Skip to main content

RETRACTED ARTICLE: Development of attributes and attribute levels for a discrete choice experiment on patients’ and providers’ choice for antiretroviral therapy service in Northwest Ethiopia

This article was retracted on 24 August 2023

This article has been updated



Discrete choice experiments (DCEs) are used to assess the strength of preferences and value of interventions. However, researchers using this approach have been criticized for not conducting or publishing rigorous studies to select the required attributes and levels. Proper specification of attributes and their levels determines the validity of DCE. Hence, our study aimed to identify and define attributes and levels for the design of a DCE to elicit patients’ and providers’ preferences for ART service in Northwest Ethiopia.


Four stages were followed to derive the final list of attributes and levels: (1) a literature review to derive conceptual attributes; (2) key informant interviews of 17 providers and in-depth interviews of 15 adult stable patients to identify context-specific attributes and attribute levels; (3) ranking survey among 31 HIV/AIDS program implementers and rating survey among 35 adult stable patients and 42 health workers providing antiretroviral therapy (ART) service to indicate participants’ preference of attributes; and (4) an expert opinion to reduce the list of attributes and levels.


First, a literature review identified 23 candidate attributes. Second, individual-level analysis of the qualitative transcripts confirmed 15 of these 23 attributes. Third, the ranking and rating surveys put the importance of the 23 ART service attributes in order of preference. Fourth, through discussions with eight experts, 17 attributes were discarded based on multiple criteria. The six retained attributes were: the location of ART refills, the frequency of receiving ART refills, the person providing ART refills, the participants/others seen at the same ART refill visit, medication refill pick-up/delivery times, and the total cost of the visit during antiretroviral (ARV) medication refill. Finally, levels were assigned to these 6 attributes based on data from the literature, transcripts, and knowledge of the Ethiopian context.


This detailed description illuminates the attribute development process and provides the reader with a basis for evaluating the rigor of this phase of DCE construction. This paper contributes empirical evidence to the limited methodological literature on attributes and levels of development for DCE, thereby providing further empirical guidance on ART service preference, specifically among patients of low- and middle-income countries.


The use of discrete choice experiments (DCEs) in health services research is gaining popularity, primarily to assess patient stated preferences and willingness to pay for various models of healthcare service delivery [1,2,3,4,5]. A DCE is a quantitative approach used to determine preferences for the characteristics (attributes) of goods or services [6]. In a DCE, respondents make choices between hypothetical services (e.g., antiretroviral therapy) described by a set of attributes (e.g., location, provider, frequency of visits). Each attribute can have a number of defined dimensions called “attribute levels”. By varying the levels of each question, we can analyze the trade-offs respondents make between them [7].

Typically, experimental designs are used to combine the characteristics of the interventions and their assigned levels to construct a set of hypothetical choice alternatives [8, 9]. Following this, respondents are shown a sequence of two or more of these competing choice possibilities and asked to select their preferred option [2, 10]. The attribute levels establish the utility respondents will attach to a specific intervention trait, and as a result, their preferences or choices [10].

Discrete Choice Experiments have been applied within low- and middle-income countries (LMICs), particularly in sub-Saharan Africa (SSA) to elicit preferences of patients toward differentiated antiretroviral therapy [11,12,13,14,15]. Differentiated care or differentiated service delivery is “a client-centered approach that simplifies and adapts HIV services across the cascade to reflect the preferences and expectations of various groups of people living with HIV (PLHIV) while reducing unnecessary burdens on the health system” [16]. Well-known models of differentiated care have focused on antiretroviral therapy (ART) delivery to clients who are clinically stable [17] and have largely been implemented in high-prevalence countries in sub-Saharan Africa. Differentiated ART interventions are conducted both within health facilities and the community and rely on formally trained healthcare workers (HCWs), peers, and community health workers (CHWs) [18].

A DCE has four main stages: identifying and defining attributes and levels, the experimental design, the data collection survey, and the analysis and interpretation of results [8, 19].

As an attribute-based experiment, the validity of a DCE largely depends on the researchers’ ability to appropriately specify attributes and their levels [20]. Misspecification of the attributes and attribute levels has great negative implications for the design and implementation of DCEs and the risk of producing erroneous DCE results, which can misinform policy implementation.

To reduce the likelihood of researcher bias, attribute development has to be rigorous, systematic, and transparently reported [21]. In this regard, the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Good Research Practices for Conjoint Analysis Task Force’s checklist for conjoint analysis applications in health care asks three inter-related questions: (1) was attribute identification supported by evidence (literature reviews, focus groups, or other scientific methods)?; (2) was attribute selection justified and consistent with theory?; and (3) was the level selection for each attribute justified by the evidence and consistent with the study perspective and hypothesis? [1]. Furthermore, Hollin et al. developed guidelines for a conceptual overview of reporting formative qualitative research for the design of preference study protocols and corresponding instruments [22].

Various methods have been applied to the development of DCE attributes. These include literature reviews, existing conceptual and policy-relevant outcome measures, theoretical arguments, expert opinion review, professional recommendations, patient surveys, nominal group ranking techniques, and qualitative research methods [10, 21, 23, 24]. Despite the need to accurately describe the process used in deriving attributes and levels, there is a paucity of such descriptions in the existing literature, in both high and low-income countries including Ethiopia.

Hence, our study aimed to identify and define attributes and levels for the design of a DCE to elicit patients’ and providers’ preferences for ART service in Northwest Ethiopia.


Conceptual framework for developing attributes and attributes levels

There is a growing consensus in the literature that credible attributes and attribute-levels for a DCE must be policy-relevant, important to the study population, and consistent with the random utility theoretical foundation of DCE [10, 21, 25]. Policy-relevant attributes and attributes-levels are those that adequately reflect the essential dimensions or characteristics of the product or intervention that will be evaluated by potential beneficiaries in the DCE [26]. This implies that the identification of such attributes and levels should be guided by appropriate conceptual and theoretical explanatory models and empirical literature on the policy issue. A rigorous literature review on the policy topic can, therefore, lead to the identification of a comprehensive list of conceptual attributes, which can potentially, but not necessarily, be included in a relevant DCE.

According to Coast et al. [21], identifying attributes and their levels exclusively based on a literature review may be easier to implement, but may also lead to the non-inclusion of some important attributes. To be included in the DCE, the conceptual attributes must be considered important by the target population, whose preferences will be elicited in the final DCE, and reflect the needs of their local context. This requires a rigorous qualitative study within the local context [21, 23]. The attributes and levels derived from such a qualitative study are considered demand-driven [10], reflective of local perspectives, understandable to respondents and thereby, plausible within the study context [21]. Deriving attributes from a qualitative study can, therefore, improve the content validity of a DCE study [20]. A qualitative study is also capable of picking up other context-specific and policy-relevant attributes which might not exist in the literature, and hence, can potentially reduce the risk of omitting relevant attributes and attribute levels.

Lastly, the context specific attributes and attribute-levels must be framed in a manner that allows for efficient elicitation and analysis of preferences, according to random utility theory, which is the theoretical foundation of DCE [26]. In this case, DCE attributes (and most particularly levels) must be exhaustive and measurable [10]. The attributes and their levels must be unambiguously framed [27] and appear both cognitively (perceptually) and statistically uncorrelated in the choice sets [6]. Additionally, attributes must be experimentally manipulatable [6], and defined in a manner that gives room for trading between attribute-level alternatives [21]. To ensure these, expert opinion and additional pilot studies within the study area are also recommended [20, 21]. If the number of possible attributes exceeds what one may find possible to pilot in a DCE analysis, it may prove beneficial to use other types of rating and/or ranking exercises (often referred to as compositional approaches) to assess the importance of attributes and to facilitate the construction of the final list of attributes to be included [1].

We adopted the Helter and Boehler framework [28] to rigorously conduct and report the process of attribute development and level selection for a DCE to elicit the preferences of healthcare providers and PLHIV for the attributes of ART service. The process included raw data collection, data reduction, removal of inappropriate attributes, and wording of attributes. We initially identified policy-relevant conceptual attributes from a literature review. We used these conceptual attributes and potential attribute levels as a basis for designing a qualitative study to identify context-specific attributes, as those deemed directly by respondents to be most important. To scale down the context-specific attributes to a number manageable within a DCE, we have undertaken ranking and rating surveys. To ensure that the final attributes and levels conformed to the theoretical postulations of a DCE, we elicited expert opinion. Figure 1 provides an overview of the study’s methodological steps.

Fig. 1
figure 1

Overview of the development of attributes and levels

Initial literature review

In line with recent methodological recommendations [1, 7, 20, 21], the attribute development process began with systematic and scoping reviews aimed at identifying conceptual attributes relevant to antiretroviral therapy in the available literature. Our systematic review included studies of qualitative, quantitative, and mixed methods. PubMed, Web of Science, Embase, and Cumulative Index to Nursing and Allied Health Literature (CINAHL) databases, and Google and Google Scholar search engines were searched. The detailed search strategy, data extraction, and results are available elsewhere [29]. Similarly, our scoping review used PubMed, Web of Science Core Collection, Embase, Scopus, CINAHL, and Global Health databases together with Google and Google Scholar search engines. The detailed search strategy, data extraction, and results are available elsewhere [30].

Guided by these insights from the literature, all authors derived a comprehensive list of conceptual attributes and potential attribute levels. Then, the conceptual attributes and their potential levels were used to guide the design of data collection tools for the qualitative study.

Identification of context-specific attributes through the qualitative study

Study population and sampling

Qualitative data for the development of context-specific attributes and attribute levels were collected from July 2021 to September 2021. In-depth interviews with 15 adults (≥ 18 years old) on ART who were virally suppressed and eligible for differentiated service delivery on the date of data collection [2] and key informant interviews with 17 ART providers recruited from 15 high-load public health facilities (hospitals and health centers) implementing DSD models (multi-month dispensing, appointment spacing model, fast track antiretroviral refill, per led community antiretroviral distribution and health extension professional led community antiretroviral distribution) were conducted. Providers were eligible for key informant interviews if they were physicians, nurses, health officers, and pharmacy professionals directly prescribing and/or dispensing ARVs at a study site and had worked in the ART clinic for at least 6 months and/or as focal person of ART clinic during the study implementation. Purposive sampling was used to select both patients and health workers, and the overall sample size was determined by the expected saturation point [31]. Interviews were guided by semistructured questions and a list of fixed topics that had to be addressed. Thematic analysis was employed in both patient and provider-based interviews. A detailed description is available elsewhere [32, 33].

Ranking and rating surveys

A ranking survey was undertaken among HIV/AIDS program implementers to evaluate the importance (from most (1) to least (23)) of attributes which affect the choices for ART service. The participants were approached purposively via email communication after preparing a list of potential implementers in Ethiopia. The questionnaires were sent through their email address including what the survey is all about, its purpose and how to do the ranking activity. A phone call and further email communication were done in case of clarity issues. About 30 respondents planned to be engaged in this study.

The attributes have a mix of the structure, process, and outcome domains of ART service delivery. The structure attributes include the source of information on ART, location of ART service, the person providing ART refills, total cost of the visit, time spent seeing a healthcare provider, time spent at clinics in ART medication refill, time of health facility operation, frequency of receiving ART refills, ARV drug pill burden, availability of ARV drug dosage forms, ART packaging and self-repackaging, convenience (with social life, food requirement, time in taking drugs, and child care activity), distance from a residential area to the ART service clinic, ART room labeling, buddy system(others take drugs to home in place of a client), the spatial arrangement of the ART room and novel ART delivery methods. The provider’s attitude, preferences on involvement in treatment decision-making and appointment in ART refill (individualized or group) were the process attributes. The outcome attributes include ARV drug efficacy, drug-drug interaction and ARV drug side effects.

All respondents were asked to rank attributes in order of importance to their choice without duplication. To control for ordering bias, we randomized all participants to one of five questionnaire versions that differed in terms of the ordering of the attributes participants viewed.

Moreover, we conducted a rating survey among patients and health workers providing ART service in ART clinics of Awi, East and West Gojjam zones, and Bahir Dar city administration. Respondents were asked to rate from 1 (Attribute is not at all important) to 7 (Attribute is highly important) the attributes which affects the choices for ART service. A consecutive sampling was employed to recruit both patient and provider participants. Initially, we planned to include at least 30 respondents for each category of participant.

In both ranking and rating surveys, participants were provided with a list of 23 attributes identified in both the literature review and qualitative studies. Written informed consent was obtained from the participants prior to the actual interview. The data were collected from July 1, 2022 to July 25, 2022. In both ranking and rating surveys, an Excel spreadsheet was used to calculate the respective overall mean scores for each attribute.

Expert consultation

Expert opinion is a recommended method when one needs to reduce the number of attributes and levels [1]. Too many attributes in a DCE increase complexity of the tasks for the respondents which, in turn, results in increased error variance, attribute non-attendance (a phenomenon where not all attributes are considered in reaching a decision), and inconsistent responses across choice tasks [6, 34, 35].

To reduce the list of attributes and levels, we engaged a purposively selected panel of 8 experts. The experts comprised of one general practitioner (trained in ART, providing ART service at a tertiary hospital and trainer of differentiated service delivery), one degree nurse prescribing ARV drugs, one pharmacist engaged in dispensing ARV drugs, one HIV/AIDS program focal person at the Amhara regional Health Bureau and one national HIV/AIDS program coordinator working at the Federal Ministry of Health of Ethiopia, two master’s holder public health professionals (who are expert in the field of a discrete choice experiment) and one Ph.D. holder in reproductive health (who is well experienced in HIV/AIDS research). The experts were eligible for the expert consultation process if they were healthcare providers directly prescribing and dispensing antiretroviral drugs, coordinators for HIV/AIDS program, researchers on HIV/AIDS, and experts in the discrete choice experiment approach. They were recruited by email for online consultation from the set of experts by informing the study’s purpose and expectations from the authors to receive their written reports.

The experts provided valuable feedback on the attributes and levels that would mirror those of DCE respondents. The experts screened all the attributes and levels generated from the previous stages. They used multiple criteria such as salience, plausibility, the capability of being traded, and relevance to study objectives and decision context [1, 28, 36, 37].

Researchers’ judgement

The authors held two meetings to review the decisions of the experts. They also agreed on an interim list of attributes and levels to be included in a discrete choice experiment study.



For a systematic review, 23 papers have been finally included and the review has been published [29]. Similarly, there were 57 articles which reported findings on the implementation and scale-up of DSD-related barriers and facilitators in our published scoping review study [30].


A total of 17 ART providers (ten of them were focal persons for ART service) and 15 PLHIV who were on ART (mean duration of ART intake was 10.1 ± 1.43 years) were interviewed to elicit their opinions and perspectives. Fifteen attributes namely buddy system, ART refill (individualized or group), ART packaging and labeling, drug formulation and administration, ART room labeling, distance, location of service, preferences on involvement in treatment decision-making, the person providing ART refills, provider’s attitude, spatial arrangement of ART room, time of health facility operation, time spent at clinics, and total cost of the visit were identified. These attributes are also determined by the previous evidence synthesis. Participants in both types of interviews did not discover new attributes and levels from the literature review.

Ranking and rating surveys

A ranking survey was done among 31 professionals who had experience working in HIV/AIDS programs at Amhara Regional Health Bureau, zonal health department and Woreda health office, and NGOs in Ethiopia (ICAP, USAID, and CDC) and Debre Markos University. In the ranking task, participants voted on 23 attributes. Detailed information on the voting results is shown in Table 1. The top seven most important attributes include the distance from the residential area to the ART clinic, buddy system (others take drugs to home in place of a client), location of ART service, time spent at clinics in ART pick-ups, ART refill (individualized or group), ARV drug administration and provider’s attitude.

Table 1 Ranked mean score of attributes by experts in antiretroviral therapy service in Northwest Ethiopia, 2022 (n = 31)

During the same period, a rating survey was undertaken among 42 healthcare providers at the ART clinics of public hospitals and health centers and 35 PLHIV using ART service in Awi, East and West Gojjam zones, and Bahir Dar city administration (Table 2). The overall mean score for each attribute varies from the ranking and rating surveys and between patient and provider participants of the rating exercise.

Table 2 Rating of attributes for antiretroviral therapy service among adult people living with HIV and providers in Northwest Ethiopia, 2022

Expert opinion

The final list of all possible attributes and levels was reviewed by an expert panel.

Through further discussions with experts, 17 inappropriate attributes were discarded based on the multi-criteria mentioned above. The 6 retained attributes were: location of ART refills, frequency of receiving ART refills, the person providing ART refills, participants/others seen at same ART refill visit, medication refill pick up/delivery times and total cost of visit during antiretroviral (ARV) medication refill. Table 3 presents the expert panel’s comments and decisions on attributes and levels of antiretroviral therapy service.

Table 3 Expert panel’s comments and decisions on attributes and levels for antiretroviral therapy service in Northwest Ethiopia, 2022

Realistic, relatable, and understandable levels were assigned to the 6 attributes based on data from literature, transcripts, and knowledge of the Ethiopian context. We considered a realistic level to be actionable by the policy. Attributes and levels were purposively kept simple and unambiguous, ensuring that all respondents could complete the survey with minimum effort. As suggested by Bridges and colleagues, we avoided ambiguous wording and tried to keep the number of levels to a minimum [1]. The levels were a mix of both categorical (location, provider, mode of appointment and medication refill time) and continuous (frequency of refills and cost of visit) in nature. Regarding the cost attribute, we determined cost levels after collecting preliminary data in real-life settings in Ethiopian health centers and hospitals. For the frequency of refills attribute, we adopted the levels from the WHO building blocks of differentiated service delivery [38]. All attributes and assigned levels to be used in the discrete choice experiment study are listed in Table 4.

Table 4 Identified attributes and levels for antiretroviral therapy to be included in a discrete choice experiment in Northwest Ethiopia, 2022


In the current study, we followed the recommendations of existing literature [1, 7, 20,21,22, 28]. This study was the first attempt to develop attributes and attribute levels of ART service for the Ethiopian population.

The whole process resulted in six ART service attributes to be included in the main DCE, namely, location of ART refills, frequency of receiving ART refills, the person providing ART refills, participants/others seen at same ART refill visit, medication refill pick up/delivery times and total cost of visit during antiretroviral (ARV) medication refill. These resulting attributes correspond to those that were applied in earlier DCEs investigating ART preferences in different settings [11,12,13,14,15].

Our methodology is based on substantial and carefully planned primary research (qualitative study, ranking, and rating surveys) in addition to secondary literature, following best practices recommended in low-income settings [20], making sure that all included features are pertinent and applicable.

This study has strengths. First, it shows how to rigorously and systematically conduct and report the process of deriving attributes and levels. This improves transparency and makes it reproducible. Secondly, our study used a multi-methods approach to develop attributes and levels. The four methods have pros and cons, and they complement each other, especially the analysis of qualitative data led to a deeper and broader understanding of attributes and levels.

However, it is practically impossible to include all potentially relevant attributes in a DCE because doing so would increase the required cognitive burden due to the amount and type of attributes chosen. Idealistically, these should be realistic and restricted to the essential ones. To allow for a few and manageable amounts of included attributes, this calls for a cautious and considered reduction approach. This procedure inevitably involves trade-offs and the exclusion of attributes that could be important for a sizeable section of the target population. In this study, trade-offs were done by considerations of a country’s legal, policy, technology and HIV treatment service contexts compared to participants’ perspectives and experiences.

While DCEs are a reliable and widely acknowledged method for preference exploration, their fundamental concentration on a small number of variables limits their ability to capture more comprehensive factors that affect preferences for ART services. This implies that our discrete choice experiment findings need to be followed up by qualitative and mixed-method research that will concentrate on understanding the individual and contextual factors contributing to certain preferences to completely understand the topic. Besides, the ranking exercise of 23 attributes by HIV/AIDS program implementers might create inaccuracy in the complete set of ranking orders. Future similar studies should break down the list of attributes into manageable groups for correct judgements by participants.


This study complements existing literature on DCE attribute development, by providing a detailed account of the careful application of recently recommended approaches to attribute and attribute-level development and reporting. Our applied approach is based on the adoption of literature as the starting point, to inform comprehensive field qualitative data collection, followed by a rigorous analytical approach, supported by a series of triangulation and validation exercises. Then, priorities setting methods (ranking and rating surveys) as well as expert consultations were done to explore the importance of attributes and levels and scale down them to a manageable number respectively. Finally, the location of ART refills, frequency of receiving ART refills, the person providing ART refills, participants/others seen at the same ART refill visit, medication refill pick up/delivery times and total cost of visit during antiretroviral (ARV) medication refill were the six attributes selected for the main discrete choice experiment in the Ethiopian context. As such, our study provides additional empirical guidance on the methodological processes of developing attributes and attributes levels for DCEs specifically within LMICs. A transparent description of the attribute development process of DCEs provides useful grounds for the assessment of the rigor of this process in DCEs, and hence, should receive more attention in future DCE studies.

Availability of data and materials

The datasets generated from the current study will be available from the corresponding author at a reasonable request.

Change history



Acquired immunodeficiency syndrome


Antiretroviral therapy




Discrete choice experiment


Differentiated service delivery


Human immunodeficiency virus


International society for pharmacoeconomics and outcomes research


Low and middle income countries


People living with HIV


Sub-Saharan Africa


  1. Bridges JF, Hauber AB, Marshall D, Lloyd A, Prosser LA, Regier DA, Johnson FR, Mauskopf J. Conjoint analysis applications in health—a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. J Value Health. 2011;14(4):403–13.

    Article  Google Scholar 

  2. de Bekker-Grob EW, Ryan M, Gerard K. Discrete choice experiments in health economics: a review of the literature. J Health Econ. 2012;21(2):145–72.

    Article  Google Scholar 

  3. Ryan M, Gerard K, Amaya-Amaya M. Using discrete choice experiments to value health and health care, vol. 11. Berlin: Springer; 2007.

    Google Scholar 

  4. Ryan M, Gerard K. Using discrete choice experiments to value health care programmes: current practice and future research reflections. J Appl Health Econ Health Policy. 2003;2(1):55–64.

    Google Scholar 

  5. Soekhai V, de Bekker-Grob EW, Ellis AR, Vass CM. Discrete choice experiments in health economics: past, present and future. J Pharmacoecon. 2019;37(2):201–26.

    Article  Google Scholar 

  6. Hensher D, Rose J, Greene W. Applied choice analysis. 2nd ed. Cambridge: Cambridge University Press; 2015.

    Book  Google Scholar 

  7. World Health Organization. How to conduct a discrete choice experiment for health workforce recruitment and retention in remote and rural areas: a user guide with case studies. Geneva; 2012. Accessed 2 Dec 2022.

  8. Johnson FR, Lancsar E, Marshall D, Kilambi V, Mühlbacher A, Regier DA, Bresnahan BW, Kanninen B, Bridges JF. Constructing experimental designs for discrete-choice experiments: report of the ISPOR conjoint analysis experimental design good research practices task force. J Value Health. 2013;16(1):3–13.

    Article  Google Scholar 

  9. Louviere JJ, Pihlens D, Carson R. Design of discrete choice experiments: a discussion of issues that matter in future applied research. J Choice Model. 2011;4(1):1–8.

    Article  Google Scholar 

  10. Kjaer T. A review of the discrete choice experiment-with emphasis on its application in health care. Odense: Syddansk Universitet Denmark; 2005.

    Google Scholar 

  11. Dommaraju S, Hagey J, Odeny TA, Okaka S, Kadima J, Bukusi EA, Cohen CR, Kwena Z, Eshun-Wilson I, Geng E. Preferences of people living with HIV for differentiated care models in Kenya: a discrete choice experiment. J PLoS ONE. 2021;16(8):e0255650.

    Article  CAS  Google Scholar 

  12. Eshun-Wilson I, Mukumbwa-Mwenechanya M, Kim H-Y, Zannolini A, Mwamba CP, Dowdy D, Kalunkumya E, Lumpa M, Beres LK, Roy M. Differentiated care preferences of stable patients on antiretroviral therapy in Zambia: a discrete choice experiment. J Acquir Immune Defic Syndr. 2019;81(5):540.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Opuni M, Bishai D, Gray GE, McIntyre JA, Martinson NA. Preferences for characteristics of antiretroviral therapy provision in Johannesburg, South Africa: results of a conjoint analysis. J AIDS Behav. 2010;14(4):807–15.

    Article  Google Scholar 

  14. Strauss M, George G, Mantell JE, Mapingure M, Masvawure TB, Lamb MR, Zech JM, Musuka G, Chingombe I, Msukwa M. Optimizing differentiated HIV treatment models in urban Zimbabwe: assessing patient preferences using a discrete choice experiment. J AIDS Behav. 2021;25(2):397–413.

    Article  Google Scholar 

  15. Zanolini A, Sikombe K, Sikazwe I, Eshun-Wilson I, Somwe P, Bolton Moore C, Topp SM, Czaicki N, Beres LK, Mwamba CP. Understanding preferences for HIV care and treatment in Zambia: evidence from a discrete choice experiment among patients who have been lost to follow-up. J PLoS Med. 2018;15(8):e1002636.

    Article  Google Scholar 

  16. Grimsrud A, Bygrave H, Doherty M, Ehrenkranz P, Ellman T, Ferris R, Ford N, Killingo B, Mabote L, Mansell T. Reimagining HIV service delivery: the role of differentiated care from prevention to suppression. J Int AIDS Soc. 2016;19(1):21484.

    Article  PubMed  PubMed Central  Google Scholar 

  17. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach. 2nd ed. Geneva: World Health Organization; 2016.

    Google Scholar 

  18. Okere NE, Meta J, Maokola W, Martelli G, van Praag E, Naniche D, Gomez GB, Pozniak A, Rinke de Wit T, de Klerk J. Quality of care in a differentiated HIV service delivery intervention in Tanzania: a mixed-methods study. J PLoS ONE. 2022;17(3):e0265307.

    Article  CAS  Google Scholar 

  19. Lancsar E, Louviere J. Conducting discrete choice experiments to inform healthcare decision making. J Pharmacoecon. 2008;26(8):661–77.

    Article  Google Scholar 

  20. Mangham LJ, Hanson K, McPake B. How to do (or not to do)… Designing a discrete choice experiment for application in a low-income country. J Health Policy Plan. 2009;24(2):151–8.

    Article  Google Scholar 

  21. Coast J, Al-Janabi H, Sutton EJ, Horrocks SA, Vosper AJ, Swancutt DR, Flynn TN. Using qualitative methods for attribute development for discrete choice experiments: issues and recommendations. J Health Econ. 2012;21(6):730–41.

    Article  Google Scholar 

  22. Hollin IL, Craig BM, Coast J, Beusterien K, Vass C, DiSantostefano R, Peay H. Reporting formative qualitative research to support the development of quantitative preference study protocols and corresponding survey instruments: guidelines for authors and reviewers. Patient-Patient-Centered Outcomes Res. 2020;13:121–36.

    Article  Google Scholar 

  23. Coast J, Horrocks S. Developing attributes and levels for discrete choice experiments using qualitative methods. J Health Serv Res. 2007;12(1):25–30.

    Article  Google Scholar 

  24. Hiligsmann M, van Durme C, Geusens P, Dellaert BG, Dirksen CD, van der Weijden T, Reginster J-Y, Boonen A. Nominal group technique to select attributes for discrete choice experiments: an example for drug treatment choice in osteoporosis. J Patient Preference Adherence. 2013;7:133.

    Article  Google Scholar 

  25. Ryan M, Scott DA, Donaldson C. Valuing health care using willingness to pay: a comparison of the payment card and dichotomous choice methods. J Health Econ. 2004;23(2):237–58.

    Article  PubMed  Google Scholar 

  26. Louviere JJ, Hensher DA, Swait JD. Stated choice methods: analysis and applications. Cambridge: Cambridge University Press; 2000.

    Book  Google Scholar 

  27. Gates R, McDaniel C, Braunsberger K. Modeling consumer health plan choice behavior to improve customer value and health plan market share. J Bus Res. 2000;48(3):247–57.

    Article  Google Scholar 

  28. Helter TM, Boehler CEH. Developing attributes for discrete choice experiments in health: a systematic literature review and case study of alcohol misuse interventions. J Subst Use. 2016;21(6):662–8.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Belay YA, Yitayal M, Atnafu A, Taye FA. Patients’ preferences for antiretroviral therapy service provision: a systematic review. J Cost Eff Resour Alloc. 2021;19(1):1–25.

    Google Scholar 

  30. Belay YA, Yitayal M, Atnafu A, Taye FA. Barriers and facilitators to the implementation and scale up of differentiated service delivery models for HIV treatment in Africa: a scoping review. J BMC Health Serv Res. 2022;22(1):1–23.

    Google Scholar 

  31. Guest G, Bunce A, Johnson L. How many interviews are enough? An experiment with data saturation and variability. J Field Methods. 2006;18(1):59–82.

    Article  Google Scholar 

  32. Belay YA, Yitayal M, Atnafu A, Taye FA. Patient experiences and preferences for antiretroviral therapy service provision: implications for differentiated service delivery in Northwest Ethiopia. J AIDS Res Ther. 2022;19(1):1–16.

    Google Scholar 

  33. Belay YA, Yitayal M, Atnafu A, Taye FA. Experiences and preferences of providers towards differentiated antiretroviral therapy service delivery in Northwest Ethiopia. 2022.

  34. Heidenreich S, Watson V, Ryan M, Phimister E. Decision heuristic or preference? Attribute non-attendance in discrete choice problems. J Health Econ. 2018;27(1):157–71.

    Article  Google Scholar 

  35. Lagarde M. Investigating attribute non-attendance and its consequences in choice experiments with latent class models. J Health Econ. 2013;22(5):554–67.

    Article  Google Scholar 

  36. Kløjgaard ME, Bech M, Søgaard R. Designing a stated choice experiment: the value of a qualitative process. J Choice Model. 2012;5(2):1–18.

    Article  Google Scholar 

  37. Obadha M, Barasa E, Kazungu J, Abiiro GA, Chuma J. Attribute development and level selection for a discrete choice experiment to elicit the preferences of health care providers for capitation payment mechanism in Kenya. J Health Econ Rev. 2019;9(1):1–19.

    Google Scholar 

  38. World Health Organization. Consolidated guidelines on HIV prevention, testing, treatment, service delivery and monitoring: recommendations for a public health approach. Geneva: World Health Organization; 2021.

    Google Scholar 

Download references


We acknowledge the University of Gondar for funding this study.


This study was supported by the University of Gondar in Ethiopia.

Author information

Authors and Affiliations



YAB took part in the planning, conception, analysis, and writing of the manuscript. MY and AA participated in the design, conception, and analysis phases of this manuscript. FAT engaged in the design, analysis and manuscript preparation. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Yihalem Abebe Belay.

Ethics declarations

Ethics approval and consent to participate

This study obtained ethical approval from the Institutional Review Board of the University of Gondar and a formal letter obtained from Amhara Public Health Institute. Participants provided written informed consent documenting their willingness to participate in interviews.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This article has been retracted. Please see the retraction notice for more detail:

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Belay, Y.A., Yitayal, M., Atnafu, A. et al. RETRACTED ARTICLE: Development of attributes and attribute levels for a discrete choice experiment on patients’ and providers’ choice for antiretroviral therapy service in Northwest Ethiopia. AIDS Res Ther 20, 33 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: