Treatment Uptake Indicators (2022)

In Australia, assessment by colposcopic examination of the cervix is an integral part of the diagnostic pathway once a person is deemed to be at a sufficient underlying risk of a high-grade cervical abnormality through the screening process. Factors that may impact an individual’s attendance for diagnostic assessment include whether they received and understand their screening result, understand the need for further assessment and are referred in a timely way, their ability to access colposcopy (service availability, cost, transport, time) and the acceptability of the procedure to that person. Australia’s cervical screening registries have long played a role as a safety net to support participant and clinician’s awareness of results requiring further follow up through systematic reminders and escalating alerts where the recommended pathway has not been followed. Assessment at colposcopy is necessary to identify lesions (via biopsy of suspicious areas) requiring treatment, for which there is a 2030 WHO scale up target of 90% of identified lesions treated. High rates of timely attendance for colposcopy are needed to ensure that screening is effective in the Australian setting. Prior to the transition to HPV based screening in Australia, it was recognised that in the first screening round of the new program, a higher than historically observed rate of referral to colposcopy services would require services to scale up to meet this demand (46). 

Of the people aged 25–74 years referred for colposcopy in 2020, 79.2% had a colposcopy within 12 months of the test that led to their referral, according to the NCSR data. There was little variation by socioeconomic status but there was variation by remoteness (lower for remote and very remote residents, highest for inner regional), jurisdiction (highest in Qld and lowest in the NT), and reason for referral (highest for primary screening result HPV (not 16/18) with high grade cytology and lowest for referral indication of repeat positive HPV at 12 months follow-up in those initially classified as intermediate risk). This variation was also observed in relation to colposcopy within 3 months (a period that better captures timeliness of colposcopy; Figure 9.1). Overall, 55.0% had a colposcopy within 3 months. The proportion having a colposcopy by 3 months varied by the reason for referral, with those with a screening result of HPV positive (not 16/18) and high-grade cytology having the highest proportion of colposcopy within 3 months (78.6%), followed by the HPV 16/18 positive women (63.7%), then women who were referred following repeated HPV positive screening episodes (52.3%). Socioeconomic status was also associated with attendance by 3 months, with those in the lowest socioeconomic status group least likely to have a colposcopy in this time frame (50.7% of women in the lowest quintile compared to 57.2% in the highest).

Women in the youngest age group (25-29 years) were less likely to have colposcopy within 3, 6  or 12 months if they were referred with HPV16/18 (at 57.1%, 74.5% and 83.6% respectively) or after repeat HPV positive tests (at 47.5%, 62.4% and 72.3% respectively) but had comparable rates of colposcopy where the referral indication was HPV (not 16/18) with high-grade cytology (at 76.6%, 89.2% and 94.1% respectively) (data not shown). The age groups with the highest proportion of women attending within 3 months were those aged 65-69 with HPV 16/18 (68.0%), aged 45-59 for HPV (not 16/18) with high-grade cytology and (58.5%) (data not shown).

The 2018 referral cohort (Figure 9.1a) provides an indication of whether those referred ever reach colposcopy. By 31 months follow up, 92.1% had a record of colposcopy attendance according to the NCSR (with 96.6% of those referred with HPV (not 16/18) and high-grade cytology ever attending). There was minimal difference by socioeconomic status over this longer timeframe, but a persistently lower ever attendance by those in remote and very remote areas and in the NT compared to other jurisdictions. Of note, a further 6.5% of women reached colposcopy in the period between 12 and 31 months follow up and this proportion was highest for those referred with repeat positive HPV (10.7%), resident in remote and very remote areas (9.5 and 8.9% respectively) and for those resident in the NT (9.5%).

Comparing colposcopy attendance over time, whilst noting that the 2018 has the longest follow up time and that data from the 2020 cohort may be less complete, timeliness in reaching colposcopy appears to have declined slightly over time. Three-month attendance was 61% in 2018, 55.5% in 2019 and 55.0% in 2020, with 6- and 12-month attendance showing similar patterns. Important context is that there was a large variation in the number of referrals and reason for referral across the three years. In 2018, the first year of the HPV screening program, of just under 58,000 colposcopy referrals in women aged 25-74, most (56.2%) were for 16/18, followed by 33.0% for HPV positivity on a follow-up test (many of which would be after a low-grade cytology in the cytology program) and 10.7% for non 1618 with high grade cytology. In 2019, numbers of referrals dramatically increased to over 95,600, with referrals for repeat positive HPV dominating (63.0%), followed by 16/18 (31.4%) and HPV (not 16/18) with high grade (5.7%). In 2020 numbers of referrals were similar to 2019 at just over 96,000.  Repeat positive tests again increased as the reason for referral (80.9%), followed by HPV16/18 (15.7%) and then not 16/18 with high grade cytology (3.4%). This increase in referrals associated with testing HPV positive on a follow-up test is consistent with fewer people being expected to attend for routine screening in 2020.

Source: Analysis of NCSR data (NCSR RDE 3.4.1. 08/01/2022 for 2020 results; NCSR RDE 3.4.1 07/08/2021 for 2018 and 2019 results); data prepared by the Screening Analysis and Monitoring Unit of the Australian Institute of Health and Welfare.

Includes women classified as “higher risk”; women referred for colposcopy outside screening pathways (for example due to symptoms) without an explicit risk classification are excluded. The closest proxy available to assess whether a woman ‘ever’ had a colposcopy was to include all data as recorded on the NCSR at close of the available dataset, which allowed at least 31 months follow up for all women referred in 2018 (to end July 2021 as per August 2021 RDE), at least 21 months for all women referred in 2019 (to end July 2021 as per August 2021 RDE) and at least 12 months for all women referred in 2020 (to end Dec 2021 as per January 2022 RDE) (therefore no ‘ever’ category available). Data presented are as per the methodology of the AIHW National Cervical Screening Program monitoring report 2020 (3). Time to colposcopy is taken from the date of a person’s first higher risk screening episode. Rates are reported for women in the screening age range 25-74 years.

Completeness of colposcopy data within the NCSR is unknown, with reports submitted by clinicians supplemented by the use of Medicare rebate data and biopsy/histology data to capture the fact of a colposcopy event (although colposcopy may also be used for other indications with no unique MBS code). Thus the estimates represent minimum values based on notifications to the NCSR, and colposcopy data may be more complete for earlier years than for more recent years. Only some higher risk groups have nationally stated targets for time to further assessment (being 2 weeks for women suspected to have cancer on reflex cytology (not separated in routine data) and 8 weeks for HPV positive women with high-grade cytology) within the national guidelines (47) (reported for those with HPV (not 16/18) detected, but not for those with HPV16/18 detected). These findings include women classified as higher risk, and thus represent colposcopy attendance through the screening pathway (including 12-month follow-up); timeliness of attendance for colposcopy outside screening pathways (for example due to symptoms) is also important for early detection of cancer, but is not captured here.

Timely colposcopy attendance appears to vary by socioeconomic status and remoteness, which may relate to access barriers and service availability. This is of concern given that women residing in remote and very remote areas and who reside in areas of lower socioeconomic status are known to have higher rates of cervical cancer (3) (see indicator 1). Previous data from Queensland suggested that barriers to timely colposcopy follow up may impact Indigenous women to a greater extent than non-Indigenous women (48).

The available data suggest that many women may be experiencing suboptimal times to further assessment in this period of high colposcopy demand during transition to HPV based screening and from the 2-year to 5-year interval, but that prioritisation of women by referral indication (and possibly age group) is occurring. The COVID-19 pandemic is also likely to have increased waiting times from 2020 onwards across much of Australia. Given that over 80% of people referred in 2020 were referred due to repeat HPV positivity, the following policy change is important for reducing referral numbers, and thus waiting times, in future. In February 2021, on the basis of program data indicating that most people with a repeat non-16/18 positive result at 12 months were at a very low risk of high-grade disease or cancer (49), the program follow-up pathway was adjusted to recommend that most people now have a third HPV test at 24 months, allowing further time for clearance, prior to referral to (50,51).  It is also anticipated that subsequent screening rounds will detect lower rates of HPV positivity (and thus need for referral) than the initial screening round in which existing prevalent disease was detected due to the move to a more sensitive screening test.

Cervical screening is important to expedite elimination of cervical cancer, as it will take decades for the impact of HPV vaccination to be reflected in overall cervical cancer incidence (2,32). For screening to prevent cervical cancer, treatment of screen-detected high-grade abnormalities is essential.

A target has been set in the WHO strategy to accelerate elimination of cervical cancer for treatment of women with identified disease (2).

In 2019, there were 19,583 women diagnosed with histologically-confirmed high-grade cervical disease (18,093 aged 25-74 years) in Australia. Overall, 57.9%, 85.1%, and 88.5% were treated within 8 weeks, 6 months, and 12 months, respectively (Figure 10.1). In 2020, there were 18,423 women diagnosed with high-grade cervical disease (17,312 aged 25-74 years), and of these 60.9%, 85.8%, and 89.1% were treated within 8 weeks, 6 months, and 12 months, respectively.  Nationally, there was a small increase in the proportion treated within in each timeframe in 2020 compared to 2019.

The target of 90% treated was not met within 8 weeks in any group, but was met by 6 months in women diagnosed with CIN3 or AIS, and in several age groups (40-44 and 50-64 years in 2019; 35-64 and 70-74 years in 2020) (Figure 10.2).  By 12 months, the 90% target was additionally met by SA, WA and Tasmania, in inner regional areas, and in most age groups eligible for screening (in both 2019 and 2020), and additionally in ACT (2019 only), and Queensland, outer regional areas, and in SES quintiles 2 and 3 (2020 only). The proportion treated was consistently lower in the Northern Territory (79.5% within 12 months in 2019; 75.2% in 2020); in women outside the screening age range (<80% in those aged <25 or 75+); and in those whose high-grade was HSIL not otherwise specified. Among those with high-grade disease detected in 2020 specifically, the proportion treated was also lower among those living in very remote areas (79.3% within 12 months; a drop of 9.8 percentage points compared to 89.1% for high-grade disease detected in 2019).  In many other groups, the proportion treated within 12 months was close to 90% (>85%).

In most groups, the proportion of those treated within a given timeframe (8 weeks, 6 months or 12 months) was similar or slightly higher for those whose high-grade was detected in 2020, compared to those with a high-grade detected in 2019 (Figure 10.3). The exceptions to this pattern were those living in the very remote areas, NT, or in the ACT (although in the case of the ACT, this was from a starting point in 2019 that was higher than most other jurisdictions).

Source: Analysis of NCSR data (data extracted on 17th May 2022), provided by the NCSR.

High-grade abnormality is defined as HSIL NOS (S3.1), HSIL-CIN2 (S3.2), HSIL-CIN3 (S3.3), or AIS (E3.2, E3.3). A record of ablative or excisional treatment is identified through any of colposcopy records, histology records, or MBS claims for ablative or excisional treatment or a hysterectomy. Date of diagnosis is defined as the biopsy sample collection date and allows for treatment on the same date as biopsy. The proportion of people treated is reported as a crude rate across all ages, as for most demographic groups, age-specific data were not available in sufficient detail to allow for age-standardisation or restriction to ages 25-74.

This is the first time that this indicator has been reported on in Australia. A combination of colposcopy and histology records, and MBS claims for treatment (including hysterectomy) were used to identify if treatment had occurred. While this was as complete as could be identified using NCSR data, it is possible that some ablative treatments were missed if they were not recorded on colposcopy forms and also not associated with an MBS claim.

The 2030 WHO target does not stipulate a timeframe over which to measure whether treatment has occurred. We have presented results at multiple timepoints to enable better visibility of whether treatment does eventually occur (by 12 months) as well as whether it occurs in a shorter timeframe, more consistent with best practice (8 weeks; 6 months), but it is not clear to which of the timeframes we have reported the 90% target should apply. With the exception of those aged less than 25 and those living in very remote areas, most of those with high-grade cervical disease detected were treated within 8 weeks, and a very large proportion of the remainder had been treated by 6 months.  There was relatively little increase in the proportion of people treated between 6 and 12 months. Even over the relatively long timeframe of 12 months, however, there were still many groups in Australia where the 90% target was not met, and Australia did not meet the target nationally. In some cases, there may have been explicit choice to undergo a period of observation prior to treatment, for example young women or those who had not completed childbearing with CIN2, which the NCSP Guidelines allow for (52).

Reassuringly, for many groups there was an increase in the proportion with a high-grade who were treated for a high-grade diagnosed in 2020, compared to 2019, even though some parts of Australia were affected by lockdowns due to COVID-19 in 2020 and 2021. Lockdowns in 2021 would be relevant to the results for 2020, as the follow-up periods for some high-grades – particularly over 6 and 12 months – would stretch into 2021. In spite of this potential for disruptions, the proportion of people treated was similar or higher in every timeframe (8 weeks, 6 months, 12 months) for most groups, suggesting that in most cases treatment for high-grade disease was relatively unaffected by COVID-19-related factors.  The exceptions to this, however, were treatment for people living in remote areas and the Northern Territory, and in the ACT. Treatment for people in remote areas and the NT may have been more affected by COVID-19 than other areas, due to restrictions on people entering or leaving remote communities, especially Indigenous communities. Treatment for people living in the ACT may in some cases have been delayed for people who would otherwise have received treatment in NSW, but could not due to border closures. The lack of an effect in areas with lengthy strong lockdowns, like Victoria and NSW (particularly affecting Melbourne and Sydney), may have been because precancer treatment services were available within areas where people were allowed to and could feasibly travel.

Data are available by Indigenous status for this indicator from the NCSR, but were not available for publication in this report. Data by country of birth are still very incomplete on the NCSR.

Management of invasive cervical cancer in accordance with accepted guidelines is an important pillar of cervical cancer elimination (26). Incidence nationally was 6.3 per 100,000 in the period 2012-16, and notably higher amongst Aboriginal and Torres Strait Islander people and other population groups (see Indicator 1), despite the national cervical screening program. Given that the full impact of HPV vaccination in reducing cervical cancer incidence remains decades away, continued screening and early detection, in conjunction with optimal stage-appropriate treatment of cervical cancer, is vital to reduce suffering and mortality from invasive cervical cancer in the Australian population.

In Queensland, a Cancer Quality Index was developed to track progress in delivering safe, quality cancer care within health care services using State level linked data (53). In the periods 2010-2014 and 2015-2019, age adjusted cervical treatment rates (defined as at least one of major resection (hysterectomy), radiotherapy or chemotherapy) were 90% or higher in all areas of residence. (Table 11.) (54,55). In major cities and regional areas, rates reached 95% and 96% respectively in 2015-2019. The number of patients that received treatment within 30 days has declined slightly over time and was just under one third (29%) in 2015-2019. In the 2015-2019 period, there was an ongoing large disparity in the proportion of patients receiving treatment within 30 days by use of public vs private facilities but only minor differences by remoteness or older age. The proportion of patients receiving treatment within 30 days was higher amongst all First Nations People (31%) than among Non-First Nations people receiving treatment in public facilities (19%) but non First Nations People treated in private facilities had a greater proportion treated within 30 days (42%). There was also a gradient by socioeconomic status, whereby rates of treatment receipt within 30 days increased with higher socioeconomic status. The proportion of cervical cancer patients that were documented to have been reviewed by a multidisciplinary team (MDT) in 2015-2019 was low at 11% (decreased from 24% in 2010-2014), noting however that MDT review was only able to be documented through use of a specific platform that not all facilities use.

Victoria replicated the methods used in Queensland to produce a Victorian Cancer Quality Index (published in 2018 for data from 56). In Victoria in 2012-2015, rates of cervical cancer treatment were estimated to be lower than in Queensland, with 86% of inner regional women documented as receiving treatment in the linked dataset and < 80% in both metro (79%) and remote/outer regional areas (78%). It is unclear if this is a true difference or could be a result of difference in quality/completeness of underlying data sets, linkage methodology, or potentially the proportion of patients with early-stage disease (who may have been cured with local surgery only). However, rates for each type of treatment were similar to Queensland.

A 2018 study of 1,274 patients with cervical cancer utilized data from the South Australian Clinical Cancer Registry from four major public hospitals in South Australia (57).In these data from 1984-2012, 96.3% of patients received treatment for their cervical cancer. Over half of patients (53%) had Stage I disease, 23% Stage II, 16% Stage III and 10% Stage IV. Age was identified as a contributing factor to treatment rates, with those aged over 80 least likely to receive treatment (77.1%). Among Stage IV patients, 82.5% received treatment.

Source: There are no routinely published national data available reporting cervical cancer treatment rates. Therefore, data presented are from State level linked data and one published South Australian study (57). The Queensland Oncology Repository links patient data from the Queensland Cancer Register, Queensland Hospital Admitted Data Collection, on surgery, radiation therapy, and IV systemic therapy and data collected from multidisciplinary team meetings primarily in the public sector (54,55). The Victorian Department of Health linked Victorian Cancer Registry data and administrative datasets, including Victorian Admitted Episodes Dataset, the Victorian Radiotherapy Minimum data set and the Victorian Death Index (56).

Identification of suitable data sources and development of a methodology to routinely report treatment rates, or ideally optimal care treatment rates, are now required in order for Australia to report against current practice and monitor against the WHO 2030 scale up treatment indicator.

Although it is likely that over 90% of cervical cancer patients in Australia receive treatment, and the Queensland data provide a notable precedent for both feasible methods and the importance of monitoring cancer care data, the available data suggests that it is not always optimal. For example, a 2017 study reviewed data held in a Clinical Cancer Registry for 208 cervical cancer patients treated in southwest and central Sydney between July 2005 and December 2011 (58). Adherence to 10 widely accepted Clinical Practice Guidelines was higher for chemotherapy (97%) and for surgery (74-83%), but more variable for radiotherapy (47–100%). Of 133 patients with assessable data, 54% received overall guideline adherent care.

Previous research has suggested disparities in cervical cancer care for Indigenous women (59-61). Access to treatment and to culturally appropriate treatment may be substantial barriers for some patients (62). It is unknown how many patients in Australia currently receive cervical cancer care in a tertiary referral centre or according to the optimal care pathway for cervical cancer (63) which includes care informed by a multidisciplinary team.

Whilst some tertiary gynaeoncology centres do routinely publish numbers of cervical cancer patients seen, stage, demographics and some quality indicators (64-67), particular challenges in developing a national method for monitoring cervical treatment rates and proportion receiving optimal care include:

1. Lack of collection of consistent staging information across cancer registries at a State and Territory level and delay in national cancer incidence data reporting
2. Lack of explicit national level guidelines for the treatment of cervical cancer against which to determine if optimal treatment by stage is being received
3. Lack of availability of existing national and State level linked data sets to map optimal care pathways against including cancer registry data, hospitalisation data, primary care data, palliative care data, MBS/PBS data, and radiotherapy data

Encouragingly the National Gynae-Oncology Registry, a clinical quality registry (https://ngor.org.au), is aiming to eventually invite all newly diagnosed cervical cancer patients to participate and may, in doing so, overcome many of the key barriers identified above (67). This initiative may help facilitate the development of a high-quality measure to monitor cervical cancer treatment over time. Such a measure is required in order to identify and address any deficiencies and ensure optimal care is delivered equitably to cervical cancer patients across Australia.

Information on cervical cancer survival provides an indication of cancer prognosis at the individual level and the effectiveness of treatment. Relative survival indicates the percentage of people who are still alive for a specific amount of time (typically 5-years) following a cervical cancer diagnosis, relative to the expected rate in women in the general population and is expressed as a percentage.

A range of factors influence survival, including characteristics of the patient (such as age, comorbidities, and functional status), the nature of the tumour (such as stage at diagnosis and histology type) and the health system (access to services, screening behaviour, diagnostic and treatment facilities, follow-up services) (60,68).

Overall, 5-year relative cervical cancer survival for Australia was 74.2% in 2014–2018, showing minor improvements over time, rising from 71.9% in the period 1989–1993. In the period 2014–2018, 5-year survival was highest in the ACT (81.8%) and lowest in the NT (62.6%) (Figure 12.1).

Relative survival rates decreased with increasing age from 91.7% among women diagnosed in the age-group 20–24 years to 26.7% and 26.6% at 80-84 and 85+ years, respectively (Figure 12.2).

Relative survival rates were lower for Indigenous women (54.4%), women living in the lowest socioeconomic quintile (65.0%) and very remote areas (54.3%). Rates were similar for Australian-born and overseas-born women, noting that these high-level groupings may obscure diversity in survival for different culturally and linguistically diverse groups (Figure 12.3).

Source: 2018 Australian Cancer Database, with additional analyses of Cancer data provided by AIHW.

Relative survival rates provide an objective measure of survival probability from cancer, controlling for differences in mortality from other causes. It is defined as the ratio of the observed survival rate in a group of patients during a specified follow-up period, to the expected survival rate in a comparable population. Calculations of expected survival rates require information on the probability of a comparable population living to or dying at a particular age (i.e., life tables). Relative survival for Australia overall and by age group was calculated using national life tables, whereas relative survival by jurisdiction was calculated using jurisdiction-specific life tables. Life tables were not available by Indigenous status, socioeconomic status, remoteness area, or country of birth and observed (crude) survival rates are presented. Rates for women under the age of 20 years were not published due to an insufficient number of cases to calculate a stable estimate of survival.

Australian survival rate is above the international benchmarks, which generally show a range in 5-year survival between 50–70% (69). Crude survival data show disparities by Indigenous status, socioeconomic status and remoteness, which may be partially explained by diagnosis at a later stage (16,59,70).  However, comparisons of crude survival rates across the different sub-population groups should be interpreted with caution as they do not control for underlying chance of survival without cervical cancer.

Cancer stage is a strong predictor of survival, but stage at diagnosis is only routinely collected by the NSW Cancer Registry.

Data from the previous Pap based screening program shows that overall cervical cancer survival is notably higher for women with an adequate screening history, compared with lapsed or never screened women (71). This demonstrates the greater survival benefits for women for whom the cancer was screen-detected which is likely due to cervical cancers detected through screening being diagnosed at an earlier stage and be slower developing cancers than cancers diagnosed in women who were under or never screened.

Finally, the Indigenous status variable is not of sufficient quality for analysis for records from SA, Tas or the ACT. The survival statistics presented are for NSW, Vic, Qld, WA and NT combined.