Abstract
Background: Acute cough, often caused by a viral respiratory infection, is a common symptom in primary care. Although clinical guidelines recommend symptomatic treatment for acute cough, antibiotics are frequently prescribed. Objective: To determine antibiotic prescribing for acute cough at the initial consultation and to follow subsequent medical consultations and use of medications. Methods: The study population included all adult patients with acute cough who visited general practitioners from one health centre (HC) during four months. Information was gathered from medical charts and telephone interviews conducted two weeks later. Results: Fifty six of three hundred and thirty eight participants (16.6%) received antibiotics at the initial visit. Eighty three participants made subsequent visits to the HC, 40 participants visited physicians outside the HC and nine participants visited both. During two weeks after the initial visit, 35 participants were prescribed antibiotics (eight in the HC, 27 outside the HC). Total antibiotic use rose to 27% (91/338) during the study period. At that time 98 (29%) of the participants reported they were still ill. Multivariate analysis showed that expectation to receive antibiotics was reported at a higher rate by the participants who received it, as compared to those who did not (32.2% versus 13.2%, OR: 2.3; 95% CI: 1.2–4.8). Receiving antibiotics was also associated with use of health services (20.3% versus 9.9%, OR: 2.7; 95% CI: 1.2–6.2).
Conclusions: Patient activism during the course of acute cough is associated with increased antibiotic use.
Keywords::
KEY MESSAGE(S):
Antibiotic use for acute cough is determined by the behaviour of both physicians and patients.
Patients who continue to seek care after an initial visit for acute cough contribute to an increase in antibiotic prescribing.
When acute cough lasts long, patients may initiate further care including taking antibiotics by their own decision.
INTRODUCTION
Bacterial resistance to antibiotics is a worrisome phenomenon that is related to overuse (Citation1,Citation2). Misuse of antibiotics occurs commonly in the treatment of acute cough (Citation3). Several factors play a role in antibiotic prescribing for acute cough. The diagnosis of the cause of acute cough relies mainly on exclusion criteria and physicians make management decisions under conditions of uncertainty (Citation4). When physicians prescribe antibiotics, the explicit requests of patients are taken into account (Citation5–7) or the perceived expectations of the patients (Citation8,Citation9). Furthermore, although patients tend to understand the appropriate role of antibiotic treatment, patients often expect to have antibiotics prescribed for acute respiratory symptoms (Citation10).
The recommendation to avoid prescribing antibiotics to patients with acute cough is well-established (Citation11,Citation12). Interventions to reduce antibiotic prescriptions for acute cough have generally focused on the behaviour of physicians rather than the behaviour of patients (Citation13–16).
An education project focusing on the care of patients with acute cough was conducted in an urban health centre (HC) in northern Israel. The study that accompanied the project focused on patient behaviour. Its main objective was to examine the association between antibiotic prescription in the HC and utilization of health services and medications during the disease period. The aims of the study were: (a) to describe antibiotic prescription by general practitioners (GPs) at the first visit of patients with acute cough and the association with demographics and morbidity parameters; (b) to describe medication use (antibiotics and non-antibiotics) during two weeks following the initial visit; (c) to describe the use of health care services during the two weeks after the first visit for acute cough: return visits to the HC and visits to physicians outside the HC; and (d) to analyse the association between utilization of health care services and medication use and satisfaction at the first visit.
METHODS
Study design
This prospective cohort study took place in an urban public HC belonging to Clalit Health Services. The participants were followed during a two-week period after the first visit for acute cough to the HC. The research was approved by the local Ethics Committee. The study was conducted by the medical staff, which has access to all medical records in the HC.
The Health Centre
The HC is situated in the centre of a small town in northern Israel. Care is provided to 8750 adult patients by five GPs. Patients are assigned to GPs. In the lists of the GPs, the rates of elderly and chronic ill patients are similar; all GPs have worked at the HC for more than five years. Physician visits in the practice are free; however, visits to physicians after hours and self-referrals to hospital emergency rooms are charged. The nearest hospital is 15 km away. All visits are documented in electronic medical records, from which diagnoses of chronic diseases are available.
The education project
The education project was aimed to implement appropriate prescribing for patients with acute respiratory diseases. A seminar was held on acute respiratory tract infections and treatment of acute respiratory infections. During this seminar guidelines for the treatment of acute cough were presented. The importance of reliable and accurate medical records was emphasized.
The participating patients
The study included all adult patients who visited the general practice during a period of four months between January and April 2004 and presented with acute cough as the main complaint. Patients who were diagnosed with pneumonia at some time during the study and patients with chronic lung disease (asthma and COPD) were excluded. We wished to minimize interference with the routine work of the GPs and we avoided the extra work of filling out questionnaires, thus we extracted data from the medical records. Enrolment was accomplished by each GP from the eligible patients on the daily consulting list. A validity check for inclusion and exclusion criteria was made by reading a sample of 100 medical records (50 of the included and 50 excluded patients each).
Data collection and measurements
Within the HC, information collected from the participants’ medical records included demographic data, co-morbidities, and details about the visit, including prescriptions, referrals, sick leave, and repeat visits for acute cough.
A telephone interview was conducted with the participants two weeks after the initial practice visit by one of the authors (SCH). After giving verbal consent, the participants were asked how they felt (on a 1–5 scale: very bad to excellent), recalled satisfaction with treatment during the first visit (on a 0–3 scale: dissatisfied to very satisfied), whether or not the participants had expected to be treated by antibiotics at the first visit, whether or not the patients subsequently visited physicians outside the practice for the same problem, and the use of antibiotics and non-antibiotic medications. The participants also reported on work absenteeism: days of absenteeism were split at three days, which is the limit from which sick days are paid regularly.
Data on the purchase of medications was collected from the central computerized system from all pharmacies of Clalit Health Services. In Israel the purchase of antibiotics is only permitted with a prescription by a physician.
Statistical analysis
Data analysis was performed using SPSS 15 (SPSS, Inc., Chicago, IL, USA). The association between the categorical dependant variables, i.e. antibiotic use and utilization of health care services with categorical participant characteristics and behaviour parameters (i.e. gender, education, age (≤ 45 years and > 45 years), marital status, co-morbidities, fever, purulent phlegm, expected antibiotics, and satisfaction) were examined using chi-square tests. Multivariate logistic models with a stepwise procedure were used to test for the effect of participant characteristics and behaviour on antibiotic use and utilization of health care services respectively. Odds ratios and 95% confidence intervals were calculated from the models. All P-values were two-sided, and statistical significance was defined as a P < 0.05.
RESULTS
Four hundred and sixty one patients were eligible for the study; 76 of them were not contacted by telephone. Of 385 patients who were reached by telephone, 47 declined an interview and 338 (overall response rate 73%) agreed to an interview (). The mean age of the participants was 47.3 ± 18.6 years, 70% were female, and nearly 50% had co-morbidities (). Data were collected on all chronic diseases as defined by the GPs and grouped according to physiologic systems. Hypertension, diabetes mellitus, and heart diseases were analysed separately.
Figure 1. Flow chart of participants’ inclusion and their use of health care services. aNine participants visited doctors in and outside the HC.
Table 1. Socio-demographic and disease-related characteristics of the study population.
Use of health care services
The participants initially visited the general practice after an average of five days following onset of the cough. Cough was often accompanied by other complaints (). During the telephone interview two weeks after the first visit, 113 (33.4%) of the participants reported that they still felt ill. Sickness reports for longer than three days were given to 140 of the employed participants (78%). The rate of work absenteeism of more than three days was similar in patients who had received versus had not received antibiotics (40.7% versus 42.3%, P = 0.88).
Table 2. Morbidity characteristics of the study population.
One hundred and fourteen patients (33.7%) had sought further care: 83 (24.6%) from their physicians within the HC and 40 (11.8%) outside the HC. Of this latter group, 22 consulted physicians after hours, seven visited the hospital emergency room, five consulted private physicians, and six did not provide details. Nine patients sought further care both within and outside the HC. None of the participants who went to the hospital emergency room were hospitalized.
Use of health care services after the first visit was associated with older age, an initial expectation to receive antibiotic treatment, and feeling ill two weeks after the first visit ().
Table 3. Characteristics of participants who used health care services after the first visit.
Antibiotic prescribing within and outside the HC
At the initial visit antibiotics were prescribed to 56 participants (16.6%). At the repeat visits within the HC (n = 83), antibiotics were prescribed to eight additional patients (2.4%). A report from the pharmacy centre confirmed that antibiotics were purchased by all of the participants for whom they were prescribed. In the validity check of the medical records we found that 48 of 50 antibiotics were correctly included and 46 of 50 antibiotics were correctly excluded. The inclusion errors were that the participants were included later from initial visit (at their second visit to the practice).
Twenty seven participants (8%) received antibiotics outside the HC: 5 (1.5%) from the after-hours physician; 4 (1.2%) at the hospital emergency room; 16 (4.7%) took antibiotics based on their decision; and two participants did not provide an explanation. The number of participants who took antibiotics against the advice of a physician was too small for analysis,
At the end of the follow-up period of two weeks after the initial visit, 91 patients (27%) had received antibiotics: 64 of them (70.3%) within the HC (56 at the first visit and eight during later visits), and 27 of them (29.7%) from sources outside the HC.
Participants who had received antibiotics within the HC were compared to participants who had not received antibiotics (). The logistic regression analysis shows that being divorced or widow(er), expectations to be treated by antibiotics and use of health services were associated with antibiotic prescribing by GPs.
Table 4. Logistic regression analysis of variables associated with receiving antibiotics within the Health Centre, adjusted odds ratio (OR) with 95% confidence interval (CI).
Use of non-antibiotic medications
At the first visit, 31 patients were prescribed both antibiotics and symptomatic therapy (9.2%) and 25 patients (7.4%) received antibiotics only. Symptomatic therapy alone was prescribed to 244 participants (72.2%) and 38 participants (11.2%) received no medications. Two hundred and four (63%) participants purchased over-the-counter (OTC) medications for symptom relief. Purchase of OTC medications was not associated with antibiotic prescriptions. More participants who had received symptomatic therapy from their physicians bought additional OTC medications than those who had not (192/305 (63%) versus 8/24 (33.3%), P = 0.004). The three most frequently prescribed medications for symptom relief were cough medications, analgesics, and nose drops: 351/544 (64.5%); 104/544 (19.2%); and 78/544 (14.3%) respectively. The order of use frequency for OTC drugs was analgesics, nose drops, and cough medications: 207/497 (41.6%); 111/497 (22.3%); and 110/497 (22.1%) respectively.
DISCUSSION
Our study shows that the rate of antibiotic prescribing for acute cough at the first visit did not reflect prescribing for the entire duration of the acute cough episode. The first visit to the GPs was not the last visit for one-fourth of the participants. As more physicians were visited, the rate of antibiotic prescribing increased.
Antibiotic prescribing within and outside the HC
Most of the participants complained of a cough accompanied by other symptoms and came to the physician after several days of disease. Physicians prescribed antibiotics more often to patients who expected antibiotics, but the overall rate of prescribing antibiotics exceeded the rate of expectation.
It is known that physicians prescribe antibiotics more often when they believe that the patients expect to receive them, (Citation17,Citation18) but they often misinterpret their patients’ expectations (Citation8). Patients have been found to believe that antibiotics are indicated in more severe disease (Citation19). Patients who visited the practice after five days of acute respiratory symptoms may well believe that antibiotics are effective at that stage of the disease (Citation20). As has been reported in previous publications, antibiotic therapy did not lead to a shorter course of illness (Citation3,Citation12) and did not reduce work absenteeism. In addition, antibiotic prescribing did not contribute to participant satisfaction from the visit.
Visits to the general practice and other health services
What made the participants return for medical advice after the first visit? It is possible that unmet expectations led them to visit further physicians. The proportion of participants who initially received antibiotics was not lower in participants who sought further care. Indeed, still feeling ill was associated with the subsequent utilization of health services for acute cough after the first visit. An alternative explanation for the repeated use of health care services and medications can be provided by a search for symptom relief, and not necessarily by the search for antibiotics. This is strengthened by the data on the purchase of additional OTC medications.
The interpretation of the study results in relation to the existing literature
Patients with prolonged cough sometimes take matters in their own hands, as follows: revisiting the general practice; consulting other physicians outside the general practice; purchasing medications for symptom relief; or taking antibiotics on their own. Use of antibiotics for viral infections is common, and antibiotics are mainly prescribed by medical staff (Citation21). Patients taking antibiotics without the advice of a physician have been described previously. Patients have reported keeping left-over antibiotics for future use (Citation22) and have used the antibiotics for exchange within the family and between friends (Citation23); such use is suspected as one cause of antibiotic resistance (Citation24).
Strengths and limitations of the study
Our efforts to avoid giving the GPs extra work led to some limitations of the study. The enrolment of participants in the study by the GPs could contribute to bias in the study population. Collecting data retrospectively from patient interviews could contribute to recall bias relating to the expectation of participants to receive antibiotics and satisfaction with the visit. This was the case in our study: conducting the telephone interview two weeks after the first visit enabled us to collect data on subsequent visits, but was subject to recall bias. The patients were enrolled by five GPs. Although having similar patient populations by age and chronic disease distribution, as well as a similar experience in practice, variation in the behaviour of physicians could bias the results.
The study was conducted in winter 2004. According to recent publications on the issue of antibiotic use in acute respiratory diseases it is still relevant. In spite of the time passed since the start of the study, our data is still relevant and contribute to existing knowledge by including the sequence of events after the patient left the physician's office. We tried not to intervene in the judgment of GPs, but to focus on the outcomes of the behaviour of patients. The data on the extended use of OTC medications and initiating antibiotic treatment by self-decision point at the importance of patients initiative during a prolong acute disease.
Implications for clinical practice
A recent Cochrane review stressed the possible harm in the use of antibiotics for acute respiratory infections (Citation12). Antibiotic prescribing in Israel is higher than in western European countries (Citation25). Our study described a local limited experience, which cannot be readily generalized to Israeli or European populations. Our study has added knowledge to the issue of antibiotic misuse by describing the behaviour of patients during the period with an acute cough. The study provided satisfying results compared to the literature with respect to the rates of antibiotic prescriptions. The interviews added data emphasizing the importance of patient behaviour, indicating a need to put greater stress on patient education.
CONCLUSION
The management decisions of physicians were often not the final act during episodes of acute respiratory disease presenting as a cough. Patients continue to act by using more medications, including antibiotics, and health care services. Our results indicated a need for more emphasis on educating the patient.
ACKNOWLEDGEMENT
The authors should like to thank the GPs in Migdal Ha’Emek Health Centre for their cooperation.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Related Research Data
REFERENCES
- Wenzel RP, Edmond MB. Managing antibiotic resistance. N Eng J Med. 2000;343:1961–3.
- Costelloe C, Metcalfe C, Lovering A, Mant D, Hay AD. Effect of antibiotic prescribing in primary care on antimicrobial resistance in individual patients: Systematic review and meta-analysis. Br Med J. 2010;340:c2096.
- Butler CC, Hood K, Verheij T, Little P, Melbye H, Nattall J, et al. Variation in antibiotic prescribing and its impact on recovery in patients with acute cough in primary care: Prospective study in 13 countries. Br Med J. 2009:338:b2242.
- Coenen S, Van Royen P, Vermeire E, Hermann I, Denekens J. Antibiotics for coughing in general practice: A qualitative decision analysis. Fam Pract. 2000;17:380–5.
- Belongia EA, Schwartz B. Strategies for promoting judicious use of antibiotics by doctors and patients. Br Med J. 1998;317: 668–71.
- Chan CS. What do patients expect from consultations for upper respiratory tract infections? Fam Pract. 1996;13:229–35.
- Coenen S, Michiels B, Van Royen P, Van der Auwera JC, Denekens J. Antibiotics for coughing in general practice: A questionnaire study to quantify and condense the reason for prescribing. BMC Fam Pract. 2002;3:16.
- Snell LM, Wilson RP, Oeffinger KC, Sargent C, Chen O, Corey KM. Patient and physician explanatory models for acute bronchitis. J Fam Pract. 2002;51:1035–40.
- Coenen S, Michiels B, Renard D, Denekens J, Van Royen P. Antibiotic prescribing for acute cough: the effect of perceived patient demand. Br J Gen Pract. 2006;56:183–90.
- Cals JW, Boumans D, Lardinois RJ, Gonzales R, Hopstaken RM, Butler CC, et al. Public beliefs on antibiotics and respiratory tract infection: an internet-based questionnaire study. Br J Gen Pract. 2007;57:942–7.
- Snow V, Mottur-Pilson C, Gonzales R. Principles of appropriate antibiotic use for treatment of acute bronchitis in adults. Ann Intern Med. 2001;134:518–20.
- Smith SM, Fahey T, Smucny J, Becker LA. Antibiotic treatment for people with a clinical diagnosis of acute bronchitis. The Cochrane Libr 2012; DOI: 10.1002/14651858.CD000245.pub2.
- Hueston WJ, Hopper JE, Dacus EN, Mainous AG. Why are antibiotics prescribed for patients with acute bronchitis? A postintervention analysis. J Am Board Fam Med. 2000;13:398–402.
- Gonzales R, Steiner JF, Maselli J, Lum A, Barrett PH Jr. Impact of reducing antibiotic prescribing for acute bronchitis on patient satisfaction. Eff Clin Pract. 2001;4:105–11.
- Gonzales R, Steiner JF, Lum A, Barrett PH Jr. Decreasing antibiotic use in ambulatory practice: impact of a multidimensional intervention on the treatment of uncomplicated acute bronchitis in adults. J Am Med Assoc. 1999;281:1512–9.
- Coenen S, Van Royen P, Michiels B, Denekens J. Optimizing antibiotic prescribing for acute cough in general practice: A cluster-randomized controlled trial. J Antimicrob Chemother. 2004;54:661–72.
- Yoy JH, Yau B, Choi KC, Chau CT, Huang QR, Lee SS. Public knowledge, attitudes and behavior on antibiotic use: A telephone survey in Hong Kong. Infection 2008;36:153–7.
- Macnamara J, Harrington P, Walsh M, Burke M, Walsh N. Antibiotics for sore throat: Impact of feedback to patients on the probability of bacterial infection. Ir Med J. 2000;93:211–2.
- Ong S, Nakase J, Moran GJ, Karras DJ, Kuehnert MJ, Talan DA. Antibiotic use for emergency department patients with upper respiratory infection; prescribing practices, patient expectations, and patient satisfaction. Ann Emerg Med. 2007,50:213–20.
- Weiss MC, Deave T, Peters TJ, Salisbury C. Perception of patient expectation for antibiotic: A comparison of walk-in centre nurses and GP's. Fam Pract. 2004;21:492–9.
- Eurosurveillance editorial team. Eurobarometer on antimicrobial resistance highlights areas for action. Eurosurveill. 2010;15: pii–19540.
- McNulty CA, Boyle P, Nichols T, Clappison P, Davey P. The public's attitudes to and compliance with antibiotics. J Antimicrob Chemother. 2007;60(Suppl. 1):i63–8.
- Tsiligiani IG, Delgatty C, Alegakis A, Lions C. A household survey on the extent of home medication storage. A cross- sectional study from rural Crete, Greece. Eur J Gen Pract. 2012; 18:3–8.
- Dryden MS, Cooke J, Davey P. Antibiotic stewardship—more education and regulation not more availability? J Antimicrob Chemother. 2009;64:885–8.
- Coenen S, Adriaenssens N, Versporten A, Muller A, Minalu G, Faes C, et al. ESAC Project Group European Surveillance of Antimicrobial Consumption (ESAC): Outpatient use of tetracyclines, sulphonamides and trimethoprim, and other antibacterials in Europe (1997–2009). J Antimicrob Chemother. 2011;66(Suppl. 6):vi57–70.