- Email: [email protected]
Good continuity of care may improve quality of life in Type 2 diabetes
Diabetes Research and Clinical Practice 51 (2001) 21 – 27 www.elsevier.com/locate/diabres
Good continuity of care may improve quality of life in Type 2 diabetes Jouko Ha¨nninen a,*, Jorma Takala b, Sirkka Keina¨nen-Kiukaanniemi c a Health Centre of Mikkeli, Kiiskinma¨enk. 5 – 7, FIN-50130 Mikkeli, Finland Department of Public Health and General Practice, Uni6ersity of Kuopio, Kuopio, Finland c Department of Public Health Science and General Practice and Oulu Uni6ersity Hospital, Uni6ersity of Oulu, Oulu, Finland b
Received 1 February 2000; received in revised form 10 May 2000; accepted 17 July 2000
Abstract Some features of diabetes care and diabetes treatment regimen which may have an impact on health-related quality of life (HRQOL) in people with diabetes were studied cross-sectionally using the SF-20 questionnaire. Of the 381 subjects with Type 2 diabetes aged under 65 years, 260 (68%) participated in the study. On univariate analysis, HRQOL was associated with regular clinical review (check-up at least twice a year) and continuity of care (the same GP for at least 2 years), education by a diabetes nurse, and satisfaction with diabetes education. No associations were found between the HRQOL dimensions and home glucose monitoring, participation in educational courses, or satisfaction with care. On logistic regression analysis only good continuity of care was significantly associated with the better well-being dimensions of the SF 20 (ORs 2.5–6.0). However, good continuity of care was also associated with less satisfactory glucose control (HbA1c 8.992.0 (9SD) vs 8.392.0%, P=0.04). It is concluded that a permanent physician–patient relationship may improve HRQOL in subjects with Type 2 diabetes, but further prospective studies are needed to confirm this finding. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Type 2 diabetes; Quality of life; Diabetes care; Blood glucose control
1. Introduction People with chronic diseases, such as Type 2 diabetes, have to face many problems which may have an impact on their health-related quality of life (HRQOL). Chronic diseases are often lifelong with uncertain prognosis and despite continuing * Corresponding author. Tel.: +358-15-20111; fax + 35815-210693. E-mail address: [email protected] (J. Ha¨nninen).
treatment may cause symptoms and acute complications. In addition, they are often accompanied by other chronic diseases. These diseases require regular visits to several health care professionals, utilization of the service of clinics and outpatient centres, patient education, and technical devices [1]. Consequently, one of the primary objectives in the treatment of chronic diseases is the improvement of the patient’s HRQOL [2]. Routine diabetes care includes regular checkups by a physician and a diabetes nurse, diabetes
0168-8227/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 0 ) 0 0 1 9 8 - 4
22
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
education, compliance with diet, exercise, glucoselowering agents and self blood-glucose monitoring. This process and the physician – patient relationship are of importance for good results. Presumably, regular check-ups and laboratory monitoring, proper diabetes education, and good continuity of care could also improve HRQOL. In previous studies, people with Type 2 diabetes have been reasonably satisfied with their care [3 – 5], although the reported clinical and laboratory examination rates have been quite low [5,6]. There are only a few studies of the impact of the health care system and diabetes care on HRQOL [4,7–9]. In a prospective trial of monthly contacts with a diabetes nurse to provide patient education and reinforce compliance, glycaemic control was slightly improved but there was no improvement in HRQOL [7]. In the prospective studies, initiation of insulin therapy in Type 2 diabetic subjects tended to impair HRQOL slightly, while disappearance of hyperglycaemic symptoms improved HRQOL [8,9]. Regular blood-glucose monitoring at home had no association with good HRQOL [4]. The objective of the present study was to attempt to identify the factors in diabetes care associated with HRQOL.
2. Materials and methods The study population consisted of people with Type 2 diabetes aged under 65 years and living in the Mikkeli district (population 53 000) in eastern Finland. The formation of the study population has been discussed in detail previously [5]. HRQOL and the rate of depressive symptoms were assessed with the well-validated Short Form20 General Health Survey (SF-20) and the Zung Self-Rating Depression Scale [10,11]. In the Health Centre of Mikkeli, the population was identified from electronic data records. Using either a diagnosis of Type 2 diabetes and/or repeated fasting blood glucose concentrations]6.7 mmol/l, 381 eligible diabetic subjects aged under 65 years were identified, and 260 subjects (68%) participated in the study. The participants were interviewed and examined clinically by one of the
researchers, blood and urine samples were obtained for laboratory tests, and the patients later completed the SF-20 measure in the waiting-room or at home. The study protocol has been described previously in detail [5,12]. Diabetes care was assessed for the following factors: regularity of care (check-ups at least twice a year), continuity of care (seeing the same GP for at least 2 years), diabetes education by a diabetes nurse (participated/not participated), and attendance at educational courses (participated/not participated), blood-glucose monitoring at home (yes/no) and satisfaction with care and diabetes education (yes/no). The definitions of these variables have been discussed previously [5]. In our health care system, people can freely choose and change their GP. In the statistical analyses, scores of the dimensions of the SF-20 were tested in univariate analyses according to different diabetes care factors with Wilcoxon’s test. Those parameters that had significant associations with the HRQOL dimensions in univariate statistics were examined further with multivariate analysis. A logistic regression model was used to predict the odds ratios for the diabetes care factors associated with the different HRQOL dimensions. In the regression model, the dimensions of HRQOL were divided into thirds (with the exception of role functioning) and logistic analysis was carried out by comparing the lowest and highest third of each dimension. The model was modified from our previous results [12,13] by adding independent factors stepwise, and the improvement of the model was assessed with the change of 2 log likelihood\ 3.
3. Results The study population, and the quality of diabetes care have been discussed elsewhere [5]. The non-responders were more often men (65 vs 54%), had not been diagnosed for diabetes (54 vs 30%), and consequently they had less severe diabetes (50 vs 71% were treated with oral hypoglycaemic agents and/or insulin). Of the diabetic subjects, 62% had good continuity and regularity of care.
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
Eighty-five percent had received education from a diabetes nurse, but only 10% had participated in a diabetes education course. Most of the diabetic subjects were satisfied with their diabetes education (87%) and care (86%). Blood glucose was self-monitored by 31% of the patients. The characteristics of the diabetic patients are presented in Table 1 according to the factors of diabetes care. As it can be seen, during the first
23
years of their diabetes career. the patients were reviewed less frequently, obtained less diabetes education and tended to be dissatisfied with their diabetes education and care. These short-duration patients also seemed to have better glucose control than those with diabetes of a longer duration, and had no other treatment except a special diet. Furthermore, the subjects with a long history of diabetes did not have a permanent physician–pa-
Table 1 Characteristics of the diabetic patients according to factors of diabetes care. Significance of differences were tested with Chi square, Wilcoxon’s test and Student’s t-test (N= 260) Males/females
Median duration of diabetes (range) (years)
Mean HbA1c 9S.E. (%)
Mean BMI9S.E. (kg/m2)
Treated with diet only
Re6iew by a GP At least twice a year Less than twice a year P
80/73 54/40 NS
7.5 (1–27) 3 (0–33) B0.001
9.1 90.2 8.090.2 B0.001
30.8 9 0.5 29.6 9 0.5 0.086
21/153 (14) 46/94 (49) B0.001
The same GP For at least 2 years For less than 2 years P
82/79 58/39 NS
6 (0–27) 6 (0–33) NS
8.990.2 8.390.2 0.041
30.5 9 0.4 30.3 9 0.6 NS
40/161 (25) 35/97 (36) 0.054
Education by a diabetes nurse Yes No P
115/106 26/12 0.061
7 (0–33) 4 (0–23) 0.040
8.6 90.1 8.8 90.4 NS
30.2 9 0.4 31.290.9 NS
56/221 (25) 19/38 (50) 0.002
Satisfaction with diabetes education Yes No P
119/106 21/12 NS
6 (0–33) 4 (1–19) 0.038
8.790.1 8.8 90.4 NS
30.2 90.4 31.9 9 1.2 NS
58/225 (26) 17/33 (52) 0.002
Satisfaction with diabetes care Yes No P
120/103 18/13 NS
6 (0–33) 4 (0–26) 0.027
8.7 90.1 8.890A NS
30.4 90.4 30.6 9 1.2 NS
61/224 (27) 12/31 (39) NS
Self blood-glucose monitoring Yes No P
46/33 95/85 NS
9 (1–27) 5 (0–33) B0.001
9.090.2 8.5 90.2 0.037
29.1 9 0.6 30.9 90.4 0.017
8/79 (10) 67/180 (37) B0.001
Participation in a diabetes educational course Yes No P
10/15 131/103 NS
11 (3–22) 6 (0–33) B0.001
9.6 9 0.3 8.6 9 0.1 0.003
31.69 1.1 30.3 9 0.4 NS
2/25 (8) 73/234 (31) 0.015
221 38
Education by a diabetes nurse Yes No P
a
64.6 96.8 62.1 92.0 NS
66.0 9 3.4 60.8 92.3 NS
62.7 9 2.1 56.7 95.7 NS
64.3 9 2.0 49.5 9 5.6 0.013
64.1 92.0 52.6 95.3 0.047
65.6 92.4 56.3 9 3.2 0.017
59.2 92.6 68.3 92.9 0.034
Physical
Significance of differences were tested with Wilcoxon’s test (N =260).
25 234
79 180
Self blood-glucose monitoring Yes No P
Participation in an educational course Yes No P
224 31
Satisfaction with diabetes care Yes No P
225 33
161 97
The same GP For at least 2 years For less than 2 years P
Satisfaction with diabetes education Yes No P
153 94
Re6iew by a GP At least twice a year Less than twice a year P
N
Mean dimension score ( 9S.E.) of the SF-20
54.0 9 10 52.0 93.2 NS
49.3 9 5.6 53.5 93.6 NS
51.4 93.3 55.0 9 8.4 NS
53.5 93.2 41.9 98.7 NS
53.5 9 3.3 44.49 8.2 NS
55.4 93.8 46.2 95.0 NS
49.3 9 4.0 58.5 9 5.1 NS
Role
82.49 3.5 78.4 91.8 NS
80.09 2.7 78.3 9 2.1 NS
80.49 1.7 70.0 9 6.0 NS
79.1 9 1.8 76.39 5.2 NS
79.5 91.8 74.6 9 4.9 NS
80.6 9 2.1 75.59 2.7 0.045
79.5 9 2.1 79.19 2.9 NS
Social
63.29 3.8 66.791.3 NS
65.79 2.2 66.79 1.6 NS
67.3 91.3 61.594.3 NS
67.291.3 59.9 9 4.0 NS
66.89 1.3 63.99 3.8 NS
68.89 1.6 62.19 2.0 0.004
65.591.6 68.892.1 NS
Mental health
45.794.2 45.091.4 NS
45.1 92.3 45.09 1.7 NS
45.8 1.5 40.1 4.3 NS
45.89 1.4 39.4 9 4.3 NS
45.5 9 1.5 42.39 3.9 NS
48.99 1.8 38.39 2.0 B0.001
43.291.6 48.89 2.4 0.081
Health perception
Table 2 Differences in mean dimension scores of the SF–20 General Health Survey in subjects with Type 2 diabetes according to factors of diabetes carea
34.094.8 42.09 1.9 NS
35.69 2.7 43.59 2.2 0.083
40.6 9 1.8 45.09 6.2 NS
42.19 1.9 35.69 5.2 NS
42.391.9 34.994.6 0.099
37.691.8 47.89 2.6 0.004
41.59 2.3 38.79 2,9 NS
Pain
24 J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27 Table 3 The odds ratios and 95% confidence intervals of Type 2 diabetic subjects with good continuity of care to have better quality of life (highest vs lowest third) adjusted for existing ischaemic heart disease and depression in logistic regression analysis (n = 212 in the model) Dimension of HRQOL
OR
95% CI
Physical functioning Role functioning Social functioning Mental health Health perception Painlessness
1.99 1.46 1.81 2.48 3.45 6.02
0.94–4.20 0.79–2.67 0.84–3.91 1.16–5.32 1.52–7.87 1.82–19.92
tient relationship more often than those with a short history. However, the patients who had been treated by the same GP for at least 2 years had clearly less satisfactory blood glucose control (P= 0.041). Table 2 shows that HRQOL was associated with regular care, continuity of care, visits to a diabetes nurse, and satisfaction with diabetes education. Review by a GP less than twice a year was associated with improved physical functioning. Good continuity of care yielded five positive associations: with physical and social functioning, mental health, health perception and pain. Visits to a diabetes nurse and satisfaction with the education were both associated with better physical functioning. No differences were found in any HRQOL dimensions for patients using/not using self blood-glucose monitoring, those who had or had not participated in educational courses, or who were or were not satisfied with their care. The odds ratios of good continuity of care for the best vs worst third of HRQOL, adjusted for depression and coronary heart disease, are given in Table 3. The addition of other factors did not improve the fit of the model, judging by the log likelihood. Good continuity of care seemed to be associated with better well-being, but the HbA1c level was significantly higher compared with the group with poor continuity of care. There were no differences in the duration of diabetes, gender or age between the groups with good and poor continuity of care.
25
4. Discussion The main result of the present study is that continuity of care seems to be an important factor of good HRQOL in people with Type 2 diabetes. The diabetic subjects who had been treated by the same GP for at least 2 years seemed to have better mental health and less pain and they felt more healthy in themselves than those who did not have a permanent physician–patient relationship. The present cross-sectional study cannot reveal causal relationships, and the results should be interpreted with caution. Furthermore, only 56% of the total Type 2 diabetic population were included in the regression analysis. A notable proportion of the non-responders did not have a diagnosis of diabetes in their patient records, and they usually had less severe diabetes than the subjects who participated in the study. This implies that the results can be most safely generalized to middle-aged subjects with previously known Type 2 diabetes. There are several possible explanations for the connection between the continuity of care and good HRQOL. Firstly, the differences in HRQOL and the continuity of care were caused by factors that were not examined. It is possible, for example, that the connection between poor continuity of care and impaired HRQOL was due to socioeconomic factors, such as people who had moved to the district recently and who had less social support [14]. In another investigation of the same study population, only eight persons moved during the 5-year followup, which means that the population in the study area was quite stable [15]. Secondly, the diabetic subjects with poor HRQOL were dissatisfied with their care, and had changed their GP more often. The reasons for the dissatisfaction with care were not asked in the present study. Thirdly, a good patient–physician relationship really improved the patient’s well-being. Finally, it may be that the patients preferred to be treated by GPs who accepted less strict glucose control, and better well-being correlated with fewer demands in diabetes care. The re-
26
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
markable mean difference in HbAlc levels may support this explanation. It was shown in the UKPDS that Type 2 diabetes is a progressive disease, and multiple therapies were hence needed in the long term [16]. The present study revealed no difference in the duration of diabetes, in view of the continuity of care, and the clinically significant difference in HbAlc levels cannot be explained by differences in the severity of the disease. The present study suggests that good continuity of care may improve the treatment satisfaction and HRQOL of patients with chronic conditions, but there is a risk that the quality of medical care is not so good with a single doctor’s input. This issue needs to be examined in future prospective studies [17]. Another important result concerns the role of diabetes nurses and HRQOL. Diabetes education was associated with improved physical functioning and the patients who were satisfied with their education had remarkably higher scores in physical functioning. This may imply that diabetes education encouraged diabetic subjects to do more exercise [7], or else that the patients with poor physical functioning were dissatisfied with the demands to do more physical exercise. At any rate, it has been shown that increased exercise improves HRQOL independently of weight loss [18]. In line with the study of Wredling et al., blood-glucose self-monitoring was not associated with any dimensions of HRQOL [4]. Nor did participation in educational courses have an impact on HRQOL. As a conclusion, a permanent physician – patient relationship in some cases improves HRQOL, but this improvement turned out to be a double-edged sword as in these cases glucose control was less satisfactory. Prospective studies are needed to confirm this finding.
Acknowledgements We thank Ulla Rajala for help in regression modelling and ADP designer Paavo Ma¨kinen for help with the statistical analyses.
References [1] L.A. Strauss, J. Corbin, S. Fagerhaugh, et al., Chronic Illness and the Quality of Life, The C.V. Mosby Company, St Louis, MO, 1984, pp. 11 – 16. [2] K.G.M.M. Alberti, F.A. Gries, Management of non-insulin-dependent diabetes mellitus in Europe: a consensus view, Diabet. Med. 5 (1988) 275 – 281. [3] M. Kamien, A. Ward, F. Mansfield, B. Fatowich, C. Mather, K. Anstevm, Type 2 diabetes. Patient practices and satisfaction with GP care, Aust. Fam. Physician 24 (1995) 1043 – 1049. [4] R. Wredling, J. Stalhammar, U. Adamson, C. Berne, Y. Larsson, J. O8 stman, Wellbeing and treatment satisfaction in adults with diabetes: a Swedish population-based study, Qual. Life Res. 4 (1995) 515 – 522. [5] J. Ha¨nninen, S. Keina¨nen-Kiukaanniemi, J. Takala, Population-based audit of noninsulin-dependent diabetic patients aged under 65 years in primary health care, Scand. J. Prim. Health Care 16 (1998) 227 – 232. [6] P.E. Wa¨ndell, B. Brorsson, H. A, berg, Diabetic patients in primary health care — quality of care three years apart, Scand. J. Prim. Health Care 16 (1998) 44 – 49. [7] M. Weinberger, M.S. Kirkman, G.P. Samsa, E.A. Shortliffe, P.B. Landsman, P.A. Cowper, D.L. Simel, J.R. Feussner, A nurse-coordinated intervention for primary care patients with non-insulin-dependent diabetes mellitus: impact on glycemic control and health-related quality of life, J. Gen. Int. Med. 10 (1995) 59 – 66. [8] J.J. de Sonnaville, F.J. Snoek, L.P. Colly, W. Deville, D. Wijkel, R.J. Heine, Well-being and symptoms in relation to insulin therapy in type 2 diabetes, Diabetes Care 21 (1998) 919 – 924. [9] P.P. Goddijn, H.J. Bilo, E.J. Feskens, K.H. Groeniert, K.I. van der Zee, B. Meyboom-de Jong, Longitudinal study on glycaemic control and quality of life in patients with type 2 diabetes mellitus referred for intensified control, Diab. Med. 16 (1999) 2330. [10] A.L. Stewart, R.D. Hays, J.E. Ware, Jr, The MOS Shortform General Health Survey. Reliability and validity in a patient population, Med. Care 26 (1988) 724 – 735. [11] W.W.K. Zung, Self-Rating Depression Scale, Arch. Gen. Psychiatry 12 (1965) 63 – 70. [12] J. Ha¨nninen, J. Takala, S. Keina¨nen-Kiukaanniemi, Quality of life in NIDDM patients assessed with the SF-20 questionnaire, Diabetes Res. Clin. Pract. 42 (1998) 17 – 27. [13] J.A. Ha¨nninen, J.K. Takala, S.M. Keina¨nen-Kiukaanniemi, Depression in subjects with type 2 diabetes. Predictive factors and relation to quality of life, Diabetes Care 22 (1999) 997 – 998. [14] A.-M. Aalto, A. Uutela, T. Kangas, Health behaviour, social integration, perceived health and dysfunction. A comparison between patients with type I and II diabetes and controls, Scand. J. Soc. Med. 24 (1996) 272 – 281. [15] J. Ha¨nninen, J. Takala, S. Keina¨nen-Kiukaanniemi, Al-
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27 buminuria. and other risk factors for mortality in patients with NIDDM aged under 65 years, Diabetes Res. Clin. Pract. 43 (1999) 121–126. [16] R.C. Turner, C.A. Cull, V. Frighi, R.R. Holman, Glycernic control with diet. sulphonylurea, metformin or insulin in patients with type 2 diabetes mellitus: progressive requirements for multiple therapies (UKPDS 49). UK Prospective Diabetes Study Group, J. Am. Med. Assoc. 281 (1999) 2005 – 2012.
.
27
[17] A.-L. Kinmonth, A. Woodcock, S. Griffin, N. Spiegal, M.J. Campbell, Randomised controlled trial of patient centred care of diabetes in general practice: impact on current wellbeing and future disease risk, Br. Med. J. 317 (1998) 1202 – 1208. [18] R.M. Kaplan, S.L. Hartwell, D.K. Wilson, J.P. Wallace, Effects of diet and exercise interventions on control and quality of life in non-insulin-dependent diabetes mellitus, J. Gen. Intern. Med. 2 (1987) 220 – 228.
Good continuity of care may improve quality of life in Type 2 diabetes Jouko Ha¨nninen a,*, Jorma Takala b, Sirkka Keina¨nen-Kiukaanniemi c a Health Centre of Mikkeli, Kiiskinma¨enk. 5 – 7, FIN-50130 Mikkeli, Finland Department of Public Health and General Practice, Uni6ersity of Kuopio, Kuopio, Finland c Department of Public Health Science and General Practice and Oulu Uni6ersity Hospital, Uni6ersity of Oulu, Oulu, Finland b
Received 1 February 2000; received in revised form 10 May 2000; accepted 17 July 2000
Abstract Some features of diabetes care and diabetes treatment regimen which may have an impact on health-related quality of life (HRQOL) in people with diabetes were studied cross-sectionally using the SF-20 questionnaire. Of the 381 subjects with Type 2 diabetes aged under 65 years, 260 (68%) participated in the study. On univariate analysis, HRQOL was associated with regular clinical review (check-up at least twice a year) and continuity of care (the same GP for at least 2 years), education by a diabetes nurse, and satisfaction with diabetes education. No associations were found between the HRQOL dimensions and home glucose monitoring, participation in educational courses, or satisfaction with care. On logistic regression analysis only good continuity of care was significantly associated with the better well-being dimensions of the SF 20 (ORs 2.5–6.0). However, good continuity of care was also associated with less satisfactory glucose control (HbA1c 8.992.0 (9SD) vs 8.392.0%, P=0.04). It is concluded that a permanent physician–patient relationship may improve HRQOL in subjects with Type 2 diabetes, but further prospective studies are needed to confirm this finding. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Type 2 diabetes; Quality of life; Diabetes care; Blood glucose control
1. Introduction People with chronic diseases, such as Type 2 diabetes, have to face many problems which may have an impact on their health-related quality of life (HRQOL). Chronic diseases are often lifelong with uncertain prognosis and despite continuing * Corresponding author. Tel.: +358-15-20111; fax + 35815-210693. E-mail address: [email protected] (J. Ha¨nninen).
treatment may cause symptoms and acute complications. In addition, they are often accompanied by other chronic diseases. These diseases require regular visits to several health care professionals, utilization of the service of clinics and outpatient centres, patient education, and technical devices [1]. Consequently, one of the primary objectives in the treatment of chronic diseases is the improvement of the patient’s HRQOL [2]. Routine diabetes care includes regular checkups by a physician and a diabetes nurse, diabetes
0168-8227/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 0 ) 0 0 1 9 8 - 4
22
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
education, compliance with diet, exercise, glucoselowering agents and self blood-glucose monitoring. This process and the physician – patient relationship are of importance for good results. Presumably, regular check-ups and laboratory monitoring, proper diabetes education, and good continuity of care could also improve HRQOL. In previous studies, people with Type 2 diabetes have been reasonably satisfied with their care [3 – 5], although the reported clinical and laboratory examination rates have been quite low [5,6]. There are only a few studies of the impact of the health care system and diabetes care on HRQOL [4,7–9]. In a prospective trial of monthly contacts with a diabetes nurse to provide patient education and reinforce compliance, glycaemic control was slightly improved but there was no improvement in HRQOL [7]. In the prospective studies, initiation of insulin therapy in Type 2 diabetic subjects tended to impair HRQOL slightly, while disappearance of hyperglycaemic symptoms improved HRQOL [8,9]. Regular blood-glucose monitoring at home had no association with good HRQOL [4]. The objective of the present study was to attempt to identify the factors in diabetes care associated with HRQOL.
2. Materials and methods The study population consisted of people with Type 2 diabetes aged under 65 years and living in the Mikkeli district (population 53 000) in eastern Finland. The formation of the study population has been discussed in detail previously [5]. HRQOL and the rate of depressive symptoms were assessed with the well-validated Short Form20 General Health Survey (SF-20) and the Zung Self-Rating Depression Scale [10,11]. In the Health Centre of Mikkeli, the population was identified from electronic data records. Using either a diagnosis of Type 2 diabetes and/or repeated fasting blood glucose concentrations]6.7 mmol/l, 381 eligible diabetic subjects aged under 65 years were identified, and 260 subjects (68%) participated in the study. The participants were interviewed and examined clinically by one of the
researchers, blood and urine samples were obtained for laboratory tests, and the patients later completed the SF-20 measure in the waiting-room or at home. The study protocol has been described previously in detail [5,12]. Diabetes care was assessed for the following factors: regularity of care (check-ups at least twice a year), continuity of care (seeing the same GP for at least 2 years), diabetes education by a diabetes nurse (participated/not participated), and attendance at educational courses (participated/not participated), blood-glucose monitoring at home (yes/no) and satisfaction with care and diabetes education (yes/no). The definitions of these variables have been discussed previously [5]. In our health care system, people can freely choose and change their GP. In the statistical analyses, scores of the dimensions of the SF-20 were tested in univariate analyses according to different diabetes care factors with Wilcoxon’s test. Those parameters that had significant associations with the HRQOL dimensions in univariate statistics were examined further with multivariate analysis. A logistic regression model was used to predict the odds ratios for the diabetes care factors associated with the different HRQOL dimensions. In the regression model, the dimensions of HRQOL were divided into thirds (with the exception of role functioning) and logistic analysis was carried out by comparing the lowest and highest third of each dimension. The model was modified from our previous results [12,13] by adding independent factors stepwise, and the improvement of the model was assessed with the change of 2 log likelihood\ 3.
3. Results The study population, and the quality of diabetes care have been discussed elsewhere [5]. The non-responders were more often men (65 vs 54%), had not been diagnosed for diabetes (54 vs 30%), and consequently they had less severe diabetes (50 vs 71% were treated with oral hypoglycaemic agents and/or insulin). Of the diabetic subjects, 62% had good continuity and regularity of care.
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
Eighty-five percent had received education from a diabetes nurse, but only 10% had participated in a diabetes education course. Most of the diabetic subjects were satisfied with their diabetes education (87%) and care (86%). Blood glucose was self-monitored by 31% of the patients. The characteristics of the diabetic patients are presented in Table 1 according to the factors of diabetes care. As it can be seen, during the first
23
years of their diabetes career. the patients were reviewed less frequently, obtained less diabetes education and tended to be dissatisfied with their diabetes education and care. These short-duration patients also seemed to have better glucose control than those with diabetes of a longer duration, and had no other treatment except a special diet. Furthermore, the subjects with a long history of diabetes did not have a permanent physician–pa-
Table 1 Characteristics of the diabetic patients according to factors of diabetes care. Significance of differences were tested with Chi square, Wilcoxon’s test and Student’s t-test (N= 260) Males/females
Median duration of diabetes (range) (years)
Mean HbA1c 9S.E. (%)
Mean BMI9S.E. (kg/m2)
Treated with diet only
Re6iew by a GP At least twice a year Less than twice a year P
80/73 54/40 NS
7.5 (1–27) 3 (0–33) B0.001
9.1 90.2 8.090.2 B0.001
30.8 9 0.5 29.6 9 0.5 0.086
21/153 (14) 46/94 (49) B0.001
The same GP For at least 2 years For less than 2 years P
82/79 58/39 NS
6 (0–27) 6 (0–33) NS
8.990.2 8.390.2 0.041
30.5 9 0.4 30.3 9 0.6 NS
40/161 (25) 35/97 (36) 0.054
Education by a diabetes nurse Yes No P
115/106 26/12 0.061
7 (0–33) 4 (0–23) 0.040
8.6 90.1 8.8 90.4 NS
30.2 9 0.4 31.290.9 NS
56/221 (25) 19/38 (50) 0.002
Satisfaction with diabetes education Yes No P
119/106 21/12 NS
6 (0–33) 4 (1–19) 0.038
8.790.1 8.8 90.4 NS
30.2 90.4 31.9 9 1.2 NS
58/225 (26) 17/33 (52) 0.002
Satisfaction with diabetes care Yes No P
120/103 18/13 NS
6 (0–33) 4 (0–26) 0.027
8.7 90.1 8.890A NS
30.4 90.4 30.6 9 1.2 NS
61/224 (27) 12/31 (39) NS
Self blood-glucose monitoring Yes No P
46/33 95/85 NS
9 (1–27) 5 (0–33) B0.001
9.090.2 8.5 90.2 0.037
29.1 9 0.6 30.9 90.4 0.017
8/79 (10) 67/180 (37) B0.001
Participation in a diabetes educational course Yes No P
10/15 131/103 NS
11 (3–22) 6 (0–33) B0.001
9.6 9 0.3 8.6 9 0.1 0.003
31.69 1.1 30.3 9 0.4 NS
2/25 (8) 73/234 (31) 0.015
221 38
Education by a diabetes nurse Yes No P
a
64.6 96.8 62.1 92.0 NS
66.0 9 3.4 60.8 92.3 NS
62.7 9 2.1 56.7 95.7 NS
64.3 9 2.0 49.5 9 5.6 0.013
64.1 92.0 52.6 95.3 0.047
65.6 92.4 56.3 9 3.2 0.017
59.2 92.6 68.3 92.9 0.034
Physical
Significance of differences were tested with Wilcoxon’s test (N =260).
25 234
79 180
Self blood-glucose monitoring Yes No P
Participation in an educational course Yes No P
224 31
Satisfaction with diabetes care Yes No P
225 33
161 97
The same GP For at least 2 years For less than 2 years P
Satisfaction with diabetes education Yes No P
153 94
Re6iew by a GP At least twice a year Less than twice a year P
N
Mean dimension score ( 9S.E.) of the SF-20
54.0 9 10 52.0 93.2 NS
49.3 9 5.6 53.5 93.6 NS
51.4 93.3 55.0 9 8.4 NS
53.5 93.2 41.9 98.7 NS
53.5 9 3.3 44.49 8.2 NS
55.4 93.8 46.2 95.0 NS
49.3 9 4.0 58.5 9 5.1 NS
Role
82.49 3.5 78.4 91.8 NS
80.09 2.7 78.3 9 2.1 NS
80.49 1.7 70.0 9 6.0 NS
79.1 9 1.8 76.39 5.2 NS
79.5 91.8 74.6 9 4.9 NS
80.6 9 2.1 75.59 2.7 0.045
79.5 9 2.1 79.19 2.9 NS
Social
63.29 3.8 66.791.3 NS
65.79 2.2 66.79 1.6 NS
67.3 91.3 61.594.3 NS
67.291.3 59.9 9 4.0 NS
66.89 1.3 63.99 3.8 NS
68.89 1.6 62.19 2.0 0.004
65.591.6 68.892.1 NS
Mental health
45.794.2 45.091.4 NS
45.1 92.3 45.09 1.7 NS
45.8 1.5 40.1 4.3 NS
45.89 1.4 39.4 9 4.3 NS
45.5 9 1.5 42.39 3.9 NS
48.99 1.8 38.39 2.0 B0.001
43.291.6 48.89 2.4 0.081
Health perception
Table 2 Differences in mean dimension scores of the SF–20 General Health Survey in subjects with Type 2 diabetes according to factors of diabetes carea
34.094.8 42.09 1.9 NS
35.69 2.7 43.59 2.2 0.083
40.6 9 1.8 45.09 6.2 NS
42.19 1.9 35.69 5.2 NS
42.391.9 34.994.6 0.099
37.691.8 47.89 2.6 0.004
41.59 2.3 38.79 2,9 NS
Pain
24 J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27 Table 3 The odds ratios and 95% confidence intervals of Type 2 diabetic subjects with good continuity of care to have better quality of life (highest vs lowest third) adjusted for existing ischaemic heart disease and depression in logistic regression analysis (n = 212 in the model) Dimension of HRQOL
OR
95% CI
Physical functioning Role functioning Social functioning Mental health Health perception Painlessness
1.99 1.46 1.81 2.48 3.45 6.02
0.94–4.20 0.79–2.67 0.84–3.91 1.16–5.32 1.52–7.87 1.82–19.92
tient relationship more often than those with a short history. However, the patients who had been treated by the same GP for at least 2 years had clearly less satisfactory blood glucose control (P= 0.041). Table 2 shows that HRQOL was associated with regular care, continuity of care, visits to a diabetes nurse, and satisfaction with diabetes education. Review by a GP less than twice a year was associated with improved physical functioning. Good continuity of care yielded five positive associations: with physical and social functioning, mental health, health perception and pain. Visits to a diabetes nurse and satisfaction with the education were both associated with better physical functioning. No differences were found in any HRQOL dimensions for patients using/not using self blood-glucose monitoring, those who had or had not participated in educational courses, or who were or were not satisfied with their care. The odds ratios of good continuity of care for the best vs worst third of HRQOL, adjusted for depression and coronary heart disease, are given in Table 3. The addition of other factors did not improve the fit of the model, judging by the log likelihood. Good continuity of care seemed to be associated with better well-being, but the HbA1c level was significantly higher compared with the group with poor continuity of care. There were no differences in the duration of diabetes, gender or age between the groups with good and poor continuity of care.
25
4. Discussion The main result of the present study is that continuity of care seems to be an important factor of good HRQOL in people with Type 2 diabetes. The diabetic subjects who had been treated by the same GP for at least 2 years seemed to have better mental health and less pain and they felt more healthy in themselves than those who did not have a permanent physician–patient relationship. The present cross-sectional study cannot reveal causal relationships, and the results should be interpreted with caution. Furthermore, only 56% of the total Type 2 diabetic population were included in the regression analysis. A notable proportion of the non-responders did not have a diagnosis of diabetes in their patient records, and they usually had less severe diabetes than the subjects who participated in the study. This implies that the results can be most safely generalized to middle-aged subjects with previously known Type 2 diabetes. There are several possible explanations for the connection between the continuity of care and good HRQOL. Firstly, the differences in HRQOL and the continuity of care were caused by factors that were not examined. It is possible, for example, that the connection between poor continuity of care and impaired HRQOL was due to socioeconomic factors, such as people who had moved to the district recently and who had less social support [14]. In another investigation of the same study population, only eight persons moved during the 5-year followup, which means that the population in the study area was quite stable [15]. Secondly, the diabetic subjects with poor HRQOL were dissatisfied with their care, and had changed their GP more often. The reasons for the dissatisfaction with care were not asked in the present study. Thirdly, a good patient–physician relationship really improved the patient’s well-being. Finally, it may be that the patients preferred to be treated by GPs who accepted less strict glucose control, and better well-being correlated with fewer demands in diabetes care. The re-
26
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27
markable mean difference in HbAlc levels may support this explanation. It was shown in the UKPDS that Type 2 diabetes is a progressive disease, and multiple therapies were hence needed in the long term [16]. The present study revealed no difference in the duration of diabetes, in view of the continuity of care, and the clinically significant difference in HbAlc levels cannot be explained by differences in the severity of the disease. The present study suggests that good continuity of care may improve the treatment satisfaction and HRQOL of patients with chronic conditions, but there is a risk that the quality of medical care is not so good with a single doctor’s input. This issue needs to be examined in future prospective studies [17]. Another important result concerns the role of diabetes nurses and HRQOL. Diabetes education was associated with improved physical functioning and the patients who were satisfied with their education had remarkably higher scores in physical functioning. This may imply that diabetes education encouraged diabetic subjects to do more exercise [7], or else that the patients with poor physical functioning were dissatisfied with the demands to do more physical exercise. At any rate, it has been shown that increased exercise improves HRQOL independently of weight loss [18]. In line with the study of Wredling et al., blood-glucose self-monitoring was not associated with any dimensions of HRQOL [4]. Nor did participation in educational courses have an impact on HRQOL. As a conclusion, a permanent physician – patient relationship in some cases improves HRQOL, but this improvement turned out to be a double-edged sword as in these cases glucose control was less satisfactory. Prospective studies are needed to confirm this finding.
Acknowledgements We thank Ulla Rajala for help in regression modelling and ADP designer Paavo Ma¨kinen for help with the statistical analyses.
References [1] L.A. Strauss, J. Corbin, S. Fagerhaugh, et al., Chronic Illness and the Quality of Life, The C.V. Mosby Company, St Louis, MO, 1984, pp. 11 – 16. [2] K.G.M.M. Alberti, F.A. Gries, Management of non-insulin-dependent diabetes mellitus in Europe: a consensus view, Diabet. Med. 5 (1988) 275 – 281. [3] M. Kamien, A. Ward, F. Mansfield, B. Fatowich, C. Mather, K. Anstevm, Type 2 diabetes. Patient practices and satisfaction with GP care, Aust. Fam. Physician 24 (1995) 1043 – 1049. [4] R. Wredling, J. Stalhammar, U. Adamson, C. Berne, Y. Larsson, J. O8 stman, Wellbeing and treatment satisfaction in adults with diabetes: a Swedish population-based study, Qual. Life Res. 4 (1995) 515 – 522. [5] J. Ha¨nninen, S. Keina¨nen-Kiukaanniemi, J. Takala, Population-based audit of noninsulin-dependent diabetic patients aged under 65 years in primary health care, Scand. J. Prim. Health Care 16 (1998) 227 – 232. [6] P.E. Wa¨ndell, B. Brorsson, H. A, berg, Diabetic patients in primary health care — quality of care three years apart, Scand. J. Prim. Health Care 16 (1998) 44 – 49. [7] M. Weinberger, M.S. Kirkman, G.P. Samsa, E.A. Shortliffe, P.B. Landsman, P.A. Cowper, D.L. Simel, J.R. Feussner, A nurse-coordinated intervention for primary care patients with non-insulin-dependent diabetes mellitus: impact on glycemic control and health-related quality of life, J. Gen. Int. Med. 10 (1995) 59 – 66. [8] J.J. de Sonnaville, F.J. Snoek, L.P. Colly, W. Deville, D. Wijkel, R.J. Heine, Well-being and symptoms in relation to insulin therapy in type 2 diabetes, Diabetes Care 21 (1998) 919 – 924. [9] P.P. Goddijn, H.J. Bilo, E.J. Feskens, K.H. Groeniert, K.I. van der Zee, B. Meyboom-de Jong, Longitudinal study on glycaemic control and quality of life in patients with type 2 diabetes mellitus referred for intensified control, Diab. Med. 16 (1999) 2330. [10] A.L. Stewart, R.D. Hays, J.E. Ware, Jr, The MOS Shortform General Health Survey. Reliability and validity in a patient population, Med. Care 26 (1988) 724 – 735. [11] W.W.K. Zung, Self-Rating Depression Scale, Arch. Gen. Psychiatry 12 (1965) 63 – 70. [12] J. Ha¨nninen, J. Takala, S. Keina¨nen-Kiukaanniemi, Quality of life in NIDDM patients assessed with the SF-20 questionnaire, Diabetes Res. Clin. Pract. 42 (1998) 17 – 27. [13] J.A. Ha¨nninen, J.K. Takala, S.M. Keina¨nen-Kiukaanniemi, Depression in subjects with type 2 diabetes. Predictive factors and relation to quality of life, Diabetes Care 22 (1999) 997 – 998. [14] A.-M. Aalto, A. Uutela, T. Kangas, Health behaviour, social integration, perceived health and dysfunction. A comparison between patients with type I and II diabetes and controls, Scand. J. Soc. Med. 24 (1996) 272 – 281. [15] J. Ha¨nninen, J. Takala, S. Keina¨nen-Kiukaanniemi, Al-
J. Ha¨nninen et al. / Diabetes Research and Clinical Practice 51 (2001) 21–27 buminuria. and other risk factors for mortality in patients with NIDDM aged under 65 years, Diabetes Res. Clin. Pract. 43 (1999) 121–126. [16] R.C. Turner, C.A. Cull, V. Frighi, R.R. Holman, Glycernic control with diet. sulphonylurea, metformin or insulin in patients with type 2 diabetes mellitus: progressive requirements for multiple therapies (UKPDS 49). UK Prospective Diabetes Study Group, J. Am. Med. Assoc. 281 (1999) 2005 – 2012.
.
27
[17] A.-L. Kinmonth, A. Woodcock, S. Griffin, N. Spiegal, M.J. Campbell, Randomised controlled trial of patient centred care of diabetes in general practice: impact on current wellbeing and future disease risk, Br. Med. J. 317 (1998) 1202 – 1208. [18] R.M. Kaplan, S.L. Hartwell, D.K. Wilson, J.P. Wallace, Effects of diet and exercise interventions on control and quality of life in non-insulin-dependent diabetes mellitus, J. Gen. Intern. Med. 2 (1987) 220 – 228.