International Journal of Technology Assessment in Health Care最新文献

Medical device health technology assessment (HTA) in Australia is largely coordinated by the Medical Services Advisory Committee (MSAC). Its remit to improve the public’s health by deciding where to allocate public healthcare funding, can be enhanced by considering real world evidence (RWE). Existing data sources have limitations that can be addressed through RWE, including coverage of Australian patient populations who may not meet trial eligibility criteria, and long-term follow-up through data linkage and datasets. We partnered with a university to explore what information could be gained from an analysis of linked administrative patient data, with a view to addressing current evidence gaps and/or limitations. The findings can be used as a source of local data to define patient populations, estimate actual costs of care, and enable more comprehensive economic modeling to inform medical device HTA.

The University-developed New South Wales Cardiovascular Cohort dataset, comprising person-level longitudinal NSW administrative data for all patients admitted to hospital with a cardiovascular diagnosis from 2001 onwards, linked to national Medicare Benefits Schedule and Pharmaceutical Benefits Scheme claims data, was interrogated.

Working with RWE is resource intensive in terms of time and costs. The potential of these data was revealed as the research progressed. It was possible to continually refine the data analyzed and reported,as well as expand the data requested. Varied expertise is required to accurately analyze the administrative datasets, particularly clinical classification skills and expertise in methods for causal inference using observational data. Findings from this study will enable the refinement of information for MSAC submissions, including identifying the most relevant patient population and reporting comprehensive costs, beyond an admitted hospital setting. The data will enhance engagement with clinicians and refine messaging, for example regarding patient risk factors.

RWE enhances Australian HTA applications. Local data, extended periods of time and insights not apparent from a focus on admitted hospital episodes can be revealed. Data can be refined during the process for specificity and applicability.

Fractional Flow Reserve (FFR) is a diagnostic tool that aids decision-making in the treatment of coronary artery disease (CAD). FFR provides an objective measurement and is used as an adjunct to an angiogram. The clinical and cost-benefit of using FFR have been well established across published literature. This research was aimed at evaluating the economic impact of using FFR as an adjunct to angiogram versus an angiogram alone, in the Indian healthcare context.

A study from a tertiary care public hospital in India estimated the impact of using FFR as an adjunct to angiogram in management of CAD. This study was used to create a mathematical simulation model to estimate cost-effectiveness and economic impact of using FFR over seven years’ time horizon, from the Indian health systems perspective. A targeted literature review was performed to collect the clinical inputs for the model, and the national public health insurance program data was referenced to obtain the cost inputs.

A hypothetical cohort of 100,000 patients in the model reported 30 percent reduction in unnecessary stenting. Moreover, 14,025 deaths were averted with the adoption of FFR. In addition, there was a cost-saving of INR46,986 (574USD) per death averted and INR5,169 (63USD) per patient treated over a seven-year time horizon. The analysis demonstrated that FFR inclusion in the current clinical practice saves INR2,651 (32USD) per patient in overall upfront cost and INR2,518 (31USD) per patient in overall follow-up cost over a seven year follow-up period owing to improved diagnosis and prognosis.

In conclusion, FFR prevents unnecessary stenting, reduces overall mortality, and proves to be a cost-saving intervention in the long-term when used as a decision-making criterion in CAD patients in the Indian context.

This rapid review clarified the evidence supporting avoidance of venipuncture on the ipsilateral arm in breast cancer patients who have had sentinel lymph node biopsy (SLNB) or axillary lymph node clearance (ALNC), as a preventive measure against lymphoedema.

A systematic search was carried out for systematic reviews with the following elements:

• • Population – breast cancer patients who had SLNB or ALNC

• • Intervention – avoidance of venipuncture in the ipsilateral arm

• • Comparator –use of either arm for venipuncture

• • Outcomes – risk of lymphoedema in the ipsilateral arm

Databases searched included PubMed (MEDLINE), Epistemonikos and the Cochrane Database of Systematic Reviews. Included reviews were critically appraised with the AMSTAR2 instrument and the primary studies were extracted and tabulated in a narrative synthesis.

Six reviews were included; none of the reviews self-identified as systematic reviews in their titles/abstracts. Four reviews did report methods, including systematic search strategies and describing studies in adequate detail. However, all reviews did not meet most criteria on the AMSTAR2 checklist. The reviews concluded that the evidence base for avoiding venipuncture was inconsistent. An evidence table was consequently drawn up of the primary studies included in the reviews as a narrative synthesis of the primary evidence base.

The primary evidence base comprised 12 observational studies – six prospective cohort or descriptive studies and 6 retrospective studies. These studies were inconsistent and inconclusive; studies that found an association or reported cases following ipsilateral venipuncture were subject to recall bias or other potential confounders. Guidelines or patient information recommending avoidance of ipsilateral venipuncture do so based on expert opinion rather than consistent findings from empirical studies.

All reviews concluded that the evidence base for avoiding venipuncture was inconsistent. Review authors consistently recognized there was no strong basis for the prevalent recommendations to avoid ipsilateral venipuncture to prevent lymphoedema. Such recommendations lead to unnecessary measures that may be detrimental to patients. Stakeholders should reconsider advice to patients in the light of existing evidence and weigh up the uncertain benefits against potential harm to patients.

Despite intense efforts in development of new treatments over the last two decades, symptomatic treatments remain the only option for the vast majority patients diagnosed with dementia due to Alzheimer’s disease (AD). There remains a significant unmet need for disease modifying therapies (DMTs) to slow or stop AD progression. DMTs in development are targeting early stages of AD (pre-clinical, mild cognitive impairment and mild dementia stages), thereby creating an entirely new treatment paradigm for patients, clinicians, and payers. A key challenge will be in identifying the appropriate patient for treatment in a very heterogenous population. We have performed a literature review to better understand and define the AD population, with a view to enabling more targeted treatment in future.

Embase, MEDLINE and the Cochrane Library were searched to identify publications between 2010-2021 on observational studies reporting evidence on prevalence and subgroup identification, including clinical feasibility of identification. The search was restricted to English language.

We identified 45 studies, mostly from Europe, USA and Asia. Populations were primarily grouped based on generic demographic factors (e.g., age, sex, gender), AD staging, comorbidities or biomarkers. Prevalence data was available for six subpopulations: pre-dementia stage, mild dementia, age, Apolipoprotein E (APOE) genotype, comorbid obesity and hypertension. Across these, data on prevalence were heterogenous depending on study design and country of origin, and ranging between 66 million to 102 million for people with mild AD dementia, or as another example, ranging between 46 million to 92 million for APOE genotype carriers worldwide.

The heterogeneity and the uncertainty in prevalence of the AD population represent big challenges to clinicians and payers. Future discussions on target patient identification for new treatments should be aligned and integrated with current clinical practice e.g. leveraging validated biomarkers as diagnostic tools. Additional research on an integrated approach to identify patients who would benefit the most from DMTs will be needed.

Defining drug innovation can be challenging and there is no consensus on what a truly “innovative” medicine is. The Italian Medicine Agency (AIFA) has established an approach to assess innovativeness based on therapeutic need, added therapeutic value, and quality of evidence. However, judgment can be subjective and may not be adequate for assessment at the time of marketing authorization, when only preliminary evidence – often from non-comparative or non-randomized trials – are available. We developed a transparent methodology for early assessment of innovativeness at the time of marketing authorization, based on AIFA guidelines.

Since the perspective was the marketing authorization date, only data available at agency’s Medical Review or pivotal trial publications were considered. AIFA criteria were revisited, using oncology medicines approved in the last 10 years as a base case. Impact of preliminary evidence and inadequate study design was considered.

Each assessment should refer to the first approved specific indication and predefined clinically relevant outcomes. When more than one study was presented, best methodological quality, larger sample and/or longer follow-up was selected. Four domains were established: Therapeutical need: existence and clinical benefits of alternative therapies; Clinical benefit added when compared to those alternatives; Suitability of study design considering adequate comparator group, relevant outcome assessed and randomization; Risk of bias. For each domain, clear and specific criteria were defined in consensus by a group of experts in health technology assessment (HTA) and were applied to all cancer drugs evaluated.

Efficacy evidence available for marketing authorization are often based on preliminary data, arising from single randomized clinical trials or even non-comparative studies, which difficult early assessments of innovativeness. For this reason, transparent and reproducible methodologies can be useful not only to HTA bodies, but also for other key stakeholders in the pharmaceutical market, such as investors, researchers, doctors, and governments.

Onasemnogene abeparvovec has been approved for the treatment of spinal muscular atrophy 5q (SMA) type 1 in several countries, which calls for an independent assessment of its evidence regarding efficacy and safety.

This study results from searches conducted on databases MEDLINE, Embase, LILACS and Cochrane Library up to November 2022, supported by additional searches on registry databases and by manual searches of references listed in eligible studies. Outcomes of interest were global survival and mechanical-ventilation-free survival, improvement in motor function and treatment-related adverse events. Risk of bias was assessed via ROBINS-I and certainty of evidence via GRADE. Proportional meta-analysis models were performed when applicable.

Four reports of three open-label, non-comparative clinical trials (START, STR1VE-US and STR1VE-EU) covering 67 patients were included in review. Meta-analyses of data available in a 12-month follow-up estimate a global survival of 97.6% (95% confidence interval [CI]: 92.6, 99.9; I2 = 0%, n=67), an event-free survival of 96.5% (95%CI: 90.8, 99.5; I2 = 32%, n=66) and a CHOP-INTEND score of 40 points or less proportion of 87.3% (95%CI: 69.8, 97.8; I2 = 69%, n=67). Proportions of 61.1% (95%CI: 40, 80.2; I² = 62%, n=67) of serious adverse events and of 58.4% (95%CI: 46.5, 69.8; I2 = 78%, n=67) of treatment-related adverse events are estimated. Despite the significant effect magnitude, reviewed studies were assessed as high risk of bias and as having very low certainty of evidence due to imprecision and risk of bias.

Reduced sample size and follow-up time offer uncertainties as regards the long-term benefits of the gene therapy, which strongly calls for the monitoring and assessment of results in clinical practice.

Non-small cell lung cancer (NSCLC) constitutes 85 percent of lung cancer diagnoses and poses an economic threat to the sustainability of healthcare services. This study was conducted to estimate hospitalization costs associated with advanced NSCLC without sensitizing EGFR (epidermal growth factor receptor) and ALK (anaplastic lymphoma kinase) alterations in China and explore the potential predictors.

Data linked with patients with advanced NSCLC (stage IIIB–IV) without sensitizing EGFR and ALK alterations were obtained from the electronic medical record system of one general hospital and one cancer hospital in Jiangsu province, China, ranging from January 2017 to December 2020. We excluded patients with lung metastases from tumors elsewhere in the body. The socio-demographic characteristics, disease-related characteristics, and hospitalization cost of eligible patients were extracted. We used the generalized linear model (GLM) to assess the potential influencing factors of hospitalization cost.

Patients with advanced NSCLC (n=7,260) were included in this study. The median hospitalization cost of advanced NSCLC was USD11,540.47. The median hospitalization examination and test costs were USD1,539.46, and the median hospitalization drug cost was USD6,351.47. GLM results showed that patients aged 60 or older (95% Confidence Interval [CI]: -1019.1,128.6), who had no gene driver (95%CI: -1,681.6,-233.6) were more likely to have relatively lower hospitalization costs for advanced NSCLC. Patients treated in cancer hospital (95%CI: 1,329.1,2,620.0) and with non-squamous carcinoma (95%CI: 171.3, 1,235.4) may have higher hospitalization costs. Compared with Urban Employee Basic Medical Insurance, patients with free medical services (95%CI: 1,248.4,6,298.7) were associated with higher hospitalization costs. Patients with higher hospitalization frequency and longer length of hospital stay (p < 0.05) were linked to higher hospitalization costs.

The hospitalization costs linked to advanced NSCLC is considerable for patients, with drug costs accounting for the largest. More efforts still need to be made to alleviate the direct medical burden.

Systematic reviews are important for informing decision-making and primary research, but they can be time consuming and costly. With the advent of machine learning, there is an opportunity to accelerate the review process in study screening. We aimed to understand the literature to make decisions about the use of machine learning for screening in our review workflow.

A pragmatic literature review of PubMed to obtain studies evaluating the accuracy of publicly available machine learning screening tools. A single reviewer used ‘snowballing’ searches to identify studies reporting accuracy data and extracted the sensitivity (ability to correctly identify included studies for a review) and specificity, or workload saved (ability to correctly exclude irrelevant studies).

Ten tools (AbstractR, ASReview Lab, Cochrane RCT classifier, Concept encoder, Dpedia, DistillerAI, Rayyan, Research Screener, Robot Analyst, SWIFT-active screener) were evaluated in a total of 16 studies. Fourteen studies were single arm where, although compared with a reference standard (predominantly single reviewer screening), there was no other comparator. Two studies were comparative, where tools were compared with other tools as well as a reference standard. All tools ranked records by probability of inclusion and either (i) applied a cut-point to exclude records or (ii) were used to rank and re-rank records during screening iterations, with screening continuing until most relevant records were obtained. The accuracy of tools varied widely between different studies and review projects. When used in method (ii), at 95 percent to 100 percent sensitivity, tools achieved workload savings of between 7 percent and 99 percent. It was unclear whether evaluations were conducted independent of tool developers.

Evaluations suggest the potential for tools to correctly classify studies in screening. However, conclusions are limited since (i) tool accuracy is generally not compared with dual reviewer screening and (ii) the literature lacks comparative studies and, because of between-study heterogeneity, it is not possible to robustly determine the accuracy of tools compared with each other. Independent evaluations are needed.

In Taiwan, people with hip osteoarthritis (OA) receive a total hip arthroplasty (THA). They can apply for National Health Insurance (NHI) coverage for metal-on-polyethylene (MoP) implant or USD1,313.2 co-pay for new bearing surface materials. This study aimed to report the number of first primary THAs, and calculate the costs of THA by different choices of prothesis implant.

A retrospective cohort study of patients aged 50 years or older who had OA (as an indication for THA) from 1 January 2010 through to 31 December 2018 was established from Taiwan’s NHI Claims Data. The cohort was followed-up until 31 December 2019. THA Implant combinations were defined by bearing surface materials e.g., “ceramic-metal” into alumina or composites made from alumina and zirconia, including metal-on-polyethylene (MOP), ceramic-metal composite ceramic-on-polyethylene (c-COP), alumina ceramic-on-ceramic (a-COC), ceramic-metal composite ceramic-on-ceramic (c-COC). Since only MOP was covered by the NHI, patients who chose the COP or COC implant had to pay for additional costs. We used hospital costs comparison data to calculate the average out-of-pocket (OOP) costs for different implant combinations.

This study comprised 23,560 patients with first primary THA over 9 years. The number of patients of first primary THA increased from 1,802 in 2010 to 3,251 in 2018. The mean age of patients at baseline was 68 years, and the majority were women (70.6%). The share of users for each THA implant type were: MOP implant (49.2%), c-COC implant, (20.8%), a-COC implant, (6.5%), and c-COP implant, 5.9%. The average OOP costs of each implant were: USD3,578.60 for c-COC (SD=381.80), USD2,073.00 for a-COC (SD=279.80), and USD2,082.1 for c-COP (SD=334.1).

Although only MOP was fully covered by NHI, only about 50% of the OA patients chose this type of implant, and 26.7 percent chose alumina and zirconia ceramic composite despite this being a much higher OOP cost. Whether choosing more expensive implants would be cost-effective for THA in Taiwan’s healthcare system requires further analysis.

Non-invasive Scrambler Therapy (ST) reduces pain by attaching electrodes to neural pathways of major nerves, transmitting information along with microcurrent to the nerves to induce a painless sensation. The ST has been widely used to reduce pain for patients with musculoskeletal pain. However, little is known about the musculoskeletal pain relief effect of the ST. Therefore, this study aims to evaluate the treatment effectiveness of the ST.

A systematic literature review was conducted based on the following search strategy and databases, and all studies published before August 2021 were included in Pubmed, Embase, and Cochrane library, Ovid Medline, Koreamed, kmbase, and Science On. The subjects were patients with intractable and musculoskeletal pain, excluding cancer pain, and intervention methods included non-invasive ST alone or in combination with physical therapy. In addition, the comparative method was not limited. The outcome variables were the degree of pain relief, total analgesic use, health-related quality of life, pressure pain threshold, pain intensity and functional interference scales, and pain sensitivity. Safety-related outcome variables were all side effects. Cochrane Risk of Bias 1.0 was used to assess the risk of bias in the literature.

Two hundred forty-one articles were retrieved using a pre-determined search strategy. Of these, 15 duplicate articles, 215 articles after reviewing the abstract and title, and nine articles after checking the full text were excluded. Two studies with randomized controlled trials (RCTs) were finally selected. When comparing ST and placebo groups for musculoskeletal pain, the pain reduction effect of ST lasted for three weeks. Moreover, patients with neuropathic pain treated with ST had a lower pain intensity for one to three months compared to the drug treatment group.

Based on this systematic review, the effectiveness of ST is yet sufficient owing to small sample size and possibility of selective report bias. More studies with well-designed RCTs are required to further assess the effectiveness of the ST.