Research

Antibiotic prescribing and patient satisfaction in primary care in England: cross-sectional analysis of national patient survey data and prescribing data

Mark Ashworth, Patrick White, Hannah Jongsma, Peter Schofield and David Armstrong
British Journal of General Practice 2016; 66 (642): e40-e46. DOI: https://doi.org/10.3399/bjgp15X688105

Abstract

Background Concerns about adverse effects on patient satisfaction may be an important obstacle to attempts to curtail antibiotic prescribing.

Aim To determine the relationship between antibiotic prescribing in general practice and reported patient satisfaction.

Design and setting Retrospective cross-sectional study of general practices in England.

Method Data were obtained from the General Practice Patient Survey (GPPS) in 2012 (2.7 million questionnaires in England; 982 999 responses; response rate 36%); the national Quality and Outcomes Framework dataset for England, 2011–2012 (8164 general practices); and general practice and demographic characteristics. Standardised measures of antibiotic prescribing volumes were obtained for each practice in England during 2012–2013, together with 12 other nationally available prescribing variables. The role of antibiotic prescribing volume was identified as a determinant of GPPS scores and adjusted for demographic and practice factors using multiple linear regression.

Results The final dataset consisted of 7800 (95.5%) practices. A total of 33.7 million antibiotic prescriptions were issued to a registered population of 53.8 million patients. Antibiotic prescribing volume was a significant positive predictor of all ‘doctor satisfaction’ and ‘practice satisfaction’ scores in the GPPS, and was the strongest predictor of overall satisfaction out of 13 prescribing variables. A theoretical 25% reduction in antibiotic prescribing volume would be associated with 0.5–1.0% lower patient satisfaction scores, a drop of 3–6 centile points in national satisfaction ranking.

Conclusion Patients were less satisfied in practices with frugal antibiotic prescribing. A cautious approach to antibiotic prescribing may require a trade-off in terms of patient satisfaction.

INTRODUCTION

The cornerstone of good general practice has long been recognised as lying in the quality of the relationship between doctor and patient. This focus on the interaction between GP and patient has been further reinforced in recent years by increasing attention on the patient’s experience of healthcare encounters. Measures of this aspect of general practice activity have been elicited through patient surveys that invite patients to report on the care they have experienced from their local practice. This process has been formalised since 2007 in the annual General Practice Patient Survey (GPPS), which invites a representative sample of around 2.7 million adults registered with GPs in England to comment on the quality of their care.1 Not only does this survey provide feedback from patients to their GPs, but it has also been a component of the pay-for-performance Quality and Outcomes Framework and is currently monitored by commissioners and inspectors of primary care. GPs are incentivised to make the experience of care a good one.

However, pleasing the patient is not always consistent with providing good-quality care. GPs are well aware that patients may demand an antibiotic when it is not judged clinically appropriate. Sometimes it is the GP’s perception of that demand that seems important.2 In a survey of GPs by the organisers of the Longitude Prize (a national science prize in which the public voted for the most pressing issue facing humanity), 55% reported pressure to prescribe antibiotics, 45% had prescribed antibiotics for a viral infection knowing that they would be ineffective, and 44% admitted that they had prescribed antibiotics in order to get a patient to leave the consulting room.3 Spending more time with patients has been reported to result in fewer antibiotic prescriptions.4

Given the growing dangers from antibiotic-resistant bacteria,5 there have been many attempts to limit antibiotic prescribing in general practice. Following a national campaign in the late 1990s, GP antibiotic prescribing volumes declined by 21.7% from their peak in 1995.6 Since then, antibiotic prescribing has gradually increased with an overall 6% increase from 2010 to 2013.7 That there is scope for further reduction is suggested by evidence that many upper respiratory tract infections of a mainly viral origin continue to be treated with antibiotics, with one study reporting antibiotic prescriptions issued for 51% of patients presenting with coughs and colds.8

There may be a trade-off between the wish to nurture the doctor–patient relationship and antibiotic stewardship.4 This trade-off may be an important obstacle to the appropriate limitation of antibiotic prescribing, especially in the context of growing concerns about antibiotic resistance. Therefore, this study aimed to use the findings of the GPPS to determine whether there was an association between antibiotic prescribing and patient satisfaction in primary care. In particular, it aimed to determine whether patients registered at practices that prescribed fewer antibiotics reported lower levels of overall patient satisfaction.

How this fits in

This is the first study linking national patient experience survey data with prescribing patterns in primary care. The findings demonstrate that patients report lower levels of satisfaction if they are registered at practices that prescribe fewer antibiotics. Although observational studies cannot prove a causal relationship, these findings are consistent with other studies and suggest that frugal antibiotic prescribing is associated with modest reductions in patient satisfaction. GPs who wish to play their part in addressing issues of antibiotic resistance will need to consider alternatives to an antibiotic prescription that do not compromise patient satisfaction.

METHOD

Study design

A retrospective cross-sectional study was conducted using practice-level data from primary care databases. Regression models were constructed to identify the role of antibiotic prescribing as a determinant of patient experience (as elicited by the GPPS), adjusted for practice, structural, and demographic characteristics.

Antibiotic prescribing data

Antibiotic prescribing volumes were obtained for each practice in England during 2012–2013.9 The data were standardised such that differences in antibiotic script dose and duration were accounted for by using Average Daily Quantities (ADQs),10 and differences in the age/sex and temporary resident profile of the registered list of patients were also accounted for by using Specific Therapeutic group Age–sex Related Prescribing Units (STAR-PUs).11 The resultant measure, ADQ per STAR-PU, approximates to the number of days of antibiotic prescription at the usual prescribed dosage, which are given to a population-standardised ‘average’ patient registered at each practice.

General Practice Patient Survey

Unweighted GPPS data were obtained for the 2012 calendar year.12 During 2012, 2.7 million questionnaires were distributed, with an overall response rate of 36% (982 999 completed responses). Each year, the distribution of GPPS follows a similar pattern: questionnaires are sent to a random sample of adult patients registered with GPs in England; approximately one-half are sent in January and one-half in July, with follow-up questionnaires sent to non-responders in the subsequent 2 months. Purposive sampling compensates for lower response rates in some age, sex, practice, deprivation, and ethnicity cohorts.13

The 2012 GPPS version contained 41 questions eliciting patient experience relating to primary care, broadly covering the following domains: access, experience of GP care, experience of nurse care, out-of-hours care, and overall satisfaction. The focus of this study was on responses to the six GPPS questions in the ‘doctor care domain’ (Q21a–e, Q22) and to the two satisfaction questions: ‘overall satisfaction’ (Q28) and ‘recommending to someone else’ (Q29). Most questions eliciting patient experience offered five response options (such as ‘very good’, ‘good’, ‘average’, ‘poor’, and ‘very poor’). For the principal analysis, the two most positive response options (for example, ‘very good’ and ‘good’) were selected for each question as a percentage of all responses.

Practice data

Descriptive data were obtained for all general practices in England from the general medical services database, based on practice data obtained on 31 March 2012.9 Eight variables known to be related to patient satisfaction were obtained for inclusion in the regression models:14 registered patient list size, age and sex profile of registered patients, number of full-time equivalent GPs in each practice, the age and sex of GPs in each practice, the proportion of GPs who were non-UK qualified, and training practice status for the year 2011–2012.

Practice-level ethnicity and deprivation data for all registered patients are not available in England; therefore aggregate demographic data were used based on deprivation and ethnicity census data collated at lower layer super output area level for the practice postcode. Data were based on the 2011 national census and Index of Multiple Deprivation, IMD-2010.15,16 Three variables were obtained for inclusion in the regression models: social deprivation and the proportions of the local African Caribbean and South Asian populations.

Participants

A previously described method was used14 to match practice data (n = 8164 practices) with GPPS data (n = 8140 practices), and to exclude atypical practices with <750 patients or <500 patients per full-time equivalent GP. The initial selection consisted of 7959 practices (97.8% of all those submitting data in 2011–2012).

Analysis

A multivariable analysis was conducted to determine the relationship between patient experience factors and overall antibiotic prescribing volumes adjusted for practice and demographic factors. Separate regression models were constructed for the eight GPPS questions included in the study. GPPS scores were entered into each model as the outcome (dependent) variable.

Antibiotic prescribing volumes were entered into the regression models as one of the predictor variables. Eleven additional variables (eight demographic and three practice descriptors, as above) were tested in each model. Multicollinearity was tested for by calculating the variance inflation factor and excluding variables with variance inflation factor >10. To allow for the potential effects of multiple testing, predictor variables were only considered significant if P<0.01. The reported value for each regression model was the unstandardised B coefficient for each predictor variable. The analysis was conducted using SPSS (version 22).

Comparison with other prescribing indicators

To determine whether the volume of antibiotic prescribing was acting as a proxy for other prescribing variables, the analysis was repeated by collecting all other nationally available prescribing variables (n = 13, in addition to the antibiotic variable) and comparing the predictive power of each by adding them to the regression model for ‘overall satisfaction’ (Q28), as above. The prescribing variables were: net ingredient cost per 1000 registered patients (a measure of total prescribing cost for each practice); standardised volume of antidepressant hypnotic and antipsychotics prescribing (ADQ per STAR-PU); low-cost statins as a proportion of all statins; ezetimibe as a proportion of all lipid-lowering drugs; cephalosporins and quinolones as a proportion of all antibiotics; oral non-steroidal anti-inflammatory drugs (NSAIDs) per STAR-PU; ibuprofen and naproxen as a proportion of all NSAIDs; diclofenac and cyclo-oxygenase-2 (COX-2) inhibitors as a proportion of all NSAIDs; volume of antisecretory drugs and mucosal protectants per STAR-PU; inhaled corticosteroids per STAR-PU; and long- and intermediate-acting insulin analogues as a proportion of all insulins.9

The standardised regression coefficient, β, was used in order to test the additional prescribing indicators in the regression model. Standardised β regression coefficients were used in preference to unstandardised B coefficients, because many of the prescribing variables had differing unit denominators, making between-variable comparisons difficult to interpret. First, the β value was demonstrated for antibiotic prescribing volume as a predictor of overall satisfaction; second, the remaining prescribing indicators were added to the regression model and the β value of the additional prescribing indicators was compared with the value for antibiotic prescribing alone. Many of the prescribing variables were closely correlated with each other; therefore each regression model was restricted to a maximum of two prescribing variables (the antibiotic variable plus one of the other variables). Finally, the r2 value was derived to determine whether additional prescribing indicators strengthened the predictive power of the model. Since multiple testing was being conducted, the sensitivity analysis was confined to just one of the ‘doctor domain’ questions, ‘overall satisfaction’ (Q28).

RESULTS

Antibiotic prescribing volumes

The distribution of antibiotic prescribing volumes was skewed with several outliers. By omitting practices with the highest and lowest 1% of antibiotic prescribing volumes (n = 159), the distribution of antibiotic prescribing was normalised. All further analyses were conducted on this sample of 7800 practices (95.5% of all practices in England).

A total of 33 702 980 antibiotic prescriptions were issued to a registered population of 53 812 820. When both the volume of antibiotic prescribing and the registered population were standardised, the mean volume of antibiotic prescribing was 1.15 ADQs per antibiotic STAR-PU; with a standard deviation of 0.26 (Table 1). The mean number of antibiotic STAR-PUs per registered patient per practice was 2.15.

View this table:
Table 1.

Frequency distribution of antibiotic prescribing volume and patient experience scores in English general practices (n = 7800)

Predictors of positive patient experience scores

Positive responses to all GPPS questions included in this study, adjusted for other demographic and practice factors, were associated with higher levels of antibiotic prescribing: the regression coefficients B in the eight models ranged from 1.98 (95% CI = 1.43 to 2.53) to 4.07 (95% CI = 3.49 to 4.65) (Table 2). All models had a high predictive power ranging from an r2 value of 20.0% (Q21d) to 30.1% (Q29).

View this table:
Table 2.

Antibiotic prescribing volumes as a determinant of patient experience scores (based on General Practice Patient Survey questions)

The B regression coefficient allowed direct estimation of the degree to which patient experience varied according to the volume of antibiotic prescribing. Thus, for example, the regression coefficient for Q21a, the rating of the GP ‘giving you enough time’, was 3.24 (Table 2). A 25% reduction in antibiotic prescribing would be associated with a 0.81% (25% × 3.24) decrease in satisfaction for this item; this corresponded to a 5-point reduction in the national satisfaction rankings from the 50th to the 45th centile (Table 2).

Comparison with other prescribing indicators

In the regression model for overall satisfaction (Q28), antibiotic prescribing volume had the highest β value of all 14 prescribing variables included in the model (β = 0.08; P<0.001). All other prescribing variables had β values <0.06 (Table 3). The r2 value for the original regression model based on overall satisfaction (Q28) and including antibiotic prescribing as the only prescribing predictor variable was 27.7%; additional prescribing variables only increased the predictive power by 0.1–2.9%.

View this table:
Table 3.

Predictors of overall patient satisfaction (Question 28, General Practice Patient Survey): β values for prescribing variables included in regression modelsa

DISCUSSION

Summary

Antibiotic prescribing was a significant determinant of patient experience, as elicited by the GPPS, both for satisfaction with the GP (‘doctor domain’ questions) and, to a lesser extent, for satisfaction with the general practice (‘satisfaction domain’ questions). Reductions in patient experience scores associated with reduced antibiotic prescribing were noteworthy, especially when compared with relatively small variability in satisfaction scores. For example, practices prescribing 25% fewer antibiotics than the national mean could expect reductions in their patient satisfaction scores of between 0.5% and 1.0%. Although modest, these reductions in satisfaction were significant and consistent for all eight ‘patient satisfaction’ questions included in the GPPS. As practices have relatively similar satisfaction scores, this reduction in satisfaction has a bigger impact on the overall rankings, corresponding to a drop of 3–6 centile points in national satisfaction ranking.

Substantial reductions in antibiotic prescribing may be feasible.8 These findings suggest that practices responding to the prospect of widespread antibiotic resistance by reducing antibiotic prescribing are likely to experience reductions in their satisfaction ratings.

The comparison with other prescribing variables confirmed the importance of antibiotic prescribing as a determinant of patient satisfaction. Other prescribing variables were weaker predictors of satisfaction and added little to the overall predictive power of the regression models. The comparison may not be entirely valid because many of these prescribing variables would only have applied to a small minority of GPPS responders. Some, however, such as hypnotic or antidepressant indicators, or the total cost of all drugs prescribed, were more likely to be relevant to questionnaire responders.

Strengths and limitations

This study has demonstrated that patients registered at practices that prescribe fewer antibiotics report lower levels of overall satisfaction, and satisfaction with the care of the doctor. The regression models all had relatively high predictive power. However, a cross-sectional observational study cannot in itself demonstrate causality and corroborating evidence is needed from longitudinal evidence of practices making large changes in their prescribing patterns and from qualitative interview data. Aggregated practice-level GPPS data do not provide information on whether questionnaire responders had themselves been prescribed an antibiotic, nor whether they had personal experience of expecting an antibiotic that was not prescribed by the GP. It is estimated that just under one-third of the population are prescribed an antibiotic in any given year.17 It is likely, therefore, that most responders had not received an antibiotic in the preceding year. Antibiotic prescribing in itself may have acted as a proxy for some other aspect of the doctor–patient relationship related to satisfaction, such as the wish to please patients. Nevertheless, antibiotic prescribing was a far stronger determinant of satisfaction than the prescribing of, for example, hypnotic medication, which might also be considered a marker of the desire to please patients.

The antibiotic prescribing data obtained for this study were confined to prescriptions issued by the practice and thus omitted antibiotic prescriptions issued to registered patients attending an emergency department or out-of-hours centre. Omission of these data may have introduced confounding into the findings. It is possible that reduced volumes of antibiotic prescribing was a feature of practices with restricted access to appointments for patients who then obtained their supply of antibiotics elsewhere, and who reported subsequent dissatisfaction with their practice. A secondary analysis of the data was therefore conducted but no relationship was found between GP access (GPPS, question 12) and antibiotic prescribing volumes (results available from authors on request). It was not possible to obtain consultation rate data for each practice and it might be that practices with lower consultation rates may have ‘transferred’ their antibiotic prescribing for acute conditions to an emergency department or an out-of-hours centre.

Comparison with existing literature

This is the first national study to report the role of prescribing variables as determinants of patient experience. The associations remained when adjusted for confounding by other social and practice factors.

Decreased satisfaction has been reported following failure to prescribe antibiotics. In a study of 1014 patients with lower respiratory tract infections, patients who did not receive antibiotics that they had wanted were much more likely to express dissatisfaction.18 Patient satisfaction is often cited by GPs as a reason for prescribing antibiotics, although this perception is often overestimated.19 In another study of patients with acute bronchitis, treatment with antibiotics had no effect on patient satisfaction scores, whereas the perception of having been carefully examined was associated with increased patient satisfaction.20

Respiratory tract infections account for approximately one-half of all antibiotic prescribing in primary care and little is known about the determinants of patient satisfaction for antibiotic prescribing for other conditions. However, a Cochrane Review found no reduction in satisfaction for short courses of antibiotics in urinary tract infections.21 In a study of almost 7000 Swedish patients registered at 39 practices, those registered at practices with higher antibiotic prescribing rates (antibiotic prescribing volume per GP consultation) were more satisfied, whereas this was partially offset by lower antibiotic prescribing in practices where the GP spent more time listening to the patient.4

The current study’s findings differed and the GP characteristic of ‘listening’ was not associated with lower antibiotic prescribing. Differences in the way that the question on listening was phrased (the Swedish study emphasised the time spent listening rather than the attribute of listening) and between national and small-scale studies may account for conflicting findings.

Other studies have reported that more intensive interventions are associated with either enhancement of, or no reduction in, satisfaction following consultations for respiratory infections without an accompanying antibiotic prescription. Examples of these interventions include additional training in consultation skills,22 use of a patient booklet,22 or multiple interventions including group educational meetings, communication skills training, prescribing data feedback, pharmacist education, and educational material for patients.23 Although maintaining patient satisfaction is achievable with additional input in a research setting, it remains unclear whether these gains can be maintained in routine practice and for all types of infection.

Implications for research

Having demonstrated an association between antibiotic prescription and patient satisfaction, further research is needed to interpret the possible trade-off between these two factors. Although small-scale studies have shown that dissatisfaction about not receiving an antibiotic can be offset if the patient feels that they have been listened to or carefully examined,4 the extent to which this applies to all antibiotic prescribing is not known. This has particular importance to the growing problem of antibiotic resistance in primary care. GPs who wish to achieve more appropriate and targeted prescription of antibiotics will be supported by a better understanding of how patient satisfaction can be maintained in consultations when antibiotics are not prescribed.

Notes

Funding

This work received no external funding. This research was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health.

Ethical approval

Not applicable.

Provenance

Freely submitted; externally peer reviewed.

Competing interests

The authors have declared no competing interests.

Discuss this article

Contribute and read comments about this article: bjgp.org/letters

  • Received March 2, 2015.
  • Revision requested March 24, 2015.
  • Accepted June 5, 2015.

REFERENCES

  1. 1.↵
    1. NHS England
    GP patient survey. About the survey, http://www.gp-patient.co.uk/info/ (accessed 10 Nov 2015).
  2. 2.↵
    1. Coenen S,
    2. Michiels B,
    3. Renard D,
    4. et al.
    (2006) Antibiotic prescribing for acute cough: the effect of perceived patient demand. Br J Gen Pract 56(524):183–190.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    1. Cole A
    (2014) GPs feel pressurised to prescribe unnecessary antibiotics, survey finds. BMJ 349:g5238.
    OpenUrlFREE Full Text
  4. 4.↵
    1. Lundkvist J,
    2. Akerlind I,
    3. Borgquist L,
    4. Mölstad S
    (2002) The more time spent on listening, the less time spent on prescribing antibiotics in general practice. Fam Pract 19(6):638–640.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Davies SC
    (2013) Annual report of the Chief Medical Officer, volume two, 2011, infections and the rise of antimicrobial resistance (Department of Health, London).
  6. 6.↵
    1. Ashworth M,
    2. Latinovic R,
    3. Charlton J,
    4. et al.
    (2004) Why has antibiotic prescribing for respiratory illness declined in primary care? A longitudinal study using the General Practice Research Database. J Public Health (Oxf) 26(3):268–274.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Public Health England
    English surveillance programme for antimicrobial utilisation and resistance (ESPAUR): report 2014, https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/362374/ESPAUR_Report_2014__3_.pdf (accessed 12 Nov 2015).
  8. 8.↵
    1. Hawker JI,
    2. Smith S,
    3. Smith GE,
    4. et al.
    (2014) Trends in antibiotic prescribing in primary care for clinical syndromes subject to national recommendations to reduce antibiotic resistance, UK 1995–2011: analysis of a large database of primary care consultations. J Antimicrob Chemother 69(12):3423–3430.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Health and Social Care Information Centre
    Prescribing by GP practice, http://www.hscic.gov.uk/gpprescribingdata (accessed 18 Nov 2015).
  10. 10.↵
    1. Walley T,
    2. Roberts D
    (2000) Average daily quantities: a tool for measuring prescribing volume in England. Pharmacoepidemiol Drug Saf 9(1):55–58.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Lloyd D,
    2. Harris CM,
    3. Roberts DJ
    (1995) Specific therapeutic group age-sex related prescribing units (STAR-PUs): weightings for analysing general practices’ prescribing in England. BMJ 311(7011):991–994.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. NHS England
    The GP Patient Survey: January–September 2012, summary report, http://gp-survey-production.s3.amazonaws.com/archive/2012/December/weighted/Dec%202012%20National%20Summary%20Report.pdf (accessed 18 Nov 2015).
  13. 13.↵
    1. NHS England
    Frequently asked questions: What is weighing? Why do you weigh the data? http://www.gp-patient.co.uk/faq/weighting/ (accessed 10 Nov 2015).
  14. 14.↵
    1. Ashworth M,
    2. Schofield P,
    3. Durbaba S,
    4. Ahluwalia S
    (2014) Patient experience and the role of postgraduate GP training: a cross-sectional analysis of national Patient Survey data in England. Br J Gen Pract doi:10.3399/bjgp14X677545.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Office for National Statistics
    Who we are What we do How we live Welcome to the 2011 Census for England and Wales, http://www.ons.gov.uk/ons/guide-method/census/2011/index.html?utm_source=twitterfeed&utm_medium=twitter (accessed 10 Nov 2015).
  16. 16.↵
    1. Department for Communities and Local Government
    The English indices of deprivation 2010, http://www.communities.gov.uk/publications/corporate/statistics/indices2010 (accessed 12 Nov 2015).
  17. 17.↵
    1. Scottish Medicines Consortium,
    2. Scottish Antimicrobial Prescribing Group,
    3. NHS National Services Scotland
    Report on antimicrobial use and resistance in humans in 2012, http://www.isdscotland.org/Health-Topics/Prescribing-and-Medicines/Publications/2014-01-28/2014-01-28-SAPG-2012-Report.pdf (accessed 12 Nov 2015).
  18. 18.↵
    1. Macfarlane J,
    2. Holmes W,
    3. Macfarlane R,
    4. Britten N
    (1997) Influence of patients’ expectations on antibiotic management of acute lower respiratory tract illness in general practice: questionnaire study. BMJ 315(7117):1211–1214.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Britten N,
    2. Ukoumunne O
    (1997) The influence of patients’ hopes of receiving a prescription on doctors’ perceptions and the decision to prescribe: a questionnaire survey. BMJ 315(7121):1506–1510.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    1. van Duijn HJ,
    2. Kuyvenhoven MM,
    3. Schellevis FG,
    4. Verheij TJ
    (2007) Illness behaviour and antibiotic prescription in patients with respiratory tract symptoms. Br J Gen Pract 57(540):561–568.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    1. Lutters M,
    2. Vogt-Ferrier NB
    (2008) Antibiotic duration for treating uncomplicated, symptomatic lower urinary tract infections in elderly women. Cochrane Database Syst Rev (3):CD001535.
  22. 22.↵
    1. Tonkin-Crine S,
    2. Anthierens S,
    3. Francis N,
    4. et al.
    (2014) Exploring patients’ views of primary care consultations with contrasting interventions for acute cough: a six-country European qualitative study. NPJ Prim Care Respir Med 24:14026.
    OpenUrlPubMed
  23. 23.↵
    1. Welschen I,
    2. Kuyvenhoven MM,
    3. Hoes AW,
    4. Verheij TJ
    (2004) Effectiveness of a multiple intervention to reduce antibiotic prescribing for respiratory tract symptoms in primary care: randomised controlled trial. BMJ 329(7463):431.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

British Journal of General Practice: 66 (642)
British Journal of General Practice
Vol. 66, Issue 642
January 2016
Download PDF
Article Alerts
Email Article
Citation Tools

Antibiotic prescribing and patient satisfaction in primary care in England: cross-sectional analysis of national patient survey data and prescribing data
Mark Ashworth, Patrick White, Hannah Jongsma, Peter Schofield, David Armstrong
British Journal of General Practice 2016; 66 (642): e40-e46. DOI: 10.3399/bjgp15X688105
Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords

Advertisement

 
 

British Journal of General Practice