Incentives and disincentives to participation by clinicians in randomised controlled trials

BACKGROUND

Patients and clinicians need reliable, up-to-date information from randomised controlled trials (RCTs) on the costs and benefits of treatments. Recruitment difficulties arise when clinicians do not invite patients to participate in trials.

OBJECTIVES

Primary: to assess the evidence for the effect of disincentives and incentives on the extent to which clinicians invite eligible patients to participate in RCTs of healthcare interventions. Secondary: to assess the evidence in relation to stated willingness to invite participation.

SEARCH STRATEGY

1. The Cochrane Methodology Register and Cochrane Database of Methodology Reviews were searched in May 2006 and Cochrane Central Register of Controlled Trials, National Research Register and ClinicalTrialsGov in April 2005.2. EMBASE, MEDLINE, CINAHL, PsycINFO and AMED were searched in April 2005.3. Reference lists of included studies were checked.

SELECTION CRITERIA

Studies exploring the effect of (dis)incentives on clinicians' views and recruitment-related activity.

DATA COLLECTION AND ANALYSIS

The information about included studies was insufficient for a full assessment of quality. Data on (dis)incentives were extracted and association with recruitment tested.

MAIN RESULTS

No RCTs of interventions were identified. Eleven observational studies were included - two medical records reviews, one matched pair study, one clinician interview study, two studies documenting clinicians' decisions and five postal surveys. Three measures of recruitment were used, invitation to participate, entry into RCT and reported entry to RCT. Five studies explored the effect of patient characteristics. The effect of age and prognosis varied between trials. Six studies considered the association between clinicians' views and recruitment. Clinicians who agreed to participate because they were acquainted with the researchers were less likely to participate than those otherwise motivated (1 study, 2-sided p = 0.04 Fisher's exact test) and (Odds Ratio [OR] 0.4, 95% Confidence Interval [CI] 0.2 to 0.9, 1 study). Clinicians who had recruited were more likely to report some difficulties including "trials involve extra work" (OR 92.94, 95% CI 4.54 - 1902.11; p </= 0.01, 1 study) and "inviting patients to participate is embarrassing" (chi-square 15.55, df = 1, p < 0.0001, 1 study). The effect of the need to discuss clinical uncertainty was unclear but concern that the doctor-patient relationship would be adversely affected by participation was a deterrent (chi-square = 7.25, df = 1, p = 0.007, 1 study).

AUTHORS' CONCLUSIONS

The impact of factors varied across studies. Researchers need to be aware that aspects of the design and conduct of trials can affect clinicians' willingness to invite patients to participate. Further research is needed.

Are personality disorder criteria ethnically biased? A card-sort analysis

On the basis of methodology used in previous research on sex criterion bias, this study examined ethnicity criterion bias of personality disorders (PDs) defined in the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., Rev.) and included examination of sex as well as ethnicity. A card-sort analysis using undergraduate college students as sorters indicated that criteria for all of the PDs were applied disproportionately by ethnicity, resulting in particular ethnic groups receiving diagnoses for specific PDs. Criteria were sorted systematically such that diagnoses of antisocial and paranoid PDs were assigned to African Americans, schizoid PD was assigned to Asian Americans, and schizotypal PD was assigned to Native Americans. All other PDs were assigned to European Americans, whereas none of the criteria were sorted resulting in any PD diagnosis being applied to Latinos. Implications for clinicians, methodological considerations, and recommendations for future research are discussed.

MMPI code types and the fantasy prone personality

We administered the MMPI and the Inventory of Childhood Memories and Imagining (ICMI) to 1,200 college students. Application of diagnostic efficiency statistics for the ability of differing ICMI cutoff scores to identify college students producing a schizophrenia spectrum MMPI code type revealed that scores greater than or equal to 29 on the ICMI had good positive predictive power. Scores less than 29 on the ICMI had very good negative predictive power. ICMI scores were also used to form a group of fantasizers (n = 30) and a control group (n = 30). Fantasizers were much more likely to produce MMPI codes associated with a vulnerability to schizophrenia (70%) than were controls (3.33%). Although most controls(70%) produced non-elevated MMPI scores, 66.67% of the fantasizers produced three or more elevated clinical scales on the MMPI. The modal MMPI profile for the fantasizers was an 8-9 code, indicating that fantasizers appear at heightened risk for eccentric thinking and a Cluster A or B personality organization.

DSM-IV Cluster A personality disorder diagnoses among young adults with a 2-7-8 MMPI profile

The present study evaluated the presence of DSM-IV personality disorders among young adults from a nonclinical setting who produced an MMPI 2-7-8 profile in comparison to a group of MMPI-defined controls. Categorical and dimensional analyses of personality disorders were evaluated. Participants in the 2-7-8 group (n = 20) received significantly more personality disorder diagnoses than did controls (n = 29), and 85% of these individuals received at least one Cluster A (Paranoid, Schizoid, Schizotypal) diagnosis in contrast to only 6.9% of controls (categorical analysis). The 2-7-8 group also received significantly more Cluster A diagnoses than Cluster B or C diagnoses. When dimensional analyses were applied (subclinical diagnoses), 95% of the 2-7-8 group evidenced Cluster A features. Comorbidity patterns were also evaluated; the most frequent comorbid diagnosis for the 2-7-8 group was Avoidant Personality Disorder (n = 8), consistent with Meehl s (1962, 1989, 1990) conceptualization of schizotypy. These results support the use of the MMPI 2-7-8 profile as an indicator of schizophrenia-related pathology within nonclinical samples of young adults.

Implications of the people = male theory for the interpretation of the Draw-A-Person Test

Researchers in the language and social-cognitive fields have suggested that social mores and the use of masculine generic grammatical terms such as he and man have resulted in a people = male bias. This information processing bias causes most people to attribute male gender to a gender-unspecified person. Male gender appears to be prototypic of the person construct or category. These research findings have implications for the interpretation of the Draw-A-Person Test (Machover, 1949).

Using diagnostic efficiency statistics to evaluate the concurrent validity of the perceptual aberration scale

We evaluated the diagnostic efficiency of the 35-item Perceptual Aberration Scale (PAS; Chapman, Chapman, & Raulin, 1978) in selecting schizotypal college students using the presence of a Minnesota Multiphasic Personality Inventory (MMPI; Hathaway & McKinley, 1940) code type associated with the schizophrenia spectrum as indicative of schizotypal status. The PAS was able to reliably rule out the presence of schizotypy among women and men producing very low (< or = 3) PAS scores. Among women, only extremely high PAS (> or = 28) reliably predicted schizotypal MMPI status. Among men, however, no PAS cutting score was reliably associated with the presence of a schizophrenia-related MMPI code type, suggesting that the PAS should not be used to define schizotypy among male college students. The PAS cutoff values that maximized accurate identification of women with schizotypal features and identification of women and men without schizotypal features are considerably more conservative than those that have been traditionally used in research using the PAS as a screening device. The data presented here suggest that using traditional PAS cutoffs (scores > or = 2 SD above the mean and < or = .5 SD below the mean) may result in unacceptably high diagnostic error.

A comparison of the MMPI and the Psychosis Proneness Scales in their classification of normals controls for use in schizophrenia spectrum research

The original MMPI and five of the Chapmans' Psychosis Proneness Scales (PPS; L. J. Chapman, J. P. Chapman, & Miller, 1982) were administered to college students from a nonclinical setting. Two normal control groups (each with 50 female and 50 male subjects) were formed, one on the basis of item endorsement rates on the PPS and the other on the basis of K-corrected MMPI profiles. Because of the multidimensional nature of the MMPI, as compared to the single-sign nature of the various PPS, we expected most subjects classified as normal on the MMPI to also be classified as normal on the PPS. However, we predicted that a relatively high number of subjects classified as normal on the PPS would produce clinically elevated MMPI profiles. These predictions were supported. The convergent validity strategy revealed that 56% of PPS normal controls produced clinically elevated MMPI profiles, many of which were schizophrenia spectrum related. However, 71% of the MMPI normals had scores in the normal range on all five PPS. None of the MMPI controls produced PPS values associated with spectrum membership.

Variants of the MMPI 2-7-8 code type: schizotypal correlates of high point 2, 7, or 8

Our study sought to evaluate whether examination of the high-point scale among subjects with a 2-7-8 parent configuration on the MMPI could assist in the identification of individuals with schizotypal features. Additionally, this study compared male 2-7-8 subjects to female 2-7-8 subjects to determine whether gender mediates endorsement of items associated with these features. Subjects (N = 106) who produced a 2-7-8 parent profile were subdivided according to high point (2, 7, or 8), and the three subgroups were subsequently compared on other self-report measures associated with schizotypal attributes. Subgroup comparisons revealed that the High 8 and High 2 groups produced a pattern of responding consistent with schizotypal characteristics. In contrast, the at-risk status of the High 7 group appears doubtful. Comparisons based on gender generally revealed no differences among female and male 2-7-8 subjects.

Use of the MMPI to assess the construct validity of the revised Social Anhedonia Scale as an index of schizotypy

We evaluated the construct validity of the revised Social Anhedonia Scale (SAS; Mishlove & Chapman, 1985) through an examination of the Minnesota Multiphasic Personality Inventory (MMPI) profiles produced by extreme scorers on the SAS. The MMPI classification strategy employed by Moldin, Gottesman, and Erlenmeyer-Kimling (1987) was used to group profiles with regard to their specificity to schizophrenia spectrum disorders. Of 1,124 college students, 58 females and 60 males had elevated SAS scores. Thirty-five percent of the males and 24.14% of the females produced MMPI profiles within the Moldin et al. classification scheme. Another 27.59% of females and 23.33% of males had profiles that are sometimes associated with schizotypal attributes. Thus, 41.67% of high-SAS males and 48.28% of high-SAS females have MMPI profiles that are unlikely to be associated with a heightened risk for schizophrenia. Because only a subset of socially anhedonic subjects produced schizophrenia spectrum MMPI profiles, it appears that the SAS, in isolation, should not be used to identify individuals at risk for schizophrenia. The revised SAS, like its predecessor, does not appear uniquely related to the schizophrenia spectrum. Unlike Mishlove and Chapman (1985), we did not find a gender difference among subjects.

Concurrent validity for the Rust Inventory of Schizotypal Cognitions

The present study evaluated the concurrent validity of the Rust Inventory of Schizotypal Cognitions (RISC) using the MMPI and the Psychosis Proneness Scales. Multiple regression techniques applied to the RISC and the other measures employed offer support for the validity of this measures as a screening instrument for use in non-clinical samples.

Backward masking as a function of spatial frequency. A comparison of MMPI-identified schizotypics and control subjects

The present investigation relied upon a neurophysiological explanation of visual masking and compared the backward masking susceptibility of hypothetical schizotypal individuals with that of control subjects. Masking functions were assessed within two masking conditions: high spatial frequency (HSF) and low spatial frequency (LSF). Schizotypal subjects (those with a 2-7-8 or an 8-9 MMPI profile type) were compared with "psychiatric" control subjects (those with a spike 9 or a 4-9 profile type) and normal control subjects. Group differences were expected only in the transient-facilitating, LSF masking condition in this study's assessment of the hypothesis that a transient channel abnormality underlies the schizophrenia spectrum backward masking deficit. As predicted, schizotypal subjects displayed greater masking susceptibility in the LSF transient-facilitating condition as compared with the HSF, sustained-facilitating condition that produced no group differences. These results suggest that multichannel neurophysiological models of masking may help to direct research designed to gain an increased understanding of the specific nature of the spectrum masking deficit.

Backward masking spatial frequency effects among hypothetically schizotypal individuals

The present investigation relied upon a neurophysiological explanation of visual masking and compared the backward masking susceptibility of hypothetical schizotypal individuals to that of controls. In order to assess the relative contributions of the visual system's transient and sustained channels to the backward masking deficit characteristic of the schizophrenia spectrum, performance within low spatial frequency (LSF) and high spatial frequency (HSF) masking conditions was compared. Because this design was intended to test the hypothesis that a transient channel abnormality underlies the spectrum masking deficit, only the transient facilitating, LSF masking condition was expected to produce group differences. Although the two masking conditions were equivalent in their stimulus energies, as predicted, the at-risk subjects evidenced an LSF masking deficit, but did not differ from controls in the sustained facilitating, HSF masking condition. These results suggest that multichannel neurophysiological models of masking may help to direct research designed to gain an increased understanding of the specific nature of the spectrum masking deficit.

Visual masking and the schizophrenia spectrum: interfacing clinical and experimental methods

The present article reviews and evaluates 20 studies of susceptibility to visual masking among individuals within the schizophrenia spectrum using a neurophysiological framework provided by a multichannel model of masking. Particular emphasis is placed upon methodological considerations within the context of the current experimental visual masking literature. While there is ample evidence to suggest that individuals within the schizophrenia spectrum frequently exhibit a backward masking deficit, very little can be understood about the specific nature of the deficit. To gain increased understanding of the specific nature of this deficit, researchers need to use some contemporary theory of masking and derive a theoretical design rationale that facilitates a priori predictions in addition to the more typical post hoc theorizing.

Methodological and interpretive constraints resulting from the use of the Per-Mag classification in high-risk information-processing research

It has previously been argued that the current Per-Mag classification criteria may erroneously select some individuals who are not vulnerable to psychosis. We suggest that a second false-positive problem exists with these criteria. Specifically, actively psychotic individuals may be included in presumably prepsychotic samples. Implications of this problem for at-risk information-processing research are discussed.

Critical stimulus duration: schizophrenic trait or state?

Recent evidence challenges the conclusion of Nuechterlein and Dawson (1984) that a critical stimulus duration (CSD) may be a likely candidate for a state marker of schizophrenia. We summarize the results of three different CSD investigations which used differing schizotypic detection criteria and report that the Minnesota Multiphasic Personality Inventory (MMPI) 2-7-8 schizotypic profile identified vulnerable college students with a CSD deficit. On the basis of these data, we suggest that the CSD task should not be considered a state marker of schizophrenia and may qualify as a trait marker of a specific subgroup of schizophrenia.

The use of a backward masking paradigm to assess information-processing deficits among schizotypics. A re-evaluation of Steronko and Woods

In 1978, Steronko and Woods (J. Abnorm. Psychol., 87: 481-490, 1978) failed to find significant differences in early visual information processing between "schizotypic" and "psychiatric control" college students, as identified by the MMPI; yet these authors concluded that schizotypics suffer from information-processing deficits. The present study was designed to extend and clarify these findings by modifying the methods and procedures used by these researchers. A visual backward masking task was employed to study the information processing of individuals whose MMPI-168 profiles indicated schizophrenic tendencies in the absence of an obvious thought disorder. These schizotypic individuals were identified by the MMPI 2-7-8 code type and were compared with three other groups, also identified by their MMPI profiles: an "inflation-free" control group, an "other-inflations" control group, and a group with an 8-9/9-8 MMPI code type. The 8-9/9-8 code type has been associated with psychotic features in adolescents and adults. Two dependent measures were evaluated: critical stimulus duration in a no-mask condition and mean target identification as a function of varying interstimulus intervals. The 2-7-8 group had significantly higher critical stimulus duration values than either the inflation-free group or the 8-9 groups. The 2-7-8 group and the 8-9 groups had fewer correct identifications of target stimuli than either the inflation-free group or the other-inflations group. These results suggest that both the 2-7-8 group and the 8-9 group may be more vulnerable to the effects of the masking stimulus.(ABSTRACT TRUNCATED AT 250 WORDS)

The effect of vitamin A deficiency on the incorporation of D-glucosamine-1- 14 C into homogenates and subcellular fractions of the rat retina

The incorporation of D-glucosamine-1-14C into homogenates, smooth and rough microsomes, and the cell sap of retinas from vitamin-A-deficient and pair-fed vitamin-A-supplemented rats was studied. The results indicate that vitamin A deficiency produced a highly significant increase in the incorporation of D-glucosamine-1-14C into the trichloroacetic-acid-precipitable material from the different fractions studied. The possible significance of these results was discussed.