Telephone Consultation—What Standards Should Apply?

The article by Martina Albrecht and her co-authors (1) addresses an important aspect of patient-oriented medical decision making in Germany: to what extent does the advice dispensed by the freely accessible telephone consultation services correspond with the evidence? In other words: Can people who turn to these services understand and trust the information they are given? Only six of 293 consultations fulfilled the criteria defined by the authors—on the face of it, an alarming outcome.

A glance at the study’s methods

Such unequivocal results are rarely found in scientific studies. For precisely that reason, the methods employed by Albrecht and co-authors have to be scrutinized. Although the authors certainly deserve congratulation for shining a light on a hitherto neglected area, we wonder whether the situation is as bad as they claim. We have to ask whether the authors’ predefined standards are adequate. Or do the results say more about the quality of these standards than that of the information received by the hidden clients?

Discussion of the results also highlights a classic dilemma of the translation of the findings of evidence-based medicine into the situation of consultation: According to the strict rules applied by Albrecht and co-authors, all information must be conveyed. However, it is by no means clear whether this is always in the patient’s best interests.

Albrecht and co-authors employ two criteria that must be met for a consultation to be considered adequate:

• Correct content: the advice must correspond to the latest research findings.

• Intelligibility: the information must be conveyed in terms that the caller understands.

Both of these criteria must be fulfilled for the answer to be classified as adequate. We are not convinced it is appropriate to combine these two criteria, because if they are not met the reasons are likely very different and improvement of the two aspects of consultation may well require differing interventions.

We also deem the concrete formulation of the criteria problematic. According to the authors, only answers expressed quantitatively are readily comprehensible to the lay person. But is this always appropriate in telephone conversations, especially for decimal places? Is “almost 100%” really harder to understand than 99.5%? Such rigidity may—this cannot be assessed from the data provided—have drastic consequences for the number of answers the authors classify as intelligible Table in [1]).

With regard to empirical correctness, the most important criterion of the consultations’ foundation in evidence, Albrecht and co-authors evaluate the responses to six different medical decision-making scenarios using standards they developed themselves (eBox 2 in [1]). Apparently one false statement concerning one single aspect sufficed for the whole consultation to be classified as inadequate. The six topics chosen by the authors are of varying complexity. Lack of space prevents us discussing the appropriateness of all the criteria, but we will examine colonoscopy as an example.

Consultation on bowel cancer screening

The standards can be found in abbreviated form in the Table of the article by Albrecht et al. (1). Now, the benefit of the various programs for early detection of cancer is a controversial subject. From our perspective, however, the evidence with regard to bowel cancer screening is particularly clear (2).

The authors seem not to agree. According to their Table (1), question 3 is “Is it clear that people benefit from having this done? Does it reduce the number of people who die of bowel cancer?” The answer they are looking for is: “The benefit cannot be quantified, because no randomized controlled trials have yet been published. ”

We find this questionable. It is true that there are no randomized controlled trials (RCTs), but large, high-quality epidemiological studies on this topic, both case–control studies and cohort studies (3), were already available in the year of the survey (2013). As in many other fields of medicine, these studies do not permit conclusive quantification—incidentally, neither would one or multiple RCTs—but statements on the surmised effects can be made on the basis of the published data. Moreover, according to the advice of the German Evidence-Based Medicine Network (Deutsches Netzwerk Evidenzbasierte Medizin) on good practice in health information (4)—cited by Albrecht and co-authors—such studies should also be mentioned.

As for the risks of colonoscopy, the authors require death to be stated as a major risk without specification of frequency. According to the literature, the rate of death caused by colonoscopy is around 1 : 10 000 (5). We believe the answer demanded by the authors would serve particularly to scare patients off. Other risks such as perforation and bleeding could also, in our view, have been described in more detail.

According to the above-mentioned criteria for good practice, age- and sex-specific information should be provided. Now, we know that complications of colonoscopy increase with the patient’s age and occur more often in men. Therefore, a 56-year-old woman (the fictive patient in this scenario) does not have a 0.01 to 0.1% risk of perforation or a 0.2 to 0.3% risk of bleeding, as stated by the authors.

The correctness of these standards can thus be hotly debated (6). In many cases the advisors gave no numerically expressed answers at all, leading automatically to downgrading of the consultation by the authors. However, whether an answer such as “Bleeding may occur, but is rare” is really of no use to the caller seems dubious.

Conclusion

Albrecht and co-authors are truly to be congratulated for publishing an article showing that:

• Problems occurred during telephone consultations

• By no means all advisors gave correct answers to all questions

• The advisors not infrequently shied away from making quantitative statements

It can therefore be hoped that this study will prompt further investigation and improvement of telephone consultation services—including those offered by health insurance providers, which were unfortunately not included. We doubt, however, that Albrecht and co-authors’ harsh verdict will be unreservedly upheld.