other | 157 (1.4) | 42 (1.3) | 30 (1.3) | 32 (1.5) | 22 (1.2) | 31 (1.6) | |
| Morbiditiesa, n (%) | |||||||
| Chest problems | 2588 (20.9) | 489 (13.9) | 413 (16.6) | 487 (21.0) | 517 (25.6) | 682 (33.0) | <0.001 |
| Heart problems | 2219 (17.9) | 496 (14.1) | 366 (14.7) | 398 (17.1) | 404 (20.0) | 555 (26.9) | <0.001 |
| Deafness | 2277 (18.4) | 492 (14.0) | 395 (15.9) | 467 (20.1) | 419 (20.7) | 504 (24.4) | <0.001 |
| Eyesight problems | 2659 (21.4) | 598 (17.0) | 453 (18.3) | 495 (21.3) | 495 (24.5) | 618 (29.9) | <0.001 |
| Elevated blood pressure | 4180 (33.7) | 1087 (30.9) | 781 (31.5) | 748 (32.2) | 706 (34.9) | 858 (41.6) | <0.001 |
| Diabetes | 1061 (8.6) | 283 (8.0) | 169 (6.8) | 199 (8.6) | 190 (9.4) | 220 (10.7) | <0.001 |
aEach morbidity was analysed separately. Numbers of participants available for analysis: total, n = 12 408; SF-12 MCS, n = 10 823; SF-12 PCS, n = 10 823; BMI, n = 11 863; alcohol, n = 12 254; smoking, n = 12 286; employment status, n = 12 018; marital status, n = 12 262; socio-economic status, n = 11 574; consultation for OA, n = 9399. NorStOP: North Staffordshire Osteoarthritis Project; MCS: mental component summary; PCS: physical component summary.
There was evidence of negative linear (unadjusted) associations between MCS or PCS mean scores and NPS in all four age groups (Fig. 2). MCS mean scores decreased with increasing NPS approximately in parallel for the four age groups (Fig. 2A). These associations varied little with age group. PCS mean scores decreased strongly with increasing NPS (Fig. 2B). Differences in PCS mean scores between age groups diminished as NPS increased, with the four lines converging at ∼28 pain sites.
Fig. 2.
Unadjusted relationship between mean SF-12 MCS or PCS and number of pain sites according to age group
(A) SF-12 MCS. (B) SF-12 PCS. A lowess line was used to smooth the mean response profile in each age group. MCS: mental component summary; PCS: physical component summary; SF-12: 12-item Short Form Health Survey.
The complete case analysis and models based on imputed data yielded similar regression coefficients (data not shown), hence results from the complete case analyses are presented here. Both MCS and PCS scores decreased for every additional pain site reported (Table 2). These linear associations were independent of age group and sex. Additional adjustment for social factors, health behaviours and morbidities attenuated the strength of the associations between mental or physical HRQoL and NPS but the associations remained statistically significant. Additional adjustment for consultation for OA slightly reduced the strength of the associations further, but they remained statistically significant. The percentage of variability in both MCS and PCS explained by the unadjusted models was increased by the fully adjusted models. There was no pattern to the residuals when plotted against the predicted values, indicating no evidence of heterogeneity and a reasonable model fit for the fully adjusted models.
Table 2.
General linear models of association between the SF-12 MCS or PCS and number of pain sites
| n | β | 95% CI | P-value | Adjusted R2a | |
|---|---|---|---|---|---|
| SF-12 mental component | |||||
| Adjustments | |||||
| None | 10823 | −0.43 | −0.46, −0.40 | <0.001 | 0.076 |
| Model 1: adjusted for age and sex | 10823 | −0.42 | −0.45, −0.39 | <0.001 | 0.091 |
| Model 2: model 1 + adjusted for BMI, smoking, alcohol, employment status, marital status and socio-economic status | 9560 | −0.34 | −0.37, −0.31 | <0.001 | 0.132 |
| Model 3: model 2 + adjusted for chest problems, heart problems, deafness, eyesight problems, elevated blood pressure and diabetes | 9560 | −0.30 | −0.33, −0.26 | <0.001 | 0.152 |
| Model 4: model 3 + adjusted for consultation for OA | 7443 | −0.28 | −0.31, −0.24 | <0.001 | 0.151 |
| SF-12 physical component | |||||
| Adjustments | |||||
| None | 10823 | −0.87 | −0.90, −0.84 | <0.001 | 0.249 |
| Model 1: adjusted for age and sex | 10823 | −0.87 | −0.90, −0.84 | <0.001 | 0.345 |
| Model 2: model 1 + adjusted for BMI, smoking, alcohol, employment status, marital status and socio-economic status | 9560 | −0.72 | −0.75, −0.69 | <0.001 | 0.435 |
| Model 3: model 2 + adjusted for chest problems, heart problems, deafness, eyesight problems, elevated blood pressure and diabetes | 9560 | −0.66 | −0.69, −0.63 | <0.001 | 0.475 |
| Model 4: model 3 + adjusted for consultation for OA | 7443 | −0.63 | −0.66, −0.59 | <0.001 | 0.483 |
Regression coefficients are unstandardized. β = regression coefficient. aAdjusted R2 values are for the entire model in each case. MCS: mental component summary; PCS: physical component summary; SF-12: 12-item Short Form Health Survey.
Addition of an interaction term (age group × NPS) to the fully adjusted MCS model showed that the association between MCS and NPS was moderated by age (F3, 7414 = 12.419, P < 0.001). This significant interaction suggests the effect of NPS on MCS is different for different age groups. A similar result was observed after adding an interaction term (age group × NPS) to the fully adjusted PCS model, indicating that the overall association between PCS and NPS was also moderated by age (F3, 7414 = 6.006, P < 0.001).
Separate fully adjusted models (with no interaction term) of the association between MCS or PCS and NPS were derived for each age group (Table 3). Although some differences were observed in the associations between HRQoL and NPS according to age, the changes were modest overall for mental HRQoL. For MCS, the strength of the association increased up to age 70–79, followed by a slight decrease in strength; for PCS, the strength of the association was greater than for MCS, but changed little between ages 50 and 69 and decreased thereafter.
Table 3.
Association between the SF-12 MCS or PCS and number of pain sites, stratified by age groupa
| n | βb | 95% CI | P-value | Adjusted R2c | |
|---|---|---|---|---|---|
| SF-12 mental component | |||||
| Age 50–59 years | 2659 | −0.15 | −0.21, −0.09 | <0.001 | 0.162 |
| Age 60–69 years | 2375 | −0.30 | −0.36, −0.24 | <0.001 | 0.148 |
| Age 70–79 years | 1718 | −0.40 | −0.47, −0.32 | <0.001 | 0.156 |
| Age ≥80 years | 691 | −0.31 | −0.42, −0.20 | <0.001 | 0.145 |
| SF-12 physical component | |||||
| Age 50–59 years | 2659 | −0.64 | −0.69, −0.59 | <0.001 | 0.510 |
| Age 60–69 years | 2375 | −0.68 | −0.73, −0.62 | <0.001 | 0.453 |
| Age 70–79 years | 1718 | −0.59 | −0.66, −0.52 | <0.001 | 0.352 |
| Age ≥80 years | 691 | −0.48 | −0.58, −0.37 | <0.001 | 0.232 |
aAdjusted for sex, BMI (continuous), alcohol, smoking, employment status, marital status, individual socio-economic status, chest problems, heart problems, deafness, eyesight problems, elevated blood pressure, diabetes and consultation for OA. bA general linear model was generated for each age group separately. cAdjusted R2 values are for the entire model for each age group. MCS: mental component summary; PCS: physical component summary; SF-12: 12-item Short Form Health Survey.
Discussion
To our knowledge this is the first study to examine the relationship between HRQoL and NPS in older adults, including those aged ≥75. The hypothesis that among older people an increasing NPS is associated with poorer HRQoL was supported with a significant linear relationship between an increasing NPS and decreasing mental or physical HRQoL assessed by the SF-12. These relationships persisted after adjustment for age, sex, social factors, health behaviours, morbidities and consultation for OA. The fully adjusted models explained 15% of the variance in MCS scores and 48% of the variation in PCS scores. The second hypothesis was not fully confirmed because, although the associations between HRQoL and NPS were moderated by age, the strongest associations were not in the oldest old for either mental or physical HRQoL.
This study demonstrates a dose-response relationship between the extent of pain and both physical and mental HRQoL in older people, including those aged ≥75. It builds on results from a study of younger adults (24–76 years old) in which there was a linear relationship between a smaller range of pain sites (1–10) and psychological distress and poor general physical and psychological health [5]. The current study is also in line with studies of older populations, including those aged ≥75, which have found that physical and mental markers of geriatric syndromes [37], such as increased risk of disability [1], poorer lower extremity function [2], locomotor disability [26], cognitive complaints [18] and cognitive decline [27], are associated with increased extent of pain.
At the population level, widespread pain and NPS have been shown to be a relatively stable trait over time in adults aged up to 85 and 62, respectively [4, 28]. However, there is significant individual variation in the reporting of NPS over time. Data from studies of chronic widespread pain show that two-thirds of individuals with chronic widespread pain at baseline no longer reported it at follow-up, although half continued to report some pain, with only 15% becoming pain free [28, 38]. Furthermore, it is likely that recovery will be associated with better outcomes. With the predicted increase in the percentage of the population aged ≥50, chronic musculoskeletal pain and its main consequence, i.e. disability in later life [39], in older people will become an increasing problem for clinicians working in primary [40] and secondary care relative to that of other chronic diseases in the next 20 years [41]. Assuming that NPS is a continuum [15], then the question remains, how can we shift not only the population, but individuals as well, down the continuum whereby the impact on health-related outcomes, such as mental and physical HRQoL, is likely to be reduced?
This study has several strengths. It was a large, general population survey of older people, including a substantial number of the oldest old (36% of participants were aged ≥70 years and 12% were ≥80), with a high response to the questionnaire. Inclusion of the widely used SF-12 to measure mental and physical HRQoL allows comparison of the results with other studies. Additionally, compared with the SF-36 from which it was derived, the SF-12 has fewer items and can be completed more quickly, reducing respondent burden [11, 20, 42]. This may be an important consideration for the participation of older people in a study, particularly those aged 65 and over with existing impairments and disabilities [43]. Some authors have suggested that older person-specific measures of HRQoL would be preferable, as they may have greater validity in older adults [43, 44], although a structured review of such instruments found limited evidence for their performance [44]. Hence we cannot exclude the possibility that an older person-specific HRQoL measure may have provided a more precise picture of the association of HRQoL with NPS. Several potential confounders for the association between HRQoL and NPS, including morbidities common in those with multiple site pain [45] and of older age, were assessed. Adjustment for consultation for OA was included since symptomatic OA has been shown to be associated with reduced HRQoL [46]. However, we do not believe that NPS represents underlying OA because chronic musculoskeletal pain is not necessarily associated with advanced radiographic changes in joints in which the symptoms are located [47], chronic musculoskeletal pain commonly affects multiple (including non-joint) sites in the body [1] and the genetic factors that predispose to developing chronic musculoskeletal pain are independent of the genetic factors that predispose to developing OA [48].
There are a number of limitations to this study. The range of pain sites measured was 0–44. Inevitably, if the manikin had been divided into fewer pain sites, the prevalence of multiple site pain would have been lower; however, our aim was to use the manikin to estimate as precisely as possible the extent of pain experienced by our population. Although manikins are routinely used in population-based pain research [10, 13, 16–18, 25–33], they can be subject to missing data. In our study, of those who did not provide complete pain data, 2.2% reported pain in the past 4 weeks but did not shade pain on the manikin. However, the addition of this small extra number of participants to the total is unlikely to have influenced the results significantly. Clinically some patterns from self-completed pain diagrams compare favourably with referrals to rheumatology clinics, suggesting their potential future use in prioritising rheumatology referrals, but further study is needed [32]. The manikin used in our study potentially captures both acute and chronic pain, which may limit its clinical relevance, e.g. any acute pain included in our measure will dilute the overall effect, potentially giving an underestimation of chronic pain. However, there is evidence that a blank manikin captures worse pain (longer duration, more severe, more disability) than a pre-shaded manikin [29], which would be consistent with the characteristics of chronic, rather than acute, pain. Furthermore, the recall of pain over extended periods of time may be subject to bias. Although reported NPS remains fairly stable over time [4], our study was cross-sectional. We therefore suggest repeating our study longitudinally to determine whether decreases in NPS lead to improved HRQoL over time.
Non-respondents to the questionnaire were more likely to be male and younger than respondents. This could affect the prevalence of pain reporting, although there was a non-significant difference in pain prevalence between responders to the first mailing and late responders [16], and it is unlikely that the associations between NPS and HRQoL will be affected. Furthermore, the associations from the imputation and the complete case analyses were similar. The study was conducted in a more deprived area in terms of health, employment and education, but less deprived in terms of housing and services than in England overall [17], which may limit the generalizability of the findings. Morbidity data were self-reported, some of which may be prone to reporting bias [49–51]. However, the agreement between self-reported and medical record data has been shown to be good for diabetes, hypertension and some specific heart problems [49–51]. While we adjusted for self-reported morbidity data, we did not adjust for diagnosed morbidities (e.g. coronary heart disease, chronic obstructive pulmonary disorder), which may have explained some of the association between physical HRQoL and NPS. If this were true we would have expected to find the strongest association (unadjusted for diagnosed morbidities) between physical HRQoL and NPS in the older age groups, since the prevalence of diagnosed morbidities increases with age; however, the strongest association between physical HRQoL and NPS was in the younger age groups (ages 50–69 years). Also, there may be confounders in addition to those measured in this study that contribute to older people’s declining function (e.g. cognitive problems, anxiety, depression, sleep) and may provide further explanation of some of the associations. Lastly, the errors were non-normal, but the sampling distributions of the model parameters will be approximately normal for large sample sizes according to the central limit theorem [52].
This study has shown that both mental and physical HRQoL decrease with increasing NPS in older people, including those ≥75 years of age. Age moderates the associations between NPS and mental or physical HRQoL, although the strongest associations are not in the oldest old. NPS could provide a clear and measureable gauge and target for interventions aimed at maintaining and improving HRQoL in older individuals. Based on these data, the next step would be to conduct longitudinal analyses to understand the influence of NPS on mental and physical HRQoL over time in older people.
Rheumatology key messages.
Physical and mental HRQoL decline with an increasing number of pain sites in older people.
The impact of number of pain sites on mental HRQoL increases with age up to age 70–79 years.
Number of pain sites could provide a target for improving HRQoL in older people.
Supplementary Material
Acknowledgements
The authors would like to thank all study participants, GPs and staff at the participating general practices. The authors wish to acknowledge the contributions of all members of the NorStOP research team to the study design and data collection, Prof Peter Croft for advice on the manuscript and the administrative and informatics staff at the Research Institute for Primary Care & Health Sciences at Keele University.
Funding: This work was supported by the Medical Research Council (grants G9900220 and G0501798), Primary Care Research West Midlands North and the North Staffordshire Primary Care Research Consortium. The funders had no role in the study design, data collection, analysis, data interpretation or writing of this article.
Disclosure statement: The authors have declared no conflicts of interest.
Supplementary data
Supplementary data are available at Rheumatology Online.
References
- 1.Soldato M, Liperoti R, Landi F, et al. Non malignant daily pain and risk of disability among older adults in home care in Europe. Pain. 2007;129:304–10. doi: 10.1016/j.pain.2006.10.016. [DOI] [PubMed] [Google Scholar]
- 2.Eggermont LH, Bean JF, Guralnik JM, Leveille SG. Comparing pain severity versus pain location in the MOBILIZE Boston study: chronic pain and lower extremity function. J Gerontol A Biol Sci Med Sci. 2009;64:763–70. doi: 10.1093/gerona/glp016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Croft P, Jordan K, Jinks C. “Pain elsewhere” and the impact of knee pain in older people. Arthritis Rheum. 2005;52:2350–4. doi: 10.1002/art.21218. [DOI] [PubMed] [Google Scholar]
- 4.Kamaleri Y, Natvig B, Ihlebaek CM, Benth JS, Bruusgaard D. Change in the number of musculoskeletal pain sites: a 14-year prospective study. Pain. 2009;141:25–30. doi: 10.1016/j.pain.2008.09.013. [DOI] [PubMed] [Google Scholar]
- 5.Kamaleri Y, Natvig B, Ihlebaek CM, Benth JS, Bruusgaard D. NPS is associated with demographic, lifestyle, and health-related factors in the general population. Eur J Pain. 2008;12:742–8. doi: 10.1016/j.ejpain.2007.11.005. [DOI] [PubMed] [Google Scholar]
- 6.Leveille SG, Jones RN, Kiely DK, et al. Chronic musculoskeletal pain and the occurrence of falls in an older population. JAMA. 2009;302:2214–21. doi: 10.1001/jama.2009.1738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kamaleri Y, Natvig B, Ihlebaek CM, Bruusgaard D. Localized or widespread musculoskeletal pain: does it matter? Pain. 2008;138:41–6. doi: 10.1016/j.pain.2007.11.002. [DOI] [PubMed] [Google Scholar]
- 8.Kamaleri Y, Natvig B, Ihlebaek CM, Bruusgaard D. Does the number of musculoskeletal pain sites predict work disability? A 14-year prospective study. Eur J Pain. 2009;13:426–30. doi: 10.1016/j.ejpain.2008.05.009. [DOI] [PubMed] [Google Scholar]
- 9.Chen Q, Hayman LL, Shmerling RH, Bean JF, Leveille SG. Characteristics of chronic pain associated with sleep difficulty in older adults: the Maintenance of Balance, Independent Living, Intellect, and Zest in the Elderly (MOBILIZE) Boston study. J Am Geriatr Soc. 2011;59:1385–92. doi: 10.1111/j.1532-5415.2011.03544.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Thomas E, Peat G, Harris L, Wilkie R, Croft PR. The prevalence of pain and pain interference in a general population of older adults: cross-sectional findings from the North Staffordshire Osteoarthritis Project. Pain. 2004;110:361–8. doi: 10.1016/j.pain.2004.04.017. [DOI] [PubMed] [Google Scholar]
- 11.Busija L, Pausenberger E, Haines TP, et al. Adult measures of general health and health-related quality of life: Medical Outcomes Study Short Form 36-Item (SF-36) and Short Form 12-Item (SF-12) Health Surveys, Nottingham Health Profile (NHP), Sickness Impact Profile (SIP), Medical Outcomes Study Short Form 6D (SF-6D), Health Utilities Index Mark 3 (HUI3), Quality of Well-Being Scale (QWB), and Assessment of Quality of Life (AQoL) Arthritis Care Res. 2011;63(Suppl 11):S383–412. doi: 10.1002/acr.20541. [DOI] [PubMed] [Google Scholar]
- 12.Self A, Thomas J, Randall C Office for National Statistics. 2012. Measuring National Well-being: Life in the UK. http://www.ons.gov.uk/ons/dcp171766_287415.pdf (4 June 2013, date last accessed) [Google Scholar]
- 13.Nicholl BI, Macfarlane GJ, Davies KA, et al. Premorbid psychosocial factors are associated with poor health-related quality of life in subjects with new onset of chronic widespread pain—results from the EPIFUND study. Pain. 2009;141:119–26. doi: 10.1016/j.pain.2008.10.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Salaffi F, Sarzi-Puttini P, Girolimetti R, et al. Health-related quality of life in fibromyalgia patients: a comparison with rheumatoid arthritis patients and the general population using the SF-36 health survey. Clin Exp Rheumatol. 2009;27(Suppl 56):S67–74. [PubMed] [Google Scholar]
- 15.Croft P. The question is not “have you got it”? But “how much of it have you got”? Pain. 2009;141:6–7. doi: 10.1016/j.pain.2008.10.019. [DOI] [PubMed] [Google Scholar]
- 16.Thomas E, Wilkie R, Peat G, et al. The North Staffordshire Osteoarthritis Project—NorStOP: prospective, 3-year study of the epidemiology and management of clinical osteoarthritis in a general population of older adults. BMC Musculoskelet Disord. 2004;5:2. doi: 10.1186/1471-2474-5-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jordan KP, Thomas E, Peat G, Wilkie R, Croft P. Social risks for disabling pain in older people: a prospective study of individual and area characteristics. Pain. 2008;137:652–61. doi: 10.1016/j.pain.2008.02.030. [DOI] [PubMed] [Google Scholar]
- 18.Westoby CJ, Mallen CD, Thomas E. Cognitive complaints in a general population of older adults: prevalence, association with pain and the influence of concurrent affective disorders. Eur J Pain. 2009;13:970–6. doi: 10.1016/j.ejpain.2008.11.011. [DOI] [PubMed] [Google Scholar]
- 19.Bowling A. Research Methods in Health. 3rd edn. Maidenhead: Open University Press; 2009. [Google Scholar]
- 20.Ware JE, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34:220–33. doi: 10.1097/00005650-199603000-00003. [DOI] [PubMed] [Google Scholar]
- 21.Gandek B, Ware JE, Aaronson NK, et al. Cross-validation of item selection and scoring for the SF-12 Health Survey in nine countries: results from the IQOLA Project. International Quality of Life Assessment. J Clin Epidemiol. 1998;51:1171–8. doi: 10.1016/s0895-4356(98)00109-7. [DOI] [PubMed] [Google Scholar]
- 22.Jakobsson U. Using the 12-item Short Form health survey (SF-12) to measure quality of life among older people. Aging Clin Exp Res. 2007;19:457–64. doi: 10.1007/BF03324731. [DOI] [PubMed] [Google Scholar]
- 23.Haywood KL, Garratt AM, Fitzpatrick R. Quality of life in older people: a structured review of generic self-assessed health instruments. Qual Life Res. 2005;14:1651–68. doi: 10.1007/s11136-005-1743-0. [DOI] [PubMed] [Google Scholar]
- 24.Pettit T, Livingston G, Manela M, et al. Validation and normative data of health status measures in older people: the Islington study. Int J Geriatr Psychiatry. 2001;16:1061–70. doi: 10.1002/gps.479. [DOI] [PubMed] [Google Scholar]
- 25.Lacey RJ, Lewis M, Jordan K, Jinks C, Sim J. Interrater reliability of scoring of pain drawings in a self-report health survey. Spine. 2005;30:E455–8. doi: 10.1097/01.brs.0000174274.38485.ee. [DOI] [PubMed] [Google Scholar]
- 26.Mottram S, Peat G, Thomas E, Wilkie R, Croft P. Patterns of pain and mobility limitation in older people: cross-sectional findings from a population survey of 18,497 adults aged 50 years and over. Qual Life Res. 2008;17:529–39. doi: 10.1007/s11136-008-9324-7. [DOI] [PubMed] [Google Scholar]
- 27.Lee DM, Pendleton N, Tajar A, et al. Chronic widespread pain is associated with slower cognitive processing speed in middle-aged and older European men. Pain. 2010;151:30–6. doi: 10.1016/j.pain.2010.04.024. [DOI] [PubMed] [Google Scholar]
- 28.Papageorgiou AC, Silman AJ, Macfarlane GJ. Chronic widespread pain in the population: a seven year follow up study. Ann Rheum Dis. 2002;61:1071–4. doi: 10.1136/ard.61.12.1071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lacey RJ, Lewis M, Sim J. Presentation of pain drawings in questionnaire surveys: influence on prevalence of neck and upper limb pain in the community. Pain. 2003;105:293–301. doi: 10.1016/s0304-3959(03)00244-6. [DOI] [PubMed] [Google Scholar]
- 30.Carnes D, Parsons S, Ashby D, et al. Chronic musculoskeletal pain rarely presents in a single body site: results from a UK population study. Rheumatology. 2007;46:1168–70. doi: 10.1093/rheumatology/kem118. [DOI] [PubMed] [Google Scholar]
- 31.van den Hoven LH, Gorter KJ, Picavet HS. Measuring musculoskeletal pain by questionnaires: the manikin versus written questions. Eur J Pain. 2010;14:335–8. doi: 10.1016/j.ejpain.2009.06.002. [DOI] [PubMed] [Google Scholar]
- 32.Caines A, Samadi N, Ouimet G, Thompson A, Pope JE. The sensitivity and specificity of pain diagrams in rheumatic disease referrals. Rheumatology. 2012;51:1093–8. doi: 10.1093/rheumatology/ker422. [DOI] [PubMed] [Google Scholar]
- 33.Southerst D, Côté P, Stupar M, Stern P, Mior S. The reliability of body pain diagrams in the quantitative measurement of pain distribution and location in patients with musculoskeletal pain: a systematic review. J Manipulative Physiol Ther. 2013;36:450–9. doi: 10.1016/j.jmpt.2013.05.021. [DOI] [PubMed] [Google Scholar]
- 34.Office for National Statistics. Standard Occupational Classification 2000 (SOC2000) London, UK: Stationery Office; 2000. [Google Scholar]
- 35.Office for National Statistics. National Statistics Socio-economic Classification (NS-SEC 2002) London, UK: Stationery Office; 2002. [Google Scholar]
- 36.NHS Information Authority. The Clinical Terms Version 3 (The Read Codes) Birmingham, UK: NHS Information Authority; 2000. [Google Scholar]
- 37.Inouye SK, Studenski S, Tinetti ME, Kuchel GA. Geriatric syndromes: clinical, research, and policy implications of a core geriatric concept. J Am Geriatr Soc. 2007;55:780–91. doi: 10.1111/j.1532-5415.2007.01156.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.MacFarlane GJ, Thomas E, Papageorgiou AC, et al. The natural history of chronic pain in the community: a better prognosis than in the clinic? J Rheumatol. 1996;23:1617–20. [PubMed] [Google Scholar]
- 39. World Health Organisation: The burden of musculoskeletal conditions at the start of the new millennium. Report of a WHO Scientific Group. WHO Technical Report Series No. 919. http://whqlibdoc.who.int/trs/WHO_TRS_919.pdf (4 June 2013, date last accessed)
- 40.Jordan KP, Kadam UT, Hayward R, et al. Annual consultation prevalence of regional musculoskeletal problems in primary care: an observational study. BMC Musculoskelet Disord. 2010;11:144. doi: 10.1186/1471-2474-11-144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Jagger C, Matthews R, Matthews F, et al. The burden of diseases on disability-free life expectancy in later life. J Gerontol A Biol Sci Med Sci. 2007;62:408–14. doi: 10.1093/gerona/62.4.408. [DOI] [PubMed] [Google Scholar]
- 42.Jenkinson C, Layte R. Development and testing of the UK SF-12 (short form health survey) J Health Serv Res Policy. 1997;2:14–8. doi: 10.1177/135581969700200105. [DOI] [PubMed] [Google Scholar]
- 43.Parker SG, Bechinger-English D, et al. Factors affecting completion of the SF-36 in older people. Age Ageing. 2006;35:376–81. doi: 10.1093/ageing/afl003. [DOI] [PubMed] [Google Scholar]
- 44.Haywood KL, Garratt AM, Fitzpatrick R. Older people specific health status and quality of life: a structured review of self-assessed instruments. J Eval Clin Pract. 2005;11:315–27. doi: 10.1111/j.1365-2753.2005.00538.x. [DOI] [PubMed] [Google Scholar]
- 45.Leveille SG, Bean J, Ngo L, McMullen W, Guralnik JM. The pathway from musculoskeletal pain to mobility difficulty in older disabled women. Pain. 2007;128:69–77. doi: 10.1016/j.pain.2006.08.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Salaffi F, Carotti M, Stancati A, Grassi W. Health-related quality of life in older adults with symptomatic hip and knee osteoarthritis: a comparison with matched healthy controls. Aging Clin Exp Res. 2005;17:255–63. doi: 10.1007/BF03324607. [DOI] [PubMed] [Google Scholar]
- 47.Neogi T, Felson D, Niu J, et al. Association between radiographic features of knee osteoarthritis and pain: results from two cohort studies. BMJ. 2009;339:b2844. doi: 10.1136/bmj.b2844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Williams FM, Spector TD, MacGregor AJ. Pain reporting at different body sites is explained by a single underlying genetic factor. Rheumatology. 2010;49:1753–5. doi: 10.1093/rheumatology/keq170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Barber J, Muller S, Whitehurst T, Hay E. Measuring morbidity: self-report or health care records? Fam Pract. 2010;27:25–30. doi: 10.1093/fampra/cmp098. [DOI] [PubMed] [Google Scholar]
- 50.Okura Y, Urban LH, Mahoney DW, Jacobsen SJ, Rodeheffer RJ. Agreement between self-report questionnaires and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure. J Clin Epidemiol. 2004;57:1096–103. doi: 10.1016/j.jclinepi.2004.04.005. [DOI] [PubMed] [Google Scholar]
- 51.Simpson CF, Boyd CM, Carlson MC, et al. Agreement between self-report of disease diagnoses and medical record validation in disabled older women: factors that modify agreement. J Am Geriatr Soc. 2004;52:123–7. doi: 10.1111/j.1532-5415.2004.52021.x. [DOI] [PubMed] [Google Scholar]
- 52.Hogg RV, Tanis E. Probability and Statistical Inference. 8th edn. Harlow: Pearson; 2013. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.