The SuSPECT qualitative study
Research type
Research Study
Full title
A qualitative study exploring patients' descriptions and doctors' interpretations of upper abdominal symptoms potentially suggestive of upper GI cancers in primary care
IRAS ID
263611
Contact name
Fiona Walter
Contact email
Sponsor organisation
University of Cambridge
Clinicaltrials.gov Identifier
Insurance reference, HVS/2018/2539
Duration of Study in the UK
1 years, 7 months, 1 days
Research summary
Research Summary
The symptoms patients present with to primary care help GPs determine the most probable cause(s) of their symptoms and the diagnostic tests/referral pathways to select to confirm or refute their differential diagnosis. However, the vocabulary patients use to convey symptoms can often differ from standardised terminology used to record symptoms in medical records. As GPs translate lay language into clinically meaningful terminology, symptoms, or important nuances of those symptoms, might be over-looked or misinterpreted. The concern is that GPs' documentation of patients symptoms might not fully reflect what patients actually communicate. Since the symptoms that are recorded in medical records influence the differential diagnoses formulated, mismatches between what patients communicate and what is recorded may lead to the wrong diagnosis being pursued.
Upper gastrointestinal (GI) symptoms with/without systemic symptoms are a common reason for consulting in primary care. Whilst they more frequently indicate non-serious conditions, occasionally they indicate serious disease, including upper GI malignancy. Often vague and intermittent in the early stages of disease, they can fail to elicit suspicion of malignancy. Compounded by suboptimal information exchange during the patient history, patients may be sent down many testing/referral pathways before suspicion is raised causing diagnostic delay. By the time a correct diagnosis is reached, prognosis is often poor, particularly for some upper GI malignancies. Using archived data of primary care consultations we will explore how patients with upper GI and systemic symptoms that could indicate any upper GI cancer describe their symptoms and how GPs document them. The findings will tell us about the information GPs may integrate and omit when forming a differential diagnosis. This will guide development of interventions to optimise the diagnostic testing and referral pathway for patients with low-risk but not no-risk symptoms, leading to more timely diagnosis of upper GI cancers.
Summary of Results
Study Title: A qualitative study exploring patients’ descriptions and doctors’ interpretations of upper abdominal symptoms potentially suggestive of upper GI cancers in primary care Abdominal symptoms are common reason patients present to primary care. Infrequently these symptoms might be caused by an upper gastrointestinal (UGI) cancer. There is some evidence that patients’ descriptions of abdominal symptoms differ from the medical terminology general practitioners (GPs) use to for the same symptoms. Such differences may affect the quality of information GPs collect and record during consultations, creating potential missed opportunities for timely investigation.
To address this concern, we analysed 28 video-recordings of primary care consultations and corresponding medical record data, to explore how patients communicate abdominal symptoms potentially caused by an upper gastrointestinal cancer and how GPs document these symptoms in patient’s medical records. Data for this study was obtained from the ‘One in a Million’ primary care consultation archive, managed by the University of Bristol.
During analysis we looked for patterns in the characteristics of patient’s verbal and non-verbal descriptions of abdominal symptoms that underpinned variations in the accuracy and completeness with which they are documented. Symptoms in medical records that mirrored the content and language patients used were termed ‘alignments’. ‘Misalignments’ constituted symptoms that were documented in vocabulary that was either semantically different from what patients said, was incomplete, or altogether missing.
Our study sample comprised a total of 28 patients and 18 GPs. Patients were aged 40 to 69 years, amongst which there was a marginally higher proportion of female patients (n=16) than male patients (n=12). With regards to GPs, slightly more were male (n=10) than female (n=8). Both patients and GPs were exclusively White British, speaking English as their primary language.
Our analysis was guided by the structure of the Calgary-Cambridge Guide, which is an internationally renowned communication tool used in medical education to improve doctor’s verbal and non-verbal communication skills during different components of the consultation. The specific component upon which our analysis was structured was the ‘information gathering’ phase which breaks down different symptom features doctors should enquire about with their patients according to both the biomedical perspective and patient perspective (in other words aspects of symptoms that are medically orientated as well as those that might be important to the patient).
Across consultations, we identified a total of ten abdominal symptom features patients communicated during consultations. Seven features related to the domain: ‘biomedical perspective’ (e.g., the location and duration of symptoms) and the remaining three to the ‘patient perspective’ (e.g., effect of symptoms on life). Symptom features of each are presented in Figure 2 in the published manuscript (follow appended hyperlink). Among these features, four characteristics underpinned variations in the accuracy and completeness of GP documentation: use of ‘medicalized’ terminology (definitive vocabulary with precise hand gestures), ‘vague’ (non-definitive vocabulary with broader hand gestures) and articulation of symptoms in terms of pain and discomfort.
GPs’ documentation in medical records was typically aligned with information patients communicated about their abdominal and systemic symptoms when patients used ‘medicalized’ for symptom features or articulated ‘pain’. In contrast, symptoms communicated in ‘vague’ language or as ‘discomfort’ were typical of misalignments.
When examining the proportion of misalignments and alignments by symptom feature, misalignments occurred more often for abdominal symptoms when patients said they were ‘persistent and evolving’, communicated the ‘location’ of abdominal symptoms, and conveyed ‘discomfort’, whereas alignments were more common when patients talked about the ‘aggravating and easing factors’ and ‘pain’.
We concluded that a gap exists between patients’ and GP’s language for abdominal symptoms. Mismatches in GPs documentation may have implications for the timely investigation of upper gastrointestinal conditions (including cancers). This warrants empirical evaluation.Full details of the results of this study can be found at: https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fu2790089.ct.sendgrid.net%2Fls%2Fclick%3Fupn%3DXv3JSvJ-2B3M71ppf7N9agbYLcu6Wa5jZlZ-2Fw7JrIX24c29GL6O0rQo57g-2BW5oTxFDePTHkwyPBsHBT25cZdLWoo9Z5NFMNUpDaSwM4x9AbAI-3DEcN6_E1aO2-2BZlVOSJJV-2FajQqskegTd6IRomHYTi-2Fbt8SH3YLFbBY3TBmuoybnRA4MW7QP2sglU06g06aHItLRqNgqbpi5H2QQL7ZVX-2FtIIrQOpYiKDxOwfGzfY7qrOgAKBOnjYQQLiJ4DEuarcYJrxOjntiJMNjrzaum7QvntEYkXWI6MD2rJGsxo607uPsfLz6q03zwQl2bbws7yurpCbA91kg-3D-3D&data=05%7C01%7Capprovals%40hra.nhs.uk%7Ca855d0a084be47711c7a08db8d002314%7C8e1f0acad87d4f20939e36243d574267%7C0%7C0%7C638258803946171319%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=2ycs%2BVekq927FAM%2Fhd%2FS6N6bI3hCTT2rGTbU1z3nTf0%3D&reserved=0
This study which was led by the University of Cambridge was conducted on behalf of the CanTest Collaborative research programme, funded by a Cancer Research UK Catalyst Award.
REC name
East of Scotland Research Ethics Service REC 1
REC reference
19/ES/0057
Date of REC Opinion
14 Jun 2019
REC opinion
Further Information Favourable Opinion