TY - JOUR
AU - Struwe, J
AB - Abstract Background Systematic surveillance of surgical-site infections is not standard. The aim of this retrospective cohort study was to evaluate the feasibility of using existing national health registers for surveillance of postoperative antibiotic treatment suggestive of surgical-site infection. Methods Data from national registers on hospital admissions and drug use were combined. Antibiotic purchases by 8856 patients subject to ambulatory care for inguinal hernia repair in Sweden during 2006 were ascertained during a 30-day interval immediately after surgery (postsurgical period) and in an 11-month control period (6 months before and 5 months after the postsurgical period). Results The incidence of first purchases of skin and soft tissue antibiotics was 245 per 8697 person-months in the first postoperative month and 180 per 52 612 person-months in the preoperative control period, representing a 1-month risk difference of 2·4 (95 per cent confidence interval (c.i.) 2·0 to 2·7) per cent. Hence, a 1-month risk of 2·4 per cent could be attributed tentatively to the surgery. The rate of episodes with antibiotics used mainly for skin and soft tissue infection was sevenfold higher in the first postoperative month than in the control period (rate ratio 7·01, 95 per cent c.i. 5·94 to 8·27). Conclusion The risk of antibiotic treatment during the postsurgical period was of the same order of magnitude as infection rates reported in the Swedish Hernia Register and review studies. Surveillance of postoperative antibiotic use may be considered as a resource-saving surrogate marker for surgical-site infections or an indicator of inappropriate use. Introduction Surgical-site infections (SSIs) cause unwanted morbidity and even increase the risk of death. They have considerable financial consequences for the health system1–5. Despite this burden, most care providers have no integrated systematic registration of SSIs6,7. Obstacles to registering are many, and include the resource-intensive logistics of identifying cases manually, particularly patients with onset of infection after discharge from hospital8–10, the need for trained staff11–16, and lack of uniform criteria to allow comparisons7,17. Lack of clinician interest, poor integration of surveillance in electronic information systems and low priority among healthcare providers may also be factors contributing to poor surveillance of SSIs. The validity of electronic surveillance has been debated9. Notably, many of the existing large national nosocomial infection surveillance systems, such as the Nosocomial Infections Surveillance System in the USA, National Surveillance of Infections in Hospitals in Belgium, the Finnish Hospital Infection Program, Krankenhaus Infektions Surveillance System in Germany and the Nosocomial Infection National Surveillance Service in England, or international systems such as the European Hospitals in Europe Link for Infection Control through Surveillance within the Improving Patient Safety in Europe project, require trained staff and a substantial amount of manual work during both case identification and registration11–16. Electronic surveillance of postoperative exposure to antibiotics is one of the alternatives to manual infection surveillance, as investigated previously9. Data on postoperative antibiotic exposure have been collected from hospital databases, medical charts and health plan records7,18–23. These studies all suggested that antibiotic exposure can be a good indicator of postoperative infection, yet antibiotic exposure is not a widely used surrogate marker in this regard. Furthermore, Leveille and colleagues24 used automated pharmacy records to identify antibiotic-treated infections among postmenopausal women belonging to the same healthcare system, and came to the conclusion that this was a good alternative to costly medical record reviews for use in epidemiological research. However, despite these findings, the feasibility of register-based surveillance of postoperative purchases of antibiotics at the national level as a surrogate marker of possible postoperative infections has not been investigated. The main objective of this study was to investigate whether combining data on individuals identified in the Swedish Hospital Patient Register with data for the same individuals from the Swedish Prescribed Drug Register could be used for surveillance of postoperative antibiotic treatment suggestive of related infections. Methods In this register-based retrospective cohort study, the risk of skin and soft tissue antibiotic use attributed to ambulatory hernia surgery was estimated by comparing patients' purchases of such antibiotics during the first 30 days after surgery (postsurgical period) with their purchases during an 11-month control period (6 months before and 5 months after the postsurgical period). The risk difference, that is the risk reasonably attributable to the surgical procedure, was then compared with available data on SSIs from relevant clinical registrations. The project was granted ethical approval by the Regional Ethics Committee of Stockholm in May 2008. Outpatient hernia surgery and antibiotic prescription in Sweden Swedish hospital-based health services are almost exclusively public, financed by taxes, and run or commissioned by the county councils. A large part (more than 60 per cent) of all hernia surgery is done on an outpatient basis, typically without scheduled return visits to the operating surgeon. Currently, there are no official Swedish guidelines for antibiotic use in connection with hernia surgery, either as prophylaxis or given therapeutically to treat SSI. Although most surgeons would probably refute the idea of routine prophylaxis in hernia surgery, the Swedish Hernia Register (SHR), which covers approximately 95 per cent of all inguinal hernia operations in patients aged 15 years or more in Sweden, noted that antibiotic prophylaxis was given to 21 per cent of all male patients undergoing hernia repair in 200625. Registry and data description The Hospital Patient Register and the Prescribed Drug Register, both held by the Swedish National Board of Health and Welfare, served as data sources. Data on migration were obtained from the Migration Register at Statistics Sweden and data on deaths from population registries at the National Tax Board. Data in both the patient and drug registers contain the National Registration Number (NRN), a unique personal identifier assigned to all Swedish residents almost immediately after birth or upon immigration. Because the NRNs are used as identifiers in all Swedish health registers, precise linkage of individual records is possible. The Hospital Patient Register provides information on all inpatient and ambulatory care at Swedish hospitals, notably hospital, date, surgical codes according to the Nordic NOMESCO Classification of Surgical Procedures and diagnostic codes according to the International Classification of Diseases (ICD), 10th revision. Under-reporting to the register as a whole is estimated to be 1–2 per cent, the NRN is missing or incorrect in another 1 per cent (mainly children and expatriates), and a main diagnosis is missing for approximately 1 per cent26. The Prescribed Drug Register contains information on all dispensed prescriptions in ambulatory care to the whole population of Sweden. Among other things, it contains data on the prescribed item, amount, dose, the patient's age, sex and place of residence, and date of prescribing and dispensing27. Drugs administered in hospitals, for instance perioperative prophylactic antibiotics, are not recorded in this register. Data on operated patients' purchases of antibiotics were extracted, covering the period from 6 months before until 6 months after operation. The Anatomical Therapeutic Chemical (ATC, version 2006) classification was used (http://www.whocc.no). Definition of antibiotic prescription groups Antibiotics were grouped into three therapeutic categories: antibiotics used mainly for skin and soft tissue infections (β-lactamase-resistant penicillins and lincosamides; ATC codes J01CF and J01FF); antibiotics used mainly for urinary tract infections (pivmecillinam, trimethoprim and trimethoprim with sulphonamides, fluoroquinolones and nitrofurantoin; ATC codes J01CA08, J01E, J01MA and J01XE); and other antibiotics for systemic use (ATC code J01), excluding those mentioned above and methenamine (ATC code J01XX05). Those described as ‘other antibiotics’ are used mainly for respiratory tract infections in Sweden. After record linkages had been made, the data were anonymized. Study population All male patients aged at least 15 years who had open inguinal hernia repair by means of the prevailing method of the time, surgical procedure code JAB30 using prosthetic material, in Swedish ambulatory care during 2006 were identified in the Hospital Patient Register. Those for whom more than one hernia repair was recorded in 2006 most probably had sequential repair of a bilateral hernia. For these patients, data up to the day before the second operation were included in the analysis. This was done to avoid the influence of the second operation on the follow-up of the first, but also to preclude dependency between observations that would arise if the same individual contributed more than one observation. To be consistent, patients who had undergone hernia repair in 2005 were excluded as they were deemed to have undergone sequential repair of bilateral hernias, the operation in 2006 not being their first. After identification of patients who fulfilled the criteria for inclusion, their NRNs were used as a key to collect data from the Prescribed Drug Register. Study period A study flow chart is presented in Fig. 1. In this study, a month was defined as a 30-day interval. For each patient, the postsurgical period was defined as the first month after surgery. The control period was the 11-month period starting 6 months before surgery and ending 6 months after surgery, excluding the postsurgical period. Reasons for censoring were emigration, death or a second operation. Figure 1 Open in new tabDownload slide Study flow chart. The total postsurgical period was 8845 months and the total control period 97 043 months. *Reasons for censoring were emigration, death or a second operation Statistical analysis The number of first purchases of antibiotics per person-day was calculated for the postsurgical period, the preoperative control period and the postoperative control period. The person-time contributed by each patient ended at his first antibiotic purchase or at the time of censoring for reasons noted above. The cumulative incidence of antibiotic purchases was used as an estimate of average risk. The difference in 1-month risk of antibiotic use was then calculated, by comparing the 30-day postsurgical period with the 6-month preoperative or 5-month postoperative control period, as a measure of the risk tentatively attributable to the surgical procedure. The analysis was repeated without censoring at the time of first antibiotic purchase. However, because early changes of antibiotic regimens based on resistance pattern or owing to apparent treatment failure after a standard course would then be counted as new events, ‘antibiotic episodes’ were counted instead. Purchases of antibiotics within the same therapeutic group and within the same 14-day interval were considered to relate to the same episode. The rate in the postsurgical period was compared with that in the whole 11-month control period using the rate ratio as a measure of relative risk. Episodes extending into both postsurgical and control periods were attributed to the period in which they started. To calculate risk differences and rate ratios repeated-measures Poisson regression modelling was applied with the generalized estimating equations method. These models were fitted into the framework of generalized linear models using a log link to estimate rate ratios and an identity link to estimate risk differences. This method takes into account the correlation between antibiotic use across the three time intervals within individuals, and is not sensitive to missing periods at repeated measurements. Results After exclusion of 282 men who had a previous hernia repair in 2005, 8856 patients underwent the selected procedure in 2006 and were included in the study cohort. The total person-time in the postsurgical and control periods was 8845 and 97 043 person-months respectively. The age of the patients included in the study ranged from 16 to 96 (median 60, interquartile range 50–68) years. During the study period, the follow-up of 110 patients was censored owing to death or emigration (27) and second inguinal hernia repair (83). Fig. 2 shows the cumulative incidence of antibiotic purchases, month by month, in the control and postsurgical periods. The differences in risk of receiving antibiotics were quite similar when comparing the cumulative incidence in the postsurgical period with the estimated 1-month risk in either the preoperative or postoperative control period (Table 1). The excess risk difference with regard to antibiotics for skin and soft tissue infections and for urinary tract infections in the first postsurgical month was small, but statistically significant. For ‘other antibiotics’ the risk differences were not significant, regardless of comparison period. Figure 2 Open in new tabDownload slide Estimated 1-month cumulative incidence (‘risk’) of purchase of antibiotics, month by month, in the 6 months before surgery until 6 months after operation. The risk period for surgery-related antibiotic treatment (the first month after surgery; postsurgical period) is shown in a darker shade; 30-day intervals during the control period (6 months before and 5 months after the postsurgical period) are also shown. See Methods for definitions of groups of antibiotics Table 1 Differences in 1-month risk of receiving antibiotics between the 1-month postsurgical period and each control period (6 months before operation and 5 months after the postsurgical period) . Antibiotic purchases per person-month . Risk difference (%)* . Group of antibiotics† . Preop. control . Postsurgical . Postop. control . Postsurgical versus preop. control . Postsurgical versus postop. control . Antibiotics for skin and soft tissue infections 180 per 52 612 245 per 8697 175 per 43 389 2·4 (2·0, 2·7) 2·3 (2·0, 2·7) Antibiotics for urinary tract infections 349 per 51 973 97 per 8793 244 per 43 239 0·3 (0·1, 0·5) 0·4 (0·2, 0·7) Other antibiotics 744 per 50 812 139 per 8779 668 per 42 210 − 0·1 (−0·4, 0·2) − 0·2 (−0·5, 0·1) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 1176 per 49 377 474 per 8581 1001 per 41 209 2·4 (1·8, 2·9) 2·4 (1·8, 2·9) . Antibiotic purchases per person-month . Risk difference (%)* . Group of antibiotics† . Preop. control . Postsurgical . Postop. control . Postsurgical versus preop. control . Postsurgical versus postop. control . Antibiotics for skin and soft tissue infections 180 per 52 612 245 per 8697 175 per 43 389 2·4 (2·0, 2·7) 2·3 (2·0, 2·7) Antibiotics for urinary tract infections 349 per 51 973 97 per 8793 244 per 43 239 0·3 (0·1, 0·5) 0·4 (0·2, 0·7) Other antibiotics 744 per 50 812 139 per 8779 668 per 42 210 − 0·1 (−0·4, 0·2) − 0·2 (−0·5, 0·1) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 1176 per 49 377 474 per 8581 1001 per 41 209 2·4 (1·8, 2·9) 2·4 (1·8, 2·9) * Values in parentheses are 95 per cent confidence intervals. † See Methods for definitions of groups of antibiotics. Risk estimates are based on cumulative incidence rates. The risk differences indicate absolute 1-month risks tentatively attributable to the surgical procedure. Censoring occurred at the time of first purchase. ATC, Anatomical Therapeutic Chemical. Open in new tab Table 1 Differences in 1-month risk of receiving antibiotics between the 1-month postsurgical period and each control period (6 months before operation and 5 months after the postsurgical period) . Antibiotic purchases per person-month . Risk difference (%)* . Group of antibiotics† . Preop. control . Postsurgical . Postop. control . Postsurgical versus preop. control . Postsurgical versus postop. control . Antibiotics for skin and soft tissue infections 180 per 52 612 245 per 8697 175 per 43 389 2·4 (2·0, 2·7) 2·3 (2·0, 2·7) Antibiotics for urinary tract infections 349 per 51 973 97 per 8793 244 per 43 239 0·3 (0·1, 0·5) 0·4 (0·2, 0·7) Other antibiotics 744 per 50 812 139 per 8779 668 per 42 210 − 0·1 (−0·4, 0·2) − 0·2 (−0·5, 0·1) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 1176 per 49 377 474 per 8581 1001 per 41 209 2·4 (1·8, 2·9) 2·4 (1·8, 2·9) . Antibiotic purchases per person-month . Risk difference (%)* . Group of antibiotics† . Preop. control . Postsurgical . Postop. control . Postsurgical versus preop. control . Postsurgical versus postop. control . Antibiotics for skin and soft tissue infections 180 per 52 612 245 per 8697 175 per 43 389 2·4 (2·0, 2·7) 2·3 (2·0, 2·7) Antibiotics for urinary tract infections 349 per 51 973 97 per 8793 244 per 43 239 0·3 (0·1, 0·5) 0·4 (0·2, 0·7) Other antibiotics 744 per 50 812 139 per 8779 668 per 42 210 − 0·1 (−0·4, 0·2) − 0·2 (−0·5, 0·1) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 1176 per 49 377 474 per 8581 1001 per 41 209 2·4 (1·8, 2·9) 2·4 (1·8, 2·9) * Values in parentheses are 95 per cent confidence intervals. † See Methods for definitions of groups of antibiotics. Risk estimates are based on cumulative incidence rates. The risk differences indicate absolute 1-month risks tentatively attributable to the surgical procedure. Censoring occurred at the time of first purchase. ATC, Anatomical Therapeutic Chemical. Open in new tab Rate ratios of ‘antibiotic episodes’, comparing the rate in the first month after surgery with that in the entire 11-month control period (before and after the postsurgical period), are presented in Table 2. The rate of episodes with antibiotics used mainly for skin and soft tissue infection was sevenfold higher in the first postoperative month than in the control period (rate ratio 7·01, 95 per cent confidence interval 5·94 to 8·27), whereas the excess rate of episodes with antibiotics for urinary tract infection was only slightly increased (rate ratio 1·56, 1·28 to 1·90). No increase in the rate of episodes was found with ‘other antibiotics’; the overall excess of any antibiotic was approximately twofold. Table 2 Rate ratios: episodes of antibiotic treatment per person-month during the postsurgical period compared with the control period . Antibiotic episodes per person-month . . Group of antibiotics† . Postsurgical . Control (preop. + postop.) . Rate ratio* . Antibiotics for skin and soft tissue infections 255 per 8845 399 per 97 043 7·01 (5·94, 8·27) Antibiotics for urinary tract infections 104 per 8845 731 per 97 043 1·56 (1·28, 1·90) Other antibiotics 140 per 8845 1629 per 97 043 0·94 (0·80, 1·12) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 499 per 8845 2758 per 97 043 1·96 (1·78, 2·15) . Antibiotic episodes per person-month . . Group of antibiotics† . Postsurgical . Control (preop. + postop.) . Rate ratio* . Antibiotics for skin and soft tissue infections 255 per 8845 399 per 97 043 7·01 (5·94, 8·27) Antibiotics for urinary tract infections 104 per 8845 731 per 97 043 1·56 (1·28, 1·90) Other antibiotics 140 per 8845 1629 per 97 043 0·94 (0·80, 1·12) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 499 per 8845 2758 per 97 043 1·96 (1·78, 2·15) * Values in parentheses are 95 per cent confidence intervals. † See Methods for definitions of groups of antibiotics. Purchases of antibiotics within the same therapeutic group and within the same 14-day period were considered to belong to the same episode. No censoring took place at the time of antibiotic purchase. ATC, Anatomical Therapeutic Chemical. Open in new tab Table 2 Rate ratios: episodes of antibiotic treatment per person-month during the postsurgical period compared with the control period . Antibiotic episodes per person-month . . Group of antibiotics† . Postsurgical . Control (preop. + postop.) . Rate ratio* . Antibiotics for skin and soft tissue infections 255 per 8845 399 per 97 043 7·01 (5·94, 8·27) Antibiotics for urinary tract infections 104 per 8845 731 per 97 043 1·56 (1·28, 1·90) Other antibiotics 140 per 8845 1629 per 97 043 0·94 (0·80, 1·12) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 499 per 8845 2758 per 97 043 1·96 (1·78, 2·15) . Antibiotic episodes per person-month . . Group of antibiotics† . Postsurgical . Control (preop. + postop.) . Rate ratio* . Antibiotics for skin and soft tissue infections 255 per 8845 399 per 97 043 7·01 (5·94, 8·27) Antibiotics for urinary tract infections 104 per 8845 731 per 97 043 1·56 (1·28, 1·90) Other antibiotics 140 per 8845 1629 per 97 043 0·94 (0·80, 1·12) Any antibiotic for systemic use, excluding methenamine (ATC code J01, excluding J01XX05) 499 per 8845 2758 per 97 043 1·96 (1·78, 2·15) * Values in parentheses are 95 per cent confidence intervals. † See Methods for definitions of groups of antibiotics. Purchases of antibiotics within the same therapeutic group and within the same 14-day period were considered to belong to the same episode. No censoring took place at the time of antibiotic purchase. ATC, Anatomical Therapeutic Chemical. Open in new tab Discussion Although SSIs are common and costly, systematic surveillance of infections is not standard in most institutions. As computerized medical records are becoming increasingly common, register-based surveillance is feasible and could be particularly useful in the continuous monitoring of trends over time in large single institutions, or regionally or nationally. When a signal arises in such a monitoring system, further in-depth studies can establish whether the change is due to increases in SSIs, a greater inclination among doctors to use antibiotics, administrative or organizational changes, or any combination of these factors. The present study demonstrated the feasibility of using nationwide and population-based record linkages of population and healthcare registers to estimate rates of SSI after inguinal hernia surgery. This study supports the findings from previous studies7,18–23 that antibiotic exposure can serve as a marker for postoperative infection. Compared with traditional surveillance systems, automated surveillance programmes have been found to reduce time consumption by up to 61 per cent6. In addition, this method is well suited for postdischarge surveillance of SSIs, which may be difficult with traditional surveillance systems. The trend towards reduced length of hospital stay after operation and the increase in day-case surgery means that SSIs will increasingly occur after hospital discharge1. The rate of outpatient treatment episodes with antibiotics suited mainly for skin and soft tissue infection was increased sevenfold in the first month after inguinal hernia repair, compared with the rate in the periods before and after this. The true difference in 1-month risk of being prescribed antibiotics used mainly for skin and soft tissue infections, that is the absolute 1-month risk attributed to the surgical procedure, was estimated at between 2·0 and 2·7 per cent. In addition, a 1-month risk, most likely amounting to 0·1–0·7 per cent, of being prescribed antibiotics used mainly for urinary tract infections could also be attributed tentatively to the operation. Whether these findings truly reflect the incidence of SSIs and other surgery-related infections, or just represent overprescribing of antibiotics, cannot be confirmed by the present data. Compared with other clinical sources of information, the present 1-month attributable risk of receiving antibiotics used mainly for skin and soft tissue infections seemed to coincide reasonably with reported risks of postoperative infection after hernia surgery. According to the SHR, the frequency of postoperative infection among men having outpatient inguinal hernia repair using prosthetic material (surgical procedure code JAB30), and not also operated on in 2005, was 1·4 per cent in 2006 (B. Häggqvist, personal communication of data from the SHR). However, as there is no organized active follow-up in the SHR, important underascertainment has been noted28. In a previous study, patients undergoing any kind of inguinal hernia repair and registered in the SHR were asked to self-report complications via a postal questionnaire28. Notably, the reported overall complication rate was more than four times higher than the rate recorded in the SHR (23·8 versus 5·2 per cent). Infections were reported by 7·3 per cent of the questionnaire respondents, of whom more than half (4·5 per cent) asserted that the infection had led to healthcare visits. On the other hand, in a carefully conducted follow-up study at 32 Scottish hospitals, with repeated telephone interviews during the first month after surgery and visual inspection of the wounds in all patients with a suspected infection, 29 per cent of patients with self-reporting infections were deemed to be uninfected29. When extrapolating the false-positive rate to the SHR questionnaire rate of perceived infections leading to healthcare visits (which would be necessary to get an antibiotic prescription), the proportion with a true infection seeing a doctor would be slightly more than 3 per cent, which is in good accordance with the present data. However, infections also occur in patients who do not contact healthcare providers. In the Scottish study, the 1-month risk of a confirmed SSI after any type of hernia repair was 5·3 per cent29. However, in a Cochrane review of 12 clinical trials that evaluated antibiotic prophylaxis for hernia repair, the frequency of postoperative infection in the subgroup undergoing repair of inguinal hernia using prosthetic material was 1·4 and 3·0 per cent with and without antibiotic prophylaxis respectively30. Thus an exact standard value against which the present data could be compared cannot be derived confidently from the existing literature. Of importance for interpretation of the present results is that a perfect concordance between incidence of SSIs and prescription of antibiotics is neither expected nor desirable. Management of uncomplicated SSIs may not necessarily include antibiotic treatment. Furthermore, doctors' inclination to use antibiotics may vary with the type of operation and surgical or medical specialty. Therefore, the rate of antibiotic prescription should be lower than that of SSI. The threshold for antibiotic treatment is likely to be lower when the operation involves implantation of foreign prosthetic material, as in the present case. It is also conceivable that non-surgeons faced with a SSI are more inclined to resort to antibiotic treatment than surgeons. Consequently, the performance of this register-based surveillance scheme might differ with the character of the surgical procedure and perhaps also with the healthcare setting. Validation studies addressing these differences are warranted. Reasonable validity of the results in this study is assured through the use of nationwide registers covering essentially all hospital-based healthcare and all ambulatory care drug prescriptions with means of attaining precise record linkages. The quality of registration of surgical operations was evaluated in 198631. The surgical code was incorrect in 2 per cent and missing in 5 per cent of the records. Although new classification of surgical procedures has been introduced more recently, the authors believe that misclassification of the surgical procedures is unlikely to have distorted the present results. The accuracy of the data in the Prescribed Drug Register is dependent on the correct input of data by the pharmacies, when dispensing prescriptions. Lack of data caused by incomplete NRNs has been noted in less than 1 per cent32. The register includes only data on prescriptions that have been dispensed. Studies have shown that the proportion of men choosing not to purchase drugs they had been prescribed was approximately 2·5 per cent in 200533. In the case of hernia surgery, the care episode is almost invariably so short that no infections are expected to become clinically manifest before discharge. In addition, the present study included only operations performed in ambulatory care. Therefore, in-hospital antibiotic treatment (which will not be captured in the Prescribed Drug Register) is an unlikely source of error. However, such treatment will affect risk estimates for major procedures requiring several days of in-hospital care after surgery. Surveillance using this register-based approach is inexpensive. Using data from the Swedish registers, it can be performed at any desired level, from the operating clinic to county or national level, thus allowing bench-marking. The method can be assumed to be less time consuming and less costly than existing manual methods of surveillance of postoperative infection rates. Thus, even if surveillance of postoperative antibiotic use requires linkage between documentation of surgical procedures or discharge diagnoses and prescriptions, this does not necessarily depend on national registers and it should be considered as a function in information technology-based medical records, when permitted. Because the translation of observed attributable risks for antibiotic treatment to risks for SSIs may require establishment of correction factors (after adequate validation studies) that are specific for both the surgical procedure and the setting, the present method may be less suited for comparisons between different surgical methods, or comparisons between healthcare providers in different settings. Acknowledgements Dr Christer Norman, Professor Otto Cars and Professor Cecilia Stålsby Lundborg all contributed technical advice and valuable discussion in the early planning of this study. The study was supported financially by Strama—the Swedish Strategic Programme Against Antibiotic Resistance, the Swedish Patient Insurance, and by grants from the Scandinavian Society for Antimicrobial Chemotherapy and the Karolinska Institute. The authors' work was independent of the funders, who had no role in the study design; collection, analysis and interpretation of data; writing of the report; and in the decision to submit the article for publication. The authors declare no other conflict of interest. References 1 Petherick ES , Dalton JE, Moore PJ, Cullum N. Methods for identifying surgical wound infection after discharge from hospital: a systematic review . BMC Infect Dis 2006 ; 6 : 170 . Google Scholar Crossref Search ADS PubMed WorldCat 2 European Centre for Disease Prevention and Control . Annual Epidemiological Report on Communicable Diseases in Europe 2008: Report on the State of Communicable Diseases in the EU and EEA/EFTA countries. European Centre for Disease Prevention and Control : Stockholm , 2008 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 3 Urban JA . Cost analysis of surgical site infections . Surg Infect (Larchmt) 2006 ; 7 ( Suppl 1 ): S19 – S22 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Bratzler DW . The Surgical Infection Prevention and Surgical Care Improvement Projects: promises and pitfalls . Am Surg 2006 ; 72 : 1010 – 1016 . Google Scholar Crossref Search ADS PubMed WorldCat 5 de Lissovoy G , Fraeman K, Hutchins V, Murphy D, Song D, Vaughn BB. Surgical site infection: incidence and impact on hospital utilization and treatment costs . Am J Infect Control 2009 ; 37 : 387 – 397 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Chalfine A , Cauet D, Lin WC, Gonot J, Calvo-Verjat N, Dazza FE et al. Highly sensitive and efficient computer-assisted system for routine surveillance for surgical site infection . Infect Control Hosp Epidemiol 2006 ; 27 : 794 – 801 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Yokoe DS , Noskin GA, Cunnigham SM, Zuccotti G, Plaskett T, Fraser VJ et al. Enhanced identification of postoperative infections among inpatients . Emerg Infect Dis 2004 ; 10 : 1924 – 1930 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Biscione FM , Couto RC, Pedrosa TM. Accounting for incomplete postdischarge follow-up during surveillance of surgical site infection by use of the National Nosocomial Infections Surveillance system's risk index . Infect Control Hosp Epidemiol 2009 ; 30 : 433 – 439 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Leal J , Laupland KB. Validity of electronic surveillance systems: a systematic review . J Hosp Infect 2008 ; 69 : 220 – 229 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Tanner J , Khan D, Aplin C, Ball J, Thomas M, Bankart J. Post-discharge surveillance to identify colorectal surgical site infection rates and related costs . J Hosp Infect 2009 ; 72 : 243 – 250 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Centers for Disease Control and Prevention . National Nosocomial Infections Surveillance System (NNIS) . http://www.cdc.gov/nhsn/PDFs/dataStat/NNIS_2004.pdf [accessed 15 August 2009 ]. 12 Emori TG , Edwards JR, Culver DH, Sartor C, Stroud LA, Gaunt EE et al. Accuracy of reporting nosocomial infections in intensive-care-unit patients to the National Nosocomial Infections Surveillance System: a pilot study . Infect Control Hosp Epidemiol 1998 ; 19 : 308 – 316 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Ronveaux O , Jans B, Suetens C, Carsauw H. Epidemiology of nosocomial bloodstream infections in Belgium, 1992–1996 . Eur J Clin Microbiol Infect Dis 1998 ; 17 : 695 – 700 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Lyytikäinen O , Lumio J, Sarkkinen H, Kolho E, Kostiala A, Ruutu P; Hospital Infection Surveillance Team . Nosocomial bloodstream infections in Finnish hospitals during 1999–2000 . Clin Infect Dis 2002 ; 35 : e14 – e19 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Gastmeier P , Sohr D, Schwab F, Behnke M, Zuschneid I, Brandt C et al. Ten years of KISS: the most important requirements for success . J Hosp Infect 2008 ; 70 ( Suppl 1 ): 11 – 16 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Wilson JA , Ward VP, Coello R, Charlett A, Pearson A. A user evaluation of the Nosocomial Infection National Surveillance System: surgical site infection module . J Hosp Infect 2002 ; 52 : 114 – 121 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Bruce J , Russell EM, Mollison J, Krukowski ZH. The measurement and monitoring of surgical adverse events . Health Technol Assess 2001 ; 5 : 1 – 194 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Hirschhorn LR , Currier JS, Platt R. Electronic surveillance of antibiotic exposure and coded discharge diagnoses as indicators of postoperative infection and other quality assurance measures . Infect Control Hosp Epidemiol 1993 ; 14 : 21 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Yokoe DS , Shapiro M, Simchen E, Platt R. Use of antibiotic exposure to detect postoperative infections . Infect Control Hosp Epidemiol 1998 ; 19 : 317 – 322 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Gastmeier P , Bräuer H, Hauer T, Schumacher M, Daschner F, Ruden H. How many nosocomial infections are missed if identification is restricted to patients with either microbiology reports or antibiotic administration? Infect Control Hosp Epidemiol 1999 ; 20 : 124 – 127 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Sands KE , Yokoe DS, Hooper DC, Tully JL, Horan TC, Gaynes RP et al. Detection of postoperative surgical-site infections: comparison of health plan-based surveillance with hospital-based programs . Infect Control Hosp Epidemiol 2003 ; 24 : 741 – 743 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Miner AL , Sands KE, Yokoe DS, Freedman J, Thompson K, Livingston JM et al. Enhanced identification of postoperative infections among outpatients . Emerg Infect Dis 2004 ; 10 : 1931 – 1937 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Platt R , Kleinman K, Thompson K, Dokholyan RS, Livingston JM, Bergman A et al. Using automated health plan data to assess infection risk from coronary artery bypass surgery . Emerg Infect Dis 2002 ; 8 : 1433 – 1441 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Leveille SG , Gray S, Black DJ, LaCroix AZ, Ferrucci L, Volpato S et al. A new method for identifying antibiotic-treated infections using automated pharmacy records . J Clin Epidemiol 2000 ; 53 : 1069 – 1075 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Swedish Hernia Register . [ Svenkt Bråckregister—Verksamhetsberättelse och resultatrapport 2007 .] 2008 ; www.incanet.se/upload/Dokument/oc6/Brackregister/Rapporter/rapport081223.pdf [accessed 21 June 2010]. 26 Swedish National Board of Health and Welfare . [Official Statistics of Sweden. Inpatient Diseases in Sweden 1987–2008.] Swedish National Board of Health and Welfare : Stockholm , 2009 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 27 Wettermark B , Hammar N, Fored CM, Leimanis A, Otterblad Olausson P, Bergman U et al. The new Swedish Prescribed Drug Register—opportunities for pharmacoepidemiological research and experience from the first six months . Pharmacoepidemiol Drug Saf 2007 ; 16 : 726 – 735 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Fränneby U , Sandblom G, Nyrén O, Nordin P, Gunnarsson U. Self-reported adverse events after groin hernia repair, a study based on a national register . Value Health 2008 ; 11 : 927 – 932 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Taylor EW , Duffy K, Lee K, Hill R, Noone A, Macintyre I et al. Surgical site infection after groin hernia repair . Br J Surg 2004 ; 91 : 105 – 111 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Sanchez-Manuel FJ , Lozano-Garcia J, Seco-Gil JL. Antibiotic prophylaxis for hernia repair . Cochrane Database Syst Rev 2007 ; ( 3 ) CD003769 . Google Scholar OpenURL Placeholder Text WorldCat 31 Nilsson AC , Spetz CL, Carsjo K, Nightingale R, Smedby B. [Reliability of the hospital registry. The diagnostic data are better than their reputation.] Lakartidningen 1994 ; 91 : 598 – 605 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 32 Swedish National Board of Health and Welfare . [Official Statistics of Sweden. Pharmaceuticals—Statistics for 2008.] Swedish National Board of Health and Welfare : Stockholm , 2009 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 33 Swedish National Board of Health and Welfare . [The National Public Health Report 2009.] Swedish National Board of Health and Welfare : Stockholm , 2009 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Footnotes Preliminary results were presented to the Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, California, USA, September 2009 Copyright © 2010 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2010 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.
TI - National surveillance of surgical-site infection through register-based analysis of antibiotic use after inguinal hernia repair
JF - British Journal of Surgery
DO - 10.1002/bjs.7261
DA - 2010-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/national-surveillance-of-surgical-site-infection-through-register-mLRVQe4Fiw
SP - 1722
EP - 1729
VL - 97
IS - 11
DP - DeepDyve
ER -