Recruitment
During the 12 month study period, a total of 145 older adult patients 65 years of age and over were admitted with a diagnosis of hip fracture. We excluded those patients who had a traumatic fracture (n=3) or did not require surgical repair (n=11), those admitted over 48 hours from time of fracture (n=2), and those with unable to walk independently prior to their fracture (NMS less than 2, n=11). As this study aims to examine practices in one specific unit, any patients transferred to other units during hospitalization were excluded (n=40). We excluded one patient who died during surgery. The final analysis included of a total of 77 patients admitted to the target unit (Figure 1).
Demographics
Of the 77 patients, 32 (42%) were over the age of 90 years, and 55 (71%) were female. 35 (45%) of patients had a low pre-fracture functional ability (Table 1: Demographics). A total of 53 (68%) of patients were living independently or with a caregiver prior to admission.
Cognitive impairment (CI), documented as dementia or delirium on admission was present in 33 (43%) patients. There were 23 (30%) females and 10 (13%) males in this CI group. Post-operatively, an additional three patients with no history of cognitive impairment documented on admission did develop delirium (Table 2). Of those with cognitive impairment, 25 (76%) also had low pre-fracture functional ability. Of the 12 (16%) living in a retirement residence, eight had a cognitive impairment, and all 12 (16%) admitted from long term care had a cognitive impairment.
41 (53%) patients had a femoral neck fracture, and 36 (47%) were trochanteric fractures. The anesthetic approach included: 45 (58%) spinal, 31 (40%) general anesthetic, and 1 patient with a failed spinal converted to general. 45 (58%) of patients received a peripheral nerve block: 20 in addition to the spinal and 12 in addition to the general anesthetic.
Aim 1: Early mobility activities within the cohort
Physiotherapy assessment
On the first post-surgical day, 71 (92%) of the cohort were assessed by physiotherapy (Table 2: Mobility activities). The rate of physiotherapy assessments declined over the following post surgical days, ranging from 50-64% (Table 2: Mobility activities).
There was a similar rate of therapist assessment between the low pre-fracture function group compared to the high group; 32 (91%) and 39 (93%), respectively. As there were counts less than five (6 patients were not assessed), a Fisher's Exact Test (2 tail) was conducted. There was not a significant difference in physiotherapy assessments between those with a high pre-fracture functional ability and a low pre-fracture functional ability (p=1).
Physiotherapy assessed 40 (98%) of patients with no cognitive impairment, compared to 31 (86%) of those with a cognitive impairment. A Fisher’s Exact Test did not identify a significant difference in physiotherapy assessments between those with a cognitive impairment and those with no cognitive impairment (p=.09). Additional Fisher’s Exact tests for differences in physiotherapy assessments were insignificant for sex (p=.34), fracture location (p=.67), and presence of nerve block (p=.39).
Sitting up in chair
Achieving mobility to chair was documented consistently by both physiotherapy and nursing staff. Only 42 (54%) of patients sat in chair on the first day, with subsequent post-operative days demonstrating an increasing number of the patients being up to chair (Figure 1). When looking at patients grouped by functional status and cognition, as in Table 2, those with a high pre-fracture function with no cognitive impairment had the lowest percentage of sitting in a chair; 14 (46%). By the fourth day after surgery, this was the group with the highest percentage of sitting up in chair 17 (89%). Those with low pre-fracture function and a cognitive impairment demonstrated the highest rate of sitting up to chair on the third day 15 (83%) (Table 2).
We conducted Chi-square analyses to explore differences between groups for sitting up to chair on the first day. There was not a significant difference in sitting in chair between those with a cognitive impairment and those with no cognitive impairment (chi-square=.811, df=1, p=.36). There was not a significant difference in sitting in chair between those with high pre-fracture functional ability and those with low (chi-square=.183, df=1, p=.66). Additional Chi-square tests for differences in sitting in chair were insignificant for sex (chi-square=1.483, df=1, p=.22), fracture location (chi-square =.088, df=1, p=.77), and presence of nerve block (chi-square=0.0, df=1, p=.99).
Walking: 5m or less (within room)
On the first day, 35 (45%) of patients walked within the room (less than 5 metres). Patients with a high pre-fracture function had a higher incidence of achieving this activity on the first day; 22 (52%) compared to 13 (35%), in the low pre-fracture functional group (Figure 2). The highest percentage of walking within the room was represented by those with a high pre-fracture function with no cognitive impairment on the second day after surgery 22 (76%) (Table 2).
We conducted Chi-square analyses to explore differences between groups for walking within the room on the first day. There was not a significant difference in walking within room between those with a cognitive impairment and those with no cognitive impairment (chi-square=3.755, df=1, p=.05). There was a weak correlation (phi=-.228). There was not a significant difference in walking within room between those with high pre-fracture functional ability and those with low (chi-square=1.471, df=1, p=.22). Additional Chi-square tests for differences in mobility to chair were insignificant for sex (chi-square=3.93, df=1, p=.53), fracture location (chi-square =.00, df=1, p=.98), and presence of nerve block (chi-square=.196, df=1, p=.65).
Walking: more than 5m (out of room)
Overall, 5 (6%) of patients were able to ambulate over 5 meters on the first post-operative day (Table 2). By postoperative day five, across all groups, only two patients walked over 50 metres (one patient walked 200 metres). 9 (31%) of the patients with a high pre-fracture function and no cognitive impairment walking out of room on the second day (Table 2).
Relationship between variables
Logistic regression was performed to determine the predictive power of the independent variables on participation in mobility activities (dependent variables). These mobility activities were physiotherapy assessment, sitting up to chair, and participation in walking. The model contained five independent variables (sex, cognition, pre-fracture functional status, presence of nerve block, and fracture site). None of the full models containing all predictors was significant and no independent variables made a unique statistically significant contribution to the model. As they are insignificant, results are not reported.
Complications / length of stay
The Classification of Surgical Complications (37) was utilized to categorize postoperative complications of the cohort. Within the cohort, 44 (57%) had no complications. Twelve (16%) of patients had a documented report of pain, and ten (13%) were documented to have a medical issue leading to a deferral of physiotherapy (e.g, x-ray, telemetry, low blood pressure). Ten (13%) patients displayed a Grade 1 complication (e.g., transient confusion not requiring therapy, or requiring an anti-emetic). The most common complication was receiving a blood transfusion (Grade 2) by 16 (21%) patients. Seven patients (9%) died during their hospital stay (Grade 5). All deaths occurred in the low pre-fracture functional group; prior to their fracture three patients were admitted from long term care, one patient resided in a retirement home, and three resided in their own homes with relative or caregiver support.
As patients were discharged, the cohort size decreased, and by post operative day five 38 (49%) had been discharged (Table 2). The mean length of postoperative stay was seven days (range 1–64 days). Sixty-one (79%) patients were discharged to a rehabilitation facility, and nine (12%) were discharged directly to their previous facility (i.e., long term care).
Aim 2: Identifying the evidence-to-practice gap
Returning to the French & Green [16] method, the chart audit information can aid in identifying the existence of any evidence-to-practice gaps.
One of the first HQO (2017) recommendations is that when a patient presents to the hospital with a fragility hip fracture, surgery is to occur within 48 hours. The mean time to surgery in the cohort was 33 hours (SD:17.16), with a range from 3 to 97 hours, and 7 (9%) patients had their surgery beyond the recommended 48 hours after admission. The average time to surgery was not significantly different between those with no cognitive impairment (35 hours, with four patients after 48 hours) and those with a cognitive impairment (31 hours, with three patients after 48 hours) (p=.688).
Post-operatively, there are several other standards of care that are recommended, and these recommendations are compared to documented evidence of implementation within the chart, as outlined in Table 3.