Research and analysis

Increase in Clostridioides difficile infections (CDI): current epidemiology, data and investigations – Technical report

Published 1 May 2025

Main points

Clostridioides difficile infection (CDI) cases increased by 33% between financial year (FY) 2020 to 2021 and FY 2023 to 2024. This increase contrasts to the year-on-year decrease in rates between FY 2007 to 2008 and FY 2013 to 2014 and stable rates seen between FY 2013 to 2014 and FY 2020 to 2021.

The 2023 to 2024 FY numbers mark the highest overall case count and incidence rates since the nadir of FY 2011 to 2012, but remain substantially below levels seen in 2007.

Increases are observed in both hospital- and community-onset CDI cases, and across all age groups and sexes.

All 7 UKHSA regions have seen an increase in counts and rates between FY 2020 to 2021 and calendar year 2024 but to differing extents.

CDI immunoassay test positivity rates are relatively unchanged since FY 2012 to 2013 (when rates were first stable) and are insufficient to solely explain the large rise in CDI incidence.

CDI ribotyping does not indicate clonal expansion as the cause of the increase.

The English Surveillance Programme for Antimicrobial Utilisation and Resistance (ESPAUR) report demonstrates an increase in overall antibiotic consumption in England in 2023. However, consumption remains below pre-pandemic levels, and significantly below the consumption peak seen in 2014.

Use of broad-spectrum antibiotics is a recognised risk factor for CDI. Broad-spectrum antibiotics typically fall under the UK Access, Watch, Reserve, and Other (UK-AWaRe) classification Watch and Reserve categories. Since 2021, use of Access antibiotics as a proportion of total antimicrobial consumption has seen a gradual increase, beyond 2019 levels.

Findings from modelling analyses are in line with the known epidemiology of CDI in England and do not indicate a risk factor causing this increase.

Several data gaps have been identified for further investigation, including use of corridor care (care in unconventional places), infection prevention control resource and staffing, cleaning procedures within the hospital, sampling and testing practises and so on. We aim to collect this data through the use of surveys and existing reported data where possible.

Further work is ongoing, with next steps to include further analysis of available data, enhanced by collection of data not currently available.

Situational assessment of CDI

Data summary

C. difficile is a leading cause of healthcare-associated infections (HCAIs). Its spores are highly resistant to heat and many disinfectants, making it easily transmissible within healthcare settings. C. difficile produces 2 principal toxins, which are both enterotoxic and cytotoxic but are traditionally labelled enterotoxin A and cytotoxin B, virulence factors for disease. The toxins damage the colonic mucosa resulting in a spectrum of disease, ranging from mild diarrhoea to severe colitis, which may be associated with pseudomembranes, megacolon and perforation.

The spores of C. difficile are the transmissible part and contaminate the environment, where they can survive for long periods and from where they can be cross-transmitted to patients. Ingested spores can germinate and CDI can develop if the C. difficile produces toxins. This usually occurs once a patient has received antibiotics, which disturb the normal gut microbiota such that the C. difficile is able to dominate and colonise the large intestine.

CDI incidence

Mandatory surveillance of CDI began in FY 2007 to 2008 with analysis reported routinely as official statistics in the public domain.  

As of the 31 December 2024, we report a continuing increasing trend in CDI counts and rates in England, following the rises seen from FY 2021 to 2022.

The incidence of CDI in England for cases reported in the mandatory surveillance system can be divided into 3 distinct periods; the first, between FY 2007 to 2008 and FY 2013 to 2014, saw a rapid decline in the incidence from 107.6 cases to 24.8 cases per 100,000 population, representing a 77% decline; the second is one of continued gradual decline followed by relative stability in the incidence, with minimal fluctuation between FY 2013 to 2014 and FY 2020 to 2021. The final period, starting in FY 2021 to 2022, sees a shift in trend for the first time since the mandatory reporting began, with an increase in cases annually. Between FY 2020 to 2021 and FY 2023 to 2024, rates have increased 33% reaching 29.5 cases per 100,000 population, (Figure 1). CDI rates have now increased to levels not seen since FY 2011 to 2012. Detailed data is also available in the Annual Epidemiology Commentary, 2024 (1). Cases are further broken down into Prior Trust Exposure categories, definitions for the breakdown in healthcare exposure can be found in Appendix.

Children under 2 years of age are rarely tested for CDI due to high rates of colonisation in this age group, thus this report includes data for those aged 2 years and over.

Figure 1. Rates of (A) all reported and (B) hospital-onset (HO) CDI cases, between FY 2007 to 2008 and FY 2023 to 2024

A                                                                       B

All cases (asterisk): Mid-year population estimates for January to December 2023 onwards were unavailable at time of publication and so population data for January to December 2022 was used as a proxy.

---

Comparing January to December 2012, when rates had first stabilised, to January to December 2024 an increase of 7.7% in case counts was observed, the most recent 3 months (October to December 2024) saw a steeper increase of 13.3%, compared with the same period in 2023. We also note a much steeper incline in monthly case counts from June 2023 (Figure 2).

Figure 2. Twelve-month rolling percent change since calendar year 2012 [note 1] for CDI, until December 2024

Note 1: The calendar year 2012 was chosen as the baseline, as it marks a period when CDI rates had stabilised

---

Comparing CDI rates annually (Figure 3), not only have rates increased year on year from 2020, but the rate of change over time is accelerating. The rate rises are present in cases from both hospital- and community-onset cases. Although rates for the most recent quarter (October to December 2024) appear to decline, this is likely seasonal, and will continue to be monitored.

Figure 3A. Quarterly CDI rate from April 2018 to December 2024, by all CDI

Figure 3B. Quarterly CDI rate from April 2018 to December 2024, by hospital-onset

Figure 3C. Quarterly CDI rate from April 2018 to December 2024, by community-onset CDI

Onset type and prior trust exposure

Overall, hospital-onset, healthcare associated (HOHA) cases see the greatest proportion of cases, from October to December 2024, accounting for 46% of all reported cases. Community-onset, healthcare associated (COHA) and community-onset, community associated (COCA) cases constituted 17% and 26% of the total, respectively. Community-onset indeterminate-association (COIA) cases were 10%. These percentages have been relatively stable since April 2018, although there were increases in the CO categories in the recent quarters (Figure 4) (full definitions can be found in Appendix).

Figure 4. CDI percentage by onset type by quarter, April 2017 [note 2] to December 2024

Note 2: Data begins from 2018 when prior trust exposure questions were mandated in the surveillance programme

Age and sex distribution

Rates of CDI increase with age and are highest amongst those 75 years and over. Since the COVID-19 pandemic, incidence has increased amongst males and females in all age groups, and rates were greatest in patients aged 65 years and older (Figure 5).

Figure 5. Age and sex distribution of CDI rates by financial quarter, April 2012 to December

Note: differing scale on y axes

Vertical dashed line refers to beginning of the COVID-19 pandemic

Regional distribution

The distribution of cases between January 2024 and December 2024 was analysed by NHS commissioning region, integrated care board (ICB) and sub integrated care board (sICB). sICBs replaced the previous CCG/STP structure, with an ICB consisting of several sICBs. When analysed by NHS commissioning region, CDI rates varied across England. The highest incidence was observed in the North West, at 48.1 cases per 100,000 population. The South West and Midlands reported rates of 39.5 and 37.3 per 100,000, respectively, followed by the East of England (34.7) and North East and Yorkshire (33.3). The South East reported rates of 29.7 per 100,000 population, followed by the London region, with the lowest rate at 18.3 per 100,000 population.

Figure 6 shows ICBs with the highest rates, as Lancashire and South Cumbria ICB (49.7 cases per 100,000 population), Cornwall and the Isles of Scilly ICB (48.5 cases per 100,000 population) and Herefordshire and Worcestershire ICB (48 cases per 100,000 population). The lowest ICB rates were observed in North East London ICB (14.8 cases per 100,000 population), South East London ICB (16 cases per 100,000 population) and Frimley ICB (19 cases per 100,000 population).

After age-sex standardisation, the ICBs with the highest rates were Greater Manchester ICB (52.0 cases per 100,000 population), Lancashire and South Cumbria ICB (49.6 cases per 100,000 population) and Birmingham and Solihull ICB (44.6 cases per 100,000 population) (Figure 6).

Figure 6. Geographical distribution of crude (A) and age-sex standardised (B) CDI rates by integrated care board (ICB), FY 2023 to 2024

Regional analysis was repeated at the sICB level because patterns seen at the ICB level might not fully reflect local variations. Particularly high CDI rates were observed within the North West region. The highest rates were in Cheshire and Merseyside ICB – 12F (formerly Wirral CCG) at 76.6 per 100,000 population, followed by Lancashire and South Cumbria ICB – 02M (formerly Fylde and Wyre CCG) at 71.4 per 100,000 population (Figure 7). Four other sICBs within Cheshire and Merseyside, Lancashire and South Cumbria and Greater Manchester had rates higher than 60 per 100,000 population (Figure 7) – substantially higher than the mean sICB rate during this period, 37.1 cases per 100,000 population.

Figure 7A. Geographical distribution of CDI rates by ICB, January 2024 to December 2024

Figure 7B. Geographical distribution of CDI rates by sICB, January 2024 to December 2024

                                                                            

Change in CDI rates over time

Comparing FY 2020 to 2021 (when national rates were relatively stable and at their lowest point since surveillance began) to the most recent 12 months (January 2024 to December 2024) there were widespread increases in rates of all reported (hospital-onset and community-onset) CDI cases with regional variation. When analysed at an ICB level, the greatest absolute increases between these time periods were observed in Black Country ICB, Cheshire and Merseyside ICB and Dorset ICB, each increasing by more than 20 per 100,000 population. When analysed by percentage increase, notable increases (more than 100%) were also observed in Leicester, Leicestershire and Rutland ICB, Northamptonshire ICB and Shropshire, Telford and Wrekin ICB. Some ICBs saw much lower increases, with 4 of 5 London based ICBs seeing increases of less than 5 per 100,000 population (Figure 8).

At sICB level, there were areas of very large increases – most notably Cheshire and Merseyside ICB – 01J (formerly Knowsley CCG), which saw an absolute increase of 44.6 per 100,000 population or 213%. Nottingham and Nottinghamshire ICB – 02Q (formerly Bassetlaw CCG), saw an increase of 28 per 100,00 population (165.6%) and Mid and South Essex ICB – 99E (formerly Basildon and Brentwood CCG) increased by 28 per 100,000 population (157.7%).

Some central and Northern sub ICBs remained relatively stable, with 4 seeing decreases in cases – Cheshire and Merseyside ICB – 01V, (formerly Southport and Formby CCG) the West Yorkshire ICBs of 02T (formerly Calderdale CCG) and 03R (formerly Wakefield CCG) and North East and North Cumbria ICB – 00P (formerly Sunderland CCG) (Figure 8). It is notable that some of these decreases may be due to relatively high rates in the initial time period (FY 2020 to 2021). It must also be noted, counts will vary by ICB and sICB based on size (Figure 8).

Figure 8A. Absolute change (per 100,000 population) in CDI rates by ICB between April 2020 to March 2021 and January 2024 to December 2024  

Figure 8B. Absolute change (per 100,000 population) in CDI rates by sICB between April 2020 to March 2021 and January 2024 to December 2024   

Mortality data

Case fatality rates (CFRs) have broadly remained stable across all age groups, from April 2018 to December 2024. We observe quarterly fluctuations, and more recent years showing some decline in older age groups (Figure 9).

Figure 9. CFR by age group, from April 2018 to December 2024

Further analysis of CDI increases

Further analyses were performed which accounted for differences in factors such as: demographics, level of CDI testing (toxin immunoassay), UKHSA region, antibiotic prescribing and characteristics of the healthcare system, for example hospital activity, waiting lists, and so on.

ICB-level modelling

A linear regression model explored the relationship between either cases per 1,000 population, number of COCA cases per 1,000 population of an ICB or hospital-associated COHA and HOHA per 1,000 admissions in an ICB per quarter with a number of ICB-level variables. 

Results indicate that both increased deprivation and longer mean referral to treatment (RTT) time are most significantly associated with a higher number of both total and COCA CDI cases per 1,000 population while increased rates of testing is related to a higher number of hospital-associated CDI cases per 1,000 admissions.

Trust-level modelling

Analysis of healthcare-associated CDI cases from October 2023 to September 2024 by NHS trust did identify 3 factors which were significantly associated with increased rates in this period: higher percentage of white ethnicity in trust catchment population, increased deprivation in trust catchment population and higher CDI rate in FY 2020 to 2021 (regarded as baseline). These factors are known to be associated with CDI and will be subject to further in-depth analysis, but do not by themselves fully explain the overall increase in CDI rates which has been observed.

Individual-level modelling

A logistic regression model was used to explore the factors that affect the likelihood of an individual developing a CDI infection between 2019 to 2020 and 2023 to 2024. CDI case data was linked to hospital episode statistics (HES) to infer patient level information (age, gender, ethnicity, previous visits in the last 12 months and comorbidities listed in their ICD-10 codes). This data was combined with factors associated to the trust that this patient visited (antibiotic prescribing rate, bed occupancy, testing), and factors associated to their registered postcode (IMD, community prescribing). The reference population was a random sample of 1% of non-CDI admissions within the same time period.

Results indicate that all factors significantly increase the odds ratio (OR) of developing CDI (Table 1) except for being male, non-white ethnicities, referral to treatment time and London region, which decrease the OR, and antibiotic prescribing and IMD, which are statistically insignificant.

Due to the integration of discrete and continuous factors in this analysis, the odds ratios should be interpreted in combination with the data range. For example, considering the data range for continuous variables demonstrates that the largest OR is that representing age (OR 50 years = 4.38 versus OR 85 years = 12.33).  

The findings that older patients who are sicker, male and of white ethnicity are more likely to have CDI is in line with the known epidemiology of CDI in England. The results for hospital and community variables are organisational-level measures applied to individuals based on where they went to hospital and where they live. They are proxy measures for the environment the patient was treated in and should be interpreted in that way.

Table 1. Odds ratios from individual-level logistic regression model

Factor	Units or comparison population	OR	p-value
Age	Years	1.030	<0.001
Male	Compared to female	0.767	<0.001
Ethnicity	Compared to white ethnic group
Asian		0.769	<0.001
Black		0.529	<0.001
Mixed		0.645	<0.001
Other ethnic group		0.853	<0.001
Comorbidities
Frailty	Score from 0-17	1.119	<0.001
Renal disease	Binary data	1.333	<0.001
Cancer	Binary data	1.051	<0.001
Other comorbidities	Up to 15 other comorbidities	1.142	<0.001
Total nights in hospital in previous year (excluding current spell)	Days	1.014	<0.001
Hospital and community variables
Hospital antibiotic prescribing rate	DDDs per 1000 admissions	1.000	<0.001
Hospital bed occupancy	Percentage	1.004	<0.001
CDI testing rate	Tests per 1,000 admissions	1.004	<0.001
Community prescribing rate	Units prescribed per 1000 people	0.999	0.275
Percentage of A&E patients waiting over 4 hours	Percentage	1.002	<0.001
Mean referral to treatment time	Days	0.998	<0.001
Index of Multiple Deprivation (IMD)	Most deprived decile compared to least. Other deciles not shown.	1.054	0.014
UKHSA region	Compared to East Midlands
East of England		1.218	<0.001
London		0.729	<0.001
North East		1.168	<0.001
North West		1.211	<0.001
South East		0.989	0.643
South West		1.137	<0.001
West Midlands		1.068	0.010
Yorkshire and Humber		1.180	<0.001

Assessment of potential drivers of this increase

Established risk factors associated with CDI include circulation of highly virulent and transmissible ribotypes (for example ribotype 027), and exposure to antibiotics (some of which are deemed higher risk).

In addition, changes in sampling or testing may influence the observed changes in CDI rates, due to greater ascertainment and false positives association with testing. Evaluation of existing data in these areas does not provide a strong indication that any one of these factors is solely driving the recent increases in CDI. Details of the analysis are provided below.

Ribotyping data

Analysis of ribotyping data from the Clostridioides difficile Ribotyping Network (CDRN), linked to mandatory reported CDI cases by year does not indicate a particular clonal expansion driving the CDI increases we observe. It should be noted, there is substantial variation in isolate submission to the CDRN service by acute NHS trusts, including variation in those trusts identified as seeing greater increases, leaving gaps in this data. More details on the CDRN service and submission of isolates, including criteria for submission can be found at Clostridioides difficile ribotyping network (CDRN) report.

Figure 10 demonstrates the linked ribotypes prevalence nationally between FY 2008 to 2009 and FY 2023 to 2024. Ribotypes with a prevalence of 2% or more are shown, any valid ribotypes under 2% are grouped into ‘Other’, ‘NA’ represent sporadic ribotypes that are not commonly recognised across the CDRN labs, along with unknown and invalid data entries.

There has been a historical decline in ribotypes 001, 106 and 027 with the prevalence for 027 being under 2% in the latest FY 2023 to 2024, a steep decline from its initial prevalence of 31.7% in FY 2008 to 2009.

Figure 10. Prevalence of C. difficile ribotypes [note 3] [note 4] in England, by financial year April 2008 to March 2024, of linked CDRN DCS CDI cases

Note 3: 29.4% of CDI cases linked to a CDRN record successfully

Note 4: NA – include where ribotype was not recognised by the CDRN service

CDI testing and prevalence

C. difficile toxin immunoassay testing saw a decline between quarter 1 (April to June) 2010 and quarter 3 (October to December) 2013, which corresponded with a period of declining CDI incidence, following the peaks observed in the early 2000s. During this time the toxin immunoassay testing rates decreased from 19.2 to 14.7 per 1,000 bed-days. Following this, the testing rates remained stable, before switching to an increasing trend from quarter 3 (October to December) 2017. A more rapid increase in testing is observed from quarter 3 (October to December) 2023 (when the rate was 20.0 per 1,000 bed days) and quarter 2 (July to October) 2024 (when the rate was 23.0 per 1,000 bed days), corresponding with the increased incidence in CDI (Figure 11). Increases are more notable in the rolling percentage toxin test figure, Figure 12.  

Figure 11. Trends in the rate of C. difficile toxin immunoassay testing in England, by calendar quarter: April 2010 to September 2024

Figure 12. Twelve-month (by calendar quarter) rolling percentage change in C. difficile toxin tests: January 2020 to September 2024, England

In contrast to the increases in CDI testing, test positivity rate is relatively stable. Since quarter 4 (January to March) 2018, it has remained relatively stable between 1.9% and 2.8%, with some fluctuation. Although testing has increased by 25% from the FY 2020 to 2021 reference point, overall positivity is relatively unchanged and is insufficient to solely explain the large rises in CDI incidence, but not ruled out as a contributor (Figure 13).

Figure 13. C. difficile toxin immunoassay positivity rate in England (acute trusts), by calendar quarter: April 2010 to September 2024

Antibiotic prescribing

Available data from the English Surveillance Programme for Antibiotic Prescribing, Usage and Resistance (ESPAUR) [2] demonstrates an increase in overall antibiotic consumption in England in 2023 (compared to 2022). However, levels remain below pre-pandemic levels, and significantly below the peak in consumption seen in 2014, 22.5 DDDs per 1,000 inhabitants (Figure 14).

Broad-spectrum antibiotics typically fall under the UK-AWaRe classification Watch and Reserve categories [3], with the exception of first generation cephalosporins. Access antibiotics are those with a narrow spectrum of activity, whilst Watch antibiotics have a higher resistance potential and are broader-spectrum antibiotics, and Reserve antibiotics are ‘last resort’ antibiotics, including new antibiotics [4]. Since 2021, use of Access antibiotics as a proportion of total antimicrobial consumption has seen a gradual increase, beyond 2019 levels. Access use in primary care was consistently higher than in secondary care between 2019 and 2023. In 2023, 66.7% of antibiotics used in primary care were in the Access category, 53.4% of antibiotics in secondary care and 64.1% across both sectors.

Figure 14. Total consumption of antibiotics by calendar year from 2014 to 2023 (taken from 2023 ESPAUR report)

Waiting lists for hospital treatment

Published data reports waiting lists for hospital treatments to have reached a record high of 7.7 million in September 2023 but to have fallen to 7.5 million in December 2024 (Figure 15). However, levels are still just over 1.5 times 2020 levels. Waiting lists for treatment have been increasing annually since 2010; however, this increase is much more pronounced from 2020, coinciding with the COVID-19 pandemic which added to this increase with the cessation of routine treatment.

Consequences of long waiting times for hospital treatment include antibiotic treatment, perhaps multiple courses of treatment, pending hospital assessment for an underlying condition or a procedure such as an operation. In some cases antibiotics may be given for a prolonged period. Research showed people waiting for general surgery had an average of 2 more primary care prescriptions and 2 more secondary care contacts per year. This increased to 6 more primary care prescriptions in those waiting for gastroenterology treatment [5].

Figure 15. Waiting list for hospital treatment

Source: NHSE England (Consultant-led referral to treatment waiting times)

Accident and emergency (A&E) admissions and wait times, and use of ‘corridor care’

Attendance at A&E departments has been increasing annually since early the early 2000s, consequently leading to increased wait times, and breaches of the ‘4-hour wait’. In January 2025, 42.3% of patients waited over 4 hours (Figure 16), and data from the Royal College of Emergency Medicine shows that more than 1.5 million patients waited 12 hours or more for care in 2023, of which 65% were patients awaiting admission into an inpatient ward bed [6].

Figure 16. Patients spending over 4 hours in major A&E

Source: NHSE England (Accident and emergency attendances and emergency admissions)

Healthcare workers are therefore regularly required to provide care to patients in chairs and corridors for extended periods of time giving rise to the concept of ‘corridor care’ [7]. A survey of almost 11,000 UK frontline nursing staff revealed that 1 in 3 (37%) of them had delivered care in inappropriate settings on their last shift [8]. As corridor-care occurs when the system is under pressure, it is reasonable to assume difficulties with IPC compliance, not only due to stretched healthcare workers but also because, the tools for IPC, such as hand-wash basins, will not be as readily accessible in a non-clinical area. Further, cleaning standards for non-clinical areas less thorough than for clinical areas. Numerous factors make EDs a potential source for acquisition of CDI infection, including a crowded environment, sub-optimal clinical management, limited access to isolation rooms and difficulty with implementing cleaning and IPC measures in these areas which were not originally designed for healthcare delivery

Data is not readily available on ‘corridor care’, and the impact this may be having. A planned survey within hospital trusts is being designed to collect some of this data.

Current IPC and guidance

C. difficile causes infectious diarrhoea, which is most frequently healthcare associated, and increases morbidity, mortality and hospital length of stay. As a healthcare associated infection, it imposes a unique burden on healthcare resources because recurrences of CDI are estimated to occur in up to 25% of treated patients with 40 to 65% of patients experiencing multiple recurrences after a second or third episode respectively. Although many CDI are healthcare-associated, the potential contribution of the non-healthcare environment (for example animals, and the food chain) to C. difficile epidemiology is increasingly being recognised.

Investigators at University of Leeds and the CDRN studied antibiotic susceptibility of more than 200 C. difficile isolates from ribotypes detected more than 5 times in a 3-month period (2022) and less common ribotypes of known epidemiological significance. The study found reduced susceptibility to metronidazole, vancomycin and fidaxomicin, the antibiotics used to treat C. difficile, to be uncommon across all ribotypes tested. Oral metronidazole is no longer recommended for treatment considering gut concentrations close to minimum inhibitory concentration (MIC). Gut levels of vancomycin are far more than MICs, including those officially above the vancomycin breakpoint at 2mg/L, and therefore clinical relevance of reduced susceptibility by this breakpoint definition may be limited.

Resistance to clindamycin was most common (73%), followed by imipenem (23%), moxifloxacin (9.5%) and rifampicin (3.1%). There was no resistance to tigecycline. The most resistant isolate was the single ribotype 955 tested in this series, notably reduced susceptibility to metronidazole. This agreed with susceptibility testing of C. difficile RT955 isolates involved in the outbreak, which demonstrated similar antimicrobial resistance profiles including reduced metronidazole susceptibility.

Trends globally

Surveillance programmes vary globally, and not all nations are able to report comprehensive national data. Global data was available for some countries through a publication by Moisi et al. 2024 [9] (Figure 17A, taken Moisi et al. 2024 [9]). CDI case reporting by country is variable, with differing definitions, making direct comparisons problematic. Incidence rate across several reported countries remain relatively stable with fluctuations. Notable increases in recent years are observed in, the USA (rise from 2020 to 2021), Sweden (uptick from 2021 to 2022), and Ireland (spike in 2019), and Norway from 2019. Hospital-onset or healthcare associated rates remained across for all reported countries, except for England. Wales have reported similar rises in CDI rates, data is not presented here (Figure 17B).  

Figure 17A. Rates of CDI per 100,000 population, across multiple countries, to 2023 (taken Moisi et al. 2023 [9]) [note 5]

Figure 17B. Rates of hospital-onset or healthcare-associated CDI cases per 100,000 bed days, across multiple countries, to 2023 (taken Moisi et al. 2023 [9]) [note 5]

Note 5: Data for some countries is by financial year and others calendar year, data may not be national data for some countries and data endpoints vary

Further investigation

There are many further routes of investigation that will be addressed in future reports where possible. These include:

• analysis of geographical variation, including analysis of statistical outliers, at trust and sICB level –

  • ribotype trends within regions
  • variation in community and hospital cases
  • unwarranted variation between regions

• use of surveys and local intelligence to understand differences in hospital processes, including –

  • staffing levels and practise, including IPC and cleaning
  • sampling and testing data
  • general hospital processes including CDI testing and isolation procedures
• a deep dive on community-onset cases, to identify potential trends and drivers
• further analyses of prescribing trends and their potential relationship with changing CDI rates

References

1. UKHSA. Annual epidemiological commentary: Gram-negative, MRSA, MSSA bacteraemia and C. difficile infections, up to and including financial year 2023 to 2024 2024
2. UKHSA. English surveillance programme for antimicrobial utilisation and resistance (ESPAUR) report 2024
3. Sabine Bou-Antoine and others. ‘Adaptation of the WHO AWaRe (Access, Watch, Reserve) antibiotic classification to support national antimicrobial stewardship priorities in the UK: findings from a modified Delphi approach to achieve expert consensus’ Journal of Antimicrobial Chemotherapy 2025
4. UKHSA. UK Access, Watch, Reserve, and Other classification for antibiotics (UK-AWaRe antibiotic classification) 2025
5. James C, Denholm R and Wood R. ‘The cost of keeping patients waiting: retrospective treatment-control study of additional healthcare utilisation for UK patients awaiting elective treatment’ BMC Health Service Research 2024
6. Royal College of Emergency Medicine. RCEM explains: Long waits and excess deaths
7. Royal College of Nursing. Corridor care: RCN declares ‘national emergency’ and demands political action
8. Royal College of Nursing. Corridor care: unsafe, undignified, unacceptable. The impact on patients and staff of providing care in corridors and other inappropriate areas
9. Angulo FJ, Furtado M, Gonzalez E, Zhang P, Kelly PH, Moïsi JC. ‘Incidence of public health surveillance-reported Clostridioides difficile infections in thirteen countries worldwide: A narrative review’ Anaerobe 2024

Appendix

Prior trust exposure categories

Hospital-onset healthcare-associated (HOHA)

Date of onset is greater than 2 days after admission (where day of admission is day 1).

Community-onset healthcare-associated (COHA)

Is not categorised HOHA and the patient was most recently discharged from the same reporting trust in the 28 days prior to the specimen date (where day 1 is the date of discharge).

Community-onset indeterminate association (COIA)

Is not categorised HOHA and the patient was most recently discharged from the same reporting trust between 29 and 84 days prior to the specimen date (where day 1 is the date of discharge).

Community-onset community-associated (COCA)

Is not categorised HOHA and the patient has not been discharged from the same reporting organisation in the 84 days prior to the specimen date (where day 1 is the date of discharge).

Unknown

The reporting trust answered ‘Don’t know’ to the question regarding previous discharge in the 3 months prior to CDI case.

Not reported

The reporting trust did not provide any answer for questions on prior admission.