Renewal of the National Cervical Screening Program: health professionals’ knowledge about screening of specific populations in NSW, Australia March 2022, Volume 32, Issue 1

Introduction

The renewed National Cervical Screening Program (NCSP) was introduced in Australia from 1 December 2017, and included replacing 2-yearly cervical cytology testing via Pap smears with 5-yearly human papilloma virus (HPV) testing with partial genotyping and reflex liquid-based cytology (LBC).¹ The screening age range changed to women aged 25–74 years. The option for self-collection of a vaginal swab for HPV testing under medical supervision was also made available for under- or never-screened women, defined as women aged 30–74 years who have never been screened or are overdue for screening by 2 or more years.² It was anticipated that these changes would lead to a 24–36% reduction in cervical cancer incidence and mortality in Australia compared with the previous screening program.³

Updated accompanying screening guidelines were commissioned by the Australian Government Department of Health³ and released online in February 2017. Accredited educational activities have been delivered by various stakeholders and were available to cervical screening providers both in the lead up to, and following the program renewal, to assist with delivery of the new screening recommendations.

Correct implementation of the renewed NCSP is required for the success of the program.^4-6 Improving consumer and health professional confidence and participation in the renewed program is a key element in the elimination of cervical cancer in Australia.⁷ It is imperative that health professionals delivering cervical screening are familiar with and confident in following the current evidence-based guidelines, which will in turn ensure that women receive appropriate screening and follow-up. Elimination of cervical cancer in Australia is highly dependent on HPV-screening behaviour, particularly in the older non-HPV-vaccinated cohort.⁸ General practice is the primary setting for delivery of cervical screening, and has the capacity to provide both routine screening and opportunistic screening for under- and never-screened women, thus impacting significantly on screening and mortality rates. It is important to capture behaviours of screening providers in order to assess delivery of the program, and modify future educational interventions in response to any identified gaps to ensure ongoing success of the program.

A retrospective cross-sectional review of the first 6 months of the program showed some evidence of problems with clinician application of guidelines, specifically that requesting of a co-test (HPV test plus liquid based cytology outside of routine 5-yearly screening) for symptomatic women varied among clinicians, with anecdotal reports of over-ordering.⁹ This led to revision of the definition of signs and symptoms requiring a co-test.¹⁰ This highlights the need to continue to identify strengths and gaps in program delivery, so that an appropriate response can be undertaken.

Methods

We invited health professionals delivering cervical screening in New South Wales (NSW) to complete an online survey on the SurveyMonkey platform (San Mateo, CA: SurveyMonkey Inc). Recruitment was via email and social media platforms. Invitations to participate included an email invitation to Family Planning NSW (FPNSW) internal and external mailing lists, advertisements in electronic newsletters (Primary Health Networks and NSW Rural Doctors Network), social media postings on medical professional Facebook groups and Twitter, and postings on the Royal Australian College of General Practitioners ShareGP platform. All invitations and advertisements included a link to the online survey which was open for 3 months from September to November 2018.

Survey questions were developed based on previous research⁶, and pilot tested by the research team before implementation. Questions focused on health professional demographic characteristics, knowledge of management of symptomatic women and screening of specific populations as defined in the renewed NCSP guidelines, which include screening after total hysterectomy, screening in women who experience early sexual intercourse or victims of sexual abuse, screening in lesbian women, screening in immune-deficient women and screening in DES (diethylstilboestrol)–exposed women. The survey also asked about respondents’ cervical screening experience, and whether they had undertaken any educational activities relevant to the renewal. The questions discussed in this paper are presented in the results tables.

Once the survey was closed, survey data was exported to Microsoft Excel (Redmond, WA: Microsoft Corporation; 2010) and converted and analysed in Stata (College Station, TX: StataCorp LP; version 14.2). Inclusion criteria were: health professionals living in NSW who provided cervical screening and who completed demographic questions in the survey. Data cleaning and logic check was conducted prior to the analysis. Descriptive statistics of frequency and percentages were used to describe responses regarding uptake of the new cervical screening guidelines, and clinicians’ understanding of the updated guidelines. Chi-square testing was applied to detect differences in the category variables, e.g., by clinician type, gender, years of practice and location of practice. The significance level was set at 0.05.

Ethics approval for the study was received from the Family Planning NSW Ethics Committee (approval R2018–06).

Results

In total, 261 clinicians started the survey. Of these, 241 were included in the final data analysis. Nearly half of respondents were FPNSW staff, or recuited via external FPNSW mailing lists. Surveys were excluded if the participant lived outside NSW, did not perform cervical screening or did not complete any demographic questions. Due to the nature of social media posting, we were unable to estimate how many potential participants saw the survey invitation. Respondent demographic characteristics are outlined in Table 1: just under half (46%) were general practitioners (GPs) and just over one-third were nurses or midwives. The majority were female practitioners (96%), and from a metropolitan area (55%).

Most health professionals reported performing cervical screening at least weekly (87%) and accessing a range of educational activities as outlined in Table 2; the most common of these activities were provided by FPNSW (49%) and the Royal Australian College of General Practitioners (RACGP) (22%). One in ten clinicians had not undertaken any educational activities relating to the screening program renewal.

Table 1.     Survey respondent demographic characteristics (N = 241)

FACRRM = Fellow of Australian College of Rural and Remote Medicine; FRACGP = Fellow of Royal Australian College of General Practitioners; VR = vocationally registered; DRANZCOG = Diploma Royal Australian and New Zealand College of Obstetricians and Gynaecologists; RANZCOG = Royal Australian and New Zealand College of Obstetricians and Gynaecologists

^a The sum of each variable is not the same due to missing values
^b This category covered 45–55 years (rather than 46–55 years) due to an oversight in the survey design
^c Other included non-vocationally registered medical doctor, social worker, clinical nurse consultant or specialist

Table 2.     Cervical screening experience and related education (N = 241)

RACGP = Royal Australian College of General Practitioners
^a The sum of each variable is not the same due to missing values

Knowledge of when to test women younger than 25 years

The majority of practitioners demonstrated good knowledge of indications for testing women younger than 25 years, with 64–85% correctly identifying situations for testing women in this age group (Table 3). However, between one in five and one in three health professionals identified incorrect indications for testing women aged under 25. Knowledge of correct indications for testing under 25 was lower for clinicians from a regional centre at around 70% compared with 80% of those from a rural town/remote area and 90% from a capital city/metropolitan city area in relation to the following two statements: “If a woman has post coital bleeding and/or unexplained intermenstrual bleeding” (Χ² = 10.88, p = 0.004) and “If a woman has had an abnormality under the previous screening program and next follow up test is due” (Χ² = 10.26, p = 0.006) (see Supplementary Table 1, available from:  doi.org/10.6084/m9.figshare.14298737).

Knowledge of screening after hysterectomy

Approximately half (56%) of health professionals were aware of advice within the new guidelines¹¹ on screening requirements following hysterectomy (see Table 3). Knowledge of screening after hysterectomy was higher for respondents who had completed Cancer Council online education modules found on the guideline website (75.8%, Χ² = 6.78, p = 0.03) or educational activities from FPNSW (63.3%, Χ² = 7.01, p = 0.03) than for those who had not. A significant difference in knowledge was also observed across different health professional types (Χ² = 19.96, p = 0.03), with lower awareness among GPs with DRANZCOG or equivalent (23.5%) than among other health professional groups (see Supplementary Table 2, available from: doi.org/10.6084/m9.figshare.14298737).

Knowledge of management of symptomatic women

Approximately three-quarters of health professionals (71–78%) identified correct management of women with abnormal vaginal bleeding, including indications for co-testing and indications for referral (Table 3). Knowledge on the indication for “referral for specialist gynaecological assessment regardless of test results” was significantly lower among clinicians from a regional centre (63.6%) or rural/remote area (63%) (Χ² = 6.34, p = 0.04) compared to those from metropolitan/capital city areas (see Supplementary Table 3, available from: doi.org/10.6084/m9.figshare.14298737).

Knowledge of screening needs of other specific populations of women

Just over one-third of practitioners (37%) correctly identified that: “Immune-deficient women require 3-yearly screening” (Table 4). Knowledge of management of immune-deficient women was significantly higher among those who accessed the guidelines weekly or more frequently (50%) (Χ² = 9.99, p = 0.04) compared with those who accessed the guidelines less frequently. Most health professionals (81%) correctly identified that: “Unvaccinated women have the same screening requirements as vaccinated women”. Knowledge of screening requirements of HPV-unvaccinated women was significantly lower among respondents from a rural town or remote area (77.1%) (Χ² = 11.13, p = 0.03) compared to major city/regional area (see Supplementary Table 4, available from: doi.org/10.6084/m9.figshare.14298737).

Almost all health professionals (87%) correctly identified that: “Women still needing to complete Test of Cure after previous abnormalities on the old program should continue with Test of Cure under the new program”. Knowledge of completing the Test of Cure under the new program was significantly higher among those who performed cervical screening daily or weekly (89.2%) (Χ² = 8.04, p = 0.02) compared to those who performed screening less frequently (see Supplementary Table 4, available from doi.org/10.6084/m9.figshare.14298737). Approximately nine in 10 respondents (89%) correctly identified that: “Lesbian women who have never had heterosexual intercourse have the same screening requirements as heterosexual women”. Knowledge of screening requirements of lesbian women was high overall, however it differed significantly across health professional types, with lower knowledge among GPs with DRANZCOG or equivalent (68.8%), GPs (with FRACGP or FACRRM/vocationally registered) (87.2%) and GP registrars (89.5%), compared to other groups, where knowledge ranged between 96–100% (Χ² = 24.06, p = 0.01) (see Supplementary Table 4, available from: doi.org/10.6084/m9.figshare.14298737). Just under half of health professionals (46%) correctly identified that: “DES (diethylstilboestrol) exposed women require annual screening and specialist review”.

Knowledge regarding self-collected screening

Approximately two-thirds of practitioners correctly identified the indications for self-collected screening (see Table 4). Knowledge that self-collected screening is indicated for: “Women who have never participated in the National Cervical Screening Program and are aged 30 or over, and decline a clinician collected sample” was significantly higher among those who accessed the guidelines weekly or more (88.9%, Χ² = 6.37, p = 0.04) compared to those who accessed the guidelines less frequently. This knowledge was also significantly higher among those who had undertaken educational activities from FPNSW (73.1%, Χ² = 5.93, p = 0.02) compared with those who had not. Knowledge that self-collected screening is indicated for “Women who are overdue for cervical screening by 2 years or longer and are aged 30 or over, and decline a clinician collected sample” was significantly lower among participants from a regional centre (47.3% compared to 55.6% from rural town/remote area and 68.9% from capital city/metropolitan city, Χ² = 8.53, p = 0.01) (see Supplementary Table 5, available from: doi.org/10.6084/m9.figshare.14298737). Almost one in five respondents (17.8%) incorrectly identified that self-collected screening should be offered to any woman who requests it.

Table 3.     Participant’s knowledge of renewal guidelines^a regarding screening under age 25, post-hysterectomy, women with previous abnormalities and management of symptomatic women (N = 241)

^a Cancer Council Australia Cervical Cancer Screening Guidelines^3
b The sum of each variable is not the same due to missing values
^c For questions with multiple choice answers, percentages are calculated from total cohort (N = 241)

Table 4.     Participant’s knowledge of renewal guidelines^a relating to specific populations of women, and self-collection (N = 241)

^a Cancer Council Australia Cervical Cancer Screening Guidelines^3
b The sum of each variable is not the same due to missing values
^c For questions with multiple choice answers, percentages are calculated from total cohort (N = 241)

Discussion

Findings from this survey highlight several areas where health professionals demonstrate good knowledge of the management of specific populations in the renewed cervical screening program. Strengths include knowledge of correct indications for testing women aged younger than 25 years, knowledge of completion of the Test of Cure, and management of symptomatic women. Just under 90% of respondents reported performing cervical screening at least weekly, suggesting this survey represents a health professional population that is highly engaged with cervical screening. However several knowledge gaps were identified. These include management of immune-deficient women, management of women following hysterectomy and screening with self-collection. Gaps in knowledge were also identified to a lesser extent regarding screening women aged under 25, including incorrect assertions regarding testing being indicated in unvaccinated young women, those with a family history of cervical cancer and young women already participating in screening with a normal screening history.

Only one-third of respondents correctly identified that immune-deficient women require 3-yearly screening. Current evidence suggests clearance of oncogenic HPV may differ in immune-deficient women and the renewed guidelines advise it is considered safer to perform 3-yearly screening in these women.¹² Health professionals delivering screening need to be made aware of this shorter screening interval for immune-deficient women. Although most of the research in immunosuppression and cervical cancer has been carried out with HIV-positive women and renal transplant recipients, the definition of which patient groups are considered immune-deficient in the context of cervical screening requirements is broad. The guidelines for management of immune-deficient women apply to HIV-positive women, solid organ transplant recipients, women with congenital (primary) immune deficiency, women who are being treated with immunosuppressant therapy for autoimmune disease and allogenic bone marrow transplant recipients treated for graft versus host disease.¹²

Indications for self-collected screening represented an important knowledge gap in our study. Approximately two-thirds of health professionals correctly identified indicators for self-collection. Almost one in five participants (17.8%) incorrectly identified that self-collection could be offered to any woman who requests it. Offering HPV screening on self-collected samples has potential to improve outcomes for under- and never-screened women. Modelling of outcomes for a cohort of 100 000 never-screened women at age 30 predicted women joining the mainstream program at age 30 would have 1097 fewer cancer diagnoses compared to one round of self-collected HPV screening at age 30.¹³ Currently, self-collection is reserved as a strategy specifically for under- and never-screened women, offering them an alternative option for screening. Self-collection has been shown to be as accurate as clinician-collected samples for detecting CIN 2+ lesions when validated PCR-based assays are used.¹⁴ However, given that one in five health professionals in our survey indicated self-collection could be offered to any woman requesting it, there appears to be confusion about eligibility. A survey-based study of Australian practitioners by Sultana et al also found a gap in clinician level of comfort and confidence with recommending self-collection, more so for practitioners outside of Victoria.¹⁵

We found some areas where health professionals incorrectly identified that screening was indicated in the under 25 years population. These include 17% identifying screening was indicated in women aged under 25 with a first-degree relative with cervical cancer; 21% identifying screening was indicated if it had been 2 years since the last normal screening test; and 30% identifying screening was indicated under age 25 for unvaccinated women. This is an important knowledge gap and has potential to undermine the success of the program, considering that cervical cancer is a rare disease in the under-25 cohort, with only nine new cases in women aged 20–24 years in Australia in 2015 and zero cases in women aged under 20.¹⁶ The benefits of screening this younger cohort do not necessarily outweigh harms (such as potential future obstetric comorbidity from treatment of detected lesions which may well regress).¹⁷

We identified some significant associations between practitioner characteristics and knowledge levels, specifically practitioner location. A 2012 secondary analysis of a cross-sectional survey of Australian GPs found practising in a rural area was associated with less routine use of clinical management guidelines.¹⁸ This was reflected in our results, with health professionals from regional areas having less knowledge of indications for screening women younger than 25 years and management of symptomatic women. Those from a rural or remote area also had less knowledge of management of symptomatic women and screening requirements of unvaccinated women than those in other areas. There are implications here for future educational interventions which specifically target health professionals outside metropolitan areas.

Undertaking educational activities provided by various stakeholders was associated with health professionals having better knowledge of certain aspects of the guidelines. For example, completion of Cancer Council online learning modules was associated with better self-reported awareness of guidelines on screening after hysterectomy, and completion of a FPNSW educational activity was associated with highest knowledge of indications for self-collected screening. Significantly, accessing the guidelines weekly or more frequently was associated with improved knowledge of management of screening for immune-deficient women and knowledge of self-collected screening. All education specific to the renewed screening program should therefore encourage frequent use of the guidelines.

Limitations

Nearly half of respondents were FPNSW staff or from external FPNSW mailing list, and 87% of respondents conducted cervical screening at least weekly, meaning our study may represent a cohort of health professionals that is particularly well informed regarding cervical screening. The majority of respondents were female and from a metropolitan or capital city, and the sample of individual health professional types was relatively small. Due to the recruitment strategies used, we are unable to provide denominator data for how many health professionals received the survey invitation and therefore the survey response rate. Tables 1 and 2 have duplication of some numbers within categories which was a survey design oversight. For multiple choice questions in the survey, we were not able to ascertain if all 241 respondents answered each question, therefore the whole cohort was used to calculate outcomes. If not all respondents answered that question, this may have underestimated response percentages.

Conclusion

Health professionals in NSW demonstrated strong knowledge of key aspects of the renewed NCSP, however, our findings highlight some important knowledge gaps that may impact successful delivery of the renewed program. We also describe some significant associations between practitioner characteristics and these knowledge gaps, which will be important for education providers and stakeholders to address. Practitioner location outside a metropolitan area was associated with some knowledge gaps, while accessing the guidelines more frequently was associated with improved knowledge. Future educational interventions relating to the cervical screening program should therefore target practitioners outside metropolitan locations and encourage more frequent access of, and improved familiarity with, the guidelines. Continued evaluation of practitioner knowledge is essential to allow ongoing tailoring of educational programs as we move further into the post-renewal period.

Peer review and provenance

Externally peer reviewed, not commissioned.

Copyright:

© 2021 Sweeney et al. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Licence, which allows others to redistribute, adapt and share this work non-commercially