Psychosocial impact of epidermolysis bullosa on patients: A qualitative study.

BACKGROUND: Epidermolysis bullosa (EB) is an inherited disorder that results in painful skin blisters requiring daily wound care. The psychosocial impact of EB is one of the most significant concerns for patients, but there is minimal research addressing these concerns or ways to improve patient quality of life (QoL). OBJECTIVE: To examine the psychosocial impact of EB on affected patients and ways to improve their QoL. METHODS: Eight EB participants were selected from the 2006 DEBRA Family Conference Day in Toronto. Participants were interviewed by a social worker and a field evaluator. The transcript of each interview was assessed using qualitative content analysis. RESULTS: Four themes were identified: school interaction, daily life, family interactions, and societal interactions. Participants reported being teased and avoided by peers, and they felt their conditions were misunderstood by the general public. School absenteeism resulted in some patients falling behind in school, which may have been misinterpreted as intellectual impairment. Patients acknowledged significant dependence on others and felt they were contributing to caregiver burden. CONCLUSIONS: Our findings highlight the psychosocial impact of EB on patients. As EB awareness and resources to support patients and caregivers have improved since this study was conducted, more studies exploring the current landscape and opportunities to improve quality of life are needed.

Cancer blues? A promising judgment bias task indicates pessimism in nude mice with tumors.

In humans, affective states can bias responses to ambiguous information: a phenomenon termed judgment bias (JB). Judgment biases have great potential for assessing affective states in animals, in both animal welfare and biomedical research. New animal JB tasks require construct validation, but for laboratory mice (Mus musculus), the most common research vertebrate, a valid JB task has proved elusive. Here (Experiment 1), we demonstrate construct validity for a novel mouse JB test: an olfactory Go/Go task in which subjects dig for high- or low-value food rewards. In C57BL/6 and Balb/c mice faced with ambiguous cues, latencies to dig were sensitive to high/low welfare housing: environmentally-enriched animals responded with relative 'optimism' through shorter latencies. Illustrating the versatility of this validated JB task across different fields of research, it further allowed us to test hypotheses about the mood-altering effects of cancer in male and female nude mice (Experiment 2). Males, although not females, treated ambiguous cues as intermediate; and males bearing subcutaneous lung adenocarcinomas also responded more pessimistically to these than did healthy controls. To our knowledge, this is the first evidence of a valid mouse JB task, and the first demonstration of pessimism in tumor-bearing animals. This task still needs to be refined to improve its sensitivity. However, it has great potential for investigating mouse welfare, the links between affective state and disease, depression-like states in animals, and hypotheses regarding the neurobiological mechanisms that underlie affect-mediated biases in judgment.

The use of noninvasive imaging techniques in the diagnosis of melanoma: a prospective diagnostic accuracy study.

BACKGROUND: Early detection of melanoma is crucial to improving the detection of thin curable melanomas. Noninvasive, computer-assisted methods have been developed to use at the bedside to aid in diagnoses but have not been compared directly in a clinical setting. OBJECTIVE: We conducted a prospective diagnostic accuracy study comparing a dermatologist's clinical examination at the bedside, teledermatology, and noninvasive imaging techniques (FotoFinder, MelaFind, and Verisante Aura). METHODS: A total of 184 patients were recruited prospectively from an outpatient dermatology clinic, with lesions imaged, assessed, and excised. Skin specimens were assessed by 2 blinded pathologists, providing the gold standard comparison. RESULTS: Fifty-nine lesions from 56 patients had a histopathologic diagnosis of melanoma, whereas 150 lesions from 128 patients were diagnosed as benign. Sensitivities and specificities were, respectively, MelaFind (82.5%, 52.4%), Verisante Aura (21.4%, 86.2%), and FotoFinder Moleanalyzer Pro (88.1%, 78.8%). The sensitivity and specificity of the teledermoscopist (84.5% and 82.6%, respectively) and local dermatologist (96.6% and 32.2%, respectively) were also compared. LIMITATIONS: There are inherent limitations in using pathology as the gold standard to compare sensitivities and specificities. CONCLUSION: This study demonstrates that the highest sensitivity and specificity of the instruments were established with the FotoFinder Moleanalyzer Pro, which could be a valuable tool to assist with, but not replace, clinical decision making.

Forearm blood flow and local responses to peptide vasodilators: a novel pharmacodynamic measure in the phase I trial of antagonist G, a neuropeptide growth factor antagonist.

PURPOSE: Arg-D-Trp-NmePhe-D-Trp-Leu-Met-NH(2) (Antagonist G), a substance P (SP 6-11) analogue, inhibits mitogenesis stimulated by a broad spectrum of neuropeptides and has demonstrated antitumor activity in vitro and in vivo with IC(50) concentrations of 10-20 microM in small cell lung cancer and other cell lines. Because neuropeptides are part of complex neurohumoral pathways, we have sought to develop novel pharmacodynamic approaches as part of the early clinical development of this potential anticancer drug. EXPERIMENTAL DESIGN: A Phase I trial was performed in two stages. In stage 1, Antagonist G was administered at 3- week intervals using an accelerated dose-escalation strategy until the target maximum plasma concentration (C(max)) of 10 microM was achieved. In stage 2, dose intensity was increased to weekly, and the inhibitory effect of i.v. Antagonist G was assessed by forearm blood flow (FBF) using SP as a vasodilator, as measured by venous plethysmography. RESULTS: In stage 1, dose was escalated from 2 to 300 mg/m(2) in 12 dose levels using only 15 patients. In stage 2, nine patients were entered at three dose levels (300, 350, and 400 mg/m(2)) and a C(max) of 45 microM was achieved. Facial flushing was the only consistent toxicity but was not dose limiting. FBF studies demonstrated that Antagonist G consistently inhibited the vasodilatory effects of SP (mean, 62 +/- 2% inhibition). CONCLUSIONS: Antagonist G can be safely administered up to 400 mg/m(2), achieving C(max)s >20 microM by weekly 6-h i.v. infusion. FBF studies in patients demonstrated that Antagonist G inhibits SP vasodilatory effects in vivo at these doses in the absence of dose-limiting toxicity.

Development of a gradient elution high-performance liquid chromatography assay with ultraviolet detection for the determination in plasma of the anticancer peptide [Arg6, D-Trp7,9, mePhe8]-substance P (6-11) (antagonist G), its major metabolites and a C-terminal pyrene-labelled conjugate.

[Arg6, D-Trp7,9 mePhe8]-substance P (6-11), code-named antagonist G, is a novel peptide currently undergoing early clinical trials as an anticancer drug. A sensitive, high efficiency high-performance liquid chromatography (HPLC) method is described for the determination in human plasma of antagonist G and its three major metabolites, deamidated-G (M1), G-minus Met11 (M2) and G[Met11(O)] (M3). Gradient elution was employed using 40 mM ammonium acetate in 0.15% trifluoroacetic acid as buffer A and acetonitrile as solvent B, with a linear gradient increasing from 30 to 100% B over 15 min, together with a microbore analytical column (microBondapak C18, 30 cm X 2 mm I.D.). Detection was by UV at 280 nm and the column was maintained at 40 degrees C. Retention times varied by <1% throughout the day and were as follows: G, 13.0 min; M1, 12.2 min; M2, 11.2 min; M3, 10.8 min, and 18.1 min for a pyrene conjugate of G (G-P). The limit of detection on column (LOD) was 2.5 ng for antagonist G, M1-3 and G-P and the limit of quantitation (LOQ) was 20 ng/ml for G and 100 ng/ml for M1-3. Sample clean-up by solid-phase extraction using C2-bonded 40 microm silica particles (Bond Elut, 1 ml reservoirs) resulted in elimination of interference from plasma constituents. Within-day and between-day precision and accuracy over a broad range of concentrations (100 ng/ml-100 microg/ml) normally varied by < 10%, although at the highest concentrations of M1 and M2 studied (50 microg/ml), increased variability and reduced recovery were observed. The new assay will aid in the clinical development of antagonist G.

Dose-limiting neurotoxicity in a phase I study of penclomedine (NSC 388720, CRC 88-04), a synthetic alpha-picoline derivative, administered intravenously.

3,5-Dichloro-2,4-dimethoxy-6-(trichloromethyl)pyridine (penclomedine, NSC 338720, CRC 88-04) is an alpha-picoline derivative with anti-tumour activity in preclinical models. Penclomedine administration by 1-h intravenous infusion on 5 consecutive days was repeated 3 weekly in the absence of dose-limiting toxicity (DLT) or disease progression. Five dose levels were investigated (22.5-340 mg m(-2) day[-1]). Eight men and eight women were entered, median age 59 years (range 39-73 years), with good performance status (ECOG 0/1) in 11 patients. A total of 13 out of 16 patients had received previous chemotherapy. Common toxicity criteria grade (CTCg) II vomiting was recorded at all dose levels. Neurotoxicity (cerebellar ataxia and dizziness) was the DLT, CTCg III toxicity occurring in three out of three patients treated at 340 mg m(-2) day(-1). CTCg III dizziness was noted in one out of three patients at 250 mg m(-2) day(-1). Neurotoxicity developed during the 1-h infusion and persisted for a variable period (maximum 5 h) after infusion. Prophylactic antiemetic drugs appeared to reduce associated vomiting but did not prevent ataxia. No antiproliferative toxicities were noted and no anti-tumour responses were documented. Penclomedine pharmacokinetic studies confirmed preclinical evidence of extensive apparent distribution (93 l m[-2]) and rapid clearance (41 l h[-1] m[-2]). Purkinje cell loss has been identified in preclinical models after intraperitoneal administration (O'Reilly et al, 1996a) and further clinical development of penclomedine will focus on oral administration.

Effects of nitric oxide manipulation on the disposition of platinum in an experimental glioma model.

Rodents with striatal C6 glioma were given carboplatin (65 mg kg(-1) in a 10 mg ml(-1) solution, i.v.) after pretreatment with the NO modulating agents 3-morpholinosydnonimine (SIN-1), NG-nitro-L-arginine methyl ester (L-NAME), bradykinin or dexamethasone, to determine whether platinum disposition in the glioma and normal brain was altered. There was no significant change in mean glioma platinum disposition after 3 days of dexamethasone (32+/-9.7 microg/g). Treatment with SIN-1 (45.1+/-14.2 microg/g), L-NAME (42.9+/-4.9 microg/g) and bradykinin (45.7+/-11.3 microg/g) all resulted in increased tumour platinum concentration compared with controls (29+/-5.5 microg/g) but these results were not statistically significant. Dexamethasone significantly (p < 0.05) reduced the platinum concentration in normal brain but the other agents had no effect. Although glioma platinum concentration could be increased by some agents that alter tissue NO levels, the patterns of response were unpredictable and the magnitude (approximately 50%) of the increased platinum disposition is unlikely to be biologically significant.

Processing of [D-Arg1,D-Phe5,D-Trp7,9,Leu11]substance P in xenograft bearing Nu/Nu mice.

[D-Arg1,D-Phe5,D-Trp7,9,Leu11]Substance P is a broad-spectrum neuropeptide growth factor antagonist that has exhibited in vitro activity against a range of human cancer cell lines. The fate of this compound in vivo following i.p. administration at 12 micrograms/g to nu/nu mice bearing the H69 small-cell lung cancer xenograft has been studied. Metabolism was confined to the C-terminus producing [D-Arg1,D-Phe5,D-Trp7,9,Leu11]substance P acid and [D-Arg1,D-Phe5,D-Trp7,9]substance P(1-10). The peptide had a long half-life in plasma (45.9 min) and became widely distributed among the tissues studied with the highest accumulation observed in the liver (AUC 1102 micrograms/g x min) and the lowest in the brain (5 micrograms/g x min). Uptake into the tumor xenograft was poor (AUC 189 micrograms/g x min); however, uptake into the lungs was much greater (AUC 507 micrograms/g x min), offering encouragement that therapeutic concentrations may be targeted to primary lung tumors.

Metabolism of the broad-spectrum neuropeptide growth factor antagonist: [D-Arg1, D-Phe5, D-Trp7,9, Leu11]-substance P.

Broad-spectrum neuropeptide growth factor antagonists, such as [D-Arg1, D-Phe5, D-Trp7,9, Leu11]substance P (antagonist D) and [Arg6, D-Trp7,9, NmePhe8]substance P(6-11) (antagonist G), are currently being investigated as possible anti-tumour agents. These compounds are hoped to be effective against neuropeptide-driven cancers such as small-cell lung cancer. Antagonist D possesses a broader antagonistic spectrum than antagonist G and hence may be of greater therapeutic use. The in vitro metabolism of antagonist D has been characterised and the structures of two major metabolites have been elucidated by amino acid analysis and mass spectrometry. Metabolism was confined to the C-terminus where serine carboxypeptidase action produced [deamidated]-antagonist D (metabolite 1) and [des-Leu11]-antagonist D (metabolite 2) as the major metabolites. Biological characterisation of the metabolites demonstrated that these relatively minor changes in structure resulted in a loss of antagonist activity. These results provide some of the first structure-activity information on the factors that determine which neuropeptides these compounds inhibit and on the relative potency of that inhibition.

Characterization of the deamidase enzyme responsible for the metabolism of the anticancer peptide: H-Arg-D-Trp-NmePhe-D-Trp-Leu-Met-NH2.

H-Arg-D-Trp-NmePhe-D-Trp-Leu-Met-NH2, a broad spectrum neuropeptide growth factor antagonist (antagonist G), is soon to enter a phase I clinical trial for the treatment of small-cell lung cancer (SCLC). The pre-clinical pharmacology of this peptide has revealed that its metabolism proceeds from the C-terminus via deamidation. In this study the class of enzyme responsible for the degradation of antagonist G has been characterized. Tissue distribution studies on the enzyme have shown it to be very widespread with high specific activity being detected in the spleen, kidney, H69 SCLC xenograft and liver (12.64, 9.58, 8.00 and 6.94 nmols G/mg protein/hr, respectively). HPLC gel filtration indicated that the G-deamidase enzyme had an apparent molecular mass of 81 kDa. The sub-cellular distribution of the enzyme using differential centrifugation indicates that it is largely soluble with > 85% of the activity located in the cytosolic fraction. The distribution of activity towards antagonist G closely resembles that of esterase and acid carboxypeptidase activity, two activities, along with deamidase activity, known to be possessed by serine carboxypeptidases. Studies using a range of protease inhibitors showed clear inhibition of metabolism by phenylmethylsulphonylfluoride and benzyloxycarbonylphenylalanine chloromethylketone, indicating that the enzyme is a chymotrypsin-like serine carboxypeptidase. This knowledge of the enzyme will be invaluable in the further development of antagonist G and similar compounds. Moreover, the widespread distribution of this enzyme together with its broad specificity for C-terminal group suggests that it should be given serious consideration when designing C-terminally modified peptide drugs.

Pharmacokinetics, metabolism, tissue and tumour distribution of the neuropeptide growth factor antagonist [Arg6, D-Trp7,9, NmePhe8]- substance P(6-11) in nude mice bearing the H69 small-cell lung cancer xenograft.

BACKGROUND: [Arg6, D-Trp7,9, NmePhe8]-Substance P (6-11) (codenamed antagonist G) represents the first board spectrum antagonist of a number of neuropeptides shown to act as growth factors in small-cell lung cancer (SCLC) and is shortly to enter clinical trials. DESIGN: Pharmacokinetics, metabolism, tissue disposition have been studied in mice (nu/nu) bearing the NCI-H69 human SCLC xenograft after systemic drug adminstration at an active dose (45 mg/kg i.p.). RESULTS: The peptide exhibited relatively long half life (28.9 min; clearance 45.6 ml/min/kg) and distributed widely (volume of distribution 1490 ml/kg). Marked accumulation of antagonist G (and its metabolites) was noted in the liver (AUC5278 micrograms/g x min) and to a lesser extent the spleen (AUC 930 micrograms/g x min) but only low levels appeared to cross the blood brain barrier (AUC in brain, 20 micrograms/g x min) or be taken up into the heart (AUC 101 micrograms/g x min). Tumour uptake was intermediate in value out of the 7 tissues studied (AUC 195 micrograms/g x min). Metabolism was restricted almost exclusively to the C terminal of the peptide producing 4 major products: M1, deamidated antagonist G; M2, Harg-DTrp-NmePhe-DTrp-Leu-OH, both of which retain growth factor antagonist activity; M3, a combination of oxidised antagonist G [Met11(O)] and oxidised deamidated antagoinst G; and M4, a combination of H-Arg-DTrp-NmePhe-DTrp-OH and H-DTrp-NmePhe-DTrp-Leu-OH. Extensive biotransformation to predominately M1 and M2 occurred in most tissues including the tumour where the parent peptide accounted for only 48.5% of the total. CONCLUSION: Levels of antagonist G required to produce a small but significant effect on the growth of SCLC cell lines in vitro are in the region of 4-7 microM. Taking into account metabolites, a peak concentration of 4.1 microgram/g (4.3 microM) was achieved in the H69 xenograft. These studies reveal a favourable preclinical pharmacology profile for antagonist G and offer hope that anticancer activity may be achievable in man.

Processing of the neuropeptide growth factor antagonist [Arg6, D-Trp7.9, NmePhe8]-substance P (6-11) by a small cell lung cancer cell line (H69).

[Arg6, D-Trp7.9, NmePhe8]-substance P (6-11) (antagonist G) is a broad spectrum neuropeptide growth factor antagonist about to enter clinical trials as an anticancer drug. Its fate has been studied after incubation with two densities (5 x 10(4) cells/mL and 1 x 10(6) cells/mL) of the H69 small cell lung cancer cell line for up to 7 days at a concentration of 20 microM, corresponding to the IC50 for growth inhibition. HPLC analyses were conducted on cell pellets and media and in controls consisting of cell free media and water. Over 7 days in media containing cells a 70.4% reduction in parent peptide concentration occurred at the high density and a 44.1% reduction at low density. Despite this, there was a steady elevation in peptide associated with cells reaching a 189% increase by day 7. Oxidation of G at the C-terminal methionine residue occurred in all media studied indicative of a chemical process. The two major active metabolites of antagonist G (deamidated G and G minus Met11) were detected only in media in the presence of cells. These accumulated with time in media and cells together with oxidized products. These results reveal complex cellular pharmacology for antagonist G where H69 cells are increasingly exposed to 4 different peptide products rather than 1.

Metabolism of the anticancer peptide H-Arg-D-Trp-NmePhe-D-Trp-Leu-Met-NH2.

H-Arg-D-Trp-NmePhe-D-Trp-Leu-Met-NH2 (Antagonist G) will be the first broad-spectrum neuropeptide antagonist to enter a phase I clinical trial. Its in vitro and in vivo metabolism has been extensively characterized. The major metabolites were identified and their structures elucidated by mass spectroscopy and amino acid analysis. Metabolism occurred almost exclusively at the C-terminus and was arrested by phenylmethylsulfonylfluoride, a known serine-protease inhibitor. Biological characterization of the metabolites demonstrated that the degradation of Antagonist G produces metabolites that retain neuropeptide antagonist properties.

Stability and in vitro metabolism of the mitogenic neuropeptide antagonists [D-Arg1,D-Phe5, D-Trp7,9, Leu11]-substance P and [Arg6, D-Trp7,9, MePhe8]-substance P (6-11) characterized by high-performance liquid chromatography.

The substance P (SP) analogues [D-Arg1, D-Phe5, D-Trp7,9, Leu11]-SP and [Arg6, D-Trp7,9, MePhe8]-SP (6-11) (antagonists D and G, respectively) are under consideration as new anticancer drugs. In this report, the stability and in vitro metabolism of both antagonists in up to seven different media (water, 1 M acetic acid, human plasma, nude mouse liver and WX 322 human SCLC xenograft homogenized in either 1 M acetic acid or phosphate buffered saline (PBS), pH 7.4) have been characterized by both isocratic and gradient elution reversed-phase HPLC. Antagonist D was stable (never > 13% degradation over 24 h, at 37 degrees C) in water, 1 M acetic acid and plasma but was metabolized by PBS liver homogenates (10%, w/v) sequentially to two stable metabolites with a half life of 0.98 h at a concentration of 500 micrograms ml-1. The major pathway of degradation of antagonist G appeared to be C-terminal methionine oxidation (particularly in plasma) as well as hydrolysis, with even aqueous solutions being significantly affected at low concentrations of peptide (0.1 micrograms ml-1, half life 20.9 h at 37 degrees C). Stable metabolites of antagonist G were also detected in incubations with PBS liver homogenates (half life 1.53 h at 500 micrograms ml-1, 37 degrees C). Overall, the data presented indicate that the modifications made to SP have been relatively successful in preserving chemical and biological stability.

Determination of two neuropeptide growth factor antagonists, [D-Arg1,D-Phe5,D-Trp7,9,Leu11]-substance P and [Arg6,D-Trp7,9,N-MePhe8]- substance P(6-11), by high-performance liquid chromatography with electrochemical detection.

The neuropeptide growth factor antagonists [D-Arg1,D-Phe5,D-Trp7,9,Leu11]-substance P (D) and [Arg6,D-Trp7,9, [corrected] N-MePhe8]-substance P(6-11) (G) are currently undergoing preclinical evaluation as potential anticancer agents and clinical trials are planned for G in the near future. A reversed-phase high-performance liquid chromatographic separation has been developed which is both sensitive (limit of detection 250 pg/263 fmol for G; 500 pg/330 fmol for D) and selective, based on electrochemical detection of the two tryptophan residues present in each peptide. Two ion-pairing agents were included in the isocratic mobile phase to eliminate adsorption of the peptides onto the analytical column. Extensive sample clean-up procedures have been developed for plasma, tissue and tumour based on solid-phase extraction. Precision and accuracy of each assay was 91.3 +/- 16.9% (between-day) for G and 99.3 +/- 16.9% (between-day) for D. The assays were able to detect the intact peptides and a number of their metabolites in plasma, liver and the WX 322 SCLC human xenograft in nude mice for at least 6 hr after administration of therapeutic and pharmacological doses.

Cerebrospinal fluid leakage after sacral bar removal. A case report.

A well recognized tool in the management of sacral fractures is the use of extradural sacral bars. This is a case of a previously unreported complication in the use of sacral bars: leakage of cerebrospinal fluid after their removal from a healed fracture.

The comparative disposition of the pyrolloquinone GR63178A and its 9-hydroxy metabolite GR54374X in sensitive and resistant mouse colon adenocarcinoma.

The novel anticancer compound GR63178A is being evaluated in the clinic, having demonstrated activity against a wide range of experimental tumour systems in animals without significant toxic side-effects being apparent. In this work, we have demonstrated significant antitumour action of this compound against one murine colon cancer model (colon 38 tumour in BDF-1 mice, specific growth delay = 1.2) when given at 10 mg/kg over 21 consecutive days and in contrast shown minimal sensitivity of another similar murine colon adenocarcinoma, MAC 26, in NMRI mice with the same dose regime. We investigated the disposition of both the parent drug and the 9-OH metabolite (GR54374X) in plasma, tissues and tumours, using solid phase extraction followed by reversed-phase high performance liquid chromatography. Although plasma clearance profiles of GR63178A were similar, significant differences were seen in the disposition of the drug to major organs in two mouse strains. Noteably, the liver and kidneys of the sensitive model had higher levels of parent drug and 9-OH metabolite at both 30 min and 4 h post-injection. However, this was not apparent in the tumours themselves, and the levels of 9-OH metabolite were lower in the plasma and higher in the urine of the sensitive mice, indicating possible rapid renal clearance of this compound. Neither GR63178A nor GR54374X proved cytotoxic in in vitro experiments. The data presented here have revealed considerable variation in drug handling by these two mouse strains, but this did not produce different levels of either parent drug or GR54374X in the tumours, which are the presumed targets, suggesting that differences in disposition are probably not responsible for the different sensitivities of the two tumours. Other possible explanations include the production of a hitherto undetected ultimate cytotoxic metabolite in the sensitive, but not in the resistant, mouse/tumour combination, or differences in inherent tumour sensitivity, or in host-mediated effects. These possibilities are discussed.

The long term stability of mechlorethamine hydrochloride (nitrogen mustard) ointment measured by HPLC.

HPLC has been applied to determine the stability of mechlorethamine hydrochloride (nitrogen mustard) formulated as an ointment in white soft paraffin (10 mg drug, 50 g paraffin and 1 mL acetone). A new solubilization technique is described for extraction of the drug from the ointment for HPLC analysis which has an extraction efficiency of 76.1% with a coefficient of variation of 10.4%. Stored at 4 degrees C the drug content of the ointment remained stable for at least 84 days and stored at 37 degrees C the drug content remained stable for at least 40 days. In comparison aqueous solutions of mechlorethamine at the same concentration, fully degraded after 4 days. These data will aid hospital pharmacists decide on a standardized protocol for the controlled usage of the ointment which is both safe and cost-effective.

Effect of nonmedical factors on family physicians' decisions about referral for consultation.

OBJECTIVES: To identify nonmedical factors perceived by family physicians (FPs) and consultants as important influences on decisions about referral for consultation, to determine the relative frequency with which such factors are cited and to identify those factors ranked as most important by the FPs and consultants. DESIGN: Survey with semistructured interview between July 1989 and April 1990. PARTICIPANTS: A total of 41 FPs and 20 consultants who were practising or had practised previously in Nova Scotia. INTERVENTIONS: The questionnaire comprised 10 questions: 4 were nondirective "probes" designed to elicit responses without suggesting possible answers, 2 asked the participants to rank such responses in order of importance, and 4 were "prompts" that asked for comments about a list of factors based on a review of the literature. RESULTS: A total of 4845 discrete items were mentioned as being capable of influencing FPs' decisions about referral for consultation. Aggregation of related items resulted in a list of 35 nonmedical factors, of which 11 were identified by at least half the respondents and 14 by less than half but more than 10. These 25 factors fell into three categories: patient and family factors (e.g., patient's wishes), FP and consultant factors (e.g., FP's capabilities), and other influences (e.g., style of practice). On the basis of both frequency of identification and priority scores "patient's wishes" emerged as the most important factor. Two medical factors that were consistently cited--type of problem and age of patient--were thought to interact with the other factors. CONCLUSION: Certain nonmedical considerations may substantially affect physicians' referral practices.