Supracondylar fractures of the humerus (SCH) are the most common fractures about the elbow in children. It is accepted that Gartland type III fractures are managed by closed or open reduction and pinning. This results in a lower incidence of neurovascular complications when compared with closed management.1,2 Battaglia et al3 estimated the incidence of compartment syndrome in children with SCH to be 0.3% to 1%. A compromised limb from a neurovascular standpoint has long been seen as the most significant complication of this common injury. Timely identification and documentation of the presence and/or absence of neurovascular injury is important.
Documentation is crucial to good patient care. It also is of prime importance for justification or defense of our care should this arise. A common statement in litigation is that if you did not document it, you did not do it. The senior author was involved in the defense of 2 compartment syndrome litigations that occurred in conjunction with a SCH. One hospital was still using a paper record. The other was in transition to an electronic medical record (EMR).
The paper record was concise and easy to follow. The nursing notes for 24 hours were on a single sheet of paper and contained all the crucial information necessary to defend the case. Legibility was satisfactory but can be a problem in paper charts. In the transitional case, nursing notes were electronic while the physician notes were handwritten. Finding the nursing notes in the EMR was difficult and often contained multitudes of information resulting in useful clinical assessment information becoming buried in the minutiae.
Purpose of the study was not to look at outcomes of treatment. We did not propose a novel intervention for the treatment of this common pediatric injury. There are plenty of studies establishing closed reduction and percutaneous pinning as the standard of care. Our focus was on appropriate and complete documentation of the neurovascular examination in the perioperative and follow-up time frame. Long-term follow-up is not routine for this type of injury and would have been outside the norm and not provided any additional benefit for the research question at hand.
It was elected to review a year’s experience with documentation in our hospital system to assess the adequacy of documentation. Institution review board (IRB) approval was obtained. After the results were analyzed and showed less than acceptable documentation by nursing and physician staff, a clinical pathway was developed to provide forms to record all necessary documentation by both physicians and nurses. The pathway was designed by physicians, nurses, and administration as a performance improvement project. Acceptance of the pathway was obtained by the performance improvement committee. A prospective study was then designed to assess the results of the implementation of the clinical pathway. It was implemented after presentations to nursing staff, residents, and faculty. IRB approval was obtained for both studies. All patients for whom consent was obtained were included.
A literature search using PubMed and Google Scholar revealed only 2 studies that related to documentation and clinical pathways in SCH fractures. Cao et al4 performed a study before and after implementing note templates in the EMR. Their study focused on improvement of only physician documentation by establishing an EMR template. No assessment of nursing documentation was performed. Sung et al5 performed a study before and after implementation of a clinical pathway. Their primary outcomes were decreasing inpatient stay, improvement of patient flow, and improvement of care team satisfaction. Our study addresses combing the elements of both of these studies.
Documentation before and after the implementation of a clinical pathway was compared. A retrospective chart review was used to collect documentation information before the implementation of the clinical pathway, and a prospective study design was used to collect information after the implementation of the clinical pathway. The inclusion criteria for both cohorts were operatively-treated SCH fractures. Patients were excluded if they did not require a trip to the operating room (OR) or if family did not consent. Information was collected from our inpatient EMR (Meditech, Hospital Corporation of America) and our clinic EMR (Centricity EMR; General Electric South Burlington, VT) on patients with CPT codes 24538 and 24545.
A retrospective review of patient 40 charts was performed for the original cohort, all physician notes were handwritten, and all nursing notes were electronic. The original intent was for the investigators to review all electronic nursing notes looking for adequacy of documentation of nursing neurovascular status preoperative, in the postanesthesia care unit (PACU), and on the floor. We also sought documentation of parent education and follow-up instructions by the nursing staff, particularly with regard to monitoring neurovascular status, signs of compartment syndrome, and cast care instructions. Physician notes were reviewed for preoperative, postoperative, and clinic visits. Physician clinical follow-up was obtained from the physician postoperative clinic EMR notes. In the prospective cohort, all records were in the EMR and should have followed the clinical pathway.
The following were recorded: name, unit number, date of admission, age, sex, length of stay (LOS) in hours, side of injury, presence of ipsilateral forearm fracture, number of pins used, pin insertion sites, return to OR, pain status, presence of compartment syndrome, and LOS.
In the initial study, the nursing notes were voluminous and finding the data we sought was not easy. Multiple levels of nursing administration and patient care nursing were approached to aid in identification of the data we sought in the EMR. It was ultimately decided that a clinical nurse coordinator would print all nursing notes on all patients. The coordinator then went through each page highlighting any documentation of neurovascular status. We sought these 4 data points: preoperative, PACU, and 2 postoperative observations. We were then given only the pages that contained data thought to meet our criteria. A physician investigator went through all the nursing notes focusing on the highlighted portions looking for documentation of vascular status and neural function.
The discharge nursing notes and a discharge form were reviewed for patient/parent education regarding neurovascular status and signs of compartment syndrome. This was a paper form with a parent signature. The signed form was attached at the end of the paper chart, or, if following protocol, scanned into the EMR. The physician examinations were reviewed looking for adequacy of documentation of neurovascular status. Adequate documentation had to include observation of adequacy of circulation and note that median, radial, ulnar, and anterior interosseous nerves were checked.
We initially tried to document the pain status from the physician and nursing notes. Despite all our efforts to include this as part of our study, it was found to be too inconsistent to track and thus excluded from this manuscript.
After implementation of the clinical pathway, we sought to evaluate any improvement in standardized evaluations and documentation. This was a consecutive, prospective enrollment. Because of the clinical pathway, the data for the second study was easy to identify and evaluate.
Proportions were compared before and after the implementation of the clinical pathway using a χ2 test, or the Fisher exact test for measures in which at least 20% of the expected frequencies were <5. A 2-sided 0.05 α level was used to define statistical significance.
A total of 90 patients were included for the preclinical pathway retrospective cohort and 67 were included for the prospective cohort after the implementation of the clinical pathway. Because of the feasibility of obtaining inpatient nursing documentation in our EMR (transition between written and EMR), records were only available for 42 of the 90 patients. Therefore, the sample size for the nursing documentation is 42. The 48 patients without nursing records to verify were not excluded as we were still able to assess the presence of physician documentation.
A summary of the nursing documentation is seen in Table 1. We evaluated the presence of nursing documentation before and after the implementation of the clinical pathway. We identified a total of 42 preoperative and PACU observations. A total of 168 observations for the postoperative notes were reviewed. Of the 168 observations, 68 (40.5%) had no neurovascular examination. Five patients were discharged from the recovery room and had no recovery room notes showing any documented evaluation. Only 58 of 168 observations (34%) had both neural and vascular function recorded.
After stressing the importance of physician documentation with our faculty and residents and revealing the unsatisfactory results from the retrospective cohort, a focus on improved accountability with the clinical pathway led to significant improvement. Physician documentation was statistically significantly better for both preoperative (90% to 100%) and postoperative (62% to 100%) inpatient examinations. Although documentation trended towards improvement in the clinic setting (97% to 100%), it did not meet statistical significance (Table 2).
The documentation of proper parent education in the initial study was tracked by a single portion of the nursing data stating that it had been performed, but this did not reveal parental confirmation. Only 39 of 90 (43%) instances of nursing discharge documentation revealed cast care education. No formal compartment syndrome education confirmation was utilized at that time. The clinical pathway implemented a requirement for parents to sign forms for cast care and compartment syndrome education. In the process of looking at the data, it was difficult to find this information due to the requirement that the signed forms be scanned into an additional portion of our inpatient EMR (Hcare Portal). We were able to find 46 of 67 (69%) cast care forms and 15 of 67 (22%) compartment syndrome forms via the web-based database.
Both cohorts had patients who returned to the OR for lost reduction or inadequate reduction, Table 3. The preclinical cohort had 2 (2%), whereas the postclinical cohort had 4 (6%) return to the OR. LOS in the retrospective cohort was 22.1 hours (range, 9 to 55 h). In the prospective cohort, LOS was 18.9 hours (5.5 to 39 h). There were no compartment syndromes in either series.
The percentage of patients with nursing preoperative documentation was significantly higher following the implementation of the clinical pathway (97% after compared with 48% before implementation, P<0.0001). No other differences in nursing documentation were significant. It was evident that the paper forms documenting parent cast care and compartment syndrome education were either not being performed or done but not scanned into the EMR.
The percentage of patients with physician preoperative documentation was significantly higher following the implementation of the clinical pathway (100% after compared with 90% before implementation, P=0.011). The percentage with postoperative documentation also was higher following implementation of the pathway (100% after compared with 62% before implementation, P<0.0001). There was no significant difference in clinic examination documentation before and after implementation of the pathway (P=0.26).
Physician documentation is crucial for the justification/defense of adequate medical care. The transition to EMR-keeping has many advantages but can also have significant drawbacks. Templates have become the norm. They can simplify data entry and provide consistency in the appearance and flow of the note. One must be carefully to avoid populating whole fields with findings when some were not actually performed. As our paper is trying to emphasize proper documentation, it is crucial to express the importance of honest documentation as well. Cao et al4 concluded that use of a standardized template in the EMR can improve completeness and timing of documentation of neurological injury. Sung et al5 concluded that use of a clinical pathway enhances the treatment efficiency of SCH fractures by streamlining the process with no increase in hospital stay or cost. The author acknowledges that the study was performed in a country with universal healthcare, so these results might not be applicable in other countries. LOS in their study was 2.9±0.7 days. LOS in our retrospective study was 22.1 hours (9 to 55 h). LOS in our prospective study was 18.9 hours (5.5 to 39 h). There were no compartment syndromes in either series.
In the EMR, the nursing note fields are exhaustive, and many have little to do with direct patient care (ie, skin color appropriate for ethnicity, do not resuscitate, living will, advanced directive, etc), yet have to be populated by the nursing staff. This is a time-consuming effort. In some of our patients, there were >50 pages of nursing notes for admissions that ranged from 9 to 55 hours.
For some of our patients there were no nursing notes from the PACU. The head nurse of the PACU tried to track these notes down but found that they had been lost in a change of formatting of the EMR. Information Technology indicated the notes could be recovered if needed, but we did not pursue this.
The most useful finding in the retrospective study was an “intervention 46” that was embedded in the PACU notes. Intervention 46 is a nurse charting prompt within our EMR that requires evaluation of a neurovascular examination. It is an Association of Perioperative Registered Nurses (AORN) standard that is used in our PACU and covered all the neurovascular documentation. There was, at that time, no such standard for the floor nurses. All of these problems identified in the first study were corrected and placed in the clinical pathway.
Postoperative documentation in the clinic is crucial. It is not unusual for a nerve injury to fail to be detected until the postoperative visits, when the child is more cooperative, and the signs of the deficit are readily apparent. At a minimum, there should a postoperative and a final examination if there is any question after the first examination. It was of interest that for 1 patient we could not find any evidence of a postoperative visit with our institution and thus excluded. There were 2 other patients excluded, who had an anterior interosseous nerve palsy, which returned for normal routine postoperative visits but were lost to follow-up for their palsy with “failure to show” noted.
Patients were routinely seen at 5 to 7 days postoperatively, and the posterior splint was converted to a cast. They were then seen back 2 to 3 weeks after casting. In many cases this was the patient’s last visit if there were no problems. Parents were instructed to return if there were any problems, including loss of motion.
A secondary finding was that the average LOS was 22 hours in the first study and 18.9 hours in the second. There were no compartment syndromes in either study with a combined cohort of 157. Following surgery, each patient was observed until we were certain it was safe to discharge them. Pain control on oral analgesics, nonsteroidal or narcotics, with stable neurovascular examination compared with preoperative, and tolerating oral intake were our discharge criteria. This included examinations before discharge confirming no concerns for compartment syndrome. This resulted in a range of 9 to 55 hours. Although other studies have suggested patients should be monitored for 24 hours postoperative, our study calls in to question the necessity of that LOS requirement. It continues to be our protocol to discharge most patients the afternoon after their procedure. This often gives a LOS<24 hours.
At times, a very young child cannot or will not cooperate on an examination. All care must be taken to try to do a proper examination and if it is not possible to do so this must be documented. If templates are used, the user must be certain that all items in the field have been done. In 1 of the cases reviewed by the senior author, templates were used that covered multiple items. Depositions proved not all of the items in the template had been performed.
Documentation is crucial to good medical care and defense should any legal issues arise. The implementation of a clinical pathway demonstrated significant improvement by physicians and nurses. There were areas identified that still require improvement such as scanning any paper documents into the EMR.
Clinic pathway order set and discharge paperwork are available as appendices online, Supplemental Digital Content 1, http://links.lww.com/BPO/A220.
1. Omid R, Choi PD, Skaggs DL. Supracondylar humeral fractures in children. J Bone Joint Surg Am. 2008;90:1121–1132.
2. Pirone AM, Graham HK, Krajbich JI. Management of displaced extension-type supracondylar fractures of the humerus in children. J Bone Joint Surg Am. 1988;70:641–650; Erratum in: J Bone Joint Am
3. Battaglia TC, Armstrong DG, Schwend RM. Factors affecting forearm compartment pressures in children with supracondylar fractures of the humerus. J Pediatr Orthop. 2002;22:431–439.
4. Cao J, Farmer R, Carry PM, et al. Standardized note templates improve electronic medical record documentation of neurovascular examinations for pediatric supracondylar humeral fractures. JB JS Open Access. 2017;2:e0027.
5. Sung K, Chung C, Lee K, et al. Application of clinical pathway
using electronic medical record system in pediatric patients with supracondylar fracture of the humerus: a before and after comparative study. BMC Med Inform Decis Mak. 2013;13:87.