ABSTRACT
Objectives: Chronic obstructive pulmonary disease (COPD) disproportionately affects low- and middle-income countries. Health systems are ill prepared to manage the increase in COPD cases. Methods: We performed a pilot effectiveness-implementation randomized field trial of a community health worker (CHW)-supported, 1-year self-management intervention in individuals with COPD grades B-D. The study took place in low-resource settings of Nepal, Peru, and Uganda. The primary outcome was the St. George's Respiratory Questionnaire (SGRQ) score at 1 year. We evaluated differences in moderate to severe exacerbations, all-cause hospitalizations, and the EuroQol score (EQ-5D-3 L) at 12 months. Measurements and Main Results: We randomly assigned 239 participants (119 control arm, 120 intervention arm) with grades B-D COPD to a multicomponent, CHW-supported intervention or standard of care and COPD education. Twenty-five participants (21%) died or were lost to follow-up in the control arm compared with 11 (9%) in the intervention arm. At 12 months, there was no difference in mean total SGRQ score between the intervention and control arms (34.7 vs. 34.0 points; adjusted mean difference, 1.0; 95% confidence interval, -4.2, 6.1; P = 0.71). The intervention arm had a higher proportion of hospitalizations than the control arm (10% vs. 5.2%; adjusted odds ratio, 2.2; 95% confidence interval, 0.8, 7.5; P = 0.15) at 12 months. Conclusions: A CHW-based intervention to support self-management of acute exacerbations of COPD in three resource-poor settings did not result in differences in SGRQ scores at 1 year. Fidelity was high, and intervention engagement was moderate. Although these results cannot differentiate between a failed intervention or implementation, they nonetheless suggest that we need to revisit our strategy. Clinical trial registered with www.clinicaltrials.gov (NCT03359915).
Subject(s)
Pulmonary Disease, Chronic Obstructive , Self-Management , Humans , Developing Countries , Pilot Projects , Hospitalization , Pulmonary Disease, Chronic Obstructive/therapy , Quality of LifeABSTRACT
Importance: Most of the global morbidity and mortality in chronic obstructive pulmonary disease (COPD) occurs in low- and middle-income countries (LMICs), with significant economic effects. Objective: To assess the discriminative accuracy of 3 instruments using questionnaires and peak expiratory flow (PEF) to screen for COPD in 3 LMIC settings. Design, Setting, and Participants: A cross-sectional analysis of discriminative accuracy, conducted between January 2018 and March 2020 in semiurban Bhaktapur, Nepal; urban Lima, Peru; and rural Nakaseke, Uganda, using a random age- and sex-stratified sample of the population 40 years or older. Exposures: Three screening tools, the COPD Assessment in Primary Care to Identify Undiagnosed Respiratory Disease and Exacerbation Risk (CAPTURE; range, 0-6; high risk indicated by a score of 5 or more or score 2-5 with low PEF [<250 L/min for females and <350 L/min for males]), the COPD in LMICs Assessment questionnaire (COLA-6; range, 0-5; high risk indicated by a score of 4 or more), and the Lung Function Questionnaire (LFQ; range, 0-25; high risk indicated by a score of 18 or less) were assessed against a reference standard diagnosis of COPD using quality-assured postbronchodilator spirometry. CAPTURE and COLA-6 include a measure of PEF. Main Outcomes and Measures: The primary outcome was discriminative accuracy of the tools in identifying COPD as measured by area under receiver operating characteristic curves (AUCs) with 95% CIs. Secondary outcomes included sensitivity, specificity, positive predictive value, and negative predictive value. Results: Among 10â¯709 adults who consented to participate in the study (mean age, 56.3 years (SD, 11.7); 50% female), 35% had ever smoked, and 30% were currently exposed to biomass smoke. The unweighted prevalence of COPD at the 3 sites was 18.2% (642/3534 participants) in Nepal, 2.7% (97/3550) in Peru, and 7.4% (264/3580) in Uganda. Among 1000 COPD cases, 49.3% had clinically important disease (Global Initiative for Chronic Obstructive Lung Disease classification B-D), 16.4% had severe or very severe airflow obstruction (forced expiratory volume in 1 second <50% predicted), and 95.3% of cases were previously undiagnosed. The AUC for the screening instruments ranged from 0.717 (95% CI, 0.677-0.774) for LFQ in Peru to 0.791 (95% CI, 0.770-0.809) for COLA-6 in Nepal. The sensitivity ranged from 34.8% (95% CI, 25.3%-45.2%) for COLA-6 in Nepal to 64.2% (95% CI, 60.3%-67.9%) for CAPTURE in Nepal. The mean time to administer the instruments was 7.6 minutes (SD 1.11), and data completeness was 99.5%. Conclusions and Relevance: This study demonstrated that screening instruments for COPD were feasible to administer in 3 low- and middle-income settings. Further research is needed to assess instrument performance in other low- and middle-income settings and to determine whether implementation is associated with improved clinical outcomes.
Subject(s)
Developing Countries , Peak Expiratory Flow Rate , Pulmonary Disease, Chronic Obstructive/diagnosis , Surveys and Questionnaires , Adult , Airway Obstruction/epidemiology , Cross-Sectional Studies , Feasibility Studies , Female , Forced Expiratory Volume , Humans , Male , Middle Aged , Nepal/epidemiology , Peru/epidemiology , Predictive Value of Tests , Prevalence , Pulmonary Disease, Chronic Obstructive/classification , Pulmonary Disease, Chronic Obstructive/epidemiology , Pulmonary Disease, Chronic Obstructive/physiopathology , ROC Curve , Reference Standards , Sensitivity and Specificity , Smoking/epidemiology , Spirometry/methods , Tobacco Smoke Pollution/statistics & numerical data , Uganda/epidemiologyABSTRACT
BACKGROUND: Sparse relative effectiveness evidence is a frequent problem in Health Technology Assessment (HTA). Where evidence directly pertaining to the decision problem is sparse, it may be feasible to expand the evidence-base to include studies that relate to the decision problem only indirectly: for instance, when there is no evidence on a comparator, evidence on other treatments of the same molecular class could be used; similarly, a decision on children may borrow-strength from evidence on adults. Usually, in HTA, such indirect evidence is either included by ignoring any differences ('lumping') or not included at all ('splitting'). However, a range of more sophisticated methods exists, primarily in the biostatistics literature. The objective of this study is to identify and classify the breadth of the available information-sharing methods. METHODS: Forwards and backwards citation-mining techniques were used on a set of seminal papers on the topic of information-sharing. Papers were included if they specified (network) meta-analytic methods for combining information from distinct populations, interventions, outcomes or study-designs. RESULTS: Overall, 89 papers were included. A plethora of evidence synthesis methods have been used for information-sharing. Most papers (n=79) described methods that shared information on relative treatment effects. Amongst these, there was a strong emphasis on methods for information-sharing across multiple outcomes (n=42) and treatments (n=25), with fewer papers focusing on study-designs (n=23) or populations (n=8). We categorise and discuss the methods under four 'core' relationships of information-sharing: functional, exchangeability-based, prior-based and multivariate relationships, and explain the assumptions made within each of these core approaches. CONCLUSIONS: This study highlights the range of information-sharing methods available. These methods often impose more moderate assumptions than lumping or splitting. Hence, the degree of information-sharing that they impose could potentially be considered more appropriate. Our identification of four 'core' methods of information-sharing allows for an improved understanding of the assumptions underpinning the different methods. Further research is required to understand how the methods differ in terms of the strength of sharing they impose and the implications of this for health care decisions.
Subject(s)
Information Dissemination , Technology Assessment, Biomedical , Adult , Child , HumansABSTRACT
BACKGROUND: Empirical evidence supporting the cost-effectiveness estimates of particular health care technologies may be limited, or it may even be missing entirely. In these situations, additional information, often in the form of expert judgments, is needed to reach a decision. There are formal methods to quantify experts' beliefs, termed as structured expert elicitation (SEE), but only limited research is available in support of methodological choices. Perhaps as a consequence, the use of SEE in the context of cost-effectiveness modelling is limited. OBJECTIVES: This article reviews applications of SEE in cost-effectiveness modelling with the aim of summarizing the basis for methodological choices made in each application and recording the difficulties and challenges reported by the authors in the design, conduct, and analyses. METHODS: The methods used in each application were extracted along with the criteria used to support methodological and practical choices and any issues or challenges discussed in the text. Issues and challenges were extracted using an open field, and then categorised and grouped for reporting. RESULTS: The review demonstrates considerable heterogeneity in methods used, and authors acknowledge great methodological uncertainty in justifying their choices. Specificities of the context area emerging as potentially important in determining further methodological research in elicitation are between- expert variation and its interpretation, the fact that substantive experts in the area may not be trained in quantitative subjects, that judgments are often needed on various parameter types, the need for some form of assessment of validity, and the need for more integration with behavioural research to devise relevant debiasing strategies. CONCLUSIONS: This review of experiences of SEE highlights a number of specificities/constraints that can shape the development of guidance and target future research efforts in this area.
Subject(s)
Comparative Effectiveness Research/organization & administration , Cost-Benefit Analysis , Consensus , Decision Support Techniques , Humans , Models, Economic , Probability , Research DesignABSTRACT
BACKGROUND: Venous leg ulcers are open skin wounds on the lower leg which can be slow to heal, and are both painful and costly. The point prevalence of open venous leg ulcers in the UK is about 3 cases per 10,000 people, and many people experience recurrent episodes of prolonged ulceration. First-line treatment for venous leg ulcers is compression therapy, but a wide range of dressings and topical treatments are also used. This diversity of treatments makes evidence-based decision-making challenging, and a clear and current overview of all the evidence is required. This review is a network meta-analysis (NMA) which assesses the probability of complete ulcer healing associated with alternative dressings and topical agents. OBJECTIVES: To assess the effects of (1) dressings and (2) topical agents for healing venous leg ulcers in any care setting and to rank treatments in order of effectiveness, with assessment of uncertainty and evidence quality. SEARCH METHODS: In March 2017 we searched the Cochrane Wounds Specialised Register; the Cochrane Central Register of Controlled Trials (CENTRAL); Ovid MEDLINE; Ovid MEDLINE (In-Process & Other Non-Indexed Citations); Ovid Embase and EBSCO CINAHL Plus. We also scanned reference lists of relevant included studies as well as reviews, meta-analyses, guidelines and health technology reports to identify additional studies. There were no restrictions with respect to language, date of publication or study setting. We updated this search in March 2018; as a result several studies are awaiting classification. SELECTION CRITERIA: We included published or unpublished randomised controlled trials (RCTs) that enrolled adults with venous leg ulcers and compared the effects of at least one of the following interventions with any other intervention in the treatment of venous leg ulcers: any dressing, or any topical agent applied directly to an open venous leg ulcer and left in situ. We excluded from this review dressings attached to external devices such as negative pressure wound therapies, skin grafts, growth factors and other biological agents, larval therapy and treatments such as laser, heat or ultrasound. Studies were required to report complete wound healing to be eligible. DATA COLLECTION AND ANALYSIS: Two review authors independently performed study selection, 'Risk of bias' assessment and data extraction. We conducted this NMA using frequentist meta-regression methods for the efficacy outcome; the probability of complete healing. We assumed that treatment effects were similar within dressings classes (e.g. hydrocolloid, foam). We present estimates of effect with their 95% confidence intervals (CIs) for individual treatments focusing on comparisons with widely used dressing classes, and we report ranking probabilities for each intervention (probability of being the best, second best, etc treatment). We assessed the certainty (quality) of the body of evidence using GRADE for each network comparison and for the network as whole. MAIN RESULTS: We included 78 RCTs (7014 participants) in this review. Of these, 59 studies (5156 participants, 25 different interventions) were included in the NMA; resulting in 40 direct contrasts which informed 300 mixed-treatment contrasts.The evidence for the network as a whole was of low certainty. This judgement was based on the sparsity of the network leading to imprecision and the general high risk of bias in the included studies. Sensitivity analyses also demonstrated instability in key aspects of the network and results are reported for the extended sensitivity analysis. Evidence for individual contrasts was mainly judged to be low or very low certainty.The uncertainty was perpetuated when the results were considered by ranking the treatments in terms of the probability that they were the most effective for ulcer healing, with many treatments having similar, low, probabilities of being the best treatment. The two most highly-ranked treatments both had more than 50% probability of being the best (sucralfate and silver dressings). However, the data for sucralfate was from one small study, which means that this finding should be interpreted with caution. When exploring the data for silver and sucralfate compared with widely-used dressing classes, there was some evidence that silver dressings may increase the probability of venous leg ulcer healing, compared with nonadherent dressings: RR 2.43, 95% CI 1.58 to 3.74 (moderate-certainty evidence in the context of a low-certainty network). For all other combinations of these five interventions it was unclear whether the intervention increased the probability of healing; in each case this was low- or very low-certainty evidence as a consequence of one or more of imprecision, risk of bias and inconsistency. AUTHORS' CONCLUSIONS: More research is needed to determine whether particular dressings or topical agents improve the probability of healing of venous leg ulcers. However, the NMA is uninformative regarding which interventions might best be included in a large trial, largely because of the low certainty of the whole network and of individual comparisons.The results of this NMA focus exclusively on complete healing; whilst this is of key importance to people living with venous leg ulcers, clinicians may wish to take into account other patient-important outcomes and factors such as patient preference and cost.
Subject(s)
Anti-Ulcer Agents/therapeutic use , Bandages , Silver/therapeutic use , Sucralfate/therapeutic use , Varicose Ulcer/therapy , Wound Healing , Aged , Aged, 80 and over , Bandages, Hydrocolloid , Female , Humans , Male , Middle Aged , Network Meta-Analysis , Randomized Controlled Trials as Topic , Sensitivity and SpecificityABSTRACT
BACKGROUND: Pressure ulcers, also known as bedsores, decubitus ulcers and pressure injuries, are localised areas of injury to the skin or the underlying tissue, or both. Dressings are widely used to treat pressure ulcers and promote healing, and there are many options to choose from including alginate, hydrocolloid and protease-modulating dressings. Topical agents have also been used as alternatives to dressings in order to promote healing.A clear and current overview of all the evidence is required to facilitate decision-making regarding the use of dressings or topical agents for the treatment of pressure ulcers. Such a review would ideally help people with pressure ulcers and health professionals assess the best treatment options. This review is a network meta-analysis (NMA) which assesses the probability of complete ulcer healing associated with alternative dressings and topical agents. OBJECTIVES: To assess the effects of dressings and topical agents for healing pressure ulcers in any care setting. We aimed to examine this evidence base as a whole, determining probabilities that each treatment is the best, with full assessment of uncertainty and evidence quality. SEARCH METHODS: In July 2016 we searched the Cochrane Wounds Specialised Register; the Cochrane Central Register of Controlled Trials (CENTRAL); Ovid MEDLINE; Ovid MEDLINE (In-Process & Other Non-Indexed Citations); Ovid Embase and EBSCO CINAHL Plus. We also searched clinical trials registries for ongoing and unpublished studies, and scanned reference lists of relevant included studies as well as reviews, meta-analyses, guidelines and health technology reports to identify additional studies. There were no restrictions with respect to language, date of publication or study setting. SELECTION CRITERIA: Published or unpublished randomised controlled trials (RCTs) comparing the effects of at least one of the following interventions with any other intervention in the treatment of pressure ulcers (Stage 2 or above): any dressing, or any topical agent applied directly to an open pressure ulcer and left in situ. We excluded from this review dressings attached to external devices such as negative pressure wound therapies, skin grafts, growth factor treatments, platelet gels and larval therapy. DATA COLLECTION AND ANALYSIS: Two review authors independently performed study selection, risk of bias assessment and data extraction. We conducted network meta-analysis using frequentist mega-regression methods for the efficacy outcome, probability of complete healing. We modelled the relative effectiveness of any two treatments as a function of each treatment relative to the reference treatment (saline gauze). We assumed that treatment effects were similar within dressings classes (e.g. hydrocolloid, foam). We present estimates of effect with their 95% confidence intervals for individual treatments compared with every other, and we report ranking probabilities for each intervention (probability of being the best, second best, etc treatment). We assessed the certainty (quality) of the body of evidence using GRADE for each network comparison and for the network as whole. MAIN RESULTS: We included 51 studies (2947 participants) in this review and carried out NMA in a network of linked interventions for the sole outcome of probability of complete healing. The network included 21 different interventions (13 dressings, 6 topical agents and 2 supplementary linking interventions) and was informed by 39 studies in 2127 participants, of whom 783 had completely healed wounds.We judged the network to be sparse: overall, there were relatively few participants, with few events, both for the number of interventions and the number of mixed treatment contrasts; most studies were small or very small. The consequence of this sparseness is high imprecision in the evidence, and this, coupled with the (mainly) high risk of bias in the studies informing the network, means that we judged the vast majority of the evidence to be of low or very low certainty. We have no confidence in the findings regarding the rank order of interventions in this review (very low-certainty evidence), but we report here a summary of results for some comparisons of interventions compared with saline gauze. We present here only the findings from evidence which we did not consider to be very low certainty, but these reported results should still be interpreted in the context of the very low certainty of the network as a whole.It is not clear whether regimens involving protease-modulating dressings increase the probability of pressure ulcer healing compared with saline gauze (risk ratio (RR) 1.65, 95% confidence interval (CI) 0.92 to 2.94) (moderate-certainty evidence: low risk of bias, downgraded for imprecision). This risk ratio of 1.65 corresponds to an absolute difference of 102 more people healed with protease modulating dressings per 1000 people treated than with saline gauze alone (95% CI 13 fewer to 302 more). It is unclear whether the following interventions increase the probability of healing compared with saline gauze (low-certainty evidence): collagenase ointment (RR 2.12, 95% CI 1.06 to 4.22); foam dressings (RR 1.52, 95% CI 1.03 to 2.26); basic wound contact dressings (RR 1.30, 95% CI 0.65 to 2.58) and polyvinylpyrrolidone plus zinc oxide (RR 1.31, 95% CI 0.37 to 4.62); the latter two interventions both had confidence intervals consistent with both a clinically important benefit and a clinically important harm, and the former two interventions each had high risk of bias as well as imprecision. AUTHORS' CONCLUSIONS: A network meta-analysis (NMA) of data from 39 studies (evaluating 21 dressings and topical agents for pressure ulcers) is sparse and the evidence is of low or very low certainty (due mainly to risk of bias and imprecision). Consequently we are unable to determine which dressings or topical agents are the most likely to heal pressure ulcers, and it is generally unclear whether the treatments examined are more effective than saline gauze.More research is needed to determine whether particular dressings or topical agents improve the probability of healing of pressure ulcers. The NMA is uninformative regarding which interventions might best be included in a large trial, and it may be that research is directed towards prevention, leaving clinicians to decide which treatment to use on the basis of wound symptoms, clinical experience, patient preference and cost.
Subject(s)
Bandages , Dermatologic Agents/therapeutic use , Pressure Ulcer/therapy , Wound Healing , Alginates/therapeutic use , Bandages, Hydrocolloid , Collagenases/therapeutic use , Egg White , Gels/therapeutic use , Glucuronic Acid/therapeutic use , Hexuronic Acids/therapeutic use , Humans , Network Meta-Analysis , Ointments/therapeutic use , Pharmaceutic Aids/therapeutic use , Phenytoin/therapeutic use , Povidone/therapeutic use , Randomized Controlled Trials as Topic , Zinc Oxide/therapeutic useABSTRACT
BACKGROUND: Drawbacks exist with the standard treatment (four-layer compression bandages) for venous leg ulcers. We have therefore compared the clinical effectiveness and cost-effectiveness of two-layer compression hosiery with the four-layer bandage for the treatment of such ulcers. METHODS: We undertook this pragmatic, open, randomised controlled trial with two parallel groups in 34 centres in England and Northern Ireland. The centres were community nurse teams or services, family doctor practices, leg ulcer clinics, tissue viability clinics or services, and wound clinics. Participants were aged 18 years or older with a venous leg ulcer and an ankle brachial pressure index of at least 0·8, and were tolerant of high compression. We randomly allocated participants (1:1) to receive two-layer compression hosiery or a four-layer bandage, using a remote randomisation service and prevalidated computer randomisation program. Participants were stratified by ulcer duration and ulcer area with permuted blocks (block sizes four and six). The primary endpoint was time to ulcer healing, with a maximum follow-up of 12 months. Although participants and health-care providers were not masked to treatment allocation, the primary endpoint was measured by masked assessment of photographs. Primary analysis was intention to treat with Cox regression, with adjustment for ulcer area, ulcer duration, physical mobility, and centre. This trial is registered with the ISRCTN register, number ISRCTN49373072. FINDINGS: We randomly allocated 457 participants to the two treatment groups: 230 to two-layer hosiery and 227 to the four-layer bandage, of whom 453 (230 hosiery and 223 bandage) contributed data for analysis. Median time to ulcer healing was 99 days (95% CI 84-126) in the hosiery group and 98 days (85-112) in the bandage group, and the proportion of ulcers healing was much the same in the two groups (70·9% hosiery and 70·4% bandage). More hosiery participants changed their allocated treatment (38·3% hosiery vs 27·0% bandage; p=0·02). 300 participants had 895 adverse events, of which 85 (9·5%) were classed as serious but unrelated to trial treatment. INTERPRETATION: Two-layer compression hosiery is a viable alternative to the four-layer bandage-it is equally as effective at healing venous leg ulcers. However, a higher rate of treatment changes in participants in the hosiery group than in the bandage group suggests that hosiery might not be suitable for all patients. FUNDING: NIHR Health Technology Assessment programme (07/60/26).
Subject(s)
Compression Bandages , Varicose Ulcer/therapy , Aged , Aged, 80 and over , Compression Bandages/adverse effects , Compression Bandages/economics , Cost-Benefit Analysis , England , Female , Health Care Costs/statistics & numerical data , Humans , Kaplan-Meier Estimate , Male , Middle Aged , Northern Ireland , Quality-Adjusted Life Years , Stockings, Compression/adverse effects , Stockings, Compression/economics , Treatment Outcome , Varicose Ulcer/economics , Wound HealingABSTRACT
BACKGROUND: Network meta-analysis methods extend the standard pair-wise framework to allow simultaneous comparison of multiple interventions in a single statistical model. Despite published work on network meta-analysis mainly focussing on the synthesis of aggregate data, methods have been developed that allow the use of individual patient-level data specifically when outcomes are dichotomous or continuous. This paper focuses on the synthesis of individual patient-level and summary time to event data, motivated by a real data example looking at the effectiveness of high compression treatments on the healing of venous leg ulcers. METHODS: This paper introduces a novel network meta-analysis modelling approach that allows individual patient-level (time to event with censoring) and summary-level data (event count for a given follow-up time) to be synthesised jointly by assuming an underlying, common, distribution of time to healing. Alternative model assumptions were tested within the motivating example. Model fit and adequacy measures were used to compare and select models. RESULTS: Due to the availability of individual patient-level data in our example we were able to use a Weibull distribution to describe time to healing; otherwise, we would have been limited to specifying a uniparametric distribution. Absolute effectiveness estimates were more sensitive than relative effectiveness estimates to a range of alternative specifications for the model. CONCLUSIONS: The synthesis of time to event data considering individual patient-level data provides modelling flexibility, and can be particularly important when absolute effectiveness estimates, and not just relative effect estimates, are of interest.
Subject(s)
Compression Bandages , Leg Ulcer/therapy , Models, Statistical , Varicose Ulcer/therapy , Data Interpretation, Statistical , Humans , TimeABSTRACT
INTRODUCTION: Prior to investing in a large, multicentre randomised controlled trial (RCT), the National Institute for Health Research in the UK called for an evaluation of the feasibility and value for money of undertaking a trial on intravenous immunoglobulin (IVIG) as an adjuvant therapy for severe sepsis/septic shock. METHODS: In response to this call, this study assessed the clinical and cost-effectiveness of IVIG (using a decision model), and evaluated the value of conducting an RCT (using expected value of information (EVI) analysis). The evidence informing such assessments was obtained through a series of systematic reviews and meta-analyses. Further primary data analyses were also undertaken using the Intensive Care National Audit & Research Centre Case Mix Programme Database, and a Scottish Intensive Care Society research study. RESULTS: We found a large degree of statistical heterogeneity in the clinical evidence on treatment effect, and the source of such heterogeneity was unclear. The incremental cost-effectiveness ratio of IVIG is within the borderline region of estimates considered to represent value for money, but results appear highly sensitive to the choice of model used for clinical effectiveness. This was also the case with EVI estimates, with maximum payoffs from conducting a further clinical trial between £ 137 and £ 1,011 million. CONCLUSIONS: Our analyses suggest that there is a need for a further RCT. Results on the value of conducting such research, however, were sensitive to the clinical effectiveness model used, reflecting the high level of heterogeneity in the evidence base.
Subject(s)
Cost-Benefit Analysis/methods , Immunoglobulins, Intravenous/administration & dosage , Immunoglobulins, Intravenous/economics , Randomized Controlled Trials as Topic/economics , Shock, Septic/drug therapy , Shock, Septic/economics , Aged , Decision Support Techniques , Female , Humans , Male , Middle Aged , Sepsis/drug therapy , Sepsis/economics , Survival Rate/trends , Treatment OutcomeABSTRACT
Background. Antimicrobial resistance (AMR) is a global public health threat. The wider implications of AMR, such as the impact of antibiotic resistance (ABR) on surgical procedures, are yet to be quantified. The objective of this study was to produce a conceptual modeling framework to provide a basis for estimating the current and potential future consequences of ABR for surgical procedures in England. Design. A framework was developed using literature-based evidence and structured expert elicitation. This was applied to populations undergoing emergency repair of the neck of the femur and elective colorectal resection surgery. Results. The framework captures the implications of increasing ABR by allowing for higher rates of surgical site infection (SSI) as the effectiveness of antibiotic prophylaxis wanes and worsened outcomes following SSIs to reflect reduced antibiotic treatment effectiveness. The expert elicitation highlights the uncertainty in quantifying the impact of ABR, reflected in the results. A hypothetical SSI rate increase of 14% in a person undergoing emergency repair of the femur could increase costs by 39% (-2% to 108% credible interval [CI]) and decrease quality-adjusted life-years by 11% (0.4% to 62% CI) over 15 y. Conclusions. The modeling framework is a starting point for addressing the implication of ABR on the outcomes and costs of surgeries. Due to clinical uncertainty highlighted in the expert elicitation process, the numerical outputs of the case studies should not be focused on but rather the framework itself, illustration of the evidence gaps, the benefit of expert elicitation in quantifying parameters with limited data, and the potential magnitude of the impact of ABR on surgical procedures. Implications. The framework can be used to support research surrounding the health and cost burden of ABR in England. Highlights: The modeling framework is a starting point for assessing the health and cost impacts of antibiotic resistance on surgeries in England.Formulating a framework and synthesizing evidence to parameterize data gaps provides targets for future research.Once data gaps are addressed, this modeling framework can be used to feed into overall estimates of the health and cost burden of antibiotic resistance and evaluate control policies.
ABSTRACT
BACKGROUND: The evidence used to inform health care decision making (HCDM) is typically uncertain. In these situations, the experience of experts is essential to help decision makers reach a decision. Structured expert elicitation (referred to as elicitation) is a quantitative process to capture experts' beliefs. There is heterogeneity in the existing elicitation methodology used in HCDM, and it is not clear if existing guidelines are appropriate for use in this context. In this article, we seek to establish reference case methods for elicitation to inform HCDM. METHODS: We collated the methods available for elicitation using reviews and critique. In addition, we conducted controlled experiments to test the accuracy of alternative methods. We determined the suitability of the methods choices for use in HCDM according to a predefined set of principles for elicitation in HCDM, which we have also generated. We determined reference case methods for elicitation in HCDM for health technology assessment (HTA). RESULTS: In almost all methods choices available for elicitation, we found a lack of empirical evidence supporting recommendations. Despite this, it is possible to define reference case methods for HTA. The reference methods include a focus on gathering experts with substantive knowledge of the quantities being elicited as opposed to those trained in probability and statistics, eliciting quantities that the expert might observe directly, and individual elicitation of beliefs, rather than solely consensus methods. It is likely that there are additional considerations for decision makers in health care outside of HTA. CONCLUSIONS: The reference case developed here allows the use of different methods, depending on the decision-making setting. Further applied examples of elicitation methods would be useful. Experimental evidence comparing methods should be generated.
Subject(s)
Expert Testimony , Technology Assessment, Biomedical , Decision Making , Delivery of Health Care , Humans , Probability , UncertaintyABSTRACT
The objective was to undertake an early cost-effectiveness assessment of valoctocogene roxaparvovec (valrox; Roctavian) compared to factor (F)VIII prophylaxis or emicizumab (Hemlibra; Roche HQ, Bazel, Switzerland) in patients with severe Hemophilia A (HA) without FVIII-antibodies. We also aimed to incorporate and quantify novel measures of value such as treatment durability, maximum value-based price (MVBP) and break-even time (ie, time until benefits begin to offset upfront payment). We constructed a Markov model to model bleeds over time which were linked to costs and quality-of-life decrements. In the valrox arm, FVIII over time was estimated combining initial effect and treatment waning and then linked to bleeds. In FVIII and emicizumab arms, bleeds were based on trial evidence. Evidence and assumptions were validated using expert elicitation. Model robustness was tested via sensitivity analyses. A Dutch societal perspective was applied with a 10-year time horizon. Valrox in comparison to FVIII, and emicizumab showed small increases in quality-adjusted life years at lower costs, and were therefore dominant. Valrox' base case MVBP was estimated at 2.65 million/treatment compared to FVIII and 3.5 million/treatment versus emicizumab. Mean break-even time was 8.03 years compared to FVIII and 5.68 years to emicizumab. Early modeling of patients with HA in The Netherlands treated with valrox resulted in estimated improved health and lower cost compared to prophylactic FVIII and emicizumab. We also demonstrated feasibility of incorporation of treatment durability and novel outcomes such as value-based pricing scenarios and break-even time. Future work should aim to better characterize uncertainties and increase translation of early modeling to direct research efforts.
ABSTRACT
We can use decision models to estimate cost effectiveness, quantify uncertainty regarding the adoption decision and provide estimates of the value of further research. In many cases, the existence of only limited data with which to populate a decision model can mean that a cost-effectiveness analysis either does not proceed or may misrepresent the degree of uncertainty associated with model inputs. An example is the case of negative pressure wound therapy (NPWT) used to treat severe pressure ulceration, for which the evidence base is limited and sparse. There is, however, substantial practical experience of using this treatment and its comparators. We can capture this knowledge quantitatively to inform a cost-effectiveness model by eliciting beliefs from experts. This paper describes the design and conduct of an elicitation exercise to generate estimates of multiple uncertain model inputs and validate analytical assumptions for a decision model on the use of NPWT. In designing the exercise, the primary focus was the use of elicitation to inform decision models (multistate models), where representations of uncertain beliefs need to be probabilistically coherent. This paper demonstrates that it is feasible to collect formally elicited evidence to inform decision models.
Subject(s)
Expert Testimony , Negative-Pressure Wound Therapy/economics , Pressure Ulcer/therapy , Decision Support Techniques , Humans , Nurses , Pressure Ulcer/economics , Severity of Illness Index , Technology Assessment, Biomedical/economics , Technology Assessment, Biomedical/methodsABSTRACT
Well-established methods of economic evaluation are used in many countries to inform decisions about the funding of new medical interventions. To guide such decisions, it is important to consider what health gains would be expected from the same level of investment elsewhere in the health care system. Recent research in the United Kingdom has evaluated the evidence available and the methods required to estimate the health effects of changes in health care expenditure within the National Health Service. Because of the absence of sufficiently broad-ranging data, assumptions were required in the previously mentioned work to estimate health effects in terms of a broader measure of health (quality-adjusted life-years), which is more relevant for policy. These assumptions constitute important sources of uncertainty. This work presents an application of the structured elicitation of the judgments of key individuals about these uncertain quantities. This article describes the design and conduct of the exercise, including the quantities elicited, the individual (rather than consensus) approach used, how uncertainty in knowledge was elicited (mode and bounds of an 80% credible interval), and methods to generate group estimates. It also reports on a successful application involving 28 clinical experts and 25 individuals with policy responsibilities. Although, as expected, most experts found replying to the questions challenging, they were able to express their beliefs quantitatively. Consistent across the uncertainties elicited, experts' judgments suggest that the quality-adjusted life-year (QALY) impacts of changes in expenditure from earlier work using assumptions are likely to have been underestimated and the "central" estimate of health opportunity cost from that work (£12,936 per QALY) to have been overestimated.
Subject(s)
Cost-Benefit Analysis/methods , Expert Testimony/methods , Patient Simulation , Uncertainty , Cost-Benefit Analysis/trends , HumansABSTRACT
In recent years, Health Technology Assessment (HTA) processes specific to diagnostics and prognostic tests have been created in response to the increased pressure on health systems to decide not only which tests should be used in practice but also the best way to proceed, clinically, from the information they provide. These technologies differ in the way value is accrued to the population of users, depending critically on the value of downstream health care choices. This paper defines an analytical framework for establishing the value of diagnostic and prognostic tests for HTA in a way that is consistent with methods used for the evaluation of other health care technologies. It assumes a linked-evidence approach where modeling is required, and incorporates considerations regarding several different areas of policy, such as personalized medicine. We initially focus on diagnostic technologies with dichotomous results, and then extend the framework by considering diagnostic tests that provide more complex information, such as continuous measures (for example, blood glucose measurements) or multiple categories (such as tumor classification systems). We also consider how the methods of assessment differ for prognostic information or for diagnostics without a reference standard. Throughout, we propose innovative graphical ways of summarizing the results of such complex assessments of value.
Subject(s)
Clinical Decision-Making/methods , Decision Support Techniques , Diagnostic Techniques and Procedures/statistics & numerical data , Technology Assessment, Biomedical/methods , Decision Making , Diagnostic Techniques and Procedures/standards , Humans , Monitoring, Physiologic/methods , Monitoring, Physiologic/statistics & numerical data , PrognosisABSTRACT
BACKGROUND: The current recommendation of using transrectal ultrasound-guided biopsy (TRUSB) to diagnose prostate cancer misses clinically significant (CS) cancers. More sensitive biopsies (eg, template prostate mapping biopsy [TPMB]) are too resource intensive for routine use, and there is little evidence on multiparametric magnetic resonance imaging (MPMRI). OBJECTIVE: To identify the most effective and cost-effective way of using these tests to detect CS prostate cancer. DESIGN, SETTING, AND PARTICIPANTS: Cost-effectiveness modelling of health outcomes and costs of men referred to secondary care with a suspicion of prostate cancer prior to any biopsy in the UK National Health Service using information from the diagnostic Prostate MR Imaging Study (PROMIS). INTERVENTION: Combinations of MPMRI, TRUSB, and TPMB, using different definitions and diagnostic cut-offs for CS cancer. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Strategies that detect the most CS cancers given testing costs, and incremental cost-effectiveness ratios (ICERs) in quality-adjusted life years (QALYs) given long-term costs. RESULTS AND LIMITATIONS: The use of MPMRI first and then up to two MRI-targeted TRUSBs detects more CS cancers per pound spent than a strategy using TRUSB first (sensitivity = 0.95 [95% confidence interval {CI} 0.92-0.98] vs 0.91 [95% CI 0.86-0.94]) and is cost effective (ICER = £7,076 [8350/QALY gained]). The limitations stem from the evidence base in the accuracy of MRI-targeted biopsy and the long-term outcomes of men with CS prostate cancer. CONCLUSIONS: An MPMRI-first strategy is effective and cost effective for the diagnosis of CS prostate cancer. These findings are sensitive to the test costs, sensitivity of MRI-targeted TRUSB, and long-term outcomes of men with cancer, which warrant more empirical research. This analysis can inform the development of clinical guidelines. PATIENT SUMMARY: We found that, under certain assumptions, the use of multiparametric magnetic resonance imaging first and then up to two transrectal ultrasound-guided biopsy is better than the current clinical standard and is good value for money.
Subject(s)
Health Care Costs , Image-Guided Biopsy/economics , Magnetic Resonance Imaging/economics , Prostatic Neoplasms/diagnostic imaging , Prostatic Neoplasms/economics , Ultrasonography/economics , Cost-Benefit Analysis , Decision Support Techniques , Decision Trees , Humans , Image-Guided Biopsy/methods , Male , Models, Economic , Neoplasm Grading , Predictive Value of Tests , Prostatic Neoplasms/pathology , Prostatic Neoplasms/therapy , Quality-Adjusted Life Years , Reproducibility of Results , State Medicine/economics , Time Factors , Treatment Outcome , United KingdomABSTRACT
Cost-effectiveness analysis (CEA) models are routinely used to inform health care policy. Key model inputs include relative effectiveness of competing treatments, typically informed by meta-analysis. Heterogeneity is ubiquitous in meta-analysis, and random effects models are usually used when there is variability in effects across studies. In the absence of observed treatment effect modifiers, various summaries from the random effects distribution (random effects mean, predictive distribution, random effects distribution, or study-specific estimate [shrunken or independent of other studies]) can be used depending on the relationship between the setting for the decision (population characteristics, treatment definitions, and other contextual factors) and the included studies. If covariates have been measured that could potentially explain the heterogeneity, then these can be included in a meta-regression model. We describe how covariates can be included in a network meta-analysis model and how the output from such an analysis can be used in a CEA model. We outline a model selection procedure to help choose between competing models and stress the importance of clinical input. We illustrate the approach with a health technology assessment of intravenous immunoglobulin for the management of adult patients with severe sepsis in an intensive care setting, which exemplifies how risk of bias information can be incorporated into CEA models. We show that the results of the CEA and value-of-information analyses are sensitive to the model and highlight the importance of sensitivity analyses when conducting CEA in the presence of heterogeneity. The methods presented extend naturally to heterogeneity in other model inputs, such as baseline risk.