Compliance of Diabetic Patients . Pros and Cons

Diabetes is a major public health problem allover the world. Compliance refers to a patient both agreeing to and then undergoing some part of their treatment program as advised by their doctor or other health care workers. Aim: This study was conducted to demonstrate the pattern of compliance among diabetic patients, its determinants and reasons beyond noncompliance. Methodology: This cross section descriptive study was conducted at the outpatient Clinic of Diabetes, Kasr Al-Aini. Their compliance pattern were assessed using special scoring. Results: Improper compliance to diet was more prevalent [104 (34.7%)]. The significant determinants of improper compliance to diet were younger age, type 1 diabetes, long duration of illness, absence of complication, positive family history, receiving 3 or more drugs per day and improper compliance to drugs. Financial constraints and depression were beyond improper diet and drugs compliance. Conclusion: Compliance to appointment and drugs were better than compliance to diet. Cost and psychological depression were among the main reasons of improper compliance.


Introduction
DIABETES is a major public health problem allover the world.The World Health Organization (WHO) estimates that more than 180 million people worldwide have diabetes.This number is likely to be more than double by 2030 [1] .There is a high clinical and economic burden from the disease; people with diabetes have a two-to-four fold increased risk of cardiovascular disease compared to the general population and increased incidence of retinopathy, peripheral nerve damage and renal problems [2] .
Diabetes is a challenging disease to manage successfully.Although the care regimen is complex, patients with good diabetes self-care behaviors can attain excellent glycemic control.However, many patients do not achieve good glycemic control and continue to suffer health problems as a result [3] .
A dictionary definition of "compliance" suggests a disposition to yield to or comply with the wishes of others.Compliance in a medical context refers to a patient both agreeing to and then undergoing some part of their treatment program as advised by their doctor or other healthcare worker [4] .
Adherence has been defined as the "active, voluntary and collaborative involvement of the patient in a mutually acceptable course of behavior to produce a therapeutic result" [5] .Some years ago, diabetes educators argued that the term "adherence" be used preferentially instead of "compliance" and the term "nonadherence" be substituted for "noncompliance" [6] .However, much of the medical community especially primary care providers in active clinical practice has never adopted this concept and the term "noncompliance" remains the standard description of this adverse patient behavior [7] .
The first step to improve patient compliance is to understand why noncompliance occurs.A substantial literature has documented a number of factors related to diabetes regimen compliance problems.They include demographic, psychological and social factors, as well as health care provider, medical system and disease-and treatmentrelated factors [8] .
We selected for this study patients with diabetes because they are chronic patients required to follow a complex regimen and so they face a compliance problem.This study was conducted to demonstrate the pattern of compliance among diabetic patients, its determinants (what is with and what is against)

Study setting and design:
This cross section descriptive study was conducted at the outpatient Clinic of Diabetes, Cairo University, over a period of 4 months from June to September, 2007.

Study participants:
The study enrolled a total of 300 patients who were regularly registered and followed-up at the clinic for at least one year prior to their inclusion in the study.

Ethical considerations:
Data confidentiality was preserved in accordance with the Revised Helsinki Declaration of Bioethics [9] .All patients were informed about the aim of the study and consents were obtained from the participants.

Methods:
All the patients were interrogated by a pre-tested questionnaire covering the following items: 1-Sociodemographic characteristics.

3-Compliance pattern:
• Compliance to appointment: It was assessed by calculating the ratio of attendance to the number of pre-determined and set appointments.A score of 3 was given to patients attending the clinic 5/6 or 6/6 times in the last 6 months (proper compliance), a score of 2 was given to patients attending the clinic 3/6 or 4/6 times (fair compliance) and a score of 1 was given to patients attending the clinic less than 3/6 times (poor compliance).

• Compliance to diet regimen:
According to what was stated by the patients; a score of 3 was given to patients who said that they are strictly compliant to the regimen in the last month (proper compliance), a score of 2 was given to patients who are sometimes compliant (fair compliance) and a score of 1 was given to patients who are not compliant at all (poor compliance).

• Compliance to drug regimen:
According to what was stated by the patients; a score of 3 was given to patients who said that they are strictly compliant to the regimen in the last month (proper compliance), a score of 2 was given to patients who ignored treatment 1 to 3 times per month (fair compliance) and a score of 1 was given to patients who ignored treatment more than 3 times per month (poor compliance).
Scoring of compliance was done according to Khattab et al. [10] .
• Non compliant patients were asked about reasons of improper compliance through open ended questions.

4-Measurement of quality of care:
We based our measures on the Diabetes Quality Improvement Project (DQIP).Detailed descriptions of each measure are publicly available through the Internet [11] .The measures derived from DQIP were as follows: Medical record documentation of blood pressure measurement every time the patient attend the clinic, fasting blood sugar, an eye examination, foot examination, laboratory examination for HbA 1 c , lipid panel and urine microalbumin screen in the previous year.A scoring system was calculated to assess the quality of care.
We use the weighted care score proposed by Gulliford et al. [12] .This score takes into account the importance of glycemic and blood pressure measurements relatively to other measurements, so a score was calculated in which glucose and blood pressure measurement were given a weighted score of 4; while the other measurements were given a score of 1 (total score of 13).

5-Assesment of glycaemic control:
We depend upon the fasting blood glucose (FBS) rather than HbA 1c as it is cheaper and can be done at the outpatient clinic.Good glycemic control is defined when FBS is <130mg/dl and poor control was defined when FBS is >_ 130 [13] .

Statistical analysis:
Data were collected and analyzed using the statistical package for social sciences (SPSS) version 11.Univariate relationships between the dependent variables (degree of compliance with diet, drugs and appointment) and independent variables related to patients, disease and care characteristics were examined using the chi-square test for categorical variables and student t-test for continuous variables.A p value of 0.05 was used as the cutoff level for statistical significance.Multivariate analysis (logistic regression) was performed with non-compliance as the dichotomous outcome variable with those variables which were significant in the chi-square as predictor variables in order to calculate adjusted odds ratios (OR) and 95% confidence intervals (CI).

General characteristics of the patients (Table 1):
Three hundred patients were recruited in this study with age ranged from 16-66 years.Nearly three fourths of the patients were females.Most of the patients were married coming from urban areas and nearly half of them were illiterate.92.7% of them were suffering from type 2 diabetes with duration ranged from one to 26 years.About one third of them were suffering from different types of complications especially macrovascular complication in the form of cardiovascular and cerebrovascular diseases.More than half of the patients showed positive family history of diabetes.All of the patients received treatment for diabetes and 60% of them received treatment for other diseases as liver cirrhosis (33%), hypertension (31%) and coronary heart disease (20%).2-5):
As regard to patient characteristics: Literate, working married patients were significantly less compliant to appointment, while young aged patients were significantly less compliant to diet.Urban residence was the only significant determinant to improper compliance to drugs (Table 3).
Regarding disease characteristics: Type 1 diabetes and long duration of the disease were significant determinants to improper diet compliance.Presence of complication was significantly associated with improper compliance to appointment, on the other hand absence of complication was significantly associated with improper diet and drug compliance.Previous experience of the disease among the relatives of patients had a significant negative impact on patient's compliance to appointment and diet (Table 4).
Receiving three or more drugs per day was significantly associated with improper appointment, diet and drug compliance, while single daily dose administration of drugs was significantly associated with improper compliance to appointment only (Table 5).
Quality of care showed no significant relation to pattern of compliance among the patients.As regard to glycaemic control, patient with controlled blood glucose levels attending the clinic less regularly than those with uncontrolled levels.Patients with improper compliance to the diet and drug regimens showed high level of blood glucose (Table 5).Logistic regression analysis was done and it included only the significant determinants that result from the univariate analysis that affect appointment, diet, drugs and total compliance in order to demonstrate the significant predictors of compliance.The analysis reveled that (Table 6).
Significant predictors of improper appointment compliance were mainly related to numbers and frequency of drugs taken by the patients.Significant predictors to improper diet compliance were younger age, increase duration of the disease, +ve family history and improper drug compliance.The only significant predictor to improper drug compliance was improper diet compliance, while +ve family history was the only significant predictors to improper total compliance.
As regard to main reasons of improper compliance as stated by the patients: Felt good health was the main reason beyond improper appointment (32.1%), while financial constraints were beyond improper diet and drugs compliance (41.3% and 70.1% respectively) (Table 7).

Discussion
It has been generally acknowledged for years that noncompliance rates for chronic illness regimens and for lifestyle changes are 50% [14] .Our results are in agreement with this finding (49.3%).
As a group, patients with diabetes are especially prone to substantial regimen compliance problems.In general, research has shown that the diabetes regimen is multidimensional and compliance to one regimen component may be unrelated to compliance in other regimen areas [14,15] .This study shows better compliance to appointment and drugs rather than to diet.This fact is supported by the study conducted by Khattab et al. [10] who mentioned also that compliance with some aspect of diabetic regimen as appointment and drugs can predict compliance with other aspect as diet (odds' ratio 3.1 and 14.9 respectively).Our result showed that compliance with drugs is a good significant predictor to compliance with diet (odd's ratio=24.2) and on the other hand compliance to diet is a significant predictor to compliance to drugs (odds' ratio= 11.1).
Appointment compliance in this study is higher than that reported by Diabetes Attitudes, Wishes and Needs (DAWN) study recently interviewed over 5000 persons with diabetes in 13 countries around the world [16] .This higher figure is due to the fact that some of the drugs are dispensed free of charge every month to the diabetic patients in the outpatient clinic in Kasr El-Aini hospital.So the patients are keen to attend the clinic on regular basis every month.This finding is also supported by a higher rate of drug compliance than reported by other study [17] .
The pattern of diet compliance in this study is in concordance with that reported by Kravitz et al.
[15] (65.3% Vs. 65% respectively) but it is higher than the results of DOWN study [26] (65.3% Vs. 37% respectively).Low level of diet compliance reported by DOWN study may be due to variability of the dietary habits in different nations all over the world.
As regard to sociodemographic determinants, younger age was associated significantly with bad diet compliance, but with good appointment compliance.Literate working patients showed significant bad appointment compliance and this may be related to unsuitable clinic time as in Kasr El-Aini hospital all the clinic are working from 8am to 2am only.This time coincides with time of work for the working patients.Another explanation may be related to high socioeconomic level of the literate working patients that allow them to follow-up their health status in the health insurance system or on private basis.The literate working patients also showed bad diet and drug compliance that may be secondary to bad appointment compliance This result is in contradictory to Delamater et al. [8] who stated that demographic factors such as ethnic minority, low socioeconomic status and low levels of education have been associated with lower regimen adherence and greater diabetes-related morbidity.
As regard to disease determinants, patients with type 2 diabetes showed less compliance to appointment and drugs than patients with type 1 and this relates to the nature of type 1 diabetes which occur at young age.Research in this area has associated various psychological factors beyond this finding.These factors include encouragement and support from parents, fear of complications, will power, motivation and a sense of normality.On the other hand patients with type 1 diabetes demonstrated significantly less diet compliance than patients with type 2 diabetes as this group of patients wish to have a similar lifestyle to their peers and the other physical, emotional and social changes, can all affect the ability to manage their lifestyle properly [18] .Better adherence for type 1 than for type 2 diabetic patients across most regimen domains was also observed by DOWN study [16] .
Research has generally shown that lower regimen adherence can be expected when a health condition is chronic [21] .Longer duration of diabetes was associated with bad compliance to ap-pointment, diet and total compliance.Logistic regression analysis supported this finding especially for diet compliance.Similar result was obtained by Khattab et al. [10] who mentioned that increasing duration of diabetes was found to be the predictive of decreasing total compliance score.
In the current study, despite the presence of complication was associated with bad appointment, it was associated with good diet and drug compliance and this may be explained by: Presence of complication may limit the movement of the patients but its burden enforce the patients to follow the diet and drug regimen to decrease its severity.This finding is in agreement with that reported by Dietrich [19] who concluded that when diabetes complications started, the patients' compliance improved.Among diabetes-related factors significantly associated with low ratings of compliance was positive family history of diabetes as stated by Clarke and Goosen [20] .Similar results was drown from this study and this could be explained by; presence of diabetic relative may give the patient the sense of being more familiar with the disease and this consequently may reflect on their compliance.
As regard to care determinants, research among diabetic patients indicates that simple regimen where patients take only one type of drug, once a day, yields better compliance.Decreases in compliance of 22% for each increase in frequency of daily dosing had been found [21] .Our results support this finding as 100% of patients receiving single daily dose are good compliant to medication than those receiving 2 or 3 doses per day.
Studies of compliance with oral hypoglycemic drugs (OHDs) and insulin prescriptions showed compliance rates averaging approximately 75% for OHDs compliance and 63% for insulin compliance [21] .Our data also demonstrated higher rate of compliance to OHDs than to insulin.
Polypharmacy is the natural consequence of providing evidence-based medical care to patients with type 2 diabetes [22] .Patient compliance to prescribed medications is crucial to the goal of reaching metabolic control.Improper compliance in this study was associated with receiving 3 or more drugs per day.Logistic regression analysis demonstrated that the significant predictors to improper appointment was taking 3 or more medication per day in repeated 3 or more doses.
Good care was found to be associated with better compliance.This might be explained by the increasing satisfaction of the diabetic patients with the improving quality of care and their relationship with the outpatient clinic team, which are important determinants of good compliance [23] .Similar finding is obtained by Khattab et al. [10] .Although no significant difference was detected as regard to care score, yet lower scores were associated with improper compliance.
A wide variety of studies demonstrated that patients who fail to adhere to prescribed clinical regimens have very poor glycaemic control [24] .Similar result from our study is in concordance with these finding.
To improve patient compliance, it is important to understand why noncompliance occurs [8] .In this study, patient health (whether good or bad) accounts for 52.9% of causes of non compliance with appointment.Cost is one of the issues that affect patient compliance [25] .Patients typically cope with economic hardship by not having prescriptions filled, taking a smaller dose, or buying a cheaper over-the-counter product that is presumed to have a similar effect.In this study cost is responsible for 70.1% of causes of noncompliance to drugs and 41.3% of non compliance to diet.Although some of the drugs were dispensed free at the out patient clinic, it is still not sufficient to all patients.
Depression is at least twice as common in patients with diabetes as compared with the nondiabetic population and it affects as many as onequarter of patients with diabetes.It may contribute to problematic medication use, due to increase in forgetfulness and/or a loss of interest in protecting one's health [26] .This study shows that depression accounted for more than quarter of causes of non compliance to diet as well as nearly one fifth of causes of bad drug compliance in the form of forgetfulness and not feeling any benefits of treatment.

Limitation to the study:
We purposely chose patients with diabetes because we wanted to study people with a chronic illness who faced a challenging compliance regimen, but this leaves unresolved question about whether our findings would generalize to other patient populations in different settings.Our study highlights some determinants of compliance with some aspects of the diabetic regimen in Egypt.Further studies are needed to explore the impact of social factors, such as family cohesion and support, psychological factors and quality of life.

Conclusion:
Compliance to appointment and drugs is better than compliance to diet.Improper compliance in general is determined by: Older age, literacy, working, type 2 diabetes, long duration of the disease, absence of complication, positive family history, and polypharmacy.Cost and psychological depression are among the main reasons of improper compliance.

Recommendations:
1-The first step to improve compliance is to have some protocol of assessment of noncompliance.
2-Appointment of a behavioral health specialist to help in breaking the cycle of depression and poor compliance through cognitive-behavioural therapy.
3-Appointment of nutritionist at the diabetes clinic to enforce the dietary educational programs.
4-Organize the clinic time to become more suitable to working patients.

Table ( 1
): General characteristics of the participants.
# Mean ± SD. * p † Improper compliance include fair and poor compliance.

Table ( 2
): Pattern of compliance of diabetic patients.

Table ( 3
): Effect of patient characteristics on appointment, diet and drugs compliance.

Table ( 4
): Effect of disease characteristics on appointment, diet and drugs compliance.
(5)† Improper compliance include fair and poor compliance.Table(5): Effect of care characteristics on appointment, diet and drugs compliance.*p † Improper compliance include fair and poor compliance.

Table ( 6
): Significant predictors of improper appointment, diet, drugs and total compliance.