Author Response
Reviewer #1 (Public Review):
Ghosh and colleagues report on their multidisciplinary effort to improve cervical cancer screening attendance in the East Boston Neighborhood Health Center (March-August 2021). Specifically, the authors 1) identified using electronic medical records overdue follow-up visits, 2) scheduled screening appointments during regular clinic hours and weekends/evenings, and 3) surveyed patients on their experience. These objectives were clearly defined (although not consistently so throughout the manuscript) and data analyses/presentation were simple and straightforward, appropriate to the study design and methodology used.
Thank you for this comment. We have clarified the objectives in the revised manuscript.
Overall, it is unclear to what extent the overdue appointments were backlogs created by the COVID-19 pandemic or due to pre-pandemic factors that could have been exacerbated by the pandemic. In order to contextualize the current study and its findings, an elaboration is needed on whether the pandemic created the delays in cervical cancer screening or simply compounded the problem. For example, the authors report on page 8, lines 196-197 that in 30% of encounters (not clear how many of the 118 reviewed charts were overdue appointments) the healthcare provider did note the overdue appointments.
We have Figure 2 (now Figure 4) and added Figures 2and 5 to address this comment. In 2019, prior to the COVID-19 pandemic, approximately 70% of patients were up-to-date with cervical cancer screening, corresponding to 8467 patients overdue for screening. In 2020, the up-to-date percentage dropped to 63.5% and the overdue number increased to 8812. Figure 2 is a flowchart of the project which clarifies the “30%” mentioned in the reviewer comment
In addition, a brief description of the cervical cancer screening program in place would be informative.
We have added this in the “setting” section of the methods on page 4-5, lines 107-128)
Table 1 provides an effort versus value summary; however, these constructs are ill-defined, with few inconsistencies with what is reported in the text.
This table is intended to help inform clinics that are considering implementing quality improvement programs about the effort required and value obtained for different aspects of our program. These are based in part on proprietary cost analyses so certain details are not able to be included. We have amended the text/table to eliminate inconsistencies.
Comments specific to Aim 1:
The methodology is missing information on key elements, mainly relating to the decision-making process of establishing and defining the "validated" patient chart list (1375 overdue patients out of 6126 reviewed charts). A chart of the 1375 approached study population is also warranted (459 patients were screened, 622 could not be reached, and 203 cancelled/missed their appointments, what about the remaining 91 patients). A description of the characteristics of the study population and a comparison of the different groups (screened, not reached, cancelled/missed appointment) along these characteristics are missing.
We have added a flowchart with this information to the results section. See Figure 2.
Comments specific to Aim 2:
About 63% of the 459 scheduled screenings were done during the evening/weekend clinics, which represents a substantial gain and clearly indicates a window of opportunity to increase screening rates by pinpointing the importance of offering a convenient time to women attend screening visits. In general, and as expected, offering additional screening clinics was effective in addressing the backlog of patients, although with significant investment and resources as mentioned by the authors. How significant is significant?
We are not able to share these data publicly. We have added the following sentence: “The cost data is proprietary/not shareable but analysis by clinical leadership indicated the program was not cost-effective/sustainable.” Page 22, lines 678-80
Comments specific to Aim 3:
A more structured and detailed presentation/description of the survey instrument, its administration, response rate, and significance of results are warranted in the manuscript, albeit the joint reporting of this in the appended material.
We have added additional detail about the survey method (page 9, lines 225-6, 228-31) and results ( Page 14-5, lines 518-22, 530-3) . We also inserted the survey used in the clinics. (Figure 1).
Reviewer #2 (Public Review):
The purpose of this study is unclear from the introduction. Additionally, the methods are incomplete and did not describe how data was collected and analyzed. The results do not describe the sample. Once these are described more clearly, further comments can be made about what the authors were trying to achieve and the impact of the work on the field.
We have clarified the study purpose in the introduction: “The purpose of the project was to examine the impact of a Quality Improvement intervention on improving cervical cancer screening, as well as to evaluate the effectiveness and sustainability of different methods for addressing overdue screening.” (page 3, lines 87-90)We have also clarified the methods and results to describe data extraction more completely from electronic medical records and statistical analysis using descriptive statistics.