No long-term impact of low-energy distal radius fracture on health-related quality of life and global quality of life: a case-control study

To study one-year changes in HRQOL and GQOL in patients with low-energy distal radius fracture we applied a case-control, prospective longitudinal study design. The study was recommended by the Regional Committee for Medical Research Ethics and approved by the National Data Inspectorate.

Patients with low-energy distal radius fracture aged 50 years and older were consecutively recruited from an osteoporosis centre at a regional hospital in southern Norway in 2004 and 2005. A low-energy fracture was defined as a minimal trauma falling from standing height or less [15], and a distal radius fracture was defined as located within 3 cm of the radio-carpal joint [16]. The distal radius fractures were closed injuries, and the fractures were treated conservatively by stabilising the fracture by a plaster cast or by external fixation. Patients were assessed and data collected in median 10 days (interquartile range 13) after fracture and reassessed one year after fracture. With regard to demographical and clinical variables, HRQOL and GQOL the patients were asked to report their status prior to fracture. The patients also were asked to report their exercise habits, falls and the use of health care recourses during the year before fracture. The controls were asked about their status and habits prior to inclusion. The same data collection performed at baseline was repeated after one year.

The included patients comprised 56% of all 324 patients with low-energy distal radius fracture treated at the hospital, and 73% of 249 patients examined at the osteoporosis centre. Before inclusion in this study, we confirmed that the fracture was not a result of high-energy trauma and was caused only by minimal trauma according to the definition of low-energy fracture [17]. Patients who were excluded comprised a total of 51 patients with confusion or dementia, serious infection, patients not capable of giving informed consent, patients not capable of speaking Norwegian or tourists and 92 patients who did not want to participate in the study.

At baseline 181 patients with distal radius fracture were included along with 181 age- and sex-matched controls. The age- and sex-matched controls were randomly allocated from the national registry for the catchment area and invited by mail to participate in the study. The controls were identified consecutively along with patient recruitement. If a potential control refused to participate or did not respond to the invitation, a new control was invited. Overall, 131 potential controls refused to participate or did not respond to the invitation. We aimed an age match of ± 1 year in the patients with distal radius fracture; however, this was a challenge for some of the patients aged 80 years and older. In these patients we accepted a match of ± 5 years, except for one woman aged 96 years who was matched with an 86 years old control.

Demographical and clinical data (listed in table 1) were collected, and included also exercise, smoking habits, medication, previous fracture, number of falls the year before the fracture, and co-morbidity. Furthermore, patients and controls reported their use of health care resources; like visiting general practitioners, medical specialists, physiotherapists, and hospitalization the year prior to the fracture or prior to inclusion in the control group. Regular exercise was defined as walking or doing more intensive exercise more than 30 minutes three times a week. Previous fracture was defined as a low-energy trauma fracture after the age of 50. Co-morbidity included heart diseases, pulmonary diseases, neurological disorders, urogenital disorders, gastrointestinal disorders, endocrine disorders, inflammatory joint disorders and connective tissue disorders, cancer, mental disorders. For co-morbidity, we also computed a sum score of the number of diseases in each patient and control, which was used in the multivariate analyses.

Standardized BMD measurements at lumbar spine L2-4 and both hips were performed by four trained nurses using the same dual energy X-ray absorptiometry (DXA) equipment (General Electric, Lunar Prodigy) at baseline and at one year follow-up. The machine was stable over the entire measurement period. Long term spine phantom in-vitro coefficient of variation (CV) for the whole period was 0.62%. The in-vivo CV for the measurement procedure was 1.19% at lumbar spine L2-4, 0.95% at right total hip and 0.89% at left total hip. The BMD measurements were expressed as T-scores (SD) calculated on the basis of the reference value in the DXA machine provided by the manufacturer. Osteoporosis was defined as T-score ≤ -2.5 SD, osteopeniae as T-score > -2.5 and < -1.0 and normal BMD as T-score > -1.0, according to the WHO definition for osteoporosis [17].

Modified Health Assessment Questionnaire (MHAQ) measures a patients ability to perform activities of daily living [18, 19]. Although primarily developed as a measure for use in rheumatoid arthritis, MHAQ has been used across a variety of diseases [20]. The MHAQ consists of 8 items covering daily activities including skills that demand a good arm function e.g. dressing, lift a full cup or glass to the mouth, wash and dry the entire body [18, 19]. The total mean scores range from 1–4, with 1 representing "without any difficulty". For incomplete questionnaires, the missing values were replaced with the mean value of the answered questions of the respondent when at least 6 out of 8 items had valid response, which is based on the scale instructions given by the developers of the questionnair [20, 21]. At baseline all the the patients and controls had valid responses. At one year follow-up 1,5% of the patients and 1% of the controls had one or two missing responses. In the multivariate analyses, which were performed to identify if a low-energy distal radius fracture was a predictor of worsened arm functions, we rescaled MHAQ from 0 to 100, with 100 representing "without any difficulty" in accordance with prior studies [21, 22].

The Short- Form 36 (SF-36) was used to assess HRQOL (physical and mental health) [23, 24]. The SF-36 includes eight domains (general health, bodily pain, physical functioning, physical role limitations, mental health, vitality, social functioning, and emotional role limitations), which can be combined into a physical health summary scale and a mental health summary scale. The physical component summary (PCS) and mental component summary (MCS) scales were used in this study. The SF-36 scales were scored according to published scoring procedures, and each was expressed as a value from 0 to 100, with 100 representing "excellent health". For incomplete questionnaires substitution of missing values is based on the scale instructions given by the developers of the questionnaire [23, 24]. At baseline 5.6% of the patients and 13.5% of the controls had one or more missing responses. At one year follow-up 18.8% of the patients and 14% of the controls had one or more missing responses. The questionnaire has been thoroughly tested for psychometric properties in other studies, within several countries, including Norway [23‐26].

The Quality of Life Scale (QOLS), a 16-item, domain-specific instrument adapted by Burckhardt et al. for people with chronic conditions, was used to assess GQOL [9, 27, 28]. In this questionnaire GQOL is understood as a broad range of human experiences related to one's overall well-being and satisfaction. The QOLS is a self-administered questionnaire [27, 29]. The items are rated at a 7-point satisfaction scale. For incomplete questionnaires, the missing values were replaced with the mean value of the answered questions of the respondent when at least 80% of the items had a valid response. The substitution of missing values is based on the scale instructions given by the developers of the questionnaire [9, 27]. At baseline 26% of the patients and 23% of the controls had one or more missing responses. At one year follow-up 35% of the patients and 33% of the controls had one or more missing responses. The items with most missing responses were QOLS item number four (having and rearing children) and item five (close relationship with spouse or other significant other).

The questionnaire is scored by adding up the items to obtain a total score from a minimum of 16 to a maximum of 112. Higher scores indicate better GQOL. Burckhardt et al. [28] suggested that the QOLS comprising three sub-dimensions: relationship and marital well-being (items 3, 4, 5, 6, and 14); health and functioning (items 1, 2, 11, 15, and 16); and personal, social, and community commitment (items 7, 8, 9, 10, 12, and 13) [28, 30]. The three dimensions are scored by summing the scores for each item in the dimension. The questionnaire has been thoroughly tested for psychometric properties in other studies, within several countries [28, 30‐32].

Statistical analyses were carried out using the Statistical Package for Social Sciences (SPSS) for Windows (version 16.0). Chi-square tests and t-tests were used to compare differences between subgroups. Wilcoxon rank tests were used to compare continuous health care resources data between inclusion and one year follow-up, and paired samples t-tests were used to compare HRQOL and GQOL at inclusion and one year follow-up within the patients with distal radius fracture and within the controls. Furthermore, standard difference scores (s-scores) were calculated by subtracting the mean MHAQ, SF-36 or QOLS scores at baseline from the mean score of one year follow-up, and then dividing by the standard deviation (SD) at baseline [33]. To estimate the proportion of patients and controls with clinically significant changes in HRQOL and GQOL, we also identified participants with modest changes (between -5 and -10%), moderate changes (between -10 and -20%) and substantial changes (more than -20%) between baseline and one year follow-up [33, 34]

Multiple linear regression analyses (procedure GLM in the SPSS) were used to identify significant predictors of worsened HRQOL (delta total mean MHAQ, SF-36; delta PCS and delta MCS) or GQOL (delta QOLS) in the study-population (both patients and controls). The regression analyses were adjusted for baseline total mean MHAQ (rescaled), PCS, MCS or QOLS respectively, at inclusion. The independent variables in the multiple regression analyses were selected based on results from earlier studies which show that age, sex, education level, marital status, BMD, falls, BMI, co-morbidity and osteoporotic fractures appear to be associated with HRQOL and/or GQOL, and these variables were all included in the regression model [14]. To test if the effects of predictors of change in our dependent variables were significantly different for patients and controls, interaction terms involving the patient/control dichotomy and each of the predictors were entered one pair at a time, while retaining main effects in the model. The level of significance was set at 0.05.

The patients in the study were significantly (p < 0.001) younger (67 ± 9 years) than the excluded patients (76 ± 12 years) and those who did not want to participate (72 ± 11 years). Among the 181 patients with distal radius fracture and 181 controls included at baseline, 160 patients and 169 controls attended the osteoporosis-centre at one year follow-up. There were minor differences between participants at one year follow-up and participants who were lost to follow-up. Among patients with distal radius fracture a statistically significant difference was only reported for gastrointestinal disorders (p = 0.015). Among controls, those who were lost to follow-up were significantly older (p = 0.046), and reported significantly lower SF-36 score in vitality (p = 0.014).

Socio-demographical and clinical characteristics at baseline of those participants who completed both baseline and one year follow-up assessments are shown in table 1.

Patients with distal radius fracture more often were living alone (p = 0.011), had fewer years of education (p = 0.011), had lower BMI (p = 0.027), and were more frequently classified with osteoporosis (p < 0.001) compared to the controls. The distal radius fracture occurred indoors in 31 (19%) patients and outdoors in 129 patients (81%). Mean age in patients whose fracture occurred indoors was 70 ± 11 years old and outdoors 66 ± 9 years old (p = 0.074).

At one year follow-up, the patients with distal radius fracture compared with controls were also more frequent user of calcium and/or vitamin D treatment (103 vs. 88, p = 0.024) and antiresorbtive treatment (ART) (52 vs. 27, p < 0.001). Four patients with distal radius fracture and no controls got a new fracture between inclusion and one-year follow-up (p = 0.069).

During the one year follow-up, patients with distal radius fracture on average visited their general practitioner more frequently than the year before fracture (3.9 vs. 3.4, p = 0.006). There were no significant differences with regard to the number of visits to other health care providers, like medical specialists (p = 0.083), physiotherapist (p = 0.139) or number of days hospitalized the last year (p = 0.581), the year following the fracture compared to number of visits the year before fracture. There were no significant changes in the controls.

No significant changes in arm function as assessed by MHAQ were identified in patients with distal radius fracture at one year follow-up compared to the baseline (prior to fracture) assessment (p = 0.202) (table 2). Only ten patients (6%) with a distal radius fracture did not attain their pre-fracture arm functions one year after fracture.

Furthermore, no significant changes were identified in HRQOL as assessed by the SF-36; physical health (p = 0.209) and mental health (p = 0.840) from pre-fracture to one year after the fracture in the patients with a distal radius fracture. The same pattern was seen in controls (table 2).

With regard to GQOL, the patients with distal radius fracture reported significantly lower total GQOL score (p < 0.001, s-score = -0.4) and for the sub-dimensions; relationship and marital well-being (p = 0.015, s-score = -0.2), health and functioning (p = 0.001, s-score = -0.2) and personal, social and community commitment (p < 0.001, s-score = -0.3) at one year follow-up compared to the baseline assessment (table 3). In the controls we also found significant changes in GQOL scores within both the overall score and the three sub-dimensions (p < 0.001) one year after inclusion, with s-score = -0.6 in QOLS, s-score = -0.3 for relationship and marital well being, s-score = -0.4, s-score = -0.4 for health and functioning and s-score = -0.6 in personal, social and community commitment (table 3).

Modest (-5 to -10) or moderate (-10 to -20) worsening arm functions between baseline and one year follow up were reported by 2.5% of the patients and 3% of the controls and substantial change (-20 or more) in 2 patients and one control. Modest or moderate worsening of physical health was reported by 20% of the patients and 10% of the controls, and substantial changes in one patient. Modest or moderate worsening of mental health was reported by 19% of the patients and 14% of the controls and substantial changes in four patients and in one control. Furthermore, modest or moderate worsening of GQOL was reported by 35% of the patients and 42% of the controls.

No significant differences between the patients with distal radius fracture and controls were identified in HRQOL and GQOL at one year follow-up.

A low-energy distal radius fracture did not predict worsened HRQOL or GQOL one year after inclusion, and few predictors of changes were identified. Worsened arm function was predicted by low BMI (B = -0.20, p = 0.019) at baseline, worsened physical health was predicted by low education (B = 1.37, p = 0.017) at baseline, and living with someone predicted worsened mental health (B = 2.85, p = 0.009) (table 4).

Interaction terms between pairs of each independent variable and the patients/controls dichotomy (tested one pair at a time, with main effects retained) revealed no significantly different effects between the patients with wrist fracture and the controls in the regression analyses.