Healthcare workers and immunity to infectious diseases
Sanjyot Vagholkar, GP Unit, Fairfield Hospital, PO Box 5, Fairfield, NSW 1860. Fax: (02) 9616 8400; e-mail: email@example.com
Objective: In 2002, New South Wales (NSW) Health introduced an updated policy for occupational screening and vaccination against infectious diseases. This study describes healthcare worker (HCW) immunity to hepatitis B, measles, mumps, rubella (MMR) and varicella based on serological screening, following introduction of this policy.
Methods: HCW screening serology performed at two healthcare facilities in south western Sydney (Bankstown and Fairfield) was extracted for the period September 2003 to September 2005. Immunity to hepatitis B, MMR and varicella was quantitated and cross-tabulated against age, sex and staff risk category.
Results: A total of 1,320 HCWs were screened. Almost two thirds were immune to hepatitis B while immunity to MMR and varicella ranged from 88% to 94%. Age stratification showed lower levels of measles immunity in those born after 1965.
Conclusions: Despite availability of vaccination for over two decades, a significant proportion of HCWs at these two facilities were non-immune to hepatitis B. This is of concern for those non-immune staff involved in direct clinical care, who are at risk of blood and body fluid exposures. The small group of HCWs non-immune to MMR and varicella pose a risk to themselves and others in the event of an outbreak.
Implications: There is a need for improved implementation of the occupational screening and vaccination policy, including better education of HCWs about the risks of non-immunity to vaccine preventable diseases. The revised 2007 NSW Health policy may assist this process and will need evaluation to determine whether HCW immunity improves in the coming years.
Hospital and community-based healthcare settings pose potential risks to healthcare workers (HCW) for acquiring infectious diseases. Ensuring HCW immunity against infections such as hepatitis B, measles and varicella has been recognised as important in contributing to a safe environment for staff and patients.1 Outbreaks of these infections have the potential to cause significant morbidity and have workforce and cost implications for healthcare organisations. Over the past two decades health care authorities have developed guidelines to ensure protection for HCWs against vaccine preventable diseases.2–4
Studies from overseas have however shown HCW immunity to vaccine preventable diseases is not optimal. Immunity to measles and varicella has varied from 92% to 98% of HCWs in hospital settings.5–7 For hepatitis B, the levels are lower, commonly 65%-75%.8,9 Most Australian research exploring HCWs and immunity to infectious diseases is based on self-reported surveys, with limited published data based on serological evidence. In New South Wales (NSW) a survey of all public and private hospitals showed widespread hepatitis B vaccination among medical staff but low rates of screening and vaccination for measles, mumps, rubella and varicella.10 Despite this, one-third were not immune to hepatitis B based on a negative or incomplete vaccination history. The evidence to date is that even though there are occupational screening and vaccination programs offered by healthcare facilities, uptake of these programs is variable and universal protection of HCWs is yet to be achieved.
In 2002, NSW Health released an updated policy for occupational screening and vaccination against infectious diseases.11 This classified staff into four risk categories (A-D) based on their work duties (Table 3 Notes [a]) and recommended screening and vaccination requirements for hepatitis B, measles, mumps, rubella, varicella, tetanus, pertussis, influenza and tuberculosis. Two healthcare facilities in Sydney South West Area Health Service (SSWAHS), Bankstown and Fairfield, commenced implementation of the policy in late 2003 and provided screening and vaccination for staff at general practice clinics. The Bankstown program was provided by the Eldridge Road Medical Centre and the GP Unit, Fairfield Hospital provided a service for Fairfield staff. Both healthcare facilities are located in south western Sydney and provide acute hospital and community health services. The staff reflect the ethnic diversity of this region of Sydney with many born overseas in Asia, South America and the Middle East. Staff numbers are in constant flux but during the study period Bankstown had approximately 1,100 employees and Fairfield 800.
Table 3. Fairfield hepatitis B immunity and staff category.
Screening data from these two centres in SSWAHS provided an opportunity to examine HCW immunity in an Australian setting, based on serological results. The aim of this study was to describe the uptake of occupational screening by staff of the two facilities over a two-year period (2003-05) and report the level of immunity at the time of screening to the following infectious diseases: hepatitis B, measles, mumps, rubella (MMR) and varicella.
This was a descriptive study involving a retrospective analysis of data collected by the Fairfield and Bankstown staff screening and vaccination centres. De-identified data about patients’ age, sex and immunity to hepatitis B, measles, mumps, rubella and varicella was entered into an Access database. At Fairfield staff risk category (Table 3 Notes [a]) was also available. Data was extracted for the period 1 September 2003 to 31 August 2005 inclusive.
Serology tests for the screening program had been conducted by two accredited Australian pathology providers, both of which take part in an ongoing quality assurance program. Results were based on established commercial assays used for routine clinical purposes. Immunity to the infectious diseases was defined according to the following results: Hepatitis B: anti-HBs >10 mIU/ml; Measles: IgG positive; Mumps: IgG positive; Rubella: IgG >10 IU/ml; Varicella: IgG positive. These were the standard cut-offs provided by the laboratories.
The de-identified data was analysed by the Statistical Package for the Social Sciences (SPSS). Demographic data and immunity levels are presented quantitatively. Cross tabulation was performed of immunity against age, sex and staff risk category. Age stratification for MMR immunity was based on two dates — pre 1966 before which exposure to wild virus was expected and 1980 when measles-mumps vaccination became established as part of the national childhood immunisation schedule.
All staff who undertook screening provided written consent to the process. This consent included the proviso that the results would be accessible to clinical staff of the Area Health Service where relevant to the screening and vaccination program. Four of the investigators (SV, JN, JB, NZ) were clinicians involved in the provision of the occupational screening program at Fairfield and Bankstown. The study received ethical approval from the Sydney South West Human Research Ethics Committee to analyse aggregated de-identified data from the program. Findings of the study were provided to the SSWAHS Executive prior to submission for publication.
From September 2003 to August 2005, 743 staff at Bankstown (67% of employees) and 577 at Fairfield (72% of employees) participated in the occupational screening and vaccination program. The mean age of staff screened at both centres was similar (Bankstown 42.3 years, range 16-84 years, and Fairfield 41.9 years, range 19-75 years). The large age range was due to the presence of elderly volunteer staff. At both centres the majority of employees screened were female (Fairfield 88.9% and Bankstown 80.2%).
Levels of immunity to hepatitis B, measles, mumps, rubella and varicella at both centres are presented in Table 1. Combining data from the two facilities gave the following overall rates of immunity: hepatitis B 64.1% (846/1320), measles 91.5% (1208/1320), mumps 88.7% (1171/1320), rubella 91.1% (1203/1320) and varicella 89.1% (1176/1320). Age stratification showed lower levels of measles and mumps immunity in those born after 1965 (Table 2). Rubella immunity was fairly consistent across the age bands. There was no significant change in immunity levels for each of the infectious diseases between staff screened in 2003/04 and 2004/05.
Table 1. Serology results Bankstown and Fairfield Health Services.
Table 2. MMR serologya and period of birth.
At both centres close to a third of staff screened did not have immunity to hepatitis B. At Fairfield, where stratification by staff category was possible (Table 3), this figure remained similar for those in the highest risk category (category A) which consists predominantly of doctors and nurses. Among those with slightly less risk of patient exposure (category B) the level of non-immunity rose further to just over 40%. A small number (less than five) of staff were found to be chronic hepatitis B carriers as a consequence of undergoing screening.
There was some variation in the completeness of the data due to differences in the way the health policy was implemented at each site. At Fairfield, those staff that had no patient exposure (category D) were not tested for hepatitis B and varicella immunity as per the policy guidelines. At Bankstown, all staff were screened for immunity to these viruses.
At Fairfield, staff with a definite clinical history of varicella were not tested, again in keeping with the policy. Those who had a clinical history together with those who had positive serology were classified as immune.
Immunity data were sex stratified and there were no notable differences. This data is therefore not presented.
This study analysed two years of occupational screening data from two health care facilities in Sydney. At both locations approximately two thirds of staff were screened thus providing a substantial sample of serological results and risk category breakdown from Fairfield (Table 3) showed almost two-thirds of those screened were in the highest risk group. All public healthcare facilities in NSW were required to implement the 2002 health policy,11 however, the manner in which implementation took place was left up to individual healthcare facilities. At Bankstown and Fairfield facilities, although staff were made aware of the policy requirements and directed to undertake the screening process, the program relied on staff voluntarily presenting for assessment. Therefore, there was not universal coverage of all staff.
In both centres immunity to hepatitis B was not high, with a third of the staff screened non-immune. Previous research in Australia and overseas has found similar findings.8–10 At Fairfield where stratification by staff risk category was provided there was a difference in immunity levels between category A and B. This is not surprising given doctors and nurses (category A) are more likely to have been offered hepatitis B vaccination during training or on employment. The poor level of immunity among laboratory staff cannot be easily explained but given the small number (six) it may not be representative of this group across the AHS.
There are a number of reasons why HCWs may be non-immune to hepatitis B. Some of those tested in this program may have completed a course of vaccinations in the past and the low titre of anti-HBs may reflect waning immunity. It is now accepted that if they are exposed to hepatitis B, an anamnestic response will protect those who sero-converted after a previous course of vaccinations, and booster doses are not recommended.12,13 Some of the apparently non-immune in our sample may therefore be clinically protected. A proportion of the non-immune would be non-responders to vaccination and a small number of staff were found to be chronic hepatitis B carriers as a consequence of undergoing screening.
Previous research has shown that the majority of non-immune are people who have failed to be vaccinated or had an incomplete course of vaccination.9,10 Given the availability of an effective vaccination for hepatitis B for over two decades the number of non-immune staff is both disappointing and concerning and raises issues for the occupational safety of staff as well as for the delivery of HCW vaccination programs. Reasons for not undertaking vaccination were not explored in this analysis. Previous studies have found that the reasons staff opts not to be vaccinated or fail to complete a course for hepatitis B and other vaccine preventable diseases are both HCW and systems dependent. Some HCWs choose not to be vaccinated based on concerns about side-effects and efficacy of the vaccine and perception that they are not at risk.14,15 Implementation of vaccination policies by healthcare facilities is often inadequate with problems related to access to occupational health services, inadequate education of staff about vaccination policies and costs of vaccination and follow-up.14–16
Measles immunity at the two locations was 93.7% (Bankstown) and 88.7% (Fairfield). Age stratification showed that the younger age groups (born post 1965) demonstrated lower levels of immunity, particularly at Fairfield where 20% were non-immune. This is consistent with the history of expected exposure to wild virus if born before 1966. However, it is of note that even in the older age group there was a small percentage that was non-immune. A previous study in a Sydney hospital had found higher levels of immunity (98.3%) although this was a population of healthcare workers in a children's hospital.17 Overseas studies have also demonstrated slightly better rates of immunity with only 3%-7% non-immune but similarly showed that this was generally in those under 40 years of age.5,6
Mumps immunity showed a similar pattern to measles immunity with a drop in the younger population although not as marked as that seen with measles. Rubella immunity at the two centres was more consistent across the age bands with a 6%-7% non-immune group apart from those born after 1980 at Fairfield who were all immune. Given rubella vaccination was established for high school girls in 1971 it is a little surprising that the levels of immunity were not higher as the demographics of the sample were predominantly female. A possible contributing factor to the levels of non-immunity could be the ethnic diversity of the staff. A number of the staff originate from Asia and in many of these countries measles vaccine alone is administered, not a combination MMR vaccine.18
Similar to MMR immunity, a pool of screened staff appeared non-immune to varicella in both Bankstown (6.5%) and Fairfield (7.1%). These figures were slightly higher than overseas findings, which showed non-immune rates among HCWs of 1.6-4.1%.6,7,19 In our analysis there was a difference between the two centres in who was screened for varicella immunity. At Bankstown, all staff had serological testing for varicella, while Fairfield immunity figures were based on a combination of positive serology or clinical history. This was a reflection of the fact that implementation of the policy was left to each facility in the AHS. Previous studies have shown clinical history is strongly associated with immunity,7,19–21 therefore it is reasonable to assume the group classified as immune based on history were serologically immune. NSW Health policy is based on this premise and stipulates that staff with a definite clinical history can be assumed to be immune to varicella. Category D staff (no clinical contact) at Fairfield were not tested according to policy recommendations.
Given the finding that up to 10% of staff at Bankstown and Fairfield was non-immune to measles, mumps, rubella or varicella there is clearly a pool of unprotected HCWs, which may be clinically significant in the event of a single exposure or outbreak. In the younger cohort, the finding of 20% non-immunity to measles at Fairfield was particularly concerning. Measles transmission in healthcare settings has previously occurred both in Australia and overseas,22–24 resulting in infected HCWs and patients and highlights the problem of non-immune staff. There is a need to be vigilant about checking measles vaccination history and immunity in those born after 1965. Research examining the comparative costs of dealing with an outbreak versus an active screening and vaccination program for staff has shown the latter approach can save money and prevent morbidity.25,26
This analysis was limited by the fact that it did not have serological data for all HCWs employed by the two centres. The program relied on staff voluntarily presenting for screening following general directives from the employer and department managers, it was not therefore a random sample of staff. This means there may be bias in our findings as those screened may have a greater commitment to ensuring their immunity to infectious diseases. Rates of non-immunity among those not screened could potentially be greater than in our screened subset.
The study was also subject to limitations as this was a retrospective analysis of working clinical databases, which meant there was some missing data. The data was also influenced by the different ways in which the policy was implemented at each centre thus resulting in discrepancies in what serological results were available for varicella immunity and for category D staff from each location.
Despite these limitations, it is likely the findings from these two centres are transferable to other healthcare facilities in Australia, given there are no features that particularly distinguish these organisations from others nationally. Our analysis highlighted that there is a gap between occupational screening and vaccination policy and its successful implementation. Since this study was completed NSW Health has released its revised policy for occupational screening and vaccination in February 2007.2 This policy places greater responsibility on both the employer and employee to ensure protection of HCWs and there are directives to redeploy staff from designated clinical areas if they do not have immunity to vaccine preventable infectious diseases.
In order to improve staff screening and vaccination, health services need to invest more resources in this aspect of occupational health and safety. Explicit directions about how to implement the policy and consistency across organisations would be beneficial as currently there is variation in implementation between facilities, which contributes to the problems. Essential elements of any program include adequate staff to administer it, accessible services and good follow-up and documentation of screening and vaccination status of staff. An electronic database for storing this information, which can be shared across healthcare facilities, is also required. This would prevent duplication of screening processes and assist the mobile workforce in the healthcare system. Education of HCWs about policy requirements is also an important component of this process. Ensuring immunity to infectious diseases should be seen as an essential requirement of being employed such as professional registration. Redeployment has gone further than previous policies but whether there need to be penalties for failure to comply may possibly need consideration. This need for effective implementation strategies has been highlighted in previous work.10,27
The 2007 policy2 has acted as an impetus to Area Health Services to address some of the implementation issues outlined above and funding provided by NSW Health tied to targets and provision of free vaccines for 12 months will assist the program. The new policy will need to be evaluated over the next few years to see if the stronger directive results in improved levels of HCW participation and immunity. There is clearly room to improve levels of immunity although 100% coverage is an unrealistic goal. Setting achievable targets and resourcing the program adequately over the next few years would allow HCW protection to progressively increase.
We would like to acknowledge Dr Kerry Chant, Director of Health Protection & Deputy Chief Health Officer, NSW Health, who provided advice and comments for this study. Thank you to Dr Sarah Dennis, Senior Research Fellow UNSW, for statistical advice and Charmaine Rodricks, administrative assistant GP Unit, for assistance with the databases and preparation of the manuscript.