Cognitive Interviewing of the PHQ-A for Depression Screening Amongst Adolescents Living with and without HIV in Malawi

Abstract

Background:

Depression significantly affects adolescents in low- and middle-income countries (LMICs), particularly adolescents living with HIV (ALWH). In sub-Saharan Africa (SSA), where depression prevalence among ALWH ranges from 26–29%, it is essential to screen for depressive symptoms despite the challenges of limited resources. The Patient Health Questionnaire-Adolescent (PHQ-A) is a widely used tool for depression screening, but its effectiveness for adolescents in Malawi has not been thoroughly explored. This study cognitive interview tested the PHQ-A for adolescents in Malawi with the goal of ensuring cultural appropriateness and improving the tool’s sensitivity and specificity for the entire adolescent population.

Method:

The study conducted cognitive interviews to explore conceptualizations of depression among both ALWH (n=10) and adolescents without HIV (n=10) in Malawi, assess the acceptability and comprehension of PHQ-A items, and adapt the PHQ-A for local use. The study involved adolescents (age 13–19 years), half of whom were living with HIV, who were recruited from two public health centers in Lilongwe.

Results:

Findings showed that participants were able to answer all PHQ-A items, but faced challenges comprehending questions 2 and 5, using Likert scales, the response period, and truthfully answering questions about suicidality. Additionally, Malawian adolescents perceive depression through cultural expressions involving heart issues, excessive thinking, and a focus on family and friends. ALWH interpreted some symptoms of the PHQ-A differently, yet they generally understood and responded to the questions similarly to adolescents without HIV.

Conclusion:

While the PHQ-A is a widely used tool, its application among adolescents in Malawi requires further exploration and adaptation to ensure cultural relevance and accuracy in depression screening. This study underscores the importance of culturally sensitive mental health assessments to accurately screen adolescents and connect them with appropriate mental health services.

INTRODUCTION

Mental health disorders, notably depression, impose a significant global burden on adolescents with large variations observed across geographic regions and populations. Globally, the prevalence of depressive symptoms among adolescents is estimated at 10–20%. ¹Adolescents in low-middle-income countries (LMICs) and adolescents living with HIV (ALWH) are twice as likely to suffer from depression compared to their age-matched peers in high-income countries and those living without HIV.² For example, a systematic review found that the prevalence of depression among adolescents in LMICs can reach up to 39% in some settings.³ In sub-Saharan Africa (SSA), home to the largest population of ALWH in the world, depression ranges from 26 – 29%.⁴ The high prevalence of depression amongst ALWH in SSA likely reflects the numerous psychosocial and structural challenges that arise from living in a low-resource setting with a stigmatized, chronic disease during a critical developmental stage.^5,6,7

Depression in adolescents often presents as persistent sadness, irritability, loss of interest in activities, fatigue, difficulty concentrating, feelings of worthlessness, and changes in appetite or sleep patterns.⁸ These symptoms acutely disrupt academic performance, social relationships, and overall quality of life, also elevating the risk of substance abuse and suicidal behaviors.⁹ The resulting functional impairment can have long-lasting effects on development and well-being, particularly in settings such as sub-Saharan Africa, where studies show significant impacts on future educational attainment, economic stability, and social integration among depressed youth.¹⁰Depressive symptom screening is a critical first step in identifying and linking adolescents experiencing depressive symptoms to appropriate care, particularly in resource limited settings like Malawi where the prevalence of depression among ALWH ranges from 18–26%, and there are few clinicians available to conduct gold-standard clinical diagnostic interviews.¹¹ Many screening tools have been validated for use among adolescents in SSA, both with and without HIV. These tools include the Children’s Depression Inventory (CDI), the Patient Health Questionnaire for Adolescents (PHQ-A), the Patient Health Questionnaire-9 (PHQ-9), and the Center for Epidemiologic Studies Depression Scale (CESD).¹² Among these, the PHQ-A is one of the most widely used self-report screening tools specifically designed to assess depressive symptoms in adolescents. The PHQ-A is favored due to its free availability and brevity. It has demonstrated validity and reliability in various settings, including among ALWH in SSA,.¹³ It has also been used to measure the prevalence of depression in South Africa.¹⁴ The PHQ-A and PHQ-9 are widely utilized to quantify depressive symptom severity and monitor changes over time, providing a standardized approach for screening in both clinical and community settings.¹⁵ PHQ scores have been shown to correlate strongly with clinical diagnoses of depression and are effective in identifying adolescents at risk for major depressive episodes. ¹⁶ The use of PHQ-based screening tools in SSA has improved early detection and facilitated timely referrals, especially in environments with limited access to specialized mental health care. ¹⁷ Though this evidence supports the PHQ-A’s utility across SSA, its validity has not been evaluated amongst adolescents in Malawi.

Validating the PHQ-A amongst adolescents in Malawi is important for contextual and measurement reasons. Despite the widespread application of depressive symptom screeners in SSA, previous work indicates that they may not perform well among individuals with lower education levels or those who did not learn English as a first language.^18,19 Furthermore, responses may not fully reflect depressive symptoms due to cultural differences; for example, research has demonstrated that participants in Ethiopia found it easier to endorse items like feeling tired or having little energy, while others, potentially influenced by cultural values and factors like stigma, struggled with questions about suicidal thoughts.²⁰ These cultural differences highlight the importance of frameworks from cross-cultural psychiatry and cognitive anthropology, such as the ‘cultural concepts of distress,’ which has long recognized that locally salient idioms and expressions may shape how depression symptoms are experienced and reported in different settings. ²¹

The methods used to validate screeners for depressive symptoms in SSA vary significantly in their rigor. Typically, validation begins by translating the screener from English to the local language to improve comprehension and reliability. Some studies take an additional step by translating the measure back into English using a different translator to ensure accuracy and consistency with the original text.^20,22 Other studies go further by combining translation, back-translation, and qualitative analysis of participant interviews to ensure cultural relevance and accuracy.^23,24 When adapting tests for use in different languages or cultures, it is essential to gather evidence that the content and measurement properties remain comparable, as recommended by the Standards for Educational and Psychological Testing.²⁵ Only a few validation efforts have used cognitive interviewing, a specific qualitative method employed during questionnaire development, refinement, and translation. Cognitive interviewing is a technique used to understand how respondents perceive and interpret items on questionnaires. It is designed to provide information about misunderstandings, biases, and ambiguities that interviewees encounter when responding that researchers can then use to develop adapted questionnaires with improved validity and reliability.²⁶ Unlike in-depth interviews, which explore broader personal experiences, cognitive interviewing is specifically designed to probe the mental processes respondents use when answering survey questions, making it the most effective method for identifying and correcting item-level misunderstandings and cultural misalignments in psychological measures.^26,27 It is based on Tourangeau’s four-stage model of survey responses which systematically probes how participants comprehend questions, retrieve relevant information, make judgments about their answers, and select a response option to identify and address potential sources of response error. Researchers administer cognitive interviews in one-to-one sessions, asking respondents to provide their thoughts about questionnaire items and also probing (i.e., how did you arrive at the answer) and paraphrasing (i.e., can you repeat in your own words) to broaden information about respondents’ thoughts. Cognitive interviewing is frequently used in the development and adaptation of questionnaires as it provides information that can help increase the likelihood that individuals from diverse backgrounds interpret questionnaire items as intended.²⁶

This approach is particularly appropriate for pre-testing and adapting survey instruments, such as the PHQ-A, because the process helps ensure that survey items and responses accurately capture the intended information, identify potential sources of response error, and clarify how respondents understand the questions. Cognitive interviewing studies of the PHQ-9 and PHQ-A, analyzed through Tourangeau’s four-stage model (i.e., comprehension, retrieval, judgment, response), reveal that respondents commonly struggle with comprehension of multi-part items and mapping their experiences onto fixed response options. ^27,28 Difficulties at the retrieval and judgment stages are also reported, as participants may not recall symptoms in the specified timeframe or may hesitate to disclose sensitive information, leading to potential under-reporting or misclassification.^26,27 However, problems noted with recalling symptoms in specific timeframes are often interpreted a problem associated with primarily about how comprehension and developing a response that fits into the options rather than an actual information retrieval problem.²⁹ Across multiple cognitive interviewing studies of the PHQ-9, the main cognitive challenges consistently identified are comprehension of item wording and answer mapping, rather than memory retrieval. For example, studies in a range of settings including UK, Cameroon, and Kenya, ^28,30,31 as well as broader survey methodology research.^26,27 report persistent issues with interpreting questions and fitting experiences into response options, but do not highlight retrieval as a significant barrier. Collectively, these findings suggest that, while retrieval is part of the cognitive process, comprehension and response mapping are the most critical elements to address in PHQ-9 cognitive interviewing. Studies have found few significant differences in PHQ-9 scores or in the types of cognitive response issues (comprehension, decision, response) between women and adolescents living with HIV, rather participant performance may reflect broader challenges with the PHQ-9’s fit to local language, culture, and the lived experience.^28,32,33 Understanding whether these differences exist between ALWH and adolescents in Malawi is essential for accurate depression screening and referral.

This study aimed to conduct cognitive interviews to qualitatively explore conceptualizations of depressive symptoms among ALWH and adolescents without HIV in Malawi; assess acceptability and comprehension of PHQ-A items; and adapt the PHQ-A to appropriately capture depressive symptoms, with the eventual goal of creating a tool applicable to all Malawian adolescents. Ensuring the cultural appropriateness of the PHQ-A will facilitate future Malawian Ministry of Health efforts to integrate mental health screening into HIV services, other chronic care services e.g., Sickle Cell clinics, child and adolescent oncology care and support other future child and adolescent mental health research. The study aims to offer insights into how participants conceptualized depressive symptoms and responded to screening tool items, ultimately leading to improved PHQ-A sensitivity and specificity in this setting.

METHODS

Study Sample and Recruitment

Data are from the cognitive interview phase (N=20) of a study to adapt and validate the PHQ-A for adolescents living in Malawi. The study was conducted at two comparable public health centers that provided primary care services and HIV treatment services in Lilongwe, Malawi. Eligible adolescents were aged 13–19 and willing to provide consent (age 18 or 16–17 years old and married (emancipated minors per Malawi law)) or assent with parental consent (age 13–17). ALWH (n=10) were recruited from the ART clinics at the health centers and adolescents without HIV (n=10) were recruited from the primary care clinics at the health centers. All ALWH accessing HIV treatment services were eligible including ART initiators, ART re-initiators, and established patients.

PHQ-A Translation

The study coordinator, a trained nurse with more than 15 years of experience coordinating mental health research, translated the PHQ-A from English into Chichewa, the most widely spoken ethnic language in Malawi. The translation was based on previous work conducted by the study team amongst adults in Malawi and reviewed for accuracy by two other Chichewa-speaking study team members.³⁴ It was then back translated into English and reviewed by mental health professionals in Malawi and the United States to ensure that the translated items aligned with the meaning of the original English items.

Data Collection Procedures

Study participants were led through a series of semi-structured prompts that guided reflection on each PHQ-A item.^35,36 To promote standardization and minimize measurement error, all cognitive interviews were conducted by two experienced interviewers with training and practice using a semi-structured interview guide. Each interviewer followed a consistent protocol for introducing the PHQ-A, reading each item aloud in Chichewa, and providing the same instructions and response options to all participants. Both scripted and spontaneous probes were used to flexibly explore participants’ comprehension, decision-making, and response processes, with some variation in follow-up questions between interviewers to address unique responses and clarify emerging issues, consistent with best practices in cognitive interviewing. The interviewer offered participants focused probes to gain specific insights related to question comprehension (e.g., “What does “feeling down” mean to you?”), paraphrasing (e.g., “Can you repeat the question in your own words?”), social desirability (e.g., “How would you describe this feeling or situation to a friend? What specific words would you use?”), difficulty (e.g., “Was it difficult to respond to this question? Why?), and recall (e.g., “Can you remind me again how often you were feeling this way in the past two weeks?”). Additionally, interviews elicited information about the adolescent’s conceptualization of depression and perceptions of available mental health services, mental health stigma, and appropriateness of questions related to suicide.. All interviews were conducted one-on-one in a private setting, audio recorded, and transcribed verbatim for quality assurance and fidelity checks. In addition, periodic supervision and debriefing sessions were held with the interviewers to reinforce adherence to the protocol and address any procedural inconsistencies. These procedures helped ensure that, despite some natural variation in probing, the assessment process and test administration remained as standardized and reliable as possible. Interviews lasted approximately 60 minutes and were digitally audio recorded. Digital interview files were transcribed verbatim and then translated into English. Study staff referred all participants screening positive for probable depression (PHQ-A ≥ 5) to available Ministry of Health operated outpatient psychiatry services which are staffed by master’s level nurses.

Ethical Approval

The Institutional Review Boards of the University of North Carolina at Chapel Hill and the Malawi National Health Sciences Research Committee (NHSRC) approved this study. Before study enrollment, research staff engaged participants in a consent comprehension activity, which asks a series of questions to ensure participant understanding of the study. All participants (and guardian if present) received travel reimbursement of around 17,000 Malawi Kwacha equivalent to 10 USD.

Sample Size Considerations and Data Saturation

The sample size of 20 participants was determined based on established cognitive interviewing guidelines, which suggest that 15–30 interviews are typically sufficient to identify major comprehension and response issues in questionnaire adaptation. Throughout data collection and analysis, we employed an iterative approach using the Framework Method, with two independent reviewers extracting and comparing themes across participants. After approximately 16–18 interviews, no new themes or comprehension issues were identified, indicating that thematic saturation had been reached.

Data Analysis

Quantitative analyses focused on summarizing the sample using descriptive statistics (StataSE, version 19 (College Station, TX). Qualitative analyses were conducted using the Framework Method to identify trends and consistencies in participants’ responses using a meta-matrix. ^37–39 In this approach, two study team members read through each transcript and extracted text related to each PHQ-A item and discussion question into a matrix (e.g., an Excel file) with each row in the matrix representing a different participant, including their key demographics, and each column representing a PHQ-A item or discussion question. The coders also consulted with local Malawian mental health experts to further refine the coding process and interpretation. This approach enabled study team members to summarize key findings and identify patterns, themes, and illustrative quotes by each PHQ-A item and across participants in the data. Our analytic approach to interpreting cultural idioms of distress (such as “thinking too much” and “heart pain”) was grounded in the ‘cultural concepts of distress’ framework from cross-cultural psychiatry and cognitive anthropology.^8,21 This framework recognizes that culturally salient expressions of psychological suffering may overlap with, but are not identical to, Western psychiatric constructs. Accordingly, we included deductive codes for established idioms of distress and mapped participants’ narratives onto both PHQ-A constructs and culturally specific idioms. The qualitative analyses aimed to understand key themes related to PHQ-A item comprehension, depression conceptualization, comfort with acknowledging suicidal ideation, appropriateness of the item measurement, and the decision and response processes ²⁷

RESULTS

Participant Characteristics

The study included 20 participants purposively sampled to be living with HIV (50%) and without HIV (50%). Gender (50% female, 50% male) and age distribution (50% aged 13–14, 25% aged 15–16, and 25% aged 17–18) were consistent across both HIV status groups. Thirty percent of participants in both groups were orphans. Educational attainment varied slightly; 45% with no schooling, 40% completed primary education, and 15% completed secondary education. Food insecurity was more prevalent among ALWH participants (80%) compared to participants without HIV (50%). Financial insecurity was higher among ALWH, with 50% reporting having financial troubles compared to 10% among participants without HIV. The prevalence of probable depression (PHQ-A > 5) was 25% overall, 20% among ALWH, and 30% among participants without HIV. All ALWH participants were on antiretroviral therapy (ART), and 80% of ALWH acquired HIV through mother-to-child transmission (MTCT). (Table 1)

Table 1.

Participant Characteristics

		Total (N = 20)		HIV Positive (n = 10)		HIV Negative (n = 10)
		N	%	n	%	n	%
Age
	13–14	10	50%	5	50%	5	50%
	15–16	5	25%	2	20%	3	30%
	17–18	5	25%	3	30%	2	20%
Sex
	Female	10	50%	5	50%	5	50%
	Male	10	50%	5	50%	5	50%
Any Orphanhood
	Yes	6	30%	3	30%	3	30%
	No	14	70%	7	70%	7	70%
Education
	None	9	45%	4	40%	5	50%
	Completed Primary	8	40%	4	40%	4	40%
	Completed Secondary	3	15%	2	20%	1	10%
Food Insecurity
	Yes	13	65%	8	80%	5	50%
	No	6	30%	2	20%	4	40%
	Don’t Know	1	5%	0	0%	1	10%
Financial Insecurity
	Yes	6	30%	5	50%	1	10%
	No	10	50%	3	30%	7	70%
	Don’t Know	3	15%	1	10%	2	20%
Depression
	Yes (PHQ-A ≥ 5)	5	25%	2	20%	3	30%
	No	15	75%	8	80%	7	70%
ART
	Yes	10	50%	10	100%	--	--
	No	0	0%	0	0%	--	--
HIV Acquisition
	MTCT	8	40%	8	80%	--	--
	Don’t Know	2	10%	2	20%	--	--

Conceptualization of Depressive Symptoms

Participants were asked about their conceptualization of depressive symptoms. Three primary themes emerged from the response: cultural idioms, orientation towards friends and family, and variations based on HIV status (Table 2)

Table 2.

Summary of Key Themes Conceptualization of Depression

Key theme	Description of theme	Examples
Cultural idioms	Depression is described through cultural idioms, specifically issues related to the heart (n=14) and excessive thinking (n=10).	“Because I did not think about it a lot. Because when you were thinking about those things a lot your heart aches.” “I can tell her that I have no peace of mind because I am thinking about the same thing over and over again.”
Orientation towards friends and family	Strong awareness of how their actions, thoughts, feelings, and experiences related to depression could impact people in their lives (n=8).	“I cannot share this with my friends because it is difficult for me. Moreover, they would probably have their own problems, telling them could just compound their problems. On top of their problems, they will have to start feeling sorry for me too. In the end they could get sick, overthinking leading to headaches, so that is why I cannot share to my friends.” “I cannot share these feelings to my friends. They will tell me I should not think these thoughts.” “It could be that you have done something that your family doesn’t know about and what they wouldn’t expect from you. So, you feel that if they find out, things will not be good for you.”
HIV Status	Participants who were living with HIV often framed their responses to certain PHQ-A questions through the lens of their lived experience with HIV (n=8).	“Like if the drugs have finished, you are not happy and because you don’t tell anyone, you don’t feel happy.” “I can tell my friend that ‘my friend, the past five days, I did not sleep because at the clinic they said that if we don’t take the drugs we can die.’”

Cultural Idioms: the heart and excessive thinking

Many participants described depressive symptoms through cultural idioms, specifically issues related to the heart and excessive thinking. In addition to these idioms, several participants described their idioms in physical or bodily terms (i.e., pain in heart, thinking too much leading to headaches, heart aches, etc.) potentially reflecting somatization of psychological symptoms. Participants used heart-related idioms to describe feelings of sadness (n=3), lack of interest (n=3), sleep disturbances (n=2), failure (n=2), concentration issues (n=3), and suicidal thoughts (n=1). For instance, feeling down was often articulated through phrases such as “don’t feel well in the heart,” “heart not at peace,” and “anger in heart.” For example, when one participant was asked to describe depression, the participant said, “[depression is] not being happy and having something stuck in your heart.”

Similarly, lack of interest was described with expressions like “heart aches,” “heart is not accepting doing things,” and “heart is broken.” One participant explained a lack of interest by stating, “Because I did not think about it a lot. Because when you were thinking about those things a lot your heart aches.” Concentration issues were also linked to the heart, with participants mentioning the need to do something with their whole heart or lacking concentration because the heart was unwilling. For example, one participant said, “I wanted to read a book, and I failed to read the book because my heart didn’t feel like doing it.” In addition, one participant noted that the solution to suicide was related to calming one’s heart. The participant stated, “[a friend] should be patient. They should be able to calm their hearts down when something has happened. They shouldn’t hurry into thinking about committing suicide.” The idiom of excessive thinking was also used to describe depression, as reflected in participants’ descriptions of feelings of sadness (n=3), sleep problems (n=3), concentration difficulties (n=2), and a sense of failure (n=2). Participants frequently mentioned “thinking too much” to articulate their depressive symptoms. For example, one respondent said, “It can be difficult to explain because I can be thinking about many things, and I cannot manage to explain to a friend.” Another respondent described depression as, “… thinking about the same thing over and over again.” The idiom of thinking too much was also articulated in the context of sleep problems. A participant described why he had trouble falling asleep for 3 days by stating, “I feel like it was happening because I was thinking too much because when I look at my situation, I just talk to myself that if I was at a such a place, this thing wouldn’t have happened.” Another said, “I thought of things like when you do something wrong when you go to bed you keep on thinking about that same issue over and over again thereby failing to fall asleep.” Participants also described symptoms of depression by discussing issues related to their mind. For instance, when asked question 6 (feeling bad about yourself, feeling like a failure, or letting yourself or your family down), one participant described a situation in which he felt he had brought shame to his family by damaging something that belonged to someone else and being confronted about it at home in front of his parents. He explained that, as a result, he could express this feeling of failure or shame to a friend by saying, “I can tell them…. I had no peace of mind.” Another participant noted for question 3 (trouble falling asleep, staying asleep, or sleeping too much), “I had no peace of mind and that made me have sleepless nights.”

Orientation towards friends and family

Throughout the cognitive interviews, participants consistently viewed their experiences with depression symptoms through the lens of their relationships with friends and family. They demonstrated a strong awareness of how their actions, thoughts, feelings, and experiences related to depression symptoms could impact and be impacted by people in their lives. A few areas where participants’ orientation towards friends and family appeared were on PHQ-A questions related to feeling down (n=4), lack of concentration (n=8), and restlessness (n=3) among others. For example, several participants linked symptoms of depression with being gossiped about and negative peer interactions. One participant said for question 1 (feeling down, depressed, irritable, or hopeless) “For the past two weeks, I have been troubled because of being gossiped about by many people.” A different participant spoke about the pain she felt when a boy spread false rumors about her at school, stating, “This boy was always coming in my class, and I could see him, and always I could remember that he said some things to my mom, and these things angered my mom. So, because of that, I also got angry.” In response to the same question, a third individual explained how sadness leads to isolation and difficulty being around friends. The participant stated, “…I have no cause to be happy. So, I tend to be alone, angry, and unhappy. I can’t even laugh at jokes with friends. That is how one gets when one is sad.” Yet another participant described how his struggle with mental health could impact his friends in a somaticized way, stating, “it could compound their problems.…on top of their problems, they will have to start feeling sorry for me too. In the end, they could get sick, overthinking leading to headaches, so that is why I cannot share with my friends.”

Additionally, for question 8 (moving or speaking slowly, or being fidgety/restless), one participant described the symptoms of restlessness in the context of hiding secrets from family members. The participant said, “It could be that you have done something that your family doesn’t know about and what they wouldn’t expect from you. So, you feel that if they find out, things will not be good for you.” For question 7 (trouble concentrating on things), one participant described concentration as “It means you do not stay stable when reading or playing with your friends.” Another said, “because I had headache maybe because of the sun and because of what my friends did to me.”

In addition to participants’ references to family and friends in their explanations of PHQ-A symptoms, many participants seemed to misinterpret the interview prompt “how would you describe this feeling or situation to a friend.” Although this prompt was intended to elicit participants subjective conceptualization of depressive symptoms and to assess the impact of social desirability, many participants seemed to orientate their responses more towards the social relationship itself than to their own internal experiences. As a result, their answers often revealed concerns about the social desirability of disclosing depressive symptoms to friends, rather than providing as much insight into their personal understanding of those symptoms. For example, several participants responded by stating that they would never be able to tell their friends about experiences of depression symptoms, highlighting both a reluctance to disclose negative feelings and the challenge of articulating such experiences in a social context. Even when participants attempted to address the subjective aspect of the question, their difficulty in explaining symptoms seemed closely tied to the imagined act of telling a friend, suggesting that the presence of a social audience made it more cognitively and emotionally challenging to describe their feelings. One participant, for instance, stated that describing a lack of interest to a friend was “proving to be hard for me,” and explained, “It can be difficult to explain [to a friend] because I can be thinking about many things, and I cannot manage.” Another participant misunderstood the prompt entirely, interpreting it as a question about how to help a friend, rather than how to describe their own experience. While not all participants experienced this misunderstanding, it was present across multiple questions. This dynamic is further illustrated in the following exchange: when asked how he could describe the feeling of low energy to a friend, one participant replied, “…this can be very difficult for me to share with a friend…because you can’t share everything with friends…” When prompted to imagine sharing the feeling, he was able to articulate what he might say, but agreed with the interviewer’s summary that “the challenge itself is sharing with a friend, not the content of what is shared.”

HIV Status influence on depression experience

Participants who were living with HIV often framed their responses to certain PHQ-A questions through the lens of their lived experience with HIV (n=8). For example, one participant equated feeling hopeless with “having nothing to rely on” and linked it to the availability of ART drugs, stating, “Like if my drugs are finished while at the clinic, they told us that we need to be taking them daily.” Lack of interest and pleasure was similarly tied to ART availability, with one participant explaining, “Like if the drugs have finished, you are not happy and because you don’t tell anyone, you don’t feel happy.” Sleep problems were also discussed in the context of ART adherence, with one participant noting, “I can tell my friend that ‘my friend, the past five days, I did not sleep because at the clinic they said that if we don’t take the drugs we can die.” A participant also described concentration in the context of ARTs in the following way, “like taking ARTs on daily basis, it is difficult. Sometimes they have to remind me.” Regarding suicidal thoughts, one participant reported, “I thought that if I cannot take my ARTs for one month and I become very sick, I can think that it is better to die than suffering like this.”

Item Comprehension

Comprehension process is the method by which respondents understand and interpret survey questions, including the intent of the question and the meaning of specific terms. Assessing comprehension involves determining what the respondent believes the question is asking and how they interpret the words and phrases used in the question.²⁷ Overall, participants generally understood the PHQ-A questions. However, some exhibited difficulties in comprehending questions related to question 2 (little interest or pleasure in doing things), question 3 (trouble falling asleep, staying asleep, or sleeping too much) and question 5 (feeling tired or having little energy). (Table 3)

Table 3.

Matrix of Primary and Secondary Cognitive Response Issues for Each PHQ-A Items

PHQ-A Item	Comprehension	Decision	Response
1. Feeling down, depressed, or hopeless		X	x
2. Little interest or pleasure in doing things	X	x
3. Trouble falling/staying asleep, sleeping too much			X
4. Feeling tired or having little energy		X
5. Poor appetite or overeating	X	x
6. Feeling bad about self, failure, letting self/family down		X	x
7. Trouble concentrating	X
8. Moving/speaking slowly or being restless
9. Thoughts of self-harm or suicide		X	x

^*notes: Comprehension – Understanding the question and its intent. Decision – Assessing and evaluating the recalled information to form an answer. Response – Mapping the answer onto the provided response options and communicating it. X = Primary issue (uppercase bold). x = Secondary issue (lowercase regular font)

For question 2 (little interest or pleasure in doing things), participants (n=7) were confused on the components of pleasure, specifically distinguishing between having “little pleasure” and “pleasing others.” For instance, one participant remarked, “I was doing something which was not pleasing my friends. I was losing temper and being angry.” Additionally, some participants struggled to differentiate between situational interest and disengagement from daily activities. These participants appeared to view interest as fulfilling obligations or engaging in activities out of expectation, rather than pursuing things they genuinely desired to do. For example, one participant explained interest by stating, “Maybe you wanted to sweep but instead of you sweeping, someone else sweeps on your behalf. It means you are not interested because you were supposed to sweep and if someone has swept for you, you are no longer interested.”

For question 3 (trouble falling asleep, staying asleep, or sleeping too much), participants (n=3) were confused by the different variations of sleep-related issues. One participant suggested that a practitioner could clarify the question if he were asking it by saying, “I can use the word sleeping too much and not able to sleep because if I can say trouble falling asleep will be difficult for him to understand what am trying to say.”

For question 5 (feeling tired or having little energy), the interviewer needed to differentiate between having no food and having no interest in food when it is available for participants (n=8). For example, when asked about having no appetite, one participant responded, “It means when there is no food at home for three days, you have appetite for food.” Additionally, there was minimal mention of overeating among participants.

Decision Process

Decision process is the method by which respondents decide how to answer survey questions, influenced by their motivation, sensitivity to social desirability, and other social biases. Assessing decision processes involves determining whether the respondent devotes sufficient mental effort to answer accurately and whether they are inclined to respond truthfully or in a way that makes them look a particular way.²⁷ Questions related to suicidality and concerns about judgement and shame from peers emerged as significant issues that seemed to affect participants’ decision process. (Table 3). To assess participants general comfort disclosing suicidality, the interviewers asked “How comfortable do you think the youth would be answering this question honestly” 13 out of 20 participants said they did not think youth would be comfortable answering truthfully. One interviewer asked a participant whether other young people would respond honestly if questioned in the same manner. The participant’s reply was as follows: “they cannot be honest…. they may think that it is a dangerous issue to be known that you had suicidal thoughts, and they may think that you would be reported to police.” Another said “no, he cannot respond that way [honestly] because if he is to disclose that he wants to commit suicide you [the interviewer] are going to entice him with some promises or else telling him other things. So, he cannot be honest.” The participants’ discomfort sharing suicidality may have contributed to the occurrence that only 1 participant acknowledged suicidality on the PHQ-A. Of the few participants who stated they would be comfortable with sharing suicidality, the reasons they provided were “because they would want to be counselled and live normal life again.” Another participant also said, “because they want to get the support there so they will reveal what they are thinking.” Many participants expressed discomfort in sharing suicidality unless confidentiality and trust were assured (n =5). For instance, one participant indicated that before describing suicidality to a friend, he would first “see if the person can keep confidentiality for me.” Others suggested that individuals experiencing suicidality should “talk to an elder or anybody he or she trusts about the problems being faced.”

Participants also faced challenges in acknowledging their symptoms of depression due to concerns about peer judgment and shame (n=3). This issue was particularly evident in responses to question 1 (feeling down, depressed, irritable, or hopeless) and question 9 (thoughts of being better off dead or of hurting yourself in some way). For example, one participant stated, “I cannot share these feelings [suicidal ideation] with my friends. They will tell me I should not think these thoughts.” Another participant noted “they may feel ashamed to tell other people” to describe the reason why participants would likely not tell the interviewer if they had suicidal ideation. Some participants did not seem to have a well-developed idea of how the interviewer would regard them if they disclosed suicidal ideation. For example, one participant said participants would not tell the interviewer because “…of fear of being shouted at because those are bad thoughts.” Many of the issues related to shame and peer judgement came from question 6 which asked respondents. “How many days have you been bothered by feeling bad about yourself or that you’re a failure or have let yourself or your family down.” For this question one participant said that letting his/herself or family down means “doing unacceptable things that can bring shame to the family.” Another said, it is “doing something that makes your parents feel ashamed.” When asked to describe what the symptoms in question 6 meant to participants, most (n = 13) interpreted as relating to feelings of shame and being ashamed. In addition, 6 of the 9 participants who acknowledged symptoms for question 6 specifically pointed out having feelings of shame/embarrassment.

Response Process

Response process is the method by which respondents interpret and answer survey questions, ensuring their internally generated answers align with the provided response categories.²⁷ Participants struggled with certain aspects of the questionnaire, notably the Likert scale format and double-barreled questions. (Table 3). The two-week timeframe of the PHQ-A presented challenges for participants, who encountered difficulties understanding the measurement period and articulating the exact duration of their symptoms. Despite the specified timeframe, several participants referenced symptoms extending well into the past. For instance, when asked to define “the past two weeks,” some responded with “the past days” or even “the past two months.” One participant stated that there are “5 days in a week…7 counting Saturday and Sunday.” Another, acknowledging the difficulty with the PHQ-A’s format, suggested, “It would be good to ask the participant how many days are in a week, then in two weeks.” Beyond varied interpretation of the timeframe, participants also tended to select extreme responses on the Likert scale, most often choosing either 0 days (the shortest timeframe) the least or 13 −14 (the longest timeframe).

Question 3 (trouble falling asleep, staying asleep, or sleeping too much) highlighted the challenges posed by double-barreled questions. It asked, “How many days have you had trouble falling asleep, staying asleep, or sleeping too much?” When asked to rephrase the question in their own words, many participants omitted certain aspects. One participant simplified it to, “In the past three days, how many days have I had sleepless nights?” Another rephrased it as, “It means how many days I did not sleep or how many days I slept too much.”

DISCUSSSION

This study aimed to examine how Malawian adolescents, living with and without HIV, conceptualized depressive symptoms, responded to PHQ-A screening tool items, and provides recommendations for revising the PHQ-A based on these results. Conceptualizations of depressive symptoms included cultural idioms, reflected an orientation for friends and family, and were viewed in the context of HIV. Overall, study participants found the PHQ-A to be an acceptable screening tool but reported some challenges in answering items related to appetite, pleasure, sleep, and feelings of failure. Immediate comprehension of the response options for the screening tool was sometimes an issue and some participants reported hesitancy in acknowledging their depressive symptoms or discussing suicidality due to concerns about peer judgment, shame, and illegality of suicide (see Table 4). These findings highlighted opportunities for revising the PHQ-A items for this population (See Table 5a for a suggested revision of the PHQ-A for adolescents in Malawi), developing a visual guide for the response options, and proposed the creation of a two-question screening tool comprised of items that captured culturally and contextually relevant expressions of depressive symptoms (See Table 5b for pilot supplementary screener example)

Table 4.

Suggested Changes to PHQ-A Items

Original items	Issue (i.e., decision, comprehension, response problems)	Final suggested PHQ-A item
1. How many days have you been bothered by feeling down, depressed, irritable, or hopeless?	Decision problems: Concern about answering because it could lead to feelings of shame. (n=2)	No change
2. How many days have you had little interest or pleasure in doing things?	Comprehension problems: confused on the components of pleasure, specifically distinguishing between having “little pleasure” and “pleasing others (n=7).	How many days have you lost interest in doing things that you are usually interested in?
3. How many days have you had trouble falling asleep, staying asleep, or sleeping too much?	Response problem: Double barreled. Confused about the variations of sleep (n=3)	How many days have you had trouble sleeping? Trouble sleeping includes trouble falling asleep, trouble staying asleep, or trouble sleeping too much?
4. How many days have you been bothered by feeling tired, fatigued, or having little energy?	Decision problems: Participants engaged the question around the terminology of weakness (n=2)	How many days have you been bothered by feeling tired, fatigued, weak, or having little energy?
5. How many days have you experienced poor appetite, weight loss, overeating, or weight gain?	Comprehension problems: participants did not differentiate between having no food in the home and no appetite for food. (n=8)	how many days have you experienced poor appetite, weight loss, overeating, or weight gain - when food is available
6. How many days have you been bothered by Feeling bad about yourself or that you’re a failure or have let yourself or your family down	Decision problem: Participants engaged around the terminology of shame (n=3).	How many days have you been bothered by feeling bad or shameful about yourself or that you’re a failure or have let yourself or your family down?
7. How many days have you had trouble concentrating on things such as reading, talking with friends, or schoolwork?	None	No change
8. How many days have you been bothered by moving or speaking so slowly that other people could have noticed. Or the opposite – being so fidgety or restless that you have been moving around a lot more than usual	None	No change
9. How many days have you been bothered by thoughts that you would be better off dead or hurting yourself in some way?	Decision problems: Concern about answering because could lead to feelings of shame and legality (n=5)	No change

Table 5a.

Suggested revision of the PHQ-A for adolescents in Malawi

PHQ-A
Name:	Clinician:		Date:
Instructions: How often have you been bothered by each of the following symptoms during the past two weeks?
	Not at all	Several days	More than half the days	Nearly every day
1. How many days have you been bothered by feeling down, depressed, irritable, or hopeless?	0	1	2	3
2. How many days have you lost interest in doing things that you are usually interested in?	0	1	2	3
3. How many days have you had trouble sleeping? Trouble sleeping includes trouble falling asleep, trouble staying asleep, or trouble sleeping too much?	0	1	2	3
4. How many days have you been bothered by feeling tired, fatigued, weak, or having little energy?	0	1	2	3
5. How many days have you experienced poor appetite, weight loss, overeating, or weight gain - when food is available	0	1	2	3
6. How many days have you been bothered by feeling bad or shameful about yourself or that you’re a failure or have let yourself or your family down?	0	1	2	3
7. How many days have you had trouble concentrating on things such as reading, talking with friends, or schoolwork?	0	1	2	3
8. How many days have you been bothered by moving or speaking so slowly that other people could have noticed. Or the opposite – being so fidgety or restless that you have been moving around a lot more than usual	0	1	2	3
9. How many days have you been bothered by thoughts that you would be better off dead or hurting yourself in some way?	0	1	2	3

Table 5b.

Pilot supplementary screener example

Pilot Supplementary Screener
Name: __________________		Date: _______________
Over the last 2 weeks, how often have you been bothered by any of the following problems? (circle to indicate your answer)
	Note at all	Several days	More than half days	Nearly every day
1. Experienced difficulties with your heart, such as having anger in your heart, your heart aching, your heart not at peace, and or it your heart being unwilling.	0	1	2	3
2. Thinking too much, thinking about the same thing repeatedly, or having no peace of mind?	0	1	2	3
	Add columns
	TOTAL:

Themes related to the conceptualization of depressive symptoms.

The use of cultural idioms like “pain in my heart” and “thinking too much” by adolescents in Malawi provides critical insights into the somaticized experience of depressive symptoms and the interconnectedness of physical and mental health, aligning with existing research among both adults with and without HIV. In Malawi, research on postnatal depression have documented “pain in my heart” as a prevalent expression of emotional suffering and potential mental health issues.^40,41 Similarly, “thinking too much,” identified in various health contexts including a hypertension study, reflects cognitive burden and stress that may exacerbate physical health conditions.⁴² “Thinking too much” is linked to distress, depression, and anxiety and can arise from multiple overlapping conditions such as hardship, personal loss, spiritual beliefs, traumatic events, serious illnesses like HIV, social stressors, and economic factors.⁴³ “Thinking too much” shares certain features with the psychiatric concept of rumination (i.e., persistent, repetitive, and distressing thoughts), illustrating some overlap between the two experiences.⁴⁴ However, it is important to recognize that this idiom cannot be reduced to a single psychiatric construct, as it carries unique cultural significance. Idioms like “thinking too much” and “pain in my heart” are often less stigmatizing than clinical terms. Using these expressions alongside psychiatric labels, rather than replacing or strictly conforming them to clinical terminology, may help reduce stigma, enhance communication between clinicians and patients, and promote more culturally sensitive therapeutic interventions. ⁴⁵

In our study, these idioms were together used to describe a wide variety of symptoms in 6 of the 9 PHQ-A items. Participants used heart-related idioms to describe feelings of sadness, lack of interest, sleep disturbances, failure, concentration issues, and suicidal thoughts. Excessive thinking and rumination were used to describe depression, feelings of sadness, sleep disturbances, concentration issues, and a sense of failure. Therefore, it is not immediately clear which items to recommend adding or adjusting by incorporating culturally relevant examples (i.e., expanding question 1 “Feeling down, depressed, irritable, or hopeless” with examples like “thinking too much” or “no peace in your heart.” Furthermore, the PHQ-A is grounded in the diagnostic criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-T) and the International Classification of Diseases, 11th Revision (ICD-11). While these frameworks are primarily influenced by the perspectives of the Global North and may overlook other cultural expressions of distress, they provide a standardized basis for diagnosing depression. Adding more questions to the PHQ-A could compromise this diagnostic validity and might lead to identifying conditions that no longer align with the original definition of depression.

Given the pervasiveness of these cultural idioms in the conceptualization of depression, patient use of phrases like “pain in my heart” and “thinking too much” during a primary or HIV care visit could alternatively provide a useful signal to screen for depressive symptoms. Future research is needed to develop these idioms into a culturally adapted screening tool which could identify concepts like overall distress. This approach is consistent with recommendations from other studies on the integration of idioms into mental health treatment and assessment.⁴⁶ (table 5b)

Issues related to decision processes.

Results were consistent with research showing that questions about suicidal ideation are difficult to answer.^20,47 Societal stigma and the illegality of suicide in Malawi further exacerbates stigma and affected participants’ willingness to answer question 9, which asked about being bothered by thoughts of death or self-harm. They were hesitant to disclose such thoughts to a trained evaluator and friends. The need for confidentiality and trust to feel comfortable discussing suicidality was often brought up by participants. This finding was unsurprising given numerous studies that show stigma affects both suicidal individuals and their bereaved families, hindering access to care and raising suicide risk due to fear of discrimination.⁴⁷ In this setting, it may be important for healthcare providers to reassure patients of their confidentiality with respect to disclosure of suicidal ideation or behaviors, allowing time for discussion and explanation of procedures for managing and referring patients following disclosure.⁴⁸ Developing adolescent specific procedures for explaining confidentiality, management and referral in response to disclosure of suicidal ideation a behavior, may in turn help reduce patient fear and address stigma (Table 5c)

Table 5c.

Additional instructions for the interviewer

Problem	Proposed solution	Example
Hesitance to acknowledge suicidality	Script to explain suicidality and confidentiality before question 9	Sometimes, when people feel really sad or upset, they might have thoughts about hurting themselves. It’s important to talk about these feelings with a trusted adult, because help is available. If you tell me you’re having these thoughts, I’ll listen carefully and help you find the support you need, which might include talking to other caring adults or getting extra help from a doctor. Anything you tell me will stay between us, unless you’re in danger of hurting yourself or someone else. What questions do you have?
Difficulties understanding the measurement period and duration of symptoms	Script to explain 2 weeks.	The questions ask about how you’ve been feeling over the last two weeks. Two weeks is like if you go to church, think about how you’ve felt since the last two Sundays. It’s okay if you don’t remember every single day—just try your best to think about how you’ve felt most of the time.”
	Practice example with interviewer selecting Likert item	Let’s practice a question together. Suppose the question is: ‘In the past two weeks, how often have you felt tired?’ Let’s say you felt tired a couple of times last week. Which answer would you choose? (point to example)
	Image of bottles or baskets

This study found that shame and embarrassment can hinder participants’ willingness to disclose symptoms. This was evident in their general reluctance to share symptoms with friends and their particular discomfort discussing suicidal ideation, even with the interviewers. Furthermore, participants’ explanations of shame revealed a misunderstanding of the confidential nature of the therapeutic relationship. Some feared the interviewer might share their information with the community, much like a friend, while others worried about punitive responses to disclosure. Shame was especially prominent in responses to item 6 of the measure (“How many days have you been bothered by feeling bad about yourself or that you’re a failure or have let yourself or your family down?”), which strongly resonated with participants’ experiences of shame. Given the prevalence of shame and embarrassment observed in the cognitive interviews, particularly regarding item 6, we recommend revising it to: “How many days have you been bothered by feeling bad or shameful about yourself or that you’re a failure or have let yourself or your family down?”

Issues related to item comprehension.

Many participants in our sample confused the PHQ-A item related to lack of appetite with lack of food in the home. This confusion potentially reflects the high prevalence of food insecurity in Malawi, particularly amongst families where an individual is living with HIV.⁴⁹ A study conducted in Ethiopia similarly found that participants perceived poor appetite as related to the availability of food rather than the desire for eating.⁵⁰ Given the participants’ confusion in the current study, it is unclear whether the question as written measures depressive symptoms related to food insecurity or decreased appetite related to mood—an important distinction both conceptually and for intervention purposes. For instance, addressing depressive symptoms related to food insecurity stemming from food insecurity might involve offering increased instrumental support, such as providing food. This aligns with findings that instrumental support had a greater buffering effect on depressive symptoms for Ugandan PLWH than emotional social support.⁵¹ Conversely, the treatment approach for an individual experiencing decreased appetite due to low mood might involve offering emotional and social support. To reduce measurement error and comprehension challenges, the study team suggested adding “- when food is available” to distinguish between lack of appetite and lack of food for this PHQ-A item. In food insecure settings, mental health assessments may need to expressly distinguish between appetite loss due to food scarcity and mood-related appetite changes. Screening items should be refined to clarify these causes, allowing for tailored interventions that either provide support to alleviate food insecurity or address mood-related issues.

Participants in our sample interpreted the concept of pleasure within the framework of relationships, often confusing their own personal pleasure with pleasing others. This may suggest that participants were describing a loss of pleasure due to their inability to satisfy others. Their perception of pleasure appears to be more ubuntu humanism—viewing pleasure as a social experience derived from the happiness of others, rather than an individualistic perspective where pleasure is associated with the activity itself, regardless of company⁵². This aligns with cross-cultural studies on well-being, which introduce the concept of interdependent happiness as a more relationship-oriented view of happiness, emphasizing harmony with others, a sense of calm, and ordinariness.⁵³ To reduce measurement error and potential comprehension issues, the study team recommends changing question number 2 (little interest or pleasure in doing things) to the following, “how many days have you lost interest in doing things that you are usually interested in?”

Though HIV status impacted how participants conceptualized depressive symptoms, we did not find that HIV status significantly influenced participant’s responses or ability to engage with the items. These results align with a study that determined that the PHQ-9 and PHQ-2 were valid and reliable for assessing DSM-IV depressive disorders and depression severity among adults living with HIV in western Kenya.⁵⁴ Participants drew upon their experiences managing HIV to articulate their responses for the PHQ-9. For example, some participants with HIV reported experiencing bodily weakness. Additionally, feelings of hopelessness were notably linked to HIV status, often due to challenges in obtaining medication. Unlike food or other resources, the need for medication can create a heightened sense of urgency and hopelessness, not only because of the severe consequences of not obtaining the medication but also due to limited social support and the stigma associated with seeking help and living with the condition. The study shows that the PHQ-A effectively assesses depression in both adolescents with and without HIV, indicating it is a robust tool for diverse adolescent populations regardless of HIV status. Importantly, while chronic illness may shape the experience of depression, it does not seem to compromise the validity of the PHQ-A assessment.

Response processes.

Our study findings have significant implications for the development and implementation of scales with Likert response options generally. Participants showed confusion when selecting between several response options, suggesting that the use of visual aids could enhance the screening process. Studies have indicated that cultural factors might explain some of the differences in performance on Likert scales, particularly due to collectivist versus individualist response patterns and tendencies to express specific emotions.⁵⁵ Additionally, studies have demonstrated that low literacy levels can diminish the effectiveness of Likert scales.⁵⁶ Visual aids have been employed in other studies to increase the utility of the PHQ-9. For example, they have demonstrated that pictorial illustrations of Likert-type scales are culturally useful and may reduce errors associated with Westernized self-report measures in non-Western settings.⁵⁷

Moreover, our study highlights a critical issue with administering the PHQ-A in low-literacy contexts. Although the PHQ-A was adapted for provider administration in this study to address literacy challenges, it was assumed that participants understood the time frame specified in the questions. With 45% of adolescents in our study having no formal education, it is not surprising that the concept of ‘two weeks’ might not be consistently understood. Consequently, it is recommended that the definition of a ‘two-week’ period be clearly emphasized or clarified in writing at the beginning of the questionnaire. If feasible and not overly cumbersome, administrators may also assess whether participants understand this concept and provide a verbal explanation as needed. In addition, incorporating a visual aid, such as bottles or grain baskets filled to different levels, to anchor questions to the Likert scale (0 days, 1–6 days, 7–12 days, 13–14 days) could enhance comprehension. Finally, providing participants with a brief example demonstrating how to select a response based on a two-week symptom timeframe before administering the PHQ-A could prove beneficial (Table 5c).

The study also suggests revising question 3 (trouble falling asleep, staying asleep, or sleeping too much) to address its double-barreled nature, which has proven problematic for participants in several studies, along with questions 5 (feeling tired or having little energy) and 8 (moving or speaking slowly, or being fidgety/restless). While some studies have attempted to resolve this by adding instructions like, “Please select an answer choice regardless of which of these problems you may have had.”²⁸ This study recommends revising the question itself to: “How many days have you had trouble sleeping? Trouble sleeping includes trouble falling asleep, trouble staying asleep, or trouble sleeping too much?”

Limitations

This study has several limitations. It is possible that participants misinterpreted the PHQ-A items due to cultural differences that were not fully appreciated by the study. Additionally, the qualitative nature of this study means that we cannot comment on the psychometric validity and reliability of the adapted PHQ-A compared to a diagnostic gold standard. Moreover, there are limitations regarding generalizability. Another key limitation of this study is the modest sample size (N=20), which, while consistent with cognitive interviewing best practices and sufficient for identifying major comprehension issues, may not fully capture the range of variability in comprehension across all demographic subgroups, such as education level or urban/rural background. Although we purposively sampled for diversity in HIV status, gender, and age, and observed thematic saturation during analysis, our ability to detect rare or subgroup-specific comprehension challenges is limited. Future research with larger and more demographically diverse samples will be important to further explore potential differences in item interpretation and acceptability across subgroups. Due to the small, purposively selected sample, our findings cannot be generalized to a broader population. Since adolescents were recruited from peri-urban clinics in Lilongwe, these results may not represent the experiences of adolescents across the entire country, especially in rural areas. An additional limitation is that the transcripts were translated from Chichewa to English. Therefore, it is important to note that the psychometric properties of the modified questionnaire will need to be specifically tested to ensure its accuracy and appropriateness for the target population. As with any translation, some degree of information and understanding is inevitably lost. To minimize this loss, the transcripts were reviewed by team members who conducted the interviews, ensuring transcription and translation accuracy, as they were fluent in both Chichewa and English and present for the interviews. The study proposed changes to the PHQ-A to address issues such as double-barreled questions, which can confuse participants and lead to inaccurate responses. However, implementing these changes and translating them back into Chichewa could introduce additional errors, potentially affecting the tool’s reliability and validity. Despite these limitations, this study offers valuable insights into the adaptation of the PHQ-A to Malawian youth and their experiences of depression.

In future research, we plan to examine the psychometric properties, and test the diagnostic performance, of the revised PHQ-A by comparing it to clinical interview and other validated measures of depression. Pending validation, we hope to pilot the revised instrument in a larger and more demographically diverse sample.

CONCLUSION

In conclusion, cognitive interviewing revealed that Malawian adolescents, both with and without HIV, conceptualize depression through cultural idioms and orientation towards friend and family, often linking physical and mental health with expressions like ‘pain in my heart’ or ‘thinking too much. Participants showed a clear hesitancy to discuss suicidality. Adolescents living with HIV drew from their personal experiences to understand mental health issues. While participants generally found the PHQ-A to be an acceptable screening tool, some experienced difficulties with specific items and understanding response options. These findings suggest opportunities to revise PHQ-A items for this demographic, develop visual guides for response options, and incorporate culturally relevant items. Overall, the results underscore the importance of culturally nuanced mental health assessments.