Abstract
Purpose
Day-zero ambulation may enable patients to recover and leave hospital quicker following total hip replacement (THR). The present randomised control feasibility study investigated the efficacy of day-zero ambulation as a physiotherapeutic intervention.
Methods
Thirty-six non-blinded adults aged 44–85 (Mean 67.1; SD 9.6 years) undergoing primary, uncomplicated THR were block randomized to either a control group (n = 18) with standard post-operative physiotherapy or an intervention group (n = 18) incorporating walking on the same day as the operation. Outcomes were length of hospital stay (LOS), time to reach functional milestones and achieve all physiotherapy discharge criteria, post-operative pain scores, complications and patient experience.
Results
Participants treated with day-zero ambulation had reduced median hospital LOS of 1 day (p = .096), and median reduced times to reaching functional milestones of 39.7 h quicker to transfer to a chair (p < .001), 24.5 h quicker to walk 10 m (p = .009) and 26.4 h quicker to independently ascend and descend stairs (p = .01). Participants in the intervention group were deemed physiotherapy ready to leave hospital significantly earlier than control group (1.04 days, p = .015).
Conclusions
Day-zero ambulation appears safe and may have clinically relevant effects in speeding patient functional recovery and facilitating earlier discharge from hospital.
Day-zero ambulation following total hip replacement (THR) appears safe.
Preliminary data suggest that day-zero mobilisation following THR could be efficacious and support the need for a fully powered randomised controlled trial.
There may be a clinically relevant effect in speeding patient functional recovery and facilitating an earlier discharge from hospital.
Implications for Rehabilitation
Introduction
It was estimated that 2.12 million people were living in the UK with osteoarthritis of the hip joint, with increasing prevalence in the population over the age of 45 [Citation1]. Updated figures in 2019 estimate this number to have risen to 2.76 million [Citation2]. One of the most successful treatments for osteoarthritis of the hip is joint replacement surgery or total hip replacement (THR). Factors such as advancing age and obesity have been linked with an increased need for joint replacement surgery [Citation3], and forecasts suggest that the UK population could have 11 million more obese adults by 2030 [Citation4] and double the number of adults over the age of 80 by 2030 [Citation5]. Subsequently, UK orthopaedic services can anticipate a significant increase in the demand for hip replacement surgery, and the efficiency of these services will be critical to maintaining clinical standards. This is a trend already seen in the UK, with the number of primary hip replacement procedures carried out increasing by 62.9% from 2006 to 2018 [Citation6].
Consequently, investigating ways of improving efficiency of recovery for patients undergoing THR surgery and improving clinical outcomes would provide better public value, future-proofing and patient-centred quality improvements for UK orthopaedic services. Post-operative physiotherapy is a key component in assisting the patient to reach functional independence post operatively, with a view to facilitating a safe discharge from hospital and a key component in current guidelines on post-operative rehabilitation [Citation7]. There is evidence that providing more inpatient physiotherapy sessions allows earlier achievement of functional milestones [Citation8] and poorer probability of early discharge where physiotherapy is not provided over weekends [Citation9]. This suggests that enacting physiotherapy treatment earlier in the patient’s post-operative recovery could result in a reduction in the time the patient is required to remain in hospital following their THR.
Several published papers have examined ambulation on the day of surgery as a component in an enhanced recovery pathway following major joint arthroplasty and shown reductions in length of hospital stay (LOS) [Citation10–15], no increases in post-operative complications [Citation16–22] and reductions in post-operative pain [Citation13,Citation23,Citation24].
Despite these studies giving promising indications, there were various methodological drawbacks, such as concomitant pathway changes (including analgesia and surgical techniques), convenience sampling with no randomisation and amalgamation of results across different types of lower limb arthroplasty. These drawbacks limit their transferability into determining the efficacy of day-zero ambulation specific to THR. In addition, outcome measures such as patient experience have seen minimal inclusion in these studies.
The aim of this study was to give an indication of the effect of day-zero ambulation within a randomised controlled trial (RCT) methodology specific to THR and provide data for use in sample size calculations to inform a fully powered RCT. This feasibility study has been written according to CONSORT guidance.
Methods
This feasibility study was conducted using a single-centre, non-blinded, parallel group RCT design. The trial was approved by the Hampshire B research ethics committee and hospital research and development prior to commencement and was conducted within the guidance of the declaration of Helsinki [Citation25]. The trial was registered with ClinicalTrials.Gov under identifier NCT02428829. A feasibility study type was selected to identify methodological amendments and inform sample size calculations required for the success of a fully powered RCT.
Thirty-six adults who were undergoing primary unilateral THR were recruited from a single orthopaedic ward specialised in major joint replacement surgery. Participants were identified from theatre lists and provided with the study information approximately 1 week prior to their surgery. All participants gave their written and informed consent prior to anaesthesia and were enrolled following return from theatre according to the inclusion and exclusion criteria detailed in .
There were no differences in theatre scheduling between groups. Randomisation was carried out by the study sponsor using a non-stratified, computer-generated block randomisation method. Participants were randomised on a 1:1 ratio into either “standard post-operative physiotherapy” or “rapid ambulation physiotherapy” groups using a sequential sealed envelope method resulting in 18 participants in each group.
Treatment regimens were the same for the two groups with the exception of ambulation beginning on day 0 in the Rapid group, and on day 1 in the Control group in line with standard care. All participants attended a pre-admission education session detailing the recovery from surgery and what to expect. There was no pre-operative rehabilitation programme or walking aid teaching provided for either group. All participants underwent their arthroplasty via posterolateral approach, under spinal anaesthetic and were subject to standard posterior approach anti-dislocation precautions. A standardised analgesic protocol was used in both groups, and there was no use of local infusion analgesia or wound drainage catheters. All participants underwent standard thromboprophylaxis of sub-cutaneous Dalteparin for 35 post-operative days and wore compression stockings for 6 weeks post-operatively. Full treatment protocols are supplied within the study Supplementary material.
Baseline data on age, gender and number on the day’s theatre list were collected for reporting and baseline comparison. Outcome measures were LOS in days, time to reach functional milestones shown in measured from the time returned to the ward from theatre, incidence of post-operative complications, post-operative pain and patient experience measured using a numerical survey of individual questions (supplied within Supplementary material).
Data were analysed using Statistical Package for the Social Sciences Version 20 [Citation26]. The data were analysed for normality using the Shapiro–Wilk test and found to be non-normally distributed. Descriptive statistics were used to summarise the results as medians and interquartile ranges (IQR). Mann–Whitney tests were used to give an indication of statistical significance, but not to draw conclusions as this is a feasibility study and expected to be underpowered.
Results
Participant characteristics
The baseline characteristics of the sample are shown in . There were no significant inter-group differences in age or number on scheduled theatre list. There were a higher proportion of females seen in both groups. These baseline characteristics were consistent with UK national data for THR, both for age and gender [Citation27].
Length of hospital stay and functional milestones
LOS within the RAPID group was observed as a median of 1 day shorter than observed within the control group, but as expected within a feasibility study this did not reach statistical significance (p = .096). Time until participants were deemed to have achieved all physiotherapy goals for discharge indicated a clinically and statistically significant median difference with this milestone being reached 1.04 days quicker in the RAPID group than in the control group (p = .015).
For first hip joint ROM, there was no significant difference between groups (p = .302). This would be expected due to no interventional differences between groups at this milestone. However, all other functional milestones were achieved quicker within the RAPID group, with all differences reaching statistical significance as shown in .
Post-operative pain
Overall, higher pain numerical rating scores (PNRS) were observed on the day-of-surgery for participants within the RAPID group when compared to controls. However, this trend was reversed on post-operative day 1 both at rest and on hip joint range-of-motion. Comparison testing however indicated no significant differences between groups.
When comparing the first time participants walked post-operatively, in the RAPID group, mean PNRS scores were observed as 1.16 points lower compared to the control group; however, this did not reach statistical significance (p = .582).
Post-operative complications
Post-operative complications were classified for severity using the Clavien–Dindo classification system [Citation28]. Incidences of post-operative complications are shown in .
There were three incidences of post-operative complications classified as higher than grade I. One instance of pulmonary embolism, one instance of low haemoglobin requiring blood transfusion and one instance of apnoea that required intubation and ventilation, with an admission to intensive care. All of these serious complications were observed within the control group.
Orthostatic hypotension (defined using the Freeman et al. [Citation29] published consensus definition of a sustained reduction in systolic blood pressure of at least 20 mmHg or diastolic blood pressure of at least 10 mmHg within 3 min of postural challenge) was the most common complication in both groups, seen in nine different participants in total, with one patient in the intervention group experiencing three instances during their inpatient hospital stay. Although the consensus definition states that orthostatic hypotension may be symptomatic or asymptomatic, in this study only participants who experienced symptoms of orthostatic hypotension were investigated and the complication confirmed with non-invasive blood pressure readings.
Post-operative nausea and vomiting was the next most common complication to affect both groups, being observed in four of the control group and two of the intervention group, one of which experienced two episodes.
There were two incidences of post-operative wound ooze within the RAPID group, while none observed within the control group; both instances led to the delayed discharge of these individual subjects.
Patient experience
There was no missing data, with questionnaires completed for all 34 participants. All of the numerical questionnaire responses suggested no significant differences between intervention and control groups, as shown in .
Discussion
Length of stay, time to physiotherapy complete and functional milestones
This randomised control feasibility study indicated that there could be a clinically significant difference in speed of recovery when day-zero ambulation is employed. While this study was not powered to draw definitive conclusions, time to physiotherapy complete to leave hospital and time to functional milestones showed a statistically significant difference between groups, with LOS coming close to statistical significance, giving a strong indication that proceeding to a fully powered study would be valuable. The effect direction observed in the present study is in agreement with many published works [Citation10–15,Citation23,Citation30,Citation31]. Within these studies, however, there exists significant heterogeneity within both results and methodologies. The effect sizes seen vary significantly, with the highest reduction in length of stay reported at 6.45 days [Citation11], while other studies have shown much smaller effect sizes such as 0.41 days [Citation31] and 0.28 days [Citation30].
Pain numerical rating scale scores
The present study indicated no significant differences between groups in post-operative pain scores. However, being able to provide a faster and more efficient recovery without any increase in post-operative pain would be considered a clinically relevant finding. This is not supported within other published works, with several studies all observing reductions in pain within pathways which included day-zero mobilisation [Citation13,Citation23,Citation24]; however, both Bottros et al. and Raphael et al. employed altered analgesic pathways within their intervention group.
Post-operative complications
This study gives an indication that day-zero ambulation is a safe intervention to employ within this patient cohort with no high severity complications within the intervention group. This is supported by several highly powered observational studies which have shown no significant differences in mortality rates, post-op complication rates or readmissions [Citation20,Citation32–34] with Khan et al. also showing reductions in incidence of myocardial infarction and post-operative blood transfusion rate.
With respect to low severity complications, most complications were seen in fairly equal representations in both groups. However, two previously published studies that employed day-zero ambulation reported increased rates of orthostatic intolerance [Citation35] and an increased incidence of post-operative nausea and vomiting [Citation23], respectively. With this feasibility study, there is not the power in these results to refute the findings of these publications.
The current study did observe two incidences of post-operative wound ooze and bruising around the wound site in the intervention group, with no incidences in the control group. This should be considered when examining the results of a fully powered study as theoretically day-zero ambulation could affect early wound healing and be a direct cause of wound ooze problems. In addition Amlie et al. [Citation36] published findings of increased incidence of early revision arthroplasty due to peri-prosthetic infection. With post-operative wound discharge a known risk factor for prosthetic joint infection [Citation37,Citation38], a fully-powered study would give more confidence as to whether post-operative wound ooze incidence is higher in patients treated with day-zero ambulation.
Patient experience
The present study suggests that treating participants using day-zero ambulation did not adversely affect their experience of inpatient recovery following THR. A fully powered study examining patient experience would be valuable as there are very few publications regarding patient experience of this intervention. Of the studies that have been published, all have summarised experiences across both THR and total knee replacement, and examine experience across an entire enhanced recovery pathway rather than focussing on day-zero ambulation as an intervention [Citation39–43].
Sample size calculation
Using the data gathered within this feasibility study, a sample size calculation was completed following statistician advice and using the software package G*Power 3.1 [Citation44,Citation45]. Sample-size calculation indicates a requirement of 160 participants and is based on a clinically significant difference in LOS showing a median reduction of 1 day.
Limitations
The present study was only sensitive to post-operative complications occurring within the inpatient period following surgery. Some of the common post-operative complications expected within this patient cohort have expected incidence time-courses outside of the inpatient period [Citation46]. Any fully powered study would be strengthened by including a longer follow-up period.
Although participants were all subjected to a standardised post-operative analgesia regime, there was scope within this regime for participants to request and nursing staff to administer pharmacological analgesia appropriate to patient’s pain scores. Consequently, analgesic consumption could have been a hidden confounder in this study’s pain score data. Similarly, while the inclusion and exclusion criteria were deliberately left broad, to enable a cohort representative of the target population, pre-existing medical conditions could be a confounder for some of the outcome measures used. Huang et al. [Citation47] presented increasing numbers of pre-operative co-morbidities as positively correlated with LOS. Patients with a Charlson comorbidity index [Citation48] of ≥2 needed an extra 1.61 days in hospital. The presence of co-morbidity has also been shown to increase the overall care episode cost to THR [Citation49] and increase the risk of moderate to severe post-operative pain related to both medical co-morbidities and anxiety and depression [Citation50].
Data collected from the participant experience questionnaires provided some useful quantitative data for comparing groups across specific questions, with the purpose of generalisability. However, this method of measuring patient experience lacks the richness to be able to draw conclusions from the data, clarify participant thoughts and feelings and subsequently add context to the quantitative findings. Supplementation of quantitative findings with research based on a qualitative methodology would provide better detail and allow more confidence on findings relating to the patient experience.
Conclusions
As a feasibility study, the information in this study cannot be used to draw conclusions. However, there are indications that there could be clinically relevant findings down this avenue of investigation. This study can recommend conducting a fully powered study and provides robust sample size calculations to influence the design of any subsequent study. In conjunction, this study would recommend a longer follow-up period for post-operative complications, measurement or accounting for confounders such as analgesia consumption and pre-operative co-morbidities and exploring qualitative avenues for the investigation of patient experience of this intervention.
Standard_anaesthetic_and_analgesic_pathway_Suppl_4.docx
Download MS Word (187.5 KB)Local_Thromboprophylaxis_Protocol_Suppl_3.docx
Download MS Word (34.8 KB)Supplementary_Material_2.docx
Download MS Word (58.8 KB)Supplementary_Material_1.docx
Download MS Word (18.6 KB)Acknowledgements
The authors would like to thank study participants, clinical orthopaedic staff at the Royal Bournemouth Hospital and acknowledge support from the Faculty of Environmental and Life Sciences, University of Southampton.
Disclosure statement
The author(s) reported there is no funding associated with the work featured in this article.
References
- Arthritis Research UK [Internet]. Osteoarthritis in general practice. Vol. 222. Chesterfield: The Medical Press; 2013 [cited 2014 Jun 24]. p. 253–258. Available from: https://www.versusarthritis.org/media/2115/osteoarthritis-in-general-practice.pdf
- Versus Arthritis [Internet]. The state of musculoskeletal health. Vol. 91. Chesterfield (UK): Versus Arthritis; 2019. p. 31–32.
- Harms S, Larson R, Sahmoun AE, et al. Obesity increases the likelihood of total joint replacement surgery among younger adults. Int Orthop. 2007;31(1):23–26.
- Wang YC, McPherson K, Marsh T, et al. Health and economic burden of the projected obesity trends in the USA and the UK. Lancet. 2011;378(9793):815–825. Available from:(11)60814-3
- Cracknell R [Internet]. The ageing population. Key Issues for the New Parliament in 2010; 2010 [cited 2014 May 6]. p. 1. Available from: http://www.parliament.uk/documents/commons/lib/research/key_issues/Key-Issues-The-ageing-population2007.pdf
- National Joint Registry for England, Wales NI and the I of M [Internet]. Types of primary hip replacements undertaken. National Joint Registry Reports. London (UK): National Joint Registry; 2019 [cited 2020 Aug 12]. Available from: https://reports.njrcentre.org.uk/hips-all-procedures-activity/H03v2NJR?reportid=FFCEA144-54BC-486D-81A6-C6A58DDCA079&defaults=DC__Reporting_Period__Date_Range=%22MAX%22,JYS__Filter__Calendar_Year__From__To=%22min-max%22,R__Filter__Country=%22All%22,H. __Fil
- National Institute for Health and Care Excellence (NICE) [Internet]. Joint replacement (primary): hip, knee and shoulder. NICE Guideline [NG157]. London (UK): NICE; 2020 [cited 2020 Aug 12]. Available from: www.nice.org.uk/guidance/ng157
- Stockton KA, Mengersen KA. Effect of multiple physiotherapy sessions on functional outcomes in the initial postoperative period after primary total hip replacement: a randomized controlled trial. Arch Phys Med Rehabil. 2009;90(10):1652–1657.
- Husted H, Holm G, Jacobsen S. Predictors of length of stay and patient satisfaction after hip and knee replacement surgery: fast-track experience in 712 patients. Acta Orthop. 2008;79(2):168–173.
- Larsen K, Sørensen OG, Hansen TB, et al. Accelerated perioperative care and rehabilitation intervention for hip and knee replacement is effective: a randomized clinical trial involving 87 patients with 3 months of follow-up. Acta Orthop. 2008;79(2):149–159.
- den Hertog A, Gliesche K, Timm J, et al. Pathway-controlled fast-track rehabilitation after total knee arthroplasty: a randomized prospective clinical study evaluating the recovery pattern, drug consumption, and length of stay. Arch Orthop Trauma Surg. 2012;132(8):1153–1163.
- Robbins CE, Casey D, Bono JV, et al. A multidisciplinary total hip arthroplasty protocol with accelerated postoperative rehabilitation: does the patient benefit? Am J Orthop. 2014;43(4):178–181.
- Bottros J, Klika AK, Milidonis MK, et al. A rapid recovery program after total hip arthroplasty. Curr Orthop Pract. 2010;21(4):381–384.
- Tayrose G, Newman D, Slover J, et al. Rapid mobilization decreases length-of-stay in joint replacement patients. Bull NYU Hosp Jt Dis. 2013;71(3):222–226.
- Husted H, Otte KS, Kristensen BB, et al. Fast-track revision knee arthroplasty. A feasibility study. Acta Orthop. 2011;82(4):438–440.
- Berger RA, Sanders SA, Thill ES, et al. Newer anesthesia and rehabilitation protocols enable outpatient hip replacement in selected patients. Clin Orthop Relat Res. 2009;467(6):1424–1430.
- Wellman SS, Murphy AC, Gulcynski D, et al. Implementation of an accelerated mobilization protocol following primary total hip arthroplasty: impact on length of stay and disposition. Curr Rev Musculoskelet Med. 2011;4(3):84–90.
- Gulotta LV, Padgett DE, Sculco TP, et al. Fast track THR: one hospital’s experience with a 2-day length of stay protocol for total hip replacement. HSS J. 2011;7(3):223–228.
- Banerjee P. The efficacy of multimodal high-volume wound infiltration in primary total knee replacement in facilitating immediate post-operative pain relief and attainment of early rehabilitation milestones. Eur J Orthop Surg Traumatol. 2014;24(4):571–577.
- Khan SK, Malviya A, Muller SD, et al. Reduced short-term complications and mortality following enhanced recovery primary hip and knee arthroplasty: results from 6,000 consecutive procedures. Acta Orthop. 2014;85(1):26–31.
- Klein GR, Posner JM, Levine HB, et al. Same day total hip arthroplasty performed at an ambulatory surgical center: 90-day complication rate on 549 patients. J Arthroplasty. 2017;32(4):1103–1106.
- Pollock M, Somerville L, Firth A, et al. Outpatient total hip arthroplasty, total knee arthroplasty, and unicompartmental knee arthroplasty. JBJS Rev. 2016;4(12):1.
- Raphael M, Jaeger M, van Vlymen J. Easily adoptable total joint arthroplasty program allows discharge home in two days. Can J Anaesth. 2011;58(10):902–910.
- Smith TO, McCabe C, Lister S, et al. Rehabilitation implications during the development of the Norwich Enhanced Recovery Programme (NERP) for patients following total knee and total hip arthroplasty. Orthop Traumatol Surg Res. 2012;98(5):499–505.
- World Medical Association [Internet]. Declaration of Helsinki. Vol. 353. Ethical Principles for Medical Research Involving Human Subjects. Helsinki; 1974. Available from: http://www.wma.net/en/30publications/10policies/b3/index.html
- IBM Corp. Released. IBM SPSS Statistics for Windows, Version 20.0. 2011.
- National Joint Registry for England, Wales NI and the I of M [Internet]. Hip replacement edition - PUBLIC AND PATIENT GUIDE. London (UK): National Joint Registry; 2017 [cited 2018 May 1]. Available from: http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/PPG/09736NJRPPG-HIPS2018WEBSPREADS.pdf?ver=2018-02-08-112731-437
- Dindo D, Demartines N, Clavien P-A. Classification of surgical complications. Ann Surg. 2004;240(2):205–213.
- Freeman R, Wieling W, Axelrod FB, et al. Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome. Clin Auton Res. 2011;21(2):69–72.
- Juliano K, Edwards D, Spinello D, et al. Initiating physical therapy on the day of surgery decreases length of stay without compromising functional outcomes following total hip arthroplasty. Arch Orthop Trauma Surg. 2011;7(1):16–20.
- Okamoto T, Ridley RJ, Edmondston SJ, et al. Day-of-surgery mobilization reduces the length of stay after elective hip arthroplasty. J Arthroplasty. 2016;31(10):2227–2230.
- Jorgensen CC, Kehlet H. Role of patient characteristics for fast-track hip and knee arthroplasty. Br J Anaesth. 2013;110(6):972–980.
- Malviya A, Martin K, Harper I, et al. Enhanced recovery program for hip and knee replacement reduces death rate. Acta Orthop. 2011;82(5):577–581.
- Savaridas T, Serrano-Pedraza I, Khan SK, et al. Reduced medium-term mortality following primary total hip and knee arthroplasty with an enhanced recovery program. A study of 4,500 consecutive procedures. Acta Orthop. 2013;84(1):40–43.
- Jans Ø, Bundgaard-Nielsen M, Solgaard S, et al. Orthostatic intolerance during early mobilization after fast-track hip arthroplasty. Br J Anaesth. 2012;108(3):436–443.
- Amlie E, Lerdal A, Gay CL, et al. A trend for increased risk of revision surgery due to deep infection following fast-track hip arthroplasty. Adv Orthop. 2016;2016:7901953.
- Peel TN, Dowsey MM, Daffy JR, et al. Risk factors for prosthetic hip and knee infections according to arthroplasty site. J Hosp Infect. 2011;79(2):129–133.
- Kapadia BH, Berg RA, Daley JA, et al. Periprosthetic joint infection. Lancet. 2016;387(10016):386–394.
- Husted H, Hansen HC, Holm G, et al. What determines length of stay after total hip and knee arthroplasty? A nationwide study in Denmark. Arch Orthop Trauma Surg. 2010;130(2):263–268.
- Specht K, Kjaersgaard-Andersen P, Pedersen BD. Patient experience in fast-track hip and knee arthroplasty – a qualitative study. J Clin Nurs. 2016;25(5–6):836–845.
- Berg U, Berg M, Rolfson O, et al. Fast-track program of elective joint replacement in hip and knee - patients’ experiences of the clinical pathway and care process. J Orthop Surg Res. 2019;14(1):186.
- Department of Health [Internet]. NHS Reference costs 2013-2014. www.gov.uk. 2014. p. 5. Available from: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/380322/01_Final_2013-14_Reference_Costs_publication_v2.pdf
- Sjøveian AKH, Leegaard M. Hip and knee arthroplasty - patient’s experiences of pain and rehabilitation after discharge from hospital. Int J Orthop Trauma Nurs. 2017;27:28–35.
- Faul F, Erdfelder E, Buchner A, et al. Statistical power analyses using G*power 3.1: tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149–1160.
- Faul F, Erdfelder E, Lang A-G, et al. G*Power 3 : a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175–191.
- White RH, Romano PS, Zhou H, et al. Incidence and time course of thromboembolic outcomes following total hip or knee arthroplasty. Arch Intern Med. 1998;158(14):1525–1531.
- Huang CS, Cheu YD, Ying J, et al. Association between provider volume and comorbidity on hospital utilization and outcomes of total hip arthroplasty among national health insurance enrollees. J Formos Med Assoc. 2011;110(6):401–409.
- Charlson M, Pompei P, Ales K, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383.
- Rosas S, Sabeh KG, Buller LT, et al. Medical comorbidities impact the episode-of-care reimbursements of total hip arthroplasty. J Arthroplasty. 2017;32(7):2082–2087.
- Singh JA, Lewallen DG. Medical and psychological comorbidity predicts poor pain outcomes after total knee arthroplasty. Rheumatology. 2013 [cited 2018 Apr 27]52(5):916–923. Available from: https://watermark.silverchair.com/kes402.pdf?token=AQECAHi208BE49Ooan9kkhW_Ercy7Dm3ZL_9Cf3qfKAc485ysgAAAbkwggG1BgkqhkiG9w0BBwagggGmMIIBogIBADCCAZsGCSqGSIb3DQEHATAeBglghkgBZQMEAS4wEQQMfSlgI6K61wp0qFFmAgEQgIIBbCL9hRJZP4xcy_B1FgWG9tp2YMYx8c3JKtZ6CrwTgX337PA5