Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

The role of the general practitioner in cancer care and the effect of an extended information routine

The role of the general practitioner in cancer care and the effect of an extended information... ORIGINAL PAPER The role of the general practitioner in cancer care and the effect of an extended information routine 1 1 1 2 1 Birgitta Johansson , Gunilla Berglund , Katarina Hoffman , Bengt Glimelius and Per-Olow Sjo¨de´n Section of Caring Sciences, Department of Public Health and Caring Sciences, Uppsala University and Department of Oncology, Radiology and Clinical Immunology, University Hospital, Uppsala, Sweden. Scand J Prim Health Care 2000;18:143 – 148. ISSN 0281-3432 patient, GPs’ potential or actual possibilities to support the patient, desired and received information from the specialist clinic. Objecti7e – To describe the role of the General Practitioner (GP) in Results and conclusions – GPs are commonly involved in the care of the care of one specified cancer patient per GP, and to explore the cancer patients, particularly in the diagnostics of the disease but also GP’s knowledge about that patient’s disease and treatments, and during the periods of treatment and follow-up. The information from what information she:he wanted versus received from the specialist the specialist clinic to the GP is insufficient in standard care. The clinic. A further aim was to evaluate the effects of an Extended extended information routine increased the GPs’ knowledge of the Information Routine (EIR), including increased information from the disease and treatments, and facilitated their possibilities to determine specialist clinic to the GP. patients’ need for support. Design – Semi-structured interviews with GPs about a patient ran- domised between an extended information routine and standard Key words: family physician, neoplasms, information management. information from the specialist clinics. Settings – Primary Health Care. Birgitta Johansson, Department of Public Health and Caring Sci - Subjects – 20 GPs, 10 who received extended information about the ences, Uppsala Uni7ersity, Uppsala Science Park, SE -751 83 Upp - specified patient and 10 who did not. sala, Sweden. Main outcome measures – The extent of GPs’ contact with the developed for diagnoses such as diabetes, hyperten- The role of general practitioners (GPs) in the care of sion and asthma, and there are some examples of cancer patients during treatment and follow-up peri- shared oncological care (13,15,16). ods has rarely been investigated. The fact that GPs The Family Doctor Act was implemented in Swe- are involved in the diagnostics of a large proportion den in January 1994, and in September 1994 about of cancer diseases has been amply documented (1 – 3), 81% of the population had chosen a family doctor and their role in palliative care has been demon- (17). By mid-1995 this reform had been abolished, strated, especially in studies from the UK (4 – 6). partly due to problems establishing co-operation with Some studies suggest that GPs are willing to become other caregivers in the community (17). The present engaged in the follow-up of cancer patients in remis- study started in August 1994, after the Family Doc- sion, and that hospital follow-up provides no advan- tor Act was implemented, and was completed in tages compared to follow-up in primary care settings October 1996. It has several purposes: To describe (7,8). Yet, there is no consensus on this subject and the extent to which GPs in a Swedish county were several barriers have been identified compromising involved in the care of one specified cancer patient the possibilities of GPs to take a greater responsibil- per GP at diagnosis, during treatments and follow- ity in the follow-up of their cancer patients (9,10). up, and to describe the possibilities available to the Several studies indicate that replies to referrals and GP to support the patient during the course of the discharge letters from specialists to GPs are insuffi- disease. Another purpose was to explore what infor- cient in regard to both the content and timing of mation the GP desired versus received from the spe- information. This has been shown to be true in the case of information from oncologists as well as from cialist clinic, and her:his knowledge about the patient’s disease and treatments. A final aim was to other specialists (11 – 13). On the other hand, medical compare GPs who received extended information specialists play an important role in disseminating information to GPs, and by providing them with from the specialist clinic with those who did not with respect to their contacts with and possibilities to up-to date knowledge on patient management (14). Experience, internationally, is increasing concern- support the patient, desired information from the ing shared care programmes, i.e. GPs and specialists specialist clinic, their satisfaction with such informa- sharing disease management in accordance with dis- tion and their knowledge of the patient’s disease and tinct protocols. Models of shared care have been treatments. Scand J Prim Health Care 2000; 18 144 B. Johansson et al. MATERIAL AND METHODS versus men and catchment area (town, city versus Extended information routine rural) was selected from the 17 GPs who did not The present study is one part of a ‘‘Support-Care-Re- receive extended information. Subsequently, another habilitation’’ project in which a consecutive sample of group of 10 GPs, matching the former with regard to gender, catchment area and diagnosis, was selected newly diagnosed cancer patients was included (18). from the 83 who received extended information Half of the patients were randomly assigned (com- (Table I). Thus, a total of 20 GPs was selected for a puter-generated allocation schedule) to an individual semi-structured, open-ended question interview about support intervention, including increased information one specified patient per GP. from the specialist clinic to the patient’s GP (ex- The GPs were told before the interview which tended information routine, EIR) (18). patient was to be discussed, and they were asked to The EIR meant that the GP received copies of the have the patient’s medical record available during the medical record each time the patient was discharged interview. All GPs agreed to participate and the from hospital after a period of inpatient care or had interviews (lasting between 15 and 45 min) were visited a specialist outpatient clinic. This routine was performed by telephone by the first author and au- maintained for all patients for a period of 2 years diotaped, with the exception of one which took place after inclusion. in the GP’s office, when notes were taken. All inter- Two-hundred-and-sixty (85%) of a total of 305 views covered the following areas: The extent of the randomised patients reported a personal GP. A total GP’s contacts with the patient during the last 3 years, of 100 GPs were reported, 83 of whom received potential or actual possibilities to support the patient, extended information about at least one patient desired information from the specialist clinic about (mode  1, n  48 GPs; range 1 – 5); 17 received only cancer patients in general and received information standard information about his:her patients. about the specified patient. The Research Ethics Committee at the Faculty of Medicine, Uppsala University, approved the ‘‘Sup- Analysis port-Care-Rehabilitation’’ project and the present Content analysis was used to analyse the interviews. study. This method is widely used in communication re- search and is appropriate for analysing open-ended questions (19). The method includes data reduction Sample of GPs and the inter6iew and transformation of data into a form required for Only GPs reported by included patients were consid- analysis. This is done through development of vari- ered. The effects of a possible diffusion of the inter- ables, which makes it possible for independent raters vention between groups were limited by the selection to partition the recording units (the interviews in the of GPs from two distinct groups: Those who received present study) into mutually exclusive classes. The extended information (EIR GPs) and those who did following analyses are concerned with identification not for any patient (controls). Thus, a sample of 10 of patterns that are noteworthy, statistically signifi- GPs matching the total population of GPs in the cant or otherwise may account for the results of the county with regard to the proportions of women analysis (19). All audiotaped interviews were transcribed verba- Table I. Sample of GPs (n). tim and carefully read through several times by the first author to determine variables relevant to the aim Extended No extended information information of the study. The variables in the present analysis were constituted as nominal or ordinal scales, which Gender could take on between two and five values (Tables II, Female 4 4 III, IV and V). Two independent raters (GB and KH) Male 6 6 categorised each interview into one category per vari- GPs’ catchment area able and noted this on a separate rating sheet for Rural 4 4 each interview. The raters were not informed in ad- Town, city 6 6 vance whether or not each GP received extended Cancer diagnoses information about his:her patient, but they were Breast 6 5 aware of the fact that half of the GPs had received Colorectal 2 2 such information. In cases of disagreement between Prostate 2 3 coders concerning the ‘‘correct category’’, this was Time from diagnosis to inter6iew determined through discussions between the two Range (months) 14–24 13–26 coders and the first author after determination of Mean (months) 20 19 inter-rater reliability. Scand J Prim Health Care 2000; 18 Role of the GP in cancer care 145 Table II. GP contacts with the patient and involvement in the care of the cancer. All GPs (n) EIR (n) Control (n) IRR Contacts during the last 3 years No contacts 31 2 Only before cancer diagnosis 43 1 k  0.90 Only after cancer diagnosis 01 0 Before and after cancer diagnosis 12 5 7 Diagnostics of the cancer Involved in the diagnostics 13 7 6 k  0.79 Not involved in the diagnostics 73 4 Treatment and follow -up period Not involved at all 6 3 3 Blood tests only, no contacts 11 0 Contacts for other reasons only 21 1 k  0.62 Contacts for other reasons, cancer-related 64 2 problems also addressed Contacts occasioned by the cancer 51 4 Follow -up of the cancer Responsible for follow-up 1 1 0 k  0.64 Not responsible for follow-up 19 9 10 Table III. GP possibilities to follow the patient through the course of the disease and to determine needs for support. All GPs (n) EIR (n) Control (n) IRR Stat. test GPs’ possibilities to follow the patient through the course of the disease GP had satisfying possibilities 13 10 3 F ex pr. GP did not have satisfying possibilities 70 7 k  0.57 pB0.01 GPs’ possibilities to determine patient needs for and :or to support the patient GP was not informed about the disease or the 4 0 4 treatments GP stated that the patient would have benefited 2 0 2 k  1.00 from improved support compared to the support given GP’s opinion is that the patient did not need 4 4 0 his:her support GP offered the patient contact but the patient 1 1 0 did not respond GP stated that she:he supported the patient by 9 5 4 addressing the disease and:or related problems RESULTS The inter-rater reliability (IRR) was determined by GP’s contacts with the patient Kappa (k) or percent agreement (pa) if k  undefined Table II presents the data concerning GPs’ contacts (20) (Tables II, III, IV and V). The IRR was poor with the specified patients and their involvement in (k  0.05 – 0.20) for the variables concerned with the the care of the cancer. A majority of the 20 GPs were desired information from the specialist clinic. After involved in the diagnostics of the cancer as well as in overlapping categories were joined and redefined the treatment and follow-up periods. The results sug- through discussions between the raters and the first gest no differences between the comparison groups author, the IRR reached fair to almost perfect agree- with respect to these variables. ment (k  0.33 – 0.88, pa  95%) (20). When appropriate, Fisher’s exact probability test (F ex pr.) and the Mann-Whitney U test (MWU) (21) Possibilities for patient follow -up and determination were used to compare between GPs who had experi- of support needs enced the EIR and those who had not (Tables II, III, The results of the analyses of the GP’s possibilities to IV and V). A p-value of B 0.05 was considered to be follow the patient, and to determine patient needs for statistically significant. support, are presented in Table III. All GPs in the Scand J Prim Health Care 2000; 18 146 B. Johansson et al. EIR group declared that they were satisfied with the knowledge about the patients’ disease and treatments possibilities for follow-up, compared with only three as well as their satisfaction with the information from in the control group (F ex pr. pB 0.01). The results the specialist clinics. also suggest that the EIR increased the GP’s possibil- ities to determine patients’ needs of support com- pared to GPs in the control group. DISCUSSION The majority of the GPs were involved in the diag- Desired information from the specialist clinics about nostics of the cancer. It was also common that GPs cancer patients in general stayed in contact with the patient during the period The answers to the question about what information of treatment and follow-up and that cancer-related GPs want from specialist clinics about cancer patients problems were addressed. in general were defined through three hierarchical The information from the specialist clinic seems to variables. Thus, the GPs’ responses were judged as to be insufficient in standard care. Four control GPs whether or not they gave examples of contents of were not informed at all, even though two of them information, important situations for information referred their patient to the specialist clinic owing to and the significance of such information. If they did, a suspected cancer. The EIR led to all GPs being the next step concerned whether or not she:he gave informed about the disease and treatments and in examples of the underlying variables (Table IV). general satisfied with this information. Their possibil- The results suggest that information about the ities to follow the patient increased, and appeared to patient’s cancer and other diseases and information facilitate determination of the patient’s need for sup- about future planning and about the GPs’ role in the port. The extent of contacts with the patients was not care of the patient were considered as the most affected by the EIR. One explanation for this may be important. The results suggest no differences between that GPs who are not properly informed about medi- GPs who did experience the EIR and those who did cal examinations and cancer treatments are more not. likely to stay in contact with the patient owing to an uncertainty about whether or not the patient is well Recei6ed information and knowledge about the cared for. The present result, showing that four EIR specified cancer patient GPs expressed the opinion that the patient did not Table V presents the data concerning the GPs’ opin- need his:her support, endorses such an explanation. ions about received information from the specialist Another possible explanation may pertain to the clinic and their knowledge about the disease and reorganisation of the GPs’ work due to the imple- treatments. mentation and rapid abolition of the Family Doctor The results suggest that EIR GPs received informa- Act during the study period. This may have meant tion from the specialist clinic to a greater extent than that the GPs possibilities for extra contacts were did the control group (F ex pr. pB 0.05). Further- limited. more, the results indicate differences between EIR The analysis of the desired information and the GPs and the control group with regard to their agreement between wanted and received information Table IV. Desired information from the specialist clinic about cancer patients in general. All GPs (n) EIR (n) Control (n) IRR The GP ga6e examples of contents of information 20 10 10 100% Information about the cancer, other diseases and impairments 19 10 9 k  0.64 Information about the future planning and the GP’s role in the care of 20 10 10 95% the patient Information about a specified specialist to contact if needed 2 1 1 k  0.77 The GP ga6e examples of important situations when information was desired 62 4 k  0.88 When the GP had referred a patient to a specialist clinic 3 0 3 k  0.33 When a patient is discharged after a period of inpatient care 2 2 0 k  0.66 When a patient consults a specialist clinic without being referred by the 1 01 k  0.57 GP The GP stressed the significance of information from the specialist clinic The information is important for correct judgement of the patient’s 17 9 8 85% symptoms and difficulties, and to determine the role of the GP in the care of the patient Scand J Prim Health Care 2000; 18 Role of the GP in cancer care 147 Table V. Received information from the specialist clinic and the GP’s knowledge about the specified cancer patient. All GPs (n) EIR (n) Control (n) IRR Stat. test Number of GPs who recei6ed information from the Fexpr. specialist clinic The GP received information 16 10 6 k  0.86 pB0.05 The GP did not receive information 4 0 4 GPs’ opinions of the contents of the information Bad, most medical information missing 1 0 1 Not especially good, some medical information 30 3 MWU  13.0 missing Medical information correct, misses information 44 0 k  0.54 pB0.06 about GP’s role and:or about other patient problems, such as psycho-social status Good, does not miss any information 75 2 Very good 1 0 GPs’ opinions of timing of information Too early, would like continuous information 1 0 1 Too late, would like continuous information 1 0 1 k  1.00 Good, did receive continuous information 6 6 0 Received information too often 33 0 Agreement between desired and recei6ed information MWU  9.0 from the specialist clinic No or very bad agreement 6 0 6 Moderate agreement 4 1 3 k  0.86 pB0.01 Quite good agreement 44 0 Good agreement 6 5 1 GPs’ knowledge about the disease and treatments GP has no knowledge about the disease 2 0 2 GP did refer the patient due to a suspected cancer 2 0 2 MWU  10.0 but has no further information GP is informed about the diagnosis and has limited4p 0 4 k  0.68 B0.01 information about treatments GP is informed about the diagnosis and is well 12 10 2 informed about treatments Five GPs did not express an opinion on timing of information. also indicate that EIR GPs’ needs were not fully findings. There is also a certain risk for Type II met. Three GPs reported that they had received errors, since more powerful parametric statistics information too often, and four reported a lack of could not be used. On the other hand, all GPs information about the GP’s role in the care of the approached agreed to participate, and the present patient’s cancer and about other patient problems. analyses could identify differences between the two Thus, the EIR led to an improvement compared to comparison groups, a fact that supports the asser- standard care but it did not result in an optimal tion that the EIR actually made a difference. Still, information routine to satisfy all GPs’ needs for the presented results must be considered only to information. give a preliminary description of this topic. Further Since the main threat to the internal validity of research is warranted to explore the role of Swedish the conclusions was considered to be a possible dif- GPs in the care of cancer patients, both from their fusion of the intervention between groups, the con- point of view and from that of their patients. trol group was selected from those 17 GPs who did In conclusion, Swedish GPs are commonly in- not receive extended information for any patient. volved in the care of cancer patients, not only in The fact that 48% of the GPs cared for only one the diagnostics of the disease, but also during the project patient suggests that it cannot be considered periods of treatments and follow-up. The informa- exceptional that these 17 GPs did not care for any tion from the specialist clinic to the GP is insuffi- patient. It is therefore unlikely that the selection of cient in standard care. The EIR intervention the control GPs has introduced a bias. increased the GPs’ knowledge about the disease and However, the small sample of GPs in the com- the treatments and appeared to facilitate their possi- parison groups threatens the external validity of the bilities to determine the patients’ need for support. Scand J Prim Health Care 2000; 18 148 B. Johansson et al. ACKNOWLEDGEMENTS 10. Wood ML, McWilliam CL. Cancer in remission. Challenge in collaboration for family physicians and oncologists. Can This study was financed by the Swedish Cancer Soci- Fam Phys 1996;42:899 – 910. ety and by the Swedish Foundation for Health Care 11. Jacobs LGH. Referral letters and replies from orthopedic Sciences and Allergy Research. departments; opportunities missed. BMJ 1990;301:470 – 3. 12. Tattersall MH, Griffin A, Dunn SM, Monaghan H, Scat- shard K, Butow PN. Writing to referring doctors after a new patient consultaion. What is wanted and what was contained in letters from one medical oncologist? Aust NZ REFERENCES J Med 1995;25:479 – 82. 1. Beaulieu MD, Be ´land F, Roy D, Falardeau M, He ´bert G. 13. Hampson JP, Roberts RI, Morgan DA. Shared care: a Factors determining compliance with screening mammog- review of the literature. Fam Pract 1996;13:264 – 79. raphy. Can Med Assoc J 1996;154:1335 – 43. 14. Cullen R. The medical specialist: information gateway or 2. Kiernan GN, Frame PS. Cancer occurrence and screening gatekeeper for the family practitioner. Bull Med Libr As- in family practice. A 20-year experience. J Fam Pract soc 1997;85:348 – 55. 1996;43:49 – 55. 15. Orton P. Shared care. Lancet 1994;344:1413 – 5. 3. Olsson P. Distriktsla ¨ karens roll i uppta ¨ ckt och diagnostik 16. Dahler-Eriksen K, Nielsen JD, Lassen JF, Olesen F. av cancer [The role of the district phyicians in the detection Tvaersektorielle behandlingsprogrammer – et eksempel pa ˚ and diagnosis of cancer]. La ¨ kartidningen 1991;88:2685 – 9. det samarbejdene sundhedsvaesen [Shared care pro- 4. Fakhoury W, McCarthy M, Addington Hall J. Determi- grammes – an example of a cooperating health care sys- nants of informal caregivers’ satisfaction with services for tem]. English summary. Ugeskr Laeger 1998;160:5021 – 4. dying cancer patients. Soc Sci Med 1996;42:721 – 31. 17. National Board of Health and Welfare. Sjukva ˚ rden i 5. Hinton J. Services given and help perceived during home Sverige 1995 [Medical services in Sweden 1995]. Stock- care for terminal cancer. Palliat Med 1996;10:125 – 34. holm: National Board of Health and Welfare; 1995. 6. Grande GE, Todd CJ, Barclay SIG, Doyle JH. What 18. Johansson B, Berglund G, Glimelius B, Holmberg L, Sjo ¨- terminal ill patients value in the support provided by GPs, de ´n P-O. lIntensified primary cancer care: a randomised district nurses and Macmillan nurses. Int J Palliat Nurs study on home care nurse contacts. J Adv Nurs 1996;2:138 – 43. 1999;30:1137 – 346. 7. Grunfeld E, Mant D, Yudkin P, Adewuyi Dalton R, Cole 19. Krippendorff K. Content analysis. An introduction to its D, Stewart J, et al. Routine follow up of breast cancer in methodology. Newbury Park, CA: Sage; 1980. primary care: randomised trial. BMJ 1996;313:665 – 9. 20. Flately Brennan P, Hays BJ. The Kappa statistic for estab- 8. Worster A, Bass MJ, Wood ML. Willingness to follow lishing interrater reliability in the secondary analysis of breast cancer. Can Fam Phys 1996;42:263 – 8. qualitative clinical data. Res Nurs Health 1992;15:153 – 8. 9. Grunfeld E. Specialist and general practice views on rou- 21. Siegel S, Catstellan NJJ. Nonparametric statistics for the tine follow-up of breast cancer patients in general practice. behavioral sciences. 2nd ed. New York: McGraw-Hill; Fam Pract 1995;12:60 – 5. 1988. Scand J Prim Health Care 2000; 18 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scandinavian Journal of Primary Health Care Taylor & Francis

The role of the general practitioner in cancer care and the effect of an extended information routine

Loading next page...
 
/lp/taylor-francis/the-role-of-the-general-practitioner-in-cancer-care-and-the-effect-of-f1VNquU00e

References (28)

Publisher
Taylor & Francis
Copyright
© 2000 Informa UK Ltd All rights reserved: reproduction in whole or part not permitted
ISSN
0281-3432
DOI
10.1080/028134300453331
Publisher site
See Article on Publisher Site

Abstract

ORIGINAL PAPER The role of the general practitioner in cancer care and the effect of an extended information routine 1 1 1 2 1 Birgitta Johansson , Gunilla Berglund , Katarina Hoffman , Bengt Glimelius and Per-Olow Sjo¨de´n Section of Caring Sciences, Department of Public Health and Caring Sciences, Uppsala University and Department of Oncology, Radiology and Clinical Immunology, University Hospital, Uppsala, Sweden. Scand J Prim Health Care 2000;18:143 – 148. ISSN 0281-3432 patient, GPs’ potential or actual possibilities to support the patient, desired and received information from the specialist clinic. Objecti7e – To describe the role of the General Practitioner (GP) in Results and conclusions – GPs are commonly involved in the care of the care of one specified cancer patient per GP, and to explore the cancer patients, particularly in the diagnostics of the disease but also GP’s knowledge about that patient’s disease and treatments, and during the periods of treatment and follow-up. The information from what information she:he wanted versus received from the specialist the specialist clinic to the GP is insufficient in standard care. The clinic. A further aim was to evaluate the effects of an Extended extended information routine increased the GPs’ knowledge of the Information Routine (EIR), including increased information from the disease and treatments, and facilitated their possibilities to determine specialist clinic to the GP. patients’ need for support. Design – Semi-structured interviews with GPs about a patient ran- domised between an extended information routine and standard Key words: family physician, neoplasms, information management. information from the specialist clinics. Settings – Primary Health Care. Birgitta Johansson, Department of Public Health and Caring Sci - Subjects – 20 GPs, 10 who received extended information about the ences, Uppsala Uni7ersity, Uppsala Science Park, SE -751 83 Upp - specified patient and 10 who did not. sala, Sweden. Main outcome measures – The extent of GPs’ contact with the developed for diagnoses such as diabetes, hyperten- The role of general practitioners (GPs) in the care of sion and asthma, and there are some examples of cancer patients during treatment and follow-up peri- shared oncological care (13,15,16). ods has rarely been investigated. The fact that GPs The Family Doctor Act was implemented in Swe- are involved in the diagnostics of a large proportion den in January 1994, and in September 1994 about of cancer diseases has been amply documented (1 – 3), 81% of the population had chosen a family doctor and their role in palliative care has been demon- (17). By mid-1995 this reform had been abolished, strated, especially in studies from the UK (4 – 6). partly due to problems establishing co-operation with Some studies suggest that GPs are willing to become other caregivers in the community (17). The present engaged in the follow-up of cancer patients in remis- study started in August 1994, after the Family Doc- sion, and that hospital follow-up provides no advan- tor Act was implemented, and was completed in tages compared to follow-up in primary care settings October 1996. It has several purposes: To describe (7,8). Yet, there is no consensus on this subject and the extent to which GPs in a Swedish county were several barriers have been identified compromising involved in the care of one specified cancer patient the possibilities of GPs to take a greater responsibil- per GP at diagnosis, during treatments and follow- ity in the follow-up of their cancer patients (9,10). up, and to describe the possibilities available to the Several studies indicate that replies to referrals and GP to support the patient during the course of the discharge letters from specialists to GPs are insuffi- disease. Another purpose was to explore what infor- cient in regard to both the content and timing of mation the GP desired versus received from the spe- information. This has been shown to be true in the case of information from oncologists as well as from cialist clinic, and her:his knowledge about the patient’s disease and treatments. A final aim was to other specialists (11 – 13). On the other hand, medical compare GPs who received extended information specialists play an important role in disseminating information to GPs, and by providing them with from the specialist clinic with those who did not with respect to their contacts with and possibilities to up-to date knowledge on patient management (14). Experience, internationally, is increasing concern- support the patient, desired information from the ing shared care programmes, i.e. GPs and specialists specialist clinic, their satisfaction with such informa- sharing disease management in accordance with dis- tion and their knowledge of the patient’s disease and tinct protocols. Models of shared care have been treatments. Scand J Prim Health Care 2000; 18 144 B. Johansson et al. MATERIAL AND METHODS versus men and catchment area (town, city versus Extended information routine rural) was selected from the 17 GPs who did not The present study is one part of a ‘‘Support-Care-Re- receive extended information. Subsequently, another habilitation’’ project in which a consecutive sample of group of 10 GPs, matching the former with regard to gender, catchment area and diagnosis, was selected newly diagnosed cancer patients was included (18). from the 83 who received extended information Half of the patients were randomly assigned (com- (Table I). Thus, a total of 20 GPs was selected for a puter-generated allocation schedule) to an individual semi-structured, open-ended question interview about support intervention, including increased information one specified patient per GP. from the specialist clinic to the patient’s GP (ex- The GPs were told before the interview which tended information routine, EIR) (18). patient was to be discussed, and they were asked to The EIR meant that the GP received copies of the have the patient’s medical record available during the medical record each time the patient was discharged interview. All GPs agreed to participate and the from hospital after a period of inpatient care or had interviews (lasting between 15 and 45 min) were visited a specialist outpatient clinic. This routine was performed by telephone by the first author and au- maintained for all patients for a period of 2 years diotaped, with the exception of one which took place after inclusion. in the GP’s office, when notes were taken. All inter- Two-hundred-and-sixty (85%) of a total of 305 views covered the following areas: The extent of the randomised patients reported a personal GP. A total GP’s contacts with the patient during the last 3 years, of 100 GPs were reported, 83 of whom received potential or actual possibilities to support the patient, extended information about at least one patient desired information from the specialist clinic about (mode  1, n  48 GPs; range 1 – 5); 17 received only cancer patients in general and received information standard information about his:her patients. about the specified patient. The Research Ethics Committee at the Faculty of Medicine, Uppsala University, approved the ‘‘Sup- Analysis port-Care-Rehabilitation’’ project and the present Content analysis was used to analyse the interviews. study. This method is widely used in communication re- search and is appropriate for analysing open-ended questions (19). The method includes data reduction Sample of GPs and the inter6iew and transformation of data into a form required for Only GPs reported by included patients were consid- analysis. This is done through development of vari- ered. The effects of a possible diffusion of the inter- ables, which makes it possible for independent raters vention between groups were limited by the selection to partition the recording units (the interviews in the of GPs from two distinct groups: Those who received present study) into mutually exclusive classes. The extended information (EIR GPs) and those who did following analyses are concerned with identification not for any patient (controls). Thus, a sample of 10 of patterns that are noteworthy, statistically signifi- GPs matching the total population of GPs in the cant or otherwise may account for the results of the county with regard to the proportions of women analysis (19). All audiotaped interviews were transcribed verba- Table I. Sample of GPs (n). tim and carefully read through several times by the first author to determine variables relevant to the aim Extended No extended information information of the study. The variables in the present analysis were constituted as nominal or ordinal scales, which Gender could take on between two and five values (Tables II, Female 4 4 III, IV and V). Two independent raters (GB and KH) Male 6 6 categorised each interview into one category per vari- GPs’ catchment area able and noted this on a separate rating sheet for Rural 4 4 each interview. The raters were not informed in ad- Town, city 6 6 vance whether or not each GP received extended Cancer diagnoses information about his:her patient, but they were Breast 6 5 aware of the fact that half of the GPs had received Colorectal 2 2 such information. In cases of disagreement between Prostate 2 3 coders concerning the ‘‘correct category’’, this was Time from diagnosis to inter6iew determined through discussions between the two Range (months) 14–24 13–26 coders and the first author after determination of Mean (months) 20 19 inter-rater reliability. Scand J Prim Health Care 2000; 18 Role of the GP in cancer care 145 Table II. GP contacts with the patient and involvement in the care of the cancer. All GPs (n) EIR (n) Control (n) IRR Contacts during the last 3 years No contacts 31 2 Only before cancer diagnosis 43 1 k  0.90 Only after cancer diagnosis 01 0 Before and after cancer diagnosis 12 5 7 Diagnostics of the cancer Involved in the diagnostics 13 7 6 k  0.79 Not involved in the diagnostics 73 4 Treatment and follow -up period Not involved at all 6 3 3 Blood tests only, no contacts 11 0 Contacts for other reasons only 21 1 k  0.62 Contacts for other reasons, cancer-related 64 2 problems also addressed Contacts occasioned by the cancer 51 4 Follow -up of the cancer Responsible for follow-up 1 1 0 k  0.64 Not responsible for follow-up 19 9 10 Table III. GP possibilities to follow the patient through the course of the disease and to determine needs for support. All GPs (n) EIR (n) Control (n) IRR Stat. test GPs’ possibilities to follow the patient through the course of the disease GP had satisfying possibilities 13 10 3 F ex pr. GP did not have satisfying possibilities 70 7 k  0.57 pB0.01 GPs’ possibilities to determine patient needs for and :or to support the patient GP was not informed about the disease or the 4 0 4 treatments GP stated that the patient would have benefited 2 0 2 k  1.00 from improved support compared to the support given GP’s opinion is that the patient did not need 4 4 0 his:her support GP offered the patient contact but the patient 1 1 0 did not respond GP stated that she:he supported the patient by 9 5 4 addressing the disease and:or related problems RESULTS The inter-rater reliability (IRR) was determined by GP’s contacts with the patient Kappa (k) or percent agreement (pa) if k  undefined Table II presents the data concerning GPs’ contacts (20) (Tables II, III, IV and V). The IRR was poor with the specified patients and their involvement in (k  0.05 – 0.20) for the variables concerned with the the care of the cancer. A majority of the 20 GPs were desired information from the specialist clinic. After involved in the diagnostics of the cancer as well as in overlapping categories were joined and redefined the treatment and follow-up periods. The results sug- through discussions between the raters and the first gest no differences between the comparison groups author, the IRR reached fair to almost perfect agree- with respect to these variables. ment (k  0.33 – 0.88, pa  95%) (20). When appropriate, Fisher’s exact probability test (F ex pr.) and the Mann-Whitney U test (MWU) (21) Possibilities for patient follow -up and determination were used to compare between GPs who had experi- of support needs enced the EIR and those who had not (Tables II, III, The results of the analyses of the GP’s possibilities to IV and V). A p-value of B 0.05 was considered to be follow the patient, and to determine patient needs for statistically significant. support, are presented in Table III. All GPs in the Scand J Prim Health Care 2000; 18 146 B. Johansson et al. EIR group declared that they were satisfied with the knowledge about the patients’ disease and treatments possibilities for follow-up, compared with only three as well as their satisfaction with the information from in the control group (F ex pr. pB 0.01). The results the specialist clinics. also suggest that the EIR increased the GP’s possibil- ities to determine patients’ needs of support com- pared to GPs in the control group. DISCUSSION The majority of the GPs were involved in the diag- Desired information from the specialist clinics about nostics of the cancer. It was also common that GPs cancer patients in general stayed in contact with the patient during the period The answers to the question about what information of treatment and follow-up and that cancer-related GPs want from specialist clinics about cancer patients problems were addressed. in general were defined through three hierarchical The information from the specialist clinic seems to variables. Thus, the GPs’ responses were judged as to be insufficient in standard care. Four control GPs whether or not they gave examples of contents of were not informed at all, even though two of them information, important situations for information referred their patient to the specialist clinic owing to and the significance of such information. If they did, a suspected cancer. The EIR led to all GPs being the next step concerned whether or not she:he gave informed about the disease and treatments and in examples of the underlying variables (Table IV). general satisfied with this information. Their possibil- The results suggest that information about the ities to follow the patient increased, and appeared to patient’s cancer and other diseases and information facilitate determination of the patient’s need for sup- about future planning and about the GPs’ role in the port. The extent of contacts with the patients was not care of the patient were considered as the most affected by the EIR. One explanation for this may be important. The results suggest no differences between that GPs who are not properly informed about medi- GPs who did experience the EIR and those who did cal examinations and cancer treatments are more not. likely to stay in contact with the patient owing to an uncertainty about whether or not the patient is well Recei6ed information and knowledge about the cared for. The present result, showing that four EIR specified cancer patient GPs expressed the opinion that the patient did not Table V presents the data concerning the GPs’ opin- need his:her support, endorses such an explanation. ions about received information from the specialist Another possible explanation may pertain to the clinic and their knowledge about the disease and reorganisation of the GPs’ work due to the imple- treatments. mentation and rapid abolition of the Family Doctor The results suggest that EIR GPs received informa- Act during the study period. This may have meant tion from the specialist clinic to a greater extent than that the GPs possibilities for extra contacts were did the control group (F ex pr. pB 0.05). Further- limited. more, the results indicate differences between EIR The analysis of the desired information and the GPs and the control group with regard to their agreement between wanted and received information Table IV. Desired information from the specialist clinic about cancer patients in general. All GPs (n) EIR (n) Control (n) IRR The GP ga6e examples of contents of information 20 10 10 100% Information about the cancer, other diseases and impairments 19 10 9 k  0.64 Information about the future planning and the GP’s role in the care of 20 10 10 95% the patient Information about a specified specialist to contact if needed 2 1 1 k  0.77 The GP ga6e examples of important situations when information was desired 62 4 k  0.88 When the GP had referred a patient to a specialist clinic 3 0 3 k  0.33 When a patient is discharged after a period of inpatient care 2 2 0 k  0.66 When a patient consults a specialist clinic without being referred by the 1 01 k  0.57 GP The GP stressed the significance of information from the specialist clinic The information is important for correct judgement of the patient’s 17 9 8 85% symptoms and difficulties, and to determine the role of the GP in the care of the patient Scand J Prim Health Care 2000; 18 Role of the GP in cancer care 147 Table V. Received information from the specialist clinic and the GP’s knowledge about the specified cancer patient. All GPs (n) EIR (n) Control (n) IRR Stat. test Number of GPs who recei6ed information from the Fexpr. specialist clinic The GP received information 16 10 6 k  0.86 pB0.05 The GP did not receive information 4 0 4 GPs’ opinions of the contents of the information Bad, most medical information missing 1 0 1 Not especially good, some medical information 30 3 MWU  13.0 missing Medical information correct, misses information 44 0 k  0.54 pB0.06 about GP’s role and:or about other patient problems, such as psycho-social status Good, does not miss any information 75 2 Very good 1 0 GPs’ opinions of timing of information Too early, would like continuous information 1 0 1 Too late, would like continuous information 1 0 1 k  1.00 Good, did receive continuous information 6 6 0 Received information too often 33 0 Agreement between desired and recei6ed information MWU  9.0 from the specialist clinic No or very bad agreement 6 0 6 Moderate agreement 4 1 3 k  0.86 pB0.01 Quite good agreement 44 0 Good agreement 6 5 1 GPs’ knowledge about the disease and treatments GP has no knowledge about the disease 2 0 2 GP did refer the patient due to a suspected cancer 2 0 2 MWU  10.0 but has no further information GP is informed about the diagnosis and has limited4p 0 4 k  0.68 B0.01 information about treatments GP is informed about the diagnosis and is well 12 10 2 informed about treatments Five GPs did not express an opinion on timing of information. also indicate that EIR GPs’ needs were not fully findings. There is also a certain risk for Type II met. Three GPs reported that they had received errors, since more powerful parametric statistics information too often, and four reported a lack of could not be used. On the other hand, all GPs information about the GP’s role in the care of the approached agreed to participate, and the present patient’s cancer and about other patient problems. analyses could identify differences between the two Thus, the EIR led to an improvement compared to comparison groups, a fact that supports the asser- standard care but it did not result in an optimal tion that the EIR actually made a difference. Still, information routine to satisfy all GPs’ needs for the presented results must be considered only to information. give a preliminary description of this topic. Further Since the main threat to the internal validity of research is warranted to explore the role of Swedish the conclusions was considered to be a possible dif- GPs in the care of cancer patients, both from their fusion of the intervention between groups, the con- point of view and from that of their patients. trol group was selected from those 17 GPs who did In conclusion, Swedish GPs are commonly in- not receive extended information for any patient. volved in the care of cancer patients, not only in The fact that 48% of the GPs cared for only one the diagnostics of the disease, but also during the project patient suggests that it cannot be considered periods of treatments and follow-up. The informa- exceptional that these 17 GPs did not care for any tion from the specialist clinic to the GP is insuffi- patient. It is therefore unlikely that the selection of cient in standard care. The EIR intervention the control GPs has introduced a bias. increased the GPs’ knowledge about the disease and However, the small sample of GPs in the com- the treatments and appeared to facilitate their possi- parison groups threatens the external validity of the bilities to determine the patients’ need for support. Scand J Prim Health Care 2000; 18 148 B. Johansson et al. ACKNOWLEDGEMENTS 10. Wood ML, McWilliam CL. Cancer in remission. Challenge in collaboration for family physicians and oncologists. Can This study was financed by the Swedish Cancer Soci- Fam Phys 1996;42:899 – 910. ety and by the Swedish Foundation for Health Care 11. Jacobs LGH. Referral letters and replies from orthopedic Sciences and Allergy Research. departments; opportunities missed. BMJ 1990;301:470 – 3. 12. Tattersall MH, Griffin A, Dunn SM, Monaghan H, Scat- shard K, Butow PN. Writing to referring doctors after a new patient consultaion. What is wanted and what was contained in letters from one medical oncologist? Aust NZ REFERENCES J Med 1995;25:479 – 82. 1. Beaulieu MD, Be ´land F, Roy D, Falardeau M, He ´bert G. 13. Hampson JP, Roberts RI, Morgan DA. Shared care: a Factors determining compliance with screening mammog- review of the literature. Fam Pract 1996;13:264 – 79. raphy. Can Med Assoc J 1996;154:1335 – 43. 14. Cullen R. The medical specialist: information gateway or 2. Kiernan GN, Frame PS. Cancer occurrence and screening gatekeeper for the family practitioner. Bull Med Libr As- in family practice. A 20-year experience. J Fam Pract soc 1997;85:348 – 55. 1996;43:49 – 55. 15. Orton P. Shared care. Lancet 1994;344:1413 – 5. 3. Olsson P. Distriktsla ¨ karens roll i uppta ¨ ckt och diagnostik 16. Dahler-Eriksen K, Nielsen JD, Lassen JF, Olesen F. av cancer [The role of the district phyicians in the detection Tvaersektorielle behandlingsprogrammer – et eksempel pa ˚ and diagnosis of cancer]. La ¨ kartidningen 1991;88:2685 – 9. det samarbejdene sundhedsvaesen [Shared care pro- 4. Fakhoury W, McCarthy M, Addington Hall J. Determi- grammes – an example of a cooperating health care sys- nants of informal caregivers’ satisfaction with services for tem]. English summary. Ugeskr Laeger 1998;160:5021 – 4. dying cancer patients. Soc Sci Med 1996;42:721 – 31. 17. National Board of Health and Welfare. Sjukva ˚ rden i 5. Hinton J. Services given and help perceived during home Sverige 1995 [Medical services in Sweden 1995]. Stock- care for terminal cancer. Palliat Med 1996;10:125 – 34. holm: National Board of Health and Welfare; 1995. 6. Grande GE, Todd CJ, Barclay SIG, Doyle JH. What 18. Johansson B, Berglund G, Glimelius B, Holmberg L, Sjo ¨- terminal ill patients value in the support provided by GPs, de ´n P-O. lIntensified primary cancer care: a randomised district nurses and Macmillan nurses. Int J Palliat Nurs study on home care nurse contacts. J Adv Nurs 1996;2:138 – 43. 1999;30:1137 – 346. 7. Grunfeld E, Mant D, Yudkin P, Adewuyi Dalton R, Cole 19. Krippendorff K. Content analysis. An introduction to its D, Stewart J, et al. Routine follow up of breast cancer in methodology. Newbury Park, CA: Sage; 1980. primary care: randomised trial. BMJ 1996;313:665 – 9. 20. Flately Brennan P, Hays BJ. The Kappa statistic for estab- 8. Worster A, Bass MJ, Wood ML. Willingness to follow lishing interrater reliability in the secondary analysis of breast cancer. Can Fam Phys 1996;42:263 – 8. qualitative clinical data. Res Nurs Health 1992;15:153 – 8. 9. Grunfeld E. Specialist and general practice views on rou- 21. Siegel S, Catstellan NJJ. Nonparametric statistics for the tine follow-up of breast cancer patients in general practice. behavioral sciences. 2nd ed. New York: McGraw-Hill; Fam Pract 1995;12:60 – 5. 1988. Scand J Prim Health Care 2000; 18

Journal

Scandinavian Journal of Primary Health CareTaylor & Francis

Published: Jan 1, 2000

There are no references for this article.