Chaplin, Wendy
(2025)
Investigating musculoskeletal pain and frailty in an ageing population.
PhD thesis, University of Nottingham.
Abstract
Background: Frailty is defined as a vulnerability state with decreased physiological reserve observed in older people. Frailty may be characterised by a loss of homeostatic resilience due to multi-organ, age-associated decline. People with frailty have an elevated susceptibility to stressors and/or disproportionate response to challenges; this leads to a significant functional decline and increased risk of adverse health outcomes.
Chronic pain is long-term or persistent pain which lasts for an extended period, typically beyond the normal healing time of an injury or illness; it is generally defined as pain that remains unresolved for three months.
While previous studies have touched upon the association of pain and frailty, my research delves deeper into this relationship, offering insight into this under-recognised association. This thesis proposes that chronic pain might make the transitions from non-frail to frail states more likely or less likely. Moreover, the presence of frailty might hinder the improvement of pain. The current evidence implies a potential bidirectional relationship between pain and frailty, but most of the previous research has examined each direction separately.
Pain is not usually accounted for in frailty classification; however, it is important to consider why pain is linked to frailty. People who are classified as frail may have more painful morbidities than those classified as non-frail. Additionally, it is possible that central aspects of pain could be evidence of a dysfunctional central nervous system. Such dysfunction may result in an overall increase in pain sensitivity. Central aspects of pain factor (CAPf) are considered to be associated with increased pain hypersensitivity and have been shown to predict future knee pain. Each of the 8 characteristics associated with the underlying CAP factor (anxiety, depression, catastrophising, pain distribution, neuropathic-like pain, cognitive impact, sleep, and fatigue) have been associated with frailty. Central pain mechanisms may explain the association of pain and frailty, but other sources of pain and pain mechanisms should also be investigated.
Understanding the association between these two common conditions. Furthermore, increasing awareness of this relationship may allow the implementation of actions to reduce the burden on health services and individuals.
Aims:
[1] to examine cross-sectional and longitudinal associations of pain with frailty in a cohort study. To investigate whether there is a unidirectional or bidirectional relationship between pain and frailty.
[2] to examine the extent to which the association of chronic pain with frailty might be attributed to morbidities.
[3] to investigate whether Central Aspects of Pain explain the association between chronic pain and frailty.
[4] to design the ACHING study, which aims to measure and assess the association of frailty with central pain mechanisms, alongside other potential causes of pain and pain severity in individuals with knee pain. This included preparation and training to collect some of the main physiological measures.
Methods:
Data were drawn from Investigating Musculoskeletal Health and Wellbeing, a UK-based cohort (n=2185). Participants were aged 60 years and over and either had or were at risk of musculoskeletal problems or frailty.
The main variables for this thesis were frailty, classified as present/absent using the FRAIL questionnaire, and average joint pain severity over the previous month was assessed using an 11-point numerical rating scale (NRS pain).
To confirm the association between pain and frailty, baseline and 1-year data were used for cross-sectional analysis and regression. Subsequently, this led to assessing directionality using longitudinal data and employing cross-lagged path analysis.
To assess chronic pain related to morbidities, pain from any source was reported using the McGill Pain Rating Index (PRI). Morbidities were classified as painful/non-painful using the International Association for the Study of Pain criteria. A modified FRAIL was employed in which the ‘illness item’ was omitted to remove the overlap with morbidity. These cross-sectional analyses used standardised variables in regression and Z-tests to assess the degrees of the association of pain (McGill Pain Rating Index), painful and non-painful morbidity counts with frailty (modified FRAIL).
Analyses of pain mechanism data were conducted in a subgroup of people with knee pain (NRS≥1) (n=639). This used the Central Aspects of Pain in the Knee questionnaire to calculate a modified Central Aspects of Pain score in which the ‘fatigue item’ was omitted to remove the overlap with the ‘fatigue item’ in FRAIL.
The ACHING study planned to recruit 122 participants to a case-control study. Participants classified as frail would be age- and sex-matched with people classified as robust. Measures were selected to explore multiple causes of pain, including peripheral indices and central aspects of pain. Additionally, selecting measures to explore multiple possible causes of frailty. The reliability of some of the main physiological measures was established in a group of healthy volunteers (n=20).
A statistical plan was devised for the ACHING study. The reliability between raters was assessed using Bland-Altman plots, calculating intraclass correlation coefficients and concordance correlation coefficients.
Results:
The cross-sectional data were drawn from 2,185 participants from the Investigating Musculoskeletal Health and Wellbeing cohort. Of those, 55% were female, with a median age of 74 (range 60 to 96) years. FRAIL classified 438 (20%) of participants as frail. Mean (SD) NRS pain was 5.5 (2.5). NRS pain was associated with frailty at baseline (aOR1.68, (95%CI 1.57 to 1.79), p<0.001).
Longitudinal data were from 1,179 participants who completed both baseline and 1-year measures. At baseline, 176 (15%) participants were classified as frail, and the mean (SD) NRS pain score was 5.2 (2.5). In logistic regression, NRS pain was associated with frailty at baseline (aOR 1.72, (95%CI 1.56 to 1.92), p<0.001). Additionally, NRS pain at baseline was associated with 1-year frailty (aOR 1.28, (95%CI 1.15 to 1.43), p<0.001). In cross-lagged path analysis, higher baseline pain strongly predicted 1-year frailty [β =0.25, (95%CI 0.14 to 0.36), p<0.001] and baseline frailty predicted higher 1-year pain [β =0.06, (95%CI 0.003 to 0.11), p=0.040].
When the influence of morbidities on pain and frailty was examined, PRI pain (aOR 2.07, (95%CI 1.83 to 2.33) and ‘any’ morbidity (aOR 1.74, (95%CI 1.54 to 1.97) were both significantly associated with frailty. When morbidity was subclassified, painful (aOR 1.48, (95%CI 1.30 to 1.68) and non-painful (aOR 1.39, (95%CI 1.24 to 1.56) morbidities each were associated with frailty, and PRI pain also remained associated (aOR 2.07, (95%CI 1.83 to 2.34), p<0.001).
In the subgroup of people with knee pain at baseline (n=639), 26% of participants were classified as frail, and the mean (SD) NRS pain was 6.0 (2.1). At 1-year, there were data from 343 participants with a mean (SD) NRS pain of 5.5 (2.3). A higher modified Central Aspects of Pain factor was associated with frailty (aOR 1.53, (95%CI 1.41 to 1.66), p<0.001) at baseline. When both modified Central Aspects of Pain factor (aOR 1.37, (95%CI 1.26 to 1.50) and NRS pain (aOR 1.54, (95%CI 1.33 to 1.78) were included in the same model, they were both significantly associated with frailty classification, (p<0.001). At 1-year, the modified Central Aspects of Pain factor did not predict future frailty, but NRS pain at baseline did predict 1-year frailty (aOR 1.33, (95%CI 1.05 to 3.79), p =0.016).
The ACHING study protocol was developed and prepared. An ethics application was submitted and received ethical approval from the Health Research Authority. The training and collection of the main physiological measures showed acceptable reliability.
Conclusion:
There is a bidirectional relationship between chronic pain and frailty. This could lead to a vicious cycle in which each accelerates the other’s progression. Pain could be the key driver of this association, as implied by its stronger coefficient from the analyses performed. Focusing on pain management as an intervention pathway could mitigate the effect of chronic pain upon frailty. Future studies might focus on comprehensive assessments of pain mechanisms to determine which causes of pain might be most important as future treatment targets. Given the ageing population in many countries, it is increasingly important to address frailty which disproportionately affects older individuals and to ensure that we manage chronic pain.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
