A comprehensive review of predictive and prognostic composite factors implicated in the heterogeneity of treatment response and outcome across disease areas

AIM

To assess and present the current body of evidence regarding composite measures associated with differential treatment response or outcome as a result of patient heterogeneity and to evaluate their consistency across disease areas.

METHODS

A comprehensive review of the literature from the last 10 years was performed using three databases (PubMed, Embase and Cochrane). All articles that met the inclusion/exclusion criteria were selected, abstracted and assessed using the NICE level-of-evidence criteria.

RESULTS

Forty-nine studies were identified in the data abstraction. Approximately one-third focused on existing composite measures, and the rest investigated emerging composite factors. The majority of studies targeted patients with cancer, cardiovascular disease or psychological disorders. As a whole, the composite measures were found to be disease-specific, but some composite elements, including age, gender, comorbidities and health status, showed consistency across disease areas. To complement these findings, common individual factors found in five previous independent disease-specific literature assessments were also summarised, including age, gender, treatment adherence and satisfaction, healthcare resource utilisation and health status.

CONCLUSIONS

Composite measures can play an important role in characterising heterogeneity of treatment response and outcome in patients suffering from various medical conditions. These measures can help clinicians to better distinguish between patients with high likelihood to respond well to treatment and patients with minimal chances of positive therapeutic outcomes. Herein, the individual factors identified can be used to develop novel predictive or prognostic composite measures that can be applicable across disease areas. Reflecting these cross-disease measures in clinical and public health decisions has the distinctive appeal to enable targeted treatment for patients suffering from multiple medical conditions, which may ultimately yield significant gains in individual outcomes, population health and cost-effective resource allocation.

Temporal trends in first-line chemotherapy treatment among elderly stage IIIB/IV non-small cell lung cancer (NSCLC) patients in the United States: Evidence from SEER-Medicare data

17002

Background: Data regarding first-line chemotherapy treatment among advanced-staged NSCLC patients is lacking. The purpose of this analysis was to assess first-line treatment patterns over time among elderly patients with Stage IIIB/IV NSCLC. Methods: Patients aged 65 years and older diagnosed with Stage IIIB/IV NSCLC between 1997 and 2002 were identified and followed over time using the SEER-Medicare database to evaluate temporal trends in first-line chemotherapy treatment. Results: Chemotherapy use increased from approximately 28% of Stage IIIB/IV patients diagnosed in 1997 to 36% of patients diagnosed in 2002. Among identifiable first-line treatments, cisplatin or carboplatin and taxane (C/CT) doublet therapy was the most common treatment in all years (approximately 50% or greater). The use of single-agent taxanes and gemcitabine increased over time (taxane use increased from 3.3% in 1997 to 7.6% in 2002; gemcitabine use increased from 0.6% in 1997 to 5.1% in 2002), while use of single agent cisplatin or carboplatin (C/C) declined (from 10.0% in 1997 to 3.1% in 2002). Use of doublet therapy with C/C and either a taxane or gemcitabine also increased over time (with the largest increase for the gemcitabine combination from 0.3% in 1997 to 11.8% in 2002). Correspondingly, use of doublet C/C and any other therapy declined markedly (from 18.9% in 1997 to 3.7% in 2002). Use of 3 or more agents as first-line treatment was infrequent across study years (<1.0%). Treatment discontinuation was substantial (>60% for all regimens). Conclusions: Our findings indicate relatively low but increasing use of first-line chemotherapy treatment among elderly Stage IIIB/IV NSCLC patients.

[Table: see text]

Association between clinically depressed subgroups, type of treatment and patient retention in the LIDO study

BACKGROUND

Few data are available with which to evaluate the association between depressed subgroups, type of treatment and patient retention during episodes of major depression.

METHOD

This observational study followed 1117 depressed patients over a 12-month period in the primary care setting of six different international sites. The patients were divided into three severity-linked subgroups: moderate to severe depression; moderate depression co-morbid with serious medical conditions; and mild depression.

RESULTS

In general, a low dropout rate was found, with significant differences in the rates across the six sites. However, while there was no statistical significance in the association between the three subgroups of depression and overall dropout rates, we did find that older patients were less likely to drop out, more depressed patients were more likely to drop out, and if patients were on antidepressants they were less likely to drop out. Among the three subgroups of depression, patients with moderate depression co-morbid with serious medical conditions received the lowest amount of antidepressants and had the lowest quality of life.

CONCLUSION

Although the overall dropout rate in this study was found very low and did differ between the six sites, an association between the use of antidepressants and patient retention was seen. The group of patients with serious co-morbid medical conditions received fewer antidepressants even when the level of their depressive states was taken into consideration. This group was the least satisfied with treatment and had the lowest self-reported quality of life.

Predictors of a medical-offset effect among patients receiving antidepressant therapy

OBJECTIVE

Characteristics of patients receiving antidepressant therapy were examined to identify factors that may be associated with a medical-offset effect.

METHOD

In a retrospective study, the authors analyzed claims data from a large health insurer in New England. The study subjects included 1,661 persons initiating treatment for depression with selective serotonin reuptake inhibitors or tricyclic antidepressants between July 1991 and June 1993.

RESULTS

Patients with anxiety disorders, coronary heart disease, cancer, and chronic fatigue syndrome and those remaining on their initial regimens of antidepressant therapy for at least 6 months were more likely to experience significant reductions in the costs of medical care services. The number of visits to mental health providers had no effect on the costs of medical services.

CONCLUSIONS

Specific comorbid conditions and sustained use of antidepressant drugs may be associated with a medical-offset effect for patients receiving treatment for depression.

Primary medical care and health outcome

In order to examine the relationship between primary medical care and health outcome, a systemic sample of 485 persons in a typical metropolitan Midwest county was surveyed. Primary care was defined in functional terms including concepts of ready access, continuity over time, comprehensiveness, coordination, and personalized care. Health outcome encompassed self-reported health status, disability, patient satisfaction, and medical costs. Persons who had high levels of primary care tended to have more chronic medical problems and saw physicians more frequently, especially their own principal physician. They clearly were more satisfied with their medical care, but incurred higher medical costs. They did not differ from persons with a low level of primary care in current health status or disability. It was not possible to demonstrate that quality primary care led to better health outcome except on the parameter of patient satisfaction.

The role of a medical education center in location decisions of practicing physicians

The role of a medical education center in the decisions of physicians on where to practice was examined for all physicians locating in Rockford, Illinois, during 1978-1980. Questions were asked concerning faculty status, factors influencing practice location decisions, and perceived advantages of association with a medical education center. The results indicated that a majority of the 57 responding physicians did take the center into account when deciding to locate in Rockford; however, location in a mid-sized city and an opportunity for joint practice proved to be more influential factors than the possibility of a faculty appointment. Thus, while seldom the prime factor, the possibility of a faculty appointment. Thus, while seldom the prime factor, the medical education center was an important secondary factor and perhaps provided the margin of influence necessary to decide to locate in Rockford for those physicians inclined toward a smaller community.