Quality in Primary Care Open Access

  • ISSN: 1479-1064
  • Journal h-index: 27
  • Journal CiteScore: 6.64
  • Journal Impact Factor: 4.22
  • Average acceptance to publication time (5-7 days)
  • Average article processing time (30-45 days) Less than 5 volumes 30 days
    8 - 9 volumes 40 days
    10 and more volumes 45 days
Reach us +32 25889658

Research Article - (2014) Volume 22, Issue 4

Guidelines adherence to lower urinary tract infection treatment in out-of-hours primary care in European countries

Hilde Philips

Department of Primary and Multidisciplinary Care, General Practice, University of Antwerp, Belgium

Linda Huibers

Radboud University Medical Centre, Nijmegen, Scientific Institute for Quality of Healthcare, Netherlands

Elisabeth Holm Hansen

National Centre for Emergency Primary Health Care, Uni Research, Uni Health, Norway; Associate Professor, Haraldsplass Deaconess University College

Morten Bondo Christensen

Research Unit and Department for General Practice, Aarhus University, Denmark

Ru¨ diger Leutgeb

Department of General Practice and Health Services Research, University of Heidelberg, Germany

Zalika Klemenc-Ketis

Department of Family Medicine, Medical School, University of Maribor, and Department of Family Medicine, Medical School, University of Ljubljana, Slovenia

Corinne Chmiel

Institute of General Practice and Health Services Research University of Zurich, Switzerland

Miguel-Angel Mun˜oz

Primary Healthcare University Research Institute IDIAP-Jordi Gol and Catalan Institute of Health, Barcelona, Spain

Katarzyna Kosiek

Department of Family and Community Medicine, Medical University of Lodz, Poland

Roy Remmen

Department of Primary and Multidisciplinary Care, General Practice, University of Antwerp, Belgium

*Corresponding Author:
Hilde Philips
Department of Primary and Multi-disciplinary Care
General Practice, University of Antwerp, Universiteitsplein 1
gebouw R, 3de Verd., B-2610 Wilrijk, Belgium.
Email: Hilde.philips@uantwerpen.be

Received date: 11 April 2014 Accepted date: 15 July 2014

Visit for more related articles at Quality in Primary Care

Keywords

after hours care, anti-infective agents,  guideline adherence, primary healthcare, urinary tract infection

How this fits in with quality in primary care

What do we know?

National health systems support guidelines on many clinical issues. LUTIs are the most common infections in primary care and out-of-hours (OOH) services, and a substantial reason for prescribing antibiotics. Many countries have adopted large-scale general practitioner-led services, responsible for the provision of OOH care.

What does this paper add?

OOH services are a convenient setting to study guideline adherence and offers a basis for comparison between countries. Even though European guidelines on the treatment of LUTIs are very similar, adherence differs substantially between European countries.

Introduction

Lower urinary tract infections (LUTIs) account for about 2% of the workload during regular daytime practice.[1,2] In out-of-hours (OOH) primary care 5.8% of diagnoses are urological (International Classifi-cation of Primary Care (ICPC) 2 chapter ‘U’) and 2% of all diagnoses concern LUTIs (ICPC code U71).[3] The prevalence of urological complaints in OOH care does not seem to depend on the type of setting, as an equal prevalence of urological complaints was found in an emergency department (5.1%) and a general practitioners’ cooperative (GPC) (5.3%).[4] Being the second most common bacterial infections in general practice, LUTI is a frequent reason for prescribing antibiotics.[5]

Practice guidelines are common tools for improv-ing the quality of healthcare in numerous (European) countries. In general, primary care associations have established evidence-based guidelines for diagnosis and treatment of uncomplicated LUTIs. Theoretically, only variation in type and resistance of bacteria and availability of medication should alter treatment strat-egies. There is some evidence that differences in guideline adherence exist, in particular concerning the prescribing behaviour of antibiotics.[6]

Guideline adherence has been studied in relation to prescription and use of antibiotics regarding treat-ment of uncomplicated LUTIs, because of the risk of bacterial resistance and patient safety.[7,8] Adherence to LUTI guidelines has been assessed for daytime care where suboptimal guideline adherence has shown to be less cost-effective in the treatment of uncom-plicated LUTIs.[914] In OOH care, first research results on guideline adherence in Belgium were published in 2012.[15] This study found that a simple, multifaceted intervention improved quality of antibiotic prescrib-ing for LUTI.

International comparison of LUTI guideline adher-ence will increase insight into performance and pro-vide knowledge for potential improvement. Observing differences in guideline adherence in regions of dif-ferent countries might help us learn more about reasons or causes for differences in prescribing behav-iour and guideline adherence itself. Comparison be-tween different settings also provides information about feasible and realistic goals in guideline adher-ence. Many countries have been starting large-scale OOH primary care services.[16,17] These services offer a convenient platform to study behaviour of GPs as they tend to collect routine data in large databases.[3] Fur-thermore, OOH care entails specific challenges due to the acute character of healthcare problems. In most countries the same clinical guidelines are used during daytime practice as during OOH care. Therefore, we studied the feasibility of assessing guideline adherence using routine data from OOH services.

Methods

Study design

We performed a cross-sectional observational multi-centre study. For the preparation of this study, we collected national LUTI guidelines from the nine participating European countries and compared con-tents. Secondly, we measured guideline adherence to LUTI treatment in the participating OOH primary care services, by comparing the actual performance with the country specific guideline, using routine data.

Settings and study population

EurOOHnet is a European research network for OOH

primary healthcare, which aims to study OOH pri-mary care. (www.euroohnet.eu).[18,19] As we did a convenience sample, all nine EurOOHnet members were invited to participate and all representatives were prepared to do so. The participating countries were Belgium, Denmark, Germany, the Netherlands, Norway, Poland, Slovenia, Spain and Switzerland. For the first study part, these national representatives completed a structured form reporting the content of their national guidelines for diagnosis and treatment of uncomplicated LUTI in primary healthcare, if possible specifically for OOH care. We invited each of these countries to participate in the observational part of the study.

In the second part, we aimed to collect data on at least 100 consecutive LUTI contacts from one or more regional OOH primary care services per participating country. Data originated from regular electronic, registration systems of the OOH primary care services, recording patient age, type of contact (i.e. telephone consultations, clinic/practice consultation, home visit), reported symptoms during medical history-taking, clinical examination, technical examination (e.g. ni-trite test), test results, and treatment provided (i.e. type of antibiotic, dose, and duration). Participation to this part of the study is voluntary. Details of the services are provided in Table 1. We included contacts of women aged 20 to 80 years, who were diagnosed with a first episode of LUTI (ICPC2 code U71).[20] These inclusion criteria (gender, age) were based on the target patient group mentioned in the guidelines for uncomplicated LUTI. Consequently, men, preg-nant women, children, patients with recurrent epi-sodes of LUTI (based on the patient’s story), and patients with a suspected complicated LUTI were excluded.

Data analysis

We performed a descriptive analysis of defined el-ements of the national LUTI guidelines (diagnosis and treatment). Consequently, we assessed guideline ad-herence with national recommendations for diagnosis and treatment of uncomplicated LUTI. To assess guideline adherence on diagnosis, we used registration on ‘medical history-taking’, ‘clinical examination’, and ‘technical examination’ (dipstick, laboratory testing). Due to differences in registration of diagnostic criteria (Table 2), these data were only used descriptively. Most registration systems only contained data on symp-toms, clinical and technical examinations, in case they were positive. For this reason, we could not distin-guish between symptoms or results which were not registered (but present) and symptoms which were absent (but examined). To estimate guideline adherence for treatment, we used two measures: ‘recommended type of medication’ and ‘recommended dose and duration’. ‘Full guideline adherence on treatment’ was present if type of medication and dose and duration were in accordance with the guidelines recommen-dations. For Belgium and Germany we were not able to estimate the ‘full guideline adherence on treat-ment’, because the dose and duration were not regis-tered in these regions.

This study was approved as an international multi-centre study by the ethical committee of the University of Antwerp (reference number A10–60). If required, participating countries applied for and became ap-proval of local ethical committees as well.

Results

Guidelines in participating countries

Most countries had national guidelines which are commonly used in primary care. We included guide-lines from Belgium,[21] Denmark,[22] Germany,[23] the Netherlands,[24] Norway,[25] Slovenia,[26] Spain and Switzerland (Table 2).[27,28] In Poland, no national guidelines exist, but they used the Scottish Intercol-legiate Guidelines Network (SIGN) guideline approach.[29] In Spain, GPs use different Spanish guidelines. For the purpose of this paper, we considered the guidelines from Fisterra and from the Catalan Association of Family and Community Medicine (in the region of Catalonia), which are most commonly used in the region included in our study. Overall, the nine countries used nine different guidelines, which were published be-tween 2000 (Belgium) and 2010 (Switzerland). Only Denmark had a specific guideline for OOH primary care.

Comparison of guidelines: diagnosis and treatment of LUTI

For all guidelines the diagnosis of LUTI is based on the history-taking, needing one or two of the typical symptoms (e.g. dysuria, pollakisuria (= frequent and small amounts)) (Table 2). Only two national guide-lines (Belgium, Slovenia) recommended clinical examination with finding of painful suprapubic pal-pation contributing to diagnose LUTI. In general, additional technical examination was not recommended in case of a typical history of LUTI. A dipstick with nitrite test was often mentioned in case of unclear history (i.e. Belgium, Germany, The Netherlands, Slovenia, and Switzerland).

In six guidelines, trimethoprim was the ‘drug of choice’, sometimes in combination with sulfamethoxazol or sulfamethizol. Nitrofurantoin or sulfamethizol in mono-therapy were first choice in the others (Table 3). Treatment with trimethoprim was always recom-mended for three days, mostly 300 mg a day. For nitrofurantoin the duration of treatment varied from three to seven days, with daily doses between 150 mg and 300 mg.

Guideline adherence

Due to organisational problems, three representatives (Poland, Spain, and Switzerland) were not able to deliver electronic data within the deadline. The other representatives provided data of at least 100 OOH patient contacts, except Slovenia due to insufficient data at deadline (n = 54). Two representatives delivered more data than requested (the Netherlands (n = 494, registration in three regional OOH services), and Norway (n = 196, registration in two regional OOH services) (Table 4).

Guideline adherence for diagnosis

Although most guidelines did not recommend ad-ditional clinical or technical examination, these were frequently performed. Technical examinations were performed in 16% (Denmark) to 99% (Norway) of the contacts (Belgium 42%, the Netherlands 91%, Slovenia 72%, no data in Germany, Poland, Spain, Switzerland).

Guideline adherence for treatment

Adherence to national guidelines for type of anti-biotics shows a variation between the national OOH services (Table 4). In Denmark and Norway guideline adherence with type of antibiotics was respectively 100% and 99.5%. In The Netherlands and Slovenia the adherence was above 70%, whereas in Belgium and Germany it was less than 40%. We were able to calculate the ‘full guideline adherence on treatment’ (i.e. recommended type, dose, and duration of treat-ment) for seven regions (in four countries). The Dutch regions had full adherence in 72.7% of the contacts, the Danish in 40.0%, the Norwegian regions in 38.3%, and the Slovenian in 22.2% of the contacts. Table 4 also shows us that in the Dutch regions, non-adherence mainly is caused by prescribing other types of antibiotics, whereas in the other regions, most non-adherence is found in dose and duration.

Discussion

Main findings

This study showed that it is possible to collect and use routine data from OOH services of different European regions. Our data showed that treatment adherence to the national guideline varied from 25% to 100% for type of antibiotics between regions of the participating countries and full guideline adherence on treatment also varied greatly, from 22% to 73%. Use of routine data opens opportunities for studying differences in prescribing behaviour and guideline adherence and may assist to improve prescribing behaviour of GPs.

Interpretation of findings on guideline adherence

Although data collection was feasible, coding and registration of variables differed between regions, which limited access to full valid data for some. However, our results on treatment match the con-clusions of recent national studies on primary care. In The Netherlands, prescribing antibiotics at GPCs during OOH is in accordance with guidelines in 69% to 71% of contacts.[30] A recent study in Belgium found only 34.5% guideline adherence for antibiotic treatment of patients with rhino-sinusitis in daytime care.[31] A Norwegian study showed that GPs followed national guidelines for UTI well (94% correct type of treatment), only the duration of treatment was often too short (in 32%).[32] In Spain, guideline adherence has shown to be low, with a low utilisation of first-choice antibiotics.[33] In Slovenia, guideline adherence was 72% in primary care settings, for the correct type of drug therapy.[34] These results of earlier research largely correspond to our results, which supports generalisability.

In our study, Denmark, The Netherlands, Norway, and Slovenia show greater adherence with treatment recommendations (for type of antibiotics), especially compared to Belgium and Germany. In the Dutch regions we found that once that type of antibiotic drug is correct, most GPs also prescribe it with respect to the guideline concerning dose and duration. This is less observed in the regions of other countries. In most other countries, the treatment duration is the item with the largest non-adherence (Table 4).

 Uncomplicated LUTI can be diagnosed using history-taking alone, in women without symptoms suspicious for complications. Guidelines do not recommend clinical or technical examinations (e.g. dipstick tests) in case of a typical history. Nevertheless, additional examinations were performed in 16% (Denmark) to 99% (Norway) of OOH contacts. Physicians probably perform a clinical examination during a face-to-face contact to rule out complicated LUTI or to comply with patients’ expectations. This might be unnecess-arily time-consuming or even superfluous, when we notice that in Denmark and The Netherlands un-complicated LUTIs are often treated in telephone consultations. When the patient is seen on a face-to-face contact, chances are great that she will be examined. Whether or not a patient will have a doctor’s contact depends on different explanations: possibility of telephone consult, practice nurses who treat the patients for LUTI, and telephone triage.

Differences in registration have influenced our results concerning history-taking, clinical and techni-cal examination. As most registration systems ‘in our study’ only contained data on symptoms which were positively confirmed by the patient, we could not distinguish between symptoms which were not regis-tered (but present) and symptoms which were absent (but asked). For clinical and technical examinations the same problem was encountered. In our database, however, we also found that GPs tend to record negative signs, when they are important to rule out complications (i.e. back pain). Consequently, as symp-toms and examinations probably are under-regis-tered, we can presume that our results are a minimal estimate of the tests performed. Therefore, we suspect that routine OOH data appear to be not suitable for valid decisions concerning guideline adherence for diagnostic testing in case of LUTI, unless obligatory registration fields are implemented in the software systems.

Recommendations

A prospective study design, although they tend to overestimate adherence, including a uniform regis-tration of clinical data, will be necessary to decide upon guideline adherence for diagnosis of LUTI during OOH. Routine registrations are often feasible for data gathering on treatment, but to assess diag-nostic adherence to guidelines (using data on history-taking) other study designs are necessary.

Differences in healthcare systems should be con-sidered, because these also seem to influence treat-ment routines (e.g. possibility of a telephone consult, free access, gatekeepers’ role of GP, the role of emerg-ency departments, tradition of guideline-based working). Therefore, when analysing availability and implemen-tation of guidelines, a comparison of healthcare sys-tems is crucial. In the regions of some countries, quasi full adherence already exists, and identification of contributing factors might be helpful to implement system changes in countries with lower adherence. In this respect, benchmarking of countries might be a strong incentive for quality improvement.[35]

Limitations

This European study, on guideline adherence in OOH primary care, including different regions of nine countries in the EurOOHnet network. Although par-ticipation was voluntary, selection bias is a concern as interested regions might perform better on regis-tration of data. Three OOH services were not able to deliver valid data for the study and only three from nine countries who expressed interest, were able to produce the required amount of data. This illustrates that differences in OOH services between regions and countries limits the external validity of the results. For this reason, comparisons between countries have to be interpreted with caution.

Moreover, concluding upon guideline adherence on a national level is not possible as we only had selected regional data. On the other hand, our results seem to match data from other research, suggesting that our findings are valid and not that different from daytime care settings.

Differences in registration systems in the partici-pating regions is an issue of special attention. Some electronic registration systems include fields that are obligatory to fill out. For instance in The Netherlands several signs and symptoms are listed and the caregiver has to check each box. Most other patient records record history-taking and clinical or technical exam-ination results in free text, leading to large differences in registration.

Conclusions

Using OOH routine data to assess guideline adherence for LUTI treatment is feasible. Moreover it seems to provide an opportunity to perform national, compar-ative, observational research. Provided some points of attention in registration methodology, even inter-national comparative studies will be feasible. The setting of OOH care is less suitable for interpreting guideline adherence on history-taking and diagnostic testing, due to variability in registration. Well moti-vated key persons who encourage correct registration by GPs and uniform software packages might improve quality of data during OOH.

We found relevant differences in applying the recommendations, although the content of the guide-lines was to a large extent comparable. This highlights the room for improvement for guideline adherence in some countries.[15]

List of abbreviations

LUTI: lower urinary tract infection

EurOOHnet: European research network for out-of-hours primary care

OOH: out-of-hours GP: general practitioner

GPC: general practitioner cooperative U71: acute cystitis (ICPC2 diagnostic code)

ICPC: International Classification of Primary Care

Acknowledgements

We thank Armando Casado and Jose Luis del Val for their contribution in collecting guideline content from Spain, Janko Kersnik from Slovenia for his help in conducting the study in Slovenia, Rok Macek et al from Slovenia for their help in collecting data, and Gunter Laux for his contribution in collecting and processing data from Germany.

References

Funding

This study was funded by the University of Antwerp, General Practice Unit, Department of Primary and Multidisciplinary Care.

Ethical Approval

This study was approved as an international multi-centre study by the ethical committee of the University of Antwerp (reference number A10–60). If required, participating countries applied for and became ap-proval of local ethical committees as well.

Author's Contributions

HP contributed to the conception and design of the study, coordinated the data collection, analysed and interpreted the data, and drafted the manuscript. LH participated in the planning of the study, collected national data, interpreted the data, and drafted the manuscript. EHH and MBC participated in the plan-ning of the study, collected national data, and revised the article. CC, RL and ZKK collected national data and revised the article. All authors contributed to collecting the content of the guidelines and national data. All authors have read and approved the final version of the manuscript. RR contributed to the conception and design of the study and drafted the manuscript.

Peer Review

Not commissioned; externally peer reviewed.

Conflicts of Interest

None.