Stool submission by general practitioners in SW England - when, why and how? A qualitative study

We know little about when and why general practitioners (GPs) submit stool specimens in patients with diarrhoea. The recent UK-wide intestinal infectious disease (IID2) study found ten GP consultations for every case reported to national surveillance. We aimed to explore what factors influence GP's decisions to send stool specimens for laboratory investigation, and what guidance, if any, informs them.

We used qualitative methods that enabled us to explore opinions and ask open questions through 20 telephone interviews with GPs with a range of stool submission rates in England, and a discussion group with 24 GPs. Interviews were transcribed and subjected to content analysis.

Interviews: GPs only sent stool specimens to microbiology if diarrhoea persisted for over one week, after recent travel, or the patient was very unwell. Very few had a systematic approach to determine the clinical or public health need for a stool specimen. Only two GPs specifically asked patients about blood in their stool; only half asked about recent antibiotics, or potential food poisoning, and few asked about patients' occupations. Few GPs gave patients advice on how to collect specimens.Results from interviews and discussion group in relation to guidance: All reported that the HPA stool guidance and patient collection instructions would be useful in their clinical work, but only one GP (an interviewee) had previously accessed them. The majority of GPs would value links to guidance on electronic requests. Most GPs were surprised that a negative stool report did not exclude all the common causes of IID.

GPs value stool culture and laboratories should continue to provide it. Patient instructions on how to collect stool specimens should be within stool collection kits. Through readily accessible guidance and education, GPs need to be encouraged to develop a more systematic approach to eliciting and recording details in the patient's history that indicate greater risk of severe infection or public health consequences. Mild or short duration IID (under one week) due to any cause is less likely to be picked up in national surveillance as GPs do not routinely submit specimens in these cases.