Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “confidence limits”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 379 records · Page 21Linked to original sources
PubMedDate not supplied

Validity of urine glucose measurements for estimating plasma glucose concentration.

Both physicians and diabetic patients have traditionally relied on measurement of glycosuria as an indirect method of estimating plasma glucose concentration to guide adjustment of insulin and diet therapy. Our observations on the correlation between mean plasma glucose concentration with simultaneous urine glucose concentration or excretion rate re-emphasize the limitations of this approach. Although our observations show a significant correlation (P less than 0.0001) between plasma glucose concentration and urine glucose concentration or urine glucose excretion rate, the wide confidence limits [95% confidence limits (minimum) +/- 150 mg/dl] on plasma glucose concentration estimated from urine glucose measurements limit the clinical applicability of such estimates. Differences among subjects in the renal resorption of glucose contribute to the wide variance of estimates. However, significant variability in renal glucose resorption within individuals is documented, further reinforcing the limitations of urine glucose determinations for reliable estimates of plasma glucose concentrations. Diabetologists need to reconsider the applicability of urine glucose measurements in evaluation of adequacy of therapy and in adjustment of insulin dosage.

Adolescent↗
PubMed1994

Clinical evaluation of nitroimidazoles as modifiers of hypoxia in solid tumors.

Hypoxic modification by nitroimidazoles has been explored in a large number of clinical studies. Nine different drugs (misonidazole, metronidazole, benznidazole, desmethylmisonidazole, etanidazole, pimonidazole, nimorazole, ornidazole, and RSU 1069) have reached clinical evaluation. Phase I and II trials have demonstrated a relationship between drug tolerance and expected hypoxic modification. Thus, the most tolerable drugs, viz., 5-nitroimidazoles, are those with the smallest preclinical activity. However, they may be the most clinically active, due to higher plasma and tumor concentrations. The clinical evaluation of hypoxic modification is difficult, since clinically important hypoxia can be observed only indirectly. There is, however, indirect evidence of substantial hypoxia in human tumors, although with considerable heterogeneity among individual tumors. More than 7000 patients have been included in 50 randomized trials. A meta-analysis showed that modification of tumor hypoxia significantly improved the loco-regional tumor control after radiotherapy with an odds ratio of 1.17 (95% confidence limits 1.06 to 1.28). The treatment benefit could mostly be related to an improved response in head and neck with odds ratio 1.23 (95% confidence limits 1.09 to 1.37), and to a lesser extent in bladder tumors; no significant effect was observed in other tumor sites (cervix, lung, central nervous system, and esophagus). Similarly to the local control benefit, the overall survival rate was improved with an overall odds ratio of 1.13 (95% confidence limits 1.03 to 1.23). The clinical trials showed no evidence of significant chemosensitization or direct bioreductive effect. Although 50 randomized clinical trials were evaluated, the median number of patients was only 97 (range 17-620). Clinical trials of this size are not likely to detect the observed differences, which may explain the lack of significant improvement in most of the individual studies. However, the overall results indicate that the biological issue related to hypoxia appears to be a sound rationale, especially with regard to head and neck and bladder carcinoma. Future studies related to the hypoxic problem should therefore preferably be aimed towards such tumors.

Antineoplastic Agents↗
PubMed1991

Meta-analysis: methods for combining data to improve quantitative risk assessment.

Although individual studies that constitute the data base for a risk assessment are each evaluated quantitatively as well as qualitatively, assessment of the total data base frequently results in selection of data from particular studies, rather than an effort to combine data quantitatively. Meta-analysis, or the analysis of analyses, provides an approach for the joint evaluation of the results of several studies. In this report, two very similar cancer bioassays of trichloroethylene were used to illustrate some simple meta-analytic techniques and to evaluate the validity and value of these procedures. The results demonstrate that concepts such as the upper 95% confidence limit are highly dependent upon assumptions if several data sets are involved. Use of a Monte Carlo procedure resulted in an increase in the 95% upper confidence limit relative to the value determined by EPA, but by determining the variance of the maximum likelihood estimate of the linear coordinate of the estimated dose-response curve, the 95% upper confidence limit of the cancer potency factor was reduced.

Dose-Response Relationship, Drug↗
PubMed1987

Moderate alcohol consumption and the risk of breast cancer.

In 1980, 89,538 U.S. women 34 to 59 years of age, with no history of cancer, completed an independently validated dietary questionnaire that included the use of beer, wine, and liquor. During the ensuring four years, 601 cases of breast cancer were diagnosed among cohort members. Among the women consuming 5 to 14 g of alcohol daily (about three to nine drinks per week), the age-adjusted relative risk of breast cancer was 1.3 (95 percent confidence limits, 1.1 and 1.7). Consumption of 15 g of alcohol or more per day was associated with a relative risk of 1.6 (95 percent confidence limits, 1.3 and 2.0; Mantel extension chi for linear trend = +4.2; P less than 0.0001). Adjustment for known breast cancer risk factors and a variety of nutritional variables did not materially alter this relation. Significant associations were observed for beer and liquor when considered separately. Among women without risk factors for breast cancer who were under 55 years of age, the relative risk associated with consumption of 15 g of alcohol or more per day was 2.5 (95 percent confidence limits, 1.5 and 4.2). These prospective data derived from measurements of alcohol intake recorded before the diagnosis of breast cancer confirm the findings of several previous case-control studies. Viewed collectively, they suggest that alcohol intake may contribute to the risk of breast cancer.

Adult↗
PubMed2000

Visual function thresholds in children.

PURPOSE: The aim of the present study was to establish a range of values for individual visual function measurements in Primary 1 schoolchildren. METHODS: 93 visually normal children with a mean age of 5.4 +/- 0.3 years were recruited with parental consent. Measures of visual acuity obtained included 1) High Contrast (CAT) 2) Low Contrast (LC) 3) Single Letter (SLA) and 4) Repeat Letter Acuity (RRL). These measurements were randomised and obtained during the same visit. Hyperacuity thresholds were obtained at a separate visit, using both Vernier and oscillatory motion displacement stimuli (OMDT). Re-test data for all tests was collected from a random sample of 30 children after a period of at least two weeks. RESULTS: For all measures the distribution of scores was found to be normal. All visual acuity scores are in log units. Mean ( +/- SD) values for each test were as follows; CAT = 0.11 +/- 0.07; LC = 0.38 +/- 0.08; RRL = 0.04 +/- 0.07; SLA = -0.02 +/- 0.09; Vernier threshold = 69 +/- 21 arcsec; OMDT = 60 +/- 9 arc sec. The confidence limits of the test-retest discrepancies were as follows; CAT = +/- 0.05; LC = +/- 0.07; RRL = +/- 0.04; SLA = +/- 0.04; Vernier = +/- 17 arcsec; OMDT = +/- 6 arcsec. The confidence limits of the interocular discrepancies were as follows; CAT = +/- 0.08; LC = +/- 0.08; RRL = +/- 0.08; SLA = +/- 0.11; Vernier = +/- 14 arcsec; OMDT = +/- 11 arcsec. CONCLUSIONS: To fully describe an individual's visual capacity it is important that new tests of visual function are developed based on emerging knowledge of visual physiology. In this study confidence limits can be constructed for each visual function measure, test-retest and interocular discrepancies using the normal distributions found above. This will provide the clinician with further age-matched values and contribute to our understanding of functional visual development.

Child, Preschool↗
PubMed1996

Physiological variation of serum prostate specific antigen in the 4.0 to 10.0 ng./ml. range in male volunteers.

PURPOSE: Because some patients show a surprising variation in serial serum prostate specific antigen (PSA) values, we determined the intra-individual or physiological variation in serum PSA by collecting sera 2 to 3 week apart without any prostatic manipulation. MATERIALS AND METHODS: Because 4.0 to 10.0 ng./ml. PSA is the critical range for decision making, we asked all men with a PSA in this range to return 2 to 3 weeks later for a second measurement. Total serum PSA was determined by the Hybritech Tandem-R, automated Tosoh AIA-600 and Delfia section immunoassays. Free and complexed serum PSA was determined by the Delfia assays. Between assay variation (first blood specimen retested on a separate day with the second blood specimen) was compared to the physiological variation (first versus second blood specimens). RESULTS: Mean coefficient of variation (95% confidence limits) was 10.5% for between assay and 23.5% for physiological evaluations. The preferred analysis of ratio difference variation provided a factor of 0.138 (between assay) and 0.298 (physiological) for 95% confidence limits. Changes in free or complexed PSA were not the cause of physiological variation. CONCLUSIONS: The intra-individual physiological variation is 2 to 3 times the between assay variation for sera drawn 2 to 3 weeks apart with a PSA of 4 to 10 ng./ml. A serum PSA of 4.0 ng./ml. can increase to 5.2 ng./ml. (4.0 x 0.298) and still be within the physiological variability for 95% confidence limits.

Blood Specimen Collection↗
PubMed1987

Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-up Study.

A cohort of 2270 white women, aged 40-69 years at baseline, were followed for an average of 8.5 years in the Lipid Research Clinics Program Follow-up Study. There were 44 deaths due to cardiovascular disease among the 1677 nonusers of estrogens and six cardiovascular disease deaths among the 593 estrogen users. The age-adjusted relative risk (RR) of cardiovascular disease deaths in users compared with nonusers was 0.34 (95% confidence limits 0.12 to 0.81). After multivariable adjustment for potential confounding factors (age, blood pressure, and smoking), the estimated RR for estrogen use was 0.37 (95% confidence limits 0.16 to 0.88). Analyses were done to explore whether these results could be due to selection bias for estrogen use. However, the prevalence of cardiovascular disease at baseline was slightly higher in estrogen users (12%) than in nonusers (10%); furthermore, the exclusion of all women with prevalent cardiovascular disease at baseline did not alter the apparent protective effect of estrogen use on cardiovascular disease mortality (RR = 0.42, 95% confidence limits 0.13 to 1.10). Additional analyses examining the complex association between estrogen use, lipoprotein levels, and cardiovascular disease mortality suggest that the protective effect of estrogen is substantially mediated through increased high-density lipoprotein levels.

Adult↗
PubMed1991

The subsequent risk of internal cancer with Bowen's disease. A population-based study.

OBJECTIVE: To address the subsequent risk of internal cancer in a population-based cohort of patients with Bowen's disease. DESIGN: Incident cases of skin cancers other than malignant melanoma occurring in a defined population were classified as basal cell carcinoma, squamous cell carcinoma, or Bowen's disease. Through medical records, all patients were followed up for the development of subsequent internal cancer until they died, moved from Rochester, Minn, or January 1, 1986, whichever came first. Incidence rates of skin cancer and subsequent cancer were computed; the subsequent rate of internal cancer was compared with that prevailing in the community. PATIENTS: Enrolled were all permanent residents in the population base of Rochester, Minn, who developed basal cell carcinoma (n = 657), squamous cell carcinoma (n = 169), or Bowen's disease (n = 71) on the basis of pathologic examination and clinical presentation from 1976 through 1984. MAIN OUTCOME MEASURE: The relative risk of subsequent internal cancer among patients with Bowen's disease compared with that of the population base from which they arose was 1.1 (95% confidence limits, 0.5, 1.6). RESULTS: Annual incidence rates, adjusted to the 1980 US white population, were 14.9 per 100,000 for Bowen's disease, 38.8 for squamous cell carcinoma, and 146 for basal cell carcinoma. The estimated relative risk for subsequent cancer was 0.9 (95% confidence limits, 0.5, 1.6) among patients with squamous cell carcinoma and 1.0 (95% confidence limits, 0.7, 1.3) for patients with basal cell carcinoma. These risks were not significantly different for various durations of follow-up or for sun-exposed vs sun-protected sites. CONCLUSIONS: We find no evidence in these population-based cohort data of and increased risk of subsequent internal cancer associated with Bowen's disease or other forms of nonmelanomatous skin cancer.

Adult↗
PubMed1984

Haemoglobinopathies, malaria, and other interferences with HBA1 assessment.

Total glycosylated haemoglobin (HbA1) was determined by a rapid minicolumn chromatography technique in 438 diabetic patients and correlated with the mean of fasting and post-prandial blood glucose values for the preceding six weeks. In 360 of them, free of congenital haemoglobinopathies and other detected causes of HbA1 mis-interpretation (reference group), a significant correlation was established between the HbA1 and glucose values: y = 0.54 X + 4.91; r = 791; (p less than 0.01). In 28 of the 29 patients with heterozygous haemoglobinopathies (Hb S = 17; Hb C = 8; Hb D Pundjab = 1; Hb E = 2) the apparent HbA1 values were inappropriately low. The apparent HbA1 value was above the 95% confidence limits in the 29th patient, with beta thalassaemia. In 10 out of 14 diabetics with recurrent hypoglycemic attacks, the HbA1 value was lower than the 95% confidence limits of expected values. Out of 21 diabetics with a shortened red cell lifespan (occult blood losses: 10; haemolysis: 11) 15 displayed a lower than expected HbA1 value. Among these was a diabetic patient with malaria and severe anaemia. Out of 14 diabetics with severe chronic renal failure only 3 presented with apparent HbA1 values above the 95% confidence limits.

Adult↗
PubMed1988

Aortic coarctation with hypoplastic aortic arch. Results of extended end-to-end aortic arch anastomosis.

Between 1980 and 1986, 80 infants (less than or equal to 3 months old) with symptomatic aortic coarctation and associated severe tubular hypoplasia of the transverse aortic arch underwent surgical treatment. Extended end-to-end aortic arch anastomosis was used in an attempt to correct both the isthmic stenosis and the hypoplasia of the transverse arch. After complete excision of the coarctation tissue, a long incision was made in the inferior aspect of the aortic arch, which was then anastomosed to the obliquely trimmed distal aorta. Pure coarctation was present in 17 patients (group I); 24 infants had an additional ventricular septal defect (group II), and 39 patients had associated complex heart disease (group III). The overall early mortality rate was 26% (confidence limits 21% to 32%) (18% in group I, 17% in group II, and 36% in group III). The early risk declined with time and was 18% (confidence limits 12% to 26%) for the last 2 years (seven deaths in 39 patients). Follow-up was 100% for a mean of 19 months. Actuarial survival rate at 3 years was 82% for group I, 78% for group II, and 32% for group III. Recurrent coarctation (gradient greater than or equal to 20 mm Hg) occurred in six operative survivors (10%, confidence limits 6% to 16%) and necessitated reoperation in three. Freedom from recoarctation at 4 years was 88%. Because extended end-to-end aortic arch anastomosis provides adequate correction of the aortic obstruction and entails a low risk of restenosis, it is our procedure of choice in infants with coarctation and severe hypoplasia of the aortic arch.

Actuarial Analysis↗
PubMedDate not supplied

An atlas of forecasted molecular data. 2. Vibration frequencies of main-group and transition-metal neutral gas-phase diatomic molecules in the ground state.

This atlas of diatomic-molecular vibration frequencies parallels the previously offered Atlas of Internuclear Separations. The Atlas was produced by mining the data from Huber and Herzberg and training neural network software to forecast new data. New protocols were employed with the powerful software, which was originally designed for forecasting the financial markets. The Atlas presents 1920 additional vibration frequencies for use until critical tables are available to fill the needs more precisely. The precision of the predictions is characterized by the average fractional 1% confidence limit, that is, 10.66%. The accuracies of the predictions are determined in two ways. First, 221 of the 224 Huber and Herzberg data values used for training and validation fall within the prediction confidence limits or fall outside by less than 10% of the Huber and Herzberg values, and 181 values agree (within the limits). Second, 87 of 101 comparison data values, consisting of literature data and some additional Huber and Herzberg values, fall within the prediction confidence limits or fall outside by less than half the prediction values, and 44 of the 101 values agree (within the limits).

Journal Article↗
PubMed2002

Comparison of vaccination strategies for the control of dog rabies in Machakos District, Kenya.

Demographic and epidemiological field data were used in a deterministic model to describe dog rabies transmission in Machakos District, Kenya and to predict the impact of potential vaccination strategies for its control. The basic reproduction number (R0) was estimated to be 2.44 (1.52-3.36, 95% confidence limits). There were three key model predictions. The first was that a threshold dog density (K(T)) of 4.5 dogs km(-2) (3.8-5.2 dogs km(-2), 95% confidence limits) was required to maintain transmission. The second was that the estimated annual vaccination rate of 24% failed to decrease incidence and actually increased the stability of transmission and may be counter-productive. Thirdly, to control rabies, it was predicted that 59% (34%-70%, 95% confidence limits) of dogs should be vaccinated at any one time. This requires approximately 70% coverage for annual but only 60% coverage for semi-annual vaccination campaigns. Community-level vaccination trials are needed to test these predictions.

Animals↗
PubMed2000

The long-term fluctuation of the visual field in stable glaucoma.

PURPOSE: To determine, in stable glaucoma, the characteristics of the between-examination variability of the visual field recorded with the Humphrey Field Analyser (HFA; Humphrey Systems, Dublin, CA) using the homogeneous, LF(Ho), and heterogeneous, LF(He), components of the long-term fluctuation (LF), thereby providing a technique for separating progressive loss from fluctuation in sensitivity. METHODS: The LF components were calculated using a two-factor analysis of variance (ANOVA) with replications and were determined between each pair of three successive HFA program 30-2 fields for each patient from two groups, each containing 30 patients with primary open-angle glaucoma. The interval between examinations for the first group was 6 to 9 months and for the second group was 3 weeks. RESULTS: The group mean values for LF(Ho) ranged from 1.50 to 2.19 dB and for LF(He) from 1.70 to 2.05 dB. The average difference between examinations was within +/-0.35 dB for each component, and the 95% limits of agreement for the two groups, respectively, were +/- 2.31 and +/- 2.39 dB for the LF(Ho) and +/- 2.36 and +/- 2.09 dB for the LF(He). The estimate of the 90% confidence limit for the LF(Ho) was 3.30 dB and for the LF(He), 3.60 dB. Little relationship was present between the LF components and the modulus differences in mean deviation (MD), the corrected pattern SD (CPSD), or the mean MD, mean short-term fluctuation, and mean CPSD, of the two fields. CONCLUSIONS: Estimation of the LF components and of the corresponding confidence limits yields an expression of the normal between-examination variability of two consecutive fields that can be used as a reliability index. A value outside the confidence limits indicates the necessity for a confirmatory follow-up field.

Aged↗
PubMed1984

The incidence of primary cardiac arrest during vigorous exercise.

To examine the risk of primary cardiac arrest during vigorous exercise, we interviewed the wives of 133 men without known prior heart disease who had had primary cardiac arrest. Cases were classified according to their activity at the time of cardiac arrest and the amount of their habitual vigorous activity. From interviews with wives of a random sample of healthy men, we estimated the amount of time members of the community spent in vigorous activity. Among men with low levels of habitual activity, the relative risk of cardiac arrest during exercise compared with that at other times was 56 (95 per cent confidence limits, 23 to 131). The risk during exercise among men at the highest level of habitual activity was also elevated, but only by a factor of 5 (95 per cent confidence limits, 2 to 14). However, among the habitually vigorous men, the overall risk of cardiac arrest--i.e., during and not during vigorous activity--was only 40 per cent that of the sedentary men (95 per cent confidence limits, 0.23 to 0.67). Although the risk of primary cardiac arrest is transiently increased during vigorous exercise, habitual vigorous exercise is associated with an overall decreased risk of primary cardiac arrest.

Adult↗
PubMed1987

A case-control study of breast cancer and hormonal contraception in Costa Rica.

By 1981, 11% of married women in Costa Rica ages 20-49 years had used depot-medroxyprogesterone acetate (DMPA) and 58% had used oral contraceptives (OCs). Since 1977, the Costa Rican Ministry of Health has maintained a nationwide cancer registry. These circumstances provided an opportunity for a population-based, case-control study of DMPA, OCs, and breast cancer in Costa Rica. Cases were 171 women ages 25-58 years with breast cancer diagnosed between 1982 and 1984; controls were 826 women randomly chosen during a nationwide household survey. Cases and controls were interviewed with the use of a standard questionnaire covering their reproductive and contraceptive histories. Logistic regression methods were used to adjust for confounding factors. While few cases or controls had ever used DMPA, DMPA users had an elevated relative risk (RR) estimate of breast cancer of 2.6 (95% confidence limits = 1.4-4.7) compared with never users. However, no dose-response relationship was found; even the group of women who had used DMPA for less than 1 year had an elevated RR estimate (RR = 2.3; 95% confidence limits = 1.0-5.1). In contrast, OC users had no elevation in RR compared with never users (RR = 1.2; 95% confidence limits = 0.8-1.8). The results of the DMPA analysis are inconclusive. Before decisions are made on whether to continue providing this effective contraceptive method, other ongoing studies will need to confirm of refute these findings.

Adult↗
PubMed1988

Paget's disease of the anal margin.

A follow-up study of 22 patients with Paget's disease of the anal margin was conducted to determine the prevalence of invasive disease and coexisting visceral carcinomas, cure and survival rates. Seventy-three per cent of the patients suffered from persistent pruritus ani. Seven (32 per cent) patients had malignancy (invasive Paget's disease (n = 5) and coexisting visceral carcinoma (n = 2] and six (27 per cent) developed malignancy (invasive Paget's disease (n = 4) and coexisting visceral carcinoma (n = 2] during follow-up, giving an overall occurrence of malignancies of 59 per cent (95 per cent confidence limits: 36-79). The median follow-up period was 9 (range 0.5-25) years. No difference in length of history could be found between patients with or without malignancy. The estimated cure rate by actuarial analysis among radically treated patients was 64 per cent (95 per cent confidence limits: 43-91) at 1 year and 45 per cent (95 per cent confidence limits: 18-72) at 8 years after primary treatment. The 5- and 10-year crude survival rates of 54 per cent and 45 per cent, respectively, were significantly lower than the corresponding values of 84 per cent and 60 per cent for the normal population (P less than 0.01). Patients with persistent pruritus ani and a perianal skin lesion should be biopsied frequently. If Paget's disease is diagnosed, physical examination of the patient and anorectal region must be done carefully and repeatedly.

Aged↗
PubMed1981

Decrease in serum free triiodothyronine after jejuno-ileal bypass for obesity.

Seventeen patients with stable body-weights who had undergone a jejuno-ileal bypass for obesity, and 25 obese patients matching in sex, age and height were studied. Median serum concentration of free triiodothyronine (3.7 pmol/l) in the jejuno-ileal bypass patients was significantly (P less than 0.001) decreased by 18 per cent (95 per cent confidence limits: 11-27 per cent). Correspondingly median serum concentration of total triiodothyronine (1.47 nmol/l) was significantly (P less than 0.0001) decreased by 18 per cent (95 per cent confidence limits:11-27 per cent). The medium serum concentrations of free thyroxine (27.5 pmol/l) and total thyroxine 94 nmol/l) were of equal size in jejuno-ileal bypass patients and obese patients. In the bypass patients the median serum concentration of thyroxine-binding prealbumin (3.9 micromol/l) ws significantly (P less than 0.05) decreased by 13 per cent (95 per cent confidence limits: 7-27 per cent), whereas no significant changes could be demonstrated in thyroxine-binding globulin.

Adult↗
PubMed1989

A controlled trial of 3-month, 4-month, and 6-month regimens of chemotherapy for sputum-smear-negative pulmonary tuberculosis. Results at 5 years. Hong Kong Chest Service/Tuberculosis Research Centre, Madras/British Medical Research Council.

Of 1,710 Chinese patients with radiologically active pulmonary tuberculosis but with sputum negative for acid-fast bacilli on four or more initial microscopic examinations who were studied for 5 yr, 592 (35%) had one or more initial sputum cultures positive for Mycobacterium tuberculosis. These 592 patients were randomly allocated to receive streptomycin, isoniazid, rifampin, and pyrazinamide daily for 4 months or 3 times a week for either 4 or 6 months. The remaining 1,118 patients with all their initial cultures negative were randomly allocated to receive the same four drugs daily for 3 months or 3 times a week for either 3 or 4 months. There were no bacteriologic failures during chemotherapy, and the relapse rates for the 4-month regimens during the 5 yr were 2% in 293 patients with drug-susceptible cultures initially (95% confidence limits, 1 to 5%); 8% in 59 patients with cultures resistant to isoniazid, streptomycin, or both drugs, but susceptible to rifampin initially; and 4% in 325 patients with all their cultures negative initially (95% confidence limits, 1 to 7%). The combined relapse rate for the 3-month regimens was 7% in 709 patients with all their cultures negative initially (95% confidence limits, 5 to 9%). In Hong Kong, 4 months of chemotherapy is now used routinely in the treatment of patients with smear-negative pulmonary tuberculosis, whether their initial sputum cultures are positive or negative.

Adolescent↗