Variation in microbiological test requisitions among general practices in the North Denmark Region

The use of microbiological services has been closely examined since the 1970s [1]. When an infection is suspected during the diagnostic process, microbiological tests may be requested to identify the pathogen, confirm or exclude diagnostic hypotheses and guide treatment decisions. However, general practitioners (GPs) may vary in how often they use these services, occasionally requesting tests for reassurance or to address clinical uncertainties [2].

Diagnostic stewardship, which promotes the appropriate use of laboratory testing, has gained increasing attention because the clinical value of some microbiological test results remains unclear, and some studies have shown that only a small percentage of microbiological test results lead to treatment changes [2-6]. Moreover, diagnostic stewardship is closely linked with antibiotic stewardship, with clinical microbiology laboratories playing a key role in guiding therapy [7]. Despite concerns that many laboratory tests may be unnecessary, the volume of GP requests has been steadily increasing in Denmark [8].

In Denmark, healthcare is publicly funded, and each resident is assigned to a GP clinic. GPs provide first-line services and act as gatekeepers to secondary healthcare [9]. Importantly, microbiological services are available free of charge at the point of use to GPs and their patients. GP clinics are embedded in a universal, tax-funded healthcare system and receive a standard fee from the region for performing microbiological tests [9, 10]. In Denmark, healthcare services are administered by five regions, which are political and administrative authorities financed through national government allocations without an independent taxation authority. The North Denmark Region is one of these. This study aimed to describe the patterns and variety of microbiological test requisitions from GP clinics in the North Denmark Region, with a focus on identifying outliers with high requisition rates in specific test categories.

Methods

Setting

The North Denmark Region is the smallest of the five Danish regions, with a stable population of approximately 590,000 inhabitants. The region is served by two main hospital organisations - Aalborg University Hospital and the North Denmark Regional Hospital - which together operate across 11 hospital sites located in various cities. The Department of Clinical Microbiology (DCM) at Aalborg University Hospital provides microbiological services to all hospitals and GP clinics in the region.

Study design and data sources

This descriptive study covered the period from January 2019 to December 2023. From the microbiological database hosted at the DCM, we extracted all requisitions for microbiological tests submitted by GP clinics using their respective Danish National Health Service provider numbers. As more than one doctor may be affiliated with a single clinic [9], data on the number of health-insured individuals registered with each clinic under the Danish health insurance system were obtained from the regional Department of Business Intelligence.

For analytical purposes, all specimens performed at the DCM were grouped into nine mutually exclusive test categories: urine cultures, faecal specimens, PCR specimens, serology specimens, fungal specimens, respiratory specimens, parasitological specimens, sexually transmitted disease (STD) and gynaecological specimens, and bacterial swabs (Table 1). Each specimen was assigned to a single category. PCR tests for SARS-CoV-2 (COVID-19) were excluded due to expected large temporal fluctuations during the study period. PCR tests for influenza A/B and respiratory syncytial virus (RSV) were included only when performed as stand-alone assays. Multiplex PCR assays combining influenza and/or RSV with SARS-CoV-2 were also excluded, as these tests were introduced during the COVID-19 pandemic and exhibited substantial temporal variation related to testing strategies and diagnostic platforms.

Data analysis

GP clinics that were active for at least three consecutive years (≥ 36 months) during the study period were included, whereas clinics that were active for fewer than three years were excluded due to limited data. For each clinic, annual requisition rates were calculated using the clinic’s registered patient population for the corresponding year and expressed as specimens per 1,000 individuals across the nine test categories, rounded to the nearest integer. All analyses were performed at the clinic level, and rates were averaged across clinics’ participation years. For each test category, we calculated minimum, maximum, median and IQR, reflecting the non-normal data distribution. Data were visualised using violin plots with boxplot overlay, showing medians, IQRs and the upper threshold (Q3 + 1.5 × IQR). Outliers were defined as GP clinics with an annual specimen count per 1,000 individuals exceeding the upper threshold [11]. The study was approved as a quality project by the North Denmark Region (ID number: K2022-004).

Trail registration: not relevant.

Results

General practitioner clinics and patients

We identified 192 GP clinics in the region during the study period. A total of 39 were excluded due to insufficient data, defined as fewer than two years with specimen requisitions. In total, 153 GP clinics were included. Among these, 119 were active throughout the full five-year period, 14 for four years and 21 for three years. The number of affiliated residents averaged 548,307 annually, corresponding to approximately 92.9% of the regional population. The mean number of affiliated patients per GP clinic was 3,584, ranging from 622 to 10,177.

Microbiological specimens

During the five-year study period, a total of 690,559 specimens were requisitioned from GP clinics, corresponding to an annual average of 138,112. These comprised 245,821 urine cultures (49,164 annually), 162,787 STD and gynaecological specimens (32,557 annually), 156,852 bacterial swabs (31,370 annually), 36,393 faecal specimens (7,279 annually), 32,214 PCR tests (6,443 annually), 23,181 serology specimens (4,636 annually), 22,043 fungal specimens (4,409 annually), 7,098 parasitological specimens (1,420 annually) and 4,170 respiratory specimens (834 annually). Standardised requisition rates per 1,000 individuals are summarised in Table 2. Requisition patterns are shown in Figure 1. Urine cultures had the highest median rate of 89 (IQR: 57-123) specimens per 1,000 individuals and the widest range (0-291), followed by bacterial swabs (median 54, IQR: 39-74) and STD/gynaecological specimens (median 45, IQR: 33-67). In the STD category, PCR for Chlamydia trachomatis and Neisseria gonorrhoeae accounted for approximately 75% of the requisitions. In comparison, parasitological and respiratory specimens were requested least frequently, with annual medians of 2 (IQR: 1-3) and 1 (IQR: 0-2), respectively.

Outliers

Among the 153 GP clinics included, 54 (35.3%) were identified as outliers in at least one test category, meaning that their requisition rates exceeded the upper threshold (Table 2, Figure 1). Among these, 38 clinics were outliers in a single category, nine in two categories, five in three categories and two in four and five test categories.

Large-volume test categories had relatively few outliers: urine cultures (n = 2) and bacterial swabs (n = 4). Similarly, few outliers were observed for faecal and fungal specimens (four and five clinics, respectively). In contrast, smaller-volume categories such as respiratory specimens and parasitological specimens had the highest number of outliers, with 17 and 13 clinics, respectively, indicating that a limited number of GP clinics accounted for a disproportionately large share of requisitions in these categories.

Discussion

This study examined microbiological test requisitions from GP clinics in the North Denmark Region in a five-year period. The findings reveal considerable variability in test use, with urine cultures being the most frequently requested, whereas respiratory and parasitological specimens were the least commonly requested categories. A significant proportion of GP clinics (35.3%) were identified as outliers based on requisition patterns, consistently requesting more microbiological tests than other clinics and thus exceeding the upper threshold. This suggests that these clinics were responsible for a disproportionate number of requisitions in specific categories.

Urine cultures were the most commonly performed test, with a median of 89 annual requisitions per 1,000 individuals and the greatest variability in frequency. This high prevalence reflects the common presentation of urinary tract infections in general practice. The variability in urine culture requisitions likely stems from differing diagnostic approaches, such as urine dipstick testing, microscopy and culture, which may be performed on-site or sent to the DCM [10, 12, 13]. Accordingly, very low or zero laboratory-based urine culture requisition rates do not necessarily indicate underuse, as some clinics may perform urine testing in-house and submit samples to the DCM, primarily in cases of treatment failure, diagnostic uncertainty, or pregnancy.

Testing for STD accounted for approximately 75% of requisitions in the STD/gynaecological category, predominantly among adolescents and young adults. Higher requisition rates may reflect clinics serving relatively younger populations or diagnostic practices such as self-testing for chlamydia and gonorrhoea.

Requisitions for bacterial swabs, as well as faecal and fungal specimens, were more evenly distributed with relatively few outliers in each category. Bacterial swabs are useful for identifying bacterial infections and guiding antibiotic treatment, but in clinically obvious cases, they are not always necessary. The requisition rate for faecal specimens was relatively low (13 per 1,000 individuals annually), and diagnostic testing is often unnecessary as acute diarrhoea is typically self-limiting [7]. Moreover, according to international guidelines, diagnostic tests should be reserved for patients with community-acquired diarrhoea lasting seven days or more, those with a recent travel history, or those presenting with warning signs for severe disease, such as fever, bloody diarrhoea, severe abdominal pain, dehydration or immunocompromise [7].

Fungal specimen requisitions were relatively uncommon, likely reflecting the lower prevalence of mycoses compared to bacterial and viral infections. Testing is typically reserved for clinically suggestive cases (e.g., tinea or onychomycosis).

Smaller test categories, including PCR tests, serology tests, parasitological specimens and respiratory specimens, had the highest number of outliers, with 12, 11, 13 and 17 GP clinics, respectively. This suggests that a limited number of clinics accounted for a disproportionate share of requisitions in these categories, potentially because of specific medical considerations. For instance, sputum specimens for bacterial cultures are not routinely indicated for community-acquired pneumonia in general practice. However, certain patient groups (e.g., those with chronic lung disease or recurrent infections) may warrant such testing. We were unable to determine whether such patients were concentrated in particular clinics.

The observed variation may reflect differences in clinical practice, diagnostic confidence and familiarity with available testing modalities. Some clinics may use microbiological testing more frequently to address diagnostic uncertainty, whereas others may rely more on clinical judgement. Adherence to general practice clinical recommendations issued by the Danish College of General Practitioners may also influence requisition patterns. Interestingly, some clinics were outliers in multiple categories, whereas others were outliers in only one category, suggesting that variability may not always be systematic. The Danish healthcare context should also be considered. Denmark has a similar number of GPs per inhabitant as Sweden and Iceland, but fewer than Norway [14]. Furthermore, the North Denmark Region consistently had the highest number of patients per GP (1,750-1,815) throughout the study period [15]. Therefore, we cannot rule out that increased patient loads may have affected diagnostic responsibilities, particularly in clinics employing other healthcare professionals (e.g., nurses) who handle part of patient consultations.

The study has several strengths. It provides a comprehensive analysis of microbiological test requisitions across GP clinics in the North Denmark Region. The use of standardised data per 1,000 individuals enhances the comparability and offers a nuanced understanding of requisition patterns. However, we did not assess trends over time, and the impact of the COVID-19 pandemic on healthcare practices and patient behaviour may have introduced fluctuations, making trend analyses unreliable during this period. Several limitations should be acknowledged, including exclusion of COVID-19 testing, use of data from a single region only and lack of adjustment for patient demographics at the clinic level and for other clinic characteristics. Furthermore, the study did not account for changes in clinical behaviour or diagnostic technology over time.

Conclusions

This study highlights significant variability in microbiological test requisitions among GP clinics in the North Denmark Region, identifying 35% of clinics as outliers. These findings have implications for diagnostic stewardship and healthcare resource management. Understanding test use patterns may inform targeted interventions to optimise testing. Further research is needed to explore the drivers of this variability, including clinic-level influences, patient demographics and the impact of changes in the healthcare setting.