What is the impact of laboratory testing rates on disease notifications in NSW?

Introduction

Surveillance of notifiable infectious diseases provides information about the number of new diagnoses of disease. Data on the level of testing is useful to indicate whether an increase in notifications may be due to increased testing rather than increased disease incidence.

The NSW Denominator Data Project commenced in January 2012 to collect the total number of tests performed per month (the denominator data) for 10 selected notifiable conditions for which the testing rate might impact the notification rate. Data for sexually transmitted infections (chlamydia and gonorrhoea), vector borne infections (Ross River fever and Barmah Forest virus infection), pertussis, enteric diseases (salmonellosis, shigellosis, cryptosporidiosis and giardiasis), and HIV are reported each month from 14 public and private laboratories. These laboratories account for 88% of the total notifications for the selected conditions in NSW

The testing data is collated to give monthly aggregated data per condition. Comparison with notifications of case reports may help to inform whether an increase in notification for a condition is a result of increased testing, and to provide an indication of a trend in incidence to inform public health action.

Methods

The laboratory denominator data is submitted by the 14 laboratories quarterly to a web-based password protected secure site to ensure confidentiality of the information. No demographic information is provided. The reported denominator data is interpreted per laboratory to account for various testing methods with the summary test information collated to provide monthly aggregated data (Appendix 1).

Under the NSW Public Health Act 2010, pathology laboratories are required to report certain conditions to NSW Health. Overall positive result data was taken from these reports for people resident in NSW diagnosed with one of 10 selected notifiable conditions. Data on the 10 notifiable conditions were analysed for the period between 1 January and 31 December 2013, based on the specimen date, for the selected conditions. The positive notification rate (per 1,000 tests performed) for all reported conditions was defined as the total number of notifications (regardless of source of notification) divided by the number of tests reported from the 14 participating laboratories. HIV was not included in the analysis due to complexities in the notification process of excluding repeat positive tests.

Results

The total number of tests reported in 2013 was higher for chlamydia, gonorrhoea, HIV, and parasitic enteric conditions (cryptosporidiosis and giardiasis) compared to 2012, but was lower for Ross River virus and Barmah Forest virus infections, and bacterial enteric conditions (salmonellosis and shigellosis). Testing for pertussis was 30% lower in 2013 compared with 2012.

The positive notification rate for included conditions ranged from 52.7 for chlamydia to 0.8% for shigellosis (Table 2). There was a significant increase in positivity for Barmah Forest virus infection (36.2 per 1,000 tests in 2013 compared to 25.3 per 1,000 tests in 2012), and significant decreases noted in positivity of pertussis (22.5 per 1,000 tests in 2013 compared to 39.4 per 1,000 tests in 2012) and chlamydia (52.7 per 1,000 tests in 2013 compared to 56.4 per 1,000 tests in 2012). Notifications for chlamydia and gonorrhoea were correlated with testing, while notification rates of enteric conditions showed a seasonal trend rather than impacts of testing patterns.

Table 2 : Number and positive notification rate of tests performed for 10 selected notifiable conditions at 14 participating laboratories for 2012.

Condition Test Number of tests Positive notification rate (per 1,000 tests)
Chlamydia C trachomatis NAT 377,745 56.4
Gonorrhoea N gonorrhoea NAT & culture 459,751 9.0
HIV# Serology 18,171 Not applicable
Ross River virus infection Serology 13,981 33.1
Barmah Forest virus infection## Serology 147,272 25.3
Pertussis NAT, serology, culture 147,272 39.4
Salmonellosis NAT, culture 193,444 15.2
Shigellosis NAT, culture 193,444 0.7
Cryptosporidiosis Antigen, microscopy 157,554 4.3
Giardiasis Antigen, microscopy 157,554 12.8

Table 3 : Number and positive notification rate of tests performed for 10 selected notifiable conditions at 14 participating laboratories for 2013.

Condition Test Number of tests Positive notification rate
(per 1,000 tests)
Chlamydia C trachomatis NAT 395,131 52.7
Gonorrhoea N gonorrhoea NAT & culture 491,080 8.6
HIV# Serology 444,551 Not applicable
Ross River virus infection Serology 15,237 33.1
Barmah Forest virus infection## Serology 11,978 36.2
Pertussis NAT, serology, culture 102,646 22.5
Salmonellosis NAT, culture 188,060 18.3
Shigellosis NAT, culture 188,060 0.8
Cryptosporidiosis Antigen, microscopy 167,224 6.6
Giardiasis Antigen, microscopy 167,224 13.2

# HIV notifications were excluded due to the complexity of the notifications process and exclusion of repeat tests.
## Notifications for Barmah Forest virus infection for the period October 2012 – July 2013 should be interpreted with caution due to known false positive readings from the only available commercial IgM assay.

Vaccine preventable disease - Pertussis

The total number of tests for pertussis was calculated using a combination of serology, NAT and culture. Historically there has been a pattern of more pertussis diagnosis in summer months, but more pertussis testing occurring in winter. The number of tests in 2013 (102,646) declined significantly compared with 2012 (147,272). The positive notification rate also declined to 22.5 per 1,000 tests (range 13.0 to 45.5 per 1,000 tests) in 2013 compared with 39.4 per 1,000 tests (range 22.4 to 88.6 per 1,000 tests) in 2012 (Figure 1).

Figure 1: Comparison of pertussis notifications, tests performed, and positive notification rate per 1,000 tests - NSW 2012 - 2013.

Comparison of pertussis notificatifcations, tests performed. Text version in Appendix 2Comparison of pertussis notificatifcations, positive notification rate per 1000 tests .Text version in Appendix 2 

Dotted lines denote 95% confidence interval for positivity

Sexually transmitted infections – Chlamydia and Gonorrhoea

The total number of tests for chlamydia and gonorrhoea was calculated using NAT screening, with additional cultures performed for gonorrhoea. The positive notification rate for chlamydia was the highest among the diseases analysed in this period, with 52.7 positive tests notified across NSW per 1000 tests performed by the 14 laboratories in 2013 but was significantly lower compared with 2012 (56.7 per 1,000 tests) (Figure 2). For gonorrhoea, 8.6 positive tests were notified across NSW for every 1,000 tests performed by the 14 laboratories in 2013, similar to 2012 when 9.0 positive tests were notified for every 1,000 tests performed (Figure 3). Overall, positive notification rates declined over the period.

Figure 2: Comparison of chlamydia notifications, tests performed, and positive notification rate per 1,000 tests - NSW, 2012 –2013.

Comparison of chlamydia notifications,tests performed. Text version in appendix 2Comparison of chlamydia notifications, positive notification rate per 1,000 tests. Text version in Appendix 2 

Dotted lines denote 95% confidence interval for positivity

Figure 3: Comparison of gonorrhoea notifications, tests performed, and positive notification rate per 1,000 tests - NSW, 2012- 2013
Comparison of gonorrhoea notifications, tests performed. Text version in Appendix 2Comparison of gonorrhoea notifications, positive notification rate per 1,000 tests.Text version provided in Appendix 2

Dotted lines denote 95% confidence interval for positivity

Vector borne disease – Ross River Virus and Barmah Forest virus infections

The number of tests for Ross River virus (RRV) and Barmah Forest virus (BFV) were similar as testing is predominantly done in combination. There was a decline in testing for both infections in 2013 (15,237 RRV tests, 11,978 BFV tests) compared with 2012 (18,171 RRV tests, 13,981 BFV tests). There was a seasonal trend in testing and positivity in both 2012 and 2013, with more testing and notifications during summer and autumn.

For RRV infection, compared to the previous year, there was a significant increase in positivity in May 2013 without a corresponding increase in testing (Figure 4). For BFV infection, an increase in notifications from October 2012 was noted (Figure 5) which had also been reported nationally. The increase continued into 2013 and has been attributed to false positive results from the only available commercial IgM assay. This assay was adjusted in mid 2013, and a subsequent decline in notifications has been noted. Due to this issue, 2013 data for BFV should be interpreted with caution.

Figure 4: Comparison of Ross River notifications, tests performed, and positive notification rate per 1,000 tests - NSW, 2012- 2013.
Comparison of Ross River notifications, tests performed. Text version in Appendix 2Comparison of Ross River notifications, positive notifications rate per 1,000 tests. Text version in Appendix 2

Dotted lines denote 95% confidence interval for positivity

Figure 5: Comparison of Barmah Forest notifications, tests performed, and positive notification rate per 1,000 tests# - NSW, 2012- 2013.
Comparison of Barmah Forest notifications, tests performed. Text version in appendix 2Comparison of Barmah Forest notifications, positive notification rate per 1,000 tests. Text version in Appendix 2

Dotted lines denote 95% confidence interval for positivity

# Positivity results for Barmah Forest Virus should be interpreted with caution. An issue with false positive rates in the only available commercial IgM assay was noted in October 2012 and the assay adjusted in mid 2013.

Parasitic enteric diseases – Cryptosporidiosis and Giardiasis

Cryptosporidiosis and giardiasis are screened using the same test, so the denominator was the same for both conditions. In 2013, the total numbers of tests per month were stable across the period, while the notifications for both diseases varied with higher notifications in the summer months. There were significant spikes in notifications and positivity rates for both diseases during February 2013 (Figures 6 and 7).

Figure 6: Comparison of cryptosporidiosis notifications, number of faecal parasitic screening and antigen tests performed and positive notification rate per 1,000 tests - NSW, 2012- 2013.
Comparison of cryptosporidiosis notifications,tests performed. Text version in Appendix 2Comparison of cryptosporidiosis notifications, positive notification rate per 1,000 tests. Text version in appendix 2

Dotted lines denote 95% confidence interval for positivity

Figure 7: Comparison of giardiasis notifications, total number of faecal parasitic screening and antigen tests performed and positive notification rate per 1,000 tests - NSW, 2012- 2013.

Comparison of giardiasis notifications, tests performed. Text version in Appendix 2Comparison of giardiasis notifications, positive notification rate per 1,000 tests. Text version in Appendix 2 

Dotted lines denote 95% confidence interval for positivity

Bacterial enteric diseases - Salmonellosis and Shigellosis

Salmonellosis and shigellosis are screened using the same test, so the denominator was the same for both conditions. Similar to the parasitic enteric conditions, the total number of tests remained stable over the period. The notifications for salmonellosis followed a seasonal pattern, with high notifications and positivity rates in summer months (Figure 8). The relatively small number of shigellosis notifications shows little change over time (Figure 9).

Figure 8: Comparison of salmonellosis notifications, total number of faecal NAT and cultures performed and positive rate per 1,000 tests - NSW, 2012- 2013.

Comparison of salmonellosis notifications, tests performed. Text version in Appendix 2Comparison of salmonellosis notifications, positive notification rate per 1,000 tests. Text version in Appendix 2 

Dotted lines denote 95% confidence interval for positivity

Figure 9: Comparison of shigellosis notifications, total number of faecal NAT and cultures performed and positivity rate per 1,000 tests - NSW, 2012- 2013.

Comparison of shigellosis notifications,tests performed. Text version in Appendix 2Comparison of shigellosis notifications, positive notification rate per 1,000 tests. Text version in Appendix 2  

Dotted lines denote 95% confidence interval for positivity

Discussion

The NSW Denominator Data Project, based on ongoing collection of testing data, aims to provide timely information to aid interpretation of trends in notification data. The data has shown that the positive notification rate for selected conditions in 2013 ranged from 52.7 per 1,000 tests for chlamydia to 0.8 per 1,000 tests for shigellosis. The similar pattern of testing and disease notifications, and subsequent stable positive testing rates noted for sexually transmitted infections, suggest (as with other similar studies) that chlamydia and gonorrhoea notifications are highly correlated with testing levels.

In contrast, testing throughout the year of enteric conditions, both parasitic and bacterial, suggests that the incidence reflects seasonality in patterns of infection rather than changes to rates of testing. In 2013, while pertussis testing increased over winter months (likely in response to increased cough illness), positivity rates remained low. This indicates a true reduction of pertussis circulating in the community in 2013.

This analysis is subject to several limitations. Firstly, the testing data were collected from 14 participating laboratories and is not a total account of all testing in NSW. However, it is estimated that the laboratories included account for around 88% of the total notifications for the selected conditions. Secondly, the analysis only included 24 months of data, limiting the ability to interpret longer term trends. Thirdly, analysis of age-specific rates or specific rates per geographic area could not be calculated as demographic data including age, sex, or postcode was not collected. Lastly, there may be some duplication in test counts for conditions where more than one method of testing may be used (pertussis, gonorrhoea and enteric conditions). In addition, multiple tests could be conducted on an individual for a single condition.

Conclusion

Overall the project has shown value in interpretation of notification data and the impact of testing. For example, in understanding the impact that targeted testing campaigns may have on increased testing and subsequent notifications for STIs. Further work is underway to enhance the usefulness of the NSW denominator data project including: consideration of adding some demographic data, addition of remaining NSW laboratories which are not currently involved in the project, and a review of the selected conditions.

Acknowledgements

Health Protection NSW would like to acknowledge the work of the following laboratories in providing data for the project:

Pathology North (Hunter Area Pathology Service, Pacific Laboratory Medicine Services, Royal North Shore Hospital, Central Coast); Pathology West; South Eastern Area Laboratory Services - North and Central, SEALS South; Sydney South West Pathology Services (SSWPS) (Concord Laboratory, Liverpool Laboratory, Royal Prince Alfred Hospital Laboratory); Children’s Hospital Westmead; Douglass Hanly Moir Pathology; Healthscope; Laverty Pathology; SydPath.

Appendix 1: Selected notifiable conditions and tests included in the denominator data project

Notifiable Conditions Tests performed and collected Disease type / conditions
Chlamydia screening for Chlamydia trachomatis using NAT (single or combo) Sexually transmitted infections
Gonorrhoea screening for Neissera gonorrhoea using NAT (single or combo) and/or culture Sexually transmitted infections
HIV serology (not confirmatory testing) Blood borne virus/ Sexually transmitted infections
Pertussis screening for Bordetella pertussis using NAT, serology and/or culture Vaccine preventable diseases
Ross River virus serology (not confirmatory testing) Vector borne diseases
Barmah Forest virus serology (not confirmatory testing) Vector borne diseases
Shigellosis faecal specimens examined for Shigella and Salmonella by culture or NAT Bacterial enteric conditions
Salmonellosis faecal specimens examined for Shigella and Salmonella by culture or NAT Bacterial enteric conditions
Giardiasis faecal specimens examined by OCP screening and/or antigen testing Parasitic enteric conditions
Cryptosporidiosis faecal specimens examined by OCP screening and/or antigen testing Parasitic enteric conditions

NAT: nucleic acid test; OCP: ova cysts and parasites

Appendix 2: Number of tests, notifications and positive notification rates per 1,000 tests for selected notifiable conditions in NSW, January to December 2013

Cryptosporidium

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 116 201 240 174 113 52 41 19 17 24 62 52 1111
Number of tests 12116 14202 14460 12812 14174 12149 13790 14733 13869 15530 16284 13105 167224
Positive percent 9.6 14.2 16.6 13.6 8.0 4.3 3.0 1.3 1.2 1.5 3.8 4.0 6.6

Giardia

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 192 273 216 204 204 153 161 158 154 172 180 138 2205
Number of tests 12116 14202 14460 12812 14174 12149 13790 14733 13869 15530 16284 13105 167224
Positive percent 15.8 19.2 14.9 15.9 14.4 12.6 11.7 10.7 11.1 11.1 11.1 10.5 13.2

Salmonella

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 410 378 344 352 290 201 196 173 172 322 278 328 3444
Number of tests 14776 15591 16263 14792 15543 13782 15124 16157 15367 16988 18913 14764 188060
Positive percent 27.7 24.2 21.2 23.8 18.7 14.6 13.0 10.7 11.2 19.0 14.7 22.2 18.3

Shigella

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 14 10 13 6 11 5 13 8 12 20 13 18 143
Number of tests 14776 15591 16263 14792 15543 13782 15124 16157 15367 16988 18913 14764 188060
Positive percent 0.9 0.6 0.8 0.4 0.7 0.4 0.9 0.5 0.8 1.2 0.7 1.2 0.8

Ross River

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 41 46 31 59 100 50 34 22 27 32 34 28 504
Number of tests 1432 1555 1505 1294 1285 1007 1063 1102 1159 1202 1331 1302 15237
Positive percent 28.6 29.6 20.6 45.6 77.8 49.7 32.0 20.0 23.3 26.6 25.5 21.5 33.1

Barmah Forest

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 39 53 47 51 61 44 24 26 26 15 23 25 434
Number of tests 1037 1191 1111 1008 1005 803 861 857 951 914 1155 1085 11978
Positive percent 37.6 44.5 42.3 50.6 60.7 54.8 27.9 30.3 27.3 16.4 19.9 23.0 36.2

HIV

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Number
of tests
35557 37374 36764 36763 40891 35520 38500 37619 36284 38666 38868 31745 4445

Chlamydia

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 1780 1862 1719 1659 1875 1562 1789 1727 1573 1818 1901 1559 20824
Number of tests 32943 33315 33246 32028 35496 30507 33959 34066 31022 33805 34396 30348 395131
Positive percent 54.0 55.9 51.7 51.8 52.8 51.2 52.7 50.7 50.7 53.8 55.3 51.4 52.7

Gonorrhoea

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 400 356 352 369 340 319 384 367 318 342 359 328 4234
Number of tests 39731 41713 42188 39377 44104 37835 42107 42245 38897 42005 43182 37696 491080
Positive percent 10.1 8.5 8.3 9.4 7.7 8.4 9.1 8.7 8.2 8.1 8.3 8.7 8.6

Pertussis

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Positive cases 280 219 134 160 210 122 184 175 193 207 221 200 2305
Number of tests 6155 6650 7236 7616 9802 9597 11301 13410 10917 8116 6483 5363 102646
Positive percent 45.5 32.9 18.5 21.0 21.4 12.7 16.3 13.0 17.7 25.5 34.1 37.3 22.5

Prison health

Dr Marianne Gale: Public Health Officer Trainee

Ms Denise Monkley: Service Director Population Health, Justice Health and Forensic Mental Health Network

Dr Penny Abbott: Senior Lecturer, Department of General Practice, University of Western Sydney

In Australia, there are approximately 33 000 people in prison custody, one third of whom are in NSW.1 More than 90% of all people in custody are males, and a high proportion come from an Aboriginal or Torres Strait Islander background (23% of adult inmates, 48% of adolescents).
Prison facilities have an inherently high potential for communicable disease transmission. This article summarises the key contributing factors.

Inmate health status pre-incarceration

A high proportion of the prison inmate population comes from disadvantaged groups in the community in which the prevalence of both communicable and non-communicable diseases is higher than in the general population. A 2009 survey of 996 adult inmates in NSW showed that a high proportion of inmates have significant physical health issues related to both communicable and non-communicable disease.2 The high prevalence of mental health and substance misuse problems among prison inmates can make addressing physical health issues even more challenging.

Table 1 displays a snapshot of some of the indicators of the health status of NSW prison inmates. Data is displayed as unadjusted proportions based on self-reported information from prison inmates, except for Hepatitis B and C, which are based on serological testing.

Table 1: Snapshot of selected health conditions of NSW prison inmates, 2009 for NSW Custodial Population

Males Females
Heart Disease 19% 24%
Current tobacco smoker 75% 80%
Evidence of current or past Hepatitis B 28% 34%
Evidence of current or past Hepatitis C 28% 45%
Sexually transmitted infection (ever) 41% 47%
Mental health problem 47% 54%

Source: 2009 NSW Inmate Survey

Access to health services in prison

Justice Health and Forensic Mental Health Network (JH&FMHN) are a statewide service that provides health care for people in contact with the forensic mental health and criminal justice systems. JH&FMHN provide services in over 100 sites across NSW including inpatient facilities, custody facilities and the community.

A key part of reducing the risk of communicable disease transmission in prison is to have effective systems to screen inmates on arrival, diagnose communicable disease early, and provide timely access to appropriate treatment and management.

JH&FMHN offer voluntary testing for blood borne viruses to all people in prison custody with access to specialist medical management. JH&FMHN also offer screening and treatment for other conditions to all prison entrants including sexually transmitted infections and tuberculosis.

Risk behaviour in prison

Risk behaviour in prison can be a driver of communicable disease transmission in the absence of effective harm minimisation programs. In NSW, JH&FMHN and Corrective Services NSW work together to implement a range of harm minimisation strategies in custodial settings. These include information and education, condom availability in vending machines, bleach and detergent availability, opiate substitution therapy and training for inmates on safe barbering and hairdressing.

Table 2 displays some of the risk behaviours self reported by NSW inmates. Practices like injecting drugs and tattooing do occur despite being banned activities in the prison setting.

Table 2: Snapshot of risk behaviours reported by NSW prison inmates, 2009 for NSW Custodial Population

Males Females
Ever injected drugs in prison 16% 16%
Used injecting equipment after another person >65% >65%
At least one tattoo obtained in prison 39% 21%
At least one piercing obtained in prison 14% 14%
Engaged in sexual activity with another inmate 2% 12%
Tried to access condoms/dams in prison 16% 12%

Source: 2009 NSW Inmate Health Survey

Prison conditions

The environment of the prison itself carries inherent risks for communicable disease transmission. In many countries, poor water and sanitation, overcrowding and poor food safety standards have led to outbreaks of infectious diseases in prison.

In NSW, rigorous cleaning and infection control standards for prisons are in place. In addition, JH&FMHN conduct environmental monitoring activities that include a safe drinking water program, a food safety program, environmental health assessments and building safety. As the prison population in NSW continues to grow, the potential for overcrowding is a public health risk that requires close monitoring. The frequent transfer of prison inmates between different facilities across the state is another important reality of the prison environment that represents challenges to public health.

Conclusion

Prison facilities by their very nature have a high potential for the transmission of communicable diseases. Ongoing vigilance and support for the activities in place by JH&FMHN are necessary to maintain low rates of disease transmission in prisons and therefore also in the broader community.

References

  1. Australian Bureau of Statistics, Prisoners in Australia 2013. ABS Dec 2013
  2. Indig D, Topp L, Ross B et al. 2009 NSW Inmate Health Survey: Key Findings Report. Justice Health. Sydney
​​
Page Updated: Tuesday 30 September 2014
Contact page owner: Health Protection NSW