- Email: [email protected]
Postdischarge Unscheduled Care Burden After Lower Limb Arthroplasty
The Journal of Arthroplasty xxx (2018) 1e7
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Postdischarge Unscheduled Care Burden After Lower Limb Arthroplasty Adam Tucker, MPhil, MRCS, MB a, *, Andrew Walls, MB, BCh, MRCS b, Beverley Leckey, BSc b, Janet C. Hill, PhD, MEng, BSc a, Glenn Phair, BSc, PgDip c, Damien B. Bennett, MB, MPH, PhD d, Seamus O’Brien, PhD a, David E. Beverland, MD, FRCS (Orth) b a
Department of Orthopaedics, Outcomes Unit, Musgrave Park Hospital, Belfast, County Antrim, United Kingdom Department of Orthopaedics, Musgrave Park Hospital, Belfast, County Antrim, United Kingdom c Northern Ireland Clinical Trials Unit (NICTU), Royal Victoria Hospital, Belfast, County Antrim, United Kingdom d Public Health Agency, Belfast, County Antrim, United Kingdom b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 March 2018 Received in revised form 18 April 2018 Accepted 23 April 2018 Available online xxx
Background: In contrast to postdischarge arthroplasty readmission rates, the unscheduled reattendance burden to primary care is under-reported. Understanding reasons for reattendance would allow for implementation of strategies to reduce this burden. The present study aims to quantify the out-of-hours (OOH) general practitioner and emergency department (ED) service reattendance burden and readmission rate after primary total hip arthroplasty and total knee arthroplasty, with estimation of the associated costs. Methods: This is a prospective consecutive cohort study. A prospective audit of all total hip arthroplasty and total knee arthroplasty patients in 2016 in a single high-volume UK arthroplasty unit was performed. Incidence and reasons for reattendance to OOH and ED service, as well as readmission rates, at both 30 and 90 days following discharge are reported. A multivariate analysis was performed to determine patient characteristics, which results in increased reattendance and readmission rates. Results: A total of 2351 procedures resulted in 374 attendances of OOH service and 665 to ED with a total estimated cost of £190,000 within 90 days. The readmission rate was 6.8%. Risk factors for reattendance and readmission were increasing age and a prolonged length of stay. The use of a 5-day postdischarge phone call and a dedicated Arthroplasty Care Practitioner favors reduced reattendances but not the readmission rate, with the additional benefit of being cost-effective. Conclusion: The postdischarge arthroplasty reattendance burden is associated with significant costs, and strategies to reduce this should be developed. Further research is required to assess the effectiveness and cost-effectiveness of multicomponent strategies to reduce reattendance operating at scale. © 2018 Elsevier Inc. All rights reserved.
Keywords: arthroplasty reattendance readmission unscheduled care reduction strategies
As numbers of hip and knee arthroplasties increase in the UK [1] and worldwide, so does economic pressure on health-care systems. Hospital readmission rates at 30 or 90 days after surgery are
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2018.04.041. * Reprint requests: Adam Tucker, MPhil, MRCS, MB, Department of Orthopaedics, Outcomes Department Unit, Musgrave Park Hospital, Stockmans Lane, Belfast BT9 7JB, County Antrim, United Kingdom. https://doi.org/10.1016/j.arth.2018.04.041 0883-5403/© 2018 Elsevier Inc. All rights reserved.
increasingly recognized as key care indicators. Although a reduced length of stay has definite financial benefits [2], this should not be at the cost of increased readmission rates. Readmission rates have been reported between 2.2% and 9.0% after lower limb arthroplasty, with higher rates observed after knee arthroplasty [2e8]. It has been estimated that up to 80% of hospital readmissions are potentially avoidable [9], with substantial cost savings [10]. Multidisciplinary enhanced recovery and facilitated discharge programs have been developed to avoid unplanned readmissions [11,12]. Direct comparisons of readmission rates between different hospitals or health-care systems are complicated by differences in criteria used for example, 30- or 90-day or other readmission rates
2
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
[13e15], with similar limitations reported in recent meta-analyses [16]. Also some count the interval from the date of surgery and others from the date of discharge [13e15]. Although readmission rates are widely reported [4,5,12e17], rate of attendance at OOH and emergency department (ED) is not. Understanding the reasons for these unscheduled health-care contacts could allow targeted changes to patient pathways, which could reduce these attendances and costs, and improve patient care and outcomes. The following are the aims of this audit: (1) To report the incidence of, and reasons for, unscheduled contact with OOH and ED attendances and hospital readmissions within 30 and 90 days from the date of surgery in a large consecutive cohort of lower limb arthroplasties. (2) To assess the impact of proactive Arthroplasty Care Practitioner (ACP) telephone support on unscheduled attendance and readmission rates compared with helpline access and usual care. (3) To compare estimated costs between these care models. We also discuss how unscheduled contacts could be further reduced. Methods A single regional orthopedic service in a high-volume arthroplasty unit was evaluated. We collected demographic and surgical information as well as incidence of, and reasons for, OOH and ED attendances and hospital readmissions within 30 and 90 days from the date of discharge in 2351 consecutive primary total hip and knee arthroplasties, performed under the care of 25 different consultants from January 1 2016 to December 31, 2016. This service uses the Belfast Orthopaedic Information System, which collects patient demographics, intraoperative data, and episode-specific outpatient data after lower limb arthroplasty. The electronic care record (ECR) records patient contact across health-care facilities in Northern Ireland, except for scheduled attendances to the patients' own general practitioner. Initial points of contact for patients experiencing postoperative difficulties are their base hospital, their general practitioner (GP), OOH, or ED. Within our hospital, primary arthroplasty is performed under the care of 25 different consultants with variation in management on advice to patients following discharge if they experience problems. This falls broadly into 3 follow-up groups: group A in which patients routinely receive a 5day phone call with advice to contact a “helpline” phone number; group B in which they can access a helpline only; and group C in which they have neither resource and are advised to contact their GP. The unit helpline and 5-day phone calls are performed by a specialist ACP nurse. These nurses function as part of the wider orthopedic multidisciplinary team, collecting follow-up and outcomes data, facilitating nurse-led review, and providing a helpline service for patients with postoperative problems (Supplementary Appendix A). As part of a unit audit process, Belfast Orthopaedic Information System and ECR were interrogated. Local audit approval was gained (BHSCT Orthopaedic Outcomes Ref 5626). Local imaging systems (picture archiving communication systems) were interrogated for episodes of hip dislocation and suspected and/or confirmed venous thromboembolic events (VTEs) (ultrasound Doppler for deep vein thrombosis [DVT] and computed tomography pulmonary angiogram [CTPA] for pulmonary embolism [PE] investigations). This facilitated comprehensive collection of complications including mortality, which were cross-referenced with the ECR, within 90 days of surgery. Health economic analysis was performed using unit costs per attendance at both OOH GP and ED services. Patients use of hospital services were combined with
corresponding unit costs to estimate local costs per attendance for patient follow-up groups A, B, and C. Unit costs were obtained from publicly available sources set at 2015/2016 prices [18]. Statistical Analysis Continuous data were tested for normality using the Kolmogorov-Smirnov test. Differences between groups were assessed using Student's t test or analysis of variance test for continuous variables and chi-square, Kruskal-Wallis, or Mann Whitney U tests for nonparametric data. Multivariate logistic regression was used to investigate which variables were significantly associated with attendance at unscheduled cares services or readmission. For multivariate analysis, variables with moderate associations (P value < .20) on univariate analysis or of a priori importance (eg, body mass index) were included. For all tests, a P value < .05 was considered statistically significant. All analyses were performed using SPSS v22 for Mac (IBM Inc., Armonk, NY). Results Patient Characteristics There were 1341 total hip arthroplasties (THAs) and 1010 total knee arthroplasties (TKAs) (1 simultaneous bilateral). Patient characteristics are shown in Table 1. Procedures were performed under the care of 25 consultant orthopedic surgeons and prospectively followed-up to 90 days from the date of surgery. Chemical VTE prophylaxis was predominately aspirin or enoxaparin with 489 patients receiving aspirin; 1858, enoxaparin; 1, warfarin; and 3 patients receiving no chemical prophylaxis. All patients received mechanical prophylaxis with calf pumps. Average hospital length of stay was 4.4 ± 2.9 days. Mean time to first unscheduled contact was 25.4 ± 24.6 days. There were 2 deaths within 90 days (0.09%), both occurring in female TKA patients with American Society of Anaesthesiologists (ASA) grade 2 who were discharged on day 3 with no initial complications. One died 10 days postoperatively from a perforated abdominal viscus following emergency laparotomy. The second died of PE 55 days postoperatively, having been on aspirin VTE prophylaxis for 6 weeks postoperatively. Unscheduled Care Attendance Out-of-Hours General Practitioners There were 284 (12.1%) patients who were attended by OOH GPs within 90 days of discharge, creating 374 OOH attendances.
Table 1 Demographic Data. Variable
THA
TKA
N Male Female Age (mean ± SD) BMI (kg/m2) (mean ± SD) ASA 1 2 3 4
1341 592 (44.1%) 749 (55.9%) 66.7 ± 11.7 29.8 ± 7.3
1010 411 (40.7%) 599 (59.3%) 69.0 ± 9.3 32.2 ± 5.9
127 1032 177 5
53 797 153 7
(9.5%) (76.9%) (13.2%) (0.4%)
P Value
(5.2%) (78.9%) (15.1%) (0.7%)
.102 <.001* <.001* <.001*
ASA, American Society of Anaesthesiologists; BMI, body mass index; SD, standard deviation; THA, total hip arthroplasty; TKA, total knee arthroplasty. * Denotes significant differences at P < .05 level.
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
3
Table 2 Reasons for Reattendance and Readmission at 90 d. OOH (n ¼ 284)
Reason THA Medical Suspected VTE Wound Surgery related Nonsurgery related Other musculoskeletal Total P value
85 (55.9%) 46 (30.3%) 9 (5.9%) 8 (5.3%) 1 (0.7%) 3 (2.0%) 152 (11.3%) .936
ED (n ¼ 353)
TKA 66 47 8 9 2
(50.0%) (35.6%) (6.1%) (6.8%) (1.5%)
0 132 (13.1%)
THA
TKA
59 (31.2%) 110 (58.2%) 4 (2.1%) 8 (4.2%) 1 (0.5%) 7 (3.7%) 189 (14.1%) .749
50 87 8 8 1 10 164
Readmission (n ¼ 159)
(30.5%) (53.0%) (4.9%) (4.9%) (0.6%) (6.1%) (16.2%)
THA
TKA
45 (56.9%) 15 (19.0%) 7 (8.9%) 9 (11.4%) 1 (1.3%) 2 (2.5%) 79 (5.9%) .873
49 14 10 6 1 0 80
(61.3%) (17.5%) (12.5%) (7.5%) (1.3%) (7.9%)
ED, emergency department; OOH, out-of-hours; THA, total hip arthroplasty; TKA, total knee arthroplasty.
The proportions attending OOH GPs did not differ significantly after THA (11.3%, 152/1341) compared with TKA (13.1%, 132/1010, P ¼ .201) (Table 2). Similarly, there was no difference between reasons for OOH GP reattendance between surgical procedures (P ¼ .936) (Table 2). Of patients seen by OOH GPs, 54/284 (19%) were immediately referred to EDs. Thirty had suspected VTE, comprising 26/30 (86.7%) DVT and 4/30 (13.3%) PE, 19 medical problems, 4 wound infections, and 1 dislocation of THA. Emergency Departments There were 353 (15.0%) patients who attended EDs, accounting for 665 attendances, within 90 days. The proportions attending EDs did not differ significantly after THA (14.1%, 189/1341) compared with TKA (16.2%, 164/1010, P ¼ .150) (Table 2). There was no difference between reasons for ED reattendance between surgical procedures (P ¼ .749) (Table 2). Reasons for reattendance and readmission are reported in Table 2. Within both OOH and ED groups, there was a significantly higher proportion of >75-year-olds reattending. However, on univariate analysis, there was no difference in any of the other
Table 3 Proportion of Patients Reattending ED and OOH. Variable
ASA 1 2 3 4 N P value (within group) BMI <25 25-30 31-35 >35 Missing data N P value (within group) Age <55 56-64 65-74 >75 N P value (within group) Male Female P value (within group)
n
ED (%)
OOH (%)
P Value (Between Group)
181 1827 329 8 2345
21 (12.1%) 268 (14.7%) 62 (19.5%) 2 (25%) 353 .127
25 (13.8%) 212 (11.6%) 44 (14.1%) 3 (37.5%) 284 <.001*
.441
259 987 642 446 3 2334
37 (14.3%) 150 (15.2%) 99 (15.4%) 64 (14.3%) 3 350 .946
36 (13.9%) 118 (12.0%) 71 (11.1%) 57 (13.2%) 0 282 .532
.670
311 499 863 678 2351
34 (10.9%) 67 (13.4%) 121 (14.0%) 131 (19.3%) 353 .001* 162 (16.2%) 191 (14.2%) .194
35 (11.3%) 57 (11.4%) 89 (10.3%) 103 (15.2%) 284 .028* 111 (11.1%) 173 (12.8%) .183
.661
1003 1348
.099
ASA, American Society of Anaesthesiologists; BMI, body mass index; ED, emergency department; OOH, out-of-hours. * Denotes significant differences at P < .05 level.
variables between patients who attended OOH and those who attended EDs (Table 3). Readmissions There were 159 (6.8%) patients readmitted within 90 days of discharge, with 98/2351 (4.2%) readmitted within 30 days. Mean time to readmission was 30.4 ± 26.3 days (median 23 days; interquartile range 7.0-45.2 days, range 1-87 days). There was no significant difference in 30- or 90-day readmission rates between THA (3.8% and 5.9%) and TKA (4.7% and 7.9%) patients (P ¼ .052). There was no difference in 90-day readmission rates between males (7.3%, 73/1003) and females (6.4%; 85/1348, P ¼ .391). Patients readmitted had a longer initial length of stay (5.1 ± 3.1 vs 2.9 ± 4.4 days; P ¼ .001). Reasons for readmission are given in Table 2. A return to theater within 90 days occurred in 20 cases, with surgery-related complications (infection and dislocation) accounting for 11/20 (55.0%) of cases (Table 4). Of note, the reported dislocation rates are only for those returning to theater for open/closed reduction, within 90 days. On multivariate logistic regression only age > 75 years (odds ratio 2.367 [95% confidence interval 1.253-4.472]; P ¼ .008) was significantly associated with readmission. ASA grade, gender, body mass index, length of stay, and all other age bands were not significantly associated with an increased risk of readmission (Table 5). Service Model Comparison There were 3 groups of patients who experienced different service models for both procedures (THA, TKA). Group A (n ¼ 560) received a 5-day phone call postdischarge and had access to a telephone helpline; group B (n ¼ 647) had access to the helpline only; and group C (n ¼ 1144) had neither. Length of stay was significantly shorter in group A compared with both groups B and C (3.6 ± 2.6 vs 4.5 ± 3.2 and 4.8 ± 2.8 days, respectively; P < .001). Group A had lower OOH and ED reattendance rates, without increased readmission rates at both 30 and 90 days postoperatively (Table 6). Overall, 30.5% of patients in group C who experienced
Table 4 Reasons for Return to Theater Within 90 d. Variable
THA (n ¼ 1341)
Washout for infection Dislocation Periprosthetic fracture Other surgical intervention not related to primary surgery Total
2 4 1 1
(0.15%) (0.30) (0.07%) (0.07%)
8 (0.60%)
THA, total hip arthroplasty; TKA, total knee arthroplasty.
TKA (n ¼ 1010) 5 (0.50%) 0 3 (0.30%) 4 (0.4%) 12 (1.19%)
4
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
Table 5 Multivariate Regression Analysis for Unscheduled Reattendance to OOH and ED and for Readmission. Variable
OOH
ED
OR (95% CI) Gender Male Female Age overall <55 55-65 66-75 >75 BMI overall <25 25-30 31-35 >35 Length of stay ASA Surgery Hip Knee Follow-up group A B C
P
1
Readmission
OR (95% CI)
P -
OR (95% CI)
P
1 0.790 1.028 1 1.051 1.359 2.367 0.993 1 1.030 0.991 1.176 1.035 1.372
(0.568-1.100) (1.010-1.047)
.162 .003*
(0.523-2.114) (0.725-2.549) (1.253-4.472) (0.966-1.021)
.889 .339 .008* .626
(0.590-1.799) (0.549-1.791) (0.626-2.210) (0.984-1.088) (0.978-1.926)
.917 .977 .614 .184 .067
(0.869-1.456) (0.993-1.019)
.372 .630
(0.634-1.573) (0.563-1.321) (0.808-1.949) (0.970-1.014)
.997 .497 .312 .453
1 0.802 1.014 1 1.253 1.277 1.648 0.996
(0.573-1.303) (0.503-1.226) (0.589-1.518) (1.017-1.098) (0.747-1.286)
.485 .288 .816 .005* .883
1.042 1.045 0.993 1.033 1.129
1 0.886 (0.683-1.150)
.364
1 0.886 (0.699-1.123)
.318
1 0.804 (0.575-1.125)
.203
1 1.246 (0.853-1.819) 1.477 (1.054-2.069)
.255 .023*
1 1.124 (0.799-1.581) 1.330 (0.984-1.798)
.501 .064
1 0.845 (0.536-1.333) 0.877 (0.587-1.310)
.470 .521
1.12 1.006 1 0.999 0.862 1.255 0.992 1 0.864 0.786 0.945 1.057 0.980
(0.635-1.013) (1.001-1.026)
.064 .028*
(0.799-1.964) (0.840-1.941) (1.067-2.546) (0.977-1.015)
.326 .252 .024* .676
(0.699-1.553) (0.684-1.596) (0.627-1.574) (1.029-1.105) (0.885-1.442)
.839 .839 .976 <.001* .329
ASA, American Society of Anaesthesiologists; BMI, body mass index; CI, confidence interval; ED, emergency department; OOH, out-of-hours; OR, odds ratio. * Denotes significant differences at P < .05 level.
standard care (no 5-day phone call or helpline access) either attended by OOH GPs/EDs or were readmitted to hospital within 90 days of discharge. Unscheduled care attendance was significantly reduced for patients with helpline access (group B; 25.8%) and further reduced for those receiving a 5-day phone call along with helpline access (group A; 21.6%, P ¼ .002) (Table 6). In multivariate analysis, the likelihood of OOH GP attendance was significantly higher for group C than for group A (odds ratio ¼ 1.5, 95% confidence interval 1.05-2.07). There was no difference in the likelihood of attendance at EDs or for readmission between the 3 groups (Tables 5 and 6). This suggests that the service received by
group A reduced the risk of unscheduled OOH GP attendance without increasing the risk of ED attendance or readmission (Tables 5 and 6). Estimated Costings for Unscheduled Care Attendance The average cost per patient of the service models ranged from £37.03 in group A to £51.42 in group C. The lowest total cost and lowest cost per patient was in group A where patients receive a 5day follow-up phone call and have helpline access. This was due to reduced unscheduled care attendance. We also estimated the costs of running the ACP-led 5-day follow-up phone call and helpline
Table 6 Thirty- and 90-d Readmission Rates by Follow-Up Group (Unit Cost per OOH £83; Unit Cost per ED Attendance £238). Variable
n
Group A (n ¼ 560)
Group B (n ¼ 647)
Group C (n ¼ 1144)
Male (%) Female (%) Age BMI ASA I II III IV Length of stay (mean meSD) At 30 d OOH attendances OOH cost ED attendances ED cost Readmissions At 90 d OOH attendances Cost ED attendances Cost Combined OOH/ED reattendance burden Total cost for OOH þ ED Average cost per patient Readmissions
2351
220 (39.3%) 340 (60.7%) 68.7 ± 9.7 31.0 ± 9.2
274 (42.3%) 373 (57.7%) 67.1 ± 11.7 31.2 ± 6.4
509 (44.5%) 635 (55.5%) 67.6 ± 10.7 30.5 ± 5.5
.122
38 458 62 2 3.6 ± 2.7
47 485 111 4 4.5 ± 3.2
95 886 157 6 4.8 ± 2.8
.090
37 (6.6%) £3071 40 (7.1%) £9520 29 (5.2%)
55 (8.5%) £4565 50 (8.5%) £11,190 21 (3.2%)
111 (9.7%) £9213 128 (9.7%) £30,464 48 (4.2%)
52 (9.3%) £4316 69 (12.3%) £16,422 121 (21.6%) £20,738 £37.03 42 (7.5%)
76 (11.8%) £6308 91 (14.1%) £21,658 167 (25.8%) £27,966 £43.22 41 (6.4%)
156 (13.6%) £12,948 193 (16.9%) £45,934 349 (30.5%) £58,882 £51.42 76 (6.6%)
2351 2351 2351
2351 203 218 98 284 353 637
159
ASA, American Society of Anaesthesiologists; BMI, body mass index; ED, emergency department; OOH, out-of-hours; SD, standard deviation. * Denotes significant differences at P < .05 level.
P Value
.031* .084
<.001* .101 .007* .245 .033* .035* <.001*
.707
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
service. Using unit data (Appendix Supplementary File A), we calculated the cost of the 5-day phone call at £5.73 per patient. The cost of the helpline service was £5.26 per patient, based on an average 1.37 phone calls to the helpline per patient and £3.84 unit cost per phone call. Thus, the total cost of running the service received by group C patients was £10.99 per patient. To calculate the total cost of the service models, we added the costs of unscheduled care attendance to the costs associated with that service. Adding the cost of running the service to the cost of unscheduled care attendance, the total costs of the service model experienced by group A was £48.02, which is less than that of group C (£51.42). The cost of running the service for group B patients (helpline only) was £48.48, which is marginally greater than that of for group A but less than that for group C. Venous Thromboembolic Events Of 298 unscheduled attendances for suspected VTE up to 90 days from surgery, 195 (65.4%) had Doppler ultrasound scanning for DVT with the majority performed after discharge (96.9%, 189/195). Fifty patients attended more than once for therapeutic prophylaxis due to the absence of a same-day Doppler service. Patients were readmitted in 30.2% of cases (59/195) with DVT confirmed by Doppler in 10.8% (21/195) of cases. There was no difference in the proportions with confirmed DVT between THA (1.04%, 14/1341) and TKA patients (0.69%, 7/1010, P ¼ .508) or between those receiving enoxaparin (16/1858) or aspirin (5/489, P ¼ .787). Pulmonary Embolus was suspected in a further 53 patients with 39.6% (21/53) investigated after discharge. All patients had CTPA with 17% (9/53) positive for PE. There was no significant difference in the proportions of patients with a confirmed PE diagnosed on CTPA who had undergone THA (0.30%, 4/1341) or TKA (0.50%, 5/ 1010, P ¼ .511). There was also no statistically significant difference between the choice of VTE prophylaxis using enoxaparin 6/1858 (0.32%) vs aspirin 3/489 (0.61%); P ¼ .406. Discussion This is the first report of unscheduled contact with OOH GPs and EDs in the postoperative period, alongside hospital readmissions, in a large consecutive cohort of UK lower limb arthroplasty patients. The burden of unscheduled contact was high with 15% of patients attending EDs and 12% attending OOH GPs within 90 days of surgery. Readmission rates were similar to previous studies [12e17]. We have shown that a service model involving ACP telephone contact on day 5 postoperatively was associated with significantly lower attendance at OOH GPs and that attendance at EDs and readmission rates were not higher in this group. We estimate this model delivered cost savings of £3.40 per patient, based on reduced unscheduled care use, compared with usual care. Postsurgical readmission rates are key quality and service performance measures and have been adopted as National Health Service (NHS) quality indicators [19]. The cost of 30-day readmissions after surgery in the NHS has been estimated at £1.6 billion annually [20]. Reports of readmission rates vary, ranging from 3.3% at 30 days to 9.7% at 90 days [13,14,16,21,22], but there is a lack of consistency in reporting with, for example, some 30- and 90-day readmission rates measured from the day of surgery and others from the day of discharge [13e16]. In our study, readmission rates at both 30 and 90 days after operation were similar to previous studies with 90-day readmission rates of 5.9% for THA and 7.9% for TKA patients [13,14,16,21,22]. Examining reasons for readmission is important to inform attempts to reduce their occurrence, and previous studies have reported reasons for readmission after THA that were predominantly medical (35%), while those after TKA (75%) were mainly surgery related [13,14]. However, in our study
5
readmissions were mainly for non-VTE related medical reasons with rates similar for THA (57.0%) and TKA (61.3%) patients. However, previous studies have mainly reported readmission rates and ignored the use of unscheduled care by lower limb arthroplasty patients in the early postoperative period. The present study reports both unscheduled care attendances and readmissions and thus comprehensively describes the lower limb arthroplasty postdischarge burden and its distribution in a large-scale study. Unscheduled care attendances were high compared to readmission rates with 14.1% of THA patients and 16.2% of TKA patients attending EDs and 11.3% of THA patients and 13.1% of TKA patients attending OOH GPs within 90 days of surgery, respectively. Considering both OOH GP and ED attendance together, 25.4% of patients of THA patients and 29.3% of TKA patients attended EDs or OOH GPs within 90 days of surgery, which underlines the substantial burden on unscheduled care services. In the present study, 6.8% of patients required readmission in the first 90 days. The most common causes of hospital readmissions were medical (59.1%), followed by suspected VTE (18.2%), wound problems (10.7%), related to surgery (9.4%), unrelated to surgery (1.3%), and finally, other musculoskeletal injury (1.3%). Reasons for attendance at OOH GPs were similar to those to those for readmission to hospital with 53.1% being for medical reasons and 32.7% for suspected VTE. However, for those attending EDs, the most common cause was suspected VTE (55.8%) followed by medical reasons (30.9%). There is conflicting reports on the effect of earlier discharge on complication or readmission rates. Sibia et al [15] reported that 30day (all cause) and 90-day (wound related) readmission rates were not different between those discharged on day 1 and day 2 postoperatively. However, a prolonged hospital length of stay has been shown to be associated with discharge to a nursing facility, compared with discharge to the patients' home with self-care [23]. Our study found that longer length of stay was a significant predictor of both OOH GP and ED attendance but not of readmission to hospital. Increasing age has also been associated with increased risk of readmission [17]. Older age was significantly associated with both ED attendance and hospital readmission in our study, with patients older than 75 years over one and half times more likely to attend an ED and over twice as likely to be admitted to hospital within 90 days of surgery compared with younger patients (<55 years). This is not surprising as multimorbidity rises with age, although we included ASA grade, which is often used a proxy measure of comorbidity, as a covariate in our analysis. OOH GPs are often the initial point of contact for patients postoperatively because of the difficulty in accessing daytime GP appointments. The financial burden of these attending OOH GPs was not inconsequential with an estimated cost of £31,000 for the 284 (12.1%) patients who were attended by OOH GPs on 374 occasions. However, the cost of ED attendances was much greater, with an estimated cost of £158,000 for the 353 (15%) patients who were attended by OOH GPs on 665 occasions. VTE is a major cause of morbidity and mortality worldwide which can prolong inpatient stay [24e26]. DVT is more common than PE, which itself has a reported in-hospital mortality of 5%-10% [27,28]. Up to 24% of VTE cases are estimated to occur postoperatively, with lower limb arthroplasty considered a high-risk procedure [29]. Symptomatic VTE is reported at up to 14%, but this still amounts to a significant workload to primary care [30,31]. In the present study, symptomatic VTE rates for DVT (THA 1.04% vs TKA 0.69%) and PE (THA 0.3% vs KA 0.50%) were comparable to previous reports [32e36]. Although only a small proportion of VTE cases are symptomatic, these can lead to ED attendances and readmissions in up to 6.3% of cases [32,37,38]. Readmissions due to suspected VTE can potentially be reduced using ambulatory
6
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
pathways with substantial cost savings [39,40]. For example, in our study, over a 12-month period, 29 suspected DVT/PE were admitted to hospital overnight for investigation at an estimated direct cost of almost £25,000 (£17,860 bed costs and £7000 for ED attendances). This is the cost equivalent of Doppler outpatient examinations for 350 patients. A dedicated daily outpatient Doppler service would allow timely investigation of up to 93 OOH and 197 ED presentations. In this audit, these patients accounted for approximately 530 attendances, which if avoided could save £108,000 per annum in ED attendances and further save £125,000 if hospital admissions were prevented. Reduced OOH and ED attendances would also be more convenient for patients. A variety of measures, such as preadmission discharge planning, preoperative education [41e43], preoperative screening using the likes of the Risk Assessment and Prediction Tool [44,45], Multidisciplinary Early Supported Discharge pathways, and early clinical review for high-risk patients have been suggested as methods to reduce costs and readmission rates [41,42,46e52]. Overall, the majority of fatal or near fatal complications after lower limb arthroplasty occurs in the first 4 days after index surgery [53]. Our study has shown that a service model involving proactive postoperative telephone contact by ACPs on day 5 significantly reduced attendance at OOH GPs and represents a promising model. The ACP 5-day call and helpline service in this study significantly reduced OOH GP attendance was found to be cost-effective. The ACP, who could be based at a surgical site or ED, provides advice, facilitates early review, triages for unscheduled review, and expedites urgent referral to suitable specialties, if appropriate [54]. Implementation of an ACP-led 5-day call and helpline access, from a hospital service perspective, could save £8016 per year in our unit (based on 2351 patients in 2016) but more importantly would reduce the burden on OOH GPs and EDs as well potentially improving patient experience. The total number of patients attending ED increased 10.1% between January 2015 and January 2016, with corresponding emergency admissions increasing 4.1% [54,55]; the aforementioned strategies may be an attractive option. The NHS is the UK public health-care system. It is a pillar of the welfare state, being directly funded from the general finances of the state, and while working citizens make contributions to this, medical care is provided completely free for all. It is a stretched system, with increasing costs year on year due to an aging population and increasing costs of drugs, equipment, and treatment options. This clearly differs significantly from privatized healthcare systems, and this must be considered when analyzing our results and their validity to other health-care systems. A strength of the present study is its prospective follow-up of a large consecutive cohort of THA and TKA patients who had surgery at a single regional UK center. We report attendances at OOH GPs and EDs alongside hospital readmission rates. To the authors' knowledge, this is the first study to comprehensively quantity the postdischarge burden of primary THA and TKA on unscheduled care services. As outlined in our methodology, we were unable to measure daytime patient attendances at their own GP as our ECR system cannot access this data. We also did not capture the small number of patients who, following helpline calls, were reviewed by an ACP in an outpatient setting, which probably prevented OOH GP or ED attendance. Economic analysis was from a hospital perspective only, with social and community care costs not considered and no quality of life (eg, EQ5D) data collected. The length of stay for the present study may seem excessive. Postoperative length of stay is falling in the UK, but it is generally longer than in the US. One major problem in the UK is the almost absence of preadmission discharge planning and frequently less focus on preoperative patient education. This results in a culture that encourages patient dependence
rather than independence. The study was based in a regional orthopedic unit over a 1-year period and captured data from patients under the care of 25 orthopedic consultants. However, this is probably representative of elective lower limb orthopedic surgery in other UK regions.
Conclusion Attendances at OOH GPs and EDs and hospital readmissions were considerable with 1 in 4 patients attending one of these services within 90 days of operation. The main reasons for attendance at OOH GPs and EDs and for hospital readmissions were medical problems or suspected VTE. We have shown that a service model involving a dedicated ACP help line and proactive patient contact 5 days postoperatively significantly reduced attendance at OOH GPs and was cost-effective. Further studies are required to assess the impact of other interventions, such as dedicated VTE pathways, on unscheduled care use and hospital readmission rates.
References [1] National Joint Registry 13th Annual Report. http://www.njrreports.org. uk/Portals/0/PDFdownloads/NJR%2013th%20Annual%20Report%202016.pdf. [Accessed 6 September 2017]. [2] Avram V, Petruccelli D, Winemaker M, de Beer J. Total joint arthroplasty readmission rates and reasons or 30-day hospital readmission. J Arthroplasty 2014;29:465e8. [3] Cullen C, Johnson DS, Cook G. Re-admission rates within 28 days of total hip replacement. Ann R Coll Surg Engl 2006;88:475e8. [4] Bini SA, Fithian DC, Paxton LW, Khatod MX, Inacio MC, Namba RS. Does discharge disposition after primary total joint arthroplasty affect readmission rates? J Arthroplasty 2010;25:114e7. [5] Vorhies JS, Wang Y, Herndon J, Maloney WJ, Huddleston JI. Readmission and length of stay after total hip arthroplasty in a national Medicare sample. J Arthroplasty 2011;26:119e23. [6] Saucedo JM, Marecek GS, Wanke TR, Lee J, Stulberg SD, Puri L. Understanding readmission after primary total hip and knee arthroplasty: who's at risk? J Arthroplasty 2014;29:256e60. [7] Clement RC, Derman PB, Graham DS, Speck RM, Flynn DN, Levin LS, et al. Risk factors, causes, and the econmimc implications of unplanned readmissions following total hip arthroplasty. J Arthroplasty 2013;28:7e10. [8] Canadian Institute for Health Information 2012: all-cause readmission to acute care and return to the emergency department. https://secure.cihi.ca/ free_products/Readmission_to_acutecare_en.pdf. [Accessed 7 September 2017]. [9] MedPAC Report to the congress: promoting greater efficiency in medicare. http://www.medpac.gov/-documents-/reports. [Accessed 21 November 2017]. [10] Bosco III JA, Karkenny AJ, Hutzler LH, Slover JD, Iorio R. Cost burden of 30-day readmissions following medicare total hip and knee arthroplasty. J Arthroplasty 2014;29:903e5. [11] Raphael M, Jaeger M, van Vlymen J. Easil adoptable total joint arthroplasty programme allows discharge home in two days. Can J Anaesth 2011;58: 902e10. [12] Husted H, Otte SO, Kristensen BB, Orsnes T, Kehlet H. Readmission after fasttrack hip and knee arthroplasty. Arch Orthop Trauma Surg 2010;130: 1185e91. [13] Schairer WW, Sing DC, Vail TP, Bozic KJ. Causes and frequency of unplanned hospital readmission after total hip arthroplasty. Clin Orthop Relat Res 2014;472:464e70. [14] Schairer WW, Vail TP, Bozic KJ. What are the rates and causes of hospital readmission after total knee arthroplasty? Clin Orthop Relat Res 2014;472: 181e7. [15] Sibia US, Waite KA, Callanan MA, Park AE, King PJ, MacDonald JH. Do shorter lengths of stay increase readmissions after total joint replacements? Arthroplasty Today 2016;3:51e5. [16] Ramkumar PN, Chu CT, Harris JD, Athoviraham A, Harrington MA, White DL, et al. Causes and rates of unplanned readmissions after elective primary total joint arthroplasty: a systematic review and meta-analysis. Am J Orthop (Belle Mead NJ) 2015;44:397e405. [17] Bernatz JT, Tueting JL, Anderson PA. Thirty-day readmission rates in orthopaedics: a systematic review and meta-analysis. PLoS One 2015;10:e0123593. [18] Curtis L. Burns a on behalf of the Personal Social Services Research Unit (PSSRU): unit costs of health and social care 2016. http://www.pssru.ac.uk/ pub/uc/uc2016/full.pdf?uc¼2016-full. [Accessed 21 November 2017]. [19] NHS digital: Clinical Commissioning Group Outcomes Indicator Set (CCGOIS): emergency readmissions within 30 days of discharge from hospital. 2015.
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
[20]
[21] [22]
[23] [24]
[25]
[26]
[27] [28] [29]
[30]
[31]
[32]
[33]
[34]
[35]
[36] [37]
https://data.england.nhs.uk/dataset/ccgois-3-2-emergency-readmissionswithin-30-days-of-discharge-from-hospital. [Accessed 30 November 2017]. Robinson P. CHKS hospital readmissions and the 30 day threshold 2010. http://www.chks.co.uk/userfiles/files/CHKS%20Report%20Hospital%20read missions.pdf. [Accessed 21 November 2017]. Zmistowski B, Hozak WJ, Parvizi J. Readmission rates after total hip arthroplasty. JAMA 2011;306:825. Mednick RE, Hasham MA, Krishnan V, Lovecchio F, Manning DW. Factors affecting readmission rates following primary total hip arthroplasty. J Bone Joint Surg Am 2014;96:1202e9. Nichols CI, Vose JG. Clinical outcomes and costs within 90 days of Primary or revision total joint arthroplasty. J Arthroplasty 2016;31:1400e1406.e3. Raskob GE, Angchaisuksiri P, Blanco AN, Buller H, Gallus A, Hunt BJ, et al. Thrombosis: a major contributor to the global disease burden. J Thromb Haemost 2014;12:1580e90. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-based clinical practice guidelines, 8th edition. Chest 2008;133: 381Se453S. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: The Seventh ACCP Conference on antithrombotic and Thrombolytic Therapy. Chest 2004;126:338Se400S. Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep vein thrombosis. Lancet 2012;379:1835e46. Alikhan R, Peters F, Wilmott R, Cohen AT. Fatal pulmonary embolism in hospitalised patients: a necropsy review. J Clin Pathol 2004;57:1254e7. Heit JA, O’Fallon WM, Petterson TM, Lohse CM, Silverstein MD, Mohr DN, et al. Relative impact of risk factors for deep vein thrombosis and pulmonary embolism: a population-based study. Arch Intern Med 2002;162:1245e8. Stulberg BN, Insall JN, Williams GW, Ghelman B. Deep-vein thrombosis following total knee replacement. An analysis of six hundred and thirty-eight arthroplasties. J Bone Joint Surg Am 1984;66:194e201. Ginsberg JS, Turkstra F, Buller HR, MacKinnon B, Magler D, Hirsch J. Postthrombotic syndrome after hip or knee arthroplasty: a cross-sectional study. Arch Intern Med 2000;160:669e72. Samama C-M, Ravaud P, Parent F, Barre J, Mertl P, Mismetti P. Epidemiology of venous thromboembolism after lower limb arthroplasty: the FOTO study. J Thromb Haemost 2007;5:2360e7. Fujita Y, Nakatsuka H, Namba Y, Mitani S, Yashotake N, Sugimoto E, et al. The incidence of pulmonary embolism and deep vein thrombosis and their predictive risk factors after lower extremity arthroplasty: a retrospective analysis based on diagnosis using multidetector CT. J Anesth 2015;29:235e41. Ogonada L, Hill J, Doran E, Dennison J, Steven M, Beverland D. Aspirin for thromboprophylaxis after primary lower limb arthroplasty: early thromboembolic events and 90 day mortality in 11,459 patients. Bone Joint J 2015;98-B:341e8. Cusick LA, Beverland DE. The incidence of fatal pulmonary embolism after primary hip and knee replacement in a consecutive series of 4253 patients. J Bone Joint Surg Br 2009;91:645e8. Allen D, Sale G. Lower limb joint replacement in patients with a history of venous thromboembolism. Bone Joint J 2014;96-B:1515e9. Planes A, Vochelle N, Darmon JY, Fagola M, Bellaud M, Huet Y. Risk of deepvenous thrombosis after hospital dis-charge in patients having undergone total hip replacement: double-blind randomised comparison of enoxaparin versus placebo. Lancet 1996;348:224e8.
7
[38] Merkow RP, Ju MH, Chung JW, Hall BL, Cohen ME, Williams MV, et al. Underlying reasons associated with hospital readmission following surgery in the United States. JAMA 2015;313:483e95. [39] All-Party Parliamentary Thrombosis Group (APPTG): improving outcomes, saving lives, saving money: innovative DVT pathways. http://apptg.org.uk/ wp-content/uploads/2016/12/NHS-Innovation-Showcase.pdf. [Accessed 21 December 2017]. [40] Sampson FC, Goodacre S, Kelly AM, Kerr D. How is deep vein thrombosis diagnosed and managed in UK and Australian emergency departments? Emerg Med J 2005;22:780e2. [41] Ibrahim MS, Khan MA, Nizam I, Haddad FS. Peri-operative interventions producing better functional outcomes and enhanced recovery following total hip and knee arthroplasty: a evidence based review. BMC Med 2013;11:37. [42] Keeney JA, Nam D, Johnson SR, Nunley RM, Clohisy JC, Barrack RL. The impact of risk reduction initiatives on readmission: THA and TKA readmission rates. J Arthroplasty 2015;30:2057e60. [43] Edwards PK, Mears SC, Barnes L. Preoperative education for hip and knee replacement: never stop learning. Curr Rev Musculskelet Med 2017;10: 356e64. [44] Oldmeadow LB, McBurney H, Robertson VJ. Predicting risk of extended inpatient rehabilitation after hip of knee arthroplasty. J Arthroplasty 2003;18: 775e9. [45] Hansen VJ, Gromov K, Lebrun LM, Rubash HE, Malchau H, Freiberg AA. Does the risk assessment and prediction took predict discharge disposition after joint replacement? Clin Orthop Relat Res 2015;473:597e601. [46] Briggs T. A national review of adult elective orthopaedic services in England: getting it right first time. 2015. https://www.boa.ac.uk/wp-content/uploads/ 2015/03/GIRFT-National-Report-Mar15.pdf. [Accessed 24 November 2017]. [47] Mechanic RE. Mandatory medicare bundled payment- Is it ready for prime time? N Engl J Med 2015;373:1291e3. [48] Bradley EH, Curry L, Horwitz LI, Sipsma H, Wang Y, Walks MN, et al. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circ Cardiovasc Qual Outcomes 2013;6:444e50. [49] Kripilani S, Theobold CN, Anctil B, Vasilevskis EE. Reducing hospital readmission: current strategies and future directions. Annu Rev Med 2014;65: 471e85. [50] Hume K, Tomsik E. Enhancing patient education and medication reconciliation strategies to reduce readmission rates. Hosp Pharm 2014;49: 112e4. [51] Tayne S, Merill CA, Smith EL, Mackey WC. Predictive risk factors for 30-day readmissions following primary total joint arthroplasty and modification of patient management. J Arthroplasty 2014;29:1938e42. [52] Jordan CJ, Goldstein RY, Michaels RF, Hutzler L, Slover JD, Bosco III JA. Comprehensive program reduces hospital readmission rates after total joint arthroplasty. Am J Orthop (Belle Mead NJ) 2012;41:E147e51. [53] Parvizi J, Mui A, Purtill JJ, Sharkey PF, Hozack WJ, Rothman RH. Total joint arthroplasty: when do fatal or near-fatal complications occur? J Bone Joint Surg Am 2007;89:27e32. [54] National Institute for Clinical Excellence (NICE) UK 2015: safe staffing for nursing in A&E departments. https://www.nice.org.uk/guidance/gid-sgwave0762/docu ments/accident-and-emergency-departments-draft-guideline2. [Accessed 2 November 2017]. [55] England NHS. A&E attendances and emergency admissions 2016. https:// www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2015/08/MonthlyAE-Report-January-16.pdf. [Accessed 28 November 2017].
7.e1
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
Appendix A. Arthroplasty Care Practitioner Nurse Roles The ACP represents multidisciplinary health professionals involved in the care of patients with a joint arthroplasty. They provide care at one or more points in patient pathway from referral for surgery, through preoperative screening, perioperative and postoperative care, and throughout long-term follow-up. They work along side orthopedic surgeons, running parallel review clinics to consultant-led new patient clinics. This minimizes review back log and allows discussion with a consultant of any problems in the parallel review clinic. Additionally, validated outcome score (Oxford Scores) are collected at review for hip and knee arthroplasty. In their role, ACPs conduct a 5-day follow-up telephone call. During this consultation, the patients are asked about: (1) Pain using visual analog scale 0-10. (2) Inquire about swelling, increasing pain at rest, and ability to weight bear ± use of walking aides and if patient is experiencing falls/ trauma since discharge. (3) Inquire about the range of movement. (4) Wound issuesdensure no hematoma, sign of infection, or wound dehiscence. If any of these are suspected, then the patient is recalled for review. (5) Oozedprolonged ooze for >10 days or new ooze from previously dry wound necessitates review. (6) VTE screening for DVT (calf pain swelling, redness) and PE (pleuritic chest pain, shortness of breath, hemoptysis). (7) Screening for suggestion of nerve injurydneuralgic pain, paresthesia/anesthesia and so forth (8) Inquire about any readmissions/reattendances to hospital. (9) General patient impression that they are improving. (10) Remind them of routine review appointments (6 weeks for THA and 3 months for TKA). (11) Reminder of contact number for helpline for any concerns in interim.
Costings for Helpline Access and 5 d Follow-Up Phone Call.
No of patient No of calls Calls per patient Average call duration (min) Calls per hour Total time (h) Nursing staffing cost (£) Hourly cost (£) Cost per patient (£) Band 6 ACP annual salary
Helpline
5-d Phone Call
2351 3232 1.37 10 6 539 12370 22.95 5.26
560 560 1 15 4 140 3210 22.93 5.73 £44,876
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Postdischarge Unscheduled Care Burden After Lower Limb Arthroplasty Adam Tucker, MPhil, MRCS, MB a, *, Andrew Walls, MB, BCh, MRCS b, Beverley Leckey, BSc b, Janet C. Hill, PhD, MEng, BSc a, Glenn Phair, BSc, PgDip c, Damien B. Bennett, MB, MPH, PhD d, Seamus O’Brien, PhD a, David E. Beverland, MD, FRCS (Orth) b a
Department of Orthopaedics, Outcomes Unit, Musgrave Park Hospital, Belfast, County Antrim, United Kingdom Department of Orthopaedics, Musgrave Park Hospital, Belfast, County Antrim, United Kingdom c Northern Ireland Clinical Trials Unit (NICTU), Royal Victoria Hospital, Belfast, County Antrim, United Kingdom d Public Health Agency, Belfast, County Antrim, United Kingdom b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 March 2018 Received in revised form 18 April 2018 Accepted 23 April 2018 Available online xxx
Background: In contrast to postdischarge arthroplasty readmission rates, the unscheduled reattendance burden to primary care is under-reported. Understanding reasons for reattendance would allow for implementation of strategies to reduce this burden. The present study aims to quantify the out-of-hours (OOH) general practitioner and emergency department (ED) service reattendance burden and readmission rate after primary total hip arthroplasty and total knee arthroplasty, with estimation of the associated costs. Methods: This is a prospective consecutive cohort study. A prospective audit of all total hip arthroplasty and total knee arthroplasty patients in 2016 in a single high-volume UK arthroplasty unit was performed. Incidence and reasons for reattendance to OOH and ED service, as well as readmission rates, at both 30 and 90 days following discharge are reported. A multivariate analysis was performed to determine patient characteristics, which results in increased reattendance and readmission rates. Results: A total of 2351 procedures resulted in 374 attendances of OOH service and 665 to ED with a total estimated cost of £190,000 within 90 days. The readmission rate was 6.8%. Risk factors for reattendance and readmission were increasing age and a prolonged length of stay. The use of a 5-day postdischarge phone call and a dedicated Arthroplasty Care Practitioner favors reduced reattendances but not the readmission rate, with the additional benefit of being cost-effective. Conclusion: The postdischarge arthroplasty reattendance burden is associated with significant costs, and strategies to reduce this should be developed. Further research is required to assess the effectiveness and cost-effectiveness of multicomponent strategies to reduce reattendance operating at scale. © 2018 Elsevier Inc. All rights reserved.
Keywords: arthroplasty reattendance readmission unscheduled care reduction strategies
As numbers of hip and knee arthroplasties increase in the UK [1] and worldwide, so does economic pressure on health-care systems. Hospital readmission rates at 30 or 90 days after surgery are
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2018.04.041. * Reprint requests: Adam Tucker, MPhil, MRCS, MB, Department of Orthopaedics, Outcomes Department Unit, Musgrave Park Hospital, Stockmans Lane, Belfast BT9 7JB, County Antrim, United Kingdom. https://doi.org/10.1016/j.arth.2018.04.041 0883-5403/© 2018 Elsevier Inc. All rights reserved.
increasingly recognized as key care indicators. Although a reduced length of stay has definite financial benefits [2], this should not be at the cost of increased readmission rates. Readmission rates have been reported between 2.2% and 9.0% after lower limb arthroplasty, with higher rates observed after knee arthroplasty [2e8]. It has been estimated that up to 80% of hospital readmissions are potentially avoidable [9], with substantial cost savings [10]. Multidisciplinary enhanced recovery and facilitated discharge programs have been developed to avoid unplanned readmissions [11,12]. Direct comparisons of readmission rates between different hospitals or health-care systems are complicated by differences in criteria used for example, 30- or 90-day or other readmission rates
2
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
[13e15], with similar limitations reported in recent meta-analyses [16]. Also some count the interval from the date of surgery and others from the date of discharge [13e15]. Although readmission rates are widely reported [4,5,12e17], rate of attendance at OOH and emergency department (ED) is not. Understanding the reasons for these unscheduled health-care contacts could allow targeted changes to patient pathways, which could reduce these attendances and costs, and improve patient care and outcomes. The following are the aims of this audit: (1) To report the incidence of, and reasons for, unscheduled contact with OOH and ED attendances and hospital readmissions within 30 and 90 days from the date of surgery in a large consecutive cohort of lower limb arthroplasties. (2) To assess the impact of proactive Arthroplasty Care Practitioner (ACP) telephone support on unscheduled attendance and readmission rates compared with helpline access and usual care. (3) To compare estimated costs between these care models. We also discuss how unscheduled contacts could be further reduced. Methods A single regional orthopedic service in a high-volume arthroplasty unit was evaluated. We collected demographic and surgical information as well as incidence of, and reasons for, OOH and ED attendances and hospital readmissions within 30 and 90 days from the date of discharge in 2351 consecutive primary total hip and knee arthroplasties, performed under the care of 25 different consultants from January 1 2016 to December 31, 2016. This service uses the Belfast Orthopaedic Information System, which collects patient demographics, intraoperative data, and episode-specific outpatient data after lower limb arthroplasty. The electronic care record (ECR) records patient contact across health-care facilities in Northern Ireland, except for scheduled attendances to the patients' own general practitioner. Initial points of contact for patients experiencing postoperative difficulties are their base hospital, their general practitioner (GP), OOH, or ED. Within our hospital, primary arthroplasty is performed under the care of 25 different consultants with variation in management on advice to patients following discharge if they experience problems. This falls broadly into 3 follow-up groups: group A in which patients routinely receive a 5day phone call with advice to contact a “helpline” phone number; group B in which they can access a helpline only; and group C in which they have neither resource and are advised to contact their GP. The unit helpline and 5-day phone calls are performed by a specialist ACP nurse. These nurses function as part of the wider orthopedic multidisciplinary team, collecting follow-up and outcomes data, facilitating nurse-led review, and providing a helpline service for patients with postoperative problems (Supplementary Appendix A). As part of a unit audit process, Belfast Orthopaedic Information System and ECR were interrogated. Local audit approval was gained (BHSCT Orthopaedic Outcomes Ref 5626). Local imaging systems (picture archiving communication systems) were interrogated for episodes of hip dislocation and suspected and/or confirmed venous thromboembolic events (VTEs) (ultrasound Doppler for deep vein thrombosis [DVT] and computed tomography pulmonary angiogram [CTPA] for pulmonary embolism [PE] investigations). This facilitated comprehensive collection of complications including mortality, which were cross-referenced with the ECR, within 90 days of surgery. Health economic analysis was performed using unit costs per attendance at both OOH GP and ED services. Patients use of hospital services were combined with
corresponding unit costs to estimate local costs per attendance for patient follow-up groups A, B, and C. Unit costs were obtained from publicly available sources set at 2015/2016 prices [18]. Statistical Analysis Continuous data were tested for normality using the Kolmogorov-Smirnov test. Differences between groups were assessed using Student's t test or analysis of variance test for continuous variables and chi-square, Kruskal-Wallis, or Mann Whitney U tests for nonparametric data. Multivariate logistic regression was used to investigate which variables were significantly associated with attendance at unscheduled cares services or readmission. For multivariate analysis, variables with moderate associations (P value < .20) on univariate analysis or of a priori importance (eg, body mass index) were included. For all tests, a P value < .05 was considered statistically significant. All analyses were performed using SPSS v22 for Mac (IBM Inc., Armonk, NY). Results Patient Characteristics There were 1341 total hip arthroplasties (THAs) and 1010 total knee arthroplasties (TKAs) (1 simultaneous bilateral). Patient characteristics are shown in Table 1. Procedures were performed under the care of 25 consultant orthopedic surgeons and prospectively followed-up to 90 days from the date of surgery. Chemical VTE prophylaxis was predominately aspirin or enoxaparin with 489 patients receiving aspirin; 1858, enoxaparin; 1, warfarin; and 3 patients receiving no chemical prophylaxis. All patients received mechanical prophylaxis with calf pumps. Average hospital length of stay was 4.4 ± 2.9 days. Mean time to first unscheduled contact was 25.4 ± 24.6 days. There were 2 deaths within 90 days (0.09%), both occurring in female TKA patients with American Society of Anaesthesiologists (ASA) grade 2 who were discharged on day 3 with no initial complications. One died 10 days postoperatively from a perforated abdominal viscus following emergency laparotomy. The second died of PE 55 days postoperatively, having been on aspirin VTE prophylaxis for 6 weeks postoperatively. Unscheduled Care Attendance Out-of-Hours General Practitioners There were 284 (12.1%) patients who were attended by OOH GPs within 90 days of discharge, creating 374 OOH attendances.
Table 1 Demographic Data. Variable
THA
TKA
N Male Female Age (mean ± SD) BMI (kg/m2) (mean ± SD) ASA 1 2 3 4
1341 592 (44.1%) 749 (55.9%) 66.7 ± 11.7 29.8 ± 7.3
1010 411 (40.7%) 599 (59.3%) 69.0 ± 9.3 32.2 ± 5.9
127 1032 177 5
53 797 153 7
(9.5%) (76.9%) (13.2%) (0.4%)
P Value
(5.2%) (78.9%) (15.1%) (0.7%)
.102 <.001* <.001* <.001*
ASA, American Society of Anaesthesiologists; BMI, body mass index; SD, standard deviation; THA, total hip arthroplasty; TKA, total knee arthroplasty. * Denotes significant differences at P < .05 level.
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
3
Table 2 Reasons for Reattendance and Readmission at 90 d. OOH (n ¼ 284)
Reason THA Medical Suspected VTE Wound Surgery related Nonsurgery related Other musculoskeletal Total P value
85 (55.9%) 46 (30.3%) 9 (5.9%) 8 (5.3%) 1 (0.7%) 3 (2.0%) 152 (11.3%) .936
ED (n ¼ 353)
TKA 66 47 8 9 2
(50.0%) (35.6%) (6.1%) (6.8%) (1.5%)
0 132 (13.1%)
THA
TKA
59 (31.2%) 110 (58.2%) 4 (2.1%) 8 (4.2%) 1 (0.5%) 7 (3.7%) 189 (14.1%) .749
50 87 8 8 1 10 164
Readmission (n ¼ 159)
(30.5%) (53.0%) (4.9%) (4.9%) (0.6%) (6.1%) (16.2%)
THA
TKA
45 (56.9%) 15 (19.0%) 7 (8.9%) 9 (11.4%) 1 (1.3%) 2 (2.5%) 79 (5.9%) .873
49 14 10 6 1 0 80
(61.3%) (17.5%) (12.5%) (7.5%) (1.3%) (7.9%)
ED, emergency department; OOH, out-of-hours; THA, total hip arthroplasty; TKA, total knee arthroplasty.
The proportions attending OOH GPs did not differ significantly after THA (11.3%, 152/1341) compared with TKA (13.1%, 132/1010, P ¼ .201) (Table 2). Similarly, there was no difference between reasons for OOH GP reattendance between surgical procedures (P ¼ .936) (Table 2). Of patients seen by OOH GPs, 54/284 (19%) were immediately referred to EDs. Thirty had suspected VTE, comprising 26/30 (86.7%) DVT and 4/30 (13.3%) PE, 19 medical problems, 4 wound infections, and 1 dislocation of THA. Emergency Departments There were 353 (15.0%) patients who attended EDs, accounting for 665 attendances, within 90 days. The proportions attending EDs did not differ significantly after THA (14.1%, 189/1341) compared with TKA (16.2%, 164/1010, P ¼ .150) (Table 2). There was no difference between reasons for ED reattendance between surgical procedures (P ¼ .749) (Table 2). Reasons for reattendance and readmission are reported in Table 2. Within both OOH and ED groups, there was a significantly higher proportion of >75-year-olds reattending. However, on univariate analysis, there was no difference in any of the other
Table 3 Proportion of Patients Reattending ED and OOH. Variable
ASA 1 2 3 4 N P value (within group) BMI <25 25-30 31-35 >35 Missing data N P value (within group) Age <55 56-64 65-74 >75 N P value (within group) Male Female P value (within group)
n
ED (%)
OOH (%)
P Value (Between Group)
181 1827 329 8 2345
21 (12.1%) 268 (14.7%) 62 (19.5%) 2 (25%) 353 .127
25 (13.8%) 212 (11.6%) 44 (14.1%) 3 (37.5%) 284 <.001*
.441
259 987 642 446 3 2334
37 (14.3%) 150 (15.2%) 99 (15.4%) 64 (14.3%) 3 350 .946
36 (13.9%) 118 (12.0%) 71 (11.1%) 57 (13.2%) 0 282 .532
.670
311 499 863 678 2351
34 (10.9%) 67 (13.4%) 121 (14.0%) 131 (19.3%) 353 .001* 162 (16.2%) 191 (14.2%) .194
35 (11.3%) 57 (11.4%) 89 (10.3%) 103 (15.2%) 284 .028* 111 (11.1%) 173 (12.8%) .183
.661
1003 1348
.099
ASA, American Society of Anaesthesiologists; BMI, body mass index; ED, emergency department; OOH, out-of-hours. * Denotes significant differences at P < .05 level.
variables between patients who attended OOH and those who attended EDs (Table 3). Readmissions There were 159 (6.8%) patients readmitted within 90 days of discharge, with 98/2351 (4.2%) readmitted within 30 days. Mean time to readmission was 30.4 ± 26.3 days (median 23 days; interquartile range 7.0-45.2 days, range 1-87 days). There was no significant difference in 30- or 90-day readmission rates between THA (3.8% and 5.9%) and TKA (4.7% and 7.9%) patients (P ¼ .052). There was no difference in 90-day readmission rates between males (7.3%, 73/1003) and females (6.4%; 85/1348, P ¼ .391). Patients readmitted had a longer initial length of stay (5.1 ± 3.1 vs 2.9 ± 4.4 days; P ¼ .001). Reasons for readmission are given in Table 2. A return to theater within 90 days occurred in 20 cases, with surgery-related complications (infection and dislocation) accounting for 11/20 (55.0%) of cases (Table 4). Of note, the reported dislocation rates are only for those returning to theater for open/closed reduction, within 90 days. On multivariate logistic regression only age > 75 years (odds ratio 2.367 [95% confidence interval 1.253-4.472]; P ¼ .008) was significantly associated with readmission. ASA grade, gender, body mass index, length of stay, and all other age bands were not significantly associated with an increased risk of readmission (Table 5). Service Model Comparison There were 3 groups of patients who experienced different service models for both procedures (THA, TKA). Group A (n ¼ 560) received a 5-day phone call postdischarge and had access to a telephone helpline; group B (n ¼ 647) had access to the helpline only; and group C (n ¼ 1144) had neither. Length of stay was significantly shorter in group A compared with both groups B and C (3.6 ± 2.6 vs 4.5 ± 3.2 and 4.8 ± 2.8 days, respectively; P < .001). Group A had lower OOH and ED reattendance rates, without increased readmission rates at both 30 and 90 days postoperatively (Table 6). Overall, 30.5% of patients in group C who experienced
Table 4 Reasons for Return to Theater Within 90 d. Variable
THA (n ¼ 1341)
Washout for infection Dislocation Periprosthetic fracture Other surgical intervention not related to primary surgery Total
2 4 1 1
(0.15%) (0.30) (0.07%) (0.07%)
8 (0.60%)
THA, total hip arthroplasty; TKA, total knee arthroplasty.
TKA (n ¼ 1010) 5 (0.50%) 0 3 (0.30%) 4 (0.4%) 12 (1.19%)
4
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
Table 5 Multivariate Regression Analysis for Unscheduled Reattendance to OOH and ED and for Readmission. Variable
OOH
ED
OR (95% CI) Gender Male Female Age overall <55 55-65 66-75 >75 BMI overall <25 25-30 31-35 >35 Length of stay ASA Surgery Hip Knee Follow-up group A B C
P
1
Readmission
OR (95% CI)
P -
OR (95% CI)
P
1 0.790 1.028 1 1.051 1.359 2.367 0.993 1 1.030 0.991 1.176 1.035 1.372
(0.568-1.100) (1.010-1.047)
.162 .003*
(0.523-2.114) (0.725-2.549) (1.253-4.472) (0.966-1.021)
.889 .339 .008* .626
(0.590-1.799) (0.549-1.791) (0.626-2.210) (0.984-1.088) (0.978-1.926)
.917 .977 .614 .184 .067
(0.869-1.456) (0.993-1.019)
.372 .630
(0.634-1.573) (0.563-1.321) (0.808-1.949) (0.970-1.014)
.997 .497 .312 .453
1 0.802 1.014 1 1.253 1.277 1.648 0.996
(0.573-1.303) (0.503-1.226) (0.589-1.518) (1.017-1.098) (0.747-1.286)
.485 .288 .816 .005* .883
1.042 1.045 0.993 1.033 1.129
1 0.886 (0.683-1.150)
.364
1 0.886 (0.699-1.123)
.318
1 0.804 (0.575-1.125)
.203
1 1.246 (0.853-1.819) 1.477 (1.054-2.069)
.255 .023*
1 1.124 (0.799-1.581) 1.330 (0.984-1.798)
.501 .064
1 0.845 (0.536-1.333) 0.877 (0.587-1.310)
.470 .521
1.12 1.006 1 0.999 0.862 1.255 0.992 1 0.864 0.786 0.945 1.057 0.980
(0.635-1.013) (1.001-1.026)
.064 .028*
(0.799-1.964) (0.840-1.941) (1.067-2.546) (0.977-1.015)
.326 .252 .024* .676
(0.699-1.553) (0.684-1.596) (0.627-1.574) (1.029-1.105) (0.885-1.442)
.839 .839 .976 <.001* .329
ASA, American Society of Anaesthesiologists; BMI, body mass index; CI, confidence interval; ED, emergency department; OOH, out-of-hours; OR, odds ratio. * Denotes significant differences at P < .05 level.
standard care (no 5-day phone call or helpline access) either attended by OOH GPs/EDs or were readmitted to hospital within 90 days of discharge. Unscheduled care attendance was significantly reduced for patients with helpline access (group B; 25.8%) and further reduced for those receiving a 5-day phone call along with helpline access (group A; 21.6%, P ¼ .002) (Table 6). In multivariate analysis, the likelihood of OOH GP attendance was significantly higher for group C than for group A (odds ratio ¼ 1.5, 95% confidence interval 1.05-2.07). There was no difference in the likelihood of attendance at EDs or for readmission between the 3 groups (Tables 5 and 6). This suggests that the service received by
group A reduced the risk of unscheduled OOH GP attendance without increasing the risk of ED attendance or readmission (Tables 5 and 6). Estimated Costings for Unscheduled Care Attendance The average cost per patient of the service models ranged from £37.03 in group A to £51.42 in group C. The lowest total cost and lowest cost per patient was in group A where patients receive a 5day follow-up phone call and have helpline access. This was due to reduced unscheduled care attendance. We also estimated the costs of running the ACP-led 5-day follow-up phone call and helpline
Table 6 Thirty- and 90-d Readmission Rates by Follow-Up Group (Unit Cost per OOH £83; Unit Cost per ED Attendance £238). Variable
n
Group A (n ¼ 560)
Group B (n ¼ 647)
Group C (n ¼ 1144)
Male (%) Female (%) Age BMI ASA I II III IV Length of stay (mean meSD) At 30 d OOH attendances OOH cost ED attendances ED cost Readmissions At 90 d OOH attendances Cost ED attendances Cost Combined OOH/ED reattendance burden Total cost for OOH þ ED Average cost per patient Readmissions
2351
220 (39.3%) 340 (60.7%) 68.7 ± 9.7 31.0 ± 9.2
274 (42.3%) 373 (57.7%) 67.1 ± 11.7 31.2 ± 6.4
509 (44.5%) 635 (55.5%) 67.6 ± 10.7 30.5 ± 5.5
.122
38 458 62 2 3.6 ± 2.7
47 485 111 4 4.5 ± 3.2
95 886 157 6 4.8 ± 2.8
.090
37 (6.6%) £3071 40 (7.1%) £9520 29 (5.2%)
55 (8.5%) £4565 50 (8.5%) £11,190 21 (3.2%)
111 (9.7%) £9213 128 (9.7%) £30,464 48 (4.2%)
52 (9.3%) £4316 69 (12.3%) £16,422 121 (21.6%) £20,738 £37.03 42 (7.5%)
76 (11.8%) £6308 91 (14.1%) £21,658 167 (25.8%) £27,966 £43.22 41 (6.4%)
156 (13.6%) £12,948 193 (16.9%) £45,934 349 (30.5%) £58,882 £51.42 76 (6.6%)
2351 2351 2351
2351 203 218 98 284 353 637
159
ASA, American Society of Anaesthesiologists; BMI, body mass index; ED, emergency department; OOH, out-of-hours; SD, standard deviation. * Denotes significant differences at P < .05 level.
P Value
.031* .084
<.001* .101 .007* .245 .033* .035* <.001*
.707
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
service. Using unit data (Appendix Supplementary File A), we calculated the cost of the 5-day phone call at £5.73 per patient. The cost of the helpline service was £5.26 per patient, based on an average 1.37 phone calls to the helpline per patient and £3.84 unit cost per phone call. Thus, the total cost of running the service received by group C patients was £10.99 per patient. To calculate the total cost of the service models, we added the costs of unscheduled care attendance to the costs associated with that service. Adding the cost of running the service to the cost of unscheduled care attendance, the total costs of the service model experienced by group A was £48.02, which is less than that of group C (£51.42). The cost of running the service for group B patients (helpline only) was £48.48, which is marginally greater than that of for group A but less than that for group C. Venous Thromboembolic Events Of 298 unscheduled attendances for suspected VTE up to 90 days from surgery, 195 (65.4%) had Doppler ultrasound scanning for DVT with the majority performed after discharge (96.9%, 189/195). Fifty patients attended more than once for therapeutic prophylaxis due to the absence of a same-day Doppler service. Patients were readmitted in 30.2% of cases (59/195) with DVT confirmed by Doppler in 10.8% (21/195) of cases. There was no difference in the proportions with confirmed DVT between THA (1.04%, 14/1341) and TKA patients (0.69%, 7/1010, P ¼ .508) or between those receiving enoxaparin (16/1858) or aspirin (5/489, P ¼ .787). Pulmonary Embolus was suspected in a further 53 patients with 39.6% (21/53) investigated after discharge. All patients had CTPA with 17% (9/53) positive for PE. There was no significant difference in the proportions of patients with a confirmed PE diagnosed on CTPA who had undergone THA (0.30%, 4/1341) or TKA (0.50%, 5/ 1010, P ¼ .511). There was also no statistically significant difference between the choice of VTE prophylaxis using enoxaparin 6/1858 (0.32%) vs aspirin 3/489 (0.61%); P ¼ .406. Discussion This is the first report of unscheduled contact with OOH GPs and EDs in the postoperative period, alongside hospital readmissions, in a large consecutive cohort of UK lower limb arthroplasty patients. The burden of unscheduled contact was high with 15% of patients attending EDs and 12% attending OOH GPs within 90 days of surgery. Readmission rates were similar to previous studies [12e17]. We have shown that a service model involving ACP telephone contact on day 5 postoperatively was associated with significantly lower attendance at OOH GPs and that attendance at EDs and readmission rates were not higher in this group. We estimate this model delivered cost savings of £3.40 per patient, based on reduced unscheduled care use, compared with usual care. Postsurgical readmission rates are key quality and service performance measures and have been adopted as National Health Service (NHS) quality indicators [19]. The cost of 30-day readmissions after surgery in the NHS has been estimated at £1.6 billion annually [20]. Reports of readmission rates vary, ranging from 3.3% at 30 days to 9.7% at 90 days [13,14,16,21,22], but there is a lack of consistency in reporting with, for example, some 30- and 90-day readmission rates measured from the day of surgery and others from the day of discharge [13e16]. In our study, readmission rates at both 30 and 90 days after operation were similar to previous studies with 90-day readmission rates of 5.9% for THA and 7.9% for TKA patients [13,14,16,21,22]. Examining reasons for readmission is important to inform attempts to reduce their occurrence, and previous studies have reported reasons for readmission after THA that were predominantly medical (35%), while those after TKA (75%) were mainly surgery related [13,14]. However, in our study
5
readmissions were mainly for non-VTE related medical reasons with rates similar for THA (57.0%) and TKA (61.3%) patients. However, previous studies have mainly reported readmission rates and ignored the use of unscheduled care by lower limb arthroplasty patients in the early postoperative period. The present study reports both unscheduled care attendances and readmissions and thus comprehensively describes the lower limb arthroplasty postdischarge burden and its distribution in a large-scale study. Unscheduled care attendances were high compared to readmission rates with 14.1% of THA patients and 16.2% of TKA patients attending EDs and 11.3% of THA patients and 13.1% of TKA patients attending OOH GPs within 90 days of surgery, respectively. Considering both OOH GP and ED attendance together, 25.4% of patients of THA patients and 29.3% of TKA patients attended EDs or OOH GPs within 90 days of surgery, which underlines the substantial burden on unscheduled care services. In the present study, 6.8% of patients required readmission in the first 90 days. The most common causes of hospital readmissions were medical (59.1%), followed by suspected VTE (18.2%), wound problems (10.7%), related to surgery (9.4%), unrelated to surgery (1.3%), and finally, other musculoskeletal injury (1.3%). Reasons for attendance at OOH GPs were similar to those to those for readmission to hospital with 53.1% being for medical reasons and 32.7% for suspected VTE. However, for those attending EDs, the most common cause was suspected VTE (55.8%) followed by medical reasons (30.9%). There is conflicting reports on the effect of earlier discharge on complication or readmission rates. Sibia et al [15] reported that 30day (all cause) and 90-day (wound related) readmission rates were not different between those discharged on day 1 and day 2 postoperatively. However, a prolonged hospital length of stay has been shown to be associated with discharge to a nursing facility, compared with discharge to the patients' home with self-care [23]. Our study found that longer length of stay was a significant predictor of both OOH GP and ED attendance but not of readmission to hospital. Increasing age has also been associated with increased risk of readmission [17]. Older age was significantly associated with both ED attendance and hospital readmission in our study, with patients older than 75 years over one and half times more likely to attend an ED and over twice as likely to be admitted to hospital within 90 days of surgery compared with younger patients (<55 years). This is not surprising as multimorbidity rises with age, although we included ASA grade, which is often used a proxy measure of comorbidity, as a covariate in our analysis. OOH GPs are often the initial point of contact for patients postoperatively because of the difficulty in accessing daytime GP appointments. The financial burden of these attending OOH GPs was not inconsequential with an estimated cost of £31,000 for the 284 (12.1%) patients who were attended by OOH GPs on 374 occasions. However, the cost of ED attendances was much greater, with an estimated cost of £158,000 for the 353 (15%) patients who were attended by OOH GPs on 665 occasions. VTE is a major cause of morbidity and mortality worldwide which can prolong inpatient stay [24e26]. DVT is more common than PE, which itself has a reported in-hospital mortality of 5%-10% [27,28]. Up to 24% of VTE cases are estimated to occur postoperatively, with lower limb arthroplasty considered a high-risk procedure [29]. Symptomatic VTE is reported at up to 14%, but this still amounts to a significant workload to primary care [30,31]. In the present study, symptomatic VTE rates for DVT (THA 1.04% vs TKA 0.69%) and PE (THA 0.3% vs KA 0.50%) were comparable to previous reports [32e36]. Although only a small proportion of VTE cases are symptomatic, these can lead to ED attendances and readmissions in up to 6.3% of cases [32,37,38]. Readmissions due to suspected VTE can potentially be reduced using ambulatory
6
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
pathways with substantial cost savings [39,40]. For example, in our study, over a 12-month period, 29 suspected DVT/PE were admitted to hospital overnight for investigation at an estimated direct cost of almost £25,000 (£17,860 bed costs and £7000 for ED attendances). This is the cost equivalent of Doppler outpatient examinations for 350 patients. A dedicated daily outpatient Doppler service would allow timely investigation of up to 93 OOH and 197 ED presentations. In this audit, these patients accounted for approximately 530 attendances, which if avoided could save £108,000 per annum in ED attendances and further save £125,000 if hospital admissions were prevented. Reduced OOH and ED attendances would also be more convenient for patients. A variety of measures, such as preadmission discharge planning, preoperative education [41e43], preoperative screening using the likes of the Risk Assessment and Prediction Tool [44,45], Multidisciplinary Early Supported Discharge pathways, and early clinical review for high-risk patients have been suggested as methods to reduce costs and readmission rates [41,42,46e52]. Overall, the majority of fatal or near fatal complications after lower limb arthroplasty occurs in the first 4 days after index surgery [53]. Our study has shown that a service model involving proactive postoperative telephone contact by ACPs on day 5 significantly reduced attendance at OOH GPs and represents a promising model. The ACP 5-day call and helpline service in this study significantly reduced OOH GP attendance was found to be cost-effective. The ACP, who could be based at a surgical site or ED, provides advice, facilitates early review, triages for unscheduled review, and expedites urgent referral to suitable specialties, if appropriate [54]. Implementation of an ACP-led 5-day call and helpline access, from a hospital service perspective, could save £8016 per year in our unit (based on 2351 patients in 2016) but more importantly would reduce the burden on OOH GPs and EDs as well potentially improving patient experience. The total number of patients attending ED increased 10.1% between January 2015 and January 2016, with corresponding emergency admissions increasing 4.1% [54,55]; the aforementioned strategies may be an attractive option. The NHS is the UK public health-care system. It is a pillar of the welfare state, being directly funded from the general finances of the state, and while working citizens make contributions to this, medical care is provided completely free for all. It is a stretched system, with increasing costs year on year due to an aging population and increasing costs of drugs, equipment, and treatment options. This clearly differs significantly from privatized healthcare systems, and this must be considered when analyzing our results and their validity to other health-care systems. A strength of the present study is its prospective follow-up of a large consecutive cohort of THA and TKA patients who had surgery at a single regional UK center. We report attendances at OOH GPs and EDs alongside hospital readmission rates. To the authors' knowledge, this is the first study to comprehensively quantity the postdischarge burden of primary THA and TKA on unscheduled care services. As outlined in our methodology, we were unable to measure daytime patient attendances at their own GP as our ECR system cannot access this data. We also did not capture the small number of patients who, following helpline calls, were reviewed by an ACP in an outpatient setting, which probably prevented OOH GP or ED attendance. Economic analysis was from a hospital perspective only, with social and community care costs not considered and no quality of life (eg, EQ5D) data collected. The length of stay for the present study may seem excessive. Postoperative length of stay is falling in the UK, but it is generally longer than in the US. One major problem in the UK is the almost absence of preadmission discharge planning and frequently less focus on preoperative patient education. This results in a culture that encourages patient dependence
rather than independence. The study was based in a regional orthopedic unit over a 1-year period and captured data from patients under the care of 25 orthopedic consultants. However, this is probably representative of elective lower limb orthopedic surgery in other UK regions.
Conclusion Attendances at OOH GPs and EDs and hospital readmissions were considerable with 1 in 4 patients attending one of these services within 90 days of operation. The main reasons for attendance at OOH GPs and EDs and for hospital readmissions were medical problems or suspected VTE. We have shown that a service model involving a dedicated ACP help line and proactive patient contact 5 days postoperatively significantly reduced attendance at OOH GPs and was cost-effective. Further studies are required to assess the impact of other interventions, such as dedicated VTE pathways, on unscheduled care use and hospital readmission rates.
References [1] National Joint Registry 13th Annual Report. http://www.njrreports.org. uk/Portals/0/PDFdownloads/NJR%2013th%20Annual%20Report%202016.pdf. [Accessed 6 September 2017]. [2] Avram V, Petruccelli D, Winemaker M, de Beer J. Total joint arthroplasty readmission rates and reasons or 30-day hospital readmission. J Arthroplasty 2014;29:465e8. [3] Cullen C, Johnson DS, Cook G. Re-admission rates within 28 days of total hip replacement. Ann R Coll Surg Engl 2006;88:475e8. [4] Bini SA, Fithian DC, Paxton LW, Khatod MX, Inacio MC, Namba RS. Does discharge disposition after primary total joint arthroplasty affect readmission rates? J Arthroplasty 2010;25:114e7. [5] Vorhies JS, Wang Y, Herndon J, Maloney WJ, Huddleston JI. Readmission and length of stay after total hip arthroplasty in a national Medicare sample. J Arthroplasty 2011;26:119e23. [6] Saucedo JM, Marecek GS, Wanke TR, Lee J, Stulberg SD, Puri L. Understanding readmission after primary total hip and knee arthroplasty: who's at risk? J Arthroplasty 2014;29:256e60. [7] Clement RC, Derman PB, Graham DS, Speck RM, Flynn DN, Levin LS, et al. Risk factors, causes, and the econmimc implications of unplanned readmissions following total hip arthroplasty. J Arthroplasty 2013;28:7e10. [8] Canadian Institute for Health Information 2012: all-cause readmission to acute care and return to the emergency department. https://secure.cihi.ca/ free_products/Readmission_to_acutecare_en.pdf. [Accessed 7 September 2017]. [9] MedPAC Report to the congress: promoting greater efficiency in medicare. http://www.medpac.gov/-documents-/reports. [Accessed 21 November 2017]. [10] Bosco III JA, Karkenny AJ, Hutzler LH, Slover JD, Iorio R. Cost burden of 30-day readmissions following medicare total hip and knee arthroplasty. J Arthroplasty 2014;29:903e5. [11] Raphael M, Jaeger M, van Vlymen J. Easil adoptable total joint arthroplasty programme allows discharge home in two days. Can J Anaesth 2011;58: 902e10. [12] Husted H, Otte SO, Kristensen BB, Orsnes T, Kehlet H. Readmission after fasttrack hip and knee arthroplasty. Arch Orthop Trauma Surg 2010;130: 1185e91. [13] Schairer WW, Sing DC, Vail TP, Bozic KJ. Causes and frequency of unplanned hospital readmission after total hip arthroplasty. Clin Orthop Relat Res 2014;472:464e70. [14] Schairer WW, Vail TP, Bozic KJ. What are the rates and causes of hospital readmission after total knee arthroplasty? Clin Orthop Relat Res 2014;472: 181e7. [15] Sibia US, Waite KA, Callanan MA, Park AE, King PJ, MacDonald JH. Do shorter lengths of stay increase readmissions after total joint replacements? Arthroplasty Today 2016;3:51e5. [16] Ramkumar PN, Chu CT, Harris JD, Athoviraham A, Harrington MA, White DL, et al. Causes and rates of unplanned readmissions after elective primary total joint arthroplasty: a systematic review and meta-analysis. Am J Orthop (Belle Mead NJ) 2015;44:397e405. [17] Bernatz JT, Tueting JL, Anderson PA. Thirty-day readmission rates in orthopaedics: a systematic review and meta-analysis. PLoS One 2015;10:e0123593. [18] Curtis L. Burns a on behalf of the Personal Social Services Research Unit (PSSRU): unit costs of health and social care 2016. http://www.pssru.ac.uk/ pub/uc/uc2016/full.pdf?uc¼2016-full. [Accessed 21 November 2017]. [19] NHS digital: Clinical Commissioning Group Outcomes Indicator Set (CCGOIS): emergency readmissions within 30 days of discharge from hospital. 2015.
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
[20]
[21] [22]
[23] [24]
[25]
[26]
[27] [28] [29]
[30]
[31]
[32]
[33]
[34]
[35]
[36] [37]
https://data.england.nhs.uk/dataset/ccgois-3-2-emergency-readmissionswithin-30-days-of-discharge-from-hospital. [Accessed 30 November 2017]. Robinson P. CHKS hospital readmissions and the 30 day threshold 2010. http://www.chks.co.uk/userfiles/files/CHKS%20Report%20Hospital%20read missions.pdf. [Accessed 21 November 2017]. Zmistowski B, Hozak WJ, Parvizi J. Readmission rates after total hip arthroplasty. JAMA 2011;306:825. Mednick RE, Hasham MA, Krishnan V, Lovecchio F, Manning DW. Factors affecting readmission rates following primary total hip arthroplasty. J Bone Joint Surg Am 2014;96:1202e9. Nichols CI, Vose JG. Clinical outcomes and costs within 90 days of Primary or revision total joint arthroplasty. J Arthroplasty 2016;31:1400e1406.e3. Raskob GE, Angchaisuksiri P, Blanco AN, Buller H, Gallus A, Hunt BJ, et al. Thrombosis: a major contributor to the global disease burden. J Thromb Haemost 2014;12:1580e90. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-based clinical practice guidelines, 8th edition. Chest 2008;133: 381Se453S. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: The Seventh ACCP Conference on antithrombotic and Thrombolytic Therapy. Chest 2004;126:338Se400S. Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep vein thrombosis. Lancet 2012;379:1835e46. Alikhan R, Peters F, Wilmott R, Cohen AT. Fatal pulmonary embolism in hospitalised patients: a necropsy review. J Clin Pathol 2004;57:1254e7. Heit JA, O’Fallon WM, Petterson TM, Lohse CM, Silverstein MD, Mohr DN, et al. Relative impact of risk factors for deep vein thrombosis and pulmonary embolism: a population-based study. Arch Intern Med 2002;162:1245e8. Stulberg BN, Insall JN, Williams GW, Ghelman B. Deep-vein thrombosis following total knee replacement. An analysis of six hundred and thirty-eight arthroplasties. J Bone Joint Surg Am 1984;66:194e201. Ginsberg JS, Turkstra F, Buller HR, MacKinnon B, Magler D, Hirsch J. Postthrombotic syndrome after hip or knee arthroplasty: a cross-sectional study. Arch Intern Med 2000;160:669e72. Samama C-M, Ravaud P, Parent F, Barre J, Mertl P, Mismetti P. Epidemiology of venous thromboembolism after lower limb arthroplasty: the FOTO study. J Thromb Haemost 2007;5:2360e7. Fujita Y, Nakatsuka H, Namba Y, Mitani S, Yashotake N, Sugimoto E, et al. The incidence of pulmonary embolism and deep vein thrombosis and their predictive risk factors after lower extremity arthroplasty: a retrospective analysis based on diagnosis using multidetector CT. J Anesth 2015;29:235e41. Ogonada L, Hill J, Doran E, Dennison J, Steven M, Beverland D. Aspirin for thromboprophylaxis after primary lower limb arthroplasty: early thromboembolic events and 90 day mortality in 11,459 patients. Bone Joint J 2015;98-B:341e8. Cusick LA, Beverland DE. The incidence of fatal pulmonary embolism after primary hip and knee replacement in a consecutive series of 4253 patients. J Bone Joint Surg Br 2009;91:645e8. Allen D, Sale G. Lower limb joint replacement in patients with a history of venous thromboembolism. Bone Joint J 2014;96-B:1515e9. Planes A, Vochelle N, Darmon JY, Fagola M, Bellaud M, Huet Y. Risk of deepvenous thrombosis after hospital dis-charge in patients having undergone total hip replacement: double-blind randomised comparison of enoxaparin versus placebo. Lancet 1996;348:224e8.
7
[38] Merkow RP, Ju MH, Chung JW, Hall BL, Cohen ME, Williams MV, et al. Underlying reasons associated with hospital readmission following surgery in the United States. JAMA 2015;313:483e95. [39] All-Party Parliamentary Thrombosis Group (APPTG): improving outcomes, saving lives, saving money: innovative DVT pathways. http://apptg.org.uk/ wp-content/uploads/2016/12/NHS-Innovation-Showcase.pdf. [Accessed 21 December 2017]. [40] Sampson FC, Goodacre S, Kelly AM, Kerr D. How is deep vein thrombosis diagnosed and managed in UK and Australian emergency departments? Emerg Med J 2005;22:780e2. [41] Ibrahim MS, Khan MA, Nizam I, Haddad FS. Peri-operative interventions producing better functional outcomes and enhanced recovery following total hip and knee arthroplasty: a evidence based review. BMC Med 2013;11:37. [42] Keeney JA, Nam D, Johnson SR, Nunley RM, Clohisy JC, Barrack RL. The impact of risk reduction initiatives on readmission: THA and TKA readmission rates. J Arthroplasty 2015;30:2057e60. [43] Edwards PK, Mears SC, Barnes L. Preoperative education for hip and knee replacement: never stop learning. Curr Rev Musculskelet Med 2017;10: 356e64. [44] Oldmeadow LB, McBurney H, Robertson VJ. Predicting risk of extended inpatient rehabilitation after hip of knee arthroplasty. J Arthroplasty 2003;18: 775e9. [45] Hansen VJ, Gromov K, Lebrun LM, Rubash HE, Malchau H, Freiberg AA. Does the risk assessment and prediction took predict discharge disposition after joint replacement? Clin Orthop Relat Res 2015;473:597e601. [46] Briggs T. A national review of adult elective orthopaedic services in England: getting it right first time. 2015. https://www.boa.ac.uk/wp-content/uploads/ 2015/03/GIRFT-National-Report-Mar15.pdf. [Accessed 24 November 2017]. [47] Mechanic RE. Mandatory medicare bundled payment- Is it ready for prime time? N Engl J Med 2015;373:1291e3. [48] Bradley EH, Curry L, Horwitz LI, Sipsma H, Wang Y, Walks MN, et al. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circ Cardiovasc Qual Outcomes 2013;6:444e50. [49] Kripilani S, Theobold CN, Anctil B, Vasilevskis EE. Reducing hospital readmission: current strategies and future directions. Annu Rev Med 2014;65: 471e85. [50] Hume K, Tomsik E. Enhancing patient education and medication reconciliation strategies to reduce readmission rates. Hosp Pharm 2014;49: 112e4. [51] Tayne S, Merill CA, Smith EL, Mackey WC. Predictive risk factors for 30-day readmissions following primary total joint arthroplasty and modification of patient management. J Arthroplasty 2014;29:1938e42. [52] Jordan CJ, Goldstein RY, Michaels RF, Hutzler L, Slover JD, Bosco III JA. Comprehensive program reduces hospital readmission rates after total joint arthroplasty. Am J Orthop (Belle Mead NJ) 2012;41:E147e51. [53] Parvizi J, Mui A, Purtill JJ, Sharkey PF, Hozack WJ, Rothman RH. Total joint arthroplasty: when do fatal or near-fatal complications occur? J Bone Joint Surg Am 2007;89:27e32. [54] National Institute for Clinical Excellence (NICE) UK 2015: safe staffing for nursing in A&E departments. https://www.nice.org.uk/guidance/gid-sgwave0762/docu ments/accident-and-emergency-departments-draft-guideline2. [Accessed 2 November 2017]. [55] England NHS. A&E attendances and emergency admissions 2016. https:// www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2015/08/MonthlyAE-Report-January-16.pdf. [Accessed 28 November 2017].
7.e1
A. Tucker et al. / The Journal of Arthroplasty xxx (2018) 1e7
Appendix A. Arthroplasty Care Practitioner Nurse Roles The ACP represents multidisciplinary health professionals involved in the care of patients with a joint arthroplasty. They provide care at one or more points in patient pathway from referral for surgery, through preoperative screening, perioperative and postoperative care, and throughout long-term follow-up. They work along side orthopedic surgeons, running parallel review clinics to consultant-led new patient clinics. This minimizes review back log and allows discussion with a consultant of any problems in the parallel review clinic. Additionally, validated outcome score (Oxford Scores) are collected at review for hip and knee arthroplasty. In their role, ACPs conduct a 5-day follow-up telephone call. During this consultation, the patients are asked about: (1) Pain using visual analog scale 0-10. (2) Inquire about swelling, increasing pain at rest, and ability to weight bear ± use of walking aides and if patient is experiencing falls/ trauma since discharge. (3) Inquire about the range of movement. (4) Wound issuesdensure no hematoma, sign of infection, or wound dehiscence. If any of these are suspected, then the patient is recalled for review. (5) Oozedprolonged ooze for >10 days or new ooze from previously dry wound necessitates review. (6) VTE screening for DVT (calf pain swelling, redness) and PE (pleuritic chest pain, shortness of breath, hemoptysis). (7) Screening for suggestion of nerve injurydneuralgic pain, paresthesia/anesthesia and so forth (8) Inquire about any readmissions/reattendances to hospital. (9) General patient impression that they are improving. (10) Remind them of routine review appointments (6 weeks for THA and 3 months for TKA). (11) Reminder of contact number for helpline for any concerns in interim.
Costings for Helpline Access and 5 d Follow-Up Phone Call.
No of patient No of calls Calls per patient Average call duration (min) Calls per hour Total time (h) Nursing staffing cost (£) Hourly cost (£) Cost per patient (£) Band 6 ACP annual salary
Helpline
5-d Phone Call
2351 3232 1.37 10 6 539 12370 22.95 5.26
560 560 1 15 4 140 3210 22.93 5.73 £44,876