Fracture definitions in observational osteoporosis drug effects studies that leverage healthcare administrative (claims) data: a scoping review.

Healthcare administrative (claims) data are commonly utilized to estimate drug effects. We identified considerable heterogeneity in fracture outcome definitions in a scoping review of 57 studies that estimated osteoporosis drug effects on fracture risk. Better understanding of the impact of different fracture definitions on study results is needed. PURPOSE: Healthcare administrative (claims) data are frequently used to estimate the real-world effects of drugs. Fracture incidence is a common outcome of osteoporosis drug studies. We aimed to describe how fractures are defined in studies that use claims data. METHODS: We searched MEDLINE (Ovid), Embase (Ovid), CINAHL (EBSCO), and gray literature for studies published in English between 2000 and 2020 that estimated fracture effectiveness (hip, humerus, radius/ulna, vertebra) or safety (atypical fracture of the femur, AFF) of osteoporosis drugs using claims data in Canada and the USA. Literature searches, screening and data abstraction were completed independently by two reviewers. RESULTS: We identified 57 eligible studies (52 effectiveness, 3 safety, 2 both). Hip fracture was the most common fracture site studied (93%), followed by humerus (66%), radius/ulna (59%), vertebra (61%), and AFF (9%). Half (n = 29) of the studies did not indicate specific data sources, codes, or cite a validation paper. Of the papers with sufficient detail, heterogeneity in fracture definitions was common. The most common definition within each fracture site was used by less than half of the studies that examined effectiveness (12 definitions in 29 hip fracture papers, 8 definitions in 17 humerus papers, 8 definitions in 13 radius/ulna papers, 9 definitions in 15 vertebra papers), and 3 definitions among 4 AFF papers. CONCLUSION: There is ambiguity and heterogeneity in fracture outcome definitions in studies that leverage claims data. Better transparency in outcome reporting is needed. Future exploration of how fracture definitions impact study results is warranted.

Estimating chronic hepatitis C prognosis using transient elastography-based liver stiffness: A systematic review and meta-analysis.

Chronic hepatitis C (CHC) is a leading cause of hepatic fibrosis and cirrhosis. The level of fibrosis is traditionally established by histology, and prognosis is estimated using fibrosis progression rates (FPRs; annual probability of progressing across histological stages). However, newer noninvasive alternatives are quickly replacing biopsy. One alternative, transient elastography (TE), quantifies fibrosis by measuring liver stiffness (LSM). Given these developments, the purpose of this study was (i) to estimate prognosis in treatment-naïve CHC patients using TE-based liver stiffness progression rates (LSPR) as an alternative to FPRs and (ii) to compare consistency between LSPRs and FPRs. A systematic literature search was performed using multiple databases (January 1990 to February 2016). LSPRs were calculated using either a direct method (given the difference in serial LSMs and time elapsed) or an indirect method given a single LSM and the estimated duration of infection and pooled using random-effects meta-analyses. For validation purposes, FPRs were also estimated. Heterogeneity was explored by random-effects meta-regression. Twenty-seven studies reporting on 39 groups of patients (N = 5874) were identified with 35 groups allowing for indirect and 8 for direct estimation of LSPR. The majority (~58%) of patients were HIV/HCV-coinfected. The estimated time-to-cirrhosis based on TE vs biopsy was 39 and 38 years, respectively. In univariate meta-regressions, male sex and HIV were positively and age at assessment, negatively associated with LSPRs. Noninvasive prognosis of HCV is consistent with FPRs in predicting time-to-cirrhosis, but more longitudinal studies of liver stiffness are needed to obtain refined estimates.

Use of instrumental variables in the analysis of generalized linear models in the presence of unmeasured confounding with applications to epidemiological research.

A major, often unstated, concern of researchers carrying out epidemiological studies of medical therapy is the potential impact on validity if estimates of treatment are biased due to unmeasured confounders. One technique for obtaining consistent estimates of treatment effects in the presence of unmeasured confounders is instrumental variables analysis (IVA). This technique has been well developed in the econometrics literature and is being increasingly used in epidemiological studies. However, the approach to IVA that is most commonly used in such studies is based on linear models, while many epidemiological applications make use of non-linear models, specifically generalized linear models (GLMs) such as logistic or Poisson regression. Here we present a simple method for applying IVA within the class of GLMs using the generalized method of moments approach. We explore some of the theoretical properties of the method and illustrate its use within both a simulation example and an epidemiological study where unmeasured confounding is suspected to be present. We estimate the effects of beta-blocker therapy on one-year all-cause mortality after an incident hospitalization for heart failure, in the absence of data describing disease severity, which is believed to be a confounder.

Impact of antibiotic administrative restrictions on trends in antibiotic resistance.

CONTEXT: In March 2001, in response to concerns about increasing resistance to fluoroquinolone (FQ) antibiotics, the Ontario Drug Benefit (ODB) program limited reimbursement of FQs to ODB beneficiaries defined as high risk or in whom other therapies are not tolerated. OBJECTIVE: To analyze the impact of the limited use (LU) policy changes on antibiotic resistance rates in Ontario, focussing on community-acquired pathogens. DESIGN: Ontario data submitted to the Canadian Bacterial Surveillance Network (CBSN) between January 1, 1998 and June 30, 2002 were analyzed for rates of resistance in various pathogen-antibiotic combinations. The effect of the LU policy on the level and rate of change of antibiotic resistance was estimated using time series models. RESULTS: Resistance rates for S. pneumoniae were 10-12% for penicillin, erythromycin and trimethoprim sulfamethoxazole (TMP/SMX) and less than 3% for amoxicillin and all three FQs tested. There was a statistically significant increasing trend in resistance rates of S. pneumoniae to amoxicillin and levofloxacin throughout the study period. Antibiotic resistance of S. pneumoniae to ciprofloxacin indicated a statistically significant decreasing trend over the study period with a statistically significant increase in the level of antibiotic resistance at the time of the LU policy implementation. No other indication of any statistically significant decrease in resistance rates associated with the LU policy was found. CONCLUSIONS: Although no direct cause and effect can be proven with these observational data, there is no evidence that the limited use policy to restrict fluoroquinolones decreased antibiotic resistance in any of the pathogen-antibiotic combinations tested.

Impact of reference-based pricing of nitrates on the use and costs of anti-anginal drugs.

BACKGROUND: Reference-based pricing limits reimbursement for a group of drugs that are deemed therapeutically equivalent to the cost of the lowest-priced product within that group. We estimated the effect of reference-based pricing of nitrate drugs used for long-term prophylaxis on prescribing of and expenditures on nitrates and other anti-anginal drugs dispensed to senior citizens in British Columbia. METHODS: We assessed trends in the monthly volume of prescriptions of anti-anginal drugs and the associated drug ingredient cost paid by the province's publicly funded drug subsidy program, Pharmacare, and by the patients themselves for the period April 1994 to May 1999. Trends in monthly rates of nitrate expenditures per 100,000 senior citizens before the introduction of reference-based pricing were extrapolated to infer what expenditures would have been without the policy. RESULTS: During the 3 1/2 years after reference-based pricing was introduced, Pharmacare expenditures on nitrates prescribed to senior citizens declined by $14.9 million (95% confidence interval $10.7 to $19.1 million). Most of these savings were due to the lower prices that Pharmacare paid for sustained-release nitroglycerin tablets and the nitroglycerin patch, which were the 2 most frequently prescribed nitrates before the introduction of reference-based pricing; $1.2 million (8%) of the savings represented expenditures by senior citizens who purchased drugs that were only partially reimbursed. There were no compensatory increases in expenditures for other anti-anginal drugs. Use of sublingual nitroglycerin--a marker for deteriorating health in patients with angina--did not increase after the introduction of reference-based pricing. The nitroglycerin patch is now the most frequently prescribed nitrate, owing to the fact that Pharmacare resumed the provision of full subsidies for the drug after its manufacturers voluntarily reduced retail prices. INTERPRETATION: Evidence to date suggests that reference-based pricing of nitrates has achieved its primary goal of reducing drug expenditures. The effects of this policy on patient health, associated health care costs and administrative costs remain to be investigated.

On the evaluation of drug benefits policy changes with longitudinal claims data: the policy maker's versus the clinician's perspective.

Cost containment in pharmaceutical-benefit plans are often controversially debated for their potential of unintended consequences on health and overall expenditures. Thorough evaluations are needed but hypotheses and design considerations are complex. Our objective is to provide a structured framework for the evaluation of drug-benefit changes using longitudinal claims data. Differential cost sharing (DCS) will serve as a recent example. Benefit-plan managers are mainly interested in the overall performance of their plan. In a policy model, any observed policy-related effects may be compared with what would have happened had the intervention not been implemented by extrapolating the pre-policy trend from the same patients. These estimates will reflect the global consequences of the policy maker's decision. However, such estimates represent summary effects of benefits and harms, separately identifiable in those complying with the intended policy and those not complying. Results from a policy model apply only to a specific policy implementation and tend to underestimate effects when non-compliance is high. Clinical-decision makers and patients, by contrast, are interested in the consequences of patients' actual compliance to the policy. A clinical model assesses the effects of DCS depending on the actual treatment in contrast to the treatment intended by the policy. However, this model must sometimes make, unprovable assumptions about the appropriate control of selection factors. In conclusion, both policy and clinical models should be tested with a clear understanding of their perspectives, hypotheses, and interpretations, using quasi-experimental time-series designs to evaluate the effects of drug cost-containment policies.

Health Utilities Index Mark 3: evidence of construct validity for stroke and arthritis in a population health survey.

BACKGROUND: The Health Utilities Index Mark 3 (HUI3) is a comprehensive, compact health status classification and health state preference system. The HUI3 system has been included in 4 Canadian population health surveys and numerous clinical trials. OBJECTIVES: To evaluate the construct validity of the HUI3 for the measurement of health-related quality of life (HRQL) and attribute-specific morbidity in respondents to the 1990 Ontario Health Survey reported to have arthritis or stroke. The authors assessed (1) whether those with stroke, arthritis, and both conditions had lower HRQL scores than those with neither condition and (2) whether HUI3 detects morbidity in specific health attributes affected by arthritis and stroke. Stroke (but not arthritis) were expected to affect speech and cognition; arthritis (but not stroke) to affect pain; both to affect mobility, dexterity, and emotion; and neither to affect vision and hearing. RESEARCH DESIGN: Linear regression models of HRQL and attribute-specific utilities were estimated as a function of 3 indicator variables of health problem (stroke only, arthritis only, both) and variables included to reduce confounding. RESULTS: Subjects with stroke, arthritis, and both conditions had substantially lower HRQL than those with neither condition. Stroke subjects had greater morbidity in speech and cognition than arthritis subjects; somewhat surprisingly, pain morbidity was only slightly higher among arthritis subjects; neither condition affected vision or hearing. These associations were robust to various model specifications. CONCLUSIONS: The HUI3 system appears valid for measuring health status and HRQL for stroke and arthritis in the context of a noninstitutionalized population health survey.

Proportion of osteoporotic post-menopausal women at increased risk for upper GI adverse events associated with bisphosphonate therapy.

Background- The bisphosphonate alendronate has been associated with higher rates of adverse oesophageal effects when used in the community setting compared to what was observed in the clinical trials. Patients with a history of gastroesophageal problems or who are concurrently using an NSAID therapy may be at increased risk for the gastroesophageal problems associated with alendronate use. This study assesses the proportion of post-menopausal women in the community with osteoporosis that are at increased risk for gastroesophageal adverse effects associated with alendronate.Methods- The administrative database for the Quebec government drug benefit program was used to identify a cohort of 5400 post menopausal women aged 65 years or older who were using the bisphosphonate etidronate for the treatment of osteoporosis. Patients were evaluated for the presence of either risk factor, chronic GI drug therapy use (a marker for prior gastroesophageal problems) or chronic NSAID use.Findings- 31% of women taking etidronate were also chronically using GI drug therapies and 50% were using NSAIDs; 18% of the women were using all three drugs.Interpretation- Many osteoporosis patients in the community setting who are candidates for bisphosphonate therapy might be considered at increased risk for alendronate's gastroesophageal adverse effects. This may account for differences in pre-marketing and postmarketing event rates. Copyright (c) 2000 John Wiley & Sons, Ltd.

Health care policy evaluation using longitudinal insurance claims data: an application of the panel Tobit estimator.

The British Columbia Ministry of Health provides enhanced prescription drug insurance coverage to residents aged 65 and older. This exogenous change in the effective price of prescription drugs is used to investigate aspects of the drug use by seniors. Three sets of issues are of interest. First, what is the effect of enhanced insurance coverage on drug use and programme costs once drugs are provided free of charge? Second, is this effect permanent, or transitory? Third, are any increases in use observed concentrated among those with lower incomes? Longitudinal administrative claims payment data on 18,000 seniors over the period 1985-92 are used. All individuals in the sample turned 65 at some point and therefore became eligible for subsidized prescription drugs. Health status information is not collected; instead, health status is treated as an individual-specific fixed endowment, subject to a common rate of decay. Estimation is complicated by censoring of real drug expenditures for those under 65, rendering 'first differencing' methods invalid. A semi-parametric fixed effects Tobit estimator is used instead. For most individuals, the extension of insurance does not permanently increase drug use. Males with lower income were the exception. Little evidence of transitory effects to insurance coverage was found. Finally, the extension of insurance has made only a minor contribution to growth in seniors' drug use, relative to secular growth in drug use over time.

On becoming 65 in Ontario. Effects of drug plan eligibility on use of prescription medicines.

OBJECTIVES: The authors assess (1) the effects of first-dollar prescription drug insurance coverage provided by the Ontario Drug Benefit plan at age 65 on prescription drug use by seniors, and (2) the differential effects of this coverage on prescription drug use by seniors with varying levels of health status. METHODS: The authors modeled self-reported prescription drug use contained in the 1990 Ontario Health Survey as a function of eligibility for coverage, controlling for health status and other factors. The two-part model was used and was estimated by maximum likelihood. RESULTS: The provision of first-dollar prescription drug insurance coverage at age 65 is associated with an increase in drug use. Increases in drug use are, however, concentrated primarily among individuals with lower levels of health status. Most of the increased use occurs among individuals already under physician supervision, ie, an increase in the level of use among drug users rather than an increase in the probability of use. CONCLUSIONS: As Ontarians turn age 65 and become eligible for publicly subsidized prescription drugs, their use increases but the effect appears to be restricted mainly to persons with lower levels of health status. Given a growing trend toward reduction of public subsidy and increased reliance on patient cost sharing, more research is needed to quantify the use and health effects of such initiatives.

Does it matter whom and how you ask? inter- and intra-rater agreement in the Ontario Health Survey.

A large amount of information in the 1990 Ontario Health Survey (OHS) was collected from proxy respondents using questions administered in face-to-face interviews. Can this type of information represent candid self-reported measures of health status? Inter-rater agreement was assessed using Cohen's kappa statistic for responses to questions that were answered both by individuals about themselves and by proxies on their behalf. Intra-rater agreement, assessing the effect of mode of survey administration (in-person interviews versus self-completed written questionnaires) on the responses, was also investigated using the kappa statistic. We conclude that: (1) proxy responses in the OHS for impairments of emotion and pain are not reliable indicators of self-response (kappa < 0.32) because proxy respondents consistently under-report the burden of morbidity; (2) levels of morbidity reported by subjects to interviewer-administered questionnaires may underestimate morbidity, relative to morbidity reported by subjects using self-administered questionnaires completed in privacy. We also hypothesize that the relative magnitudes of inaccuracy introduced by interviewer administration relative to proxy reporting depends on the phenomenon being measured. When assessing pain, mode of administration is quantitatively a more important source of disagreement than type of respondent.

Cost analysis of dialysis treatments for end-stage renal disease (ESRD).

The cost of alternative dialysis modalities for the treatment of end-stage renal disease (ESRD) was evaluated, using a societal viewpoint, in a regional nephrology program in south-western Ontario. The dialysis treatments compared were hospital hemodialysis, home hemodialysis, self-care hemodialysis, and continuous ambulatory peritoneal dialysis (CAPD). The participants were all patients treated by the same dialysis modality for the fiscal year April 1990 to March 1991. Fully allocated costs are expressed in 1993 Canadian dollars. The average costs per patient year were $88,585 for hospital hemodialysis, $55,593 for self-care hemodialysis, $44,790 for CAPD, and $32,570 for home hemodialysis. The dialysis treatment costs were $54,929 for hospital hemodialysis, $43,313 for self-care hemodialysis, $31,918 for CAPD, and $26,048 for home hemodialysis. These data quantify the magnitude of the differences between fully-allocated costs among the dialysis modalities in a regional nephrology program in Canada. The methodology used in this economic analysis can be applied to programs which differ in structure and scale. The breakdown of dialysis treatment costs into overhead, support department, personnel, supplies, and medication identifies potential areas for cost reduction strategies.

A comparison of alternative models of prescription drug utilization.

The two-part estimation technique has been advocated for estimating models using individual level health care utilization data characterised by a large proportion of non-consumption and small proportions of heavy users. This paper compares the two-part model to several other estimators, including the Poisson, negative binomial and 'zero altered' negative binomial models on the basis of within-sample forecasting accuracy and non-nested model selection tests. The empirical model estimates the differential effect of the removal of copayments for prescription medicines on the prescription drug utilization by older adults with differing levels of health status. The two-part estimator of this model is found to dominate the competitors. Results from this model indicate that utilization increases appear to be higher among individuals with lower levels of health status.