- Piero Impicciatorea, research fellow,
- Chiara Pandolfini, research fellowa,
- Nicola Casella, research fellowa,
- Maurizio Bonati, heada
- a Laboratory for Mother and Child Health Istituto di Ricerche Farmacologiche “Mario Negri”, Via Eritrea 62, 20157 Milan, Italy
- Correspondence to: Dr Bonati Mother_Child@irfmn.mnegri.it
- Accepted 28 May 1997
Objective: To assess the reliability of healthcare information on the world wide web and therefore how it may help lay people cope with common health problems.
Methods: Systematic search by means of two search engines, Yahoo and Excite, of parent oriented web pages relating to home management of feverish children. Reliability of information on the web sites was checked by comparison with published guidelines.
Main outcome measures: Minimum temperature of child that should be considered as fever, optimal sites for measuring temperature, pharmacological and physical treatment of fever, conditions that may warrant a doctor's visit.
Results: 41 web pages were retrieved and considered. 28 web pages gave a temperature above which a child is feverish; 26 pages indicated the optimal site for taking temperature, most recommending rectal measurement; 31 of the 34 pages that mentioned drug treatment recommended paracetamol as an antipyretic; 38 pages recommended non-drug measures, most commonly tepid sponging, dressing lightly, and increasing fluid intake; and 36 pages gave some indication of when a doctor should be called. Only four web pages adhered closely to the main recommendations in the guidelines. The largest deviations were in sponging procedures and how to take a child's temperature, whereas there was a general agreement in the use of paracetamol.
Conclusions: Only a few web sites provided complete and accurate information for this common and widely discussed condition. This suggests an urgent need to check public oriented healthcare information on the internet for accuracy, completeness, and consistency.
Fever in children is a common problem, and accurate information on home management of feverish children could be useful for parents
A systematic search on the world wide web for such parent oriented information retrieved 41 web pages, but only four adhered closely to published guidelines for home management of childhood fever
These findings suggest the urgent need to check public oriented healthcare information on the internet for accuracy, completeness, and consistency
Information on the internet should not be a substitute for routine care by family doctors
The rapid growth of computer based electronic communication and the shift toward giving patients more responsibility for their healthcare decisions are likely to induce cultural changes in the delivery of care.1 2 Until recently, it was extremely difficult for lay people to search for healthcare information on the internet, but applications such as the world wide web are making it more accessible.3 4 5 This changing nature of information distribution has important implications for health care: issues such as the quality of care, the validity and consistency of available information, and the effects on the doctor-patient relationship will be major concerns.6 7
Patient oriented medical information on the world wide web could provide healthcare professionals with the opportunity to learn more about patients' and relatives' concerns and to refer them to these sources of information when appropriate.8 However, little has been done to assess, control, and assure the quality of this medical information. We therefore made a systematic search of parent oriented web pages relating to the home management of children with fever in order to assess the reliability of the information.
Fever is one of the most common medical problems experienced by children and is often caused by relatively harmless, self limiting viral diseases that are manageable by parents on their own.9 However, parents' fear and misconceptions about fever often lead them to unnecessarily aggressive management and inappropriate calls to their doctor.10 Parental education programmes on fever have been shown to decrease misuse of antipyretic drugs and requests for doctors to visit.11 12 13 The availability of accurate information on the internet about the home management of the feverish child on the internet could thus be useful in counteracting parents' worries about fever and, more generally, in optimising healthcare delivery.
In December 1996, we searched the world wide web with the browser Netscape Navigator (Netscape Communication Corporation, Mountain View, CA) and located relevant web sites by using the advanced search options (with boolean operators) of the search engines Yahoo14 and Excite.15 The set of keywords we entered into the query box were: “fever management” and “child” and “parent information.” To find documents written in languages other than English, and thus reduce the language bias, we also entered the search terms in French, Spanish, Italian, and German.
We prepared a checklist to gather basic information on the web sites–such as the type of organisation that created the web site, the country it operates from, and the language in which the information was offered. We also considered more specific items relating to fever and its management:
The minimum temperature considered as fever
The optimal sites for measuring temperature
Pharmacological and physical treatments of fever
Conditions that may warrant a visit by a doctor.
To assess the reliability and completeness of the web sites, we compared their information with the guidelines to parents for managing fever at home supplied by El-Radhi and Carroll.16 The main recommendations contained in these guidelines and considered in the analysis were:
To take the child's temperature in the armpit
To give paracetamol in a dose of 10-15 mg/kg every four hours
For sponging, to use tepid water that feels neutral to the touch and to always give paracetamol before sponging
To keep the child lightly dressed, offer fluids frequently, and discourage excessive activity
To call a doctor if the child is less than 6 months old; has an underlying disease; seems unwell; or has symptoms such as pain, vomiting, headache, earache, excessive sleepiness, delirium, convulsion, breathlessness, persistent fever, or has had a temperature above 40°C on several occasions.
We performed data management and analysis with the Epi-Info (version 6.04b) software package.17
Our computer search retrieved 41 web pages. Of these, 32 were developed by commercial ventures, and the remaining nine were produced by individual practitioners, clinics, academic institutions, or other organisations with educational purposes. Most of the web pages (31) were created by centres in the United States; centres in Canada, the next most frequent source, created three. English alone was used in 32 of the documents, English was used together with another language in six, Spanish was used in two, and Italian was used in one.
Twenty eight web pages gave a specific temperature above which a child is considered to have fever. This ranged from 37.4°C to 38.3°C for rectal measurement (mean 38°C (SD 0.2°C)), 37.4 to 37.8°C for oral measurement (37.6°C (0.2°C)), and 37.0 to 37.6°C for armpit measurement (37.2°C (0.2°C)).
Twenty six web pages indicated the optimal sites for measuring children's temperature. The rectal method was the most widely recommended independently of age, and forehead strips were the most discouraged (see table 1). Nine web pages explained the correct way to take a child's temperature. All of these described the rectal method, while the oral and axillary methods were described in six and five web pages respectively.
Of the 34 documents that mentioned drug treatment, paracetamol was recommended in 31, ibuprofen in 14, and dipyrone in one, while aspirin was recommended in three and discouraged in 22. Of the web pages that recommended paracetamol, eight suggested the dose, which ranged from 10 to 15 mg/kg at intervals of either four or six hours.
Non-drug remedies were indicated in 38 documents. Increasing the intake of fluids, tepid sponging, and dressing lightly were the most commonly recommended, while sponging with alcohol and bathing in cold water were the most discouraged (table 2). Of the 22 web pages that recommended tepid sponging, seven also specified the temperature above which to begin the treatment; this ranged from 38.5°C to 40.5°C. However, only six mentioned the importance of giving an antipyretic drug before sponging.
Thirty six web pages gave at least some indication of when parents should call a doctor. Twenty seven listed warning symptoms, of which the most common were convulsions, difficulty breathing, stiff neck, and difficulty in awakening. The age of a child below which parents should call a doctor was mentioned in 21 documents and ranged from 2 to 6 months (mean 3.5 months). The temperature and duration of fever above which a doctor's visit was advisable were reported in 15 and 18 documents respectively: the temperature ranged from 38°C to 41°C (mean 39.7°C), and the duration of fever from one to three days (mean 2.3).
Table 3) summarises the results of the comparison between the contents of the web pages and the guidelines chosen as the standard. Only four web pages adhered closely to the main recommendations listed in the guidelines. The largest deviations were in the sponging procedures (only six out of 41 adhered) and in the way the temperature should be taken (16/41), whereas there was general agreement in the use of paracetamol as the antipyretic of choice (31/41).
This systematic search of web sites indicates that common medical problems, such as managing feverish children at home, are well represented on the internet. However, unlike in paper based journals, on the world wide web there is often no verification of validity–let alone peer review–of the information submitted. Difficulty in judging the validity of this information thus poses a problem for people using the internet.
As there is no consensus in every aspect of the home management of feverish children, we expected some inconsistency in the information the various documents contained. Thus, the temperature above which a child should be considered to have fever as well as the optimal site of temperature measurement were not uniform.
There is still disagreement on the optimal site of temperature measurement, and the choice depends partly on the aim of the measurement; usually, it is merely to establish the presence of fever and its approximate level.18 Most of the web pages recommended rectal measurement. Some consider this method inappropriate for parents because of the risk of breaking the thermometer, rectal injury, and cross infections. These complications, however, have been overemphasised and are too rare for rectal measurement of temperature to be discouraged.19 20 Only nine of the 24 web documents that recommended the rectal method also gave detailed instructions for taking temperature. Because parents are often reluctant to measure rectal temperature in their children, if they are to be encouraged to use this method more information is needed on the best way to perform it.21
Fever in children is rarely harmful, and treatment may not always be necessary. The temperature necessitating treatment or a doctor's consultation is arbitrary and depends on other symptoms as well as the duration of fever.22 While appropriate management of the illness must be the central part of looking after feverish children, concern for their comfort has made the use of antipyretic drugs commonplace.23
In this review paracetamol was the most widely recommended antipyretic. This drug has an established place in managing fever in children, and doctors increasingly recommend it. Although it is extensively used and has a good safety record, there have been reports that its margin of safety for frequent therapeutic doses in infants and young children is lower than previously appreciated.24 25 However, underdosing by parents is more common than overdosing and may lead to ineffective treatment of fever and unnecessary visits to the doctor's office.26 Only a quarter of the web pages that recommended paracetamol gave specific instructions about the dose and frequency of administration. Since paracetamol is sold over the counter and feverish children are often treated without medical control it is worrying that more emphasis was not placed on the need for close adherence to recommended doses.27 28
Some web pages also recommended treatment with ibuprofen. The use of this drug as an antipyretic in children is not approved in all countries, however, and its limited safety data require that paediatric ibuprofen be made available only with prescription. Although ibuprofen and paracetamol seem to be equally safe, further surveillance is needed before ibuprofen is given the confidence afforded paracetamol.29 30
Two web pages recommended aspirin for treating fever in children. The use of salicylates should be discouraged in children with viral infections, particularly upper respiratory tract infections or varicella, because of a possible association with Reye's syndrome.31 Finally, one resource considered dipyrone as an antipyretic for children. This drug was removed from the market in Britain, the United States, and many other countries long ago because of the risk of it causing agranulocytosis.32
In addition to drugs, most documents recommended physical remedies to reduce fever. Tepid sponging was the most widely suggested method, but sponging is only rarely necessary in children because antipyretic drugs are simpler to use, at least as effective in reducing body temperature, and cause less discomfort.13 Although sponging is recommended in occasional cases of very high temperature (>40-41°C, plus discomfort), some of the documents suggested sponging children with temperatures as low as 38.5°C. Moreover, when sponging is performed it is important to first give an antipyretic drug, a requirement not adequately emphasised in most of the web pages. Sponging, in fact, does not affect the thermoregulatory set point, and a feverish child may experience more discomfort as the hypothalamus attempts to offset the decreased body temperature.33
Two web pages suggested cold sponging or sponging with alcohol as remedies for fever. These methods should always be discouraged. Cold sponging opposes the physiologically raised set point in the hypothalamus, leading to shivering, which is counterproductive and actually serves to raise body temperature.34 Inhalation of alcohol during sponging may induce hypoglycaemia and coma in children.35
Variability in both content and quality of medical information to the public is not exclusive of the internet, as wide differences also exist in other forms of public communication, such as print and broadcast media. However, only a few of the web pages we reviewed gave complete and accurate information for such a common and widely discussed condition as fever in children. This suggests that there is an urgent need to check public oriented healthcare information on the internet for accuracy, completeness, and consistency.
Parents can gain from online information about managing common health problems in their children, but it may be difficult for them to put educational messages into practice when the information they receive, as revealed in this study, is often incomplete and partly misleading. Because parents are often reluctant to follow advice that clashes with their beliefs and established practices, online information, to be effective, must not only be accurate but should be developed according to parents' perceived needs and draw on their skills and experience. The internet can be a good source of information on common health problems, but advice obtained through the world wide web should not be a substitute for routine care by a family doctor. Moreover, the expansion of public oriented healthcare information on the internet should not draw resources away from other communication media.
Additional studies on different information strategies for the home management of fever in children are needed in order to establish the most rational and effective method of influencing parents' long term attitude and behaviour. In particular, it would be interesting to investigate how access to the world wide web affects parents' healthcare behaviour and to evaluate the quality and accuracy of more traditional sources of advice for parents (childcare books and pamphlets, etc), which would probably be no better than those of the web pages.
Conflict of interest: None.