Abstract
Background:
Ethnic disparities have been demonstrated in the treatment of chronic diseases, such as diabetes and heart disease. It is unclear if similar ethnic disparities appear with respect to recovery following fracture care.
Methods:
We retrospectively reviewed 496 individuals (253 whites, 100 blacks, and 143 Latinos) with a fracture of the distal part of the radius. Assessment of physical function and pain was conducted at three, six, and twelve months following treatment. The Disabilities of the Arm, Shoulder and Hand (DASH) score was used to assess physical function, and a visual analog scale was used to assess pain. Multiple linear regression was used to model physical function and pain across ethnicity while controlling for age, sex, mechanism of injury, level of education, type of fracture, type of treatment (operative or nonoperative), and Workers’ Compensation status.
Results:
Both blacks and Latinos exhibited poorer physical function and greater pain than whites did at most follow-up points. Latinos reported more pain at each follow-up point in comparison with blacks and whites (p < 0.001 at three, six, and twelve months). These significant differences remained after controlling for Workers’ Compensation status, which was also strongly associated with both pain and function.
Conclusions:
These findings suggest that recovery is different between ethnic groups following a fracture of the distal part of the radius. These ethnic disparities may result from multifactorial sociodemographic factors that are present both before and after fracture treatment.
Level of Evidence:
Prognostic Level II. See Instructions to Authors for a complete description of levels of evidence.
Fractures may be associated with diminished physical function over time. This is true of multiple fracture sites, including one of the most common, the distal part of the radius. Ethnic disparities have been documented in association with many chronic disease conditions over the last two decades1, and morbidity is disproportionately represented in black and Latino populations. To date, only one study has specifically identified worse fracture outcomes on the basis of ethnicity2. Additionally, more than twenty years ago, a study of patients with hip fractures demonstrated that blacks experienced more complications than whites did3.
The origin of differential outcomes of chronic diseases across ethnicity is likely to be multifaceted, comprising aspects of poverty, access to adequate health care and health information, psychosocial stress, cultural overtones, lifestyle behavior, and the community environment. Given the special relevance of access to quality health care in the local community during convalescence, we hypothesized that similar ethnic disparities seen in association with other chronic medical conditions would be apparent in the recovery of function following a fracture of the distal part of the radius.
A retrospective review of a prospectively collected database of patients with distal radial fractures was performed at our university-based hospital system between 2005 and 2007. The inclusion criteria were a fracture of the distal part of the radius, skeletal maturity, and the ability to obtain informed consent for the fracture registry. All patients who either presented to the emergency department and were referred to the orthopaedic surgeon on call or presented to the orthopaedic outpatient clinic with a distal radial fracture provided informed consent and were enrolled into this institutional review board-approved database. All patients underwent initial clinical and radiographic evaluation, followed by closed reduction and sugar tong splinting. A complete history and physical examination was performed for each patient, and the presence of open wounds and the neurovascular status were recorded. A standard radiographic series was made at the time of the initial presentation, including anteroposterior, lateral, and oblique views of both the injured wrist and the contralateral wrist. Fractures were classified according to the system of the Orthopaedic Trauma Association (OTA)4. Open wounds were classified according to the system of Gustilo et al.5. Trained personnel recorded baseline demographic data, medical and social history, hand dominance, injury-related information, and a baseline Disabilities of Arm, Shoulder and Hand (DASH) functional scale score6.
Stable, nondisplaced fractures were treated nonoperatively with a splint or cast. All displaced fractures were treated with closed reduction and application of a splint; a series of radiographs of the wrist was made after the reduction. Radial inclination, height, tilt, ulnar variance, and articular step-off were measured on each radiograph. Patients who met the criteria for closed treatment7,8 were discharged to follow-up in the outpatient setting within one week after presentation and were reexamined both clinically and radiographically to assess maintenance of the reduction. These criteria included <10° of residual dorsal angulation (from neutral), <2 mm of loss of height in comparison with the contralateral side, =1 mm of articular step-off, and no associated distal radioulnar joint instability. If reduction was maintained, the patient was followed weekly for three weeks with plain radiographs to monitor the maintenance of reduction7.
Patients who declined surgery were also managed with casting. Patients who underwent surgery were managed either with a volar locked plate construct or with bridging external fixation at the discretion of the surgeon. All surgical procedures were performed by three fellowship-trained orthopaedic fracture surgeons (including one of the authors [K.A.E.]) and one fellowship-trained hand surgeon, all of whom had at least ten years of clinical experience. At the time of surgery, the American Society of Anesthesiologists (ASA) classification9 of operative risk, tourniquet time, and the duration of surgery were recorded.
Patients who were managed with external fixation underwent closed reduction with the application of two pins in the base of the second metacarpal and two pins in the distal third of the radius. All pins were placed in an open manner. Following application of the external fixator (EBI Medical Systems, Parsippany, New Jersey; or Stryker, Mahwah, New Jersey), if acceptable alignment was achieved, several percutaneous Kirschner wires were placed to hold the reduction. If an acceptable reduction could not be achieved, biplanar Kirschner wires were used to manipulate fracture fragments, or one or several small incisions were made to facilitate reduction of the fracture fragments anatomically and then Kirschner wire fixation was utilized. At the end of the procedure, all traction was released from the frame and the fixator was left in place as a neutralization device7,10-12. An extended flexor carpi ulnaris approach was utilized for the patients who were managed with a volar locked plate13,14 (Hand Innovations; DePuy Orthopaedics, Warsaw, Indiana). The number of screws that were utilized was dictated by the fracture pattern and surgeon preference. Open procedures were performed under tourniquet control, whereas external fixation was not.
After operative treatment, the patient was managed with a plaster volar wrist splint with the fingers free. This splint was removed after one week, at which time the patient was fitted with a removable splint and was enrolled in an occupational therapy program under the supervision of a certified hand therapist. The standard therapy protocol consisted of early wrist, finger, and forearm range of motion supplemented by a home exercise program.
All patients were seen at standard follow-up intervals of twelve, twenty-four, and fifty-two weeks. At each visit, the ranges of motion of the wrist and the digits were measured with a hand-held goniometer. Pain was estimated on a visual analog scale from 1 to 10. Dorsal/volar tilt, radial inclination, radial length, articular step-off, and ulnar variance were measured on plain radiographs by the treating surgeon or a trained research associate. Complications and the number of physical therapy sessions attended were recorded.
The DASH survey6 was used to measure physical function, and a visual analog score (recorded in millimeters) was used to quantitate pain.
Statistical Analysis
Continuous t tests and categorical Fisher exact tests were utilized to perform the statistical analysis of data collected at the three and six-month follow-up intervals. Mean scores for the outcomes of interest (pain and physical function), adjusted for baseline measures, are presented with Bonferroni-adjusted multiple comparisons. Multiple linear regression was used to model each of the two outcomes according to ethnicity. These models controlled for age, sex, mechanism of injury, level of education, type of fracture (according to the OTA classification system), type of treatment, Workers’ Compensation status, and baseline outcomes. The level of significance was set at 0.05.
Source of Funding
There was no outside source of funding for this project.
Four hundred and ninety-six individuals were managed for a fracture of the distal part of the radius over a three-year period. One hundred patients were black, 143 were Latino, and 253 were white. The mean body mass index was higher and age was lower among blacks (p < 0.05) and Latinos (p < 0.01) relative to whites (Table I). Blacks (p < 0.01) and Latinos (p < 0.001) had lower education levels in comparison with whites, whereas Latinos demonstrated significantly lower education levels in comparison with blacks (p < 0.05). The proportion of women was lower among blacks (64.0%) and Latinos (54.2%) than among whites (73.2%) (p < 0.01 for both comparisons). The 10% difference between blacks and Latinos in terms of the proportion of women was also significant (p < 0.05). There were no significant differences across ethnicities in terms of the mechanism of injury (percentage of high-velocity trauma) or the fracture type (OTA classification). The overall rate of follow-up was 77% by one year, which was equally distributed across the ethnicity groups (78% for blacks, 73% for Latinos, and 75% for whites; p = 0.46).
Figure 1 illustrates the mean DASH scores over time, controlled for baseline function. Blacks and Latinos consistently experienced worse function than whites throughout the follow-up period, with one exception: blacks had markedly better function at one year than whites or Latinos. This function was also a significant improvement over the function experienced by blacks at six months (p < 0.05). Figure 2 shows that the pain score as measured with the visual analog scale did not improve in blacks or Latinos, whereas whites showed some improvement in the pain score over time. Pain was substantially worse in Latinos relative to whites and blacks at all time points (p < 0.001 at three, six, and twelve months).
Table II shows the differences in outcomes between blacks and whites and between Latinos and whites while controlling for age, sex, years of education, body mass index, fracture type, mechanism of injury, treatment method, and Workers’ Compensation status. The adjusted pain score on the visual analog scale was significantly greater among Latinos (regression coefficient = 1.5; 95% confidence interval, 0.73 to 2.38) but not among blacks. For DASH scores, adjusted function was significantly worse among Latinos (regression coefficient = 14.6; 95% confidence interval, 7.9 to 21.3) but not among blacks.
Age was not associated with worse pain or function. Women experienced greater pain than men by 1 point on the visual analog scale (95% confidence interval, 0.17 to 1.9) but did not show a significant difference in function. Increasing body mass index was significantly associated with greater pain (regression coefficient = 0.09; 95% confidence interval, 0.01 to 0.22) but not function. Workers’ Compensation claimants demonstrated significantly more pain (regression coefficient = 3.1; 95% confidence interval, 2.15 to 4.1) and worse function (regression coefficient = 16.5; 95% confidence interval, 8.7 to 24.3). There were significant differences with regard to the proportion of work-related injuries across ethnicities. Six percent of whites, 26% of blacks, and 14% of Latinos reported that the injury was work-related (p < 0.001). Both blacks (79%) and Latinos (79%) were also more likely to have a Workers’ Compensation claim if the injury was work-related. As noted above, the significant differences in function and pain remained after controlling for Workers’ Compensation status.
Overall, there were no differences in operative as opposed to nonoperative treatment across ethnicities (p = 0.42), nor were there differences in external fixation as opposed to open reduction and internal fixation across ethnicity groups for those who received operative care (p = 0.31). When the results were stratified according to OTA fracture classification, some disparities did emerge. Among patients with type-A fractures, whites (40%), blacks (40%), and Latinos (43%) demonstrated very little difference (p = 0.98), with all equally likely to receive operative intervention. Among patients with type-B fractures, 71% of whites, 24% of blacks, and 63% of Latinos received operative treatment (p = 0.004), whereas among patients with type-C fractures, 65% of whites, 70% of blacks, and 84% of Latinos received operative treatment (p = 0.10). When we considered the method of operative treatment, we found ethnic differences among patients with type-C fractures only, with 70% of whites, 71% of blacks, and 43% of Latinos receiving open reduction and internal fixation (p = 0.02). Controlling for fracture type did not diminish the significant differences in function and pain across ethnicities.
There were some radiographic differences, but the differences were not consistent. Whites showed significant improvement in volar tilt relative to blacks (p = 0.01) at one year, but there were no differences between whites and Latinos or between blacks and Latinos. Radial inclination was improved in whites relative to Latinos (p = 0.04) at one year, but there were no differences between whites and blacks or between blacks and Latinos. Finally, there was significant improvement in ulnar variance in whites relative to Latinos (p = 0.003) and in blacks relative to Latinos (p < 0.001), but there was no difference between whites and blacks.
Finally, blacks had a mean of 50.0 physical therapy sessions by twelve months and whites had 40.8 sessions by twelve months, whereas Latinos participated in a mean of only 34.2 sessions by twelve months. These differences were significant (p = 0.02). These findings may relate to access-to-care issues within the community.
The present report provides clinically based evidence of ethnic differences in recovery following a fracture of the distal part of the radius. In general, both blacks and Latinos exhibited poorer function and greater pain than whites did at most follow-up points. After adjusting for sociodemographic factors and injury characteristics, the disparities for Latinos persisted for both outcomes, whereas the outcome disparities for blacks were no longer significant.
To our knowledge, this is the first report to describe ethnic inequities in recovery following a fracture of the distal part of the radius. More than twenty years ago, Furstenberg and Mezey reported significantly more frequent complications following hip fracture in black patients relative to white patients3. However, that study did not examine differences in functional outcomes. Another, more recent, study examined physical and psychosocial function in patients with severe lower limb trauma2. The investigators found that patients who were of "non-white race" were three times more likely to experience poor physical function and were more than twice as likely to experience poor psychosocial function as measured with the Sickness Impact Profile.
Studies examining potential causes of differential recovery following any fracture are equally sparse. In a study of hip fractures that was published more than a decade ago, investigators evaluated the initiation and intensity of physical and occupational therapy following fracture treatment according to ethnicity15. They found that 63% of blacks and 43% of "non-blacks" received lower-intensity rehabilitative therapy after adjusting for clinical characteristics. We were not able to assess the quality of physical therapy in our patient population; however, we were able to determine the quantity. Differential access to adequate rehabilitation care following fracture treatment is one mechanism by which worse function and greater pain could manifest in blacks. Our study did not have the necessary data to address this issue directly, nor can we validly interpolate from the findings of other studies given the differences in outcomes assessment and clinical populations.
We examined the differences in treatment by fracture type and differences in Workers’ Compensation claim status across ethnicities and found significant differences in both. Nevertheless, when we adjusted our outcomes models for both of these potential confounders, the ethnic disparities remained, even in the presence of a strong Workers’ Compensation effect. Our data suggest that the change in function over time may reflect differences in the number of physical therapy sessions attended. For example, blacks and Latinos demonstrated markedly worse function than whites did at three and six months. Blacks and Latinos had roughly the same number of physical therapy sessions in the first year. At six and twelve months, blacks were attending 32% more sessions than whites per year, and both blacks and whites were attending more sessions than Latinos, who were attending roughly the same number of sessions that they had attended at three months, despite having markedly poorer function than whites and blacks.
Education, body mass index, age, and sex may in part impact the differential outcomes between certain ethnic groups. We identified significant differences in terms of education and body mass index between whites and blacks and between whites and Latinos, with white patients having more education and lower body mass indices. Both of these factors would be identified with more favorable social and personal circumstances for recovery. This potential mediation is reflected in the attenuation of outcomes disparity between blacks and whites after multivariable adjustment. However, the same was not true for Latinos, for whom the disparities persisted following adjustment for the same sociodemographic factors. Therefore, there are likely to be additional factors that are not captured in the present study that account for the poorer function and greater pain experienced by Latinos.
The present study was limited by a lack of additional demographic and health-care-access data, which would have allowed us to make more focused statements regarding the mediating factors between ethnicity and fracture outcomes. Similarly, we lacked more comprehensive data on lifestyle and social context, which would have allowed us to identify further interaction between the factors and the fracture outcomes. As such, we are limited in the extent to which we can say why these disparities exist. Adjustment for basic sociodemographic factors reduced the differential between blacks and whites but did little toward reducing that between Latinos and whites. Nevertheless, these data help us to recognize some important factors that may put blacks at higher risk for poor outcomes following a distal radial fracture. Another limitation is that it is possible that some distal radial fractures in patients who presented to our institution were missed. The present study was also limited to patients who had one year of follow-up. Patients who were lost to follow-up may have markedly better outcomes, which could result in our retaining only the patients in the worst condition. An unequal distribution of attrition across ethnicity could have resulted in spurious conclusions. However, our attrition rates were acceptable (the overall rate of follow-up was 77% by one year), and the rates of follow-up were equally distributed across ethnicities (78% for blacks, 73% for Latinos, and 75% for whites; p = 0.46).
In conclusion, the findings of the present study suggest that convalescence following a fracture of the distal part of the radius is influenced by ethnicity. Furthermore, although fracture type and employment characteristics did explain some of the poorer outcomes experienced by blacks and Latinos, those characteristics did not account for all of the ethnic disparity that was present in our patient population.
Agency for Healthcare Research and Quality. National healthcare disparities report, 2005—full report. 2005. .
Castillo
RC;
Bosse
MJ;
MacKenzie
EJ;
Patterson
BM; LEAP Study Group. Impact of smoking on fracture healing and risk of complications in limb-threatening open tibia fractures. J Orthop Trauma.
2005;19:151-7.[PubMed][CrossRef]
Furstenberg
AL;
Mezey
MD. Differences in outcome between black and white elderly hip fracture patients. J Chronic Dis.
1987;40:931-8.[PubMed][CrossRef]
Fracture and dislocation compendium. Orthopaedic Trauma Association Committee for Coding and Classification. J Orthop Trauma.
1996;10(
Suppl 1):, 1-154.[PubMed]
Gustilo
RB;
Anderson
JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am.
1976;58:453-8.[PubMed]
Hudak
PL;
Amadio
PC;
Bombardier
C. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG). Am J Ind Med.
1996;29:602-8.[PubMed][CrossRef]
Egol
K;
Walsh
M;
Tejwani
N;
McLaurin
T;
Wynn
C;
Paksima
N. Bridging external fixation and supplementary Kirschner-wire fixation versus volar locked plating for unstable fractures of the distal radius: a randomised, prospective trial. J Bone Joint Surg Br.
2008;90:1214-21.[PubMed][CrossRef]
Lafontaine
M;
Hardy
D;
Delince
P. Stability assessment of distal radius fractures. Injury.
1989;20:208-10.[PubMed][CrossRef]
Wolters
U;
Wolf
T;
Stützer
H;
Schröder
T. ASA classification and perioperative variables as predictors of postoperative outcome. Br J Anaesth.
1996;77:217-22.[PubMed]
Cooney
WP
3rd;
Linscheid
RL;
Dobyns
JH. External pin fixation for unstable Colles’ fractures. J Bone Joint Surg Am.
1979;61:840-5.[PubMed]
Dicpinigaitis
P;
Wolinsky
P;
Hiebert
R;
Egol
K;
Koval
K;
Tejwani
N. Can external fixation maintain reduction after distal radius fractures?J Trauma.
2004;57:845-50.[PubMed][CrossRef]
Egol
KA;
Paksima
N;
Puopolo
S;
Klugman
J;
Hiebert
R;
Koval
KJ. Treatment of external fixation pins about the wrist: a prospective, randomized trial. J Bone Joint Surg Am.
2006;88:349-54.[PubMed][CrossRef]
Orbay
J;
Badia
A;
Khoury
RK;
Gonzalez
E;
Indriago
I. Volar fixed-angle fixation of distal radius fractures: the DVR plate. Tech Hand Up Extrem Surg.
2004;8:142-8.[PubMed][CrossRef]
Orbay
JL;
Touhami
A;
Orbay
C. Fixed angle fixation of distal radius fractures through a minimally invasive approach. Tech Hand Up Extrem Surg.
2005;9:142-8.[PubMed][CrossRef]
Hoenig
H;
Rubenstein
L;
Kahn
K. Rehabilitation after hip fracture—equal opportunity for all?Arch Phys Med Rehabil.
1996;77:58-63.[PubMed] [CrossRef]