Implementing our feedback can enhance patient experience and care
Our study provides a unique view of the patient perspective through analyzed crowdsourced public opinion via Twitter. To our knowledge, an optimized patient-centered outpatient gynecology experience has not previously been described in the medical literature. Optimizing the found domains of hospitality, comfort measures, pain and anxiety control, privacy, and diverse representationin the outpatient gynecologic experience within the outpatient care setting may ultimately result in improved patient satisfaction, patient well-being, and adherence to care through maximizing patient-centered care. We created a checklist of suggestions, including offering analgesics during office-based procedures and tailoring the floorplan to maximize privacy (FIGURE 5), for improving the outpatient gynecology experience based on our findings.
Prior data on patient satisfaction and outcomes
Improving patient satisfaction with health care is a priority for both clinicians and hospital systems. Prior studies have revealed only variable associations between patient satisfaction, safety, and clinical outcomes. One study involving the analysis of clinical and operational data from 171 hospitals found that hospital size, surgical volume, and low mortality rates were associated with higher patient satisfaction, while favorable surgical outcomes did not consistently correlate with higher Hospital Consumer Assessment of Healthcare Provers and Systems (HCAHPS) scores.10 Smaller, lower-volume hospitals earned higher satisfaction scores related to cleanliness, quietness, and receiving help measures.10 It has also been shown that the strongest predictors of patient satisfaction with the hospital childbirth experience included items related to staff communication, compassion, empathy, and respect.11 These data suggest that patient satisfaction is likely more significantly impacted by factors other than patient safety and effectiveness, and this was supported by the findings of our analysis. The growing body of literature associating a sense of psychological and physical safety within the health care system and improved patient outcomes and experience suggests that the data gathered from public commentary such as that presented here is extremely important for galvanizing change within the US health care system.
In one systematic review, the relationship between patient-centered care and clinical outcomes was mixed, although generally the association was positive.12 Additionally, patient-centered care was often associated with increased patient satisfaction and well-being. Some studies suggest that patient well-being and satisfaction also may be associated with improved adherence and self-management behaviors.12,13 Overall, optimizing patient-centered care may lead to improved patient satisfaction and potentially improved clinical outcomes.
Additionally, increasing diverse representation in patient materials and illustrations may help to improve the patient experience. Louie and colleagues found that dark skin tones were represented in only 4.5% of 4,146 images from anatomy texts analyzed in 2018.14 Similarly, a photogrammetric analysis of medical images utilized in New England Journal of Medicine found that only 18% of images depicted non-white skin.15 More recent efforts to create a royalty-free digital gallery of images reflecting bodies with diverse skin tones, body shapes, body hair, and age as well as transgender and nonbinary people have been discussed in the lay press.9 Based on our findings, social media users value and are actively seeking diversity in representation and imagery during their outpatient gynecology experience.
Opportunities for future study
Our research utilized social media as a diverse and accessible source of information; however, there are significant opportunities to refine the methodologic approach to answering the fundamental question of creating the patient-centered gynecologic experience. This type of study has not yet been conducted; however, the richness of the information from this current analysis could be informative to survey creation. Future research on this subject outside of social media could bolster the generalizability of our conclusions and the ability to report on qualitative findings in the setting of known patient demographics.
Social media remains a powerful tool as evidenced by this study, and continued use and observation of trending themes among patients is essential. The influence of social media will remain important for answering questions in gynecology and beyond.
Our work is strengthened by social media’s low threshold for use and the ability for widespread access to a diverse group of users. Additionally, social media allows for many responses to be collected in a timely manner, giving strength to the abstracted themes. The constant production of data by X users and their accessibility provide the opportunity for greater geographic coverage in those surveyed.4 Crowdsourced public opinion also has the advantage of producing qualitative metrics in the form of likes and retweets that may provide a reliable measure of public support or engagement.1
Future studies should examine ways to implement the suggested improvements to the office setting in a cost-effective manner and follow both subjective patient-reported outcomes as well as objective data after implementation, as these changes may have implications for much broader public health crises, such as maternal morbidity and mortality.
Study limitations. Our study is limited by the inherent biases and confounders associated with utilizing data derived from social media. Specifically, not all patients who seek outpatient gynecologic care utilize social media and/or X; using a “like” as a surrogate for endorsement of an idea by an identified party limits the generalizability of the data.
The initial Twitter query specified, “I’m asking women”, which may have altered the intended study population, influenced the analysis, and affected the representativeness of the sample through utilizing non ̶inclusive language. While non-binary/transgender care and inclusivity emerged as a theme discussed with the tweets, it is unclear if this represents an independent theme or rather a reaction to the non–inclusive language within the original tweet. ●
Qualitative study methods
The data abstracted was analyzed with Dedoose1 software using a convenience sample and a mixed-methods analysis. Utilizing X (formerly Twitter and referred here as such given the time the study was conducted) for crowdsourcing functions similarly to an open survey. In the absence of similar analyses, a modified Checklist for Reporting Results of Internet E-Surveys (CHERRIES) checklist was utilized to organize our approach.2
This analysis was comprised of information freely available in the public domain, and the study was classified as IRB exempt. Ethical considerations were made for the fact that this is open access information and participants can reasonably expect their responses to be viewed by the public.3 As this question was not originally intended for research purposes, there was not a formalized development, recruitment, or consent process. The survey was not advertised beyond the original posting on Twitter, and the organic interest that it generated online. No incentives were offered to participants, and all participation was voluntary. There is no mechanism on Twitter for respondents to edit their response, although responses can be deleted. Unique visitors or viewers beyond posted impressions in response to the original tweet could not be determined.
Twitter thread responses were reviewed, and all completed and posted responses to the original Twitter query with 100 or greater “likes” were included in the analysis. These tweets were abstracted from Twitter between December 17, 2021, and December 27, 2021. At the time of tweet abstraction, engagement metrics, including the numbers of likes, retweets, and replies, were recorded. Additionally, author characteristics were abstracted, including author verification status and association with health care, as described in their Twitter profile. Definition of an individual associated with health care was broad and included physicians, advanced practice providers, nurses, first responders, and allied health professionals.
A total of 131 tweets met inclusion criteria and were uploaded for analysis using Dedoose qualitative analytic software.1 Two authors independently utilized a qualitative analysis to code the isolated tweets and identify thematic patterns among them. Uploaded tweets were additionally coded based on ranges of likes: 100-500; 501-1,000; 1,001-1,999; and >2,000. Tweets were coded for author verification status and whether or not the author was associated with the health care field. Themes were identified and defined during the coding process and were shared between the two authors. A total of 18 themes were identified, with 1 theme having 4 subdivisions. Interrater reliability testing was performed using Dedoose1 software and resulted with a pooled Cohen’s Kappa of 0.63, indicating “good” agreement between authors, which is an adequate level of agreement per the Dedoose software guidelines.
References
1. Dedoose website. Accessed July 28, 2022. https://www .dedoose.com/
2. Eysenbach G. Improving the quality of web surveys: the checklist for reporting results of internet e-surveys (CHERRIES) [published correction appears in J Med Internet Res. 2012;14:e8. doi:10.2196/jmir.2042]. J Med Internet Res. 2004;6:e34. doi:10.2196/jmir.6.3.e34
3. Townsend L, Wallace C. Social media research: a guide to ethics [University of Glasgow Information for the Media website]. Accessed March 2, 2023. https://www.gla.ac.uk /media/Media_487729_smxx.pdf