Continuity of care has been defined as sustained care between a patient and a healthcare provider through a relationship over time.1 Traditionally, following patients through all aspects of their illness and recovery has been viewed as essential to the development of competent, caring physicians.2 Further reported benefits of continuity of care include reduced hospital admissions, decreased length of stay, reduced duplication of diagnostic testing, enhanced patient satisfaction, more knowledgeable patients, and improved adherence to selected treatment regimens.3-6 This positive association between continuity of care and quality of care has been well established for patients with chronic conditions but is less clear for patients with acute conditions.1,4 The effect of continuity of care in orthopedics, where the majority of patients have acute conditions, has never been assessed.
For acute conditions, such as fractures, continuity may be sacrificed for the sake of urgency or convenience. Furthermore, the increasing size of physician groups, availability of physician assistants (PAs) and residents, access to electronic medical records, and pressure to contain healthcare costs may be reducing the likelihood of continuity of care.7 How variability in the continuity of provider care affects patients with fractures is unknown. Being treated by multiple providers may prevent delay in treatment, allow for second opinions, and create a greater likelihood that conditions will not be overlooked, which could all improve the quality of care and patient satisfaction. If communication is poor, however, important details may not be relayed to the next provider or providers may disagree, resulting in disjointed patient care and potentially reduced quality.
Surgical treatment of forearm fractures in children has increased markedly in recent years, although clinical evidence fails to support this trend.8-11 Some researchers conclude that patient and physician convenience may be a driving force in the increasing abandonment of nonoperative treatment of these fractures.8-11
Nonoperative treatment requires multiple visits to ensure ongoing maintenance of fracture alignment. Cast changes and remanipulations may be required to promote fracture healing in an acceptable alignment. The need for multiple visits can impose stress on families, so efforts to improve family convenience often result in multiple providers caring for the child at times and locations that are convenient to the family. PAs can play an important role in minimizing these stresses to the family by increasing access to and decreasing cost of follow-up care. However, family convenience must be balanced against the safety of nonoperative management—specifically the risk of the fracture healing in unacceptable alignment (malunion). This study assessed whether the type of provider (attending physician versus PA) or number of providers involved in the nonoperative management of a pediatric forearm fracture influenced the risk of that fracture healing as a malunion.
MATERIALS AND METHODS
This study was a retrospective chart review of patients ages 3 to 17 years who were seen in the orthopedics department at Cincinnati Children's Hospital Medical Center and treated for a forearm shaft fracture from February 1, 2012, to January 31, 2013. Single- and both-bone forearm shaft fractures were included. Exclusion criteria were:
- distal radius or metaphyseal fracture
- Monteggia fracture (N = 1)
- buckle or incomplete fracture
- radius or ulna fracture with a bow in the other bone
- refracture (N = 2)
- patients with poor radiographs (N = 2)
- patients with less than 1 month of radiographic follow-up (N = 5).
The final sample size was 141 patients.
Patient charts were reviewed to confirm the diagnosis and to determine the patient's age; sex; injured side; date of injury; and date, type, and any change in treatment. The name, position, and number of billing providers who treated each patient during the 1 to 2 months of their fracture treatment were recorded. Only attending physicians and PAs are billing providers in orthopedics at our hospital. Only clinic visit providers were included, not the providers who performed the fracture reduction in the ED. Information also was gathered on who signed the progress note, to determine if a PA participated in the patient's care without being the billing provider. The total number of clinic visits for each patient was recorded as well as the number of times the billing provider switched between visits and the number of times the signing provider switched between visits.
Anteroposterior and lateral radiographs taken at the time of injury were used to determine the type of fracture (greenstick or complete), the location of the fracture (distal third, middle third, proximal third), and the direction and degree of initial fracture angulation.12 The fracture location and angulation were determined using the criteria described by Bowman and colleagues.12 If the fracture location differed between the radius and ulna, the fracture location of the radius was used. If only the radius was fractured, angulation measurements were only performed on the radius and vice versa for the ulna. The final angulation was measured at the patient's latest clinic visit (1 to 2 months after injury), and the maximum angulation across both radiographic views and both bones (if available) was used to determine whether the patient had a malunion.
Malunion was determined based on the following criteria proposed by Bowman and colleagues.12,13
In girls age 8 years and younger and boys age 10 years and younger, a maximum angulation of more than 20 degrees for distal third, more than 15 degrees for middle third, and more than 10 degrees for proximal third shaft fractures was considered a malunion. In girls older than age 8 years and boys older than age 10 years, a maximum angulation greater than 10 degrees for all shaft fractures was considered a malunion. Refractures, fracture redisplacements, and changes in treatment were recorded.
Sex, fractured bones, fracture type, and whether a patient had a change in treatment were compared between patients with and without malunion using Fisher exact tests. Age, initial angulation, and length of follow-up were compared between patients with and without malunion using logistic regression. The number of clinic visits, billing providers, attending physicians, and PAs were compared between patients with and without malunion using Cochran-Armitage trend tests. Multivariate logistic regression was used to determine whether the number of providers predicted malunion when controlling for other factors (age of patient, type or location of fracture as well as direction and degree of initial angulation). For all analyses, P < .05 indicated statistical significance.
Forearm shaft fractures were identified in 141 patients (68 boys, 73 girls). Patients averaged age 8±3 years (range, 3-17 years). Left-sided fractures occurred in 70 patients and right-sided in 71. Both radius and ulna fractures occurred in 124 patients; radius-only fractures in 16 patients and an ulna-only fracture in 1 patient. Two patients had open fractures. Greenstick fractures were identified in 91 patients and complete fractures in 50 patients. The distal third of the shaft was fractured in 39 patients, the middle third in 82, and the proximal third in 20. Initial maximum angulation averaged 23±17 degrees (range, 1-86 degrees). All patients were initially treated nonoperatively with casting, with 123 (87%) undergoing closed reduction.
Malunion was identified in 42 patients (30%). The likelihood of malunion did not differ based on whether the providers included a PA (28% versus 56%, P = .13, Fisher exact test) or on the number of PA providers (P = .11, Cochran-Armitage trend test); however, both findings were not significant, although some patients had up to four PAs involved in their care. However, patients treated by two attending providers were significantly more likely to have a malunion (7/10 or 70%) than patients treated by a single attending provider (35/131 or 27%, P = .008, Fisher exact test; RR = 2.4, 95% CI: 1.1-9.1), even after controlling for initial angulation, number of clinic visits, number of PAs, and whether the patient had a change in treatment (P = .03, multivariate logistic regression). The likelihood of malunion was not significantly associated with sex, age, length of follow-up, fractured bones, fracture type, number of clinic visits, total number of providers, or whether they had a change in treatment (P > .05, Table 1).
Fractures of the radius and ulna are the most common type of fractures in children.9 About one-third of these fractures will displace during conservative treatment, leading to a risk of malunion if this redisplacement is not addressed during the course of conservative management.14 Although multiple studies have investigated risk factors for redisplacement, which factors lead to malunion have not been fully elucidated.14-16 One potential risk factor could be that redisplacement goes unnoticed because of the lack of continuity of care in medical decision making by multiple providers along the course of treatment. This potential risk factor is becoming increasingly relevant as PAs and NPs become more integral team members in the nonoperative management of common fractures. Therefore, this study sought to determine how the use of one or more PAs in the closed management of forearm fractures was associated with the risk of ultimate malunion. We found that using PAs was not associated with an increase in the malunion risk, even when the patient was seen by up to four different PAs during the course of treatment. However, we did find that changing the attending provider during the course of treatment significantly increased the association with malunion, even when controlling for fracture location, initial angulation, number of clinic visits, number of PAs, and whether the patient had a change in treatment.
The safety and efficacy of care provided by PAs has been similarly demonstrated in multiple nonorthopedic care settings. One study found equivalent outcomes of adult Medicare patients with diabetes treated by teams that included PAs and NPs and those of patients who received care from physicians only.17 Another concluded that the quality of care provided to HIV patients by PAs and NPs who worked in HIV clinics was similar to that of physician HIV experts and actually better than physician non-HIV experts.18 The use of PAs and NPs in a Level I trauma center led to a significant reduction in ICU length of stay with no increased incidence of complications.19 Likewise, the tasks and activities performed by PAs and NPs in acute care settings are similar to those performed by resident physicians, and patient outcomes are similar for both groups.20
The unanticipated finding of this study was that a change in attending physician, and thus a change in the leadership of the treatment team, was associated with an increased risk of ultimate forearm fracture malunion. One possible explanation for this finding is that unstable fractures requiring urgent treatment are transferred to another surgeon in order to expedite the required intervention, therefore introducing a selection bias. However, a change in treatment was not associated with change in attending provider on malunion risk. Also, fractures that are not healing properly might prompt families to seek a second opinion, leading to a change in attending physician for fractures already on the road to malunion. We did not specifically assess the reasons for change in attending provider in this study, as this association was unexpected. Nonetheless, further investigation into the mechanisms of this effect is warranted.
One potential mechanism deserving further investigation is variation in treatment plans between attending physicians. Studies show that physicians display random variation among treatment of similar conditions: for example, posterior malleolar ankle fracture treatment.21 Another study showed that orthopedic surgeons agreed on treatment recommendations but not on classification of proximal humerus fractures.22 An electronic survey of Australian orthopedic surgeons showed variations in the standards of care for managing ankle, scaphoid, distal radius, humeral neck, and clavicle fractures.23 The effects of these variations on the outcomes of patients with common fractures remain to be elucidated. Nonetheless, efforts to reduce random variation among physicians, such as with the use of clinical practice guidelines, deserve ongoing attention.24,25 The finding that transitions between even multiple PAs was not associated with increased malunion risk could be explained by the tendency for PAs to act out the treatment plan prescribed by the original treating physician rather than altering the treatment plan themselves.
Importantly, this study did not evaluate clinical outcomes of the patients with and without malunions. Rather, the purpose of the study was to investigate whether a relationship exists correlating the effect of type and number of provider on objective radiographic parameters at fracture healing. We used strict criteria for malunion, leading to a relatively high rate of malunion at 30%.12 Various attending physicians might disagree with the criteria for malunion used in this study, and the clinical outcomes may be substantially better than the radiographic outcomes. The literature and providers are inconsistent about remodeling potential and acceptable alignment of forearm fractures. However, we used a consistent definition of malunion across all providers rather than relying on each treating provider to determine the malunion outcome. With this strategy, we were able to evaluate factors within the dynamics of the treatment team that led to an objective radiographic outcome. Further study is required to assess the effect of change in treatment team leaders on clinical outcomes.
Additional limitations of this study include the relatively small number of patients from a single institution, as well as the study's retrospective design. Thus only associative relationships and correlations can be identified. In addition, the difficulty in standardizing radiographs inherent in a retrospective study potentially confounded the results, because the outcome (malunion) relies solely on radiographic parameters. However, the radiographs are representative of those used during clinical decision making and were performed similarly by the same radiology department for all patients. Finally, caution must be applied when generalizing the results of this study to other institutions, as the roles and skill sets of various providers will vary among teams and systems. Next, a prospective study should be conducted with a larger sample size and a more diverse population to add further clarity to this research question. However, despite these limitations, this study supports the use of PAs in the care of pediatric forearm fractures, yet calls for further investigation regarding the safety of handoffs in such care between attending physicians.
1. Cabana MD, Jee SH. Does continuity of care improve patient outcomes. J Fam Pract
2. Hutter MM, Kellogg KC, Ferguson CM, et al The impact of the 80-hour resident workweek on surgical residents and attending surgeons. Ann Surg
3. Sudarshan M, Hanna WC, Jamal MH, et al Are Canadian general surgery residents ready for the 80-hour work week? A nationwide survey. Can J Surg
4. Gill JM, Mainous AG 3rd. The role of provider continuity in preventing hospitalizations. Arch Fam Med
5. Gill JM, Mainous AG 3rd, Nsereko M. The effect of continuity of care on emergency department use. Arch Fam Med
6. hristakis DA, Mell L, Koepsell TD, et al Association of lower continuity of care with greater risk of emergency department use and hospitalization in children. Pediatrics
7. Christakis DA. Does continuity of care matter? Yes: consistent contact with a physician improves outcomes. West J Med
8. Helenius I, Lamberg TS, Kääriäinen S, et al Operative treatment of fractures
in children is increasing. A population-based study from Finland. J Bone Joint Surg Am
9. Cheng JC, Ng BK, Ying SY, Lam PK. A 10-year study of the changes in the pattern and treatment of 6,493 fractures
. J Pediatr Orthop
10. Flynn JM, Jones KJ, Garner MR, Goebel J. Eleven years experience in the operative management of pediatric forearm fractures
. J Pediatr Orthop
11. Eismann EA, Little KJ, Kunkel ST, Cornwall R. Clinical research fails to support more aggressive management of pediatric
upper extremity fractures
. J Bone Joint Surg Am
12. Bowman EN, Mehlman CT, Lindsell CJ, Tamai J. Nonoperative treatment
of both-bone forearm
in children: predictors of early radiographic failure. J Pediatr Orthop
13. Price CT. Acceptable alignment of forearm fractures
in children: open reduction indications. J Pediatr Orthop
14. Colaris JW, Allema JH, Reijman M, et al Risk factors for the displacement of fractures
of both bones of the forearm
in children. Bone Joint J
15. McQuinn AG, Jaarsma RL. Risk factors for redisplacement of pediatric
and distal radius fractures
. J Pediatr Orthop
16. Pretell Mazzini J, Rodriguez Martin J. Paediatric forearm
and distal radius fractures
: risk factors and re-displacement—role of casting indices. Int Orthop
17. Everett C, Thorpe C, Palta M, et al Physician assistants
and nurse practitioners perform effective roles on teams caring for Medicare patients with diabetes. Health Aff (Millwood)
18. Wilson IB, Landon BE, Hirschhorn LR, et al Quality of HIV care provided by nurse practitioners, physician assistants
, and physicians. Ann Intern Med
19. Gillard JN, Szoke A, Hoff WS, et al Utilization of PAs and NPs at a level I trauma center: effects on outcomes. JAAPA
20. Rudy EB, Davidson LJ, Daly B, et al Care activities and outcomes of patients cared for by acute care nurse practitioners, physician assistants
, and resident physicians: a comparison. Am J Crit Care
21. Gardner MJ, Streubel PN, McCormick JJ, et al Surgeon practices regarding operative treatment of posterior malleolus fractures
. Foot Ankle Int
22. Brorson S, Olsen BS, Frich LH, et al Surgeons agree more on treatment recommendations than on classification of proximal humeral fractures
. BMC Musculoskelet Disord
23. Ansari U, Adie S, Harris IA, Naylor JM. Practice variation in common fracture presentations: a survey of orthopaedic surgeons. Injury
24. Paxton ES, Matzon JL, Narzikul AC, et al Agreement among ASES members on the AAOS clinical practice guidelines. Orthopedics
25. Jevsevar D, Shea K, Cummins D, et al Recent changes in the AAOS evidence-based clinical practice guidelines process. J Bone Joint Surg Am