effect on the symptoms, which are best relieved by treating the underlying disorder. J P H DRENTH
Research Assistant, Faculty of Medicine, University of Limburg, Maastricht, The Netherlands
4 Alacron-Segovia D, Bag RR, Fairbairn JF, Hagedorn AB. Erythermalgia; a clue to early diagnosis of myeloproliferative disorders. Arch Intern Med 1966;117:511-5. 5 Michiels JJ, Abels J, Steketee J, Van Vliet HHDM, Vuzevski VD. Erythromelalgia caused by platelet-mediated arteriolar inflammation and thrombosis. Ann Intern Med 1985;102:466-71. 6 Michiels JJ, Lindemans J, Van Vliet HHDM, Abels J. Survival kinetics of platelets and fibrinogen in thrombocythemia related to erythromelalgia. BrJ Haematol 1982;50:691. 7 Michiels JJ, Ten Kate FWJ, Vuzevski VD, Abels J. Histopathology of erythromelalgia in
thrombocythaemia. Histopathology 1984;8:669-78.
J J MICHIELS Consultant Haematologist, University Hospital, Erasmus University, Rotterdam, The Netherlands I Mitchell SW. On a rare vasomotor neurosis of the extremities and on maladies with which it may be confounded. Amj Med Sci 1878;76:2-36. 2 Smith LA, Allen FV. Erythermalgia (erythromelalgia) of the extremities. A syndrome characterized by redness, heat and pain. Am HeartJ 1938;16:136-41. 3 Babb RR, Alacrond-Segovia D, Fairbairn JF. Erythermalgia a review of 51 cases. Circulatiwn 1964;29: 136-4 1.
8 Michiels JJ, Van Joost Th. Erythromelalgia and thrombocythaemia, a causal relation. J Am Acad Dermatol 1990;22:107-11. 9 Michiels JJ, Van Joost Th. Idiopathic erythermalgia, a congenital disorder. J Am Acad Dermatol 1989;21: 1128-30. 10 Michiels JJ, Van Joost Th. Primary and secondary erythermalgia: a critical review. Neth J Med 1988;33:205-8. 11 Brown GF. Erythromelalgia and other disturbances of the extrenmities accompanied by vasodilatation and burning. AmJMed Sci 1932;183:468-85. 12 Fairburn JF, Juergens JL, Spitell JA. Peripheral vascular diseases. Philadelphia: W B Saunders, 1972. 13 Alacron-Segovia D, Diaz-Jouananen E. Erythermalgia in systemic lupus erythematosus. AmJMed Sci 1973;266:149-551. 14 Ratz Ji, Berfield WF, Steck WD. Erythermalgia with vasculitis. J Am Acad Dermatol 1979;1: 433-50. 1 Thompson GH, Hahn G, Rang M. Erythromelalgia. Clin Orthop 1979;144:249-54. 16 Levesque H, Moore N, Wolf LM, Courtois H. Erythromelalgia induced by nicardipine (inverse Reynaud's phenomenon). BrMedj 1989;298:1252-3. 17 Drenth JPH. Erythromelalgia induced by nicardipine. BrMedJ 1989;298:1582. 18 Levesque H, Moore N. Erythromelalgia induced by nicardipine. BrMedj 1989;298:323.
General practitioners' workload Studies need to take account ofpart timers' characteristics Picture a middle-aged man who has not troubled to keep up with developments in his continuously changing profession since he qualified 20 years ago. He sees clients for two or three brief periods each week, spending the rest of his time on the golf course.
This was how the Economist began an editorial entitled "Let doctors compete" in 1987.1 The public image of the general practitioner on the golf course is tenacious. Even hospital doctors sometimes wonder what it means when they hear that the general practitioner is "on his rounds." General practitioners have fuelled these misconceptions by using terms like "part time" in a pejorative way to describe the work of women practitioners, those concerned with teaching and organisation, and almost any doctor other than themselves. Surveys of workload have added numbers to the confusion.2 What is the reality? General practitioners contract to be responsible for providing care for their patients continuously (168 hours each week).3 About 6% of doctors do work and remain on call for 130 to 168 hours a week, but most general practitioners work less than the theoretical maximum.4 They have reduced their working hours by forming groups and negotiating about their share of practice income and the hours they are prepared to contribute in return. Each group of doctors determines the hours of work rewarded by a maximum or full share of profits and the part share of profits corresponding to shorter or more convenient hours of work. This process has allowed women with young families and people past normal retirement age to continue practising. Such doctors may work fewer hours than their full profit sharing partners-and earn less-though this may be acceptable to doctors in two income families or to those receiving a pension. This does not, of course, necessarily leave these doctors with more free time; young mothers who are also general practitioners are likely to work long hours.5 The public relations problem with studies of workload is that they have tended to focus only on hours rather than the income work package. This may have been partly because the Department of Health study provided data used for the BMJ VOLUME 301
8 SEPTEMBER 1990
doctors' pay review and investigators wanted to avoid circular thinking. Or it may be that researchers feared that questions about income would reduce doctors' willingness to respond or the reliability of the data. But recent surveys by Leese and Bosanquet6 and Hooper7'8 suggest that doctors are willing to provide information about income and that income varies a great deal. Hooper found that about half of the women general practitioners who responded received a maximum or full share of practice profits-but that half received less than a maximum partners' income and worked fewer hours.78 There was some dissatisfaction with earnings among part sharing general practitioners, which may reflect lack of consensus within groups about the share of profits that should be attached to longer hours and night and weekend work. The Department of Health's study of workload found that the average general practitioner divided up the working week into 20 hours with patients in the surgery; 10 hours visiting patients in their homes; eight hours organising- and doing administration; five hours doing work (not counted as general medical service) such as continuing medical education, committee work, or teaching undergraduates; and 30 hours on call. The average of 38 hours spent on the first three activities actually represented a conflation of the hours worked by doctors with maximum shares and part sharers who worked fewer hours. General practitioners who worked fewer hours (and were likely to be part sharers) were more commonly women or over 60 years old. Yet it surely is illogical to add the hours worked by full timers with those of part timers, divide the sum by the number of workers, and present the product as an average workload. Nevertheless, when this is done the average partner worked a 73 hour week-twice the normal working week in Britain and an underestimate of the time worked by full timers. Those who do not take responsibility for sick people at night and at weekends tend to focus discussion on the figure of 20 hours spent in the surgery or the 38 hours spent on the first three activities. This total is no more than a normal working week, but focusing on this discounts and devalues the extra 35 hours spent on activities such as teaching, learning, commit-
tee work, and on call. If general practitioners ignore issues such as education, organisation, and access they will be accused of conforming to the golfer stereotype evoked in the Economist editorial. Variations in doctors' need for income and their willingness to do different aspects of the work help determine the different working patterns of general practitioners, while other features of general practice - for example, list size - are linked with the characteristics of the practitioner. Butler and Calnan found that 44% of general practitioners with a list size of less than 1500 patients were women,9 most of whom are likely, inferring from Hooper's work, to have been part timers and part sharers.'8 They also found that general practitioners with a list size of less than 1500 were more likely to be over 65 or practising in rural areas. Some older doctors are receiving a pension after a 24 hour retirement and accept less income and work. Rural doctors with a small list may balance seeing fewer patients in surgery with more time spent travelling between visits and on call. Though a small list is generally associated with low income, rural doctors may augment their income by dispensing.6 Butler and Calnan also found an association between list size and time spent with patients.'01' Doctors with smaller lists on average provided longer consultations and saw their patients more frequently. From their data I have estimated that those with fewer than 1500 patients spent on average 43 minutes with each patient each year; those with over 3000 patients spent only 20 minutes.'2 Butler and Calnan analysed their data as though list size might be a variable that determined the way in which general practitioners spena their working day. They acknowledged that implicit in their analysis was the assumption that doctors who acquired smaller lists in the course of a general reduction in list size would behave in the same way as doctors who currently have a list of that size. They reported, however, that their evidence did not suggest that reducing list sizes will produce the quantity of benefits put forward by the proponents of smaller lists. " It must be difficult to extrapolate longitudinally from the results of a cross sectional study when most small list holders were women or elderly general practitioners. Their full time colleagues might well not behave in the same way if they acquired smaller lists in the course of a gradual national reduction in list size. The income, responsibilities, and social characteristics of the groups are quite different. Some general practitioners may be more altruistic, but other factors probably create a "demand pull" for the supply of consultation time. Preston-Whyte and colleagues looked at the effect of a principal's gender on consultation patterns.'3 Women patients preferred to consult women general practitioners, especially for smear tests, contraception other than the contraceptive pill, and gynaecological problems. Buchan and Richardson found that the most time consuming physical examination was the vaginal examination.'4 This took over three minutes on average, which is three times longer than is usually spent examining the patient. There is also evidence that patients of either sex with psychological problems are more likely to consult women doctors.5 16 And such consultations take longer.'7 Women and people with psychological problems are known to consult more frequently.'8 These are all patient demand influences which might lead women doctors to provide longer and more frequent consultations. Less research has been done on the work of older doctors. It is conventional wisdom that general practitioners age with their patients, and Department of Health figures confirm that older doctors do on average have older patients (Department of Health, personal communication). It is likely that these
456
patients have more problems and consult more frequently, creating higher demand. Cochrane advocated the use of randomised controlled trials to test the effect of interventions in medicine. 9 An experiment to measure the effect of list size on general practitioners' work behaviour might require randomly allocating general practitioners to different list sizes while being careful to match doctors for demographic characteristics such as age and sex. But if only observational evidence is available, men doctors aged 40-60 could be compared in respect of list size and practice activity so that like was compared with like. Future studies might also include questions about profit shares and whether the doctor was defined as working "full" or "part" time and the criteria used by each individual general practitioner group. Butler and Calnan concluded: "To put the matter bluntly, the smaller the personal list size of doctors, the less time they appear to spend on all aspects of their work, and hence the more time they appear to spend as free time. "9 But if this free time can be accounted for in part by women with small children who accept lower profit shares this is not shameful professionally. Piecing together the evidence, it seems that women and older doctors may take lower profit shares, work fewer hours in total, and spend more time with each patient. This picture has been constructed like working on a jigsaw puzzle. Pattern recognition comes from experience, and the "facts," like pieces of the jigsaw, are drawn from research. It is difficult for observers to collect and analyse data that represent the outcome of decisions about income and the supply of services made by heterogeneous individual doctors and groups. And it is even more difficult for policy makers to use this information for national decision making that will in turn affect the workload and income of practices nationwide. In future general practitioners need to design studies that will make sense of the social differences among doctors and the effect this has on income, work distribution, and consumer demand. Negotiations about income and pay need to be analysed within practices, and terms like "part time" and "free time" clearly defined. As national policy hinges on this research evidence and inference must then be used cautiously. LEONE RIDSDALE Senior Lecturer in General Practice, United Medical and Dental School Guy's and St Thomas's Hospitals, London SEl 9RT 1 Anonymous. Let doctors compete: Britain'I general practitioners need a new prescription. Economist 1987 Nov: 19-20. 2 Ridsdale L. General practitioner workload: research and policy. _J R Coll Gen Pract 1988;38:390-1. 3 Department of Health. Terms of service for doctors in general practice. London: Department of Health, 1989. 4 Department of Health and Social Security. General medical practitioners workload. A report prepared for the doctors' and dentists' review body 1985186. London: DHSS, 1987. 5 Oakley A. The sociology ofhousework. Oxford: Blackwell, 1984. 6 Leese B, Bosanquet N. High and low incomes in general practice. BrMedJ 1989;298:932-4. 7 Hooper J. Full-time women general practitioners-an invaluable asset. J R Coll Gen Pract 1989;39:289-91. 8 Hooper J. Millman J, Schofield P, Ward G. Part-time women general practitioners-workload and remuneration. J R Coll Gen Pract 1989;39:400-3. 9 Butler JR, Calnan MW. Too many patients? A study of the economy of time and standards of care in general practice. Aldershot: Gower Publishing, 1987. 10 Calnan M, Butler JR. The economy of time in general practice: an assessment of the influence of list size. Soc Sci Med 1988;26:435-41. 11 Butler JR, Calnan MW. List size and use of time in general practice. BrMedJ 1987;295:1383-6. 12 Ridsdale L. Economics of time and truth. Practitioner 1987;233:1121. 13 Preston-Whyte ME, Fraser RC, Beckett JL. Effect of a principal's gender on consultation patterns. J R Coll Gen Pract 1983;33:657-8. 14 Buchan IC, Richardson IM. Time study of consultations in general practice. Edinburgh: Scottish Home and Health Department, 1973. (Scottish Health Service Studies, No 27.) 15 Marks HN, Goldberg DP, Hillier VF. Determinants of the ability of general practitioners to detect psychiatric illness. Psychol Med 1979;9:337-53. 16 Boardman AP. The General Health Questionnaire and the detection of emotional disorders by
general practitioners. BrJ Psychiatry 1985;147:113-9.
17 Raynes NV, Cairns V. Factors contributing to the length of general practice consultations. J R Coll Gen Pract 1980;30:496-8. 18 Royal College of General Practitioners, Office of Population Censuses and Surveys, Department of Health and Social Security 1981-82. Morbidity statistics from general practice-third national study. London: HMSO, 1986. 19 Cochrane A. Effectiveness and efficiency. London: Nuffield Provincial Hospitals Trust, 1972.
BMJ VOLUME 301
8 SEPTEMBER 1990
Research Assistant, Faculty of Medicine, University of Limburg, Maastricht, The Netherlands
4 Alacron-Segovia D, Bag RR, Fairbairn JF, Hagedorn AB. Erythermalgia; a clue to early diagnosis of myeloproliferative disorders. Arch Intern Med 1966;117:511-5. 5 Michiels JJ, Abels J, Steketee J, Van Vliet HHDM, Vuzevski VD. Erythromelalgia caused by platelet-mediated arteriolar inflammation and thrombosis. Ann Intern Med 1985;102:466-71. 6 Michiels JJ, Lindemans J, Van Vliet HHDM, Abels J. Survival kinetics of platelets and fibrinogen in thrombocythemia related to erythromelalgia. BrJ Haematol 1982;50:691. 7 Michiels JJ, Ten Kate FWJ, Vuzevski VD, Abels J. Histopathology of erythromelalgia in
thrombocythaemia. Histopathology 1984;8:669-78.
J J MICHIELS Consultant Haematologist, University Hospital, Erasmus University, Rotterdam, The Netherlands I Mitchell SW. On a rare vasomotor neurosis of the extremities and on maladies with which it may be confounded. Amj Med Sci 1878;76:2-36. 2 Smith LA, Allen FV. Erythermalgia (erythromelalgia) of the extremities. A syndrome characterized by redness, heat and pain. Am HeartJ 1938;16:136-41. 3 Babb RR, Alacrond-Segovia D, Fairbairn JF. Erythermalgia a review of 51 cases. Circulatiwn 1964;29: 136-4 1.
8 Michiels JJ, Van Joost Th. Erythromelalgia and thrombocythaemia, a causal relation. J Am Acad Dermatol 1990;22:107-11. 9 Michiels JJ, Van Joost Th. Idiopathic erythermalgia, a congenital disorder. J Am Acad Dermatol 1989;21: 1128-30. 10 Michiels JJ, Van Joost Th. Primary and secondary erythermalgia: a critical review. Neth J Med 1988;33:205-8. 11 Brown GF. Erythromelalgia and other disturbances of the extrenmities accompanied by vasodilatation and burning. AmJMed Sci 1932;183:468-85. 12 Fairburn JF, Juergens JL, Spitell JA. Peripheral vascular diseases. Philadelphia: W B Saunders, 1972. 13 Alacron-Segovia D, Diaz-Jouananen E. Erythermalgia in systemic lupus erythematosus. AmJMed Sci 1973;266:149-551. 14 Ratz Ji, Berfield WF, Steck WD. Erythermalgia with vasculitis. J Am Acad Dermatol 1979;1: 433-50. 1 Thompson GH, Hahn G, Rang M. Erythromelalgia. Clin Orthop 1979;144:249-54. 16 Levesque H, Moore N, Wolf LM, Courtois H. Erythromelalgia induced by nicardipine (inverse Reynaud's phenomenon). BrMedj 1989;298:1252-3. 17 Drenth JPH. Erythromelalgia induced by nicardipine. BrMedJ 1989;298:1582. 18 Levesque H, Moore N. Erythromelalgia induced by nicardipine. BrMedj 1989;298:323.
General practitioners' workload Studies need to take account ofpart timers' characteristics Picture a middle-aged man who has not troubled to keep up with developments in his continuously changing profession since he qualified 20 years ago. He sees clients for two or three brief periods each week, spending the rest of his time on the golf course.
This was how the Economist began an editorial entitled "Let doctors compete" in 1987.1 The public image of the general practitioner on the golf course is tenacious. Even hospital doctors sometimes wonder what it means when they hear that the general practitioner is "on his rounds." General practitioners have fuelled these misconceptions by using terms like "part time" in a pejorative way to describe the work of women practitioners, those concerned with teaching and organisation, and almost any doctor other than themselves. Surveys of workload have added numbers to the confusion.2 What is the reality? General practitioners contract to be responsible for providing care for their patients continuously (168 hours each week).3 About 6% of doctors do work and remain on call for 130 to 168 hours a week, but most general practitioners work less than the theoretical maximum.4 They have reduced their working hours by forming groups and negotiating about their share of practice income and the hours they are prepared to contribute in return. Each group of doctors determines the hours of work rewarded by a maximum or full share of profits and the part share of profits corresponding to shorter or more convenient hours of work. This process has allowed women with young families and people past normal retirement age to continue practising. Such doctors may work fewer hours than their full profit sharing partners-and earn less-though this may be acceptable to doctors in two income families or to those receiving a pension. This does not, of course, necessarily leave these doctors with more free time; young mothers who are also general practitioners are likely to work long hours.5 The public relations problem with studies of workload is that they have tended to focus only on hours rather than the income work package. This may have been partly because the Department of Health study provided data used for the BMJ VOLUME 301
8 SEPTEMBER 1990
doctors' pay review and investigators wanted to avoid circular thinking. Or it may be that researchers feared that questions about income would reduce doctors' willingness to respond or the reliability of the data. But recent surveys by Leese and Bosanquet6 and Hooper7'8 suggest that doctors are willing to provide information about income and that income varies a great deal. Hooper found that about half of the women general practitioners who responded received a maximum or full share of practice profits-but that half received less than a maximum partners' income and worked fewer hours.78 There was some dissatisfaction with earnings among part sharing general practitioners, which may reflect lack of consensus within groups about the share of profits that should be attached to longer hours and night and weekend work. The Department of Health's study of workload found that the average general practitioner divided up the working week into 20 hours with patients in the surgery; 10 hours visiting patients in their homes; eight hours organising- and doing administration; five hours doing work (not counted as general medical service) such as continuing medical education, committee work, or teaching undergraduates; and 30 hours on call. The average of 38 hours spent on the first three activities actually represented a conflation of the hours worked by doctors with maximum shares and part sharers who worked fewer hours. General practitioners who worked fewer hours (and were likely to be part sharers) were more commonly women or over 60 years old. Yet it surely is illogical to add the hours worked by full timers with those of part timers, divide the sum by the number of workers, and present the product as an average workload. Nevertheless, when this is done the average partner worked a 73 hour week-twice the normal working week in Britain and an underestimate of the time worked by full timers. Those who do not take responsibility for sick people at night and at weekends tend to focus discussion on the figure of 20 hours spent in the surgery or the 38 hours spent on the first three activities. This total is no more than a normal working week, but focusing on this discounts and devalues the extra 35 hours spent on activities such as teaching, learning, commit-
tee work, and on call. If general practitioners ignore issues such as education, organisation, and access they will be accused of conforming to the golfer stereotype evoked in the Economist editorial. Variations in doctors' need for income and their willingness to do different aspects of the work help determine the different working patterns of general practitioners, while other features of general practice - for example, list size - are linked with the characteristics of the practitioner. Butler and Calnan found that 44% of general practitioners with a list size of less than 1500 patients were women,9 most of whom are likely, inferring from Hooper's work, to have been part timers and part sharers.'8 They also found that general practitioners with a list size of less than 1500 were more likely to be over 65 or practising in rural areas. Some older doctors are receiving a pension after a 24 hour retirement and accept less income and work. Rural doctors with a small list may balance seeing fewer patients in surgery with more time spent travelling between visits and on call. Though a small list is generally associated with low income, rural doctors may augment their income by dispensing.6 Butler and Calnan also found an association between list size and time spent with patients.'01' Doctors with smaller lists on average provided longer consultations and saw their patients more frequently. From their data I have estimated that those with fewer than 1500 patients spent on average 43 minutes with each patient each year; those with over 3000 patients spent only 20 minutes.'2 Butler and Calnan analysed their data as though list size might be a variable that determined the way in which general practitioners spena their working day. They acknowledged that implicit in their analysis was the assumption that doctors who acquired smaller lists in the course of a general reduction in list size would behave in the same way as doctors who currently have a list of that size. They reported, however, that their evidence did not suggest that reducing list sizes will produce the quantity of benefits put forward by the proponents of smaller lists. " It must be difficult to extrapolate longitudinally from the results of a cross sectional study when most small list holders were women or elderly general practitioners. Their full time colleagues might well not behave in the same way if they acquired smaller lists in the course of a gradual national reduction in list size. The income, responsibilities, and social characteristics of the groups are quite different. Some general practitioners may be more altruistic, but other factors probably create a "demand pull" for the supply of consultation time. Preston-Whyte and colleagues looked at the effect of a principal's gender on consultation patterns.'3 Women patients preferred to consult women general practitioners, especially for smear tests, contraception other than the contraceptive pill, and gynaecological problems. Buchan and Richardson found that the most time consuming physical examination was the vaginal examination.'4 This took over three minutes on average, which is three times longer than is usually spent examining the patient. There is also evidence that patients of either sex with psychological problems are more likely to consult women doctors.5 16 And such consultations take longer.'7 Women and people with psychological problems are known to consult more frequently.'8 These are all patient demand influences which might lead women doctors to provide longer and more frequent consultations. Less research has been done on the work of older doctors. It is conventional wisdom that general practitioners age with their patients, and Department of Health figures confirm that older doctors do on average have older patients (Department of Health, personal communication). It is likely that these
456
patients have more problems and consult more frequently, creating higher demand. Cochrane advocated the use of randomised controlled trials to test the effect of interventions in medicine. 9 An experiment to measure the effect of list size on general practitioners' work behaviour might require randomly allocating general practitioners to different list sizes while being careful to match doctors for demographic characteristics such as age and sex. But if only observational evidence is available, men doctors aged 40-60 could be compared in respect of list size and practice activity so that like was compared with like. Future studies might also include questions about profit shares and whether the doctor was defined as working "full" or "part" time and the criteria used by each individual general practitioner group. Butler and Calnan concluded: "To put the matter bluntly, the smaller the personal list size of doctors, the less time they appear to spend on all aspects of their work, and hence the more time they appear to spend as free time. "9 But if this free time can be accounted for in part by women with small children who accept lower profit shares this is not shameful professionally. Piecing together the evidence, it seems that women and older doctors may take lower profit shares, work fewer hours in total, and spend more time with each patient. This picture has been constructed like working on a jigsaw puzzle. Pattern recognition comes from experience, and the "facts," like pieces of the jigsaw, are drawn from research. It is difficult for observers to collect and analyse data that represent the outcome of decisions about income and the supply of services made by heterogeneous individual doctors and groups. And it is even more difficult for policy makers to use this information for national decision making that will in turn affect the workload and income of practices nationwide. In future general practitioners need to design studies that will make sense of the social differences among doctors and the effect this has on income, work distribution, and consumer demand. Negotiations about income and pay need to be analysed within practices, and terms like "part time" and "free time" clearly defined. As national policy hinges on this research evidence and inference must then be used cautiously. LEONE RIDSDALE Senior Lecturer in General Practice, United Medical and Dental School Guy's and St Thomas's Hospitals, London SEl 9RT 1 Anonymous. Let doctors compete: Britain'I general practitioners need a new prescription. Economist 1987 Nov: 19-20. 2 Ridsdale L. General practitioner workload: research and policy. _J R Coll Gen Pract 1988;38:390-1. 3 Department of Health. Terms of service for doctors in general practice. London: Department of Health, 1989. 4 Department of Health and Social Security. General medical practitioners workload. A report prepared for the doctors' and dentists' review body 1985186. London: DHSS, 1987. 5 Oakley A. The sociology ofhousework. Oxford: Blackwell, 1984. 6 Leese B, Bosanquet N. High and low incomes in general practice. BrMedJ 1989;298:932-4. 7 Hooper J. Full-time women general practitioners-an invaluable asset. J R Coll Gen Pract 1989;39:289-91. 8 Hooper J. Millman J, Schofield P, Ward G. Part-time women general practitioners-workload and remuneration. J R Coll Gen Pract 1989;39:400-3. 9 Butler JR, Calnan MW. Too many patients? A study of the economy of time and standards of care in general practice. Aldershot: Gower Publishing, 1987. 10 Calnan M, Butler JR. The economy of time in general practice: an assessment of the influence of list size. Soc Sci Med 1988;26:435-41. 11 Butler JR, Calnan MW. List size and use of time in general practice. BrMedJ 1987;295:1383-6. 12 Ridsdale L. Economics of time and truth. Practitioner 1987;233:1121. 13 Preston-Whyte ME, Fraser RC, Beckett JL. Effect of a principal's gender on consultation patterns. J R Coll Gen Pract 1983;33:657-8. 14 Buchan IC, Richardson IM. Time study of consultations in general practice. Edinburgh: Scottish Home and Health Department, 1973. (Scottish Health Service Studies, No 27.) 15 Marks HN, Goldberg DP, Hillier VF. Determinants of the ability of general practitioners to detect psychiatric illness. Psychol Med 1979;9:337-53. 16 Boardman AP. The General Health Questionnaire and the detection of emotional disorders by
general practitioners. BrJ Psychiatry 1985;147:113-9.
17 Raynes NV, Cairns V. Factors contributing to the length of general practice consultations. J R Coll Gen Pract 1980;30:496-8. 18 Royal College of General Practitioners, Office of Population Censuses and Surveys, Department of Health and Social Security 1981-82. Morbidity statistics from general practice-third national study. London: HMSO, 1986. 19 Cochrane A. Effectiveness and efficiency. London: Nuffield Provincial Hospitals Trust, 1972.
BMJ VOLUME 301
8 SEPTEMBER 1990
