Challenges of Conducting an Integrative Clinic Improvement Project During COVID-19

Abstract

Objectives:

We describe navigating pandemic-related obstacles while implementing a quality improvement (QI) project for patients with osteoarthritis (OA) in an integrative health pain clinic.

Methods:

The Comprehensive Pain Center at Oregon Health & Science University delivers integrative health therapies in conjunction with mainstream medical treatments. This project aimed to collect mental and physical health data from all OA visits for a year. Assessments were to start in March 2020; however, the pandemic onset drastically altered clinical care.

Results:

Physicians and psychologists shifted to telehealth, which was not possible for acupuncture, chiropractic, massage therapy, or Rolfing therapy. Data were collected via telephone calls, which took longer than anticipated, as for many patients this interaction was their only contact with anyone outside home. COVID-19 episodes, staff shortages, and other difficulties necessitated adaptive procedures. Several adapted strategies were identified as likely to be applicable in future circumstances of similarly major disruptions in QI projects.

Conclusions:

This QI project was greatly reshaped by sudden changes in patient pathways during COVID-19.

Introduction

The COVID-19 pandemic drastically altered nearly every aspect of human life across the globe, with health care systems—including their research activities—especially severely affected. In 2019, approximately 80% of non-COVID-19 research trials were stopped or interrupted. The investigators that persisted were typically caught off-guard and forced to flexibly adapt study processes for both staff and research participants. ¹ This pattern was evident not only in clinical trials and basic science research but also in quality improvement (QI) projects. This report describes how one team navigated through multiple COVID-19-related challenges to implement a QI project designed to elucidate treatment pathways of patients with osteoarthritis (OA) in a large integrative pain clinic. A separate report describes the data outcomes of this project. ²

QI projects involve systematic, data-guided activities designed to bring about positive changes in the delivery of health care in specified settings. Teams collect data on the effects of clinical practices, assess the results, and identify ways to improve standards of care. Thus, the QI process targets change in the health care system at the local level, rather than through testing the efficacy of particular interventions, such as in clinical trials. Accordingly, QI research methods involve an iterative approach with corrections made with each step, a process that diverges somewhat from those of standard clinical research. ³

This project was developed as a collaboration between clinician-investigators based in the Comprehensive Pain Center (CPC) at Oregon Health & Science University (OHSU) and the Pfizer Medical Outcomes and Analytics team. At the time this project was implemented, the CPC was unique in delivering a wide range of integrative health modalities in conjunction with allopathic medical treatments in a large outpatient pain treatment setting; see Figure 1. ⁴ Modalities offered in the pain center included acupuncture, chiropractic care, dietary and nutritional guidance, massage, cognitive-behavioral therapy, mindfulness, naturopathic care, and Rolfing. The past several decades have witnessed an ever-growing demand for integrative health treatments by people living with chronic pain, despite such services typically being provided in outside settings rather than offered in clinic in non-military setting in the United States. ⁴

FIG. 1.

Care from multiple types of providers: a whole person approach to pain management. The graphic is an image indicating the location of hip pain with text highlighting the treatments provided at the pain center.

Our team was interested in understanding how integrative health treatments interact and may be synergistic with medication and other mainstream medical treatments for OA. The QI project was designed to collect outcome data from all CPC clinic visits (medical, psychology, and integrative modalities) on function, mental, and physical health over the course of a year. These data would then enable the team to conduct a variety of novel analyses to determine whether there are treatment trajectories that appear to be promising pathways for improved outcomes in terms of pain, mood, and daily functioning. However, as the project was launched, there was no way to know what hurdles would be faced during a major worldwide pandemic.

Methods

A multi-institutional team including OHSU faculty (an acupuncturist, a clinical psychologist, and a pain medicine physician) and pharmacists employed by the Pfizer Medical Outcomes and Analytics Team was formed.

The initial project goal was to gather a series of patient reported outcomes (PRO). Patients with OA were to be identified prior to visits and while in the waiting area for their appointments and instructed to complete PRO questionnaires using iPads. PRO tools included the Hip Dysfunction and Osteoarthritis Outcome Score (HOOS), the Knee Dysfunction and Osteoarthritis Outcome Score, the Brief Pain Inventory (BPI), and the PROMIS Global 10.^5–8 PROs were to be gathered at all visits to the clinic for 1 year. Data collected would enable providers to understand how the PROs changed over time, and after all data collection was completed, common treatment pathways were to be assessed using change scores. The OHSU Institutional Review Board (IRB) determined that the project did not need IRB oversite due to its QI design (07/30/2020, study ID 00021916). The project was designed to start collecting data in March 2020; however, due to challenges in executing the funded agreement between Pfizer and OHSU in conjunction with the onset of the COVID-19 pandemic, the plans for implementation changed drastically.

The first major implementation challenge was a lengthy administrative delay. As described above, QI research differs in some respects from clinical trials research. However, the OHSU contracting office had never before negotiated a contract with an industry sponsor to support a QI project. Ultimately it took over a year and the intervention of a high-level university administrator to obtain agreement on processing the contract. The contract was signed in the spring of 2021, but by then, COVID-19 had entirely altered clinical care at the pain center. All non-essential in-person patient visits were cancelled due to safety precautions, and the model for the project had to pivot correspondingly. Changes in treatment patterns—virtual visits for physicians and psychologists, interruptions for acupuncture, chiropractic, massage, and Rolfing—prevented PRO tools from being handed to patients prior to visits.

A multipronged approach to PRO data gathering was implemented using REDCap Cloud (nPhase, Inc., Encinitas, CA; https://www.redcapcloud.com/) rather than via the electronic health record. Because the study coordinator was no longer allowed at the pain center due to COVID restrictions, all patient data were collected via telephone calls instead of via iPads, and the coordinator used a laptop computer to implement REDCap-based surveys. This seemed the most reasonable adaptation for data collection given the OHSU-wide implementation of COVID-related restrictions. Although research on survey methods reveals that iPads offer potential advantages with respect to accuracy, speed, and data completeness in comparison to telephone calls, iPads may also present challenges related to attention and technical difficulties. In contrast, interviewer-led interactions during telephone calls can provide more detailed and nuanced data, yet also can be more time-consuming and subject to potential biases.^9,10 Generally, however, tablet and telephone administrations yield similar results, and research indicates switching between these methods during a study does not adversely affect the comparability of results. ¹¹

Data collection commenced in May 2021 with plans to continue for 1 year. However, despite the coordinator’s persistent attempts to reach out to participants via calls, COVID-19 strongly impacted the ability to recruit patients at the rate originally anticipated (for patient demographics, see the separate report describing data outcomes). Patients’ COVID-19 illness and associated depression, anxiety, and fatigue impacted data collection abilities. ¹² The coordinator had to allow more time to address patient concerns that were not project related, which often involved listening about patients’ personal issues and offering a sympathetic ear. At times, data collection was extra challenging during chaotic periods when COVID-19 illness was affecting not only patients but also staff, providers, and their families. Quite a few staff members endured up to three episodes of COVID-19, and several continue to experience fatigue and other “long COVID”-related symptoms. Childcare issues were also impactful during periods when daycare centers were shut down, thereby preventing staff and providers from being available for work. Likewise, transportation became challenging at times due to cutbacks on public transportation that made getting to work difficult.

Results

The pandemic-related interruptions in patient services necessitated a much higher volume of telephone calls to reschedule patients’ appointments, which was accompanied by a marked increase in patients expressing intense frustration and at times disrespect and anger toward staff. The accumulated emotional impact of such interactions on staff over many months took a noticeable toll, contributing to “burnout” for many and to a few staff departures. In addition, several providers reported feeling emotionally overwhelmed by the impact of the pandemic on both their personal and professional lives. ¹³ As a result, more time was needed, so the original endpoint for data collection was changed from April 30, 2022, to July 31, 2022.

Both the context and the amount of time required for data collection efforts changed dramatically. It was originally estimated that the questionnaires would take about 10 min to complete over the phone, but the average time with patients was closer to 40 min. The project coordinator administering the questionnaires was often asked questions about COVID-19, vaccinations, and masking. In many ways, these phone calls came to be experienced as a form of “treatment.” In addition, for many of these patients, the phone interaction was the only contact they had with anyone outside their home, and they often wanted to continue conversing with the project coordinator.

Discussion

As with most research efforts initiated during COVID-19, this QI project was greatly reshaped by sudden changes in patient pathways through the health care system. The project was able to proceed with the same vulnerable OA population that is at higher risk of adverse health effects from exposure to COVID-19; however, it was necessary to revise data collection methods during telehealth visits while still capturing in-person visits. Additional challenges included diminished staff numbers, increased levels of depression, anxiety, loneliness, and fatigue, and the need to develop more complicated and cumbersome study processes. ¹⁴

Many of the integrative health modalities cannot be given via telehealth (massage, chiropractic, Rolfing, and acupuncture), so the only in-person contact many of these patients experienced had been with these clinicians. Furthermore, patients were accessing more of the integrative health modalities as a percent of their usual comprehensive pain center treatment (integrative health providers compared to pain physicians and nurses). These practitioners reported that the nature of patient visits also changed. Patients were looking for more personal contact and discussion. This played out in the telephone-administered questionnaires as well.

Regarding next steps, a companion publication in this journal describes the QI data results. This team is also considering a new QI project, which we anticipate will not unfold in the midst of such an extreme disruption of health care research and services as happened due to the pandemic during this study. In this next project we will aim to streamline methods of data collection via email and/or text. Although we will seek to minimize in-person or telephone call-based data collection, we plan on alloting more time to coordinators for interacting with patients even when the interactions are not fully study-related. Moreover, we will train study coordinators more thoroughly in patient communication skills so that they are prepared to appropriately respond during such patient interactions. In addition, the goal of this new study will be to expand the range of patients in the next project to include all types of pain conditions, rather than focusing exclusively on OA.

Nonetheless, we recognize that several of the QI project adaptations that emerged from meeting the challenges of the COVID pandemic during this study would likely be applicable in future circumstances of similarly major disruptions (e.g., the aftermath of a major earthquake or hurricane). These adaptations include shifting from in-person to smartphone-based data collection; anticipating coordinators needing extra training as well as additional time and efforts for data collection; and planning on the possibility of decreased staff availability due to disruptions in their professional and personal lives.

Conclusion

In summary, given that the majority of patient care visits shifted to virtual appointments due to COVID-19, data collection was challenging. Since numerous integrative health modalities cannot be given via telehealth (massage, chiropractic, Rolfing, and acupuncture), the only in-person contact many patients with OA experienced was with the integrative health clinicians. Episodes of COVID-19, diminished staff numbers, and increased levels of depression, anxiety, loneliness, and fatigue in the patient population required the development of more complicated and extensive processes.

Authors’ Contributions

J.W.C.—author, revisions, and final approval. S.D.M.—author, revisions, statistics, and final approval. K.M.—author, revisions, and final approval. S.F.—author, revisions, and final approval.