On 11 March 2020, the World Health Organization declared SARS-CoV-2 and the resulting illness, COVID-19, a pandemic [1
]. As of January 2021, over 90 million cases and nearly 1.72 million deaths have been confirmed worldwide [2
]. However, the true toll of this pandemic is likely even higher due to the strain on worldwide healthcare systems impacting the health of patients with common illnesses [3
]. In particular, cancer patients are at increased risk of death from the virus and virus-based interruptions in care. Early data from China reported that cancer patients with COVID-19 were more likely to suffer severe sequelae compared to those without a cancer diagnosis [4
]. This was then confirmed by later international studies [5
]. In addition to deaths related to the virus itself, population-based models predicted that the pandemic would increase cancer deaths in 2020 due to delays in cancer diagnoses, secondary to healthcare systems focusing their efforts on fighting the pandemic [8
At the beginning of the pandemic, Canadian cancer centers looked to the experiences of Chinese and European centers for guidance on best pandemic practices in caring for this vulnerable population. These early recommendations largely focused on the importance of patient and staff screening, masking and disinfection, and physical distancing [9
]. Canadian provinces have since created new policies for radiation oncology departments that seek to provide the best possible care while ensuring the safety of patients and staff. These include categorizing patients based on priority to treat in order to reduce oncology’s resource load on other departments, an increased use of telemedicine and hypofractionated treatments to reduce the number of hospital visits for patients [13
]. Moreover, individual centers may enact additional policies like remote check-in services for appointments and additional waiting areas to further reduce opportunities for patient-to-patient transmission [16
]. For example, our center chose to offer patients a remote cell phone appointment check-in as well as a designated parking lot where they could wait to be notified to enter the hospital for treatment. The Italian radiotherapy experience has taught us that while some patients feel reassured by some of the measures put in place since the pandemic, many still report fears of being touched or positioned on the treatment table and increased anxiety knowing they must come to hospital for treatment [17
While cancer centers have been operating under these new policies for nearly a year now, little is known about how these policy changes have impacted patient satisfaction with care. To date, most publications addressing or surveying the cancer patient experience during the pandemic are short communications or opinion pieces [17
], studies focusing on patient perceptions of the virus [18
], or their psychological distress caused by the virus [19
]. A multi-center study conducted in Poland found that in addition to the anxiety surrounding their cancer diagnosis, cancer patients are also anxious about the virus and its impact on their treatments [21
]. This anxiety relates to delayed treatments and patient perceived loss of treatment effectiveness. Furthermore, a survey conducted by Cancer Research UK reports a significant reduction in the proportion of patients who rated their care as “very good” [22
], but it is unclear as to what components of care respondents were less satisfied with. Given that cancer centers are now treating patients during the second wave of the pandemic, without clear patient feedback on what pandemic practices do and do not work well for them, it is imperative to obtain this feedback as soon as possible in order to be able to amend policies to improve patient care.
The aim of this study was to determine patient satisfaction with their care during the pandemic in a qualitative and quantitative manner. Additionally, we evaluated patient adoption and satisfaction with our center-specific pandemic safety measures with the intention to amend policies based on patient feedback where possible.
2. Materials and Methods
2.1. Study Design and Population
This single-institution descriptive cross-sectional study consisted of a questionnaire distributed to patients treated at the McGill University Health Centre (MUHC) radiation oncology department during the first wave of the COVID-19 pandemic. Ethics approval was granted by the MUHC Research Ethics Board following a full review (project number 2021-7040). As we were interested in the new patient experience for conventional external beam treatments during the initial phase of the global pandemic, eligible patients were outpatients aged 18 years or older who conducted their initial consultation by telemedicine between 23 March and 31 May 2020 and underwent five or more fractions of radiotherapy. This was selected as the minimum number of treatments to ensure respondents had enough experience to develop informed opinions on center operations and to align with more extreme hypofractionated regimens for prostate and breast cancers that were adopted based on international recommendations [9
2.2. Survey Development
Since we could not find any existing survey tools in the literature, we designed a novel patient satisfaction survey using an evidence-based approach to assess patient satisfaction [23
]. Topics of interest including patient anxiety levels, communication of safety protocols, telemedicine experience, and in-person treatment experience were identified in planning meetings with clinical staff and used to inform survey development. As our center operates bilingually, questions were developed in tandem in English and French and carefully revised by bilingual clinicians and researchers to ensure clarity and plain language in both versions. After initial survey development was complete, a feasibility study was conducted in both languages on a small group of patients by SMS recruitment to assess the expected response rate, formatting and ease of use of the online survey instrument, and content clarity. Information from this feasibility study was used to guide minor revisions in question wording in order to increase clarity for participants.
The finalized survey consisted of 21 quantitative and qualitative questions which included categorical (yes/no/I don’t know) questions, 5-point Likert scale questions (ranging from strongly disagree to strongly agree), and short answer free-text questions. Survey instruments are available as supplementary material
2.3. Data Collection
Eligible patients were identified using our center’s oncology information system (Aria, Varian Medical Solutions) and contacted by phone for participation. The study purpose and ethical considerations were explained and clarified as needed prior to obtaining verbal consent, as approved by the Research Ethics Board. An emailed copy of the consent form was offered to participants for their records. Participants were then given the option to complete the survey by phone or online in their preferred language. Patients who selected the online option were sent the survey instrument URL and a copy of the consent form by email. Care was taken to ensure the research team member contacting patients was not involved in the patient’s current or future medical care, and this was clearly explained to participants at the time of consent.
Once eligible patients were identified and charts were reviewed, patient recruitment began in September 2020 and concluded in October 2020. Individuals who did not answer phone calls after three attempts were assumed to decline participation and were not contacted further.
2.4. Statistical Analysis
Relative frequencies of categorical and ordinal responses were calculated. For free-text qualitative data, answers were segmented into individual phrases and collaboratively codified by two researchers using a deductive coding methodology. Phrases were first codified into one of 12 common comment topics identified during the segmentation phase, then codified by commentor attitude towards the topic (positive, negative, or neutral). Chi square tests were used to assess and compare the distribution of commentor attitudes per topic to patient satisfaction scores from other questions.
Phone and online responses were compared to determine if social desirability bias resulted in phone respondents answering more positively. Mann Whitney U tests were used for comparisons of ordinal data and Chi Square tests for categorical data. Similar comparisons were also performed with patient responses from our initial feasibility study to verify if results would have been impacted had these patients been included in the main study. Statistical significance was defined as α < 0.05.
In order to assess the impact that pandemic-related changes had on patient experience in our radiotherapy department, we conducted a descriptive cross-sectional study of patients treated during the early pandemic. Overall satisfaction with care was high, with 90.7% of respondents reporting satisfaction with telehealth consults and 100% reporting satisfaction with treatment appointments. Similarly high satisfaction rates were found in an Italian center where 89.6% of patients reported their treatment quality to be good or excellent during the pandemic [24
]. While satisfaction rates reported pre-pandemic range from 76.2–95.7% [25
], comparison between these results and our own is challenging due to use of different variables in survey instruments, as well as cultural and contextual differences between centers.
A major change implemented early on in the pandemic is the shift towards telemedicine. Similar satisfaction rates (75–92%) to our 90.7% have been reported by several centers in the US [29
], with 70% of patients in one survey stating telemedicine appointments made them feel safer [29
]. However, as evidenced in a study by Zimmerman et al., only 23.3% of patients feel telemedicine services can fully replace regular standards of care, with the majority (77.1%) preferring it be used in addition to regular services [31
]. In our study, despite overall satisfaction with telemedicine, the most common complaint was poor communication quality (clarity and frequency). While this can be somewhat addressed with more frequent video-based telemedicine, it is unlikely telemedicine can fully replace all in-person appointments, as physicians are concerned about missing important clinical indicators they would otherwise perceive when in-person [29
Overall, the majority of patients were aware of and appreciated new protocols established during the pandemic with few reports of these measures worsening patient experiences. The most disliked protocol was the daily screening questionnaires, which may have been due to respondent fatigue [33
]. Increased patient anxiety due to the pandemic was similar to percentages reported in other studies [19
], with several patients noting their anxiety reduced after witnessing the protocols at their first treatment appointments. However, while new pandemic protocols did not appear to worsen patient experiences directly, several patient accounts indicate indirect effects are being felt. Reports of reduced access to supportive care, scheduling and communication issues, and difficulties building rapport with staff are suggestive of broader organizational challenges that our center needs to work on to ensure quality is maintained across the full cancer care spectrum.
To our knowledge, this is the first study conducted assessing patient satisfaction with new pandemic safety precautions at a high-volume radiation oncology center in Canada. Our center sees approximately 4000 new patients per year with over 3000 radiotherapy treatments per year. While our study provides valuable insight into patient attitudes towards departmental pandemic measures, we recognize some inherent limitations exist. Due to the eligibility criteria, sample size was limited to consenting patients who received multi-fraction treatments during the initial pandemic-related service modification period, and further limited by exclusion of those recruited to the feasibility study. While this reduced the power of the study, inclusion of the feasibility group would not have affected the main findings. The retrospective cold-call nature of the study, in addition to possibly introducing recall bias, may have also influenced sample size due to individuals screening their calls and diminished interest in providing feedback over time. The combination of these factors may have led to the exclusion of patients who had greater anxiety and/or less positive experiences than those who participated, such as individuals who chose to forego treatment specifically because of the pandemic. In addition, while question wording bias is possible, we attempted to decrease this by ensuring a balance of questions, including negative and positive responses, avoiding wording that produces emotional responses, diversifying question styles, and including an undecided option. There was also evidence of social desirability bias in the data, as online respondents were less satisfied with elements of care than phone respondents. This occurred despite efforts to mitigate the bias with non-physician surveyors. While a purely online survey may have better addressed this limitation, we were concerned about introducing a selection bias for younger patients who may be more comfortable using technology, due to age being a predictor of patient satisfaction seen in other studies [30
]. Despite these limitations, we believe that our study provides valuable insights into the patient experience during the early pandemic, particularly for areas for future improvement.