Clinical, social and relational determinants of paediatric ambulatory drug prescriptions due to respiratory tract infections in Italy

  • Marta Luisa Ciofi degli Atti
  • Marco Massari
  • Antonino Bella
  • Delia Boccia
  • Antonietta Filia
  • Stefania Salmaso
  • SPES study group
Pharmacoepidemiology and Prescription

Abstract

Objectives

Collecting information on patterns of drug prescriptions and on factors influencing prescribing decisions is fundamental for supporting the rational use of drugs. This study was aimed at investigating patterns of drug prescription in paediatric outpatients and at evaluating determinants of prescriptions for respiratory tract infections (RTIs).

Methods

We conducted a national cross-sectional survey involving primary care paediatricians and parents. Diagnoses and prescriptions made at each consultation were described. Poisson regression models were used to analyse determinants of drug and antibiotic prescriptions for visits due to RTIs.

Results

A total of 4,302 physician and parent questionnaires were analysed. These corresponded to 2,151 visits, 792 of which were due to RTIs. Drugs were prescribed in 83.4% of RTI visits, while antibiotics were prescribed in 40.4%. According to paediatricians’ perceptions, 84.2% of parents of children with a RTI expected to receive a drug prescription. Paediatricians’ perception of parental expectations was the strongest determinant for prescription of drugs and specifically of antibiotics [adjusted relative risk (RR): 1.7 and 3.6, respectively; P < 0.001]. However, in 77.1% of RTI visits, paediatricians judged themselves as not being influenced at all by parents’ expectations in their decision to prescribe.

Conclusions

This study underscores that relational factors, in particular perceived parental expectations, are one of the leading factors of drug prescriptions in paediatric ambulatory care settings, reinforcing the opinion that communication between physicians and parents can affect prescription patterns.

Keywords

Respiratory tract infections/diagnosis/drug therapy Child Parents Paediatrics Ambulatory care setting Physicians’ prescription practices Communication 

Introduction

Collecting information on patterns of drug prescriptions and on factors influencing prescribing decisions is fundamental for supporting the rational use of drugs [1]. The latter is crucial particularly for antibiotics, since inappropriate prescriptions can contribute to the emergence of antibiotic-resistant bacteria. In fact, in the last decade, over-prescription of antibiotics has been related to decreasing susceptibility of many bacteria to antimicrobial agents [2, 3], and a recent study, conducted in various European countries, showed a direct relationship between antibiotic consumption and rates of antibiotic resistance [4].

Physicians and patients both play a critical role in patterns of drug prescriptions; in fact, various studies have shown that patients’ expectations and doctors’ perception of patients’ expectations influence prescribing decisions [5, 6, 7, 8, 9]. Available data show that more than 50% of patients who consult a physician expect to receive a drug prescription [5, 6, 7, 8, 9], and perceived patient pressure has been shown to be a predictor of doctors’ prescribing behaviours [10]. Most of the above studies have considered only adult patients; nevertheless, prescriptions in children must also be evaluated for at least three reasons: (1) some drugs, particularly antibiotics, are widely used in children [11, 12]; (2) their use is often inappropriate [13, 14]; and (3) the patient-physician relationship is more difficult to investigate, because it is mediated by parents.

Most prescriptions in children below 15 years of age occur in an ambulatory setting [15]. In Italy, primary health care of children in this age group is provided free of charge by about 5,700 paediatricians contracted by the National Health System (NHS), each of whom is assigned a specific geographical area. Primary care paediatricians are paid by the NHS according to the number of children enrolled in their practices, with enrolment limited to 1,000 children/paediatrician [16]. If more than one paediatrician is present in the same area, parents can choose which paediatrician to register their child with. Alternatively, children ≥6 years of age can be registered with a general practitioner; this is the case for approximately 20% of 6- to 14-year-old children (Italian Federation of Pediatricians, personal communication).

No nationwide studies to evaluate patterns of drug prescriptions in paediatric ambulatory practices have ever been conducted in Italy. In 2002–2003, the Italian National Health Institute (Istituto Superiore di Sanità-ISS) therefore conducted a national study based on primary care paediatricians and parents of children enrolled in their practices. The main objectives of the study were: (1) to describe drug prescriptions in the paediatric ambulatory setting and (2) to evaluate determinants of prescriptions for respiratory tract infections, which are reported to be the most frequent diagnosis in paediatric outpatient visits [17]. The main results of this study are presented in this article.

Methods

Study design

The study was designed as a national cross-sectional survey. Participants were Italian National Health System primary care paediatricians who voluntarily participate in a paediatric sentinel network for the surveillance of vaccine-preventable diseases (SPES) [18], and parents of children enrolled in their practices. From a list of 482 paediatricians who participated in SPES in the year 2002, 150 paediatricians were randomly selected dand invited by mail to take part in the study; 71 physicians (47%) accepted to participate.

The study was conducted during three predefined working days, each falling in a different season (fall: 13 November 2002; winter: 19 February 2003; spring: 18 June 2003); a time window of ±1 day with respect to each established study date was allowed.

Data collection procedures

Paediatricians were asked to complete a questionnaire for each office consultation performed on the predefined study dates. Parents of children examined on the study days were also asked to complete a questionnaire. Questionnaires for physicians and parents were self-administered and had been previously pilot tested.

The paediatricians’ questionnaire (see Appendix) included questions regarding: child demographic information; date and duration of the consultation (<15 min, 15–30 min, >30 min); clinical diagnosis; presence of fever; other visits performed in the previous month for the current illness, and commercial names of drugs already in use by the child at the time of the visit; commercial names of drugs prescribed during the visit; perception of parents’ level of concern regarding their child’s illness; perception of parental expectations regarding a drug prescription; explicit requests for medicines made by parents and self-judgment regarding the degree to which paediatricians felt influenced in their treatment choices by perceived parental expectations; level of satisfaction with the amount of time allotted to the visit.

The parents’ questionnaire included questions concerning: demographic information on the child and the family; parents’ perception of the severity of their child’s clinical condition; perceived need and explicit requests for a drug prescription; level of satisfaction with the duration of the visit (Appendix). Parents filled out the questionnaire in the waiting room; the questionnaire was then placed in a sealed envelope so that the paediatrician was blinded with respect to its content.

All questionnaires were mailed to the ISS, where data were entered into a relational database developed in Microsoft Access 2000 (Microsoft, Redmond, WA, USA).

Statistical analysis

For each consultation, paediatrician and parent questionnaires were linked by a unique code identifier. Only linkable questionnaires were included in the analysis. Diagnoses were grouped into the following ten categories: (1) respiratory tract infections (subdivided into: acute otitis media (AOM), other upper respiratory tract infections and lower respiratory tract infections); (2) gastroenteritis and other gastrointestinal diseases; (3) other infectious diseases; (4) skin disorders; (5) asthma and bronchospasm; (6) allergic diseases other than asthma; (7) fever of unknown origin; (8) traumas and accidents; (9) signs and symptoms not otherwise specified; and (10) other diagnoses.

Drugs were coded according to the Anatomical Therapeutic Chemical (ATC) classification system [19].

Parents’ educational level was coded by referring to the Italian school system, as “low” if the parent had no schooling or had attended primary or middle school (≤8 years of schooling) and as “high” if the parent had attended high school or university (>8 years of schooling).

Categorical variables were compared using the χ2 test; continuous variables were compared using the Mann-Whitney test.

Analysis of prescription determinants was limited to visits due to respiratory tract infections (RTI). Two Poisson regression models with a robust error variance were used to estimate adjusted relative risks and their confidence intervals [20]. The first model assessed the association of predictor variables with overall drug prescriptions, while the second model assessed the role of these variables on antibiotic prescriptions only (ATC code J01). Factors associated with each outcome by univariate analysis (P < 0.10) were considered eligible to be included into the multivariate model and retained in the final model according to a log likelihood ratio test for goodness-of-fit.

In order to further investigate the determinants of antibiotic prescriptions in ambulatory visits due to RTIs, a multivariate model was also run for each of the three RTI diagnostic categories (i.e. AOM, other upper respiratory tract infections and lower respiratory tract infections).

All statistical analyses were carried out using STATA software version 8.2 (Stata Corporation, College Station, TX, USA). For each variable, models excluded records with missing values.

Results

Overall, 4,302 questionnaires were collected and analysed, 50% of which were filled out by paediatricians and 50% by parents. These questionnaires corresponded to 2,151 visits.

Socio-demographic characteristics of the children examined and of their parents are shown in Table 1. Parent questionnaires were filled in by the child’s mother in 91.7% of cases. Most visits occurred during the fall and winter study dates (39.3 and 36.8% in November and February, respectively, vs 23.9% in June; P < 0.0001). Consultations lasted less than 15 min in 58.1% of cases and more than 30 min in only 3.1% of cases. Most paediatricians and parents judged the time devoted to the visit as being adequate (75.2 and 81.5%, respectively).
Table 1

Socio-demographic characteristics of children and parents participating in the study; November 2002-June 2003 (total number of visits=2,151; proportions are calculated excluding missing values)

Characteristics of children

Characteristics of parents

 

Father

Mother

 

N

%

 

N

%

N

%

Mean age (years)

4.2

 

Italian nationality

1,569

97.0

2,032

97.0

Median age (years)

3

 

Level of education

    

Males

1,119

52.0

Nil/primary school

78

4.0

92

4.4

Children aged <6 years

1,480

69.1

Middle school

743

37.9

649

31.0

Siblings

  

High school

861

43.9

1,039

49.6

 0

755

35.6

University

280

14.3

312

14.9

 1

1,039

49.0

     

 2 or more

327

15.4

     

Birth order

  

Employed

1,934

98.0

1,265

60.1

 1st

1,122

53.9

     

 2nd

743

35.7

     

 3rd or higher

218

10.4

     

Diagnoses and drug prescriptions

Of the 2,151 visits, 1,323 (61.5%) were due to an acute illness, 766 (35.6%) were well child care consultations and 62 (2.9%) were vaccination visits. The distribution by diagnosis of consultations due to an acute illness is shown in Fig. 1. RTIs were the most frequent diagnosis, accounting for 36.8% of all visits (792/2,151) and for 59.9% of visits due to an acute illness (792/1,323). Each of the other diagnostic categories accounted for less than 10% of acute illness consultations.
Fig. 1

Distribution by diagnosis of visits due to acute illness (N = 1,323)

Of the 1,323 acute care visits, 980 (74.1%) were first medical consultations for the current illness. In 89.9% of cases the diagnosis was based solely on the clinical picture (1,190/1,323) and further investigations were requested in only 13.8% (164/1,190) of these cases.

Nearly half of the children examined for an acute illness (617/1,323, 46.6%) were already taking at least one medication at the time of the visit, for a total of 752 different medicinal products. The most frequently used drugs were respiratory tract medications (ATC codes R03 and R05, 260/752, 34.6%), antipyretics/analgesics (ATC code N02, 199/752, 26.5%) and antibiotics (ATC code J01, 88/752, 11.7%). Medications had been prescribed by the primary care paediatrician in 50.6% of cases (312/617), self-prescribed by the parents in 28.4% of cases (175/617), prescribed by another physician in 7.8% of cases (48/617) and recommended by a pharmacist in 3.1% of cases (19/617). This information was not available for the remaining 10.0% of cases (62/617).

At least one drug was prescribed in 75.1% of acute care visits (993/1,323), and the total number of prescribed drugs was 1,394, giving a mean of 1.4 drugs per visit that resulted in a prescription. Categories of prescribed drugs, listed according to the 14 anatomical groups of the ATC classification system, are shown in Fig. 2. Anti-infectives for systemic use (ATC class: J) were the most frequently prescribed drugs, representing 31% of total prescriptions (434/1,394). Almost all anti-infectives prescribed were antibiotics (ATC: J01) (419/434, 96%). More specifically, 28.1% of acute care visits resulted in a prescription of at least one antibiotic (372/1,323).
Fig. 2

Frequency of drug prescriptions by ATC group (N = 1,394)

Visits due to respiratory tract infections and determinants of drug prescriptions

The mean and median ages of the 792 children with a diagnosis of RTI did not significantly differ from those of children with other diagnoses (4.4 and 4 years vs 4.8 and 4 years, respectively). Of the 792 respiratory tract infections, 127 (16.1%) were diagnosed as AOM, 588 (74.2%) as other upper respiratory tract infections and 77 (9.7%) as lower respiratory tract infections.

In 89.9% of cases (712/792), the diagnosis of RTI was based solely on the clinical picture; this rate was similar to that observed in children with acute care visits due to other illnesses (478/531, 90.0%). However, further investigations were requested for only 9.0% of the children with clinically diagnosed RTIs (64/712), significantly less than that observed for other diagnoses (100/478, 20.9%; P < 0.0001).

The proportion of children with a RTI who were already taking medicines at the time of the visit was 52.3% (414/792); this proportion was significantly higher than that observed in children with other diagnoses (203/531, 38.2%; P < 0.0001).

In addition, children diagnosed with a RTI were found to have a significantly higher drug prescription rate than children with other diagnoses (661/792, 83.4% vs 332/531, 62.5%; P < 0.0001). Differences between the two groups were even more striking when considering antibiotic prescriptions (320/792, 40.4% vs 52/531, 9.8%; P < 0.0001). Antibiotics were prescribed in 31.3% of upper respiratory tract infections other than AOM (184/588), in 61.0% of lower respiratory tract infections (47/77) and in 70.1% of visits due to AOM (89/127).

According to paediatricians’ perceptions, 84.2% of parents of children with a RTI expected to receive a drug prescription (667/792); however, in 77.1% of cases (611/792), paediatricians judged themselves as not being influenced at all by parents’ expectations in their decision to prescribe a drug.

An explicit request for a drug by the parents was reported by 10.2% of paediatricians (81/792) and by 15.4% of parents (122/792; P = 0.0020).

Prescription determinants in visits due to RTI were evaluated by multivariate analysis. Variables included in the multivariate models and model results are shown in Table 2.
Table 2

Multivariate models for evaluation of prescription determinants; visits due to respiratory tract infections (total number of visits=792; proportions are calculated excluding records with missing values)

Prescription determinants

All drug prescriptions

Antibiotic prescriptions

 

Univariate

Multivariate

 

Univariate

Multivariate

Na

%

RR

P

RRadj

P

Na

%

RR

P

RRadj

P

Season

Not selected by the final model

 Fall

276

86.5

1.09

0.068

1.11

0.026

 Winter

274

83.0

1.05

0.339

1.09

0.105

 Spring

110

79.1

Ref.

Ref.

Diagnosis

 Upper RTI other than AOM

479

81.9

Ref.

Ref.

184

32.4

Ref.

Ref.

 AOM

117

92.1

1.13

<0.001

1.13

0.001

89

71.2

2.20

<0.001

2.24

<0.001

 Lower respiratory tract infections

65

84.4

1.03

0.563

1.05

0.259

47

61.0

1.88

<0.001

1.82

<0.001

Presence of fever at the visit

 No

459

81.2

Ref.

Ref.

187

34.1

Ref.

Ref.

 Yes

196

90.3

1.11

<0.001

1.10

0.003

131

60.9

1.79

<0.001

1.82

<0.001

Previously visited in the last month

Not selected by the final model

 No

451

86.1

Ref.

Ref.

 Yes

168

77.8

0.90

0.012

0.92

0.020

Perceived parents’ concern about child’s current illness

Not selected by the final model

      

 Justly worried

265

42.9

1.31

0.066

1.47

0.007

 Overly worried

34

32.7

Ref.

Ref.

 Very little worried

16

50.0

1.53

0.060

1.73

0.020

Perceived parents’ expectations of a drug prescription

 Low

46

52.9

Ref.

Ref.

15

17.4

Ref.

Ref.

 Moderate

417

85.5

1.62

<0.001

1.55

<0.001

182

38.3

2.20

0.001

2.18

0.001

 High

172

97.7

1.85

<0.001

1.72

<0.001

111

64.2

3.68

<0.001

3.57

<0.001

Influence of parents’ expectations on drug prescription as reported by the physician

Not selected by the final model

 Not at all

503

82.6

Ref.

Ref.

 Moderate/high

139

92.1

1.11

<0.001

1.10

0.001

Mother’s educational levelb

Not selected by the final model

 Low

253

88.5

1.09

0.006

1.10

0.002

 High

386

81.3

Ref.

Ref.

aNumber and % of visits with prescriptions

bLow: <8 years of schooling, high: >8 years of schooling

The strongest determinant for a drug prescription (for any drug as well as for antibiotics) was found to be paediatricians’ perceptions of parental expectations. In fact, when parents were perceived as “slightly” expecting a prescription, the adjusted relative risk (RRadj) of receiving one was 1.5 for any drug, and 2.2 for antibiotics, compared to parents who were perceived as not expecting a drug prescription. Adjusted relative risks rose to 1.7 and 3.6 if parents were instead perceived as “very much” expecting a drug prescription.

The prescription of all drugs and of antibiotics was also significantly associated with a diagnosis of AOM or of lower respiratory tract infection, as well as with the presence of fever. The association between these variables and prescriptions was stronger for antibiotics than for all drugs.

Drug prescriptions were significantly more frequent in the fall study date with respect to the spring date in children whose mothers had a low educational level, as well as in cases where paediatricians self-judged themselves as being either “slightly” or “very much” influenced in their prescribing decisions by perceived parental expectations.

On the other hand, antibiotics were significantly less frequently prescribed in cases where paediatricians perceived parents as being overly concerned about their child’s clinical condition, while they were more frequently prescribed in cases where parents were perceived as not being adequately concerned.

Results of the multivariate analysis, which evaluated antibiotic prescriptions by diagnosis, were similar to those obtained in the RTI multivariate model. In fact, perceived parental expectations remained the major determinant of antibiotic prescriptions in visits due to both AOM and other upper respiratory tract infections (Table 3). In visits due to lower respiratory tract infections, the association between perceived parental expectations and antibiotic prescription was not statistically significant, but it should be noted that the number of children in this group was small (77), and the confidence intervals (CI) of the adjusted RRs were wide (e.g. children whose parents were perceived as very much expecting an antibiotic prescription vs children whose parents were perceived as not expecting one at all: RRadj 2.7, 95% CI: 0.7–10.2).
Table 3

Multivariate models for evaluation of antibiotic prescription determinants by diagnosis; visits due to respiratory tract infections

Antibiotic prescription determinantsa

Upper RTI other than AOM (No. 588)

AOM (No. 127)

Lower respiratory tract infections (No. 77)

 

Univariate

Multivariate

 

Univariate

Multivariate

 

Univariate

Multivariate

Nb

%

RR

P

RRadj

P

Nb

%

RR

P

RRadj

P

Nb

%

RR

P

RRadj

P

Presence of fever at the visit

 No

86

22.2

Ref.

Ref.

69

68.3

Ref.

Ref.

32

53.3

Ref.

Ref.

 Yes

97

55.1

2.49

<0.001

2.29

<0.001

19

86.4

1.26

0.090

1.27

0.018

15

88.2

1.65

0.010

1.29

0.137

Perceived parents’ concern about child’s current illness

 Justly worried

150

33.3

1.23

0.274

1.67

0.004

78

73.6

1.26

0.267

1.37

0.231

37

60.6

1.33

0.508

1.37

0.411

 Overly worried

22

27.2

Ref.

Ref.

7

58.3

Ref.

Ref.

5

45.5

Ref.

Ref.

 Not adequately worried

9

40.9

1.51

0.215

2.68

0.001

3

50.0

0.86

0.999

0.99

0.980

4

100.0

2.20

0.103

1.96

0.079

Perceived parents’ expectations of a drug prescription

 Low

7

10.9

Ref.

Ref.

6

42.9

Ref.

Ref.

2

25.0

Ref.

Ref.

 Moderate

104

29.1

2.66

0.002

2.76

0.008

53

68.9

1.61

0.063

1.56

0.159

25

62.5

2.50

0.115

2.31

0.201

 High

68

56.7

5.18

<0.001

6.01

<0.001

26

86.7

2.02

0.004

1.97

0.033

17

73.9

2.96

0.032

2.71

0.141

aOther variables were not included in any of the three models

bNumber and % of visits with antibiotic prescriptions; proportions are calculated excluding records with missing values

Discussion

Italy is one of the European countries with the highest prevalence of antibiotic resistance by Streptococcus pneumoniae, one of the leading causative agents of paediatric infections [21]. It is highly probable that the excessive use of antibiotics in our country contributes to this scenario, as suggested by a study showing that the Italian regions with the highest outpatient antibiotic prescription rates were also those with the highest prevalence of antibiotic resistance [22]. For this reason, the evaluation of determinants affecting prescription of antimicrobials in the paediatric ambulatory setting may have a large clinical and public health impact.

The main strengths of this study are its nationwide scale and the involvement of both paediatricians and parents. Our results show that close to 30% of outpatient paediatric visits resulted in an antibiotic prescription. This proportion is similar to the visit-based prescribing rates reported in the USA in the late 1980s, which have, however, significantly decreased in subsequent years [23]. Since 86% of visits resulting in an antibiotic prescription were due to a RTI (320/372), the evaluation of prescription determinants related to this diagnosis can provide relevant information for designing interventions aimed at improving prescribing habits.

In this study, determinants of drug prescriptions in visits due to RTIs were found to include clinical variables (diagnosis, presence of fever), as well as environmental (season), social (maternal educational level) and relational factors. The major prescription determinant was the paediatrician’s perception of parental expectations: for all drugs, the proportion of visits in which a drug was prescribed rose from 53% in cases where parents were perceived as not expecting a drug prescription to 98% in cases where parents were perceived as strongly expecting one, with an adjusted RR of 1.7. The role of perceived parental expectations was even more clearly evident for antibiotics where prescription rates for all RTIs rose from 17 to 64%, with an adjusted relative risk of 3.6. Moreover, for upper respiratory tract infections other than AOM, where antibiotic prescriptions are often inappropriate [13, 14], prescription rates rose from 11% when parents were perceived as not expecting a prescription to 57% when they were perceived as strongly expecting one, with an adjusted relative risk of 6.0.

It should be noted that in our findings the perception of parental pressure was much greater than that reported in other studies. In fact, 84% of paediatricians reported that they perceived parents as expecting a drug prescription compared to 48% reported by US paediatricians and 56% reported by UK general practitioners [6, 24].

Italian primary health care paediatricians are paid according to the number of children enrolled in their practices and not the number of consultations performed. In addition, parents are free to change their child’s paediatrician at any time. These factors may lead to higher prescription rates as physicians may feel they should acknowledge parents’ expectations in order to maintain a positive physician-parent relationship. Nevertheless, only a minority of paediatricians self-judged themselves as being influenced by parents’ expectations in their prescribing decisions.

Previous studies have shown that explicit verbal requests for drugs occur infrequently, suggesting that perception of parental expectations is often not based on explicit verbal communication [25]. This finding is confirmed by our results, where overt parental requests were reported in <20% of visits. Interestingly, this proportion was significantly lower according to physicians than to parents (10.2 vs 15.4%, respectively; P < 0.001). Determining which behavioural pattern in the doctor-parent interaction results in physicians’ misinterpretation of parental requests would help in improving communication skills.

The mother’s educational level was also found to be a determinant of drug prescriptions, as previously reported by other studies which have documented an association between low household income or parents’ education and increased drug prescriptions in children [26, 27, 28]. In contrast, the father’s educational level was not significantly associated with prescriptions.

In our study, children were accompanied to the paediatric consultation by their mothers in >90% of cases. It is likely that paediatricians may be more cautious in prescribing drugs to a child having a more highly educated mother, because she might be better informed about the potential side effects of drugs, thus resulting in paediatricians feeling more pressured about the appropriateness of the prescription. This suggests that more vulnerable groups of the population should be taken into greater consideration when designing intervention programs.

As expected, the presence of fever and a diagnosis of AOM or lower respiratory infection were associated with an antibiotic prescription. Nevertheless, a limitation of this study is the absence of data to confirm the accuracy of the diagnoses reported by paediatricians, and therefore the appropriateness of prescriptions. There is also the potential for the so-called Hawthorne effect [29]: as paediatricians knew that we were investigating prescription determinants, we cannot be sure that paediatricians’ prescribing behaviours during the study days did not divert from their normal prescribing habits.

A further limitation is that the study was conducted with a sample of primary care paediatricians who voluntarily participate in a sentinel network for surveillance of vaccine-preventable diseases. They therefore represent a selected group of paediatricians particularly interested in public health issues. The response rate to participation of the initial sample was only approximately 50%. This response rate is similar to that reported in other studies [24, 30]; however, the national representativeness of participating paediatricians may be questionable. It should nevertheless be noted that the socio-demographic characteristics of children and their families were similar to the national figures: in particular, the birth order of children and the educational attainment of the parents was highly consistent with data obtained by the National Institute of Statistics [31].

Despite these limitations, our study underscores that relational factors, in particular perceived parental expectations, are one of the leading factors of drug prescriptions in paediatric ambulatory care settings, reinforcing the opinion that communication between physicians and parents can affect prescription patterns. To reduce unnecessary prescriptions in the paediatric ambulatory care setting, training of physicians aimed at recognising the role of relational factors in their prescribing attitudes should therefore be undertaken. In addition, proper information campaigns targeting parents on the risks of misuse of drugs and antibiotics should be implemented. This is particularly necessary in Italy, given the high rate of parental pressure perceived by paediatricians in visits due to respiratory tract infections, which constitute 60% of outpatient acute care visits in children.

References

  1. 1.
    World Health Organization (2003) Introduction to drug utilization researchGoogle Scholar
  2. 2.
    Murray B (1994) Can antibiotic resistance be controlled? N Engl J Med 330:1229–1230PubMedCrossRefGoogle Scholar
  3. 3.
    Levin BR (2001) Minimizing potential resistance: a population dynamics view. Clin Infect Dis 33 (Suppl 3):S161–S169PubMedCrossRefGoogle Scholar
  4. 4.
    Goossens H, Ferech M, Vander Stichele R, Elseviers M (2005) Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet 365:579–587PubMedGoogle Scholar
  5. 5.
    Cockburn J, Pitt S (1997) Prescribing behavior in clinical practice: patients’ expectations and doctors’ perceptions of patients’ expectations—a questionnaire study. BMJ 315:520–523PubMedGoogle Scholar
  6. 6.
    Britten N, Ukoumunne O (1997) The influence of patients’ hopes of receiving a prescription on doctors’ perceptions and the decision to prescribe: a questionnaire survey. BMJ 315:1506–1510PubMedGoogle Scholar
  7. 7.
    Himmel W, Lippert-Urbanke E, Kochen MM (1997) Are patients more satisfied when they receive a prescription? The effect of patient expectations in general practice. Scand J Prim Health Care 15:118–122PubMedCrossRefGoogle Scholar
  8. 8.
    Ramsden DJ, Quinn RF, Witham M (1998) Doctors are not pressured into giving prescriptions. BMJ 316:938PubMedGoogle Scholar
  9. 9.
    Webb S, Lloyd M (1994) Prescribing and referral in general practice: a study of patients’ expectations and doctors’ actions. Br J Gen Pract 44:165–169PubMedGoogle Scholar
  10. 10.
    Little P, Dorward M, Warner G, Stephens K, Senior J, Moore M (2004) Importance of patient pressure and perceived medical need for investigations, referral, and prescribing in primary care: nested observational study. BMJ 328:444–446PubMedCrossRefGoogle Scholar
  11. 11.
    Soumerai SB, Ross-Degnan DS (1990) Drug prescribing in pediatrics: challenges for quality improvement. Pediatrics 86:782–784PubMedGoogle Scholar
  12. 12.
    Otters HBM, van der Wouden JC, Schellevis FG, van Suijelekom-Smit LWA, Koes BW (2004) Trends in prescribing antibiotics for children in Dutch general practice. J Antimicrob Chemother 53(2):361–366PubMedCrossRefGoogle Scholar
  13. 13.
    Watson RL, Dowell SF, Jayaraman M, Keyserling H, Kolczak M, Schwartz B (1999) Antimicrobial use for pediatric upper respiratory infections: reported practice, actual practice, and parent beliefs. Pediatrics 104:1251–1257PubMedCrossRefGoogle Scholar
  14. 14.
    Finkelstein JA, Metlay JP, Davis RL, Rifas-Shiman SL, Dowell SF, Platt R (2000) Antimicrobial use in defined populations of infants and young children. Arch Pediatr Adolesc Med 154:395–400PubMedGoogle Scholar
  15. 15.
    Choonara I, Nunn AJ, Kearns G (eds) (2003) Introduction to pediatric and perinatal drug therapy. Nottingham University Press, NottinghamGoogle Scholar
  16. 16.
    del Torso S, Bussi R, DeWitt TG (1997) Primary care pediatrics in Italy: eighteen years of clinical care, research, and teaching under a national health service system. Pediatrics 99:E8PubMedCrossRefGoogle Scholar
  17. 17.
    Nyquist A, Gonzales R, Steiner JF, Sande MA (1998) Antibiotic prescribing for children with colds, upper respiratory tract infections, and bronchitis. JAMA 279:875–877PubMedCrossRefGoogle Scholar
  18. 18.
    Ciofi Degli Atti ML, Salmaso S, Bella A, Arigliani R, Gangemi M, Chiamenti G et al (2002) Pediatric sentinel surveillance of vaccine-preventable diseases in Italy. Pediatr Infect Dis J 21:763–768PubMedCrossRefGoogle Scholar
  19. 19.
    World Health Organization Collaborating Centre for Drug Statistic Methodology (1999) Guidelines for ATC classification and DDD assignment. Oslo, Norway, World Health Organization. Available at: http://www.whocc.no/atcddd/atcsystem.html
  20. 20.
    Greenland S (2004) Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies. Am J Epidemiol 160:301–305PubMedCrossRefGoogle Scholar
  21. 21.
    Bronzwaer S, Cars O, Bucholz U, Molstad S et al (2002) A European study on the relationship between antimicrobial use and antimicrobial resistance. Emerg Infect Dis 8(3):278–282CrossRefGoogle Scholar
  22. 22.
    Boccia D, Spila Alegiani S, Pantosti A, Moro ML, Traversa G (2004) The geographic relationship between the use of antimicrobial drugs and the pattern of resistance for Streptococcus pneumoniae in Italy. Eur J Clin Pharmacol 60:115–119PubMedCrossRefGoogle Scholar
  23. 23.
    McCaig L, Besser RE, Hughes JM (2002) Trends in antimicrobial prescribing rates for children and adolescents. JAMA 287:3096–3102PubMedCrossRefGoogle Scholar
  24. 24.
    Bauchner H, Pelton SI, Klein JO (1999) Parents, physicians, and antibiotic use. Pediatrics 103:395–401PubMedCrossRefGoogle Scholar
  25. 25.
    Mangione-Smith R, McGlynn EA, Elliot MN, Krogstad P, Brook RH (1999) The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics 103:711–718PubMedCrossRefGoogle Scholar
  26. 26.
    Mangione Smith R, McGlynn EA, Elliot MN, McDonald L, Franz CE, Kravits RL (2001) Parent expectations for antibiotics, physician-parent communication, and satisfaction. Arch Pediatr Adolesc Med 155:800–806PubMedGoogle Scholar
  27. 27.
    Thrane N, Olsen C, Schonheyder HC, Sorensen HT (2003) Socioeconomic factors and prescription of antibiotics in 0- to 2-year-old children. J Antimicrob Chemother 51:683–689PubMedCrossRefGoogle Scholar
  28. 28.
    Huang N, Morlock L, Lee C, Chen L, Chou Y (2005) Antibiotic prescribing for children with nasopharyngitis (common colds), upper respiratory infections, and bronchitis who have health-professional parents. Pediatrics 116:826–832PubMedCrossRefGoogle Scholar
  29. 29.
    Mangione-Smith R, Elliott MN, McDonald L, McGlynn EA (2002) An observational study on antibiotic prescribing behaviour and the Hawthorne effect. Health Serv Res 37:1603–1623PubMedCrossRefGoogle Scholar
  30. 30.
    Clavenna A, Bonati M, Rossi E, Berti A, De Rosa M (2004) Il profilo prescrittivo della popolazione pediatrica italiana nelle cure primarie. Ric Prat 20:224–244Google Scholar
  31. 31.
    Istituto Nazionale di Statistica (2003) Rapporto annuale; la situazione del Paese nel 2002. MaggioGoogle Scholar

Copyright information

© Springer-Verlag 2006

Authors and Affiliations

  • Marta Luisa Ciofi degli Atti
    • 1
  • Marco Massari
    • 1
  • Antonino Bella
    • 1
  • Delia Boccia
    • 2
  • Antonietta Filia
    • 1
  • Stefania Salmaso
    • 1
  • SPES study group
  1. 1.Reparto Malattie Infettive, Centro Nazionale di Epidemiologia, Sorveglianza e Promozione della SaluteIstituto Superiore di Sanita’RomeItaly
  2. 2.London School of Hygiene and Tropical Medicine, Department of Infectious and Tropical DiseasesClinical Research UnitLondonUK

Personalised recommendations