Clinical and psychosocial predictors of illness duration from randomised controlled trial of prescribing strategies for sore throat
BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7212.736 (Published 18 September 1999) Cite this as: BMJ 1999;319:736- P Little, MRC clinician scientist (psl3{at}soton.ac.uk)a,
- C Gould, research assistanta,
- I Williamson, senior lecturer in primary carea,
- G Warner, general practitionerb,
- M Gantley, anthropologista,
- A L Kinmonth, professor of primary medical carea
- a Community Clinical Sciences (Primary Medical Care Group), Faculty of Health, Medicine and Biological Sciences, Aldermoor Health Centre, Southampton University, Southampton SO16 5ST
- b Nightingale Surgery, Romsey, Hampshire SO51 7QH
- Correspondence to: P Little
- Accepted 29 June 1999
Sore throat is one of the commonest reasons for visiting general practitioners yet little is known about what factors are important in its natural course. This is important since people with prolonged illness after the consultation—the 36% with illness lasting more than five days after seeing the doctor (the median)—are much more likely to reattend.1 Providing patients with information about duration of illness can reduce expectation and reattendance.2 We studied factors that affect duration of sore throat and assessed whether satisfaction with the consultation independently predicts duration of illness.3
Participants, methods, and results
The methods are reported in full elsewhere.3 General practitioners documented the clinical characteristics of 716 patients presenting with sore throat, who were then randomised to one of three prescribing strategies (antibiotics, no antibiotics, delayed antibiotics). Patients recorded satisfaction with the consultation and how well their concerns had been dealt with on four point Likert scales (very, moderately, slightly, not at all) after the consultation and kept a diary of symptoms until better Resolution of symptoms was documented by 579 patients (81%).3 The Likert scales showed good test-retest reliability.3 In 75 consecutive attenders Likert scales on satisfaction and dealing with concerns both demonstrated criterion validity: rank correlations with previously validated scales of the medical interview satisfaction scale questionnaire4 were 0.56 and 0.58 for overall satisfaction and 0.63 and 0.61 for the distress-relief subscale.
Given that 36% of subjects had prolonged illness, we calculated that a total of 517 participants were required to have 80% power and 95% confidence to detect odds ratios of either 0.5 or 2.0 for prolonged illness with dichotomous variables with prevalences of 25%-75% using EPI INFO. We assessed the predictive value of variables using logistic regression models by forward selection using STATA software; variables were retained if they remained significant (5% level).
Older patients (over 12 years), those with longer duration of illness before consultation, those with cough, and those who were less satisfied were more likely to have prolonged course (table). Shorter duration of illness before seeing the doctor was more likely with higher temperature (>37.5°C; odds ratio 3.2, 95% confidence interval 1.9 to 5.5) and with the presence of three out of five of a defined complex of symptoms and signs (2.5; 1.5 to 4.2).3 Most people (69%) had their concerns very well dealt with; this was a better predictor of whether patients were very satisfied (odds ratio 88.6; 38.4 to 177.4) than whether an antibiotic was prescribed (2.97; 1.54 to 5.74). Satisfaction was not predicted by any other variable.
Comment
This trial excluded very ill patients but should inform advice to most patients. Selection and non-response bias were not significant3 so despite limitations, these data are likely to provide generalisable estimates of the predictors of duration of illness.
Patients over 12 and those with cough were more likely, and those with shorter illness before consultation less likely, to have prolonged illness. Short duration before consultation was associated with higher temperature and presence of three out of five symptoms and signs. Thus patients with a higher temperature are more likely to present sooner but also get better more quickly.
Satisfaction with the consultation predicted duration of illness independently of potential confounding variables and was more closely related to effective doctor-patient communication than to prescription of antibiotics. This supports both the preliminary analysis3 and evidence from a systematic review.5 Doctors should elicit patients' concerns and consider counselling patients—particularly those at risk of prolonged illness—about the natural course of sore throat.
Acknowledgments
We thank the following general practitioners for their help in recruitment, constructive comments, and enthusiasm: Nigel Dickson, Graham Newman, Peter Willicombe, Peter White, Sue Tippett, Richard Peace, Julie Chinn, Katie Warner, Neil Ball, Tim Taylor, Deidre Durrant, Mark Rickenbach, Sally Bacon, Tim Whelan, Peter Markby, Simon Goodison, D Traynor, R Briggs, Evelyn Beale, Fiona Bradley, Simon Smith, and Stephen Morgan.
Contributors: PL had the original idea for the study. The protocol was developed by all the authors. Data collection, day to day study management, data entry, and data cleaning were managed by CG. Analysis was performed by PL, and all authors contributed to writing the paper. PL is the guarantor.
Footnotes
-
Funding This work was supported by Wessex NHS Regional Research and Development funds. PL is currently supported by the MRC, but during the data collection for this study was supported by the Wellcome Trust.
-
Competing interests None declared.