The objective of this proposal is to estimate the frequency of Skin and Soft Tissue Infections (SSTI) seen by the General Practitioners (GP). Previous scientific publications on SSTI in the UK have been limited to intravenous drug users and other high-risk groups or are outdated. This proposal aims to update the frequency of SSTI seen by the GP by extracting the data from the Clinical Practice Research Data Link database for the period 2010-2020. The study will assess to what extent SSTI, which are mostly mild, are a common reason for visiting the GP practices, how they are treated, and if they lead to hospitalisation and can in a minority of cases evolve into more severe infections (e.g. sepsis) which might be life-threatening.
The frequency of SSTI overall and per type of SSTI will be estimated for all patients registered in the GP practices (linked with hospitalization data from the HES database) and for high-risk groups (e.g. diabetics). All SSTI recorded at GP practices, outpatient units, emergency services and in hospitals will be traced and categorised according to high-risk groups (e.g. diabetics). The frequency of incision and drainage, Emergency, and hospitalisation will be measured for SSTI episodes. Death associated with SSTI will be also estimated.
The results will be used to assess (a) the frequency of SSTI between 2010-2020, (b) the type of prescription and interventions carried out to treat SSTI; (c) how SSTI affect differently age groups, sex, diabetes and other comorbidities; and (d) how frequently SSTI recur.
Skin and Soft Tissue Infection (SSTI) can be burdensome on the health services. Even if most SSTI are mild, they can be a common reason for visiting GP practices and some are associated with possibly life-threatening complications. This study's objective is to estimate the annual incidence during 2010-2020, by SSTI type (carbuncle, furuncle, cellulitis, abscess, impetigo, unspecified local infection of skin or subcutaneous tissue, superficial injury with mention of infection, dermatitis infectiosa eczematoides, folliculitis, erysipelas, mastitis) and related complications (e.g. lymphadenitis), by age, sex, and comorbidities (e.g. diabetes, obesity).
SSTI recurrence rates will be also estimated to measure the proportion of SSTI patients developing one or more SSTI within 12-months after the index episode. The proportion of SSTI patients having recurrences, will be estimated by age, sex, and comorbidities.
Health care utilisation in the primary and secondary care settings will be measured for SSTI overall and by the aforementioned SSTI types (e.g. abscess). The following medical interventions linked to SSTI will be captured: incision & drainage, microbiological tests, antibiotics, emergency visits. Hospitalisations and deaths associated with SSTI will be also estimated.
The population of interest consists of patients of any age registered in GP practices during 2010-2020 and whose records are linked to the HES database. The data sources will therefore include the Clinical Practice Research Datalink Aurum (CPRD), the Hospital Episode Statistics (HES) and the Office of National Statistics (ONS) databases.
Data will be analysed with descriptive statistics. Frequency tables and proportions with 95% CI will be generated for categorical variables. Mean and standard deviation, median and interquartile range, minimum and maximum will be generated for quantitative variables.
Results will be used to quantify the SSTI incidence and to provide the health care utilisation and thus the basis for future estimation of the potential cost-savings associated with preventive strategies.
Primary outcomes
• Annual SSTI incidence rate. The number of episodes with a diagnosis of SSTI (carbuncle, furuncle, cellulitis, abscess, impetigo, unspecified local infection of skin or subcutaneous tissue, superficial injury with mention of infection, dermatitis infectiosa eczematoides, folliculitis, erysipelas, mastitis) and related complications (e.g. lymphadenitis), according to the codes used in the CPRD databases (Annex 1) will be divided by the PYO contributed by all the patients registered in the CPRD in each calendar year (population at risk of developing SSTI is the total CPRD population). The annual number of SSTI visits episodes per person-year of observation (PYO) will be disaggregated by (i) presence/absence complications (as per Annex I) (ii) age, (iii) sex (iv) comorbidities and (v) IMD class (as per Annex 2).
• Annual SSTI hospitalisation rate. The number of admissions with a diagnosis of SSTI from the HES database and with a CPRD registration code will be divided by the PYO contributed by the CPRD database. The SSTI (e.g. abscesses) and related complications (e.g. lymphadenitis), will be according to the codes used in the CPRD databases (Annex 1)
• Annual SSTI mortality rate. The number of deaths with a diagnosis of SSTI from the ONS database and with CRPD registration will provide the numerator while the number of PYO contributed by the CPRD database will provide the denominator for the mortality rate. The SSTI (e.g. abscesses) and related complications (e.g. lymphadenitis) will be according to the codes used in the CPRD databases (Annex 1)
Secondary outcomes
• SSTI health care utilisation: The proportion of SSTI episodes associated with:i) incision and drainage, ii) microbiological laboratory testing, iii) prescription of antibiotics iv) prescription of other drugs, v) referral to ED or specialist care, vi) hospitalisation, will be estimated for the whole study period (2011-2021).
Tertiary outcomes
• Incidence proportion and incidence rates for recurrent SSTI. For the incidence proportion the population at risk (denominator) will be the SSTI index case and the numerators will be theSSTI index cases developing recurrences within 12 months after the index diagnosis. For the recurrence rates the population at risk (denominator) will be the CPRD registered population and the numerators will be the SSTI index cases developing recurrences within 12 months after the index diagnosis.
• Incidence rate of index SSTI episodes. The number of index SSTI episodes will be divided by the PYO contributed by all the patients registered in the CPRD in specific calendar year
The definition of SSTI episode will be based on the concept of a new episode, being separated by a minimum number of days from a previous SSTI episode. This minimum number of days has varied across the literature, but the most common minimum intervals between events (episodes) have been 15+ days (Davis 2007, Szumowski 2007, Vyas 2014, May 2017), and 30+ days (Srinavasan 2009, Crum-Cianflone 2012, Hemmige 2015, Millar 2017). To be on the safe side, the 30 days will be applied, with any SSTI visit being considered a new episode if they were preceded by at least 31+ days without SSTI recorded events.
Emmanuel ARIS - Chief Investigator - GlaxoSmithKline Biologicals SA (Belgium)
Emmanuel ARIS - Corresponding Applicant - GlaxoSmithKline Biologicals SA (Belgium)
Amoolya Krishnamurthy - Collaborator - GSK India Global Service Private Limited
Dennis Robert - Collaborator - GSK India Global Service Private Limited
Dominique Derreumaux - Collaborator - GlaxoSmithKline Biologicals SA (Belgium)
Georges Van Kriekinge - Collaborator - GlaxoSmithKline Biologicals SA (Belgium)
Kajol Butala - Collaborator - GSK India Global Service Private Limited
Konstantina chatzikonstantinidou - Collaborator - GlaxoSmithKline Biologicals SA (Belgium)
Michael Scherbakov - Collaborator - GlaxoSmithKline Biologicals SA (Belgium)
Michele Pellegrini - Collaborator - GSK Vaccines S.r.l. (Italy)
Nathalie Servotte - Collaborator - GlaxoSmithKline Biologicals SA (Belgium)
Sachi Mehra - Collaborator - GSK India Global Service Private Limited
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation