OPINION letters

Please note that all letters must be ing has 196 contracts open in the Dental ers their data, and therefore their conclu- typed. Priority will be given to Practice Board (DPB) data set used by sions, are invalidated to a great extent by the those that are less than 500 words Moles et al. The actual number of individual GDP with an NHS number who maybe is long. All authors must sign the dentists in contract with the authority is only 25% reliant upon GDS earnings. letter, which may be shortened or approximately 1431 a difference of 27%. In C. P. D. P. Lister edited for reasons of space or County Durham and Darlington the health Salisbury clarity. All letters received are acknowledged. authority contracts officer reports that approximately 175-180 individual dentists The author of the paper, David Moles, work in the health authority area, again a responds: Endocarditis risks significant difference compared to 249 in We should like to thank Messers Landes and Sir,—My defence organisation once the DPB data used by Moles et al. Lister for their interest in our recently advised that the risks associated with bacte- I would suggest that the number of con- published paper (BDJ 2001; 190: 548-553). rial endocarditis were so grave that a practi- tracts for NHS care in a health authority Each correspondent questions the validity of tioner should refuse to treat a patient who may be very different to the actual number the DPB contract data used in our was willing to assume those risks by not tak- of dentists who work in an area. There are a investigation, either in terms of the actual ing a prophylactic antibiotic. number of reasons for this, contracts for number of practitioners with an active NHS The review by Seymour et al (BDJ 2000; dentists remain open for a considerable contract, or the amount of time the 189: 610-616) suggests, however, that the length of time with the DPB, after a dentist practitioners commit to NHS work. risk of death following hypersensitivity to has left an area, so their contract and that of Dr Landes presents a table of alternative penicillins may be five times greater than any replacement are reported. If there are NHS dentist to population ratios for his that due to BE itself. any chains of practices in an area it is com- region. He states ‘I cannot suggest that the The recommended prophylactic drug is mon for dentists to work throughout the Northern and Yorkshire data set is not amoxicillin and the risk of death due to chain and have an individual contract for without confounding factors’. This is not an hypersensitivity can surely be reduced by each. Thus a four practice chain, with five issue of confounding; rather the question is administration under supervision. dentists will produce 20 contracts in the whether data are biased. As Dr Landes notes, Nonetheless a significant risk has now been DPB data set. there are methodological differences between quantified and the principle of informed Consequently, there must be some doubt the data sets. The DPB data set counts consent requires disclosure of the risks and as to the robustness of the relationships contracts rather than dentists. Dr Landes’s possible outcomes of treatment. between the demographic factors and den- data was produced in 2000 and may not So how should the practitioner now pro- tist to population ratios reported by Moles correspond to the same time frame as our ceed? Are we better advised to treat at–risk et al. The only consistency between the two investigation (DPB contract data for January and informed patients without cover or data sets is that more dentists work in to March 1999). Further, he does not state the should amoxicillin cease to be the drug of authorities with a dental school, (the bor- source of his population denominators. Dr choice? ders of Gateshead and South Tyneside Landes appears to report estimated data in P. Ziar health authorities being situated within a his letter as evidenced by statements such as Penzance few miles of Newcastle Dental School). ‘When the actual number of individual I cannot suggest that the Northern and dentists in contract with the authority is Yorkshire data set is not without confound- approximately 143...’ and ‘...approximately Inequalities in avail- ing factors. But if inequalities in access to 175-180 individual dentists work in the dental care are to be addressed then it is health authority area...’ The number of ability of NHS GDPs important that reliable information is avail- contracts recorded by the DPB is a consistent Sir,—I read with interest the paper by able with which to inform decisions. and objective measurement. This is why we Moles et al (BDJ 190: 548-553). Tackling D. P. Landes chose to use it as a surrogate measure for the inequalities in the availability of dental care Durham availability of NHS dentists. for the public is an important issue, which The absolute measurement of the dentist to the dental profession needs to address. I was 1. Hull and East Riding Community Health NHS population ratio is of little interest. It is not concerned at the dentist to population Trust. Community Dental Service Oral Health the aim of this research to attempt to specify Strategy 2000-2005, p8 Hull, Hull and East ratios presented in table 1 of the paper. Riding Community Health NHS Trust 2001. an arbitrary ‘ideal’ or ‘acceptable’ ratio. In 2000, the Northern and Yorkshire Rather, the objectives were to indicate that Regional Office collated data on the dentist inequalities exist and to attempt to predict to population ratios for its health authori- Sir,—The authors are to be congratulated those inequalities. The methodological issues ties. These are given in table 1 (see opposite for opening up a debate on this subject. raised by Landes and Lister are discussed page) together with the Moles et al data set They are understandably cautious about along with other important caveats in the for the authorities concerned and the per- their figures, using the paper in part as a sta- manuscript (pages 551-552). There is no centage difference between the two data tistical exercise. reason to expect the validity of the DPB’s data sets, the lowest difference being 14% and the This paper does reveal the inadequate to be any better or worse for any particular highest 41%. When the two data sets are nature of DPB data for numbers of dentists. health authority. Thus, any potential used to rank the authorities according to It takes no account of the concept of Whole misclassification of the availability of dentists their dentist to population ratios, there is Time Equivalence (used widely elsewhere in will not be biased between health authorities. also a difference in the positions of the industry in respect of manpower planning). The usual effect of such ‘non-differential health authorities between the two, as in One wonders if their next task would be misclassification’ is to dilute (weaken) any table 2 (see opposite page). to get the DPB to supply better data in statistical associations. It is therefore, The reason for the large variations respect of numbers of dentists in relation to probable that the associations that we between the two data sets is due to the commitment to GDS work, for until then, I detected were in fact underestimates and the methodology used by Moles et al to deter- would surmise that in a climate of decreas- ‘true’ associations are in fact stronger than mine the number of NHS dentists. East Rid- ing reliance upon the GDS by practice own- those presented.

354 BRITISH DENTAL JOURNAL VOLUME 191. NO.7 OCTOBER 13 2001 OPINION letters

Table 1 Comparison of NHS (E) Northern & Yorkshire Regional Office dentist to population ratio's with Moles et al data set and percentage difference.

Health Authority NHS (E) dentist to Moles et al dentist Percentage difference population ratio to population ratio between data sets (%)

Gateshead & South Tyne 2438 2044 16 Leeds 2526 1894 25 Newcastle & North 2555 1891 26 Ty neside North Yorks 2596 2072 20 Bradford 2778 2226 20 Northumberland 2840 2205 22 Calderdale & Kirklees 2848 2165 24 Sunderland 3013 2593 14 Tees 3024 1782 41 Wakefield 3095 2601 16 North Cumbria 3225 2262 30 County Durham 3358 2443 27 East Riding 3990 2932 27

Ta ble 2 Comparison of NHS (E) Northern & Yorkshire Regional Office dentist to population ratio's with Moles et al data set for comparative ranking of health authorities.

Health Authority NHS (E) Ranking Moles et al Ranking

Gateshead & South Tyne 1 4 Leeds 2 3 Newcastle & North 3 2 Ty neside North Yorks 4 5 Bradford 5 8 Northumberland 6 7 Calderdale & Kirklees 7 6 Sunderland 8 11 Tees 9 1 Wakefield 10 12 North Cumbria 11 9 County Durham 12 10 East Riding 13 13

Understanding dental manpower and the ed with an odontogenic infection. The been detected by the additional imaging relationship between supply and demand for authors reveal that the patient developed a techniques which the authors dismissed as services is an extremely complex problem. retropharyngeal abscess which had tracked not indicated and, indeed, this could have Any model, however sophisticated, is at best a fown the cartid sheath with disastrous con- been drained. simplified representation of reality. We have sequence. They indicate that additional The authors have not emphasised the sig- considered one aspect of this complex imaging of the neck may have identified nificance of dysphagia in a patient with an problem and have shown that certain infection of the soft tissues. However, they odontogenic infection on two occasions. demographic factors from the 1991 census report that the clinical picture was such that First of all when they treated the patient and were useful predictors of inequalities in the there was no indication for further imaging. secondly when they wrote the case report. numbers of NHS contracts in 1999. We The patient, however, after initially being The ‘In Brief’ section of the report should consider that the number of contracts is treated with incision and drainage of the have included ‘do not dismiss dysphagia in a currently the most reliable surrogate measure infection and having received intravenous patient with odontogenic infection’. This is for the ‘availability of National Health antibiotics, returned two days after being probably the key point that dental practi- Service general dental practitioners.’ discharged; at that point he had not been tioners need to remember from this case. taking his antibiotics because he found it While I concur with everything else that difficult to swallow. the authors have stated in their case report, Dismissing dysphagia It is highly probable that the reason for this point is the most significant. Sir,—I was interested to read the case report this was that the patient still had a focus of D. Godden by Green et al regarding mortality associat- pus in the pharyngeal space. This may have Cheltenham

BRITISH DENTAL JOURNAL VOLUME 191. NO.7 OCTOBER 13 2001 355 OPINION letters

1. Green A W, Flower A, New N E. Mortality associated with odontogenic infection! BDJ lend themselves to paper-based learning study to the respondents and wisdom, she 2001; 190: 529-530 material, which could be used for integrated says, is such an intangible and immeasur- assessment tasks. In the event, the CAL disks able entity that Dr Nuttall believes that the The authors A. W. Green, E. A. Flower and were utilised very successfully. A total of 533 attempt was doomed to fail from the start. N. E. New respond: GDPs took part in the pilot. The topics cho- One can impart knowledge in lots of We have read your letter with interest and sen for the pilot were endodontics and cross ways, but can never impart wisdom, she now intend to address your comments. You infection control. Endodontics attracted says. I am certain that anyone with extensive mention that dysphagia is a significant 408 participants, of which number 127 experimental teaching in dentistry behind symptom. We agree that dysphagia is an completed all units. Of those who complet- them would agree wholeheartedly with that important symptom but in this case, prior to ed all units there were 5 failures, 10 received sentiment. Essentially, it comes down to the second discharge of the patient, his a passing grade and 112 passed with distinc- this, that while CAL programmes, supple- symptoms had begun to improve tion. Cross infection control attracted 279 mented by structured testing units, can be significantly; therefore it was not felt participants, of which number 136 com- an effective and efficient method of impart- appropriate to conduct any further pleted all units. Of those who completed all ing clinical information they are not neces- investigations. units there were 8 failures, 1 received a pass- sarily effective in imparting clinical wisdom. Drainage had been established and ing grade and 127 passed with distinction. But having spent a great deal of time over appropriate antibiotics were being The material used received very good assess- the years seeking such an effective method I administered. Indeed, the patient who lived ments from participants for their usefulness have to confess to failure and would be locally did not contact the hospital following for refreshing and updating knowledge and obliged to those of your readers who have discharge and was said by relatives to have for the challenging nature of the questions. convincing evidence of such a method. continued to improve at home. Routinely we The material used was also shown to be E. Renson do not request further tests on patients who appropriate to a range of approaches to London clinically appear to be improving. learning.3 I subscribe to the cautionary We understand and agree with Mr comments made by Welbury et al, in the dis- 1. Kay E J, Silkstone B, Worthington H V. Godden’s comments, however, we feel that in cussion section of their paper, because of the Evaluation of computer aided learning in developing clinical decision-making. Br Dent J this situation the management of the case low response rate. Nevertheless, it is true, as 2001; 190: 554-557. was totally appropriate and the outcome they point out, that a number of teams of 2. Welbury R R, Hobson R S, Stephenson J J, unavoidable. teachers from different dental and medical Jepson N J A. Evaluation of a computer-assisted schools have used CAL programmes for learning programme on the oro-facial signs of child physical abuse (non-accidental injury) by some years with great success, at both general dental practitioners. Br Dent J 2001; Evaluation of CAL undergraduate and graduate levels. 190: 668-670. Why then are we confronted by views 3. Renson E. Continuing Professional programmes which appear to be diametrically opposed? Development. Report of a joint project Sir,—An interesting contradiction in con- In a correspondence with Professor Kay I between Primary Dental Care and the BDJ funded by the NCCPED. Alicante: CEC Report clusion with regard to computer assisted believe that I have arrived at an answer. Pro- 2000. learning (CAL) programmes appears to be fessor Kay feels that the validity of educa- 4. Nuttall N. Clinical decision-making — can a contained in two papers published in the tional evaluations rests almost entirely on computer-aided learning package help? Br BDJ exactly one month apart. the extent to which they measure a learning Dent J 2001; 190: 545. In May, Kay et al1 describes the package programme’s objectives. This is where she they used as having ‘no effect on dentists’ thinks we often run into problems, particu- treatment decision–making behaviour’ and larly in medicine and dentistry because for Green dentistry call for ‘direct comparisons of computer many years we have confused the teaching Sir,—There does not seem to be much aided learning and traditional of knowledge and the teaching of skills (cli- scope for sustainability in the practice of education...before the ease of distribution of inical decision making and treatment plan- dentistry. Apart from cycling between home such packages causes potentially ineffective ning being skills and the subject of her and practice I cannot find many ways of educational mehods to overtake traditional evaluation). She is sure that if she had evalu- reducing the damaging impact of my prac- ones.’ In June, Welbury et al2 describe their ated the CAL package1 by measuring tice on the environment. We use a lot of dis- programme as ‘user friendly’ and say that changes in knowledge and attitudes it would posables; cotton-wool rolls, cotton packs ‘CAL has made a valuable contribution to have been shown to be effective. It could be and paper towels are fairly environment- postgraduate dentistry’. ‘CAL’, they say ‘con- argued that in such a case the CAL package friendly; but there are many plastic goods in tinues to develop as a method of self-learn- was effective. However, she was interested in the bin. I would like to reduce the amount of ing that seems to be both acceptable and whether the intervention had any impact on polythene used for labwork. A colleague attainable for the busy general dental practi- what people did (rather than said, or knew, told me recently of biodegradeable packing tioner.’ or thought). Her view is that it is the practi- (made from maize?) for dentures; but I have My own experience of CAL programmes3 cal effect of education which matters. In been unable to find a source. Can anybody would appear to support the views of Wel- Nigel Nuttall’s final comment in his sum- help? bury et al. In the joint project between Pri- mary of the paper he believes the failure of G. Balfry mary Dental Care and the British Dental the CAL programme to improve the relia- Bristol Journal on continuing professional devel- bility and validity of treatment decisions is opment (April–December 1999), a pilot more indicative of the difficulties involved Please send your letters to: project funded by the NCCPED, we made in trying to rationalise treatment decision The Editor use of the CAL disks distributed to dentists making rather than a failure of the tech- British Dental Journal in the National Health Service by the nique of computer aided learning. Professor 64 Wimpole Street Department of Health. Kay has interpreted that comment as a refer- London It was thought that the CAL disks would ence to an attempt to impart wisdom in her W1M 8AL

356 BRITISH DENTAL JOURNAL VOLUME 191 NO. 7 OCTOBER 13 2001