Management of children's urinary tract infections in Dutch family practice: a cohort study
© Harmsen et al; licensee BioMed Central Ltd. 2007
Received: 17 July 2006
Accepted: 13 March 2007
Published: 13 March 2007
Optimal clinical management of childhood urinary tract infections (UTI) potentiates long-term positive health effects. Insight into the quality of care in Dutch family practices for UTIs was limited, particularly regarding observation periods of more than a year. Our aim was to describe the clinical management of young children's UTIs in Dutch primary care and to compare this to the national guideline recommendations.
In this cohort study, all 0 to 6-year-old children with a diagnosed UTI in 2001 were identified within the Netherlands Information Network of General Practitioners (LINH), which comprises 120 practices. From the Dutch guideline on urinary tract infections, seven indicators were derived, on prescription, follow-up, and referral.
Of the 284 children with UTI who could be followed for three years, 183 (64%) were registered to have had one cystitis episode, 52 (18%) had two episodes, and 43 (15%) had three or more episodes. Another six children were registered to have had one or two episodes of acute pyelonephritis. Overall, antibiotics were prescribed for 66% of the children having had ≤ 3 cystitis episodes, two-thirds of whom received the antibiotics of first choice. About 30% of all episodes were followed up in general practice. Thirty-eight children were referred (14%), mostly to a paediatrician (76%). Less than one-third of the children who should have been referred was actually referred.
Treatment of childhood UTIs in Dutch family practice should be improved with respect to prescription, follow-up, and referral. Quality improvement should address the low incidence of urinary tract infections in children in family practice.
Awareness of the importance of timely diagnosis and treatment of childhood urinary tract infections (UTIs) is growing. Without timely treatment, renal scarring can occur , which is likely to affect approximately 5–15% of young children with a UTI [2–5]. Renal scarring is associated with serious health problems in later life, such as hypertension, complications during pregnancy, and renal failure . Optimal clinical management of childhood UTI potentiates long-term positive health effects. Therefore, guidelines across different countries advocate an active approach concerning prescription, follow-up, and referral [6–9].
The guideline on UTIs of the Dutch College of General Practitioners (DCGP) states that every childhood UTI should be treated with antibiotics because of the risk of renal scarring. Amoxicillin/clavulanic acid or co-trimoxazole are the medications of choice in an attempt to reach effective tissue levels and to maximize the chance of the most effective medicine until test results are available. The follow-up recommendation takes into account that young children may not express their complaints clearly, while they have a high risk of renal scarring. The referral recommendations are based on the patients' age and sex, which predict the probability of anatomical abnormalities of the urinary tract system .
The one-year incidence of UTIs in 0 to 6-year-old children in Dutch family practices averages for girls 25.7 and for boys 4.5 per 1000 life-years . Insight into the quality of care for UTI was limited, particularly regarding observation periods of more than a year. We have acquired the necessary prospective data to provide insight into the primary-care-based management of childhood UTIs in the Netherlands. We aimed to describe the clinical management of young children's UTIs in Dutch primary care and compare this to the national guideline recommendations.
Design and setting
A prospective cohort study in Dutch family practice was performed.
We identified all children 0–6 years old (born between 1994 and 2001) with UTI diagnosed in 2001 within the Netherlands Information Network of General Practice (LINH). The LINH network contains 120 practices and is representative of the Dutch population of patients, family practitioners, and types of practices . Informed consent was arranged within the network (general board of the National Institute for Health Services Research (NIVEL), general board of the Centre for Quality of Care Research (WOK), general board of the Dutch College of General Practitioners (NHG), and the general board of the National Association of General Practitioners (LHV)). According tothe Dutch Central Committee on Research Involving Human Subjects (CCMO) regulations only research in which the study participant has to be physically present during the study is subject to the Medical Research Involving Human Subjects Act (WMO) and therefore ethical approval is not required for studies that use patient databases.
The International Classification of Primary Care (ICPC)  defines UTI as acute pyelonephritis (ICPC code U70) or cystitis (ICPC code U71). These definitions imply that UTI was diagnosed by urine testing, not just suspected or assumed.
Indicators urinary tract infections (UTIs) in children in general practice
1. Antibiotics given
All children receiving antibiotics
2. Amoxicillin/clavulanic acid OR co-trimoxazole
Children receiving first-choice antibiotics
All children receiving antibiotics
3. Episodes with at least one follow-up contact
Number of UTI episodes with >1 contact
All UTI episodes
For all groups:
Total referred within the group
Total within the group
4. Children <1 year old
5. Boys <12 years old
6. Girls 1–4 years old with second UTI
7. Girls 5–12 years old with > 1 recurrent UTI
In the LINH network, the family practice staff routinely records the encoded patient information in electronic medical records (EMR). For the period 2001–2003 data were extracted from the EMR, concerning contacts with the family practice, prescriptions, referrals, and patient characteristics (age, sex).
Practices were excluded from the analyses if they had registered fewer than 46 weeks in 2002 or 2003. Patients who were not on the practice list and patients who had left the practice before 1 January 2004 were also excluded.
The contacts were expressed in units of episodes. Episodes were considered new episodes if the preceding contact for UTI occurred more than 28 days previously. Prescriptions and referrals within 28 days after the last contact for an episode were linked to that episode. The first UTI in 2001 was assumed to be the child's first episode ever. To be able to compare groups, four groups of children were created, based on ICPC code and number of UTI episodes: group 1 (1 episode cystitis), group 2 (2 episodes cystitis), group 3 (≥ 3 episodes cystitis) and group 4 (1 or 2 episodes acute pyelonephritis).
Descriptive statistics were applied to patient characteristics, and the numbers of children receiving medication, follow-up, or referrals. For each indicator, the number of children (or episodes in the case of follow-up) to whom family practitioners (FPs) offered the appropriate care was divided by the total number of children (or episodes in the case of follow-up) needing the provision of such care (Table 1). We calculated the percentage of children with more than one contact during a UTI episode as an indicator of follow-up. Age groups were based on age when having the first UTI.
Student's t-test or chi-square tests, as appropriate, were used to investigate whether more boys or girls were treated as recommended by the guideline, and whether younger children were treated more consistently with the guideline than older children. We also investigated whether recurrent childhood UTIs were more often treated according to the guideline than single episodes. We considered a probability level of P < 0.05 statistically significant.
Of the 284 children included, 278 (98%) were diagnosed with cystitis. Of these children, 66% had one episode (group 1), 19% had two (group 2), and 15% had 3 to 10 (group 3). Six children were diagnosed with acute pyelonephritis (2%; group 4).
During the first UTI episode, the mean age varied from 2.8 (SD 2.5) to 4.1 years (SD 1.5). About 80% of the children in group 1, 90% in groups 2 and 3, and 100% in group 4 were girls (figure 1). Because of the small numbers, no further data are presented for group 4 and for episodes 4–10 in group 3.
Prescription of antibiotics in children with urinary tract infections
Percentage (number) of children/antibiotic prescription
Overall children with antibiotics (indicator 1)
Percentage (number) of children with antibiotics/choice of antibiotic
Amoxicillin/clavulanic acid or co-trimoxazole
Overall first-choice antibiotics (indicator 2)
Ceftibuten, ofloxacin, or nitrofurantoin
Follow-up and referrals in children with urinary tract infections
Pecentage (number) of episodes with more than one contact during the episode
Overall episodes with more than one contact (indicator 3)
Percentage (number) of children with a referral
Percentage (number) of children with a referral, special groups of children
Children <1 year of age (indicator 4)
Boys overall (indicator 5)
Girls 1–4 years, second episode (indicator 6)
Girls 5–9 years, third episode (indicator 7)
Table 3 shows that overall 38 of 278 (14%) children were referred for specialist treatment (range: 8–16%). Seventy-six percent was referred to a paediatrician; 8% to a urologist; 3% to radiography; and from 13% it is not clear to what specialist they were referred.
Less than one-third of the children who should have been referred concerning the guideline on referring specific age and/or sex categories, was actually referred. Three of four children younger than one year were referred, and fewer than 25% of the boys were referred during at least one episode. In group 2, 9% of the girls aged 1–4 years were referred during the second episode, and in group 3 33%. Two girls in the 5–9 year olds group (7%) were referred during the third episode. No differences regarding age and sex subgroups were found.
This study showed that the management of childhood UTIs in the Netherlands varied substantially across patients. Only 66% of the children received antibiotics and of these 66% was prescribed first choice antibiotics. There was no follow-up in the majority of the episodes. Referral of children younger than one year was generally consistent with the guidelines, but the referral rates for boys, girls 1–4 years old with a second UTI, and girls 5–12 years old with more than one recurrent UTI should have been much higher, if we consider the guidelines.
We found that the proportion of children receiving amoxicillin decreased proportionally to the number of episodes. Perhaps FPs prescribe amoxicillin routinely because this medication is much older than the combined form with clavulanic acid, and has less side effects than amoxicillin/clavulanic acid. If amoxicillin alone does not work, they prescribe the combination. Prescribing ceftibuten, ofloxacin, or nitrofurantoin suggests that not all FPs are aware that, according to the guidelines, childhood UTI should be treated as complicated UTI. This is confirmed by the fact that, for 98% of all children, the FP had registered the ICPC for cystitis instead of the code for complicated UTI (pyelonephritis). No significant differences of age or gender might imply that GPs are unaware of the increased risk of complications or underlying pathology in boys and younger children. Such unawareness may lead to health complications when the child is older.
The LINH network provided a unique opportunity for collecting prospective data regarding clinical management in routine healthcare settings, but one can question FP registration behaviour and whether all childhood UTIs were identified with the ICPC codes 'acute pyelonephritis' and 'cystitis'. However, the incidence we found for 0 to 6-year-olds, which is 12.0 (461*1000/38408), is comparable to those in other Dutch studies: 15.1  and 13.2 . Since direct observation and hand-searching medical records are infeasible, using databases of consultation registrations seems to be the optimal method for collecting information about FP clinical behaviour.
It is difficult to compare our results on FP management with other studies because our data are prospectively collected, had a follow-up period of three years, and focussed on primary care and individual young children; this in contrast to other studies. A study by Kwok et al.  already gave some insight in the management of children's UTIs in Dutch family practice. But compared to our study, this study concentrated on a much wider age range, although the children most vulnerable to renal scarring are the younger ones. The study also had a follow-up period of only one year and did not pay attention to follow-up after the antiobiotic treatment. One British study found that 37% of children with proven UTI were sent for renal tract imaging , and another Dutch study reported 4% of the children being referred . Two other British studies found much higher rates of referral [17, 18]. However, these last two studies used postal questionnaires to measure FP behaviour, whereas our study and the first two studies used medical records. Reporting behaviour retrospectively may lead to overestimation of guideline adherence because of social desirability bias .
Improvement of professional performance might substantially improve clinical outcomes. This is demonstrated in Sweden, where a more aggressive approach led to no new cases of uraemia caused by non-obstructive pyelonephritis during the years 1986–1995 . But, development and distribution of guidelines do not necessarily lead to better patient care . Future research could focus on developing interventions to improve prescription, follow-up, and referrals, but should also consider motives for not following the guidelines. Because not many childhood UTIs appear in family practice in the Netherlands – our study saw an average of five children per practice in one year – interventions should not be too time consuming for the FPs.
In order to prevent negative health outcomes, treatment of childhood urinary tract infections in Dutch family practice should be improved with respect to prescription, follow-up, and referral. Quality improvement should address the low incidence of urinary tract infections in children in family practice.
Dutch College of General Practitioners
electronic medical record
International Classification of Primary Care
Netherlands Information Network of General Practice
urinary tract infection
The Dutch Kidney Foundation (Nederlandse Nierstichting) provided a grant for this study (project OV02). We thank Jan Mulder and Henk van den Hoogen for their help with the data analyses.
- Jacobson SH, Eklöf O, Eriksson CG, Lins LE, Tidgren B, Winberg J: Development of hypertension and ureaemia after pyelonephritis in childhood: 27 year follow up. Brit Med J. 1989, 299: 703-706.View ArticlePubMedPubMed CentralGoogle Scholar
- Pylkkänen J, Vilska J, Koskimies O: The value of level diagnosis of childhood urinary tract infection in predicting renal injury. Acta Paediatr Scand. 1981, 70: 879-883.View ArticlePubMedGoogle Scholar
- Vernon SJ, Coulthard MG, Lambert HJ, Keir MJ, Matthews JNS: New renal scarring in children who at age 3 and 4 had had normal scans with dimercaptosuccinic acid: Follow up study. Brit Med J. 1997, 315: 905-908.View ArticlePubMedPubMed CentralGoogle Scholar
- Winberg J, Andersen HJ, Bergström T, Jacobsson B, Larson H, Lincoln K: Epidemiology of symptomatic urinary tract infection in childhood. Acta Paediatr Scand. 1974, Suppl 252: 1-20.View ArticleGoogle Scholar
- Coulthard MG, Lambert HJ, Keir MJ: Occurence of renal scars in children after their first referral for urinary tract infection. Brit Med J. 1997, 315: 918-919.View ArticlePubMedPubMed CentralGoogle Scholar
- American Academy of Pediatrics: Practice Parameter: The diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. Pediatrics. 1999, 103: 843-852.Google Scholar
- Royal College of Physicians: Guidelines for the management of acute urinary tract infection in childhood. Report of a Working Group of the research unit, Royal College of Physicians. J R Coll Physicians London. 1991, 25: 36-42.Google Scholar
- Jodal U, Lindberg U: Guidelines for management of children with urinary tract infection and vesico-ureteric reflux. Recommendations from a Swedish state-of-the-art conference. Swedish Medical Research Council . Acta Paediatr Suppl. 1999, 88 (431): 87-9.PubMedGoogle Scholar
- van Haaren KAM, Visser HS, van Vliet S, Timmermans AE, Yadava R, Geerlings SE, ter Riet G, van Pinxteren B: NHG-Standaard Urineweginfecties - Tweede herziening [Guideline of the Dutch College of General Practitioners on urinary tract infections: second revision]. Huisarts Wet. 2005, 48: 341-352.View ArticleGoogle Scholar
- van der Linden MW, van Suijlekom-Smit LWA, Schellevis FG, van der Wouden JC: Tweede nationale studie naar ziekten en verrichtingen in de huisartspraktijk. Het kind in de huisartspraktijk [Second Dutch national survey of general practice. The child in general practice]. 2005, Rotterdam/Utrecht, Erasmus MC, afdeling Huisartsgeneeskunde/NIVELGoogle Scholar
- Tacken MAJB, Braspenning JCC, de Bakker DH, Jabaaij L, van den Hoogen HJM, Verheij R, Groenewegen PP, Grol RPTM: The Dutch National Information Network of General Practice (LINH) and quality-of-care data. Quality of preventive performance in general practice: The use of routinely collected data [thesis]. Edited by: Tacken MAJB. 2005, Nijmegen, Radboud University Nijmegen, 23-44.Google Scholar
- Lamberts H, Wood M: ICPC: International classification of primary care. 1987, Oxford, Oxford University PressGoogle Scholar
- Timmermans AE, Baselier PJAM, Winkens RAG, Arets H, Wiersma T: NHG-Standaard Urineweginfecties: Eerste herziening [Guideline of the Dutch College of General Practitioners on urinary tract infections: first revision]. 1999, Utrecht, Nederlands Huisartsen Genootschap (NHG)Google Scholar
- van de Lisdonk EH, Verstraeten J: Kinderen met urineweginfecties: Verwijzen of niet? Een retrospectief onderzoek in de vier CMR-praktijken [Children with urinary tract infections: To refer or not? A retrospective study in four CMR-practices]. Huisarts Wet. 2000, 43: 343-346.Google Scholar
- Kwok WY, de Kwaadsteniet MCE, Harmsen M, van Suijlekom-Smit LWA, Schellevis FG, van der Wouden JC: Incidence rates and management of urinary tract infections among children in Dutch general practice: Results from a nation-wide registration study. BMC Pediatr. 2006, 6: 10-10.1186/1471-2431-6-10.View ArticlePubMedPubMed CentralGoogle Scholar
- South Bedfordshire Practitioners' Group: How well do general practitioners manage urinary problems in children?. Br J Gen Pract. 1990, 40: 146-149.Google Scholar
- Vernon S, Foo CK, Coulthard MG: How general practitioners manage children with urinary tract infection: An audit in the former Northern Region. Br J Gen Pract. 1997, 47: 297-300.PubMedPubMed CentralGoogle Scholar
- Dighe AM, Grace JF: General practice management of childhood urinary tract infection. J R Coll Physicians London. 1984, 34: 324-327.Google Scholar
- Paulus DL: Measurement and control of response bias. Measures of personality and social psychological attitudes. Edited by: Robinson JP, Shaver PR and Wrightsman LS. 1991, San Diego (CA), Academic Press, 17-59.View ArticleGoogle Scholar
- Esbjörner E, Berg U, Hansson S: Epidemiology of chronic renal failure in children: A report from Sweden 1986-1994. Pediatr Nephrol. 1997, 11: 438-442. 10.1007/s004670050312.View ArticlePubMedGoogle Scholar
- Grol R: Beliefs and evidence in changing clinical practice. Brit Med J. 1997, 315: 418-421.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2296/8/9/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.