Annals of Tropical Medicine and Public Health

: 2017  |  Volume : 10  |  Issue : 3  |  Page : 707--714

Effect of educational intervention on self-care behaviors among patients with diabetes: An application of PRECEDE model

Seyed Ghadir Hosseini1, Davoud Shojaeizadeh2, Akram Sanagu3, Mohammad Ali Vakili4, Kamal Mirkarimi5, Reza Jahanshahi5,  
1 Department of Health Education and Promotion, School of Public Health, International Campus, Tehran University of Medical Sciences, Tehran, Iran
2 Department of Health Education and Promotion, School of Public Health, Institute of Public Health, Tehran University of Medical Sciences, Tehran, Iran
3 Nursing Research Center, Golestan University of Medical Sciences, Gorgan, Iran
4 Department of Health and Medical Sciences, Golestan University of Medical Sciences, Gorgan, Iran
5 Health Management and Social Development Research Center, Golestan University of Medical Sciences, Gorgan, Iran

Correspondence Address:
Davoud Shojaeizadeh
Department of Health Education and Promotion, School of Public Health, Institute of Public Health, Tehran University of Medical Sciences, Tehran


Introduction: Diabetes is a common noncommunicable disease across the world with a remarkable rate of early death in some countries. This survey aimed to assess the effect of educational intervention on self-care behaviors among patients with diabetes, an application of PRECEDE model. Materials and Methods: A randomized control trial study was conducted on 106 patients with type 2 diabetes who had active records in the diabetes clinics (just two clinics) in the Iranian city of Gorgan. Patients were randomly assigned into control (53 patients) and intervention (53 patients) arms. Data were collected using a three-part questionnaire including (1) a self-care behavior questionnaire based on PRECEDE model, (2) a checklist of demographic and anthropometric characteristic, and (3) a patient sheet to record glycated hemoglobin (HbA1C). To analyze data, Shapiro–Wilk, ANOVA repeated measure, Tukey tests, and linear regression model were applied. P < 0.05 was taken into account as statistically significant. Results: The mean age of patients in the control and intervention groups was 58.09 ± 1.6 and 51.55 ± 8.3 years, respectively. HbA1C and body mass index were more decreased in the intervention group as compared to the control group. At 6 months follow-up, enabling factors, knowledge, and attitude were the strongest predictors of the self-care behaviors. Conclusion: Designing an educational intervention based on PRECEDE model appeared to be likely useful to promote self-care behaviors and control diabetes among patients with type 2 diabetes.

How to cite this article:
Hosseini SG, Shojaeizadeh D, Sanagu A, Vakili MA, Mirkarimi K, Jahanshahi R. Effect of educational intervention on self-care behaviors among patients with diabetes: An application of PRECEDE model.Ann Trop Med Public Health 2017;10:707-714

How to cite this URL:
Hosseini SG, Shojaeizadeh D, Sanagu A, Vakili MA, Mirkarimi K, Jahanshahi R. Effect of educational intervention on self-care behaviors among patients with diabetes: An application of PRECEDE model. Ann Trop Med Public Health [serial online] 2017 [cited 2019 Sep 23 ];10:707-714
Available from:

Full Text


Diabetes mellitus (DM) is known as a metabolic disorder characterized by high blood sugar or glucose levels, causing by worsening insulin resistance.[1] It also is a common noncommunicable disease across the world with a remarkable rate of early death in some countries.[2] According to statistics, the number of people who have diabetes is significantly increasing.[3] Diabetes is now well documented such that from 1990 to 2010, the number of deaths resulted by diabetes doubled from 650,000 to 1.3 million all around the world.[4] It has been hypothesized that the Eastern Mediterranean region will have the prevalent rate of DM after African countries.[2] The Islamic Republic of Iran located in the region and has also the second largest population after Pakistan that predicted to reach a worrisome status of diabetes until 2030.[5] A meta-analysis of studies conducted between 1996 and 2004 reported the total prevalence of DM in the population aged over 40 years as 24%.[2]

Total costs of DM can be categorized as medical and nonmedical costs.[6] DM can cause an increased risk of complications of macrovascular (such as stroke, coronary artery disease, and myocardial infarction) and microvascular conditions (comprising nephropathy, retinopathy, and neuropathy).[7] Self-care is well documented to the successful management of all chronic conditions.[8] Scholars have found that the privileges of diabetes self-care education are as follows: increased knowledge, beneficial self-care behaviors, and improved clinical indicators.[8] In overall, diabetes self-management is a multifactorial and complex process impacted by the several factors such as individual, social, and environmental, and after the diagnosis, it is well known that the major focus is learn to manage the diabetes about self-care behavior.[9]

To achieve an effective educational program, it is essential to apply a proper health education model.[10] The PRECEDE model is one of the most usual programming models in the health education and promotion that successfully confirmed in numerous studies.[11] This model comprised two sections of PRECEDE and PROCEED [12] and was developed by Lawrence Green that is a useful theoretical model for determining needs in health education and promotion. The PRECEDE section is a planning model to achieve health promotion programs and consisted of likely effective factors on behavior including predisposing factors (knowledge, attitudes, and beliefs), reinforcing factors (influential people such as family, peers, and health professionals), and enabling factors (accessibility of resources and skills).[13] Developing an educational program for patients is given less importance compared with other clinical actions.[14]

Due to low-scale quantitative surveys being able to test the usefulness of model-based educational programs for patients with diabetes, the current study sought to identify the effect of educational intervention on self-care behaviors among patients with diabetes who attended to diabetes clinics in Gorgan city (north of Iran) using the PRECEDE model.

 Materials and Methods

Study design and procedure

A randomized control trial study (pre- and post-test with control group) was conducted on 106 patients with type 2 diabetes who had active records in diabetes clinics in the Iranian city of Gorgan. This study was granted by international campus in Tehran University of Medical Science (no. 9223489001). Furthermore, the study protocol is waiting for approval of Iranian Registry of Clinical Trial Code.

Before study, informed consent form was obtained from all participants who were adequately aware of the aim of the study. To select patients, according to field studies, there are just two diabetes clinics in the Gorgan city (governmental); therefore, these clinics were our study population and a list of patients with type 2 diabetes existed in the clinics (records) was considered. Secondary, diabetic patients were randomly assigned in the control (53 patients) and intervention (53 patients) groups. All study variables were assessed at preintervention and 1, 3, and 6 months follow-up [Figure 1].{Figure 1}


Inclusion criteria were as follows: having type 2 diabetes more than 2 years, inclines to participate in the study, age <60 years, and lack of debilitating disease. The exclusion criteria comprised treatment by insulin, disable to exercise (such as walking), gestational diabetes, cognitive disorders that may interfere with completing the questionnaire, unwillingness to participate in the study, and absence more than one session in the intervention group. In total, 106 eligible patients were included in the survey.

Variable definition

At present survey, knowledge, attitude, and behavior categorized into poor (scores <50% of the average), moderate (50%–70% of the average), and good (scores more than 75% of the average). Body mass index (BMI) was also considered normal, overweight, and obese based on the general criteria of the WHO as 18.5–24.9 for normal, 25 ≤ BMI < 30 for overweight, and BMI ≥30 for obesity.[15]

Instruments of the study

Data of the survey were collected using a three-part instrument; (1) self-care questionnaire of diabetic patients based on PRECEDE model, (2) demographic questionnaire, and (3) blood test sheet (glycated hemoglobin [HbA1C]).

Demographic characteristics were also collected using a patient sheet that included age, gender, job, education, marital status, income, BMI, history of diabetes, and place of residence. In final, blood test was carried out to measure the HbA1C.

A valid and reliable questionnaire was used to collect data in terms of self-care behaviors of patients with diabetes that used by Dizaji et al.[3] in a survey entitled effects of educational intervention based on PRECEDE model on self-care behaviors and control in patients with type 2 diabetes in 2012. This questionnaire is consisted of knowledge (eight closed questions), attitude within the form of 14 closed questions (completely agree 5, agree 4, no comment 3, disagree 2, completely disagree 1), behavior within the format of seven closed questions (yes 1, no 0, and type of behavior), enabling factors within the format of 11 closed questions (yes 1, no 0, and type of enabling factors), and in final, reinforcing factors asked by three closed questions (yes 1 and no 0).

All measurements were done in the time of preintervention and 1, 3, and 6 months after intervention. Furthermore, to blind study at the time of sample collection, interviewers and experts in the laboratory were unaware of patient's assignment in the control and intervention groups.

Sample size

According to a related study [16] with an estimation of mean and standard deviation of 9.42 ± 1.91 and 7.81 ± 1.26 of HbA1C in before and after interventions and an assumption of 1 unit decrease in HbA1C, confidence interval 95%, test power 80%, and attrition rate of 20%, a total of 106 individuals (53 patients in each group) were considered.

Intervention group

At present, all content of intervention group was presented by the workshop. In addition, lecture, enquiry, group discussion, and snowballing techniques were also used during the workshop. In the intervention group, sessions were held for 4 weeks (2 h/week) in the diabetes clinic in Gorgan city. Approximately, 25 patients took part in the each session of workshop. In the final of sessions, patients were provided by brochures and necessary guidance. To increase the reinforcing factors and family supports, patients were asked to participate in the sessions with one of their family members. In final, a single session was held for the members of patient's family to avoid ambiguity, as well. Education in this group was delivered by two persons including a general physician and a specialist in health education and promotion.

Control group

Patients in the control group were received the routine education that usually provided in the diabetes clinics. To unify the education of patients in both groups, the research team tried to consider patients who received a maximum education of 7–10 (the mean time of these sessions is usually 20–30 min in the clinics of diabetes). Education in this arm was provided by a general physician.

Data analysis

Descriptive analyses were used to explain the demographic and anthropometric characteristics such as age, gender, job, education, marital status, income, number of children, BMI, history of diabetes, place of living, and history of diabetes in family, as well as the score of the theory constructs. Shapiro–Wilk test was run to test the normal distribution in the preintervention. ANOVA repeated measure was used to observe the trend of changes during the study. To compare the two groups, Tukey test was run. To predict the effects of the theory constructs on self-care behavior, a linear regression model was utilized. P < 0.05 was considered statistically significant.


In total, in the control, 16 (30.2%) and 37 (69.8%), and, in the intervention group, 18 (34%) and 35 (66%) were, respectively, male and female. With regard to the knowledge, in the preintervention, the vast majority of patients had poor awareness 48 (90.6%) and 42 (79.2%) and poor behavior 33 (62.3%) and 29 (54.7%) in the control and intervention groups, respectively; while most of the patients revealed a moderate attitude toward self-care behaviors in the control 31 (58.5%) and intervention 35 (66.03%) arms. The mean age of patients in the control and intervention groups was 58.09 ± 1.6 (ranged 37–60) and 51.55 ± 8.3 (ranged 27–59) years, respectively. Further, the mean BMI was 28.94 ± 1.47 in the control group and 28.20 ± 1.57 in the intervention group. Homemakers 32 (60.4%) and workers 1 (1.9%) in the control group and homemakers 22 (41.5%) and workers 3 (5.7%) in the intervention group had the highest and lowest frequency in terms of job status. Given the marital status, majority of patients in the control and intervention groups were 36 (67.9%) and 31 (58.5%), correspondingly [Table 1].{Table 1}

The vast majority of patients in the control 45 (95.8%) and intervention 38 (86.4%) groups were preferred to provide their drugs from pharmacy. Health workers were the most important source of information to control diabetes preferred by the patients in the control 30 (58.8%) and intervention 24 (45.2%) groups [Table 2].{Table 2}

The main portion of participants in the control 51 (96.2%) and intervention 45 (84.9%) groups were encouraged by health workers when they used hypoglycemic drugs, diet, and did exercise. ANOVA repeated measure showed a meaningful difference between the two groups before and after interventions (P = 0.001) [Table 3].{Table 3}

[Table 4] shows a remarkable difference between the control and intervention groups such that HbA1C was further reduced in the intervention group as compared to the control group (P = 0.001). In the control group, HbA1C was reduced just at 1-month follow-up and back again to the previous value at 3- and 6-month follow-up; while its values remained even after 3 and 6 months follow-up. In terms of BMI, a remarkable difference was observed before and after interventions between the control and intervention groups (P = 0.001). After 6 months, patients in the intervention group lost more weight than the control group.{Table 4}

To predict the self-care behaviors based on the constructs of the theory, linear regression model was employed. Since the constructs of the theory had overlapping effect, we run forward method of linear regression model. As stated by [Table 5], attitude (P = 0.001, β = 0.503), enabling factors (P = 0.040, β =0.377), and knowledge (P = 0.001, β = 0.172) were, respectively, the strongest predictors of self-care behaviors in the preintervention. Moreover, 6 months after intervention, enabling factors (P = 0.001, β = 0.309), knowledge (P = 0.004, β = 0.273), and attitude (P = 0.009, β = 0.239) were also the strongest predictors of the self-care behaviors [Table 5].{Table 5}


At present, the vast majority of patients had poor awareness about diabetes. Diabetes is a disease that its main treatment process must be undertaken by patients with special emphasis on knowledge in terms of nutrition management.[3] Unfortunately, most of the current participants had poor knowledge; therefore, it necessitates researchers to implement interventional programs aimed at increasing the awareness about diabetes. In a study conducted by Katibeh et al.[2] in Iran, a remarkable relationship was found between level of education and diabetes prevalence that indicates the important role of awareness regarding diabetes. Other surveys implemented in Germany,[17] India,[18] and Saudi Arabia [19] emphasize the fact that lower education level and higher prevalence of diabetes are correlated. Since attitude and behavior will usually be increased due to the promotion of knowledge, thus, designing and developing an educational intervention to improve awareness of patients seem likely useful.

In the current exploration, applying educational intervention significantly increased knowledge, attitude, and behavior in the intervention group as compared to the control group that is in agreement with other studies in terms of knowledge,[3],[20],[21] in terms of attitude,[3],[22],[23] and in final, in terms of behavior.[24],[25]

In the current investigation, most of patients received their information about self-care behaviors from health personnel and television, respectively. A study conducted by Sharifirad et al.[11] on elderly presented mass media as the most preferred source to receive necessary information that is no consistent with our study. This may be likely caused by the fact that population study was different in two studies because patients with diabetes usually receive their information from qualified individuals at the care times, while in the Sharifirad study, stress was assessed which is a less specialized subject compared to our population survey. In agreement with our study, Dehdari et al.[26] and Lesan et al.[27] found that participants received their information from a trainer.

Scores of reinforcing factors increased significantly in the intervention arm than the control arm which is concordant with Dizaji et al.[3] In Sharifirad et al.[28] study conducted in Iranian city of Talesh, PRECEDE constructs (predisposing, enabling, and reinforcing factors) had a positive effect on the preventive behaviors of iron deficiency anemia. It should come as no surprise that score of reinforcing factors was significantly increased after intervention because we trained a member of family of patients to promote reinforcing factors status.

At the recent study, HbA1c and BMI were remarkably different between the control and intervention groups after intervention that is similar to the study conducted by Aghamolaei et al.[29] that showed a significant decrease in the mean of HbA1C, and also an increase of the personal control of weight control, blood sugar, and considering diet. These similar findings report valuable results because of the same population study. In Baghianimoghadam et al.'s study,[30] HbA1c was remarkably decreased in the intervention arm than the control arm that is in accordance with the present survey. In addition, a more decrease happens in Baghianimoghadam et al.'s survey compared to our survey that may be caused by the fact that they implemented a longer time education (5 months) than our survey (1 month). Thus, long-term educational intervention seems beneficial to reach appropriate results.

In a theory-guided study implemented by Mirkarimi et al.,[15] patients in the motivational interviewing group lost more weight and BMI and also a long-term loss weight observed compared to the control group. These similarities may cause by the fact that both studies used the interactive approaches such as group discussion and inquiry; therefore, use of aforementioned approaches seems useful to modify self-care behavior, significantly. In contrast, educational intervention executed by Dizaji et al.[3] was not effective to decrease BMI in the diabetic patients that is not chimes in with the present survey, which may be due to the time of interventions and different types of education such that we educated a member of patients family to increase enabling factors (family support and persuading patients to consume drugs regularly), and we also provided education in a long time period compared to the Dizaji et al.'s study.

The predictive power of the PRECEDE model in terms of self-care behaviors was explored by entering all the constructs in the linear regression model through the forward regression model. Next, three constructs of the PRECEDE model including enabling factors (β = 0.309), knowledge (β = 0.273), and attitude (β = 0.239) predicted self-care behaviors of patients with diabetes that presented from strongest to weakest, respectively. The strong predictive power of enabling factors may be explained that most of patients poorly acted in the field of self-care and consequently presented enabling factors such as equipment and facilities of exercise (gym and sport shoes) as most strong predictor.


Diabetes is known as a prevalent disease in the Iranian city of Gorgan that imposes a huge burden on health care system. Therefore, designing and developing a theory-guided intervention appeared to be likely beneficial to control diabetes among patients.


We wish to thank all patients who participated in the survey.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Okwechime IO, Roberson S, Odoi A. Prevalence and predictors of pre-diabetes and diabetes among adults 18 years or older in Florida: A multinomial logistic modeling approach. PLoS One 2015;10:e0145781.
2Katibeh M, Hosseini S, Soleimanizad R, Manaviat MR, Kheiri B, Khabazkhoob M, et al. Prevalence and risk factors of diabetes mellitus in a central district in Islamic Republic of Iran: A population-based study on adults aged 40-80 years. East Mediterr Health J 2015;21:412-9.
3Dizaji MB, Taghdisi MH, Solhi M, Hoseini SM, Shafieyan Z, Qorbani M, et al. Effects of educational intervention based on PRECEDE model on self care behaviors and control in patients with type 2 diabetes in 2012. J Diabetes Metab Disord 2014;13:72.
4El Bcheraoui C, Basulaiman M, Tuffaha M, Daoud F, Robinson M, Jaber S, et al. Status of the diabetes epidemic in the Kingdom of Saudi Arabia, 2013. Int J Public Health 2014;59:1011-21.
5Guariguata L, Whiting DR, Hambleton I, Beagley J, Linnenkamp U, Shaw JE. Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract 2014;103:137-49.
6Lesniowska J, Schubert A, Wojna M, Skrzekowska-Baran I, Fedyna M. Costs of diabetes and its complications in Poland. Eur J Health Econ 2014;15:653-60.
7Alwakeel JS, Sulimani R, Al-Asaad H, Al-Harbi A, Tarif N, Al-Suwaida A, et al. Diabetes complications in 1952 type 2 diabetes mellitus patients managed in a single institution in Saudi Arabia. Ann Saudi Med 2008;28:260-6.
8Gumbs JM. Relationship between diabetes self-management education and self-care behaviors among African American women with type 2 diabetes. J Cult Divers 2012;19:18-22.
9Cosma A, Baban A. Beyond self-care behaviors in diabetes: Learning from adolescents. Cognitie, Creier, Comportament/Cognition, Brain, Behavior 2015:19:233-49.
10Hosseini N, Amini A, Alamdari A, Shams M, Vossoughi M, Bazarganipour F. Application of the PRECEDE model to improve sexual function among women with hysterectomy. Int J Gynaecol Obstet 2016;132:229-33.
11Sharifirad G, Ghaffari M, Zanjani S, Hassanzadeh A. The effectiveness of educational intervention based on PRECEDE model on the level of stress among the elderly at elderly clubs. J Educ Health Promot 2013;2:3.
12Crosby R, Noar SM. What is a planning model? An introduction to PRECEDE-PROCEED. J Public Health Dent 2011;71 Suppl 1:S7-15.
13Phillips JL, Rolley JX, Davidson PM. Developing targeted health service interventions using the PRECEDE-PROCEED model: Two Australian case studies. Nurs Res Pract 2012;2012:279431.
14Deyirmenjian M, Karam N, Salameh P. Preoperative patient education for open-heart patients: A source of anxiety? Patient Educ Couns 2006;62:111-7.
15Mirkarimi K, Mostafavi F, Eshghinia S, Vakili MA, Ozouni-Davaji RB, Aryaie M. Effect of motivational interviewing on a weight loss program based on the protection motivation theory. Iran Red Crescent Med J 2015;17:e23492.
16Didarlou A, Eftekhar Ardabili H, Niknami SH, Hajizadeh E, Alizadeh M. Effective factors of self-care behaviors among women with type 2 diabetes based on the developed theory of reasoned action in Khoy city. J Sch Public Health Inst Public Health Res 2011;9:79-92.
17Heidemann C, Kroll L, Icks A, Lampert T, Scheidt-Nave C. Prevalence of known diabetes in German adults aged 25-69 years: Results from national health surveys over 15 years. Diabet Med 2009;26:655-8.
18Ajay VS, Prabhakaran D, Jeemon P, Thankappan KR, Mohan V, Ramakrishnan L, et al. Prevalence and determinants of diabetes mellitus in the Indian industrial population. Diabet Med 2008;25:1187-94.
19Al-Baghli NA, Al-Ghamdi AJ, Al-Turki KA, Al Elq AH, El-Zubaier AG, Bahnassy A. Prevalence of diabetes mellitus and impaired fasting glucose levels in the Eastern Province of Saudi Arabia: Results of a screening campaign. Singapore Med J 2010;51:923-30.
20Milenkoviæ T, Gavriloviæ S, Percan V, Petrovski G. Influence of diabetic education on patient well-being and metabolic control. Diabetol Croat 2004;33:3.
21Hazavehei S, Sharifirad G, Mohabi S. The effect of educational program based on health belief model on diabetic foot care. Int J Diabetes Dev Ctries 2007;27:1-10.
22Rezai N, Tahbaz F, Kimiagar SM, Alavi MH. The effect of nutrition education on knowledge, attitude and practice of type 1 diabetic patients from Aligoodarz 2006;10:52-9.
23Shabbidar S, Fathi B. Effects of nutrition education on knowledge and attitudes of type 2 diabetic patients. J Birjand Univ Med Sci 2007;14:9-15.
24Sharifirad GR. Effect of dietary regulation on fasting blood glucose and body mass index of patients with type 2 diabetes. Ardabil Univ Med Sci 2006;7:375-80.
25Larijani B, Fakhrzadeh H. A comparative study of the impact of education through holding group meetings and educational booklets on the rate of knowledge and practice among the patients referring to lipid clinic. J Diabetes Lipid Iran 2003;3:89-98.
26Dehdari T, Heidarnia A, Ramezankhani A, Sadeghian S, Ghofranipour F, Etemad S. Planning and evaluation of an educational intervention programme to improve life quality in patients after coronary artery bypass graft-surgery according to PRECEDE-PROCEED model. J Birjand Univ Med Sci 2008;15:27-37.
27Lesan S, GhofraniPour F, Birashk B, Faghihzadeh S. Application of PRECEDE in reducing Tehranian firemen anxiety. Iran J Psychiatry Clin Psychol 2003;9:77-83.
28Sharifirad G, Golshiri P, Shahnazi H, Shakouri S, Hassanzadeh A. PRECEDE educational model for controlling iron-deficiency anaemia in Talesh, Iran. J Pak Med Assoc 2011;61:862-5.
29Aghamolaei T, Eftekhar H, Mohammad K, Nakhjavani M, Shojaeizadeh D, Ghofranipour F, et al. Effects of a health education program on behavior, HbA1c and health-related quality of life in diabetic patients. Acta Med Iran 2005;43:89-94.
30Baghianimoghadam M. The study of BASNEF model effectiveness in controlling Yazd diabetic patients. Tehran: Tarbiatmodares University; 2001.