Preventing Mycobacterium avium complex in patients who are using protease inhibitors: a cost-effectiveness analysis

BACKGROUND

Practice guidelines recommending Mycobacterium avium complex (MAC) prophylaxis for patients with HIV disease were based on clinical trials in which individuals did not receive protease inhibitors.

OBJECTIVE

To estimate the cost-effectiveness of strategies for MAC prophylaxis in patients whose treatment regimen includes protease inhibitors.

DESIGN

Decision analysis with Markov modelling of the natural history of advanced HIV disease. Five strategies were evaluated: no prophylaxis, azithromycin, rifabutin, clarithromycin and a combination of azithromycin plus rifabutin.

MAIN OUTCOME MEASURES

Survival, quality of life, quality-adjusted survival, health care costs and marginal cost-effectiveness ratios.

RESULTS

Compared with no prophylaxis, rifabutin increased life expectancy from 78 to 80 months, increased quality-adjusted life expectancy from 50 to 52 quality-adjusted months and increased health care costs from $233000 to $239800. Ignoring time discounting and quality of life, the cost-effectiveness of rifabutin relative to no prophylaxis was $44300 per life year. Adjusting for time discounting and quality of life, the cost-effectiveness of rifabutin relative to no prophylaxis was $41500 per quality-adjusted life year (QALY). In comparison with rifabutin, azithromycin was associated with increased survival, increased costs and an incremental cost-effectiveness ratio of $54300 per QALY. In sensitivity analyses, prophylaxis remained economically attractive unless the lifetime chance of being diagnosed with MAC was less than 20%, the rate of CD4 count decline was less than 10 x 10(6) cells/l per year, or the CD4 count was greater than 50 x 10(6) cells/l.

CONCLUSION

MAC prophylaxis increases quality-adjusted survival at a reasonable cost, even in patients using protease inhibitors. When not contraindicated, starting azithromycin or rifabutin when the patient's CD4 count is between 50 and 75 x 10(6) cells/l is the most cost-effective strategy. The main determinants of cost-effectiveness are CD4 count, viral load, place of residence and patient preference.

Improved diagnostic accuracy of inferior petrosal sinus sampling over imaging for localizing pituitary pathology in patients with Cushing's disease

The majority of patients with Cushing's disease can be cured by transsphenoidal microsurgery; however, precise localization of the pituitary source of ACTH is not always possible by standard imaging techniques. Bilateral venous sampling from the inferior petrosal sinuses (IPSS) is also useful for diagnosing Cushing's disease, but the interpretation of discordant findings between IPSS and imaging remains problematic. We tested the ability of imaging and IPSS to localize an ACTH-secreting pituitary lesion in comparison to definitive histopathological examination of the pituitary in patients with Cushing's disease (n = 37). Bilateral IPS catheterization was technically feasible in 32 patients and provided evidence of lateralization in 31 patients. Histological examination confirmed a corticotropic adenoma in 28 patients and corticotropic hyperplasia in 2 patients; Crooke's hyaline change was found in 7 patients, among whom 1 subsequently was found to have an ectopic sphenoid corticotropic adenoma, and the remainder had suspected microadenomas that were not identified microscopically. Accurate localization of the pituitary lesion was more frequent when based on IPSS results than on imaging studies (70% vs. 49%, P < 0.06). The 2 tests provided directly discrepant results for 8 patients; among these, IPSS was more likely than imaging to agree with final pathology (63% vs. 13%, P < 0.10). Imaging was entirely normal for another 9 patients, in whom IPSS accurately localized the lesion for the majority (89%; 95% confidence interval: 50-99%). We suggest that IPSS is an effective tool for localizing pituitary pathology and planning surgery for patients with Cushing's disease.

The treatment of unrelated disorders in patients with chronic medical diseases

BACKGROUND

Patients can have several illnesses concurrently, yet some of these diseases may be neglected if one problem consumes attention. We conducted a population-based analysis in Ontario, Canada - where universal health insurance is provided - to determine whether unrelated disorders are less likely to be treated in patients with chronic diseases.

METHODS

We studied the 1,344,145 residents of Ontario in 1995 who were 65 or older and eligible to receive prescription medications free of charge as part of the Ontario Drug Benefit program. Patients with diabetes mellitus were identified by prescriptions for insulin, pulmonary emphysema by prescriptions for ipratropium bromide, and psychotic syndromes by prescriptions for haloperidol. For each chronic disease, we selected an unrelated treatment: estrogen-replacement therapy for patients with diabetes mellitus, lipid-lowering medications for those with pulmonary emphysema, and medical treatment of arthritis for those with psychotic syndromes.

RESULTS

The 30,669 patients with diabetes mellitus were less likely to receive estrogen-replacement therapy than the other subjects in the study (2.4 percent vs. 5.9 percent, P<0.001). The disease was associated with a 60 percent reduction in the odds of estrogen treatment (odds ratio, 0.40; 95 percent confidence interval, 0.37 to 0.43). Findings were similar for the 56,779 patients with pulmonary emphysema, who were less likely to receive lipid-lowering medications (odds ratio, 0.69; 95 percent confidence interval, 0.67 to 0.72; P<0.001), and the 17,336 patients with psychotic syndromes, who were less likely to receive medical treatments for arthritis (odds ratio, 0.59; 95 percent confidence interval, 0.57 to 0.62; P<0.001).

CONCLUSIONS

In patients 65 or older who have chronic medical diseases and who receive prescription medications free of charge, unrelated disorders are undertreated. Clinicians caring for patients with chronic diseases should remain alert to other disorders and minimize the number of missed opportunities for treating them.

Measuring injury severity: time for a change?

BACKGROUND

The Injury Severity Score (ISS) does not take into account multiple injuries in the same body region, whereas a New ISS (NISS) may provide a more accurate measure of trauma severity by considering the patient's three greatest injuries regardless of body region. The purpose of this study was to evaluate the ISS and NISS in patients with blunt trauma.

METHODS

Consecutive individuals treated from January of 1992 to September of 1996 at one institution were included if they had sustained blunt trauma and satisfied triage standards (n = 2,328). For each patient, we computed the ISS and the NISS to determine how often the two scores were identical or discrepant. Discrepant cases were then further analyzed using receiver operating characteristic curves to determine which score better predicted short-term mortality.

RESULTS

The mean ISS was 25 +/- 13, and the mean NISS was 33 +/- 18. The two predictive scores were identical in 32% of patients and discrepant in 68% of patients. Patients with identical scores had a lower mortality rate than patients with discrepant scores (10% vs. 13%; p < 0.02). In patients with discrepant scores, the area under the receiver operating characteristic curves was greater for the NISS than the ISS (0.852 vs. 0.799; p < 0.001), and greater amounts of discrepancy were associated with increasing rates of mortality (p < 0.001).

CONCLUSIONS

The NISS often increases the apparent severity of injury and provides a more accurate prediction of short-term mortality. The benefit associated with using the NISS rather than the ISS must be weighed against the disadvantages of changing a scoring system and the potential for still greater improvements.

Shopping around for hospital services: a comparison of the United States and Canada

CONTEXT

Historical comparisons indicate that US hospitals are more expensive than Canadian hospitals, but health care system reform might have changed the relative costs and timeliness of health care in the 2 countries.

OBJECTIVE

To estimate the price and convenience of selected hospital services in the United States and Canada for patients in 1997 had they paid out-of-pocket.

DESIGN

Cross-sectional telephone survey conducted May 1996 to April 1997.

PARTICIPANTS

The 2 largest acute care general hospitals from every city in the United States and Canada with a population greater than 500000.

MEASURES

Each hospital was telephoned and asked their price and waiting time for 7 services: magnetic resonance imaging of the head without gadolinium; a screening mammogram; a 12-lead electrocardiogram; a prothrombin time measurement; a session of hemodialysis; a screening colonoscopy; and a total knee replacement. Waiting times were measured in days until earliest appointment and charges were converted to American currency.

RESULTS

Overall, 48 US and 18 Canadian hospitals were surveyed. Median waiting times were significantly shorter in American hospitals for 4 services, particularly a magnetic resonance imaging of the head (3 days vs 150 days; P<.001). Median charges were significantly higher in American hospitals for 6 services, particularly for a total knee replacement ($26805 vs $10651; P<.001). Individual services showed no association between shorter waiting times and higher prices within each country, with the exception of a total knee replacement in the United States.

CONCLUSION

US hospitals still provide higher prices and faster care than Canadian hospitals for patients who pay out-of-pocket.

Interpretation and bias in case-crossover studies

The case-crossover design is an innovative epidemiologic technique with distinct strengths and limitations. We review the fundamental logic of this self-matching non-randomized design and direct attention to 15 concerns related to the available data, unavailable data, analytic technique, quantitative statistics, and etiologic model. Implications for each concern are discussed in the context of a recent report on whether cellular telephone calls are associated with an increased risk of a motor vehicle collision. We suggest that an understanding of the case-crossover design may help investigators explore selected questions in behavioral medical research.

Long-term outcomes in open pelvic fractures

BACKGROUND

Open pelvic fractures represent one of the most devastating injuries in orthopedic trauma. The purpose of this study was to document the injury characteristics, complications, mortality, and long-term, health-related quality of life outcomes in patients with open pelvic fractures.

METHODS

The trauma registry at an adult trauma center was used to identify all multiple system blunt trauma patients with a pelvic fracture from January of 1987 to August of 1995 (n = 1,179). Demographic data, mechanism of injury, and fracture type were determined from hospital records. Short-term outcome measures included infectious complications, mortality, and length of stay in hospital. Long-term outcomes of survivors were obtained by telephone interview using the SF-36 Health Survey and the Functional Independence Measure.

RESULTS

Open pelvic fractures were uncommon, occurring in 44 patients (4%). Patients with open fractures were about 9 years younger, on average, than patients with closed fractures (30 vs. 39, p < 0.001). Similarly, patients with open fractures were more likely to be male (75 vs. 57%, p < 0.02), more likely to have been involved in a motorcycle crash (27 vs. 6%, p < 0.001), and more likely to have an unstable pelvic ring disruption (45 vs. 25%, p < 0.001). Open pelvic fracture patients required more blood than closed pelvic fracture patients, both in the first day (16 vs. 4 units, p < 0.001) and during the total hospital admission (29 vs. 9 units, p < 0.001). Five patients with perineal wounds did not receive a diverting colostomy; in turn, these individuals had a total of six pelvic infectious complications (one abscess, two with osteomyelitis, and three perineal wound infections). Overall, 11 patients died, six patients were lost to follow-up, and 27 were long-term survivors (mean duration of 4 years). Chronic disability was common after a pelvic fracture, with problems related to physical role performance and physical functioning, and was particularly severe after an open pelvic fracture (p < 0.05 for both as measured by the SF-36).

CONCLUSIONS

Patients with open pelvic fractures often survive, need to be treated with massive blood transfusions, and often require a colostomy. They are frequently left with chronic pain and residual disabilities in physical functioning and physical roles, and many remain unemployed years after injury.

Long-term outcomes in blunt trauma: who goes back to work?

BACKGROUND

Trauma patients continue to improve after discharge from the trauma center, but the completeness of this recovery remains uncertain. The purpose of this study was to compare the characteristics of patients who do and who do not return to work after blunt trauma.

METHODS

Consecutive survivors of blunt trauma discharged from a regional trauma center over a 1-year interval (July of 1994 to June of 1995) were included in the study. Patients completed the SF-36 Health Survey and some additional questions related to employment status both at discharge and again after 1 year. Our principal analysis compared patients who were employed and unemployed at 1-year follow-up.

RESULTS

Complete data were available for 195 patients. The typical patient was a young man who had been in a motor vehicle collision and had an injury severity score of 25. At 1-year follow-up, 101 patients had returned to work and 94 remained unemployed. Employed individuals were younger (31 vs. 44 years, p < 0.0001), less severely injured (mean injury severity score 23 vs. 27, p < 0.001), and more likely to hold professional jobs (50 vs. 16%, p < 0.0001). Patterns of injury and operative procedures were similar for employed and unemployed patients. However, the average employed patient had fewer days in the intensive care unit (2 vs. 5 days, p < 0.001), a shorter total hospitalization (19 vs. 28 days, p < 0.01), and was more likely to be discharged to home (62 vs. 39%, p < 0.01). At discharge, those who went on to employment had marginally better SF-36 Health Survey scores on seven of the eight scales (all except general health). During the year after discharge, both groups improved significantly, although employed individuals to a greater extent on all scales of the SF-36 Health Survey.

CONCLUSIONS

Almost one half of the multiple system blunt trauma patients remain unemployed 1 year after hospital discharge. Those patients who return to work are usually young professionals with a lower severity of injury. Functional status at discharge predicts future employment status, but underestimates the extent of long-term recovery.

Primer on medical decision analysis: Part 3--Estimating probabilities and utilities

This paper describes how to estimate probabilities and outcome values for decision trees. Probabilities are usually derived from published studies, but occasionally are derived from existing databases, primary data collection, or expert judgment. Outcome values represent quantitative estimates of the desirability of the outcome states, and are often expressed as utility values between 0 and 1. Utility values for different health states can be derived from the published literature, from direct measurement in appropriate subjects, or from expert opinion. Methods for assigning utilities to complex outcome states are described, and the concept of quality-adjusted life years is introduced.

Guidelines for verbal presentations of medical decision analyses

Individuals new to decision analysis often have difficulty with oral presentations of original research projects. This article provides general guidelines on how to present effectively. Points include: 1) articulating the research issue, 2) reviewing current beliefs, 3) portraying the study question, 4) listing the main assumptions, 5) presenting the base-case analysis, 6) showing sensitivity analyses, and 7) discussing the implications. The guidelines comment on what to exclude from presentation and how best to handle audience questions. The guidelines do not replace general instruction in public speaking (or rigorous training in decision analysis), but may help students present research projects effectively.

Primer on medical decision analysis: Part 5--Working with Markov processes

Clinical decisions often have long-term implications. Analysis encounter difficulties when employing conventional decision-analytic methods to model these scenarios. This occurs because probability and utility variables often change with time and conventional decision trees do not easily capture this dynamic quality. A Markov analysis performed with current computer software programs provides a flexible and convenient means of modeling long-term scenarios. However, novices should be aware of several potential pitfalls when attempting to use these programs. When deciding how to model a given clinical problem, the analyst must weigh the simplicity and clarity of a conventional tree against the fidelity of a Markov analysis. In direct comparisons, both approaches gave the same qualitative answers.

Primer on medical decision analysis: Part 2--Building a tree

This part of a five-part series covering practical issues in the performance of decision analysis outlines the basic strategies for building decision trees. The authors offer six recommendations for building and programming decision trees. Following these six recommendations will facilitate performance of the sensitivity analyses required to achieve two goals. The first is to find modeling or programming errors, a process known as "debugging" the tree. The second is to determine the robustness of the qualitative conclusions drawn from the analysis.

Interpreting small differences in functional status: the Six Minute Walk test in chronic lung disease patients

Functional status measurements are often difficult to interpret because small differences may be statistically significant but not clinically significant. How much does the Six Minute Walk test (6MW) need to differ to signify a noticeable difference in walking ability for patients with chronic obstructive pulmonary disease (COPD)? We studied individuals with stable COPD (n = 112, mean age = 67 yr, mean FEV1 = 975 ml) and estimated the smallest difference in 6MW distances that was associated with a noticeable difference in patients' subjective comparison ratings of their walking ability. We found that the 6MW was significantly correlated with patients' ratings of their walking ability relative to other patients (r = 0.59, 95% confidence interval [CI]: 0.54 to 0.63). Distances needed to differ by 54 m for the average patient to stop rating themselves as "about the same" and start rating themselves as either "a little bit better" or "a little bit worse" (95% CI: 37 to 71 m). We suggest that differences in functional status can be statistically significant but below the threshold at which patients notice a difference in themselves relative to others; an awareness of the smallest difference in walking distance that is noticeable to patients may help clinicians interpret the effectiveness of symptomatic treatments for COPD.

Primer on medical decision analysis: Part 1--Getting started

This paper is Part 1 of a five-part series covering practical issues in the performance of decision analysis. The intended audience is individuals who are learning how to perform decision analyses, not just read them. The series assumes familiarity with the basic concepts of decision analysis. It imparts many of the recommendations the authors have learned in teaching a one-semester course in decision analysis to graduate students. Part 1 introduces the topic and covers questions such as choosing an appropriate question, determining the tradeoff between accuracy and simplicity, and deciding on a time frame.

Primer on medical decision analysis: Part 4--Analyzing the model and interpreting the results

This paper is the fourth of a five-part series that describes the principles of construction and evaluation of valid decision models. In this review, the authors describe the key principles of detecting and eliminating structural and programming errors in decision trees (debugging). In addition, they offer guidelines to facilitate the interpretation of analytic results of decision models.

Association between cellular-telephone calls and motor vehicle collisions

BACKGROUND

Because of a belief that the use of cellular telephones while driving may cause collisions, several countries have restricted their use in motor vehicles, and others are considering such regulations. We used an epidemiologic method, the case-crossover design, to study whether using a cellular telephone while driving increases the risk of a motor vehicle collision.

METHODS

We studied 699 drivers who had cellular telephones and who were involved in motor vehicle collisions resulting in substantial property damage but no personal injury. Each person's cellular-telephone calls on the day of the collision and during the previous week were analyzed through the use of detailed billing records.

RESULTS

A total of 26,798 cellular-telephone calls were made during the 14-month study period. The risk of a collision when using a cellular telephone was four times higher than the risk when a cellular telephone was not being used (relative risk, 4.3; 95 percent confidence interval, 3.0 to 6.5). The relative risk was similar for drivers who differed in personal characteristics such as age and driving experience; calls close to the time of the collision were particularly hazardous (relative risk, 4.8 for calls placed within 5 minutes of the accident, as compared with 1.3 for calls placed more than 15 minutes before the accident; P<0.001); and units that allowed the hands to be free (relative risk, 5.9) offered no safety advantage over hand-held units (relative risk, 3.9; P not significant). Thirty-nine percent of the drivers called emergency services after the collision, suggesting that having a cellular telephone may have had advantages in the aftermath of an event.

CONCLUSIONS

The use of cellular telephones in motor vehicles is associated with a quadrupling of the risk of a collision during the brief time interval involving a call. Decisions about regulation of such telephones, however, need to take into account the benefits of the technology and the role of individual responsibility.

Assessing the minimal important difference in symptoms: a comparison of two techniques

We have developed a method for estimating the minimally important difference (MID) for health status measures. Whereas the conventional approach requires patients to judge themselves relative to their memories, our method requires patients to judge themselves relative to others with the same condition. In this study we examined whether our method (based on between-patient differences) and the conventional method (based on within-patient changes) provides comparable estimates of the MID for one health status measure: the Chronic Respiratory Questionnaire. Patients with chronic obstructive pulmonary disease who were participating in a supervised respiratory rehabilitation program were included if they were in stable health (n = 112). Their mean score per question in the Chronic Respiratory Questionnaire was 4.5 (range, 1 to 7; where bigger values indicate better health). Our method estimated that the MID was 0.5 (95% confidence interval 0.4 to 0.7). This estimate was similar to the MID previously found using the conventional method. These observations support the role of the Chronic Respiratory Questionnaire for measuring patient's symptoms, the validity of our approach for assessing the MID, and an estimate on the order of 0.5 as the threshold for this particular health status measure.

Patients' memories of painful medical treatments: real-time and retrospective evaluations of two minimally invasive procedures

Patients' memories of painful medical procedures may influence their decisions about future treatments, yet memories are imperfect and susceptible to bias. We recorded in real-time the intensity of pain experienced by patients undergoing colonoscopy (n = 154) and lithotripsy (n = 133). We subsequently examined patients' retrospective evaluations of the total pain of the procedure, and related these evaluations to the real-time recording obtained during the experience. We found that individuals varied substantially in the total amount of pain they remembered. Patients' judgments of total pain were strongly correlated with the peak intensity of pain (P < 0.005) and with the intensity of pain recorded during the last 3 min of the procedure (P < 0.005). Despite substantial variation in the duration of the experience, lengthy procedures were not remembered as particularly aversive. We suggest that patients' memories of painful medical procedures largely reflect the intensity of pain at the worst part and at the final part of the experience.

Spirometry and dyspnea in patients with COPD. When small differences mean little

OBJECTIVE

To determine when a difference in FEV1 is sufficiently large to be associated with a noticeable difference in dyspnea symptoms for patients with chronic lung disease.

DESIGN

Cross-sectional analysis of 15 groups (n = 112 patients, 832 contrasts).

SETTING

Respiratory rehabilitation program.

PATIENTS

Patients with COPD (mean FEV1 = 35% predicted).

MEASURES

Patients' perspectives assessed through subjective comparison ratings of dyspnea and of overall health. Relation between the FEV1 and patients' perspectives determined the smallest difference in spirometry that was associated with a noticeable difference in patients' symptoms.

RESULTS

The FEV1 was moderately correlated with patients' ratings of dyspnea (r = 0.29; 95% confidence interval (CI), 0.22 to 0.35). In contrast, the FEV1 was minimally correlated with patients' ratings of overall health (r = 0.10; 95% CI, 0.03 to 0.17). The FEV1 needed to differ by 4% predicted for the average patient to stop rating his or her dyspnea as "about the same" and start rating his or her dyspnea as either "a little bit better" or "a little bit worse" relative to other patients (95% CI, 1.5 to 6.5). This was equivalent to the average patient's FEV1 increasing by 112 mL (starting from 975 mL and ending at 1,087 mL).

CONCLUSIONS

Some statistically significant differences in the FEV1 are so small that they may not represent important differences in symptoms for the average patient with severe COPD; an awareness of the smallest difference in FEV1 that is noticeable to patients can help clinicians interpret the effectiveness of symptomatic treatments.

On the belief that arthritis pain is related to the weather

There is a widespread and strongly held belief that arthritis pain is influenced by the weather; however, scientific studies have found no consistent association. We hypothesize that this belief results, in part at least, from people's tendency to perceive patterns where none exist. We studied patients (n = 18) for more than I year and found no statistically significant associations between their arthritis pain and the weather conditions implicated by each individual. We also found that college students (n = 97) tend to perceive correlations between uncorrelated random sequences. This departure of people's intuitive notion of association from the statistical concept of association, we suggest, contributes to the belief that arthritis pain is influenced by the weather.

In vivo responsiveness of morphological variants of growth hormone-producing pituitary adenomas to octreotide

The somatostatin analog, octreotide, is an inhibitor of growth hormone (GH) secretion that has been used to treat patients with GH-producing pituitary tumors. In this study we investigated the in vivo responsiveness to treatment with this analog in patients harboring different morphological types of GH-producing pituitary adenomas. Both GH and insulin-like growth factor I (IGF-I) plasma levels in 30 patients treated with octreotide (300 micrograms/day) for 4 months preoperatively were compared with those from 30 patients who did not receive treatment preoperatively. Tissue samples were studied using ultrastructural and immunohistochemical techniques. Amongst patients harboring densely granulated (DG) adenomas, mean GH levels were reduced to 32 +/- 9% by octreotide, to 30 +/- 7% by surgery and to 26 +/- 9% of baseline by both interventions. Surgery was equally as effective in lowering GH levels in patients with sparsely granulated (SG) adenomas as it was in those with DG adenomas; in patients with SG adenomas, GH levels were reduced by surgery alone to 37 +/- 16% and to 24 +/- 15% when performed following octreotide pretreatment. In contrast, treatment with octreotide alone in patients harbouring SG adenomas reduced GH levels to only 70 +/- 13% of baseline (p < 0.02 compared to surgery alone, or surgery and octreotide). We conclude that the GH inhibitory effects of octreotide are significantly better in patients harboring DG somatotroph adenomas compared with those harboring SG adenomas.