Asking physicians how best to implement cervical cancer prevention services in India: A qualitative study from Mysore

Cervical cancer is the second most common cancer among Indian women. Screening is an effective prevention strategy, but achieving high screening rates depend upon identifying barriers at multiple levels of healthcare delivery. There is limited research on understanding the perspectives of providers who deliver cancer prevention services. The objective of this study was to explore physician perspectives on cervical cancer prevention, barriers to effective implementation, and strategies to overcome these barriers in India. Guided by the “Multilevel influences on the Cancer Care Continuum” theoretical framework, we conducted semi-structured interviews with physicians in Mysore, India. From November 2015- January 2016, we interviewed 15 (50.0%) primary care physicians, seven (23.3%) obstetrician/gynecologists, six (20.0%) oncologists, and two (6.7%) pathologists. We analyzed interview transcripts in Dedoose using a grounded theory approach. Approximately two-thirds (n = 19, 63.3%) of the participants worked in the public sector. Only seven (23.3%) physicians provided cervical cancer screening, none of them primary care physicians. Physicians discussed the need for community-level, culturally-tailored education to improve health literacy and reduce stigma surrounding cancer and gynecologic health. They described limited organizational capacity in the public sector to provide cancer prevention services, and emphasized the need for further training before they could perform cervical cancer screening. Physicians recommend an integrated strategy for cervical cancer prevention at multiple levels of uptake and delivery with specific efforts focused on culturally-tailored stigma-reducing education, community-level approaches utilizing India’s community health workers, and providing physician training and continuing education in cancer prevention.


Thank you for your submission, and our apologies for the delay in sharing a decision. As you will see from the reviews, there was a split decision between the reviewers between Major and Minor
Response: Thank you for your comments. We hope you will note the extensive revisions to the paper, which we believe have further strengthened the connections to the conclusion and reorganized the paper. Response: We have moved this sentence to the background. Further, we have clarified that the gynecologists in the study were actually trained OB/GYNs working in the private sector. This is usually the case in the Indian public health sector, there are very few if any trained Ob/Gyns in the public sector. This is the primary reason why the public sector Ob/Gyns did not show up in our snowball sampling approach. As far as the snowball sampling diagram is concerned, we unfortunately did not keep detailed notes on how the referrals were made and instead focused on keeping records after the participants agreed to participate in the study. This is a point well taken for future studies.

In the methods section
6. Why were pathologists interviewed? They would review samples but not performing any screening. Were the primary care doctors in the public sector specialists that completed residency or medical officers that completed an MBBS (medical school plus intern year)?
Response: The screening process includes both delivering the screening tests and the delivery of the test result back to individual patients. For implementing screening programs with success, achieving efficient delivery of both these steps is essential. Based on this rationale, it was important for us to know the process details from the perspectives of the pathologists in the community. In addition, these pathologists came highly recommended by the physicians in terms of their contributions to cervical cancer prevention activities, and per the pathologists, there were several instances when the patients would come directly to a lab to ask for cervical cancer screenings, even without a physician referral. We have now added a clarifying statement, in the methods section about the pathologists. All primary care doctors in the public sector had completed their MBBS (medical school plus intern year).

Response:
We have now removed the table from the manuscript and integrated the quotations into the paper, since there are no word count limits on the manuscripts published in the journal. We have now added the job title and the sector for each supporting quote.

For Theme 5 --it would be a very concerning major gap if gynecologists reported requiring more training or didn't have a clear sense of when cervical care screenings should be done (as prevention) than pathologists or other physicians.
Response: We agree with the reviewers. The revisions to theme 5 now clearly indicate the challenges reported by the Ob/gyns in the study, which were related to reporting of Pap smears. We have also added additional quotes from Pathologists corroborating these data. More specifically, the lack of training was expressed as a gap by the primary care providers, which is also clarified in this section.

9.
A major area of revision for this paper includes: The themes are not always consistent with the quotations and would consider re-wording and re-framing a few themes. It seems like quote 3/quote 4 might have to do with cultural norms, gender inequality and stigma rather than solved by promoting health literacy. Quote 5 seems to describe difficulties with the economic repercussions of trying to access care (missing work). When going into this into the results, would consider splitting the themes and re-organizing since the text under Theme 1 jumps from a lot of different results that are not directly related to Theme 1.

Response:
We have now revised theme 1 to more accurately represent the ideas reflected in the data.
10. Quote 7 seems to deal with community influence and lack of trust in medical systems more than stigma (possibly more lack of trust in the medical system and community perceptions of what happens at the hospital rather than judgment towards the patient for having cancer). Again, quote 8 seems to be a quote more on not valuing elderly women (not valuing health care services for women and the financial burden of being diagnosed with cancer) rather than stigma towards cervical cancer.

Response:
We have now revised this section to more accurately represent the ideas noted by the reviewer and in the data.
11. Finally, it's not clear how the multilevel influences on the cancer care continuum was used to organize themes ---would do a better job of tying this framework to the results and conclusions and Figure 1 Response: For the authors of this paper, Figure 1 visually depicts the embedded nature of healthcare delivery, as outlined in the framework for multilevel influences on the cancer care continuum. We have added this to the discussion section.
Reviewer #2: 12. The concept is good however some points needs to be explained in more detailed manner. This is with regard to the process of conducting the study. Clarity is to be mentioned regarding the setting, involving opinion of community workers etc. Application and general usability to be implemented also needs to be elaborated.