If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The rise in antimicrobial resistance has highlighted the importance of surgical site infection (SSI) prevention with effective surveillance strategies playing a key role in improving patient safety.
Aim
To map national needs and priorities for SSI surveillance against current national surveillance activity.
Methods
This study analysed SSI surveillance in National Health Service (NHS) hospitals in England covering 23 surgical procedures. Data collected were: (i) annual number of procedures, (ii) SSI rates from national reports, (iii) national reporting requirement (mandatory, voluntary, not offered), (iv) priority ranking from a survey of 84 English NHS hospitals, (v) excess length of stay and costs from the literature. The relationships between estimated SSI burden, national surveillance activity, and hospital-reported priorities were explored with descriptive and univariate analyses.
Findings
Among the 23 surgical categories analysed, top priority ranking by hospitals was associated only with current surveillance (r = 0.76, P < 0.01) and mandatory reporting (33% vs 8 and 4%, P = 0.04). Percentage of hospitals undertaking surveillance, mandatory reporting, and the selection of priorities did not match SSI burden. Large bowel surgery (LBS, voluntary) and caesarean section (not offered) were the two highest contributors of total SSIs per annum, with 39,000 (38%) and 17,000 (16%) respectively, while the four orthopaedic categories (all mandatory) contributed 5000 (5%). LBS also had the highest associated costs (£119 million per annum).
Conclusion
Current surveillance and future priorities were not associated with SSI rate, volume, or cost to hospitals. The two highest contributors of SSIs and related costs have no (caesarean section) or limited (LBS) coverage by national surveillance.
GlobalSurg Collaborative Surgical site infection after gastrointestinal surgery in high-income, middle-income, and low-income countries: a prospective, international, multicentre cohort study.
Burden of six healthcare-associated infections on European population health: estimating incidence-based disability-adjusted life years through a population prevalence-based modelling study.
Surveillance of SSI is a key component of infection prevention and control (IPC) programmes as sustained surveillance has been shown to drive down infection rates [
]. The national public health institute, Public Health England (PHE), co-ordinates surveillance through the Surgical Site Infection Surveillance Service (SSISS). Prospective data collection and follow-up are performed on an individual basis by a specially trained healthcare professional. Surveillance is mandatory in National Health Service (NHS) hospitals for at least one quarter per financial year in at least one of four orthopaedic procedures, and hospitals can submit data voluntarily for 13 other surgical categories [
]. The results revealed heterogeneity in priority areas but with strong support for orthopaedic surveillance and caesarean section, the latter not currently included in the programme. However, an objective assessment of the relative burden of SSIs across different surgical categories is needed to inform distribution of surveillance resources.
This study aims to assess the consistency between the current surveillance systems, perception of priorities by NHS trusts, and the associated SSI burden in England. We mapped the estimated annual number of SSIs and associated economic burden in each surgical category, against current national surveillance activity, government priority status (mandatory vs voluntary), and the average priority ranking given by trusts in the survey [
The study included 13 clean (minimum wound class I) and 10 clean-contaminated (minimum wound class II) procedure types, including 17 procedures offered by the SSISS (four mandatory, 13 voluntary). Six additional categories for which national surveillance is not currently undertaken were included as they had an average rank in the top five in the survey [
]. Codes that could be used for more than one category were excluded. For surgical categories not included in the SSISS, the procedure types for which SSI rates were available were identified in the literature and corresponding OPCS codes were advised by medical coders (Supplementary Table A1).
Priorities and current surveillance practices were obtained from 84 responses to a survey emailed to 161 IPC teams at all acute trusts in England by PHE in 2013 [
Data on SSI rates, excess length of stay (LOS) and excess costs were obtained by conducting a literature review searched in order of applicability to the English setting (Supplementary Figure A1), beginning with national SSISS reports, and ending with single-site studies from non-OECD countries. Where research papers were the data source, only observational studies for the purpose of surveillance were included. Interventional studies and studies primarily analysing risk factors were excluded. Unpublished data on SSI rates that included full post-discharge surveillance (PDS, including community and outpatient surveillance) were provided by PHE for the categories included in the SSISS.
Estimations of parameters
Where the rates of SSI including full PDS were unavailable from the SSISS, they were estimated by scaling up the best available inpatient and readmission SSI rates in line with the proportion of infections identified by PDS reported in the literature. For example, if one study in England reported inpatient and readmission SSI rates of 3% in a category but no rate including PDS, and a French study reported an inpatient and readmission SSI rate of 8%, and 17% including PDS, we calculated ((17–8)/17)*100 = 52.9% of SSIs to be diagnosed through additional PDS, so the equivalent rate in England would be 3*((100 + 52.9)/100) = 4.59%. The annual number of SSIs per category was estimated by multiplying the annual volume of procedures by the appropriate SSI rate.
Estimation of excess costs
Papers reporting excess costs per SSI were quality-assessed using the Newcastle–Ottawa Scale [
]. Costs were inflated to 2014–15 using the Hospital and Community Health Services pay and price inflation indices, and converted to pounds sterling where necessary using the 2014 average exchange rates from UK Forex [
]. The excess cost of SSI to hospitals annually was calculated by multiplying the number of infections among inpatients and readmissions by the inflated and converted mean cost per SSI.
Data analysis
A descriptive analysis was performed by dividing data for each factor into quartiles and assigning a number Q1 to Q4. Radar charts were constructed to explore relationships between the current surveillance arrangements, number, and cost of SSIs in each category. A narrative description of the patterns based on visual inspection was performed by two authors working independently (R.T. and G.B.), and disagreements were resolved by a third person (A.H.). Surgeries were divided by minimum wound classification (clean vs clean-contaminated). To investigate whether the choice of category for surveillance was related to any of the factors (e.g. SSI rate, perceived priority), analysis was carried out using Spearman's correlation test for continuous variables and the Kruskall–Wallis test for categorical variables. P ≤ 0.05 was considered statistically significant.
Results
The 23 surgical categories analysed in the study represented a volume of 2.01 million procedures annually in England. The SSI rate (inpatient and readmission, includes superficial, deep, and organ/space) for all procedures varied from 0.4% (95% confidence interval (CI) not available) in pacemaker surgery to 10.4% (95% CI: 9.9–10.8) for large bowel surgery (Table I) [
Infection after pacemaker implantation: infection rates and risk factors associated with infection in a population-based cohort study of 46299 consecutive patients.
]. From all evidence sources (Supplementary Table A2), it was estimated that when additional post-discharge infections (community and outpatient detection) were included, surgery in England results in 100,965 SSIs annually.
Table IFactors associated with the risk, number and cost of SSIs in England by surgical category along with average NHS trust-reported priority ranking and current surveillance arrangements
Infection after pacemaker implantation: infection rates and risk factors associated with infection in a population-based cohort study of 46299 consecutive patients.
Incidence, treatment intensity, and incremental annual expenditures for patients experiencing a cardiac implantable electronic device infection: evidence from a large US payer database 1-year post implantation.
Table IFactors associated with the risk, number and cost of SSIs in England by surgical category along with average NHS trust-reported priority ranking and current surveillance arrangements
Infection after pacemaker implantation: infection rates and risk factors associated with infection in a population-based cohort study of 46299 consecutive patients.
Incidence, treatment intensity, and incremental annual expenditures for patients experiencing a cardiac implantable electronic device infection: evidence from a large US payer database 1-year post implantation.
From the published literature, the excess LOS due to SSI was estimated to range from 1 (95% CI: –3 to 17) day for SSI occurring after cranial surgery and up to 29 days (95% CI not available) for SSI following gastric surgery [
]. Across all surgery types, excess LOS due to SSIs equates to an estimated 501,490 extra bed-days annually. Similarly, excess costs ranged from £1315 (95% CI not available) per SSI in abdominal hysterectomy to £30,171 (95% CI: 26,434–33,583) in pacemaker surgery (2014/2015) [
Incidence, treatment intensity, and incremental annual expenditures for patients experiencing a cardiac implantable electronic device infection: evidence from a large US payer database 1-year post implantation.
]. SSIs across all surgery types were estimated to cost hospitals an extra £232,866,861 annually.
Half of the 84 respondents to the survey (50%) reported hip replacement as a top future priority, followed by CABG, coronary artery bypass graft (CABG) (47%) and caesarean section (46%) [
]. Cholecystectomy, gastric, reduction of femoral neck fracture, limb amputation and bile duct, liver, pancreatic surgeries were classified as lower future priorities.
Visual analysis of radar charts of the surgical categories and assessment factors revealed mismatches in several surgical categories. Large bowel surgery and cholecystectomy had an annual number of SSIs and estimated costs in the highest quartile (Q4), whereas surveillance is voluntary (Q3) (Figure 1). Caesarean section had an annual number of SSIs in Q4, costs in Q3, and priority ranking in Q4, but is not included in the national surveillance programme. Conversely, knee replacement has SSI numbers in Q2 and costs in Q3, but the percentage of hospitals conducting surveillance and future priorities in Q4, and reporting is mandatory.
Figure 1Radar charts comparing surgical categories and quartiles in which the major factors fall relative to other categories. Shown are the four mandatory surgical site infection (SSI) surveillance categories in England and four categories in which surveillance does not match the relative burden or cost of SSIs. Chosen by independent visual analysis by two researchers (R.T. and G.B.). SSIs p.a. is overall estimated no. of SSIs per annum; current surveillance: 1, not offered; 3, voluntary surveillance; 4, mandatory surveillance; future priority based on ranking in Public Health England survey.
The survey also asked respondents whether they were currently undertaking surveillance in a given surgical category. The percentage of hospitals undertaking surveillance in a category was not associated with the excess cost to hospitals, or the number of SSIs (Table IIa). Future priorities did not correlate with annual volume of procedures, current estimated no. of SSIs, or the excess cost to hospital, but was associated with surveillance already performed (r = 0.76, P < 0.01) and mandatory surveillance (P = 0.04) (Table IIb).
Table IIaContingency table assessing factors explaining the surveillance method and the perception of priorities
Hospitals already undertaking surveillance
P-value
Procedures selected as future priority
P-value
Procedures ranked as top priority
P-value
Missing values
Economic impact (r)
Excess LOS
0.09
0.69
0.03
0.89
0.15
0.53
4
Estimated excess cost to hospitals annually
−0.22
0.33
0.01
0.95
−0.06
0.81
3
Medical and societal impact (r)
Estimated no. infections + PDS
0.20
0.34
−0.13
0.54
0.19
0.38
0
Endogenous factors (r)
Median age (years)
0.36
0.09
−0.35
0.09
0.38
0.07
0
Exogenous factors
Clean, median (Q1–Q3)
29 (22–59)
0.04
3.9 (3.1–4.3)
0.08
18 (9–26)
0.05
0
Clean-contaminated, median (Q1–Q3)
16 (9–22)
4.5 (4–4.8)
7 (2–9)
Factors impacting the surveillance method
Annual patient volume
0.13
0.54
−0.35
0.09
0.26
0.22
0
% SSIs detected by PDS
0.16
0.47
−0.26
0.23
0.26
0.24
1
National surveillance requirement
Voluntary, median (Q1–Q3)
25 (17–33)
<0.01
4.3 (3.9–4.8)
0.2
9 (7–18)
0.10
0
Mandatory, median (Q1–Q3)
74 (40.5–90.5)
3.4 (2.9–4.2)
33 (17.5–45)
Not offered, median (Q1–Q3)
8.5 (7–10)
4 (3.1–4.4)
4 (0–26)
Hospitals already undertaking surveillance
–
–
−0.57
<0.01
0.76
<0.01
0
Procedure considered as future priority
−0.57
<0.01
–
–
−0.79
<0.01
0
Procedure ranked as a top priority
0.76
<0.01
−0.79
<0.01
–
–
0
r, Spearman's correlation coefficient; LOS, length of stay; PDS, post-discharge surveillance; SSI, surgical site infection; Q, quartile.
By mapping the current available data on SSI characteristics and surveillance requirements, mismatches have been revealed between surgical categories under surveillance, the economic and disease burden represented by infections, and the identification of future priorities. In general, the risk and number of SSIs, and their excess costs seemed not to influence hospitals in the design of their surveillance strategy or their perception of priorities.
Trusts in England perceived orthopaedic surgery, CABG, and caesarean section as top priorities for SSI surveillance strategies [
]. Orthopaedic surgery has historically been included in surveillance systems as a surgical quality indicator, but a recent Australian study concluded that SSI rates are procedure specific rather than hospital specific, challenging the rationale behind this indicator [
The measurement and monitoring of safety: drawing together academic evidence and practical experience to produce a framework for safety measurement and monitoring.
]. Whereas it is important to continue surveillance in orthopaedic surgery in order to maintain these gains, there is little room for further improvement [
]. More recently, the Getting It Right First Time (GIRFT) initiative found that only four of the 50 hospitals were able to report SSI rates for general surgery [
]. Consequently, SSIs were targeted by a national audit aiming to tackle unwarranted variation in care quality and procurement, which may generate the data and inertia needed for hospitals to strategically expand surveillance into categories not currently included in national surveillance [
]. The fact that many hospitals are already undertaking in-house surveillance in categories not included in the SSISS is an encouraging first step.
The present study suggests there are opportunities to use SSI surveillance to drive improvements in patient outcomes, costs, and efficiency by targeting interventions where they can have the highest impact. A report by Lord Carter focused on savings that could be made by reducing infection rates in orthopaedic surgery [
]. However, the present study shows that SSIs in the four mandatory orthopaedic categories account for only 9% of estimated excess cost to hospitals annually, seven times less than large bowel surgery. Likewise, interventions in categories with a high associated LOS such as gastric surgery (29 days) could create more bed spaces, improve hospital efficiency and avoid patient exposure to other HCAIs [
]. Future studies may examine how these factors should be weighted, and similar studies could be replicated on a local, national, or inter-national scale.
Despite proven clinical value, surveillance methods are labour-intensive, and, due to financial constraints and staff shortages, surveillance must be prioritized carefully [
]. Automating some aspects of surveillance could help to reduce costs and subjectivity in diagnosis. Several automated or semi-automated surveillance systems have been developed and innovative systems of PDS are also being developed, including an electronic post-discharge questionnaire for the SSISS, which are especially important in surgical categories with a short postoperative length of stay [
Matching bacteriological and medico-administrative databases is efficient for a computer-enhanced surveillance of surgical site infections: retrospective analysis of 4,400 surgical procedures in a French university hospital.
]. There is growing evidence to support the utility of these systems for SSI surveillance, and growing guidance to support their design and implementation [
To our knowledge, this is the first paper to assess the fit between current and future priorities of a national SSI surveillance system and the important factors which impact patients and the health system. The strengths of this paper are its scope in including multiple surgical categories and factors, its selection of papers based on their applicability to the setting, and its use of visual and statistical analysis to enable both category-level and overall assessment of agreement.
However, there are some important limitations to this study. Costs presented were estimated using several different methodologies, some of which focus purely on costing additional LOS, potentially limiting cost comparability and applicability, though the papers included were all judged to be of a high quality using the Newcastle–Ottawa checklist (data not shown) [
]. Often the costliest facet of SSI treatment for orthopaedics is revision surgery, which is not accounted for if only additional LOS is considered. To assess the extent of this omission, a separate analysis – using the cost of revision surgery due to SSIs in hip and knee replacement, the number of joint replacements and proportion of these that were revisions due to infection – calculated excess costs due to SSIs of at least £19,844,824 and £23,230,805, respectively [
]. However, these concerns are common to all surveillance systems, particularly those that rely on manual data collection, and SSISS reports still represent the best data available for estimating SSI rates in England [
]. Not all studies were performed in England; unit costs and patient treatment pathways may vary greatly between countries, reducing external validity of such estimates [
]. Unfortunately, it was not possible to include confidence intervals or ranges in the univariate statistical analysis performed, due to lack of data availability. The lack of data on morbidity and mortality associated with SSI, and the extent to which SSIs are preventable in different categories precluded the inclusion of these factors despite their obvious importance. We recommend further studies on morbidity associated with SSI in different categories using quality-adjusted life-years, potentially through data linkage methodologies to reduce the data collection burden on the NHS.
In conclusion, this study suggests that current SSI surveillance and future hospital priorities are not targeting surgical categories with the highest burden in terms of risk, number of SSIs, and cost. Systematic priority setting could provide benefits both to patients and health systems. The methodology used in this study could be used in settings of any size when designing or redesigning surveillance strategies.
Acknowledgements
We would like to thank contributors to the SSISS, and T. Lamagni, C. Wloch, S. Elgohari, and P. Harrington at the SSISS for the provision of data and their comments on the manuscript.
Conflict of interest statement
None declared.
Funding sources
This research was partially funded by the National Institute for Health Research (NIHR) Health Protection Research Unit in Healthcare-Associated Infection and Antimicrobial Resistance at Imperial College London (grant number HPRU-2012-10047), in partnership with Public Health England (PHE), and the NIHR Imperial Patient Safety Translational Research Centre. The views expressed are those of the authors and not necessarily those of the National Health Service (NHS), the NIHR, the Department of Health, or PHE. The authors also acknowledge the UK Clinical Research Collaboration Centre for Infection Prevention and Management, Imperial College Healthcare NHS Trust, and the NIHR Imperial Biomedical Research Centre.
Appendix A. Supplementary data
The following is the supplementary data related to this article:
Surgical site infection after gastrointestinal surgery in high-income, middle-income, and low-income countries: a prospective, international, multicentre cohort study.
Burden of six healthcare-associated infections on European population health: estimating incidence-based disability-adjusted life years through a population prevalence-based modelling study.
Infection after pacemaker implantation: infection rates and risk factors associated with infection in a population-based cohort study of 46299 consecutive patients.
Incidence, treatment intensity, and incremental annual expenditures for patients experiencing a cardiac implantable electronic device infection: evidence from a large US payer database 1-year post implantation.
The measurement and monitoring of safety: drawing together academic evidence and practical experience to produce a framework for safety measurement and monitoring.
Matching bacteriological and medico-administrative databases is efficient for a computer-enhanced surveillance of surgical site infections: retrospective analysis of 4,400 surgical procedures in a French university hospital.
30320-7&id=gr1.jpg){kind=link}