You are here

  • Home
  • Archive
  • Volume 8, Issue 3
  • Do NHS GP surgeries employing GPs additionally trained in integrative or complementary medicine have lower antibiotic prescribing rates? Retrospective cross-sectional analysis of national primary care prescribing data in England in 2016

Article Text

Article menu

Download PDFPDF

XML
General practice / Family practice
Research
Do NHS GP surgeries employing GPs additionally trained in integrative or complementary medicine have lower antibiotic prescribing rates? Retrospective cross-sectional analysis of national primary care prescribing data in England in 2016
  1. Esther T van der Werf1,
  2. Lorna J Duncan1,
  3. Paschen von Flotow2,
  4. Erik W Baars3,4
  1. 1 Department of Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK
  2. 2 Sustainable Business Institute (SBI), Oestrich-Winkel, Germany
  3. 3 Louis Bolk Institute, Bunnik, The Netherlands
  4. 4 University of Applied Sciences, Leiden, The Netherlands
  1. Correspondence to Dr Esther T van der Werf; esther.vanderwerf{at}bristol.ac.uk

Abstract

Objective To determine differences in antibiotic prescription rates between conventional General Practice (GP) surgeries and GP surgeries employing general practitioners (GPs) additionally trained in integrative medicine (IM) or complementary and alternative medicine (CAM) (referred to as IM GPs) working within National Health Service (NHS) England.

Design Retrospective study on antibiotic prescription rates per STAR-PU (Specific Therapeutic group Age–sex weighting Related Prescribing Unit) using NHS Digital data over 2016. Publicly available data were used on prevalence of relevant comorbidities, demographics of patient populations and deprivation scores.

Setting Primary Care.

Participants 7283 NHS GP surgeries in England.

Primary outcome measure The association between IM GPs and antibiotic prescribing rates per STAR-PU with the number of antibiotic prescriptions (total, and for respiratory tract infection (RTI) and urinary tract infection (UTI) separately) as outcome.

Results IM GP surgeries (n=9) were comparable to conventional GP surgeries in terms of list sizes, demographics, deprivation scores and comorbidity prevalence. Negative binomial regression models showed that statistically significant fewer total antibiotics (relative risk (RR) 0.78, 95% CI 0.64 to 0.97) and RTI antibiotics (RR 0.74, 95% CI 0.59 to 0.94) were prescribed at NHS IM GP surgeries compared with conventional NHS GP surgeries. In contrast, the number of antibiotics prescribed for UTI were similar between both practices.

Conclusion NHS England GP surgeries employing GPs additionally trained in IM/CAM have lower antibiotic prescribing rates. Accessibility of IM/CAM within NHS England primary care is limited. Main study limitation is the lack of consultation data. Future research should include the differences in consultation behaviour of patients self-selecting to consult an IM GP or conventional surgery, and its effect on antibiotic prescription. Additional treatment strategies for common primary care infections used by IM GPs should be explored to see if they could be used to assist in the fight against antimicrobial resistance.

  • antimicrobial resistance
  • complementary medicine
  • integrative medicine
  • antibiotic prescription rate

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/

https://doi.org/10.1136/bmjopen-2017-020488

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • Use of National Health Service (NHS) Digital data on antibiotic prescription per Specific Therapeutic group Age–sex weighting Related Prescribing Unit (STAR-PU) provided a comprehensive insight into the prescribing practices of total antibiotics, and for respiratory tract infection and urinary tract infection separately in conventional General Practice (GP) surgeries and GP surgeries employing general practitioners (GPs) additionally trained in integrative medicine (IM GPs).

    • NHS England IM GP surgeries were comparable to conventional GP surgeries in terms of list sizes, demographics, deprivation scores and comorbidity prevalence.

    • Accessibility of IM/complementary and alternative medicine (CAM) within the NHS in General Practice in England is very limited. IM/CAM provision is currently almost exclusively private in the UK.

    • Results are limited by the lack of data on (1) number of consultations, (2) individual GP characteristics, (3) individual deprivation scores and (4) continuum of care.

    Introduction   

    Antimicrobial resistance (AMR) and the inappropriate use of antibiotics represent a serious threat to public health internationally.1 Antibiotics are currently indispensable throughout the healthcare system, and the consequences of AMR, not only in primary care, but also in major surgery and cancer treatment for example, are dire. Fortunately, reductions in antibiotic use have been shown to be associated with a reduction in some resistance,2 and the reduction in the use of antibiotics, especially in primary care, to control the development of AMR is therefore a pressing national and international priority.1 3

    In the UK, 74% of antibiotics are prescribed in primary care making this one of the most important contributors to the development of AMR.4 National Institute for Health and Care Excellence (NICE) guidelines on respiratory tract infection5 (RTI) management advise that a non-antibiotic prescribing strategy or a delayed antibiotic prescribing strategy should be considered for patients with the following conditions: acute otitis media, acute sore throat/acute pharyngitis/acute tonsillitis, common cold, acute rhinosinusitis and acute cough/acute bronchitis. Nevertheless, antibiotics are commonly prescribed for RTIs in adults and children in primary care, and are the reason for 60% of all antibiotic prescribing in general practice in the UK.6 Several studies have shown that there is substantial overprescribing of antibiotics for, often viral and self-limiting, RTIs in primary care.7–9 Consequently, there is a large potential to reduce antibiotic prescribing for RTIs, potentially by using other treatment strategies that do not increase the development or spread of AMR. Urinary tract infections (UTIs) are the most common confirmed bacterial infection, with about half of all women experiencing one of more UTIs in their lifetime.10 Most women with UTIs are currently treated with antibiotics, with longer duration and multiple courses associated with higher AMR rates.11 12 For example, NICE guideline on uncomplicated UTIs in women advise offering symptom relief and an antibiotic to all women with a suspected UTI. It states that for a woman with mild symptoms who has normal immunity, normal renal function and a normal renal tract, treatment can be delayed if the patient wishes to see if symptoms will resolve without treatment. For all other women treatment needs to start without delay.13 Therefore, overprescribing seems to be much less common for UTIs than for RTIs.14 Consequently, it may be easier to safely reduce antibiotic prescribing for RTIs than for UTIs in primary care.

    There is great variability in the use of antimicrobial medications between countries, with the lowest prescription rates reported in northern European nations, and higher rates in southern Europe and the USA.15–17 Variations in the prescription of antibiotics both within and between countries may indicate poor practice18 with inappropriate use of antibiotics which increases the risk of adverse events for the patient,12 wastes healthcare resources19 and contributes to the rise in antibiotic resistance.16 Previous studies have shown that a complex array of factors influence antibiotic prescribing, which may explain the wide variety of antibiotic usage both at the clinician level and worldwide. The attitudes of both doctor20–22 and patient23–25 are shown to be of major significance in prescribing decisions.

    General practitioners (GPs) as a professional group are expected, following the principles of evidence-based medicine, to apply best available evidence to patient’s individual situation, within the framework of national and local funding and administrative guidance. Prescription style (measured as the prevalence of prescriptions per GP) is found to be an important factor in the variation in antibiotic prescribing behaviour.18 Underlying factors for this finding might be differing views on medicalisation, differences in guidelines between countries and between specialists and knowledge and use of complementary and alternative medicine (CAM)/integrative medicine (IM).26–28 The association between the knowledge/use of IM/CAM by GPs and antibiotic prescribing has so far not been widely scrutinised.

    The aim of this study is to determine the differences in antibiotic prescription rates between ‘conventional General Practice (GP) surgeries and GP surgeries employing GPs additionally trained in IM/CAM (hereafter referred to as integrative medicine (IM) GPs), and the association between having staff with an additional training in IM/CAM and antibiotic prescription (measured as total antibiotics, RTI-specific antibiotics and UTI-specific antibiotics) within the National Health Service (NHS) in England.

    Methods

    IM GP surgeries

    To identify NHS General Practices employing an IM GP in 2016, two sequential tasks were required. First, IM GPs were identified and then a current working link was made to an NHS General Practice.

    Identification of NHS GPs registered with an additional CAM qualification

    We included the ‘Big 5’ CAM therapies as defined in a report by the House of Lords in 200029 (chiropractic, osteopathy, acupuncture, herbal medicine and homoeopathy); as well as Anthroposophic Medicine (AM). AM is an extension of conventional medicine and incorporates a holistic approach to people and nature and to illness and healing and is established in 80 countries, mostly in Central Europe.30 AM has been included so that future comparison with data in other European countries is possible.

    In the UK each of these six therapies is either state regulated (osteopathy and chiropractic) or has voluntary regulation (including a voluntary regulatory body for mainstream healthcare practitioners in the case of acupuncture and homoeopathy). The regulatory bodies were therefore initially approached by email to check the best route of establishing which practitioners on their registers were trained as a conventional GP and trained in IM/CAM as well. Details of the organisations and methods by which IM GPs were identified are indicated in table 1. Where organisations were not able to provide this information, searches were made of the online registers (between May and June 2017) for the following:

    View this table:
    Table 1

    Organisations from which details of training in integrative medicine or complementary and alternative medicine (IM/CAM) of General Practitioners (GPs) were obtained

    1. Location: the registers enabled us to search for practitioners either on a nationwide (England/UK) basis or by county (in the latter case, all English counties were checked, including recent boundary changes).

    2. Qualification: in some databases, only healthcare professionals (eg, practitioners qualified in biomedicine) were included, and GP status was specified. Where this was not the case, titles were given and the absence of the title ‘Dr’ was used to exclude practitioners from our study. Where qualifications were also identified, practitioners with PhDs but not being a GP could be excluded. And if no additional information was given on the register, online searches were made of the practitioner to establish their professional qualifications. In all cases, putative IM GPs were checked against the General Medical Council (GMC) register, to confirm whether they were currently permitted to work in medical practice in the UK.

    Identification of NHS General Practices employing IM GPs

    Practice location(s) were indicated for each GP registrant identified in the CAM registers indicated in table 1. These workplaces were then checked against both the practice websites and the NHS website (www.nhs.uk), which lists all NHS practices and gives information such as staff lists. This acted to confirm the working location of the GP and whether this practice offered provision within NHS England.

    Data

    Monthly prescribing data were obtained from NHS Digital. NHS Digital collect data and information about a wide range of General Practice (GP) services, for many different organisations and purposes. It also collates all primary care prescribing data. These data are released for monthly download via the NHS Digital website (http://digital.nhs.uk/searchcatalogue). Data are released at the specific healthcare provider level and volumes are provided by full British National Formulation code (BNF code). To determine mean antibiotic prescription rates, we used the total number of oral antibiotic prescriptions per general practice for the most recent calendar year for which antibiotic prescribing was publicly available via the NHS Digital website (January 2016 to December 2016).

    Specific Therapeutic group Age–sex weighting Related Prescribing Units

    Specific Therapeutic group Age–sex weighting Related Prescribing Units (STAR-PUs) have been used as the denominator instead of the number of registered patients as STAR-PUs allow more accurate and meaningful comparisons within a specific therapeutic group by considering the types of people who will be receiving that treatment. The amount of STAR-PU per practice was estimated by multiplying the number of patients in each age–gender category.31

    Comorbidities

    The prevalence of various comorbidities that may adversely influence the outcome of infections, based on the conditions that indicate high-risk patients who qualify for the free seasonal influenza vaccination programme,32 was also measured to identify potential case-mix differences that may explain different antibiotic prescribing rates. General practice-specific prevalence is available for the following high-risk comorbidities via the Quality Outcomes Framework indicators at the NHS Digital website: asthma, cancer, chronic kidney disease, chronic obstructive pulmonary disease, heart failure and diabetes. We extracted comorbidity prevalence for the financial year 2015–2016.

    Indices of multiple deprivation

    Previously it has been shown that indices of multiple deprivation are indicators of poor health in a population. Tosas Auguet et al 33 for example, found that more deprived areas are at higher risk of methicillin-resistant Staphylococcus aureus infection. The most recent index of multiple deprivation was calculated in 2015 and is available from the Department for Communities and Local Government.34 Deprivation scores are available at a lower-layer super output area (LSOA), which consists of approximately 1500 residents each. Linkage of the data from NHS Digital was performed using a lookup table from NHS Digital. Where a practice served multiple LSOAs, the average deprivation score for that practice was calculated, weighted by the number of patients in each LSOA.

    The final dataset included only practices that were present in both the comorbidity and deprivation, and antibiotic prescribing files. We removed outliers based on practice size, since there were some doubts about the validity of these data (eg, a practice with 157 patients registered). We removed the outer 2% of data based on practice size.

    Statistical analysis

    All analyses are performed on GP surgery level (hereafter referred to as GPs). Potential differences in antibiotic prescribing rates per STAR-PU between the IM and conventional GPs were evaluated. NHS Digital defines a prescription item as: ‘a prescription item is a single supply of a medicine, dressing or appliance written on a prescription form’. If a prescription form includes three medicines, it is counted as three prescription items. We tested for between group differences using a random effects meta-analysis model for proportions (R package ‘meta’). For continuous variables like the number of STAR-PU per practices Mann-Whitney U tests were used to test for statistically significant differences.

    The association between the IM GPs and antibiotic prescribing rates per STAR-PU was assessed using negative binomial regression models with the number of antibiotic prescriptions as the outcome and the natural logarithm of the number of STAR-PU per practice as an offset. A negative binomial regression model was used as this type of regression model can handle count data (number of antibiotic prescriptions), accounts for differences in the number of antibiotics purely caused by practice size (by including the offset) and can still provide valid results when the variance in antibiotic use does not equal the mean antibiotic use.35 Both crude results and results adjusted for additional variables to correct for potential confounding are presented.

    We additionally evaluated associations between IM GPs and antibiotics commonly used for RTI (amoxicillin, amoxicillin and enzyme inhibitor, ampicillin, clarithromycin, doxycycline, erythromycin and phenoxymethylpenicillin) and for UTI (cephalexin, cefixime, ciprofloxacin, nitrofurantoin, pivmecillinam and trimethoprim).36

    Results

    Identification of NHS IM GPs in England

    Eight hundred and fifty CAM practitioner records were checked against the various CAM registers (table 1) to identify 21 GPs who are conventionally trained as a GP and also trained in CAM at 19 NHS GP surgeries in England (table 2). It should be noted that the 850 registrants were not the total numbers of CAM practitioners on the registers, as (as previously described) some registers permitted limiting searches to medical practitioners, or more specifically to medically qualified doctors.

    View this table:
    Table 2

    Integrative medicine (IM) General Practice (GP) surgeries subdivision (based on website information)

    While some practices were publicly working in an IM structure, in others it was unknown whether the identified GP was using IM/CAM or indeed whether or the practice permitted this, as no mention was made of it on the websites (including no reference to privately available CAM clinics). General Practices were therefore classified into two subsets as indicated in table 2. There were further IM GPs on the register who were either not practising (or at least not in England), or working in private practice. As any level of IM/CAM activity in subset 2 practices could not be determined, it was decided to exclude these GP surgeries from further analysis. Nine NHS IM GP surgeries (urban (n=6), semiurban (n=2) and semirural (n=1)) were included in the analysis. table 2 also shows the CAM therapies for which each IM GP in our subsequent analysis was registered. Each practice included has at least one IM GP, as GP partner or salaried. The number of IM GPs per practice varies from a minimum of 1 IM GP in a practice with 12 GPs of whom 6 part time, to a maximum of 3 IM GPs (all GP practice partners) in a practice with 7 GPs (two full-time GPs and five part time GPs).

    Antibiotic prescription rates

    In total 7283 NHS England General Practices (N_conventional=7274/N_IMGPs=9) were included in the analyses. Table 3 presents the baseline characteristic of the NHS IM GP surgeries compared with those characteristic of NHS conventional GPs in England. It also shows the prevalence of various comorbidities that may adversely influence the outcome of infections and may consequently influence antibiotic prescribing. The patient populations of both kinds of practices were comparable for most of the listed comorbidities. Statistically significant differences (P<0.05) between the IM GP- and conventional GP surgery patient population were found in the percentage of patients with coronary heart disease, cancer and diabetes, although absolute differences were relatively small. No statistical differences (P<0.05) were found in deprivation score between IM GP surgeries and conventional GP surgeries.

    View this table:
    Table 3

    Baseline characteristics of included NHS General Practice (GP) surgeries

    IM GP surgeries and antibiotic prescription rates

    Overall within the NHS in England, the median prescription rates of IM GP surgeries were lower for ‘any antibiotic’ and for ‘RTI-specific antibiotic’ compared with the rates of the conventional GP surgeries over 2016, while the median prescription rates of ‘UTI-specific antibiotic’ per STAR-PU were comparable for the two groups (table 4).

    View this table:
    Table 4

    Median antibiotic prescription rates and RR of prescribing antibiotics in primary care England over 2016

    View this table:

    The relative risks (RRs) in table 4 were obtained using negative binomial regression models with the number of antibiotic prescriptions as the outcome and taking into account differences in practice sizes. Our analysis shows that IM GP surgeries were associated with lower prescriptions of ‘any antibiotic’ (RR 0.78, 95% CI 0.64 to 0.97) and with lower prescriptions of ‘RTI-specific antibiotic’ (RR 0.74, 95% CI 0.59 to 0.94*). Patients consulting an IM GP surgery were 22% less likely to get ‘any antibiotic’ prescription compared with those who consulted a conventional GP surgery. Receiving a RTI-specific antibiotic prescription was 26% less likely among those who consulted an IM GP surgery compared with those who consulted a conventional GP surgery. No statistically significant difference (P<0.05) was found in the number of prescriptions of UTI-specific antibiotic prescriptions per STAR-PU between IM GP surgeries and conventional GP surgeries within the NHS in England.

    Adjustment for deprivation score or diabetes resulted in virtually identical results. For ‘any antibiotic’ if adjusted for deprivation score the RR for IM GP surgeries remains virtually identical (RR 0.78, 95% CI 0.64 to 0.97), and for diabetes (RR 0.80, 95% CI 0.65 to 0.99). This was the case regardless of which confounder was added to the model, however adjusting for multiple potential confounders was not possible due to the small number of cases.

    Discussion

    There were 7283 NHS England General Practices included in our analyses. Despite the very small proportion of IM GP surgeries, our data show that significantly fewer ‘total antibiotics’ and ‘RTI-specific antibiotics’ per STAR-PU were prescribed at IM GP surgeries compared with conventional GP surgeries within NHS England over 2016. No statistically significant differences were found in median prescription rates of ‘UTI-specific antibiotics’ per STAR-PU in the two kinds of NHS GP surgeries.

    This is the first (retrospective) study comparing antibiotic prescribing rates between IM GP surgeries and conventional GP surgeries in England. However, the small proportion of NHS IM GPs in England asks for careful interpretation of the results. Accounting for one other variable (eg, deprivation or diabetes) did not change our results, but, due to the low number of cases it was not possible to similarly account for more variables.

    Lack of information on the number of consultations is the main limitation of this study. Consultation rates may explain most of the variation in antibiotic prescribing, and with the data used in this study, it is not clear whether patients consulting at IM GP surgeries consulted less in general. If this were the case, they may be equally likely to receive antibiotics when they do consult their GP. Besides, previous studies show that the consultation rate is also dependent on the previous likelihood to receive antibiotics for a RTI.37–39 Future studies should therefore include consultation behaviour/number of consultations as a confounding factor.

    Other study limitations that need to be taken into account when interpreting the results are the lack of information on (1) individual deprivation scores, (2) individual GP characteristics and (3) continuum of care. First, no statistically significant differences were found in deprivation score on practice level. However, this may be partly because deprivation scores are area based and not based on the individuals registered at the different practices. Second, our analyses are at GP practice level and include information on GP practice characteristics, such as list size and population. Data on GP characteristics on individual level are not part of NHS Digital data and are therefore not included in our analysis. However, these GP characteristics may partly explain the variation in antibiotic prescribing.40–42 Finally, the data used for our analysis are based on the number of antibiotic ‘prescriptions’, which may differ from the numbers of antibiotics ‘consumed’, and do not include information on the continuum of care of patients (eg, hospital admissions/reconsultation). Future studies using clinical practice data taking continuum of care into account are warranted.

    The lower antibiotic prescription rates of IM GPs are in line with the current national guidance aimed at reducing antibiotic usage and AMR.1 IM GPs might possible comply more closely with this guidance. However, the difference found could also be partly explained by the fact that patients who consult IM GPs might demand less for antibiotics, and that GPs in IM surgeries have other avenues to offer to patients than antibiotics or that they are more confident to delay prescriptions and to assert themselves against the wishes of those patients who appear to want antibiotics. Our results are in line with a yet unpublished pilot study in The Netherlands in which the prescription of antimicrobials for systemic use in 23 unselected anthroposophic GP surgeries was compared with the national mean GP figures for the years 2012–2014. On average AM GPs in the Netherlands prescribed less antimicrobials: −13% (2012), –10% (2013) and −7% (2014) (unpublished data).

    Despite the differences we found in RTI antibiotics, no statistically significant differences were found in UTI antibiotics prescription rates between the two kinds of NHS GP surgeries. Although it should be borne in mind that the use of prescribing data to infer the type of infection may be prone to errors, our finding reflects current UK GP clinical guidance.

    For the majority of RTIs it is recommended that antibiotics should be avoided or delayed, so that this is an area where the desired reduction in prescribing could take place. In the case of UTIs, antibiotics are advised more readily.36 For several RTIs, including common colds, sore throat, sinusitis and acute bronchitis, randomised controlled trials (RCTs) have shown that antibiotics provide no, or negligible benefit compared with placebo.43–46 As such, symptom management with paracetamol, ibuprofen or the use of CAM therapies proven to be effective and safe for RTIs may safely reduce antibiotic prescribing among patients with a low risk for pneumonia. A recent RCT comparing ibuprofen with fosfomycin treatment for UTIs indicates that it may be more difficult to safely reduce antibiotic prescribing for UTIs using a similar approach.47

    Our study shows that accessibility of IM/CAM within the NHS in General Practice is very limited, and this limited the number of IM GPs included in our analysis. IM/CAM provision is currently almost exclusively privately provided in the UK, which could be at least partly linked to the austerity measures currently imposed in the UK possibly resulting in NHS IM/CAM provision being withdrawn at a local level in recent years (and imminently nationally). It was seen on practice websites that there was a noticeable amount of ‘private’ CAM provision available at several NHS surgeries by non-GP CAM practitioners—in weekly clinics for example. However, these surgeries have not been included in our analysis as this study specifically aimed to determine the differences in antibiotic prescribing between conventional GP surgeries and GP surgeries staffing GPs who were additionally trained in IM/CAM.

    Attitudes of GPs to IM/CAM are extremely important for it to remain available within the NHS. A 2015 study of these attitudes in England, showed that, despite demand for CAM among the general public, GPs remain concerned about its limited evidence base as well as the lack of regulation of CAM practitioners. Nevertheless, those questioned continue to see a role for CAM in clinical practice.48

    In our study each of the NHS IM GP practices included at least 1 IM GP. However, as the number of (IM) GP partners and salaried (IM) GPs (full time and part time) in these practices varies, proportions of the number of IM GPs per included NHS IM GP surgery are difficult to determine and in addition, will not provide meaningful information as they do not take into account the power balance of the different GPs within these practices.

    The impact which any one IM GP could have in terms of antibiotic prescribing may vary hugely between practices partly depending on their status at the practice—as a partner or a salaried employee for example, or as a full-time or part time worker. In the presented analysis we did not include NHS GP practices that are offering NHS IM/CAM provision by a ‘non-GP NHS CAM practitioner’ or private IM/CAM practitioner. However, having even one CAM contact within a surgery might give the possibility for others to experience CAM perspectives either formally or informally from them, and for long-held attitudes to be perhaps modified. Additionally, it would be of interest to explore if patients may be independently accessing IM GPs in the private sector and then seeking antibiotics from non-IM GPs in the NHS.

    In line with Hawker et al 49 our results suggest that a further decrease in prescribing in conventional surgeries might be possible. It may be that advice should be sought from this small number of surgeries to establish whether their daily clinical practice may differ from other surgeries and whether this could be used to assist others in the fight against AMR.

    Although this study found only a small number of CAM practitioners working at NHS General Practices in England, the difference seen in antibiotic prescribing rates at IM GP surgeries warrants further study. It is very likely that, due to similarity in healthcare system (NHS) and the number of NHS IM GPs and—surgeries, our findings are generalisable to Scotland, Wales and Northern Ireland. However, differences in healthcare systems and the general level of IM practice may lead to other results in other (European) countries. Therefore, analysis in comparison with other (European) countries are indicated to prove whether in general IM GP (surgeries) tend to prescribe less antibiotics or whether other (socioeconomic) factors dominate the prescription rate for specific infections.

    As the clear majority of CAM practitioners (mainly non-GPs) work privately in England, there is also potential for research into non-antibiotic strategies in private practice, and to analysis of how these practitioners work with their patients’ NHS GPs in this regard. However, as patients who self-select to consult IM GPs might be less likely to demand antibiotics, differences in lifestyle and the ‘transferability of lifestyle skills’ need to be taken into account as well in future study design.

    Acknowledgments

    The authors gratefully acknowledge the statistical support of Dr Koen Pouwels (Senior Mathematical/Economic Modeller).

    References

    1. 1.↵
      Public Health England. English surveillance programme for antimicrobial utilisation and resistance (ESPAUR). 2016 https://phe-newsroom.prgloo.com/resources/espaur-report-2016
    2. 2.↵
      2. Butler CC ,
      3. Dunstan F ,
      4. Heginbothom M , et al
      . Containing antibiotic resistance: decreased antibiotic-resistant coliform urinary tract infections with reduction in antibiotic prescribing by general practices. Br J Gen Pract 2007;57:785–92.
      OpenUrlAbstract/FREE Full Text
    3. 3.↵
      2. Laxminarayan R ,
      3. Duse A ,
      4. Wattal C , et al
      . Antibiotic resistance-the need for global solutions. Lancet Infect Dis 2013;13:1057–98.doi:10.1016/S1473-3099(13)70318-9
      OpenUrlCrossRefPubMedWeb of Science
    4. 4.↵
      Public Health England. English surveillance programme for antimicrobial utilisation and resistance (ESPAUR. London: Public Health England, 2016.
    5. 5.↵
      NICE. Respiratory tract infections: prescribing of antibiotics for self-limiting respiratory tract infections in adults and children in primary care, 2008.
    6. 6.↵
      2. Lindbaek M
      . Prescribing antibiotics to patients with acute cough and otitis media. Br J Gen Pract 2006;56:164-6.
      OpenUrl
    7. 7.↵
      2. Dekker AR ,
      3. Verheij TJ ,
      4. van der Velden AW
      . Inappropriate antibiotic prescription for respiratory tract indications: most prominent in adult patients. Fam Pract 2015;32:401–7.doi:10.1093/fampra/cmv019
      OpenUrlCrossRefPubMed
    8. 8.↵
      2. Fleming-Dutra KE ,
      3. Hersh AL ,
      4. Shapiro DJ , et al
      . Prevalence of inappropriate antibiotic prescriptions among us ambulatory care visits, 2010-2011. JAMA 2016;315:1864–73.doi:10.1001/jama.2016.4151
      OpenUrlCrossRefPubMed
    9. 9.↵
      2. Flöistrup H ,
      3. SWARTZ J ,
      4. BERGSTROM A , et al
      . Allergic disease and sensitization in Steiner school children. J Allergy Clin Immunol 2006;117:59–66.doi:10.1016/j.jaci.2005.09.039
      OpenUrlCrossRefPubMedWeb of Science
    10. 10.↵
      2. Foxman B
      . Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Dis Mon 2003;49:53–70.doi:10.1067/mda.2003.7
      OpenUrlCrossRefPubMed
    11. 11.↵
      2. Bryce A ,
      3. Hay AD ,
      4. Lane IF , et al
      . Global prevalence of antibiotic resistance in paediatric urinary tract infections caused by Escherichia coli and association with routine use of antibiotics in primary care: systematic review and meta-analysis. BMJ 2016;352:i939.
      OpenUrlAbstract/FREE Full Text
    12. 12.↵
      2. Costelloe C ,
      3. Metcalfe C ,
      4. Lovering A , et al
      . Effect of antibiotic prescribing in primary care on antimicrobial resistance in individual patients: systematic review and meta-analysis. BMJ 2010;340:c2096.doi:10.1136/bmj.c2096
      OpenUrlAbstract/FREE Full Text
    13. 13.↵
      NICE. Urinary tract infection (lower) - women - summary. Clinical knowledge summaries, 2015.
    14. 14.↵
      Public health England. Management and treatment of common infections. Antibiotic guidance for primary care: for consultation and local adaptation, 2017.
    15. 15.↵
      2. Goossens H ,
      3. Ferech M ,
      4. Vander Stichele R , et al
      . Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet 2005;365:9–587.doi:10.1016/S0140-6736(05)70799-6
    16. 16.↵
      2. Goossens H
      . Antibiotic consumption and link to resistance. Clin Microbiol Infect 2009;15(Suppl 3):12–15.doi:10.1111/j.1469-0691.2009.02725.x
      OpenUrlCrossRefPubMedWeb of Science
    17. 17.↵
      2. Goossens H ,
      3. Ferech M ,
      4. Coenen S , et al
      . Comparison of outpatient systemic antibacterial use in 2004 in the United States and 27 European countries. Clin Infect Dis 2007;44:1091–5.doi:10.1086/512810
      OpenUrlCrossRefPubMedWeb of Science
    18. 18.↵
      2. Cordoba G ,
      3. Siersma V ,
      4. Lopez-Valcarcel B , et al
      . Prescribing style and variation in antibiotic prescriptions for sore throat: cross-sectional study across six countries. BMC Fam Pract 2015;16:7.doi:10.1186/s12875-015-0224-y
      OpenUrlCrossRefPubMed
    19. 19.↵
      2. Sirovich B ,
      3. Gallagher PM ,
      4. Wennberg DE , et al
      . Discretionary decision making by primary care physicians and the cost of U.S. Health care. Health Aff 2008;27:11.doi:10.1377/hlthaff.27.3.813
      OpenUrlFREE Full Text
    20. 20.↵
      2. Carthy P ,
      3. Harvey I ,
      4. Brawn R , et al
      . A study of factors associated with cost and variation in prescribing among GPs. Fam Pract 2000;17:36–41.doi:10.1093/fampra/17.1.36
      OpenUrlCrossRefPubMedWeb of Science
    21. 21.↵
      2. Teixeira Rodrigues A ,
      3. Ferreira M ,
      4. Roque F , et al
      . Physicians' attitudes and knowledge concerning antibiotic prescription and resistance: questionnaire development and reliability. BMC Infect Dis 2016;16:7.doi:10.1186/s12879-015-1332-y
      OpenUrl
    22. 22.↵
      2. Teixeira Rodrigues A ,
      3. Roque F ,
      4. Falcão A , et al
      . Understanding physician antibiotic prescribing behaviour: a systematic review of qualitative studies. Int J Antimicrob Agents 2013;41:203–12.doi:10.1016/j.ijantimicag.2012.09.003
      OpenUrlCrossRefPubMed
    23. 23.↵
      2. Gualano MR ,
      3. Gili R ,
      4. Scaioli G , et al
      . General population’s knowledge and attitudes about antibiotics: a systematic review and meta-analysis. Pharmacoepidemiol Drug Saf 2015;24:2–10.doi:10.1002/pds.3716
      OpenUrlCrossRefPubMed
    24. 24.↵
      2. Linder JA ,
      3. Singer DE
      . Desire for antibiotics and antibiotic prescribing for adults with upper respiratory tract infections. J Gen Intern Med 2003;18:795–801.doi:10.1046/j.1525-1497.2003.21101.x
      OpenUrlCrossRefPubMedWeb of Science
    25. 25.↵
      2. Moro ML ,
      3. Marchi M ,
      4. Gagliotti C , et al
      . Why do paediatricians prescribe antibiotics? Results of an Italian regional project. BMC Pediatr 2009;9:69.doi:10.1186/1471-2431-9-69
      OpenUrlCrossRefPubMed
    26. 26.↵
      2. Grimaldi-Bensouda L ,
      3. Bégaud B ,
      4. Rossignol M , et al
      . Management of upper respiratory tract infections by different medical practices, including homeopathy, and consumption of antibiotics in primary care: the EPI3 cohort study in France 2007-2008. PLoS One 2014;9:e89990.doi:10.1371/journal.pone.0089990
    27. 27.↵
      2. Hamre HJ ,
      3. Glockmann A ,
      4. Schwarz R , et al
      . Antibiotic use in children with acute respiratory or ear infections: prospective observational comparison of anthroposophic and conventional treatment under routine primary care conditions. Evid Based Complement Alternat Med 2014;2014:1–17.doi:10.1155/2014/243801
      OpenUrl
    28. 28.↵
      2. Taylor JA ,
      3. Jacobs J
      . Homeopathic ear drops as an adjunct in reducing antibiotic usage in children with acute otitis media. Glob Pediatr Health 2014;1:2333794X1455939.doi:10.1177/2333794X14559395
      OpenUrl
    29. 29.↵
      House of Lords. In: Ho L , ed.  Science and technology - sixth report, 2000.
    30. 30.↵
      2. Kienle GS ,
      3. Albonico HU ,
      4. Baars E , et al
      . Anthroposophic medicine: an integrative medical system originating in europe. Glob Adv Health Med 2013;2:20–31.doi:10.7453/gahmj.2012.087
      OpenUrl
    31. 31.↵
      NHS Digital. Practice level prescribing in England: a summary. http://content.digital.nhs.uk/gpprescribingdata
    32. 32.↵
      Public Health England. Chapter 19. Influenza: the Green Book, 2015.
    33. 33.↵
      2. Tosas Auguet O ,
      3. Betley JR ,
      4. Stabler RA , et al
      . Evidence for community transmission of community-associated but not health-care-associated methicillin-resistant Staphylococcus aureus strains linked to social and material deprivation: spatial analysis of cross-sectional data. PLoS Med 2016;13:e1001944.doi:10.1371/journal.pmed.1001944
      OpenUrlCrossRefPubMed
    34. 34.↵
      Department for Communities and Local Government. The english indices of deprivation 2015, 2015.
    35. 35.↵
      2. Hilbe JM
      . Negative binomial regression. Cambridge, UK: Cambridge University Press, 2012.
    36. 36.↵
      2. McNulty CA
      . Management of infection guidance for primary care for consultation and local adaptation. London, UK: Public Health England, 2017.
    37. 37.↵
      2. Ashworth M ,
      3. Charlton J ,
      4. Ballard K , et al
      . Variations in antibiotic prescribing and consultation rates for acute respiratory infection in UK general practices 1995-2000. Br J Gen Pract 2005;55:603–8.
      OpenUrlAbstract/FREE Full Text
    38. 38.↵
      2. Little P ,
      3. Gould C ,
      4. Williamson I , et al
      . Reattendance and complications in a randomised trial of prescribing strategies for sore throat: the medicalising effect of prescribing antibiotics. BMJ 1997;315:350–2.doi:10.1136/bmj.315.7104.350
      OpenUrlAbstract/FREE Full Text
    39. 39.↵
      2. Moore M ,
      3. Little P ,
      4. Rumsby K , et al
      . Effect of antibiotic prescribing strategies and an information leaflet on longer-term reconsultation for acute lower respiratory tract infection. Br J Gen Pract 2009;59:728–34.doi:10.3399/bjgp09X472601
      OpenUrlAbstract/FREE Full Text
    40. 40.↵
      2. Brookes-Howell L ,
      3. Hood K ,
      4. Cooper L , et al
      . Understanding variation in primary medical care: a nine-country qualitative study of clinicians’ accounts of the non-clinical factors that shape antibiotic prescribing decisions for lower respiratory tract infection. BMJ Open 2012;2:e000796.doi:10.1136/bmjopen-2011-000796
      OpenUrlAbstract/FREE Full Text
    41. 41.↵
      2. Butler CC ,
      3. Hood K ,
      4. Verheij T , et al
      . Variation in antibiotic prescribing and its impact on recovery in patients with acute cough in primary care: prospective study in 13 countries. BMJ 2009;338:b2242.doi:10.1136/bmj.b2242
      OpenUrlAbstract/FREE Full Text
    42. 42.↵
      2. Silverman M ,
      3. Povitz M ,
      4. Sontrop JM , et al
      . Antibiotic prescribing for nonbacterial acute upper respiratory infections in elderly persons. Ann Intern Med 2017;166:765–74.doi:10.7326/M16-1131
      OpenUrl
    43. 43.↵
      2. Kenealy T ,
      3. Arroll B
      . Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database Syst Rev 2013;6:CD000247.doi:10.1002/14651858.CD000247.pub3
      OpenUrlCrossRefPubMed
    44. 44.↵
      2. Little P ,
      3. Stuart B ,
      4. Moore M , et al
      . Amoxicillin for acute lower-respiratory-tract infection in primary care when pneumonia is not suspected: a 12-country, randomised, placebo-controlled trial. Lancet Infect Dis 2013;13:123–9.doi:10.1016/S1473-3099(12)70300-6
      OpenUrlCrossRefPubMedWeb of Science
    45. 45.↵
      2. Smith SM ,
      3. Fahey T ,
      4. Smucny J , et al
      . Antibiotics for acute bronchitis. Cochrane Database Syst Rev 2017;6:CD000245.doi:10.1002/14651858.CD000245.pub4
      OpenUrl
    46. 46.↵
      2. Spinks A ,
      3. Glasziou PP ,
      4. Del Mar CB
      . Antibiotics for sore throat. Cochrane Database Syst Rev 2013;5:CD000023.
      OpenUrl
    47. 47.↵
      2. Gágyor I ,
      3. Bleidorn J ,
      4. Kochen MM , et al
      . Ibuprofen versus fosfomycin for uncomplicated urinary tract infection in women: randomised controlled trial. BMJ 2015;351:h6544.doi:10.1136/bmj.h6544
      OpenUrlAbstract/FREE Full Text
    48. 48.↵
      2. Jarvis A ,
      3. Perry R ,
      4. Smith D , et al
      . General practitioners’ beliefs about the clinical utility of complementary and alternative medicine. Prim Health Care Res Dev 2015;16:246–53.doi:10.1017/S146342361400022X
      OpenUrl
    49. 49.↵
      2. Hawker JI ,
      3. Smith S ,
      4. Smith GE , et al
      . 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 2014;69:3423–30.doi:10.1093/jac/dku291
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors Substantial contributions to the conception and design of the work: ETvdW/LJD/PvF/EWB. Acquisition, analysis or interpretation of data for the work: ETvdW/LJD. Drafting the work and revising it critically for important intellectual content: ETvdW/LJD/PvF/EWB. Final approval of the version to be published: ETvdW/LJD/PvF/EWB. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: ETvdW/LJD/PvF/EWB.

    • Funding This research was supported in part by an unrestricted grant of WALA Heilmittel GmbH, Germany.

    • Competing interests None declared.

    • Patient consent Not required.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement Data are available through NHS digital.