The type of research used is the descriptive method. The research was conducted by collecting medical record data of diabetes mellitus patients at five community health centers in Makassar City, Indonesia, including Paccerakang (1), Andalas (2), Cenderawasih (3), Tamalanrea (4), and Jumpandang Baru (5) Community Health Centers.

The population in this study was all diabetes mellitus patients in 2023 in Makassar City, Indonesia. The sample in this study was diabetes mellitus (DM) patients in five community health centers in 2023. The sample was calculated using the Slovin formula (Alqallaf et al. 2022; Boekoesoe et al. 2023) and obtained a total sample size of 299 samples.

This study used a stratified random sampling method to select participants from various Puskesmas (community health centers) in Makassar City, South Sulawesi Province. This method was chosen to ensure that the sample includes a representative population of all relevant groups in this study, based on certain categories such as age, gender, socioeconomic status, and location of the Puskesmas. With this approach, it is expected that a more varied sample can be obtained and reflect the real conditions in the community so that the research results are more valid and can be generalized to a wider population.

After the health center was selected, participants were selected based on inclusion and exclusion criteria. Inclusion criteria included patients who had been diagnosed with type 2 diabetes mellitus (DM), were receiving regular diabetes medication therapy, and were willing to participate by providing informed consent. Meanwhile, the exclusion criteria included patients suffering from serious comorbidities that affect diabetes management, such as end-stage renal disease, and patients who were unable to communicate or understand verbal or written instructions. Participants were randomly selected from the list of patients who met the inclusion criteria at each selected health center. To ensure representativeness, selection was made based on relevant demographic categories, such as age, gender, and socioeconomic status, so that the sample reflected the diversity of the diabetes patient population in Makassar City.

Several steps were taken to minimize potential bias in participant selection and data collection. Selection bias was reduced using stratified random sampling, which ensured that each subgroup in the diabetes patient population was proportionally represented in the sample. Measurement bias was reduced by engaging trained researchers who used validated instruments to collect data on medication habits and adherence to therapy, as well as conducting structured interviews and checking the accuracy of medical records. To reduce respondent bias, data on medication adherence and drug use were verified through medical records and interviews with health workers. Finally, to reduce non-response bias, a reminder strategy was implemented to increase participation in the survey by ensuring a sufficiently large number of participants from each health center. With these measures, it is expected to obtain a representative sample and research results that better reflect the real conditions in diabetes management in the community.

This descriptive research focuses on examining the relationship between the use of diabetes mellitus (DM) medication and its alignment with the DM therapy algorithm among patients in Makassar. The independent variable in this study is the use of DM medication, where data will be collected regarding the name of the drug and the dosage prescribed to each patient. Indicators for this variable include the specific medications prescribed for managing diabetes mellitus, the frequency, and the correct dosage as per medical prescriptions. The intervening variable is the DM therapy algorithm, which serves as the standard reference for evaluating the prescribed medication’s appropriateness.

The dependent variable is the appropriateness of DM medication use based on the DM therapy algorithm, categorized as either “appropriate” or “inappropriate.” This evaluation will be conducted by comparing the patient’s medication regimen to the established algorithm. The appropriateness is assessed based on whether the prescribed drug and dosage adhere to the recommended guidelines. Additionally, the study will take into account the socioeconomic conditions of the DM patients in Makassar, which may influence the selection and adherence to prescribed therapies.

In this research, data is analyzed descriptively by collecting medical records of diabetes mellitus patients at the community health center in Makassar City. The analysis process includes the following steps:

1. Suitability categories are presented in percentage (%) form (Cojic et al. 2021; Gallardo-Gómez et al. 2024). The research data is compiled into data collection tables, followed by data processing and percentage calculation using the following formula: $\%=\frac{\text{Frequency}}{\text{Total}\text{Count}}\times 100$
2. Frequency refers to variables such as gender, age, medication use, therapy patterns, drug usage compliance based on the algorithm, drug accuracy, and dosage accuracy (Religioni and Pakulska 2020).

The data will be presented in tables that summarize the frequency and percentage of each category, allowing for a clear interpretation of patient characteristics and the appropriateness of diabetes management in relation to the therapeutic algorithm.

Based on the socio-economic data of diabetes mellitus patients (Table 1), it can be seen that the majority of patients are women with a percentage of 63.88%, while men are only 36.12% of the total 299 patients. This shows that the prevalence of female patients is higher than male. In terms of age, most patients are in the age group of 45–65 years with a percentage of 57.90%, while patients under 45 years old are only 3.68%, and patients over 65 years old reach 36.45%. This condition confirms that diabetes mellitus is more commonly found in the middle-aged to elderly age group.

The education level of patients also varied, where the majority of patients had a high school education level (equivalent to high school) at 42.48%. Patients with junior high school and elementary school education accounted for 27.76% and 11.04%, respectively, while patients with D3/D4/S1 education amounted to 14.05%. This level of education may affect patients’ knowledge about their disease and treatment. In terms of occupation, the majority of patients are housewives (IRT) with a percentage of 33.44%, followed by self-employed as much as 24.09%, and patients who do not have a job reach 21.07%. Other groups, such as laborers, civil servants, and retired civil servants, have smaller numbers.

From this picture, it can be concluded that diabetes mellitus patients at the community health center mostly consist of women aged 45–65 years, with a medium level of education (high school), and the majority work as housewives. These factors provide important insights into the population most vulnerable to the disease, as well as the potential challenges faced in terms of access and understanding of disease management.

Based on data on the distribution of the use of diabetes mellitus drugs at the community health center (Table 2, Fig. 1), it can be seen that drugs from the Biguanide group, namely Metformin, are the most widely used drugs, with a total of 231 patients or around 60.31% of all patients. Metformin is known as a first-line drug that is often prescribed for the management of type 2 diabetes mellitus. Sulfonylurea group drugs, such as glimepiride and gliclazide, were also used quite frequently, with 109 patients, or 28.46%, and two patients, or 0.52%, respectively, with a total of 111, or 30%, of patients. This drug serves to increase insulin secretion in diabetics.

Figure 1. 

Graph of the distribution of the diabetes mellitus drug’s use.

The use of rapid-acting insulin, such as insulin aspart, was recorded in nine patients, or about 2.35%. This type of insulin is usually used in patients who require rapid blood sugar control. In addition, the drug Acarbose from the Glucosidase Inhibitor class was used by seven patients, or 1.83%, which serves to slow the absorption of carbohydrates and reduce glucose levels after meals.

Long-acting insulin drugs, such as insulin glargine and insulin detemir, were used by 15 patients, or 3.92%, and four patients, or 1.04%, respectively, of the total population, which helps control blood sugar levels stably throughout the day. Meanwhile, premix insulin was used by six patients, or about 1.57%. Premixed insulin is usually used to control blood glucose levels over a long period of time while helping to control blood sugar spikes after meals.

From the data, it can be concluded that Metformin and Glimepiride dominate the management of diabetes mellitus at community health center, while the use of insulin, both rapid-acting and long-acting, is only given to a small proportion of patients, who most likely have more complex diabetic conditions.

Evaluation of the use of antidiabetic drugs is carried out to determine the suitability of antidiabetic drugs given with Perkeni 2021 standards. For the treatment of complications, no evaluation was carried out because the evaluation of conformity with the Perkeni 2021 standard was only aimed at evaluating oral antidiabetics OHO and insulin. In this study, an evaluation of the rationality of antidiabetics was carried out on the right indication, the right drug selection, the right dose, and the right patient.

Based on the evaluation of the use of antidiabetic drugs in patients at five health centers (Fig. 2, Table 3), the results showed a fairly good level of conformity in several aspects of medication rationality. First, in the aspect of appropriate indication. Appropriate indication in the treatment of diabetes mellitus is the accuracy in using antidiabetic drugs according to the doctor’s diagnosis in the medical record file based on the examination of HbA1C levels ≥ 6.5% for diabetes and 5.7%–6.4% for pre-diabetes (McCoy et al. 2020; Cojic et al. 2021). Based on the results of the analyzed medical record files, the number of patients diagnosed with diabetes mellitus was as many as 87.29% of patients, and they received antidiabetic drugs in accordance with their clinical indications. This shows that most patients have received therapy that is appropriate for their medical condition, although there are still 12.71% of patients whose treatment is not in accordance with the expected indication. This discrepancy may be due to a lack of thorough assessment of clinical conditions or limited therapeutic options at health facilities.

Figure 2. 

Graph of evaluation of rational use of DM drugs totaled from the studied community health centers based on the parameters of appropriate indication, appropriate medicine, appropriate dose, and appropriate patient.

Secondly, for appropriate drug selection, 86.96% of patients received the appropriate type of drug based on the recommended therapy algorithm. This percentage indicates that in terms of drug selection, the majority of patients received the recommended drugs, although there were 13.04% of patients who did not comply with the correct drug choice. Factors affecting this discrepancy could be related to the availability of drugs in health facilities or the presence of certain clinical preferences that are not in accordance with therapeutic guidelines.

Third, in the aspect of correct dosage, the evaluation results showed that only 69.90% of patients received the correct dosage of drugs. This means that around 30.10% of patients did not receive the appropriate dose, which could have an impact on the effectiveness of their glycemic control. This inappropriate dosage needs to be a major concern, as correct dosage adjustment is crucial in optimizing antidiabetic therapy to avoid complications due to hyperglycemia or hypoglycemia.

Lastly, in terms of patient appropriateness, 100% of patients received medication that was appropriate for their individual profile, such as age, gender, and other comorbid conditions. This shows that the administration of medication has taken into consideration the characteristics of each patient, which is an important component in personalized and comprehensive therapy.

The results of this study indicate that there is inaccuracy in dosing in diabetics in several health centers in Makassar City, which is influenced by factors related to health workers, service systems, and patients themselves. The main factor of dose inaccuracy is the limited knowledge and skills of health workers, which can be overcome by continuing education and the use of digital tools to calculate the dose precisely. Limited consultation time also affects the accuracy of dosing, so increased consultation time and additional health workers are needed.

The findings of our study showed a high level of concordance (86%) in the prescribing of diabetes medications at puskesmas in Makassar, in accordance with therapeutic algorithms. This is in line with previous studies that emphasize the importance of adherence to evidence-based guidelines in improving treatment outcomes of DM patients (Williams et al. 2014). However, the issue of dose discrepancy found in 30.10% of cases underscores a critical gap that needs to be addressed. In line with research in Said et al. 2021, which discusses the role of pharmacists in improving patient safety through better communication of medication guidelines, involvement of health workers, especially pharmacists, in dose optimization and patient counselling.

Patients’ lack of understanding of the importance of correct dosage can be addressed with intensive education programs and ongoing counseling (Gudeta and Mechal 2019). In addition, irrational drug use can be minimized by ensuring the availability of drugs in accordance with therapeutic guidelines and conducting regular drug audits. Weak surveillance systems can also be improved by strengthening surveillance and monitoring technology (Gudeta and Mechal 2019).

Social and economic factors also affect patient adherence to therapy, so drug assistance programs and social support should be provided to help patients who have difficulties in purchasing drugs or following therapy (Elwenspoek et al. 2022; Pemberton et al. 2025). The implementation of these interventions is expected to improve dose accuracy and adherence to therapeutic guidelines so that diabetes management at Puskesmas can be optimized (Puspitasari et al. 2022).