Trends in the Incidence and Treatment of Distal Radius Fractures in the United States in Privately Insured and Medicare Advantage Enrollees

Abstract

Background: This study aims to determine the trends in the treatment of distal radius fractures (DRFs) in patients aged 18 years and older. Methods: An administrative claims database of more than 100 million patients was used to identify patients aged 18 years and older with a DRF between 2005 and 2014. A total of 137 130 DRFs were identified in 135 128 patients. Results: The proportion and rate of fractures were more predominant in those aged 55 years and older compared with a decreasing incidence in patients younger than 55 years. Age-adjusted rates of surgical treatment have significantly increased in both women and men by 15.9% (absolute change, 4.8%) and 5.0% (absolute change, 1.7%) change over time, respectively. Conversely, age-adjusted rates of nonsurgical treatment have significantly decreased overtime in both women and men by 6.9% and 2.6%, respectively. Conclusions: These data provide better understanding of the epidemiology of DRF, which is important to develop preventive strategies targeting high-risk populations and to develop effective treatment strategies.

Introduction

Distal radius fractures (DRFs) are very common and represent a major expenditure to health care. They account for approximately one-sixth of all fractures, with more than 640 000 emergency department visits and an estimated treatment cost of over $2 billion annually in the United States.^1-4 Their incidence has been increasing in recent years, but there are controversies with respect to incidence trends in selected population subgroups. Notably, epidemiologic studies to date are mostly focused on trends in the elderly.^5-7 In contrast, there is little data on the incidence trends in younger adults.

Distal radius fractures typically present in a bimodal distribution, with higher rates in children and the elderly. White women above the age of 65 years sustain the highest incidence secondary to postmenopausal osteoporosis. ³ Treatment options range from nonoperative measures to a myriad of surgical options that include closed reduction and percutaneous pinning, external fixation, and open reduction internal fixation (ORIF). ⁸ Controversy exists about which treatment option is optimal, and there is no clear evidence supporting one treatment over the other. ¹ Recent studies within the Medicare population have shown that the treatment of DRF has risen over the previous decades with geographic variations in the rates of surgical treatment.^5,9,10 Chung et al reported that most DRFs were treated nonsurgically in Medicare patients between 1996 and 2005 despite the introduction of volar locking plates in the year 2000. The use of ORIF has increased while closed treatment has decreased in the same time period. These data, however, are lacking in the US population after the year 2000. In addition, there is a paucity of data in the management of DRFs in the non-Medicare elderly patient population (also known as Medicare Advantage), which represents up to 33% of the elderly population in the United States as of 2017. ¹¹ It is unknown whether this population is similar to or different from the rest of the Medicare population regarding DRF and access to care.

The objective of this study was to determine the trends in the incidence and treatment of DRFs in privately insured and Medicare Advantage enrollees within the adult population over a 10-year period (2005-2014). Given the initiatives targeting fragility fractures over the last decade, we hypothesize that there is no increase in DRF trends in both elderly male and female population.

Materials and Methods

OptumLabs Data Warehouse comprises administrative claims data from privately insured and Medicare Advantage enrollees throughout the United States. ¹² It contains information on more than 100 million enrollees from geographically diverse regions across the United States, with greatest representation from the South and Midwest. The plan provides fully insured coverage for inpatient, outpatient, and pharmacy services. Medical claims include International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes; International Classification of Diseases, Ninth Revision (ICD-9) procedure codes; Current Procedural Terminology, Version 4 (CPT-4) procedure codes; Healthcare Common Procedure Coding System procedure codes; site of service codes; and provider specialty codes. ¹³ This study involved analysis of preexisting, deidentified data and was deemed exempt from institutional review board approval.

As we have previously described, ¹⁴ all claims with a primary or secondary DRF diagnosis and treatment code for patients aged 18 years or older with or without a distal ulnar fracture between 2005 and 2014 were reviewed (Supplementary Appendix 1). The ICD-9-CM and CPT procedure codes and the ICD-9-CM diagnosis codes were used to identify the cohort. Fracture management was divided into nonsurgical (closed reduction) or surgical (closed reduction and percutaneous pinning, external fixation, and ORIF). If the patient had both nonsurgical and surgical treatments, it was recorded as a surgical procedure. Patients were also required to have continuous medical coverage for at least 30 days prior to the diagnosis. Similarly, our denominator for rate calculations also included enrollees with medical coverage for at least 30 days and age ≥18 years in the given calendar year. The algorithm used for selection of the study cohort is presented in Supplemental Figure 1. In a sensitivity analysis, we included patients without requiring the presence of a CPT procedure code.

Analysis

Patient demographics (age, sex, and geographic region) were reported using mean (SD) or count (percentage) as appropriate. Annual age-, sex-specific and overall age-sex-adjusted rates were calculated and adjusted to the 2014 population. In the sensitivity analysis, trends were adjusted to each year’s US population. Rates were expressed as numbers of DRFs per 100 000 enrollees. For the analysis of the types of treatment, age-adjusted rates were expressed as the number of procedures per 100 fractures. We performed linear regression analysis to assess for trends across study time frame 2005-2014 and reported P values for trend (trend analysis). All significance tests were 2-sided, and P < .05 was considered statistically significant.

Results

Between 2005 and 2014, we identified a total of 137 130 DRFs (135 128 unique patients). The mean (SD) age at the time of fracture increased from 48.7 (15.8) in 2005 to 57.3 (18.2) in 2014 (Table 1). Approximately two-thirds of the cohort were women (69.5%). The percentage of female patients increased from 64.1% in 2005 to 73.3% in 2014. The patients were primarily from the South (42.7%), followed by the Midwest (30.7%), West (13.7%), and Northeast (12.8%). Most fractures were treated nonsurgically (65.4%).

Table 1.

Unique Episode of Treatment Characteristics (N = 137 130).

Age	Years	Treatment type	N (%)
Mean (SD)	53.3 (17.5)	Nonsurgical	89 656 (65.4)
Median	55	Surgical	47 474 (34.6)
Q1, Q3	42.0, 65.0	Surgical treatments	N (%)
Range	(18.0-88.0)	Open	39 201 (82.6%)
Age group, y	N (%)	External fixation	3929 (8.3%)
18-24	11 545 (8.4%)	Percutaneous pinning	8319 (17.5%)
25-34	11 463 (8.4%)	Year	N
35-44	16 846 (12.3%)	2005	10 989 (8.0%)
45-54	26 541 (19.4%)	2006	11 613 (8.5%)
55-64	36 056 (26.3%)	2007	12 878 (9.4%)
65+	34 679 (25.3%)	2008	12 954 (9.4%)
Sex	N (%)	2009	13 152 (9.6%)
Female	95 298 (69.5%)	2010	14 054 (10.2%)
Male	41 832 (30.5%)	2011	14 783 (10.8%)
Patient’s region	N (%)	2012	14 878 (10.8%)
Midwest	42 102 (30.7%)	2013	16,439 (12.0%)
Northeast	17 572 (12.8%)	2014	15,390 (11.2%)
South	58 509 (42.7%)
West	18 826 (13.7%)
Unknown	121 (0.1%)

Age-specific analyses of rates of DRF per 100 000 enrollees demonstrated that those older than 55 years experienced an increased incidence of DRF. The rates increased from 2005 to 2014 by 3.4% (P = .47 for trend) in the age group 55-64 years and by 17.7% (P = .01) in the age group 65+ years. In contrast, there had been a decline in rates in patients younger than 55 years (Table 2; Figure 1). Age-adjusted analyses demonstrated that the rates significantly decreased by 16.3% from 2005 to 2014 in the male cohort (P < .01), and there was no change in the female cohort (5.4% increase, P = .16). The overall age- and sex-adjusted fracture rates remained constant over the study time frame (1.7% decrease, P = .99).

Table 2.

Rates of Distal Radius Fractures per 100 000 Enrollees.

Age-specific	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	P value	% Change a
Age-specific
18-24	52.7	57.5	60.6	56.0	62.1	58.1	56.2	54.8	52.3	47.9	.15	−9.10
25-34	38.7	36.9	39.3	38.6	36.4	36.6	34.2	32.6	32.1	30.0	<.01	−22.50
35-44	55.7	51.4	54.6	51.5	50.0	49.9	50.6	42.6	44.5	41.1	<.01	−26.20
45-54	75.9	73.8	82.7	82.2	77.4	80.4	76.4	69.7	74.4	68.2	.1	−10.10
55-64	126.1	127.1	136.7	136.8	136.7	141.0	139.8	126.6	138.6	130.4	.47	3.40
65+	201.4	181.5	229.1	238.4	232.4	236.4	244.3	239.7	243.6	237.0	.01	17.70
Age-adjusted women	116.6	109.9	127.8	130.2	128.6	131.1	130.8	123.4	130.9	122.9	.16	5.40
Age-adjusted men	56.4	55.6	59.8	57.3	55.4	55.4	54.8	50.2	49.1	47.2	<.01	−16.30
Age-sex-adjusted	86.9	83.1	94.2	94.1	92.4	93.6	93.1	87.2	90.4	85.4	.99	−1.70

^aPercentage change from 2005 to 2014.

Figure 1.

Age-specific rates of distal radius fractures per 100 000 enrollees.

In a sensitivity analysis without procedure codes, we identified 224 506 patients with DRF. The findings of this sensitivity analysis were in agreement with the primary analysis; age-specific trends and age-adjusted trends in men were similar regarding direction and statistical significance. Age-adjusted trends in women and overall age- and sex-adjusted trends have become statistically significant. This suggests that the trends over time are significant; it does not suggest any differences between age-sex groups. The point estimate for the yearly rates has nearly increased by a factor of 2 (Supplementary Appendix 2).

The proportion of patients receiving nonsurgical treatment had declined from 66.7% in 2005 to 64.1% in 2014. The proportion of patients undergoing surgical treatment had increased from 33.3% in 2005 to 35.9% in 2014. Open reduction internal fixation remained the most common surgical treatment modality across the study years and has increased from 22.7% in 2005 to 32.2% in 2014. Percutaneous pinning has declined from 11.1% in 2005 to 3.5% in 2014. Similarly, external fixation has declined from 4.5% in 2005 to 2.1% in 2014 (Table 3). The annual percentages of treatment modalities for those above and below 65 years are shown in Figure 2.

Table 3.

Proportions of Surgical and Nonsurgical Treatment of Distal Radius Fractures.

	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014
Nonsurgical	66.7	67.3	66.3	66.1	66.3	64.6	64.4	64.8	64.3	64.1
Surgical	33.3	32.7	33.7	33.9	33.7	35.4	35.6	35.2	35.7	35.9
Treatment type
Closed	66.7	67.3	66.3	66.1	66.3	64.6	64.4	64.8	64.3	64.1
Surgical: Open	22.7	23.9	26.2	27.3	27.7	29.6	30.6	30.8	31.5	32.2
Surgical: External fixation	4.5	3.7	3.4	3.2	2.8	2.8	2.5	2.3	2.2	2.1
Surgical: Percutaneous pinning	11.1	9.1	7.8	6.6	5.9	5.7	5.0	4.5	4.0	3.5

Figure 2.

Annual percentages of treatment modalities of distal radius fractures.

Age-adjusted analyses of rates of surgical treatment per 100 fractures showed that the rates significantly increased by 15.9% in women and by 5.0% in men (P < .01). Conversely, age-adjusted rates of nonsurgical treatment have significantly decreased over the study time frame by 6.9% in women and by 2.6% in men (P < .01) (Supplemental Table 1; Figure 3).

Figure 3.

Age-adjusted rates of distal radius fracture treatment per 100 fractures.

Upon further stratifying the surgically treated patients by the type of procedure performed, age-adjusted rates of ORIF per 100 fractures significantly increased by 43.9% in women and by 33.7% in men (P < .01). In contrast, age-adjusted rates of external fixation (women: −66.4%, men: −41.9%; P < .01) and percutaneous pinning (women: −74.8%, men: −72.4%; P < .01) have declined (Table 4; Figure 3). These estimates were calculated by including only patients who underwent operative management in the denominators.

Table 4.

Adjusted Rates of Open Reduction, External Fixation, and Percutaneous Pinning Treatment per 100 Surgical Fractures.

	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	P value	Absolute % change a	Relative % change a
Open reduction internal fixation
Age-adjusted women	62.2	69.3	73.3	76.9	79.3	82.1	85.1	86.9	88.0	89.5	<.01	27.3	43.9
Age-adjusted men	67.0	73.4	78.8	79.9	85.9	86.1	86.8	86.9	89.5	89.6	<.01	22.6	33.7
Age-sex-adjusted	64.9	71.2	75.6	78.6	81.1	83.2	85.5	87.3	88.2	89.8	<.01	24.9	38.4
External fixation
Age-adjusted women	14.9	10.8	10.1	9.7	8.4	7.5	6.4	6.1	5.3	5.0	<.01	−9.9	−66.4
Age-adjusted men	13.6	14.2	11.6	12.6	9.2	9.0	9.9	9.1	7.8	7.9	<.01	−5.7	−41.9
Age-sex-adjusted	14.6	11.5	10.4	9.8	8.5	8.0	7.1	6.5	6.2	5.8	<.01	−8.8	−60.3
Percutaneous pinning
Age-adjusted women	39.7	31.7	27.2	22.0	20.0	17.6	15.0	13.5	11.3	10.0	<.01	−29.7	−74.8
Age-adjusted men	34.1	27.2	21.2	20.5	13.3	13.8	12.9	11.8	9.9	9.4	<.01	−24.7	−72.4
Age-sex-adjusted	36.8	29.7	25.2	20.9	18.4	16.7	14.4	13.0	11.2	9.7	<.01	−27.1	−73.6

^aPercentage change from 2005 to 2014.

Discussion

Distal radius fractures are common injuries that occur in a bimodal distribution, namely in children and the elderly population. Osteoporosis accounts for most fractures in the latter group, whereas high-energy trauma accounts for most fractures in the younger population. ¹⁵ This results in over $2 billion of medical costs each year in their management and sequela. ¹⁵

Within this study, there was an increasing rate of DRFs in patients aged 55 years and older. The etiology of this could be multifactorial and may be driven by age-related osteoporotic changes and the patient’s increased level of physical activity. Indeed, in 2014, patients aged 65 years and older represented 36.4% of the entire DRF population compared with only 13.3% in 2005. It should be noted, however, that it is uncertain whether our findings are generalizable to the underinsured, noninsured, or general Medicare population. A previous study comparing trends in forearm fractures between 1989-1991 and 2009-2010 in Olmsted county (Rochester, Minnesota) showed decreasing rates of fractures among middle-age women and increased rates in elderly men, which is similar to the findings of this study. ¹⁶ The large sample size of this study enabled us to demonstrate the changing trends year by year over the past decade for all age groups.

Age-adjusted analyses demonstrated significantly decreased incidence of DRF in men compared with an increase in women, especially within the age group 55 years and older. This finding highlights the important role of sex in driving the rates of DRF and may be related to the preventive measures implemented^17,18 for fragility fractures in the male population over the past decade. Taking age-specific and age-adjusted analyses together, the elderly female population remains at the highest risk of DRFs. In fact, this group has demonstrated the highest increase in rates of DRF, which has increased by 17.8% (P = .01) despite the implemented national preventive measures, including “own the bone,” advocated by the American Orthopaedic Association. ¹⁸ This finding can be used to support ongoing resource allocation to prevent fragility fractures in this particular group. This is certainly important because these trends appear to have changed from the previous decades; a previous study showed increasing rates of DRF in men and no statistically significant change among the female population between 1945 and 1994. ¹⁹ It should also be noted that the increase in DRF may be partially driven by the aging population of the United States; however, our estimates were reported in terms of ratios as opposed to absolute numbers, and we conducted age-stratified analyses to isolate the impact of the aging population for each age group.

In this study, the average annual incidence of DRF in women and men was 112.6 and 52.7 per 100 000 enrollees, respectively, resulting in a female-to-male incidence ratio of 2.3:1. This ratio is consistent with a previous study based on the Swedish population which reported a ratio of 2.7:1, despite a higher annual incidence. ²⁰ This discrepancy may be accounted for by the fact that in the primary analysis, we only included patients with a documented treatment modality (surgical or nonsurgical) for their DRF and also for the inherent differences in racial composition and other demographics between the 2 populations.

The percentage of fractures treated with ORIF has increased over the years. At the same time, the percentage of fractures treated with percutaneous pinning or external fixation is decreasing (Table 3). When examining age-sex-adjusted rates, surgically treated fractures significantly increased by 14.0% (P < .01). In contrast, the age-sex-adjusted rate of nonsurgically treated fractures has decreased by 6.4% (P < .01) (Supplemental Table 1). Percutaneous pinning likely decreased in the time period 2005-2014 as the use of volar locking plate became increasingly popular. This differential trend in the type of surgical treatment is consistent with previous studies performed in Finnish and Swedish populations.^20-22 Wilcke et al also reported that 73% of adult patients were treated nonsurgically in 2010. In this study, we found that 65% of patients were treated nonsurgically in 2010. Despite the increasing trend toward surgical treatment, recent meta-analyses have shown that there has been no improvement in outcomes (Disabilities of the Arm, Shoulder, and Hand; function) between nonoperative treatment and surgical fixation in the management of DRFs in the elderly.^23,24 In a randomized controlled trial, Arora et al compared DRF outcomes between patients treated nonoperatively and with ORIF with a volar locking plate in patients aged 65 years or older. At 12-month follow-up, the 2 groups showed similar range of motion, level of pain, and functional wrist scores. Grip strength was better in the operatively treated cohort. The authors concluded that achieving anatomical reduction does not improve the ability to perform daily live activities in their cohort. ²⁵

These findings, related to the differential trends in the type of surgical treatment, seem noteworthy given that recent randomized controlled trails have demonstrated no significant difference in functional outcomes between patients treated with ORIF (with a volar locking plate) and percutaneous pinning (Krischner wires) or external fixation.^26,27 Tubeuf et al ²⁸ demonstrated significant cost savings have been shown when using percutaneous pinning compared with ORIF using a volar locking plate. It should be noted that in the light of these studies, a change in clinical practice has been observed in the United Kingdom. ²⁹ Bridging the gap between research and practice should be emphasized by continuing professional education and addressing the misconceptions associated with practicing evidence-based medicine. ³⁰

The findings of this study regarding the trends in DRF treatment among the elderly cohort appear to be consistent with previous US-based studies which demonstrated increasing rates of ORIF and decreasing rates of external fixation over time.^5,7 Chung et al reported that closed treatment remained the predominant method; however, it had declined from 82% in 1996 to 70% in 2005. The authors also reported very little variation in the treatment modality exists according to sex. As such, commercially insured and Medicare Advantage populations appear to experience a similar change in trends compared with other Medicare categories. More recent studies are warranted to further characterize the differences between these populations.

There are several limitations to this study. Databases are driven by the accuracy of the coding documentation. We attempted to mitigate this by grouping the diagnosis and treatment codes of DRF. There is a lack of granularity within database analyses; for instance, there were no details available regarding the bilaterality of the fractures. Given that the database used comprises commercially insured and Medicare Advantage enrollees, the results obtained may not be applicable to other patient populations. It is known that accounting for the true number of DRFs may be difficult due to the fragmented nature of care and coding discrepancy, which likely underestimate the typically quoted rates in the literature. ⁴ In addition, we were not able to account for clustering of patients among providers. As such, the analysis does not necessarily account for the variation in fracture management attributable to surgeons, such as experience and preferences. Moreover, our analysis did not account for fracture severity based on CPT coding. This could be a focus of a subsequent study to investigate the changes in fracture incidence and treatment type based on fracture severity. Despite limitations, this large study with more than 135 000 patients over a 10-year period from all over the United States provides comprehensive determination of trends in DRF among all age groups and both sexes.

The rates of DRF have been increasing in the elderly population and decreasing in young adults over the past decade and may reflect a change in pattern compared with the previous decades. The elderly female population increasingly incurred the highest rates of DRF. Despite a lack of superior functional outcomes within the literature, the rates of ORIF continue to increase at the expense of declining external fixation, percutaneous pinning procedures, and nonoperative management.