A total of 355 records will be assessed from the BNN. Furthermore, 381 records will be assessed from penitentiary institutions or lembaga pemasyarakatan (LPs) and community correction centers or balai pemasyarakatan (BPs). For the quantitative analysis, we used secondary data from the Indonesian government’s drug rehabilitation records. These resources originate from the BNN, which serves as the central agency for narcotics regulation; LPs, which provide drug rehabilitation services to prisoners who have been criminalized for drug-related offenses; and BPs, which offer rehabilitation for individuals preparing to reintegrate into their communities.

The database for the years 2022 to 2024 contains detailed information on individuals involved in drug misuse, the services they received, and their health status as recorded in medical records. The medical records will focus on physical health assessments, treatment interventions, and instances of relapse.

Participant eligibility criteria are presented in Textbox 1.

The sampling technique used in this study is total sampling, whereby all data that meet the inclusion and exclusion criteria will be included in the analysis.

For the qualitative data collection, participants will be selected using purposive sampling techniques. This selection will focus on individuals who have expertise or experience in dealing with drug misuse–related issues. Participants will include stakeholders such as health workers (doctors and nurses) and rehabilitation officers. Their involvement will be determined based on recommendations from the rehabilitation leader of the Provincial National Narcotics Board, ensuring that the participants have relevant knowledge and insights related to the study’s focus. The interviews will take place either in person or via videoconferencing and will last approximately 60 to 90 minutes. With participants’ consent, each interview will be audio-recorded and then transcribed for detailed analysis.

The first phase aims to explore several key aspects of drug misuse among middle-aged and older adult populations. Specifically, it seeks to identify the physical health profiles of individuals with substance use disorders within these age groups. On the basis of this objective, the study further examines participants’ perceptions of their own health status.

To understand the influence of various factors on self-health perception, we apply the health belief model. The modifying factors include demographic variables such as age, gender, education, occupation, ethnicity, marital status, living arrangements, and religion. The perceived susceptibility factors include being a drug dealer, types of drugs used, routine drug misuse, excessive drug misuse, and the impact of these behaviors on occupational functioning and interpersonal relationships. Examining these variables enables the identification of behavioral patterns and their impact on self-assessment of health. The perceived severity variables include length of conviction, recidivism status, and excessive drug misuse.

The second phase aims to examine risk and protective factors related to physical and biological health that may influence drug misuse relapse. Prediction of relapse will be assessed through readiness to change for discontinuing drug misuse. The readiness to change is determined by the application of the transtheoretical model, which explains the principle of change, specifically behavioral change [18,19]. This variable is measured by the University of Rhode Island Change Assessment Scale [20]. The University of Rhode Island Change Assessment Scale has been used to assess people who participated in the rehabilitation program at the BNN [21]. From the total score, the variable will be categorized to precontemplation (score <8), contemplation (8-11), action (11-14), and maintenance (>14).

The independent variables include the demographics of respondents, such as gender, age, education background, marital status, employment status, and religion. Other variables include drug choice; the Addiction Severity Index, Fifth Edition [22]; risk for substance use as measured by the Alcohol, Smoking, and Substance Involvement Screening Test [23]; and rehabilitation monitoring, which encompasses health status, psychological status, social status, and environmental status.

Finally, the study aims to develop a health model grounded in the health belief model to effectively prevent relapse among these vulnerable populations.

Descriptive statistics will be used to summarize demographic characteristics, substance use history, patterns of drug misuse, and health perceptions among individuals with substance use disorders, and bivariate analyses will be conducted to examine associations between independent and dependent variables. Furthermore, inferential statistical analyses, specifically logistic regression, will be performed to identify predictors of health outcomes and drug misuse relapse while controlling for demographic variables. A P value <.05 will be considered the threshold for statistical significance throughout the analysis.

Thematic analysis will be conducted to identify and analyze patterns within the qualitative data gathered from interviews. Transcripts will be coded inductively to develop themes that reflect participants’ experiences and insights. NVivo software (Lumivero) will be used to assist in the organization and analysis of qualitative data.