![[ Visit Client Website ]](images/banner.gif)
PURPOSE: To examine the nature of fatigue experienced by individuals with OA, as well as some of the measurement properties of a multi-dimensional fatigue measure, including potential overlap with pain and depression
METHODS: Individuals with symptomatic hip or knee OA (n=809) participating in an ongoing population-based, longitudinal study were interviewed by telephone. The main fatigue measure was the Multi-dimensional Fatigue Symptom Inventory-Short Form (MFSI-SF: general, physical, emotional and mental fatigue, vigour) Additional fatigue data were collected using the Fatigue Sympton Inventory (FSI) and the Profile of Mood States (POMS). Data were also collected on pain (WOMAC), depression (CES-D) and socio-demographics. Internal consistency (Cronbach’s α) was evaluated for the MFSI-SF. Correlations (Spearman’s r) between the MFSI-SF and the other fatigue measures were computed to assess convergent validity. Correlations between the MFSI-SF subscale scores and pain and depression were assessed. Correlations of individual fatigue items with pain and depression scores, as well as analogous correlations for depression and pain items were computed to examine the potential for construct overlap.
RESULTS: Participants were of mean age 76.5 years (range 64 to 100); 75.8% were female. Mean (SD) MFSI-SF scores were 12.3(5.9)/24, 4.6(3.9)/24, 3.0(3.7)/24, 1.4(2.3)/24 and 10.3(3.8)/24 for general, physical, emotional and mental fatigue and vigour, respectively (higher scores = greater fatigue/vigour). Scores for general and physical fatigue are higher than those in a published cancer sample, while those for emotional and mental fatigue are lower. Fatigue scores were generally higher for women, older people and those with less education. Internal consistency on all subscales was acceptable (α>0.7). Most correlations between subscales scores were r=0.3-0.5, indicating that although related, the MFSI-SF subscales appear to measure distinct fatigue types. This is also supported by the variation in correlations between the MFSI-SF subscales and the other fatigue measures, which ranged from r=0.1(mental fatigue) to r=0.9 (general fatigue). Moderate to high correlations (r=0.5-0.7) for fatigue with pain and depression suggest that these constructs are related, so that those with higher pain or depression levels tended to have greater fatigue, although this varied by fatigue type. There was also evidence that there may be overlap among these constructs, such that items from one measure may also capture other construct(s): 6 MFSI-SF items had correlations of ±0.6 with CES-D scores and 7 CES-D items similarly correlated with fatigue scores. CONCLUSIONS: The majority of fatigue experienced by individuals with OA appears to be characterized as general or physical fatigue. Findings suggest that fatigue in OA may be associated with pain and depression, however in order to fully understand these relationships, further research is required to better understand and account for measurement overlap among constructs.
METHODS: Individuals with symptomatic hip or knee OA (n=809) participating in an ongoing population-based, longitudinal study were interviewed by telephone. The main fatigue measure was the Multi-dimensional Fatigue Symptom Inventory-Short Form (MFSI-SF: general, physical, emotional and mental fatigue, vigour) Additional fatigue data were collected using the Fatigue Sympton Inventory (FSI) and the Profile of Mood States (POMS). Data were also collected on pain (WOMAC), depression (CES-D) and socio-demographics. Internal consistency (Cronbach’s α) was evaluated for the MFSI-SF. Correlations (Spearman’s r) between the MFSI-SF and the other fatigue measures were computed to assess convergent validity. Correlations between the MFSI-SF subscale scores and pain and depression were assessed. Correlations of individual fatigue items with pain and depression scores, as well as analogous correlations for depression and pain items were computed to examine the potential for construct overlap.
RESULTS: Participants were of mean age 76.5 years (range 64 to 100); 75.8% were female. Mean (SD) MFSI-SF scores were 12.3(5.9)/24, 4.6(3.9)/24, 3.0(3.7)/24, 1.4(2.3)/24 and 10.3(3.8)/24 for general, physical, emotional and mental fatigue and vigour, respectively (higher scores = greater fatigue/vigour). Scores for general and physical fatigue are higher than those in a published cancer sample, while those for emotional and mental fatigue are lower. Fatigue scores were generally higher for women, older people and those with less education. Internal consistency on all subscales was acceptable (α>0.7). Most correlations between subscales scores were r=0.3-0.5, indicating that although related, the MFSI-SF subscales appear to measure distinct fatigue types. This is also supported by the variation in correlations between the MFSI-SF subscales and the other fatigue measures, which ranged from r=0.1(mental fatigue) to r=0.9 (general fatigue). Moderate to high correlations (r=0.5-0.7) for fatigue with pain and depression suggest that these constructs are related, so that those with higher pain or depression levels tended to have greater fatigue, although this varied by fatigue type. There was also evidence that there may be overlap among these constructs, such that items from one measure may also capture other construct(s): 6 MFSI-SF items had correlations of ±0.6 with CES-D scores and 7 CES-D items similarly correlated with fatigue scores. CONCLUSIONS: The majority of fatigue experienced by individuals with OA appears to be characterized as general or physical fatigue. Findings suggest that fatigue in OA may be associated with pain and depression, however in order to fully understand these relationships, further research is required to better understand and account for measurement overlap among constructs.
J. Power, None; E.M. Badley, None; G.A. Hawker, None.