Medical Decision Research Articles

The impact of immune recovery and treatment duration on disseminated histoplasmosis consolidation therapy in AIDS patients

IntroductionThe present study investigated the impact of immune recovery and the duration of antifungal adherence in the consolidation phase of disseminated histoplasmosis (DH) in acquired immune deficiency syndrome (AIDS) patients living in a hyperendemic area in northeastern Brazil. Material and MethodsSixty-nine patients with DH/AIDS, admitted to the São José Hospital between 2010 and 2015, who continued histoplasmosis consolidation therapy at the outpatient clinic were studied. The follow-up duration was at least 24 months. ResultsSixty-eight patients used itraconazole 200–400 mg/day or amphotericin B deoxycholate weekly during the consolidation phase, and six patients relapsed during follow-up. The overall median duration of consolidation antifungal use was 250 days [IQR 101 - 372]. Antifungal withdrawal by medical decision occurred in 41 patients (70.7 %) after a median of 293 days [IQR 128 - 372] of use; 16 patients discontinued by their own decision, with a median of 106 days [IQR 37 - 244] of therapy; three patients had no information available, and nine continued on AF therapy. The median CD4+ T-cell count in the group without relapse was 248 cells/µL [IQR 115–355] within 6 months after admission; conversely, in the relapse group, the median cell count remained below 100 cells/µL. Irregular adherence to highly active antiretroviral therapy (HAART) was the leading risk factor associated with relapse and death (p< 0.01). DiscussionThe regular use of HAART, combined with immune recovery, proved to be highly effective in preventing relapses in DH/AIDS patients, suggesting that long-term antifungal therapy may not be necessary.

An intelligent garment for online fetal well-being monitoring

Considered as one category of wearable systems, intelligent garments represent an opportunity for meeting special needs of various populations on well-being monitoring. In this paper, we present a new wearable system based on an intelligent garment specially designed for online monitoring of fetal movements, which are usually taken as a key indicator for fetal well-being during pregnancy. This intelligent garment is composed of the following components: (1) a garment carefully designed for enhancing pregnant women’s comfort and increasing measured signal quality, (2) a network of sensors integrated into the right positions of the garment, (3) a microcontroller also integrated into the garment, permitting to merge and analyze data measured from these sensors and (4) an Android application connected to the garment, allowing the mother to visualize key information related to her baby’s health. Moreover, in the frame of the wearable system, this garment is connected to both the wearer’s smartphone which makes basic diagnosis on fetal well-being (local decision support system) and a medical expert system on the cloud computing platform with which clinicians can make advanced medical diagnosis. In this wearable system, the techniques of artificial intelligence have been used for extracting relevant features of fetal movements and supporting medical decisions of clinicians. Different from the existing wearable systems, both the electronic/signal issues and textile/garment design have been fully taken into account in the proposed intelligent garment and the capability of accurate fetal movement monitoring has been further enhanced by the well-designed algorithms.

Role of ChatGPT and Gemini in the Urology Field: A Case-Based Study

Introduction The healthcare sector is witnessing a transformation with the advent of artificial intelligence (AI), exemplified by ChatGPT and Gemini AI. These AI systems emulate human conversation and provide accurate medical responses. This study explores their integration into medical decision-making in the urology field. Methods The study presented a collection of 20 medical case scenarios, carefully crafted and revised by a team of authors in the field of urology. Each case was presented to ChatGPT and Gemini in September of 2023, and their responses were recorded and analyzed. Results Both AI tools displayed varying accuracy in diagnoses and management recommendations. ChatGPT failed in identifying congenital penile curvature, while Gemini succeeded. Conversely, ChatGPT excelled in recommending a management plan for renal artery aneurysms. Gemini outperformed in explaining iodinated contrast material toxicity. Both struggled with a bladder prolapse prevention question. Conclusion AI integration in urology is promising but has limitations. AI provides valuable insights but cannot replace human expertise. Research is vital to improve AI's role in urology. Clinicians should view AI suggestions as supplements to their judgment, fostering collaborative healthcare decisions.

Decision Support and Behavioral Health for Reducing High-Dose Opioids in Comorbid Chronic Pain, Depression and Anxiety: Stepped-Wedge Cluster Randomized Trial.

High prevalence of depression or anxiety with opioid use for chronic pain complicates co-management and may influence prescribing behaviors. Compare clinical effectiveness of electronic medical record clinical decision support (EMR-CDS) versus additional behavioral health (BH) care management for reducing rates of high-dose opioid prescriptions. Type 2 effectiveness-implementation hybrid stepped-wedge cluster randomized trial in 35 primary care clinics within a health system in LA, USA. Patients aged 18+ receiving chronic opioid therapy for non-cancer pain with depression or anxiety and matched controls. EMR-CDS included opioid risk mitigation procedures. BH care included cognitive behavioral therapy; depression or anxiety medication adjustments; and case management. Outcomes of interest included difference-in-difference (DID) estimate of changes in probability for prescribing high-dose morphine equivalent daily dose (MEDD ≥50 mg/day and MEDD ≥90), average MEDD, and rates of hospitalization, emergency department use, and opioid risk mitigation. Most participants were female with 3+ pain syndromes. Data analysis included 632 patients. Absolute risk differences for MEDD≥50 and ≥90 decreased post-index compared to pre-index (DID of absolute risk difference [95%CI]: -0.036 [-0.089, 0.016] and -0.029 [-0.060, 0.002], respectively). However, these differences were not statistically significant. The average MEDD decreased at a higher rate for the BH group compared to EMR-CDS only (DID rate ratio [95%CI]: 0.85 [0.77, 0.93]). There were no changes in hospitalization and emergency department utilization. The BH group had higher probabilities of new specialty referrals and prescriptions for naloxone and antidepressants. Incorporation of a multidisciplinary behavioral health care team into primary care did not decrease high-dose prescribing; however, it improved adherence to clinical guideline recommendations for managing chronic opioid therapy for non-cancer pain. ClinicalTrials.gov ID NCT03889418.

A hybrid machine learning-powered intelligent system for enhancing dengue patient safety and care

Dengue fever is a significant global health concern, with millions of cases reported each year, leading to considerable morbidity and mortality. Early diagnosis, patient monitoring and timely intervention are crucial for managing dengue patients. This study proposed a hybrid machine learning-powered intelligent system designed to enhance dengue patients safety and care. Utilizing data provided at enrollment, including platelet, age, white cell, genders, hematocrit and lymphocyte counts, a Weighted K Nearest Neighbor fused Gradient Boosting Decision Tree (WKNN-GBDT) was utilized to forecast the ultimate diagnosis. The study included 50 patients recruited between July to October 2019 and diagnosed with dengue infection. The collected data was preprocessed and features extracted using min max normalization and independent component analysis (ICA). The WKNN-GBDT model had an overallprecision of 0.96%, f1-score of 0.90% accuracy of 0.98%, andrecall of 0.99% in predicting the final diagnosis. As a consequence of seasonality and other variables, model results changed over time. These models might enhance medical decision-making in the field of medical care and offer passive surveillance in dengue affected areas. Considering the unexpected consequences of human-induced climate change and its impact on health, as well as the implications of seasonality and shifting disease prevalence, is crucial.

A novel deer hunting optimized cat boost approach for interactive medical decision-making system

The need for precise and effective medical decision-making mechanisms is driving constant change in the healthcare sector. In this regard, algorithms based on machine learning are becoming prominent due to their potential to help medical practitioners diagnose and cure a wide range of illnesses. All over the world, one significant public health issue is chronic kidney disease (CKD), which requires prompt and precise diagnosis to manage and cure. A new Deer Hunting Optimized CatBoost (DHO-CB) method for CKD seriousness prediction is presented in this research. Our approach takes advantage of CatBoost's built-in flexibility and resilience by incorporating an optimization framework inspired by deer hunting. This framework dynamically modifies model parameters to traverse intricate medical data environments. The standard CKD dataset is used to verify the suggested DHO-CB approach, and Python is the primary language used in this study. In terms of kappa coefficient (92.6%), f1-score metrics (97.7%), accuracy (97.2%), and specificity (95.5%), the effectiveness of the suggested DHO-CB approach is examined. Based on this analyzed data, our suggested DHO-CB approach outperforms other current approaches in CKD prediction as well as severity assessment. The suggested DHO-CB offers a viable path toward the creation of complex yet approachable medical decision support systems, which will promote increased accuracy and effectiveness in the identification and treatment of CKDs.

The Patient's Perspective: Does It Align with Dialysis Adequacy?

The concept of patient-centered care puts the individual's health needs and desired health outcomes as the driving forces behind medical decision-making and quality assessment in the healthcare system. Patients with end-stage kidney disease (ESKD) treated by hemodialysis require frequent encounters with the dialysis facility in order to survive. Therefore, their satisfaction with care and perceived patient experience are important aspects that might impact their adherence to the care regimen. The aim of this study was to evaluate patient satisfaction and its association with perceived patient experience and objective clinical quality parameters, across three hemodialysis clinics. A prospective cohort study analyzed the data of 126 patients with ESKD receiving chronic hemodialysis over 9 months in 3 different care facilities. Sociodemographic characteristics, medical history, treatment details, and dialysis adequacy (measures as STDKt/V) were collected. Perceived quality of care, patient satisfaction and clinical outcomes were assessed. Patients differed significantly between sites by age, diabetes status and biochemical parameters. Satisfaction scores varied significantly for 12/14 survey questions and at the site-level, with Site 2 scoring highest. Overall satisfaction did not correlate with Kt/V. At Site 1, a moderate negative correlation was found between satisfaction and Kt/V. Kt/V correlated positively with age but inversely with satisfaction. Hospitalization rates were similar regardless of satisfaction. Mortality trended lower in the highest Kt/V quartile. Achieving clinical quality while optimizing patient satisfaction requires multifactorial approaches tailored to the unique population of the hemodialysis facility. Further research is needed to fully understand factors influencing satisfaction and perceived quality.

Exploring the Relationship Between Psychological Constructs and Decision-Making Preferences in Psychiatric Outpatients.

This study aimed to elucidate the relationships among health locus of control, psychological reactance, attitudes toward psychiatric treatment, and patient decision-making preferences within a psychiatric outpatient population. A total of 200 consecutive psychiatric outpatients from a community mental health center in Tenerife, Spain, were approached for participation between September 2023 and March 2024. Of these, 151 patients consented to participate in this cross-sectional study. Participants were selected based on their willingness to participate and were provided with informed consent forms. Data were collected using the Patient's Health Belief Questionnaire on Psychiatric Treatment (PHBQPT) and the Control Preferences Scale (CPS). The PHBQPT evaluates health beliefs impacting adherence to psychiatric treatment, while the CPS assesses the preferred level of involvement in medical decision-making. Sociodemographic data were also collected to contextualize the findings. Significant correlations were found between patients' control preferences and their attitudes towards medication, compliance with psychiatric advice, and perceptions of treatment control. A collaborative control preference was notably associated with positive attitudes toward medication and trust in the psychiatrist. These findings suggest that tailored treatment approaches prioritizing patient involvement could enhance adherence and outcomes. The study underscores the importance of considering psychological constructs in psychiatric care to foster a holistic, patient-centered approach. Recognizing and integrating patients' control preferences, attitudes towards medication, and psychological reactance can improve the therapeutic relationship and treatment adherence. Future research should explore longitudinal and interventional studies to further understand the impact of aligning treatment approaches with patient preferences and psychological profiles.

The Accuracy of Artificial Intelligence ChatGPT in Oncology Examination Questions

The aim of this study is to assess the accuracy of Chat Generative Pretrained Transformer (ChatGPT) in response to oncology examination questions in the setting of one-shot learning. Consecutive national radiation oncology in-service multiple-choice examinations were collected and inputted into ChatGPT 4o and ChatGPT 3.5 to determine ChatGPT’s answers. ChatGPT’s answers were then compared with the answer keys to determine whether ChatGPT correctly or incorrectly answered each question and to determine if improvements in responses were seen with the newer ChatGPT version. A total of 600 consecutive questions were inputted into ChatGPT. ChatGPT 4o answered 72.2% questions correctly, whereas 3.5 answered 53.8% questions correctly. There was a significant difference in performance by question category (P < .01). ChatGPT performed poorer with respect to knowledge of landmark studies and treatment recommendations and planning. ChatGPT is a promising technology, with the latest version showing marked improvement. Although it still has limitations, with further evolution, it may be considered a reliable resource for medical training and decision making in the oncology space.

Этические и культурные вызовы внедрения искусственного интеллекта в медицинских практиках: мультикультуральный анализ

The use of AI in medical practice offers a number of significant and visible advantages. The problem of the presented research is relevant due to the growing integration of technologies into healthcare. The purpose of this study is to analyze the ethical and cultural challenges associated with integration of artificial intelligence (AI) in medical practice in various cultural contexts. The main objective of the study is to identify specific problems and suggest possible solutions to ensure effective and justifiable use of AI. To achieve this goal, a literary review, a case study, an expert interview, and a questionnaire were used. The main areas of ethical and cultural challenges include the issues of confidentiality and data protection; culturally specific attitudes towards automation of medical decision-making; the impact of algorithm bias on the diagnosis and treatment of various ethnic groups; ethical dilemmas related to access and fairness in the use of medical AI systems. The study highlights the need to develop ethical standards for the use of AI in medicine that will take into account cultural differences.

A health-conformant reading of the GDPR's right not to be subject to automated decision-making.

As the use of Artificial Intelligence (AI) technologies in healthcare is expanding, patients in the European Union (EU) are increasingly subjected to automated medical decision-making. This development poses challenges to the protection of patients' rights. A specific patients' right not to be subject to automated medical decision-making is not considered part of the traditional portfolio of patients' rights. The EU AI Act also does not contain such a right. The General Data Protection Regulation (GDPR) does, however, provide for the right 'not to be subject to a decision based solely on automated processing' in Article 22. At the same time, this provision has been severely critiqued in legal scholarship because of its lack of practical effectiveness. However, in December 2023, the Court of Justice of the EU first provided an interpretation of this right in C-634/21 (SCHUFA)-although in the context of credit scoring. Against this background, this article provides a critical analysis of the application of Article 22 GDPR to the medical context. The objective is to evaluate whether Article 22 GDPR may provide patients with the right to refuse automated medical decision-making. It proposes a health-conformant reading to strengthen patients' rights in the EU.

Analytical performance evaluation of a new integrated clinical chemistry and immunoassay analyzer

BackgroundClinical laboratories perform a wide range of tests that are used by healthcare professionals to guide medical decision making. Use of automated analyzers in the clinical laboratory can improve patient care by not only reducing the turn-around-time (TAT) of results but also improving accuracy of the reported results by reducing human error. The aim of this study was to evaluate the performance characteristics of a new automated laboratory instrument, the Atellica® CI Analyzer, Model 1900, over a 3-month period in a European laboratory setting. MethodsAnalytical performance of 17 analytes (13 chemistry and four immunochemistry) was assessed by evaluating repeatability and within-laboratory precision using anonymized remnant serum samples. Method comparison studies were performed on the Atellica CI Analyzer and the Roche cobas® 6000. ResultsExcellent precision was observed with coefficients of variation (CVs) less than 2 % for repeatability and less than 3 % within-laboratory imprecision for most analytes. Comparison of select assays with the cobas 6000 system resulted in correlation coefficients ranging from 0.980 to 1.000. ConclusionThis is the first reported evaluation of the Atellica CI Analyzer in a clinical laboratory setting. The strong analytical performance of the Atellica CI Analyzer demonstrates that this instrument is suitable for routine clinical use.

Usability and Feasibility of the Antipsychotic Medication Decision Aid in a Community Program for First-Episode Psychosis.

Although antipsychotic medications are considered first-line treatment for psychosis, rates of discontinuation and nonadherence are high, and debate persists about their use. This pilot study aimed to explore the usability, feasibility, and potential impact of a shared decision making (SDM) intervention, the Antipsychotic Medication Decision Aid (APM-DA), for decisions about use of antipsychotic medications. A pilot randomized controlled trial was conducted with 17 participants in a first-episode psychosis program. Nine participants received the APM-DA, and eight received usual care. After their appointments, intervention group participants had less decisional conflict and greater satisfaction with decisions than control group participants had. Use of the APM-DA did not increase appointment length. Comparison of the intervention outcomes with the control outcomes was limited because of the small sample. The results support the feasibility and usability of an SDM process via the use of the APM-DA in routine community psychosis care.

Evaluation of a Diazo-Based Point-Of-Care Bilirubin Assay careSTART S1 Total Bilirubin Strip.

Neonatal jaundice (NNJ) affects a significant proportion of newborns globally, with an increased burden in low-resource settings. Effective health risk management of NNJ is hindered, particularly in resource-constrained environments, where early detection and treatment are challenging. The careSTART S1 Total Bilirubin Strip, a point-of-care testing (POCT) device based on a diazo-method, offers a potential solution by enabling onsite bilirubin measurement, thus, addressing the gap in early NNJ detection and management. The current study evaluated the analytical performance of the careSTART S1 Total Bilirubin Strip for precision, linearity, method comparison, and lot-to-lot consistency following CLSI guidelines. For method comparison, 105 residual EDTA whole blood samples were analyzed with the careSTART S1 Total Bilirubin Strip and compared with reference measurements from the Roche Cobas c702 analyzer. Additionally, statistical analyses, including Passing-Bablok regression and Bland-Altman plots, were performed. The careSTART S1 Total Bilirubin Strip showed allowable (<10%) within-laboratory imprecision of 2.5%-3.6% across all levels and demonstrated linearity over the range of 4.16-439.3 μmol/L. Method comparison revealed a constant negative bias with a mean bias -4.19 μmol/L. However, the 95% confidence interval (-7.10 to -1.28 μmol/L) of the bias is covered by the prespecified allowable bias of 8.3%, at medical decision point. Lot-to-lot variation ranged from 0.14%-6.49%, and was within the acceptable critical difference of 8.3%. The careSTART S1 Total Bilirubin Strip provided accurate and reliable bilirubin measurements that could contribute to neonatal care in settings lacking central laboratory facilities.

Limited health literacy does not adversely affect compliance with postoperative restrictions, 90-day emergency department return, or opioid use following shoulder arthroscopy

BackgroundLimited health literacy can negatively impact how patients process medical information, make medical decisions, and navigate the healthcare system. The literature with regards to health literacy and its impact on both postoperative compliance and healthcare utilization remains scant. MethodsWe retrospectively analyzed the records for patients who underwent elective shoulder arthroscopy with a minimum 90-day follow-up at a single academic institution. Demographic data including age, gender, prior ipsilateral shoulder arthroscopy, body mass index (BMI) and age-adjusted Charlson Comorbidity Index (ACCI) were collected. A validated 9-item Literacy in Musculoskeletal Problems (LiMP) questionnaire to assess musculoskeletal health literacy (MHL) was administered preoperatively. Postoperative compliance with therapy and surgeon-directed immobilization restrictions, 90-day return to ED, and the number of opioid prescriptions filled within 3 months postoperatively was recorded. ResultsThere were 252 cases included in this study. Seventy-seven (31%) patients demonstrated adequate MHL (AMHL). On multivariable analysis, limited MHL (LMHL) was not significantly associated with 90-day postoperative ED return, compliance with postoperative surgeon instructions regarding shoulder motion or therapy restrictions, or obtaining ≥ 2 postoperative opioid prescriptions. ConclusionsLMHL is highly prevalent amongst patients undergoing elective shoulder arthroscopy. The lack of association between LMHL and postoperative compliance, 90-day ED return, or filling ≥ 2 postoperative opioid prescriptions suggests that further research is needed to identify more relevant modifiable risk factors that could reduce these negative clinical outcomes and healthcare utilization patterns.

Clinical Decision Support Systems and Where to Apply Them: A Systematic Review

Background: Clinical decision support systems (CDSSs) provide healthcare providers with suggestions by evaluating the situation and organizing the decision-making process to reduce their mental workload so they can make decisions more effortlessly and safely in emergencies and critical-safety situations. This article reviews the applications of CDSSs and the benefits that these systems bring to healthcare settings. Method: Scopus, PubMed/Medline, and Web of Science were selected to search high-quality CDSS-related journal articles published in English from 2000 to 2022. Two reviewers, under the supervision of the third reviewer, screened and analyzed the search results based on the inclusion criteria. Results: The required information was extracted from 78 included articles, and it was found that CDSSs have been exploited for four primary purposes, namely, accurate diagnoses, early prevention of diseases, management of clinical/medical processes, and prescriptions. Conclusion: CDSS has many benefits for each of the four applications, the most important of which is to improve patient safety. Generally, the results showed that DSSs in healthcare can positively affect medical decisions by reducing possible errors and putting forward specific medical suggestions for each patient.

Shared decision-making implementation status among dermatologists engaging in medical esthetics: a cross-sectional study in China.

Shared decision-making (SDM) is a collaborative process in which patients and healthcare providers jointly make a medical decision. This cross-sectional study aimed to identify the implementation status of shared decision-making among dermatologists engaging in medical esthetics in China and to identify factors associated with the good practice of SDM among them. From January to June 2023, a total of 1,287 dermatologists engaging in medical esthetics in China were recruited and completed the online interviews about their implementation of SDM based on the Shared Decision-Making Questionnaire for Doctors (SDM-Q-Doc). Logistic regression was used to calculate the odds ratio (OR) and 95% confidence interval (CI) to explore factors associated with the higher SDM score achievement among dermatologists with medical esthetic practice. The median value of the total SDM score was 39, and 48% (621/1278) of dermatologists with medical esthetic practice achieved at least 40 out of 45 scores. Logistic regression indicated that dermatologists aged 40-49 or ≥ 50 years and those engaging in medical esthetic practice for ≥5 years were more likely to achieve at least 40 out of 45 scores compared to dermatologists aged <30 years with less than 5 years of medical esthetic practice. The ORs were 1.82 (95% CI: 1.13-3.12), 1.94 (95% CI: 1.13-3.61), and 1.76 (95% CI: 1.34-2.31), respectively. The SDM implementation level among Chinese dermatologists engaging in medical esthetics is high, especially for those who are older age and have more years of practice. Hence, it is highly recommended to promote and enhance SDM practice among younger dermatologists engaging in medical esthetics with less working experience.

Emerging Artificial Intelligence In Therapeutic Agreements With A Medicolegal Approach

Introduction: Medical services use artificial intelligence for operating to help and even transform the healthcare system. AI innovators have developed tools to improve clinical care processes, advance medical research, and increase efficiency in medical services.Purposes of the Research: The purpose of this article is to analyse the legal validity of therapeutic agreements using AI in medical field.Methods of the Research: The research method used is a normative juridical research type with an analytical approach.Results of the Research: The results show that medical services are complex and closely related systems, and always contain risks, so they must be carried out with great care. Legal provisions governing the use of AI in therapeutic agreements with a medicolegal approach must be able to evaluate and ensure the safety and accuracy of medical decisions made by AI "thinking algorithms".

Identifying Strategies to Improve Shared Decision Making for Pregnant Patients' Decisions about Prenatal Genetic Screens and Diagnostic Tests.

Prenatal genetic screens and diagnostic tests are vital components of prenatal care. The first prenatal visit is a critical time in the decision-making process when patients decide whether to use these tests in addition to address a series of other essential prenatal care aspects. We conducted this study to examine the role of a shared decision-making (SDM) instrument to support these discussions. We conducted a cluster randomized controlled trial of patients allocated to an SDM tool or usual care at their first prenatal visit. Participants completed a baseline survey to measure decision-making needs and preferences. Direct observation was conducted and analyzed using the OPTION scale to measure SDM during prenatal genetic testing discussions. Levels of SDM were similar across groups (P = 0.081). The highest levels of SDM were observed during screening test discussions (NEST 2.4 ± 0.9 v. control 2.6 ± 1.0). Lowest levels were observed in discussions about patients' preference for risk versus diagnostic information (NEST 1.0 ± 1.1 v. control 1.2 ± 1.3). Study findings demonstrate the need for targeted patient-focused and provider-focused efforts to improve SDM to enhance patients' informed decision making about these options. Importantly, patients' baseline knowledge and attitudes need to be considered given that patients with less knowledge may need more carefully crafted communication. Choices about whether, when, and how to use prenatal genetic tests are highly preference-based decisions, with patients' baseline attitudes about these options as a major driver in health care discussions.The decision-making process is also shaped by patient preferences regarding a shared or informed decision-making process for medical decisions that are highly personal and have significant ramifications for obstetric outcomes.There is a need to develop targeted efforts to improve decision making and enhance patients' ability to make informed decisions about prenatal genetic tests in early pregnancy.

Interpreting PPV and NPV of Diagnostic Tests with Uncertain Prevalence.

Medical decision-making is often uncertain. The positive predictive value (PPV) and negative predictive value (NPV) are conditional probabilities characterizing diagnostic tests and assessing diagnostic interventions in clinical medicine and epidemiology. The PPV is the probability that a patient has a specified disease, given a positive test result for that disease. The NPV is the probability that a patient does not have the disease, given a negative test result for that disease. Both values depend on disease incidence or prevalence, which may be highly uncertain for unfamiliar diseases, epidemics, etc. Probability distributions for this uncertainty are usually unavailable. We develop a non-probabilistic method for interpreting PPV and NPV with uncertain prevalence. Uncertainty in PPV and NPV is managed with the non-probabilistic concept of robustness in info-gap theory. Robustness of PPV or NPV estimates is the greatest uncertainty (in prevalence) at which the estimate's error is acceptable. Four properties are demonstrated. Zeroing: best estimates of PPV or NPV have no robustness to uncertain prevalence; best estimates are unreliable for interpreting diagnostic tests. Trade-off: robustness increases as error increases; this trade-off identifies robustly reliable error in PPV or NPV. Preference reversal: sometimes sub-optimal PPV or NPV estimates are more robust to uncertain incidence or prevalence than optimal estimates, motivating reversal of preference from the putative optimum to the sub-optimal estimate. Trade-off between specificity and robustness to uncertainty: the robustness increases as test-specificity decreases. These four properties underlie the interpretation of PPV and NPV. The PPV and NPV assess diagnostic tests, but are sensitive to lack of knowledge that generates non-probabilistic uncertain prevalence and must be supplemented with robustness analysis. When uncertainties abound, as with unfamiliar diseases, assessing robustness is critical to avoiding erroneous decisions.