Family Practice Vol. 20, No. 5, 583-591
© Oxford University Press 2003
Health Services Research |
New drug uptake: qualitative comparison of high and low prescribing GPs’ attitudes and approach
Prescribing Research Group, Department of Pharmacology and Therapeutics, The Infirmary, 70 Pembroke Place, Liverpool L69 3GF, UK.
Correspondence to Helen Prosser; E-mail: H.Prosser{at}liverpool.ac.uk
Prosser H and Walley T. New drug uptake: qualitative comparison of high and low prescribing GPs’ attitudes and approach. Family Practice 2003; 20: 583–591.
Received 4 November 2002; Revised 25 April 2003; Accepted 19 May 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Prescribing varies considerably between GPs, and a minority of GPs account for a large proportion of new drug prescribing.
Objective. Our aim was to explore the differences in approach and attitudes towards new drug prescribing between high and low GP prescribers.
Methods. Semi-structured interviews were undertaken with 30 GPs who were either high or low prescribers of new drugs. Interview topics included reasons for prescribing new drugs; use of information sources; attitudes and approach to drug innovation; and consideration of the balance of risk and therapeutic benefit.
Results. Differences in GPs’ appraisal of new drugs enabled the development of a classification of attitudes and approaches applying to high and low prescribers. Difference in the behaviours of each group is a matter of degree of attitude or influence rather than (usually) its absence. Key dimensions in the classification are attitudes towards new drug prescribing, perception of risk and benefits, strategies adopted for risk management, information-seeking behaviour and use of the pharmaceutical industry as a significant information source.
Conclusion. Variability in GP prescribing of new drugs relates not only to levels of acquired knowledge, but also to differences in subjective and ideological beliefs.
Keywords. General practice, GP attitudes, GP behaviour, new drugs, prescribing influences.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Prescribing new drugs is a major driver in prescribing costs and may expose patients to risks of adverse drug reactions. Key factors associated with new drug uptake are doctor characteristics,1 patients2 and personal contact,3 including colleagues, hospital consultants4,5 and the pharmaceutical industry.6,7 However, the uptake of new drugs varies among GPs, with some GPs very willing to prescribe; in one study, 42% of all prescriptions for a range of new products were issued by only 10% of GPs.8 Attempts to characterize early and late prescribers have found early new drug uptake to be associated with larger list sizes and reliance on commercial sources of drug information.9 Interest in the concept of early or late adopters as a reliable categorization for all new drugs has generated inconsistent results,10,11 although late prescribers of one drug tend to be late adopters of other drugs.10 It has been suggested that the type of drug and perception of its risk also influence the process of new drug uptake. An early study showed that the use of professional information sources and the time taken to adopt a drug increase in relation to drugs categorized as low, medium and high therapeutic risk.12
Many previous studies on variation in new drug prescribing have been quantitative, while there has been little qualitative examination of whether GPs’ attitudes towards new drugs, therapeutic innovation, risk and evaluation of evidence may contribute to variation. This study describes one component of a study identifying the factors that influence new drug adoption.2 Here we explore the differences in approach and attitudes towards new drug prescribing between GPs who were either high or low prescribers of new drugs.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A list of all new drugs (i.e. those labelled with a black triangle by the Committee on Safety of Medicines) identified for use in primary care (n = 19) was compiled from those launched between January 1998 and March 1999 (Table 1
). Two health authorities provided the rate of practice prescribing per head of population by general practices. Initial sampling was based on stratifying practices into tertiles according to the level of prescribing and defined as high, medium and low prescribing practices. Fifty-four from a total number of 161 practices (30%) were chosen at random to include each tertile. All GPs from identified practices were invited to participate by letter and followed-up with a telephone call. A total of 107 GPs agreed, a participation rate of 73% of GPs and 77% of practices.
|
GPs reported how many of the listed drugs they had prescribed. To ensure maximal variation, further sampling and analysis was based on 30 interviews conducted with participating outlying prescribers, i.e. the highest prescribers (all those who reported prescribing
9 study drugs, n = 17) and lowest prescribers (all those who reported prescribing 
2, n = 13). Incidents where prescribing was based solely on others’ decisions e.g. recommendation from a hospital doctor, did not contribute to the present analysis.
Interviews and analysis
Semi-structured interviews were undertaken between August 1999 and February 2000. Interviews were tape-recorded and transcribed. The critical incident technique was used to allow doctors to describe the range of factors that determined their initial prescribing of the study drugs. This technique gives a record of the purpose of and reasons for specific behaviours from those in the best position to make the necessary observations and evaluations.13 Taking each prescribed drug in turn, doctors were asked about:
- the context in which the drug was prescribed;
- information acquisition and factors influencing prescribing; and
- reasons for prescribing the new drug rather than alternatives.
For GPs who had not prescribed any study drugs, questions included determinants in new drug usage and factors influencing the prescribing of a new drug compared with alternatives. In addition, all prescribers were asked about their use of information sources, attitudes and approach to drug innovation and consideration of the balance of risk and therapeutic benefit.
Construction of the classification model
The constant comparative method14,15 was used to develop categories and classification. This involved comparing one respondent’s views and behaviour with another’s, testing emerging explanations with the data and moving back and forth between analysis and explanation. First of all, we searched for the common features between the two groups. We then searched for unique and overlapping features of each group. An important element was to search for exceptions to determine whether the categories needed to be modified to encompass these differences. Transcripts were analysed independently by a sociologist (HP) and a random subset by a clinical pharmacologist (TW). Themes, categories and concepts were compared and subsequently agreed upon.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The categorization of high or low prescribers based on GP reported prescribing of the study drugs broadly agrees with the practice tertile stratification compiled from Health Authority prescribing data (Table 2
). High prescribers reported 173 new drug initiations (median 10), while the lowest prescribers described 19 (median 2). There were no significant differences between high and low groups in terms of gender, number of years qualified or whether they were working full/part time.
|
Table 3
compares prescribing influences. More than one influence could be cited for each incident, and the table is merely a description of the frequency of cited influences, rather than relative importance. Both groups were influenced by hospital prescribing, suboptimal response to existing therapy and patient requests. A noticeable difference, however, is that the most frequent influence for high prescribers was the pharmaceutical representative. GPs’ stated general attitude and approach to new drug prescribing concord with these influences, and the following interpretation encompasses both the specific influences GPs reported for each new drug prescribed and their general accounts of attitudes towards risk, innovation and new drug prescribing.
|
Towards a classification
In constructing the framework (Table 4
), attitudes and approaches common to both GP groups are underplayed since the objective of the study is to explore influences on prescribing variation: the classification is based, therefore, on distinguishing themes.
|
Attitudes to new drugs (Box 1
)..
Both high and low prescribing GPs reported only prescribing new drugs when they considered they offered a relative advantage over current therapy (either therapeutic or economic in the case of therapeutically equivalent alternatives) or the opportunity of treating problems where no treatment existed. Both groups also claimed responsiveness to ‘cost pressures’, recognizing that effectiveness needed to be balanced against cost, although there was no reluctance to prescribe higher cost, more effective drugs. Prescribers, however, felt that the relatively high cost of some new drugs constrained their routine use of it, e.g. rofecoxib, so that prescribing was limited to those patients in which cheaper alternatives were not tolerated or ineffective. Most GPs described themselves as cautious, although there was an apparent difference in the definition between high and low prescribers. Low prescribers’ conservative prescribing of the listed drugs concords with their overall attitude, in which cautiousness was characterized as a general reluctance to introduce new drugs early on after their launch, preferring to use an established, clinically familiar therapy. In contrast, high prescribers had less reservation about the early use of new drugs, although, within this group, GPs’ willingness to prescribe varied according to contextual factors. Eight GPs reported an overall interest and willingness to prescribe new drugs, depicting themselves as curious and willing to experiment. This group was keen to keep up to date, welcomed the potential therapeutic value of new drugs and was particularly interested in ‘novel’ products —not just additions to an existing therapeutic class or new formulations of existing drugs (me-toos). Nine high prescribers, whilst stating a generally cautious approach to new drugs, placed caveats on this, stating that they were willing to use new drugs where they felt knowledgeable (e.g. familiarity through experience with similar drugs or within an area of special interest), where previous treatments had been ineffective for a particular patient or where there was no existing treatment. Cautiousness used by high prescribers relates to prescribing under these latter conditions or, alternatively, to mean initially prescribing a new drug in a relatively small number of patients. An often primary, source of constraint on high prescribers was the relatively high cost of some new drugs. | Box 1 Attitudes towards new drugs Reluctance to prescribe/cautiousness "I’m just cautious and sceptical. I’m not in the early waves of up-takers. I think I’ve seen too many come in and go out. I do not believe that modern drugs are the best thing that have made major advances to the health of people in this country, so I’ll often wait until such evidence appears. There’s the safety, there’s the efficacy, do we actually need it?" (GP76, low prescriber) "I tend to wait for drugs to be tried and tested. In that way I tend to be rather conservative. I’d have to be convinced that it’s effective and safe and that the specialist body with an interest in that area thought it was an effective and safe drug. I mean, this raloxifene, I read quite a few bits and pieces and that, so if you look over in MEREC and all the rest of it there’s the query over that at the moment so I wouldn’t jump in and prescribe it." (GP99, low prescriber) "I think I’m probably relatively conservative. I don’t jump to try new drugs as soon as they come out." (GP106, prescriber of 10 new drugs) Willingness to prescribe "I’m very keen on keeping up with modern medical advice, treatment advances. I have got certain habits with drugs that I will use all the time, but it’s nice to try these new ones out and get success with them. I haven’t really got a lot of time for me-toos unless they’ve got something novel to offer. But novel groups of drugs such as the SERMs and Vioxx, then I’m happy to prescribe especially when I’ve got a patient base that I feel is waiting for that drug." (GP76, high prescriber) "I’m keen to try something if it appears to be a good drug." (GP32, high prescriber) "I think when a drug comes out that has got a use which is novel or significantly better than anything we’ve had before, then we’re inevitably going to start trying it soon." (GP98, high prescriber) Curiosity/experimentation "You just sometimes want to sort of try it. It’s the one most recent in your memory because the drug company reps have been round so it springs to mind." (GP36, high prescriber) Q: "Why did you prescriber Accolate?" DR: "Just to try it and see. It wasn’t terribly scientific, I could have easily used Singulair but because it was new and I’d heard about it." (GP71, high prescriber) Caveats "If it’s because the existing drugs are not fulfilling a need then obviously you are going to be more likely to try it." (GP26, high prescriber) "I don’t usually prescribe them immediately. I’m more cautious than that. I usually give it time, especially if you’ve got alternatives and good alternatives that you know work, then I tend to stick with those . . . but if it’s something like detrusitol that you’re desperate for because you haven’t got a decent alternative, then yeah, you use it much more quickly." (GP94, high prescriber) "I’ve got an interest in the asthma, the diabetes and the menopause and family planning. I take new medications with caution, but I like to try those new drugs in areas I’m familiar with and if I feel they serve a need. With Evista and Vioxx, there’s a very definite niche for them." (GP7, high prescriber) Evaluation of cost and effectiveness "I think I need to be convinced, really, that they have a role and that the usual higher cost is worth it." (GP34, low prescriber) Economic cost as a constraint on prescribing "I’m quite willing to take them on, I don’t think there’s any doubt that the newer drugs are going to be more focused for clinical activity, but unfortunately in these days of cost restraint we also get hammered with the cost of them, don’t we, they’re always more expensive. So the answer is I genuinely believe that most of the newer drugs are probably more beneficial but there is a price to be paid." (GP48, high prescriber)
|
Perception and management of risk (Box 2)..
One underlying factor appearing to differentiate the prescribing of the two groups is their perspective of the benefit–risk ratio. Low prescribers regarded new drug prescribing as experimental, and raised concerns over the risks. This was expressed as the risk of doing harm or prescribing ineffectively. They considered that a new drug’s effectiveness, its therapeutic value compared with existing therapy and, more importantly, its safety could not be fully assessed at the time of its introduction. In particular, they considered that serious adverse effects are often not detected until the drug is in clinical use. Low prescribers felt that critical appraisal at introduction was often inhibited because of a lack of independent published clinical effectiveness data or economic evaluations. Previous personal experience or anecdotes of using new drugs that had caused serious adverse effects also heightened the risk perception in some low prescribers. The primary evaluation criterion for low prescribers was established, non-commercial evidence of a drug’s effectiveness and safety.
| Box 2 Perception and management of risk Concerns over safety—a deterrent to new drug prescribing "In essence I don’t feel that my patients should be guinea pigs. If the product is a good product, and is effective and useful then over the next, sort of couple of years it will become widely used and studies will be done comparing that to previously used things and at that point I may well start to use it certainly, but I’m not going to go out on a limb and use things that I know nothing about." (GP1, low prescriber) "There has to be a risk to new products. We all know of new products that have bombed because of their side-effects. Thalidomide is a good example and that could still happen and rare side effects occur." (GP2, low prescriber) Acceptance/minimization of risks "I take on board the risks that you run in terms of early problems with interactions, unexpected events, somebody’s got to prescribe them, and in the areas that you have expertise, I think, you have a duty to try them out, so that colleagues can learn from you." (GP76, high prescriber) "Once the products are licensed there’s a definite OK from authority, yes, this is a qualified product for this particular disease, so I don’t mind trying it now." (GP5, high prescriber) Risk management "Once the drug’s been around for say 24 months, that is probably a reasonable length of time to anticipate any serious side-effects, so I’m going to feel a greater ease prescribing a new drug after it’s been used by a consultant peer group for some two years." (GP 28, low prescriber) "With these drugs, I’ve just tried them out in one or two patients. I like to get first hand knowledge of a drug first. I’m sure you’ll find the same thing in all your interviews." (GP5, high prescriber) "There have been cox II inhibitors before but this one was supposedly far more selective and in fact we have got quite a number of patients who even with the best protection in the world we would not give them non-steroidals, you know, because I won’t see them suffer. Vioxx was an alternative for those, and you know I have used it. You still take the risk of bleed, but the risk is considerably less as far as I can see." (GP76, high prescriber) "I’m prepared to prescribe new drugs without having experience of others using it, but usually only if it’s a drug class that I’m used to using and know the indications for, etc. If it was a completely novel drug then I would be somewhat cautious until I know that my peers are prescribing it." (GP106, high prescriber) Clinical experience—‘trialling’ "You have a mindset that says right, I’ve got two patients that have gone out on Pariet, bring them back, see how they got on on it, did it work, any problems? You evaluate it very much on that basis. You might think, well why not just read the papers on it, but papers don’t give you that feedback. I mean this is the problem, evidence-based medicine can only be done for those things that can be measured scientifically. So you know, I’m a great fan of evidence-based medicine, but if you were to work entirely to evidence-based medicine and guidelines, no new drug would ever get on the market. You couldn’t move forward." (GP76, high prescriber) Trialling under conditions of uncertainty "You don’t know whether they’re right in the claims that it’s got less side-effects, so you just give it to people. And people come back to you and say, ‘yeah, I could use this drug, whereas I couldn’t use the other’. So, it’s fair to believe them. I’m afraid I’m not a scientist. I’m a doctor. I deal with my patients." (GP36, high prescriber)
|
Low prescribers managed the risks by adopting a ‘wait and see’ policy, delaying the use of new drugs until there was evidence of it in peer-reviewed sources, or it was validated by consensus or expert opinion (particularly hospital doctors). However, critical incident findings indicate that such validation or evidence would not provoke prescribing where the same benefits existed in an established therapy, but usually in addition to contexts when established treatment had been suboptimal, e.g. the prescribing of Tolterodine in patients unable to tolerate the side effects of oxybutynin. Therefore, where there was professional legitimization and no alternative therapy, the benefit–risk ratio often favoured new drug use.
High prescribers on the other hand appeared either to accept uncertainty and the risks involved, or considered risks to be minimal on the grounds that the approving licensing authority had considered safety. If it was a me-too drug, familiarity gained through experience with similar drugs minimized potential risks. Coupled with a greater emphasis on the therapeutic opportunity of new drugs, this meant that high prescribers had a favourable view of the benefit–risk ratio. High prescribers managed risk by first limiting new drugs initially to a small number of patients where existing therapy was considered ineffective (where the benefit–risk ratio would be greatest); or to areas where they felt particularly knowledgeable. A new drug was often first used as a niche drug, catering to a small subset of patients with a particular condition. In these situations, prescribers considered experimentation and the risks of new drugs justified by an anticipated greater therapeutic gain.
For both groups, once a drug had been prescribed, clinical experience and increasing familiarity reduced uncertainty and was considered in establishing a case for effectiveness. ‘Trialling’ was an important aspect of whether a drug would become part of a GP’s prescribing repertoire. If the initial outcome was successful, repeat prescribing is more likely. A poor outcome reduced the likelihood of similar prescribing. Nine high prescribers, however, described an experimental approach to some of the study drugs, whereby they had ‘trialled’ a drug in order to test out its claims for effectiveness or reduced side effects. Furthermore, six high prescribers reported initial prescribing without explicit knowledge of a drug’s indications, side effects or effectiveness. In some cases, this was prompted by ‘the need to do something’ (exhaustion of other possibilities), in others by experimentation and curiosity and, in one case, patient request.
Information sources (Box 3)..
The groups differed in how they acquired evidence of a drug’s benefit. Low prescribers described a gradual, cumulative process in which evidence was often gathered from a number of sources. Whilst these GPs may become aware of a drug from commercial sources, the ultimate sanction to prescribe was usually from professional sources, particularly medical journals and hospital prescribing. An exception, however, was noted in one prescribing event in which a GP described prescribing a new ‘me-too’ drug when other alternatives had proved ineffective, based entirely on the information presented by a pharmaceutical representative. Whilst, this GP ascribed to the importance of professional sources as a general rule, this incident suggests that particular individual contexts of matching a patient to a particular drug may support a benefit–risk ratio which circumnavigates this norm. This particular example emphasizes not only the importance of idiosyncratic clinical judgement, but also the push– pull tensions involved in prescribing between cautiousness towards new drug adoption and wanting to treat patients more effectively.
| Box 3 Information-seeking behaviour Hospital prescribing "I’ll only prescribe a new drug if I know the hospital are prescribing it. After that I’ll try them out in the patients after I’ve seen how they’ve got on from the hospital." (GP89, low prescriber) Use of scientific evidence "I think I’m very sceptical. I think I need quite a lot of convincing to prescribe and I tend to wait for evidence that appears in say the Drugs and Therapeutic Bulletin or in MEREC bulletins and I don’t tend to change my prescribing until it does, really. And I think that’s fair." (GP34, low prescriber) Consensus/accumulation of evidence I tend to have a quick glance at Prescriber, at the Drug and Therapeutics Bulletin and MEREC. I like those, they give a nice, steady, consistent review of things, so I tend to look at those and then look round to the immediate hospital setting and the wider conferences that I go to." (GP3, low prescriber) "I tend not to initiate them until they have been used locally for a while." (GP88, low prescriber) Colleague or consensus not requisite "If a drug has a definite niche I’d be happy to prescribe it but I’d wait for consultant or hospital endorsement if I wasn’t sure." (GP9, high prescriber) Lack of use of independent, scientific evidence "I do tend to flick through the mailings and the adverts in the sort of comic magazines rather than the high-powered ones. I think they’re more readable and they do keep you up to date with modern things." (GP98, high prescriber) "The journals take a bit of ploughing through. I tend to look at the GP magazines, GP, Doctor, Pulse, those sort of things. The independent journals and the BMJ are a bit stale really, I have trouble managing them." (GP53, high prescriber) Innovative colleagues "Dr M is a brilliant rheumatologist. He’s very, very innovative." (GP76, high prescriber). "The new consultants are the ones that influence us most and we tend to look at what they’re doing. They keep up to date and are willing to try new things. You tend to think they know more because they keep up with research." (GP68, high prescriber) "The problem about being in specialties is that you have to be the most up-to-date cutting edge, guy." (GP2, low prescriber) Convenient and useful source of information "Reps, quite honestly, are very good and they will do the donkey work for you, because if they don’t know the answer they will contact their powers that be and come back to you with the information, which is good. I mean, I think that is useful . . . no-one else is going to tell us about new drugs." (GP98, high prescriber) Industry-provided information, but prescribing change based on peer-review evidence I: "Where do you get your information from about new drugs, new products?" R: "Probably from drug reps. I don’t see a lot I probably see about one a month. I read mainly the British Medical Journal, which does have some advertising in. There’s advertisements in the papers, the journals that we get, those are the weekly ones, I tend not to look at those very much. I think it’s mainly the BMJ and the paper journals where I get evidence from." (GP3, low prescriber) Negative attitudes towards pharmaceutical representatives "I don’t trust drug companies and I don’t see drug reps. There’s no point in wasting half an hour of your time in talking to someone who you wouldn’t believe no matter what they said, is there?" (GP2, low prescriber) "We’re a bit dubious about the glossy little things that they produce, you know, because a survey of 5 patients isn’t really a study, but they can make it sound like it is." (GP48, high prescriber) Separating the wheat from the chaff "Hopefully they’re (reps) going to give you factual information about their products, I mean obviously they’re out to sell the product so there’s always going to be bias, but I think it’s possible to get fairly factual information from them if you look hard enough and if you question them appropriately." (GP14, high prescriber)
|
Low prescribers often used hospital colleagues as prescribing guides. This was either ‘eminence-based’, i.e. influenced by the opinions of knowledgeable clinicians, or based on ‘experiential’ knowledge, i.e. experience of using a drug first recommended by a hospital doctor. Two low prescribers, however, were unwilling to prescribe any new drug unless local consultants were using it. They stressed limited time to access scientific literature and used consultant opinion as a decision-making short cut. Other low prescribers considered local consultants too innovative, prescribing without sufficient evidence of a new drug’s benefit.
High prescribers were also influenced by professional trust and experiential knowledge gained from hospital prescribing, although a favourable benefit–risk perception meant that colleague endorsement or consensus was not a requisite; one prescriber thought the local hospital consultants out of touch with pharmacological advances. On the other hand, other high prescribers were keen to follow the prescribing trends of hospital colleagues they considered innovative. They declared less use of independent information sources, and some expressed criticism of medical journals as excessively time consuming, out of date or overly complex in reporting trial and research findings. Instead, they used the GP press or relied on the pharmaceutical industry for new drug information; in some instances, prescribing had been influenced principally by advertisements or mailings.
There was a noticeable trend in differences between the two groups in the frequency of visits from pharmaceutical representatives, with high prescribers reporting greater frequency. All but one high prescriber received personal visits from representatives, these occurring at least weekly, whilst seven saw at least three a week. Four low prescribing GPs never saw representatives (criticizing information as distorted and selective and recognizing that major therapeutic innovations would be reported in medical journals), and none saw more than one representative a week. Whilst there were GPs in both groups who found representatives to be a convenient and useful source of information, especially for keeping up to date with new products, a major difference appears to be in their use of representative information. Whilst both groups criticized representative-provided information as selective and overly positive, low prescribers tended to seek additional professional information whilst high prescribers selected the information they perceived to be relevant and valid—‘separating the wheat from the chaff’—and often relied on this information for decision making.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study suggests that variability in GP prescribing of new drugs relates not only to levels of acquired knowledge, but also to differences in subjective and ideological beliefs. The classification of attitudes towards new drug prescribing between high and low prescribers stresses differences, but in reality it is not a simple dichotomy and the behaviour of each group is usually a matter of degree of attitude or influence rather than (usually) its absence. Key dimensions in the classification are attitudes to innovation, risk perception and benefit (generally perceived as favourable by high prescribers but unfavourable by low prescribers), strategies for risk management, information-seeking behaviour, and the importance of interpersonal contact in influencing prescribing, via both the pharmaceutical industry and the medical profession. Attitudes and approach towards new drug prescribing vary across a spectrum and represented here are the extremes. The manner in which each GP approaches prescribing is dynamic, involving any of the classification criteria, and an interplay of attitude and information-seeking behaviour. It is likely that most prescribers will hold a mixture of conflicting attitudes over new drug adoption, on the one hand favouring cautiousness whilst at the same time desiring to treat patients more effectively. Approach and practice are flexible and the eventual outcome determined by the type of drug and the individual patient context.
Risk perception is also dynamic, shaped by clinical experience, personal beliefs, the perceived weight of evidence and effectiveness of alternatives. Low prescribers frequently cited fear of serious adverse effects, and wanted an accumulation of information. Some wanted consensus of opinion, in particular the expert opinion evidenced by hospital prescribing. For others, evidence-based acquired knowledge was the most powerful determinant. High prescribers were more inclined to underplay the risks of new drugs, so freeing them to prescribe a new drug they believed offered improved therapeutic effectiveness or, in some cases, to satisfy personal curiosity. Both groups used strategies to minimize risk, a common thread being the influence of hospital prescribers. PCTs will need to work closely with local hospitals in order to develop joint formularies and prescribing protocols.
The frequency of pharmaceutical representative visits is a key difference between the two groups. However, whether high prescribers’ relative willingness to prescribe new drugs is shaped by the pharmaceutical industry or merely reinforced by such contact is difficult to say, but the frequency of citation of the representative as a primary information source and often an influence suggests the former. Some high prescribers held mixed views about new drug prescribing, stating cautiousness yet expressing keen interest in new drugs as therapeutic opportunities. These ‘pull–push’ factors provide tensions for prescribers, and accepting information from the pharmaceutical industry may be influential in tipping the balance in favour of prescribing. This is a cause for concern since evidence has shown that the more reliant doctors are on commercial sources of information, the less rational they are as prescribers.16
As our sample was small and purposive, the classification is tentative. Nevertheless, it offers hypotheses for further study. Whilst the study depended on GPs’ self-reporting and rationalization of events, which may have been normative, this is somewhat overcome by the validity of the critical incident technique which uses specific factual prescribing contexts focused on recently launched drugs, an interview structure that is probing and interactive and analysis based on GPs own responses. A methodological observation from the study is that we cannot make presumptions about people’s actual behaviour from accounts of their general behaviour. With some of the high prescribers, there were inconsistencies between their actual pattern of new drug prescribing and the perception of their approach as cautious. This has important implications for the design of future prescribing studies, and we believe that research should focus on actual prescribing incidents and attitudes attached to these events, despite the possibility of post hoc rationalization, rather than general accounts of attitudes and behaviour.
|
Acknowledgments |
|---|
We would like to thank all GPs who generously gave up their time to participate in this study. We would also like to thank Clare Dutton and Clare Bottomley, Health Authority pharmaceutical advisors, for providing prescribing data. This study was funded by the Department of Health Prescribing Research Initiative.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Lloyd DC, Harris CM, Roberts DJ. Specific therapeutic group age–sex related prescribing units (STAR-PUs): weightings for analysing general practices’ prescribing in England. Br Med J 1995; 311:991–994.
2 Prosser H, Walley T. Influences on GPs’ decision to prescribe new drugs—the importance of who says what. Fam Pract 2003; 20:61–68.
3 McGettigan P, Golden J, Fryer J, Chan R, Feely J. Prescribers prefer people: the sources of information used by doctors for prescribing suggest that the medium is more important than the message. Br J Clin Pharm 2000; 51:184–189.[CrossRef]
4 Strickland-Hodge B, Jepson M. The role of hospital consultants in general practice prescribing. J R Soc Med 1988; 81:207–209.[Abstract]
5 Feely J, Chan R, McManus J, O’Shea B. The influence of hospital-based prescribers on prescribing in general practice. Pharmacoeconomics 1999; 16:175–181.[CrossRef][ISI][Medline]
6 Peay MY, Peay ER. The role of commercial sources in the adoption of a new drug. Soc Sci Med 1988; 26:1183–1189.[CrossRef][ISI][Medline]
7 Greenwood J. Marketing Medicines: What Works and Why. London: REMIT Consultants; 1991.
8 Inman W, Pearce G. Prescriber profile and post-marketing surveillance. Lancet 1993; 342:658–661.[CrossRef][ISI][Medline]
9 Strickland-Hodge B, Jepson MH. Identification and characterisation of early and late prescribers in general practice. J R Soc Med 1982; 75:341–345.[Medline]
10 Steffensen FH, Sorensen HT, Olesen F. Diffusion of new drugs in Danish general practice. Fam Pract 1999; 16:407–413.
11 Jones MI, Greenfield M, Bradley C. Prescribing new drugs: qualitative study of influences on consultants and general practitioners. Br Med J 2001; 323:378–396.
12 Williamson P. How general practitioners assess risks in using new drugs. J R Coll Gen Pract 1975; 25:383–386.[Medline]
13 Flanagan JC. The critical incident technique. Psychol Bull 1954; 51:327–358[CrossRef][ISI][Medline]
14 Glaser B, Strauss A. The Discovery of Grounded Theory. Chicago: Aldine; 1967.
15 Huberman A, Miles M. Data management and analysis methods. In Denzin N, Lincoln Y (eds), Handbook of Qualitative Methods. Newbury Park: Sage; 1994: 428–444.
16 Lexchin, J. Doctors and detailers: therapeutic education or pharmaceutical promotion? Int J Health Sci 1989; 19:663–679.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. R Florentinus, R. van Hulten, M. E. Kloth, E. R Heerdink, A. F. Griens, H. G. Leufkens, P. P Groenewegen, and for the Netherlands Institute for Health Services The Effect of Pharmacotherapy Audit Meetings on Early New Drug Prescribing by General Practitioners Ann. Pharmacother., February 1, 2007; 41(2): 319 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. van Duijn, M. M. Kuyvenhoven, F. G. Schellevis, and T. J. M. Verheij Determinants of prescribing of second-choice antibiotics for upper and lower respiratory tract episodes in Dutch general practice J. Antimicrob. Chemother., August 1, 2005; 56(2): 420 - 422. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||