SEARCH

SEARCH BY CITATION

Keywords:

  • Antibiotics;
  • epidemiology;
  • point of care tests;
  • primary care;
  • Switzerland

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

Clin Microbiol Infect 2011; 17: 855–861

Abstract

The use of antibiotics is highest in primary care and directly associated with antibiotic resistance in the community. We assessed regional variations in antibiotic use in primary care in Switzerland and explored prescription patterns in relation to the use of point of care tests. Defined daily doses of antibiotics per 1000 inhabitants (DDD1000pd) were calculated for the year 2007 from reimbursement data of the largest Swiss health insurer, based on the anatomic therapeutic chemical classification and the DDD methodology recommended by WHO. We present ecological associations by use of descriptive and regression analysis. We analysed data from 1 067 934 adults, representing 17.1% of the Swiss population. The rate of outpatient antibiotic prescriptions in the entire population was 8.5 DDD1000pd, and varied between 7.28 and 11.33 DDD1000pd for northwest Switzerland and the Lake Geneva region. DDD1000pd for the three most prescribed antibiotics were 2.90 for amoxicillin and amoxicillin-clavulanate, 1.77 for fluoroquinolones, and 1.34 for macrolides. Regions with higher DDD1000pd showed higher seasonal variability in antibiotic use and lower use of all point of care tests. In regression analysis for each class of antibiotics, the use of any point of care test was consistently associated with fewer antibiotic prescriptions. Prescription rates of primary care physicians showed variations between Swiss regions and were lower in northwest Switzerland and in physicians using point of care tests. Ecological studies are prone to bias and whether point of care tests reduce antibiotic use has to be investigated in pragmatic primary care trials.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

There is growing evidence from observational studies that antibiotic overuse is an important risk factor for the emergence of antibiotic resistance [1,2]. In primary care, most antibiotics are prescribed for upper and lower respiratory tract and urinary tract infections. Antibiotic resistance has been recognized as an increasing public health problem. Therefore the European Surveillance of Antimicrobial Consumption project was established to collect data on outpatient antibiotic use in a uniform format [3]. Highly variable rates of antibiotic prescriptions were found, with higher rates of defined daily doses (DDD) in the Mediterranean countries, compared with those in the northern European countries. Striking seasonal fluctuations and higher rates of antibiotic resistance were observed in particular in high-prescribing countries [3]. Switzerland has one of the lowest rates of antibiotic consumption in Europe [4]. Contrary to many European countries, point of care tests are widely used by primary care providers in Switzerland [5]. First intervention trials indicate that point of care tests may contribute to a more diligent use of antibiotics in primary care [6,7]. However, there is conflicting evidence from epidemiological data of outpatient antibiotic use in relation to the use of such tests. Based on prescription records of the largest social health insurer, we investigated differences in outpatient antibiotic use in Switzerland in relation to regions, seasons and use of point of care tests.

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

We analysed reimbursement data of systemic antibiotic use in ambulatory care from 2007 of the Helsana Group, the largest social health insurer in Switzerland. This insurer covers, according to regions, between 11.5% and 22.8% of the Swiss population.

Information on drug prescriptions was received from invoices. Antibiotics were identified by the pharma code, an index for the identification of the drug, application form, doses and package size used in Switzerland and provided by e-mediat AG, Bern, Switzerland [8]. According to regions, between 95.8% and 97.9% of invoices are sent or processed electronically and centrally stored by the health insurer, guaranteeing high completeness of data for the particular calendar year. All antibiotics analysed have to be covered by the social health insurance, although individuals with high deductibles may not send their bills to the health insurer.

We considered only outpatient antibiotic use provided by self-dispensing physicians or pharmacies. We classified each antibiotic with the anatomic therapeutic chemical (ATC) classification codes [9]. We included only antibiotics with systemic and oral administration and excluded all antifungals, antimycobacterial drugs, and all topical antibiotics. For each ATC code and drug, we calculated defined daily doses (DDD) based on the World Health Organisation (WHO) definitions [9]. For the calculation of DDD1000 per day (pd), the study population was standardized by age and gender. The reference was the Swiss population of 2007 [10]. The number of Swiss residents being insured with the Helsana Group for the year 2007 and the total person days involved was calculated for each region. This allowed us the calculation of the DDD per 1000 persons per day (DDD1000pd) by summing up the DDD either per month or by year and dividing it by the number of insured individuals for the respective unit of time. We explored regional differences in antibiotic use based on the seven major regions as defined by the Swiss Federal Statistical Office (for a map see reference [11]).

We further analysed for each antibiotic prescription whether the reimbursement claims included compensations for office-based point of care tests, a laboratory test or radiographs ordered within 7 days prior to or 2 days following the procedure date. Point of care tests included urine analysis for leucocyte counts and urine culture, white blood cell count, C-reactive protein (CRP), pharyngeal-smear for antigen testing or culture of group A Streptococcus. Primary care physicians in Switzerland may run their own radiology facilities for conventional radiographs, or refer their patients to radiology centres.

In Switzerland deductibles are mandatory for all patients and according to individual budget and insurance package may range between Swiss Francs 300–2500 per year. Patients whose invoices are below the limit of the deductible sometimes do not send their bills to the insurance company but pay them directly. This results in incomplete information regarding their actual health care utilization. Therefore, we included for this part of the analysis only patients with data on point of care tests where the expenses exceeded the deductible. This restriction ensures that only individuals with complete information on antibiotic prescriptions and point of care tests were considered for the analysis, which reduced our number of prescriptions by 12.7%. We further restricted our analysis of point of care tests to primary care physicians and general internists (by excluding all specialists) and to physicians with at least four prescriptions per antibiotic group in order to avoid coefficients with zero values. This additional restriction reduced our initial number of prescriptions for the analysis of point of care tests by another 54.1% and resulted in a final total of 126 048 (40.0% of all prescriptions).

We used linear regression analysis to calculate the ratio between the number of prescriptions per patient group and the percentage of point of care tests for these prescriptions. All patients of one physician were divided into four patient groups according to gender and age (18–64 and >65 years). For each antibiotic group, the outcome variable (e.g. the number of prescriptions) divided by the number of patients per patient group and physician was calculated. Because our outcome variable was skewed to the left we used a log-transformation and obtained normally distributed dependent variables and residuals. In addition, the percentage of point of care tests per antibiotic prescription and region was calculated for all patients with costs in excess of the deductible. All statistical analyses were carried out with the open software R version 2.7.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

This analysis is based on data from 1 067 934 Swiss residents aged 18 or older that had mandatory health insurance with the Helsana Group, representing 17.1% of the Swiss population. In 2007, the mean age of this served population was 1.6 years above the national average age of 41.1 years, and compensation costs for this population were 7.6% above the national average [10,12]. Overall, 52.5% of the population were female and 26% were aged over 65 years. General practitioners comprised 41.5%, general internists 17.3% and specialists 41.2% of 11 208 antibiotic prescribing physicians, with a total of 314 915 prescriptions.

The rate of outpatient antibiotic prescriptions in the entire population was 8.5 DDD1000pd and varied by a factor of 1.5, with the highest rate in the region of Lake Geneva (11.33 DDD1000pd) and the lowest prescription rate in northwest Switzerland (7.28 DDD1000pd) (Fig. 1). In general, regions with higher overall prescription rates had also higher prescriptions per antibiotic group, with the exception of central Switzerland where consumption of amoxicillin and amoxicillin-calvulanate was lower (Fig. 1). In all regions, fluoroquinolones were the second most prescribed antibiotics, with DDD1000pd ranging between 1.45 and 2.28. Macrolides were the third most prescribed antibiotics, with DDD1000pd between 1.13 (east Switzerland) and 1.94 (Lake Geneva). Espace Midland, Lake Geneva and Ticino had a higher prescription rate for other betalactams (all betalactams excluding amoxicillin and amoxicillin/clavulanate), with DDD1000pd ranging between 0.69 and 1.29 when compared with ranges of prescription rates of 0.34 and 0.48 DDD1000pd in the remaining regions.

image

Figure 1.  Standardized DDD per 1000 inhabitants per day, for outpatient prescriptions in 2007 for all antibiotics and selected antibiotics for seven Swiss regions and Switzerland. Measures are standardized for age and sex to the Swiss population. NW, northwest; CH, Switzerland.

Download figure to PowerPoint

In order to explore seasonal variation we calculated mean DDD1000pd per month according to regions and antibiotic groups (Fig. 2). There was a high variation between summer and winter months for amoxicillin and amoxicillin-clavulanate for the two regions Lake Geneva and Ticino, with the highest prescription rates and less seasonal variation in the remaining regions with lower overall prescriptions for amoxicillin and amoxicillin-clavulanate. We found a very similar pattern for macrolides and other betalactam antibiotics, although the absolute difference in seasonal prescription variation was lower for theses groups of antibiotics.

image

Figure 2.  Monthly average DDD per 1000 inhabitants per day, for outpatient prescriptions in 2007 for seven regions in Switzerland. For better visibility of seasonal fluctuations the first month shown is April 2007 (A) and the last month shown is March 2007 (M). Measures are standardized for gender and age to the Swiss standard population. NW, northwest; CH, Switzerland.

Download figure to PowerPoint

Table 1 shows the use of point of care tests according to regions. We found a consistent pattern for each antibiotic, with a higher use of point of care tests in regions with lower antibiotic use. The use of all point of care tests was highest in the northwest, northeast and the Zurich regions and lowest in the Ticino and Lake Geneva regions. In a further analysis, we modelled the ratio of the number of prescriptions divided by the number of patients per physician and groups of antibiotics in relation to the use of point of care tests. The use of a point of care test was for all antibiotics consistently associated with fewer antibiotic prescriptions (Table 2). For example, if physicians conducted any point of care test the number of antibiotic prescriptions for amoxicillin and amoxicillin-clavulanate and fluoroquinolones was reduced by a factor of 0.79 (95% CI, 0.74–0.84) and 0.68 (95% CI, 0.65–0.72%), respectively. Thus, the use of point of care tests was associated with a 21% (95% CI, 16–26%) and 32% (95% CI, 28–45%) reduction in prescriptions of these antibiotics. With the exception of fluoroquinolones, female patients were less likely to receive an antibiotic prescription for each antibiotic class. Age was consistently associated with a higher use of antibiotics for each antibiotic class. The variance of antibiotic prescriptions and drug class explained by our model, however, was modest (adjusted R2).

Table 1.   Percentage of antibiotic prescriptions with point of care per region in patients with health consumptions above the deductible
Antibiotic group (number of prescriptions/number of physicians)RegionPercentage of tests
CRPLeucocyte countsPharyngeal swabaUrinebRadiographcAnye
  1. NW, northwest; CH, Switzerland.

  2. aTested for Streptococci group A either by antigen test or by culture.

  3. bUrine analysis and culture.

  4. cAny type of radiograph.

  5. dNot standardized for age and gender.

  6. ePerformed in a time window of 7 days prior to 2 days following an issued prescription.

Amoxicillin and amoxicillin-clavulanate (58 550/5274)CH totald2625480834
NW CH312938938
East CH3331391042
Zurich3331410841
Midland252347833
Central CH36344101044
Ticino151435723
Lake Geneva70835514
Other betalactams (15 100/2972)CH totald242357934
NW CH3229661040
East CH3231711745
Zurich4238691050
Midland2321571033
Central CH41416111254
Ticino201956829
Lake Geneva06755616
Doxycycline (8041/2347)CH totald21220111631
NW CH232427730
East CH2323112734
Zurich3232114642
Midland1717112628
Central CH2223112632
Ticino121207618
Lake Geneva6816914
Fluoroquinolones (76 423/5363)CH totald2424141557
NW CH2626141658
East CH2729046564
Zurich30290145663
Midland2322043457
Central CH3131140761
Ticino1818032645
Lake Geneva912033440
Macrolides (40 230/4802)CH totald3431441340
NW CH4137541446
East CH4138441347
Zurich4843551552
Midland3026431236
Central CH4742551553
Ticino2015320825
Lake Geneva78320914
Trimethoprim-sulfamethoxazole (12 295/3187)CH totald1517143352
NW CH1215143350
East CH1619045356
Zurich2022143456
Midland1216149256
Central CH2123045460
Ticino915036244
Lake Geneva57128332
Table 2.   Ratio of the number of antibiotic prescriptions and number of patients of primary care physicians for antibiotic compounds according to point of care tests and patient characteristics
 Amoxicillin and amoxicillin-clavulanateFluoroquinolonesMacrolidesOther betalactamsTrimethoprim-sulfamethoxazoleDoxycycline
Coefficients (95% CI)a [P-value]
  1. aBack-transformed coefficients with confidence intervals (CI) for antibiotic prescriptions: coefficients >1 indicate more and coefficients <1 less prescriptions.

  2. Regression model: log(Yi) ∼ β0 + β1 × AnyXRayi + β2 × AnyPocti + β3 × genderi + β4 × agei + εi, where: Yi, number of prescriptions/number of patients; i, antibiotic group, physician and patient group; AnyPocti, number POCTi/number of antibiotic prescriptionsi 0 ≤ Any POCT ≤ 1; genderi, [male, female].

  3. bAny radiographi, number of radiographs (including all radiographs)i/number of antibiotic prescriptionsi, 0 ≤ Any radiograph ≤ 1.

  4. cPOCT, point of care test.

  5. dAgei [18–64, 65+].

Any radiographb1.03 (0.91, 1.18) [0.604]1.01 (0.90, 1.12) [0.926]0.91 (0.81, 1.01) [0.087 ]0.81 (0.65, 1) [0.046 ]0.71 (0.45, 1.12) [0.142]0.95 (0.63, 1.45) [0.822]
Any POCTc0.79 (0.74, 0.84) [<0.001]0.68 (0.65, 0.72) [<0.001]0.76 (0.72, 0.81) [<0.001]0.63 (0.56, 0.71) [<0.001]0.68 (0.6, 0.77) [<0.001]0.91 (0.73, 1.13) [0.387]
Female patients0.77 (0.74, 0.79) [<0.001]1.02 (1.00, 1.05) [0.105]0.85 (0.82, 0.89) [<0.001 ]0.79 (0.73, 0.85) [<0.001]0.76 (0.68, 0.84) [<0.001]0.77 (0.68, 0.87) [<0.001]
Age groupd1.30 (1.26, 1.35) [<0.001]1.60 (1.56, 1.64) [<0.001]1.11 (1.06, 1.16) [<0.001]1.26 (1.17, 1.36) [<0.001]1.47 (1.35, 1.60) [<0.001]1.18 (1.03, 1.34) [0.014]
Adjusted R20.1100.1880.0460.1130.1590.039

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

In this study, data from roughly one-fifth of the adult Swiss population are represented. We confirm a very low rate of outpatient antibiotic prescriptions with important variation between regions within Switzerland. Overall DDD1000pd were highest for the French and Italian speaking parts (Lake Geneva and Ticino) of Switzerland. However, rates in these regions were still in the range of low prescribing European countries such as the Netherlands [3]. Similar to the pattern in other European countries, we found a higher seasonal variation of antibiotic prescriptions in high prescribing regions. Amoxicillin and amoxicillin-clavulanate continue to be the most used antibiotics for outpatients in Switzerland [4]. Use of macrolids and fluoroquinolones, however, is relatively high in Switzerland and constitutes a known risk factor for antibiotic resistance, in particular for S. pneumoniae [13–15]. Increased use of amoxicillin-clavulanate and fluoroquinolones has also been noted in other European countries [16]. We found a consistent inverse relation between use of any point of care test and antibiotic prescriptions for all classes of antibiotics that are mostly used in upper and lower respiratory tract and urinary tract infections.

This study has several strengths. Contrary to other studies, we used a single and rather comprehensive data source for antibiotic use from the reimbursement system of the largest Swiss social health insurer. This source guarantees a uniform data collection system that is based on information on drug use for antibiotics that were actually delivered to patients. Thus, our analysis is not based on data of drug sales as used in previous studies [3,4], which may introduce bias in ecological analyses [3,17]. In addition, our database allowed us to identify the prescription behaviour at the individual physician level. We used the ATC classification system and report DDD measurement units in standardized formats that allow for comparison of our data with other countries [9]. We restricted our analysis to adults in order to avoid bias, because DDDs do not take into account different doses for children.

We could show an important and statistically significant reduction in the range of 21% to 37% for prescriptions of the five most frequently used classes of antibiotics with the use of any point of care test. Although these findings are important and interesting from the public health perspective, they have to be interpreted with caution. No causal association can be drawn from our ecological study. Our findings could be confounded by other factors related to physicians or patients we were unable to measure. Physicians who use more tests may reassure patients and themselves that antibiotics are not needed. In a primary care study on patients’ expectations for antibiotics in respiratory tract infections, reassurance was more strongly related to patient satisfaction than actually receiving an antibiotic prescription [18]. Thus, the association of lower antibiotic prescription rates with the use of point of care tests could simply reflect testing patients with low probability of bacterial infection. As we lack data on the type and the severity of infections, we cannot rule out such bias. Because health insurers in Switzerland reimburse physicians for office-based laboratory tests, there exists an economic incentive to use such tests. Issues of liability due to complications may be another reason for laboratory testing or patients may request testing to exclude a serious infection. Patient pressure has been reported as an important factor in patient referral and additional clinical investigations, even when general practitioners judge the need for such activities as not high [19].

Previous observational studies from Scandinavia have shown a reduction in antibiotic prescriptions with the use of CRP in primary care [20,21]. Whether a diagnostic test leads to fewer antibiotic prescriptions without compromising patient outcome is ideally best addressed in a clinical trial evaluating two or more diagnostic strategies and their clinical consequences for patient-relevant outcomes [22]. Trials conducted in primary care and in emergency departments found no reduction of antibiotic prescriptions for respiratory tract infections with CRP testing [23–25], whereas one trial using a more complex intervention with general practitioner training and use of CRP tests did show a reduction of antibiotic prescription in patients with acute bronchitis [7].

Our study has several limitations. The population served by the Helsana Group was slightly older and generated higher costs than the average Swiss population. Therefore, antibiotic consumption eventually may be higher in this population than in the remaining Swiss population. In some areas with self-dispensing physicians, patients with high deductibles may not send the invoice to the health insurer but rather pay them directly. From previous unpublished studies, we estimate that 2–3% of all invoices are directly paid and not reimbursed by the Helsana Group [26]. Social health insurers do not get full and accurate information regarding the exact diagnosis from the physician-patient encounter. Therefore, we were unable to link antibiotic prescriptions and the use of point of care tests to a particular diagnosis. Data from a Swedish surveillance study indicate that CRP in primary care is overwhelmingly used in respiratory tract infections [27]. In addition, our study lacks detail regarding important characteristics of the study population that goes beyond crude estimates of age and gender distributions.

In conclusion, this study shows that the regions of the northwest and east of Switzerland have one of the lowest antibiotic prescription rates in ambulatory care in Europe. Swiss regions with higher prescription rates show higher seasonal variability in antibiotic use. The use of any point of care test was associated with an important reduction in the prescription of the most frequently used antibiotics. Our findings need confirmation from other countries and indicate that more research for new point of care tests and better integration of already existing point of care tests into primary care may be a promising strategy to reduce antibiotic consumption and thereby the emergence of antibiotic resistance in the community.

Transparency Declaration

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

R. Achermann and P. Gyger are employed by the Helsana Group. H. C. Bucher and K. Suter are supported by grants from santésuisse and the Gottfried and Julia Bangerter-Rhyner-Foundation. The sponsors had no role in the planning, the conduct and the submission of this manuscript. All authors declare no conflict of interests. The corresponding author had full access to all data in the study and had the final responsibility for the decision to submit the manuscript.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References
  • 1
    Seppala H, Klaukka T, Vuopio-Varkila J et al. The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. New Engl J Med 1997; 337: 441446.
  • 2
    Hay AD, Thomas M, Montgomery A et al. The relationship between primary care antibiotic prescribing and bacterial resistance in adults in the community: a controlled observational study using individual patient data. J Antimicrob Chemother 2005; 56: 146153.
  • 3
    Goossens H, Ferech M, Vander SR, Elseviers M. Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet 2005; 365: 579587.
  • 4
    Filippini M, Masiero G, Moschetti K. Socioeconomic determinants of regional differences in outpatient antibiotic consumption: evidence from Switzerland. Health Policy 2006; 78: 7792.
  • 5
    Briel M, Young J, Tschudi P et al. Prevalence and influence of diagnostic tests for acute respiratory tract infections in primary care. Swiss Med Wkly 2006; 136: 248253.
  • 6
    Briel M, Schuetz P, Mueller B et al. Procalcitonin-guided antibiotic use vs a standard approach for acute respiratory tract infections in primary care. Arch Intern Med 2008; 168: 20002007.
  • 7
    Cals JW, Butler CC, Hopstaken RM, Hood K, Dinant GJ. Effect of point of care testing for C reactive protein and training in communication skills on antibiotic use in lower respiratory tract infections: cluster randomised trial. BMJ 2009; 338: b1374.
  • 8
    e-mediat AG. Artikelstamm über Informationssysteme im Gesundheitswesen Galdat. Bern, Switzerland, 2009. Available at: http://www.e-mediat.ch/e-mediat/de/galdat.php (last accessed 30 June 2010).
  • 9
    World Health Organisation. WHO Collaborating Centre for Drug Statistics Methodology. The anatomical therapeutic chemical (ATC) and defined daily dosing (DDD) system. Oslo, Norway, 2009. Available at: http://www.whocc.no/atcddd/ (last accessed 30 June 2010).
  • 10
    Federal Statistic Office. Altersmasszahlen der ständigen Wohnbevölkerung nach Staatsangehörigkeit und Geschlecht, 2007. Bern, Switzerland, 2009. Available at: http://www.bfs.admin.ch/bfs/portal (last accessed 30 June 2010).
  • 11
    Federal Statistic Office. Die 7 Grossregionen der Schweiz. 2010. Available at: http://www.bfs.admin.ch/bfs/portal/de/index/regionen/11/geo/analyse_regionen/02a.html (last accessed 30 June 2010).
  • 12
    Gemeinsame Einrichtung Krankenversicherungsgesetz (KVG). Statistik definitiver Risikoausgleich 2007. Solothurn, Switzerland, 2009. Available at: http://www.kvg.org/default.htm (last accessed 30 March 2010).
  • 13
    Besser RE. Antimicrobial prescribing in the United States: good news, bad news. Ann Intern Med 2003; 138: 605606.
  • 14
    Klugman KP, Lonks JR. Hidden epidemic of macrolide-resistant pneumococci. Emerg Infect Dis 2005; 11: 802807.
  • 15
    Hyde TB, Gay K, Stephens DS et al. Macrolide resistance among invasive Streptococcus pneumoniae isolates. JAMA 2001; 286: 18571862.
  • 16
    Elseviers MM, Ferech M, Vander Stichele RH, Goossens H. Antibiotic use in ambulatory care in Europe (ESAC data 1997-2002): trends, regional differences and seasonal fluctuations. Pharmacoepidemiol Drug Saf 2007; 16: 115123.
  • 17
    Cars O, Molstad S, Melander A. Variation in antibiotic use in the European Union. Lancet 2001; 357: 18511853.
  • 18
    Welschen I, Kuyvenhoven M, Hoes A, Verheij T. Antibiotics for acute respiratory tract symptoms: patients’ expectations, GPs’ management and patient satisfaction. Fam Pract 2004; 21: 234237.
  • 19
    Little P, Dorward M, Warner G, Stephens K, Senior J, Moore M. Importance of patient pressure and perceived pressure and perceived medical need for investigations, referral, and prescribing in primary care: nested observational study. BMJ 2004; 328: 444.
  • 20
    Andre M, Schwan A, Odenholt I. The use of CRP tests in patients with respiratory tract infections in primary care in Sweden can be questioned. Scand J Infect Dis 2004; 36: 192197.
  • 21
    Bjerrum L, Gahrn-Hansen B, Munck AP. C-reactive protein measurement in general practice may lead to lower antibiotic prescribing for sinusitis. Br J Gen Pract 2004; 54: 659662.
  • 22
    Bossuyt PM, Lijmer JG, Mol BW. Randomised comparisons of medical tests: sometimes invalid, not always efficient. Lancet 2000; 356: 18441847.
  • 23
    Dahler-Eriksen BS, Lauritzen T, Lassen JF, Lund ED, Brandslund I. Near-patient test for C-reactive protein in general practice: assessment of clinical, organizational, and economic outcomes. Clin Chem 1999; 45: 478485.
  • 24
    Diederichsen HZ, Skamling M, Diederichsen A et al. Randomised controlled trial of CRP rapid test as a guide to treatment of respiratory infections in general practice. Scand J Prim Health Care 2000; 18: 3943.
  • 25
    Gonzales R, Aagaard EM, Camargo CA Jr. et al. C-reactive protein testing does not decrease antibiotic use for acute cough illness when compared to a clinical algorithm. J Emerg Med 2008; doi: 10.1016/j.jemermed.2008.06.021.
  • 26
    Achermann R. Abrechnungseffekte bei der Umstellung von Tiers Garant (TG) zu Tiers Payant (TP). HelsanaIS, ed. 2006.
  • 27
    Engstrom S, Molstad S, Lindstrom K, Nilsson G, Borgquist L. Excessive use of rapid tests in respiratory tract infections in Swedish primary health care. Scand J Infect Dis 2004; 36: 213218.