Use of interrupted time-series analyses in evaluating health economic outcomes following implementation of multilayer water-tight wound closure in a primary total joint arthroplasty population

To our knowledge, this is the first study to employ adjusted ITS techniques to evaluate the clinical and economic outcomes associated with advanced MLWC among patients undergoing TJA. Among 2167 patients under a single surgeon at a single center, unadjusted analyses showed that there were no significant differences in 90-day SSI rates; however, patients with MLWC had 44% lower LOS (1.4 days vs 2.5 days, p < 0.05), 40% lower discharge rates to a skilled care facility (8.5% patients vs 14.1%, p < 0.05), and 60% lower 90-day readmission rates (1.5 vs 3.8%, p < 0.05) compared with patients with CC. Multivariate analyses confirmed that CC was associated with longer LOS, discharge to skilled care, and 90-day readmission. To account for time-dependent factors, e.g., decreasing trends in LOS and shift in discharge status over the study period, ITS analyses were performed to isolate any of these effects and evaluated the direct impact of using MLWC. The adjusted ITS analyses were also performed to account for the difference in patient cohorts of MLWC and CC. The results of the adjusted ITS analyses confirmed the impact of introducing MLWC, showing a sharp decline in LOS (0.9 days) and discharge status to a skilled care facility (5.6% incidence) immediately after implementation of the MLWC approach. The sensitivity analyses showed that the unobserved confounders in this study were not sufficient to explain away the concluded difference between MLWC and CC.

Efforts to improve quality of care and control healthcare costs for THA and TKA have included penalizing hospitals for THA or TKA readmissions and consolidating THA and TKA payments. CMS’ HRRP expansion to THA and TKA was finalized in 2013 and was implemented in 2014. Then, in 2016, comprehensive care with the CJR was implemented for Medicare beneficiaries. Given that MLWC in the current study was implemented by the surgeon in 2014, it was important to elucidate whether the change in outcomes was due to the change in wound care or to other concomitant quality of care changes. ITS analysis is a valuable study design for establishing causality as it controls for pre-existing underlying short- and long-term trends by longitudinally tracking the outcome before and after an intervention [23,24]. ITS analyses showed a consistent downward trend in LOS and discharge to a skilled facility, and then a sharp decline immediately after implementation of the MLWC approach. This confirmed the important impact of introducing MLWC despite shifting trends in LOS and discharge status.

Some studies have shown mixed results when evaluating the impact of THA and TKA quality of care and cost containment strategies [27,28]. It has been suggested that readmissions for patients after elective surgery are more likely a consequence of procedure-related complications rather than failure of care coordination. Hence, systems-level interventions that have proved effective at reducing readmissions for patients with chronic medical diseases (e.g., discharge navigators, care transition programs and medication reconciliation by pharmacists) may not be generalizable to those recovering from elective surgery. Thus, in order to reduce SSI, readmission, LOS and skilled care facility discharge with THA and TKA, providers and payers may need to focus on procedure-related changes that impact complications, such as improved wound care.

The 90-day readmission rate in the current study was significantly lower for MLWC patients (1.5 vs 3.8%; p < 0.05). In 2012, the Hospital Readmissions Reduction Program was introduced by Medicare in the US to reduce the frequency of readmissions. Hospitals with above average readmissions were financially penalized. Furthermore, the Comprehensive Care for Joint Replacement bundled payment model was introduced in 2016 as a mechanism to restrain healthcare spending for total hip and knee arthroplasty (the study hospital participated in this program). Hospitals were provided a single lump-sum payment from Medicare to manage the entire episode of care for a period of 90 days. Hospitals that spent more than the target price or performed below a quality threshold were required to repay Medicare a portion of the episode spending. Therefore these findings are highly economically relevant from payer and provider perspectives. A driver of the observed differences in readmission rates could be the shorter hospital LOS in the MLWC group. A study from 2017 reported that among Medicare patients undergoing knee or hip arthroplasty, increased hospital LOS was a significant risk factor for 90-day readmission because of the increased adverse outcomes associated with increased LOS such as hospital-acquired infections [29]. However, reasons for readmission were not available in the dataset and therefore could not be examined. Regardless, readmissions represent not only an undesirable circumstance for patients, but are also an important quality target for hospitals, and viewed as an important driver of costs to the overall healthcare system.

The reduced LOS and discharge rate to SNF could be attributed to SFX KTCD which has numerous clinical advantages compared with traditional interrupted sutures including the ability to achieve better watertight closure of the wound. Furthermore, 2OPM for skin closure may also promote quicker transition to a home setting, given that it allows patients to shower immediately after procedures and obviates the need for wound dressing changes in the hospital [29–32]. The watertight closure across multiple tissue layers with SFX KTCD and comparatively greater strength as well as unique antimicrobial barrier of 2OPM represent a potential solution to wound complications such as dehiscence and drainage, which can be a retrograde infection pathway [30]. Nevertheless, further research to corroborate and identify the specific underlying causes of these findings related to hospital LOS and discharge to SNF rate would be beneficial.

Some previous studies have also shown the benefits of advanced MLWC with the innovative combination of barbed sutures and a skin closure system. Anderson et al. [33] and Herndon et al. [34] found that use of barbed sutures used in deep layers with 2OPM was associated with fewer episodes of delayed wound healing and reoperations than the standard wound dressing. The authors suggested that the possible mechanism may be that 2OPM distributes wound tension more evenly. In addition, since it remains in place longer during the immediate postoperative period, it may provide prolonged wound-edge support. Snyder and colleagues [14] described their implementation of evidence-based integrated clinical pathways incorporating watertight MLWC and suggested that the integrated clinical pathways and MLWC could synergistically improve the preventative effectiveness against joint prosthetic adverse events such as SSIs [14]. Similarly, Vendittoli and colleagues [35] used an enhanced recovery after surgery (ERAS) protocol along with MLWC that resulted in reduced hospital LOS, improved patient care, and reduced direct healthcare costs. Finally, Flener and colleagues [36] conducted a retrospective EMR review to evaluate the outcomes of 1828 patients receiving MWLC or CC for total hip or knee arthroplasties. Similar to the current study, MLWC was associated with significantly shorter inpatient LOS and increased likelihood of being discharged to home compared with CC. Adjusted mean hospital LOS was a half-day shorter for MLWC (1.10 vs 1.65 days; p < 0.001) and MLWC patients were more likely to be discharged home (odds ratio 4.61; p = 0.002). Our study findings support those of previous authors and build upon them given that ITS analysis is considered a more robust technique for evaluating healthcare quality improvement because of its ability to control for biases [20–22]. ITS controls for pre-existing underlying short- and long-term trends by longitudinally tracking the outcome before and after an intervention [23,24], enabling assessment of whether MLWC had an effect significantly greater than the underlying time trend [26].

Although not perfect in capturing all relevant data, the availability of EMR data presents an opportunity to efficiently evaluate quality of care for individuals in the real-world clinical setting. EMR datasets capture a variety of patient-level data that represent integral components of care that may not necessarily be available through other retrospective database sources such as administrative claims databases or patient registries (e.g., information on the specific type of endoprosthesis used) [37,38]. EMRs also provide efficient access to detailed, longitudinal data that would otherwise be difficult to collect [39,40]. However, limitations of EMR data include coding errors or incomplete information for clinical visits and subsequent follow-up. Also, the patients selected for this study represent a convenience sample from a single site, and patients, treatment methods, and institutional policies may differ in other settings of care, therefore limiting the generalizability of results. Finally, although the present study carefully applied statistical analysis techniques (i.e., logistic regression and ITS) to improve causal inference of comparative treatment effects, the study was still retrospective and observational in design and causality cannot be established for the observed relationships. Despite these limitations, the study utilized ITS analysis to contribute to the evidence for the potential clinical and economic benefits of MLWC in THA and TKA.

Among patients undergoing THA or TKA, MLWC was associated with at least 40% reduction in 90-day readmission rates, LOS, and discharge to a skilled facility compared with CC. ITS confirmed the impact of introducing the MLWC approach despite shifting trends in LOS and discharge status. MLWC may facilitate shorter LOS and the ability to discharge to home and may result in fewer readmission after THA and TKA. Further studies with larger cohorts of patients at multiple sites are warranted.

Z Tan, K Etter, J Tomaszewski and B Po-Han Chen and WJ were involved in the study design, implementation of clinical research data analysis, data interpretation, and manuscript review; NJ Gunja was involved in the study design, data interpretation, and manuscript writing and review.

The authos acknowledge the entire TriHealth team who contributed to this study. The authors also thank the late MA Snyder for shaping the study design and whose patient cohort was examined in this research. Thanks also go to S Wurzelbacher for providing details on the surgical approach for wound closure and to L Leibel for assisting with data extraction. The authors also thank Natalie Edwards of Health Services Consulting Corporation, Boxborough, MA, USA for editorial assistance with the manuscript.

This study was funded by the Dr Mark A Snyder Musculoskeletal Research Fund and Ethicon. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

The authors are employees of Johnson & Johnson (J&J). The authors had full control of the manuscript and provided their final approval of all content. Publication of the study results was not contingent on the sponsor's approval or censorship of the manuscript. The authors have no other competing interests or relevant affiliations with any organization/entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

The study was reviewed and approved by the TriHealth Institutional Review Board (IRB number 19-032).

This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/4.0/