Hyperprolactinaemia, a common side effect of some antipsychotic drugs, is also present in drug-naïve psychotic patients and subjects at risk for psychosis. Recent studies in non-psychiatric populations suggest that increased prolactin may have negative effects on cognition. The aim of our study was to explore whether high plasma prolactin levels are associated with poorer cognitive functioning in subjects with early psychoses. We studied 107 participants: 29 healthy subjects and 78 subjects with an early psychosis (55 psychotic disorders with <3 years of illness, 23 high-risk subjects). Cognitive assessment was performed with the MATRICS Cognitive Consensus Cognitive Battery, and prolactin levels were determined as well as total cortisol levels in plasma. Psychopathological status was assessed and the use of psychopharmacological treatments (antipsychotics, antidepressants, benzodiazepines) recorded. Prolactin levels were negatively associated with cognitive performance in processing speed, in patients with a psychotic disorder and high-risk subjects. In the latter group, increased prolactin levels were also associated with impaired reasoning and problem solving and poorer general cognition. In a multiple linear regression analysis conducted in both high-risk and psychotic patients, controlling for potential confounders, prolactin and benzodiazepines were independently related to poorer cognitive performance in the speed of processing domain. A mediation analysis showed that both prolactin and benzodiazepine treatment act as mediators of the relationship between risperidone/paliperidone treatment and speed of processing. These results suggest that increased prolactin levels are associated with impaired processing speed in early psychosis. If these results are confirmed in future studies, strategies targeting reduction of prolactin levels may improve cognition in this population.
Citation: Montalvo I, Gutiérrez-Zotes A, Creus M, Monseny R, Ortega L, Franch J, et al. (2014) Increased Prolactin Levels Are Associated with Impaired Processing Speed in Subjects with Early Psychosis. PLoS ONE 9(2): e89428. doi:10.1371/journal.pone.0089428
Editor: Joseph Najbauer, University of Pécs Medical School, Hungary
Received: October 16, 2013; Accepted: January 20, 2014; Published: February 24, 2014
Copyright: © 2014 Montalvo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants from Fundació La Marató de TV3 (092230/092231) and from Instituto de Salud Carlos III (FIS, PI10/01607). JL has received an Intensification of Research Activity grant (Programme I3SNS-INT11/323) from the Instituto de Salud Carlos III (Spain) during 2012 and 2013. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: SML has received financial support from Pfizer (formerly Wyeth) in relation to imaging studies of people with schizophrenia and bipolar disorder. SML has done consultancy work for Roche Pharmaceuticals in connection with a possible new treatment for schizophrenia. SML has also received honoraria for lectures, chairing meetings, and consultancy work from Janssen in connection with brain imaging and therapeutic initiatives for psychosis. JL has received honoraria for lectures from Janssen. IM, AGZ, MC, LO, RM, JF, EV and RMR have no biomedical financial interests or potential conflicts of interest. This does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.
Hyperprolactinaemia is a common condition in subjects with a psychotic disorder. As dopamine is the main prolactin inhibiting factor, hyperprolactinaemia is a common consequence of D2 receptor blockade in the tuberoinfundibular dopaminergic pathway ,  by antipsychotic drugs. However, increased prolactin levels have also been reported in drug-naïve patients with a first psychotic episode or at risk mental states –. The mechanisms that mediate the increase of prolactin levels in psychotic subjects not receiving antipsychotic drugs are poorly understood. Moreover, prolactin levels may be increased by stress , which may in turn contribute to the increased prolactin levels in drug-naïve psychotic populations.
The most studied consequences of hyperprolactinaemia in psychotic subjects are amenorrhoea, galactorrhoea, sexual impairment and infertility , . A recent study conducted in non-psychiatric population suggests that increased prolactin may have negative effects on cognition . This prospective study examined the cognitive changes during late pregnancy and the early postpartum period, and their possible association with fluctuating hormone levels (estradiol, progesterone, testosterone, prolactin and cortisol). A total of 55 pregnant women and 21 controls were studied, with a neuropsychological assessment during the third trimester of pregnancy and retest during the early postpartum period. They concluded that very high and very low levels of cortisol were associated with poorer performance in certain cognitive domains, but the most novel finding was that they found a negative linear association between prolactin levels and executive function scores, suggesting that higher levels of prolactin are detrimental to executive function abilities. Animal studies also support a role for prolactin in the modulation of non-spatial cognitive tasks . In this recent study, the induction of hyperprolactinaemia in male rats receiving pituitary grafts was associated with impaired object recognition. Other studies have reported an association between low gonadal steroid levels and poorer cognitive abilities , . To our knowledge, there are no studies addressing whether high prolactin levels often found in psychotic patients can contribute to the cognitive impairment of patients with a psychotic disorder.
Subjects with schizophrenia show mild to moderate cognitive impairment, and perform an average of 1.5 to 2 standard deviations below population norms , . These cognitive alterations appear before the onset of the first psychotic episode ,  and are an important determinant of functional outcome . Early intervention in psychosis is a novel approach to mental health care that includes treatment of both psychotic disorders at first years after the onset (defined as a critical period in the first 3 to 5 years after the onset) as well as subjects with prodromal symptoms who are at risk for psychosis (high-risk, HR). Early intervention services have been developed to reduce the duration of untreated psychosis, a variable that has been associated with a poorer prognosis of the illness and poorer cognitive performance .
Some studies have shown that atypical antipsychotic drugs have a better cognitive profile than typicals –, but this remains controversial. Antipsychotic drugs differ in their affinity at muscarinic receptors, with detrimental effects on cognitive abilities in those antipsychotics with higher anticholinergic activity . However, whether different antipsychotic medications exert any benefits on cognitive performance remains questionable , . Based on the degree of blockade of D2 receptors at this pathway, the risk of hyperprolactinaemia differs among different antipsychotics, being greater for typical antipsychotics and some atypicals including risperidone and paliperidone , . Also, concomitant treatment with anticholinergics or benzodiazepines can also have deleterious effects on cognition –.
The main aim of our study was to explore whether prolactin levels are associated with poorer cognitive functioning in subjects with early psychoses, including both first episode of psychosis and high-risk subjects. We hypothesized that increased prolactin is associated with poorer cognitive performance in subjects with early psychoses. We also aimed to determine whether prolactin may mediate the relationship between antipsychotic drugs and cognition while adjusting for other potential contributors (e.g. adjunctive treatment with benzodiazepines or anticholinergic drugs).
Materials and Methods
All procedures are in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Committee for Ethical Clinical Investigation of the Hospital Sant Joan de Reus. The capacity of the patients to provide informed consent was evaluated and confirmed by a psychiatrist. After complete description of the study to the subjects, written informed consent was obtained from all participants (or their guardians if the patients had a compromised capacity). All potential participants who declined to participate or otherwise did not participate were not disadvantaged in any way by not participating in the study.
We studied 78 outpatients with an early psychosis, aged between 18 and 35 years old, attending the Early Psychosis Program from Reus (HPU Institut Pere Mata, Spain). We included a control group of 29 healthy subjects (HS) that were recruited among patients’ friends and non-genetic relatives, and screened to rule out past or current history of psychiatric disorder.
Early psychosis patients were divided into two different clinical populations: 1) High risk for psychosis (HR, subjects with prodromal psychotic symptoms fulfilling set criteria for At Risk Mental State , N = 23); 2) Psychotic Disorder with less than 3 years from the onset of the illness (PD, N = 55). DSM-IV diagnoses for the PD group were: schizophreniform disorder (N = 11), schizophrenia (N = 10), schizoaffective disorder (N = 3), psychotic disorder not otherwise specified (N = 31). Exclusion criteria were: pregnancy, mental retardation, severe head injury or neurological disease, active glucocorticoid treatment, language difficulties, visual impairment and alcohol, cocaine or heroin dependence.
Characteristics of Patients
All subjects were assessed with the Schedules for Clinical Assessment in Neuropsychiatry . OPCRIT checklist v.4.0. (available at http://sgdp.iop.kcl.ac.uk/opcrit/) was used to generate DSM-IV diagnoses for psychotic disorders. HR subjects were also assessed with the Comprehensive Assessment of At Risk Mental States, to ensure that subjects met criteria for any of the three HR groups defined by At Risk Mental State criteria .
Socio-demographic and clinical variables related to psychosis (age at onset, antipsychotic treatment, substance use) were assessed by semistructured interview. Tobacco, cannabis and alcohol consumption were registered in cigarettes/day, joints/day and standard units/day respectively. Positive and Negative Symptom Scale  was administered to explore positive, negative and overall psychotic symptoms. Calgary Depression Scale  was administered to explore depressive symptoms. These scales were administered the same day of the cognitive assessment.
Psychopharmacological treatment at neuropsychological assessment was registered. Of all 78 patients, 27 (34.6%) were not taking antipsychotics, 42 (53.8%) were on antipsychotic monotherapy (risperidone [n = 20], paliperidone [n = 5], olanzapine [n = 13], quetiapine [n = 1], aripiprazole [n = 8]) and 9 (11.5%) were on polytherapy.
Each antipsychotic dose was transformed into chlorpromazine equivalents in mg/day . We recoded chlorpromazine equivalent doses into three different variables taking into account the mechanism of action of each antipsychotic and its effects on prolactin and anticholinergic activity: 1) risperidone/paliperidone (prolactin elevating without anticholinergic activity), 2) olanzapine/quetiapine/clozapine (prolactin sparing with anticholinergic activity) and 3) aripiprazole (prolactin sparing without anticholinergic activity). Benzodiazepine treatment was registered in diazepam equivalent doses. Biperiden dose was registered in mg/day. Antidepressant treatment was registered as fluoxetine equivalents in mg/day.
The MATRICS Consensus Cognitive Battery (MCCB) was administered to explore neuropsychological functioning . This cognitive battery has demonstrated practicality of administration, high test-retest reliability, small practice effects, small ceiling effects and relationship to functional outcome. The MCCB contains 10 tests to measure cognitive performance in 7 cognitive domains: speed processing, attention/vigilance, working memory, verbal learning, visual learning, reasoning and problem solving, and social cognition (Table S1). A composite Score is obtained, which combines the individual scores of the 10 tests and scores them on a normative scale to derive a T-score, where the mean is 50 and a standard deviation is 10 for the composite. Normative data for the MCCB has been obtained in Spain , suggesting that significant age, gender, and education effects are comparable to those effects described for the original standardized English version in the U.S. All neuropsychological assessments were performed in the morning, with starting times between 9 h and 12 h. Cognitive testing in PD was assessed when they were clinically stable.
A fasting blood sample was obtained in the morning between 8∶30 h and 9∶30 h in resting conditions, to determine unstimulated plasma prolactin and total cortisol levels. Participants were told to avoid stressful activities (sports, physical exercise) or breast stimulation in the 12 hours prior to blood sampling. Prolactin and cortisol concentrations were measured using the Maglumi 2000 Analyzer chemiluminescence immunoassay system (SNIBE Co, Ltd, Guandong, China). The sensitivity of the prolactin assay was 1.77 µg/L.
The SPSS version 19.0 software (SPSS Inc., Chicago, Illinois, USA) was used to carry out the statistical analyses. As prolactin levels, PANSS and CDS scores were skewed, we log transformed (ln) these variables to reduce skewness. Chi-square tests and one-way ANOVA were used to compare categorical and continuous data between groups. Post-hoc ANOVA analyses were adjusted with a Bonferroni correction. Spearman correlation was used to explore the association between continuous or ordinal variables. Significance was set at p<0.05 (two-tailed). For all statistical analyses we used MCCB T-scores corrected for age, gender and education level.
We first divided the sample by diagnostic group (HS, PD and HR). A stratified analysis by diagnostic group was conducted to explore the association between prolactin levels and cognitive measures in each subgroup of participants.
A multiple linear regression analysis was performed in all patients (including both PD and HR subjects) to explore the relationship between plasma prolactin levels (main independent variable) and MCCB cognitive domains (dependent variable) while controlling for covariables such as other psychopharmacological treatments, psychopathological status, smoking and other substance use (cannabis and alcohol) and cortisol.
Furthermore, those MCCB domains that showed a significant association with prolactin in the multivariate analyses were also tested with a mediation analysis to explore whether the effect of antipsychotic treatment on cognition could be mediated by prolactin levels and benzodiazepine treatment. We conducted a mediation analysis according to Baron and Kenny  and used bootstrapping to test the indirect effect of mediation . A description of mediation analysis is shown in Box S1.
We used a SPSS macro  that allows the inclusion of multiple mediators and covariates. In this mediation analysis, we decided to include risperidone/paliperidone dose as the main independent variable because its elevating effect on prolactin levels. MCCB cognitive T-score (e.g. Speed of Processing) was used as the dependent variable. Two potential mediators were considered: prolactin and benzodiazepine treatment. We included as covariates other antipsychotic drugs, biperiden and antidepressant treatments. Significance of the indirect effects in this model was tested by bootstrapping.
R and ggplot2 package (http://www.r-project.org/) were used to draw scatterplot figures exploring the association between prolactin and MCCB cognitive domains.
Clinical characteristics of the sample by diagnostic group are described in Table 1. As expected, we found that HS performed significantly better than PD in all cognitive domains, and better than HR in most domains (Table 1). There were no significant differences in psychopathological scales between PD and HR. PD subjects were more frequently treated with antipsychotics, when compared to HR. However, the latter group was more frequently treated with antidepressants.
In the stratified analysis by diagnostic group, prolactin levels were significantly associated with processing speed in both PD and HR subjects (Table 2). In HR subjects, prolactin levels were also related with poorer cognitive performance in reasoning and problem solving, and global cognition. A scatter plot by diagnostic group (PD vs HR) is shown in Figure 1. We also performed another analysis including both HR and PD together (Table S2). In this analysis, prolactin was positively associated with risperidone/paliperidone and benzodiazepine doses and negatively associated with speed of processing.
In the multivariate analysis conducted in both HR and PD groups, the significant negative association between prolactin and speed of processing was maintained after adjusting for psychopharmacological treatments, smoking and substance use (cannabis and alcohol consumption), severity of psychotic symptoms and cortisol levels (Table 3). As shown in this table, both prolactin and risperidone/paliperidone treatment were related to a poorer processing speed (Model 2). However, when other drugs were included in the equation (Model 3), impaired processing speed was associated with benzodiazepine treatment but not antipsychotic doses. Other MCCB cognitive domains were not associated with prolactin levels in the multiple linear regression analyses (Table S3).
In the mediation analysis, we tested two potential mediators (prolactin and benzodiazepine treatment) in the relationship between risperidone/paliperidone doses and processing speed (Figure 2). In the unadjusted model (a), risperidone/paliperidone treatment was negatively associated with speed of processing. This effect was fully mediated by both prolactin and benzodiazepine treatment (b), as the relationship between antipsychotic treatment and speed of processing lost its significance when these two mediators were included in the equation. In this mediated analysis, other antipsychotics, antidepressants and biperiden were included as covariates, thus the results are adjusted for these psychopharmacological drugs. As a multiple mediation model is analogous to conducting a regression analysis with several predictors testing the total indirect effect of the independent variable (risperidone/paliperidone treatment) on the dependent variable (speed of processing), both prolactin and benzodiazepines are independently associated with speed of processing. The indirect effects of both variables account for all the observed relationship between risperidone/paliperidone treatment and speed of processing. Bootstrap results for indirect effects were significant for both prolactin (95% CI: −0.165 to –0.001) and benzodiazepine treatment (95% CI: −0.167 to –0.002).
Log transformed (ln) levels of prolactin were used in the mediation analysis. The mediated effect (b) was adjusted for the following covariates: olanzapine/clozapine/quetiapine dose (β = 0.002, SE = 0.008, P = 0.848), aripiprazole dose (β = −0.016, SE = 0.011, P = 0.151), biperiden dose (β = −0.567, SE = 1.486, P = 0.704) and antidepressant dose (β = 0.048, SE = 0.101, P = 0.638). Abbreviations: β = unstandardized regression coefficient; SE = standard error.
Our study suggests that increased prolactin levels are associated with impaired processing speed independent of antipsychotic drugs in subjects with early psychosis. In HR subjects only, increased prolactin was also associated with impaired reasoning and problem solving and poorer general cognition. The results of the mediation analysis also showed that the effect of risperidone/paliperidone treatment on speed of processing is mediated by both prolactin levels and benzodiazepine treatment. To our knowledge, this is the first study to highlight prolactin as an important contributor to cognitive impairment in subjects with psychosis.
Regarding speed of processing, our results are consistent with previous studies reporting that processing speed is the first cognitive domain to be affected in psychotic disorders at early stages  and in at risk mental states subjects . This cognitive domain, that may be considered a core feature of schizophrenia neurocognitive impairment, is thought to mediate the relationship between cognitive symptoms and functional outcome in schizophrenia . In contrast to other cognitive domains, speed of processing is considered a “system based” domain, reflecting a process of integration and coordination between distributed brain networks . This is in line with the disconnection hypothesis of schizophrenia, which asserts that impaired communication within the brains of schizophrenia patients occurs when there is focal disruption that adversely affects the entire network. Speed of processing deficits may point to aberrant functional connectivity within and between whole brain neural systems, rather than indexing impairment in discrete neural networks .
Although antipsychotic drugs are a common cause of hyperprolactinaemia, other studies in drug-naïve patients have also shown increased prolactin levels in PD and HR subjects –. Interestingly, in our study the relationship between prolactin and processing speed was also found in HR subjects, most of whom were not receiving antipsychotic treatment. Moreover, in the multivariate analyses, the association between prolactin levels and impaired processing speed remained significant after adjusting for psychopharmacological treatments. Prolactin is a hormone that may be raised in stressful situations. We accounted for this by controlling for cortisol levels in the multivariate analysis, and the effect of prolactin on cognition was independent of hypothalamic-pituitary-adrenal axis activity.
Antipsychotic-induced hyperprolactinaemia, which is caused by tuberoinfundibular blockade of D2 receptors, may be reflecting the blockade of D2 receptors in other dopaminergic pathways including the striatum, which causes extrapyramidal symptoms, or the mesocortical pathway, that may be related to worsening in cognitive and negative symptoms. The mediated analysis suggests that the negative effect of risperidone/paliperidone on processing speed is mediated by both prolactin and benzodiazepines. This could be explained by the prolactin-elevating profile of these antipsychotics as well as the induction of extrapyramidal symptoms (e.g. akathysia) that are often treated with benzodiazepines.
Our study suggests that prolactin may be considered as a biomarker that is associated with impaired processing speed in subjects with early psychoses. This finding may have important clinical implications. Further studies are needed to explore whether a reduction in prolactin levels by optimizing psychopharmacological treatment leads to an improvement in processing speed.
The main limitation of our study is the cross-sectional design that does not allow us to infer causality. Further prospective studies may overcome this limitation by repeatedly assessing prolactin levels and cognitive performance over time to explore whether persistent hyperprolactinaemia is a risk factor for cognitive decline in subjects with psychoses. We did not control for other hormones that may affect cognition such as sex steroids. Hyperprolactinaemia inhibits the hypothalamic-pituitary-gonadal axis , and hypogonadism secondary to hyperprolactinaemia may contribute to the negative effects of prolactin on cognition. However, in order to control for these variables, larger samples are required because a sex-stratified analysis is necessary (controlling for estradiol in women and testosterone in men).
We designed a pragmatic study with consecutive sampling in an Early Intervention Service. For this reason, the sample of our study included both HR subjects and psychotic disorders at early stages, and patients were treated with different antipsychotics that were chosen by clinicians based on the clinical routine practice. HR subjects and PD patients exhibited similar PANSS scores. As our Early Intervention Service is an outpatient service, some patients who have been previously admitted to the referral Acute Psychiatric Unit are stabilized before attending our service. Moreover, patients need to be informed of the research project and must sign the informed consent; thus in most cases, PD patients are clinically stable at recruitment. As most PD patients have been treated before entering the study, the PANSS scores reflect the psychopathological state at the neuropsychological assessment (not at the acute phase of the illness). These characteristics could explain why both groups (HR and PD) exhibited similar PANSS scores. Although HR group shares some characteristics with PD group at early stages such as cognitive impairment, only 30% develop psychosis at one year . From our cross-sectional design we do not know whether prolactin could be a biomarker related to the risk of transition to psychosis. Future prospective studies are needed to clarify this question. Finally, the sample of our study was composed by outpatients, and none of whom were receiving first-generation antipsychotics. Thus, our results may not be generalizable to other populations including inpatients, psychotic patients with a longer duration of illness or patients taking typical antipsychotics. In fact, a recent study conducted in patients with chronic schizophrenia showed that switching to aripiprazole led to a decrease in prolactin levels but was not associated with cognitive improvement . Our results should be replicated in other samples to draw definite conclusions.
However, several strengths of our study need to be highlighted: 1) our study is the first to describe the potential role of prolactin on cognition in subjects with psychoses, 2) we have used a standardized neurocognitive battery (MCCB) to assess cognition, 3) the sample included subjects with a short duration of the illness, and 4) we controlled for potential confounders such as concomitant antipsychotic medication (chlorpromazine equivalents) by differentiating antipsychotics depending upon their mechanism of action, as well as controlling for benzodiazepines that have been suggested as moderating factors of processing speed , and 5) we also controlled for smoking status, which has been describe to modify prolactin levels  and to interfere with cognition in early psychosis patients .
In summary, our study suggests that increased prolactin levels are associated with impaired processing speed in early psychosis and that also mediate the negative effects of prolactin elevating antipsychotics on processing speed. If these results are replicated in further studies, hyperprolactinaemia may be considered as a potential contributor to cognitive deterioration in psychotic subjects and strategies targeting reduction of prolactin levels may improve cognition in this population.
MATRICS Consensus Cognitive Battery tests and cognitive domains.
Correlations between prolactin levels, psychopharmacological treatment, psychopathological status and MCCB cognitive domains in subjects with early psychosis.
Multiple regression analyses exploring the relationship between prolactin levels in plasma and MCCB Cognitive domains in subjects with early psychosis.
Explanation of the mediation analysis.
Conceived and designed the experiments: JL AGZ RMR SML. Performed the experiments: IM MC JF LO RM. Analyzed the data: JL IM. Contributed reagents/materials/analysis tools: RMR EV. Wrote the paper: IM RMR EV JL.
- 1. Fitzgerald P, Dinan TG (2008) Prolactin and dopamine: What is the connection? A review article. J Psychopharmacol 22: 12–19. doi: 10.1177/0269216307087148
- 2. Inder WJ, Castle D (2011) Antipsychotic-induced hyperprolactinaemia. Aust N Z J Psychiatry 45: 830–837. doi: 10.3109/00048674.2011.589044
- 3. Aston J, Rechsteiner E, Bull N, Borgwardt S, Gschwandtner U, et al. (2010) Hyperprolactinaemia in early psychosis-not only due to antipsychotics. Prog Neuropsychopharmacol Biol Psychiatry 34: 1342–1344. doi: 10.1016/j.pnpbp.2010.02.019
- 4. Garcia-Rizo C, Fernandez-Egea E, Oliveira C, Justicia A, Parellada E, et al. (2012) Prolactin concentrations in newly diagnosed, antipsychotic-naive patients with nonaffective psychosis. Schizophr Res 134: 16–19. doi: 10.1016/j.schres.2011.07.025
- 5. Riecher-Rossler A, Rybakowski JK, Pflueger MO, Beyrau R, Kahn RS, et al. (2013) Hyperprolactinemia in antipsychotic-naive patients with first-episode psychosis. Psychol Med : 1–12.
- 6. Lennartsson AK, Jonsdottir IH (2011) Prolactin in response to acute psychosocial stress in healthy men and women. Psychoneuroendocrinology 36: 1530–1539. doi: 10.1016/j.psyneuen.2011.04.007
- 7. Halbreich U, Kahn LS (2003) Hyperprolactinemia and schizophrenia: Mechanisms and clinical aspects. J Psychiatr Pract 9: 344–353. doi: 10.1097/00131746-200309000-00003
- 8. Johnsen E, Kroken R, Loberg EM, Kjelby E, Jorgensen HA (2011) Sexual dysfunction and hyperprolactinemia in male psychotic inpatients: A cross-sectional study. Adv Urol 2011: 686924. doi: 10.1155/2011/686924
- 9. Henry JF, Sherwin BB (2012) Hormones and cognitive functioning during late pregnancy and postpartum: A longitudinal study. Behav Neurosci 126: 73–85. doi: 10.1037/a0025540
- 10. Torner L, Tinajero E, Lajud N, Quintanar-Stephano A, Olvera-Cortes E (2013) Hyperprolactinemia impairs object recognition without altering spatial learning in male rats. Behav Brain Res 252: 32–39. doi: 10.1016/j.bbr.2013.05.031
- 11. Sherwin BB (2012) Estrogen and cognitive functioning in women: Lessons we have learned. Behav Neurosci 126: 123–127. doi: 10.1037/a0025539
- 12. Wolf OT, Kirschbaum C (2002) Endogenous estradiol and testosterone levels are associated with cognitive performance in older women and men. Horm Behav 41: 259–266. doi: 10.1006/hbeh.2002.1770
- 13. Bilder RM, Goldman RS, Robinson D, Reiter G, Bell L, et al. (2000) Neuropsychology of first-episode schizophrenia: Initial characterization and clinical correlates. Am J Psychiatry 157: 549–559. doi: 10.1176/appi.ajp.157.4.549
- 14. Heinrichs RW, Zakzanis KK (1998) Neurocognitive deficit in schizophrenia: A quantitative review of the evidence. Neuropsychology 12: 426–445. doi: 10.1037//0894-4188.8.131.526
- 15. Davidson M, Reichenberg A, Rabinowitz J, Weiser M, Kaplan Z, et al. (1999) Behavioral and intellectual markers for schizophrenia in apparently healthy male adolescents. Am J Psychiatry 156: 1328–1335.
- 16. Reichenberg A, Weiser M, Rapp MA, Rabinowitz J, Caspi A, et al. (2005) Elaboration on premorbid intellectual performance in schizophrenia: Premorbid intellectual decline and risk for schizophrenia. Arch Gen Psychiatry 62: 1297–1304. doi: 10.1001/archpsyc.62.12.1297
- 17. Green MF (1996) What are the functional consequences of neurocognitive deficits in schizophrenia? Am J Psychiatry 153: 321–330.
- 18. Cuesta MJ, Garcia de Jalon E, Campos MS, Ibanez B, Sanchez-Torres AM, et al. (2012) Duration of untreated negative and positive symptoms of psychosis and cognitive impairment in first episode psychosis. Schizophr Res 141: 222–227. doi: 10.1016/j.schres.2012.08.019
- 19. Bilder RM, Goldman RS, Volavka J, Czobor P, Hoptman M, et al. (2002) Neurocognitive effects of clozapine, olanzapine, risperidone, and haloperidol in patients with chronic schizophrenia or schizoaffective disorder. Am J Psychiatry 159: 1018–1028. doi: 10.1176/appi.ajp.159.6.1018
- 20. Purdon SE, Malla A, Labelle A, Lit W (2001) Neuropsychological change in patients with schizophrenia after treatment with quetiapine or haloperidol. J Psychiatry Neurosci 26: 137–149. doi: 10.1016/s0920-9964(00)90808-9
- 21. Harvey PD (2003) Ziprasidone and cognition: The evolving story. J Clin Psychiatry 64 Suppl 1933–39.
- 22. Keefe RS, Seidman LJ, Christensen BK, Hamer RM, Sharma T, et al. (2004) Comparative effect of atypical and conventional antipsychotic drugs on neurocognition in first-episode psychosis: A randomized, double-blind trial of olanzapine versus low doses of haloperidol. Am J Psychiatry 161: 985–995. doi: 10.1176/appi.ajp.161.6.985
- 23. Chew ML, Mulsant BH, Pollock BG, Lehman ME, Greenspan A, et al. (2006) A model of anticholinergic activity of atypical antipsychotic medications. Schizophr Res 88: 63–72. doi: 10.1016/j.schres.2006.07.011
- 24. Mishara AL, Goldberg TE (2004) A meta-analysis and critical review of the effects of conventional neuroleptic treatment on cognition in schizophrenia: Opening a closed book. Biol Psychiatry 55: 1013–1022. doi: 10.1016/j.biopsych.2004.01.027
- 25. Woodward ND, Purdon SE, Meltzer HY, Zald DH (2007) A meta-analysis of cognitive change with haloperidol in clinical trials of atypical antipsychotics: Dose effects and comparison to practice effects. Schizophr Res 89: 211–224. doi: 10.1016/j.schres.2006.08.021
- 26. Maguire GA (2002) Prolactin elevation with antipsychotic medications: Mechanisms of action and clinical consequences. J Clin Psychiatry 63 Suppl 456–62.
- 27. Cookson J, Hodgson R, Wildgust HJ (2012) Prolactin, hyperprolactinaemia and antipsychotic treatment: A review and lessons for treatment of early psychosis. J Psychopharmacol 26: 42–51. doi: 10.1177/0269881112442016
- 28. Harvey PD, Keefe RS (2001) Studies of cognitive change in patients with schizophrenia following novel antipsychotic treatment. Am J Psychiatry 158: 176–184. doi: 10.1176/appi.ajp.158.2.176
- 29. Keefe RS, Silva SG, Perkins DO, Lieberman JA (1999) The effects of atypical antipsychotic drugs on neurocognitive impairment in schizophrenia: A review and meta-analysis. Schizophr Bull 25: 201–222. doi: 10.1093/oxfordjournals.schbul.a033374
- 30. Carpenter WT, Gold JM (2002) Another view of therapy for cognition in schizophrenia. Biol Psychiatry 51: 969–971. doi: 10.1016/s0006-3223(02)01399-9
- 31. Yung AR, Phillips LJ, McGorry PD, McFarlane CA, Francey S, et al. (1998) Prediction of psychosis. A step towards indicated prevention of schizophrenia. Br J Psychiatry Suppl 172: 14–20.
- 32. Wing JK, Babor T, Brugha T, Burke J, Cooper JE, et al. (1990) SCAN. schedules for clinical assessment in neuropsychiatry. Arch Gen Psychiatry 47: 589–593. doi: 10.1001/archpsyc.1990.01810180089012
- 33. Kay SR, Fiszbein A, Vital-Herne M, Fuentes LS (1990) The positive and negative syndrome scale–spanish adaptation. J Nerv Ment Dis 178: 510–517. doi: 10.1097/00005053-199008000-00007
- 34. Addington D, Addington J, Schissel B (1990) A depression rating scale for schizophrenics. Schizophr Res 3: 247–251. doi: 10.1016/0920-9964(90)90005-r
- 35. Gardner DM, Murphy AL, O’Donnell H, Centorrino F, Baldessarini RJ (2010) International consensus study of antipsychotic dosing. Am J Psychiatry 167: 686–693. doi: 10.1176/appi.ajp.2009.09060802
- 36. Nuechterlein KH, Green MF, Kern RS, Baade LE, Barch DM, et al. (2008) The MATRICS consensus cognitive battery, part 1: Test selection, reliability, and validity. Am J Psychiatry 165: 203–213. doi: 10.1176/appi.ajp.2007.07010042
- 37. Rodriguez-Jimenez R, Bagney A, Garcia-Navarro C, Aparicio AI, Lopez-Anton R, et al. (2012) The MATRICS consensus cognitive battery (MCCB): Co-norming and standardization in spain. Schizophr Res 134: 279–284. doi: 10.1016/j.schres.2011.11.026
- 38. Baron RM, Kenny DA (1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 51: 1173–1182. doi: 10.1037/0022-35184.108.40.2063
- 39. Shrout PE, Bolger N (2002) Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychol Methods 7: 422–445. doi: 10.1037//1082-989x.7.4.422
- 40. Preacher KJ, Hayes AF (2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 40: 879–891. doi: 10.3758/brm.40.3.879
- 41. Dickinson D, Ramsey ME, Gold JM (2007) Overlooking the obvious: A meta-analytic comparison of digit symbol coding tasks and other cognitive measures in schizophrenia. Arch Gen Psychiatry 64: 532–542. doi: 10.1001/archpsyc.64.5.532
- 42. Riecher-Rossler A, Pflueger MO, Aston J, Borgwardt SJ, Brewer WJ, et al. (2009) Efficacy of using cognitive status in predicting psychosis: A 7-year follow-up. Biol Psychiatry 66: 1023–1030. doi: 10.1016/j.biopsych.2009.07.020
- 43. Ojeda N, Pena J, Sanchez P, Elizagarate E, Ezcurra J (2008) Processing speed mediates the relationship between verbal memory, verbal fluency, and functional outcome in chronic schizophrenia. Schizophr Res 101: 225–233. doi: 10.1016/j.schres.2007.12.483
- 44. Dickinson D (2008) Digit symbol coding and general cognitive ability in schizophrenia: Worth another look? Br J Psychiatry 193: 354–356. doi: 10.1192/bjp.bp.108.049387
- 45. Kelleher I, Murtagh A, Clarke MC, Murphy J, Rawdon C, et al. (2012) Neurocognitive performance of a community-based sample of young people at putative ultra high risk for psychosis: Support for the processing speed hypothesis. Cogn Neuropsychiatry.
- 46. Fusar-Poli P, Bonoldi I, Yung AR, Borgwardt S, Kempton MJ, et al. (2012) Predicting psychosis: Meta-analysis of transition outcomes in individuals at high clinical risk. Arch Gen Psychiatry 69: 220–229. doi: 10.1001/archgenpsychiatry.2011.1472
- 47. Lee BJ, Lee SJ, Kim MK, Lee JG, Park SW, et al. (2013) Effect of aripiprazole on cognitive function and hyperprolactinemia in patients with schizophrenia treated with risperidone. Clin Psychopharmacol Neurosci 11: 60–66. doi: 10.9758/cpn.2013.11.2.60
- 48. Stewart SA (2005) The effects of benzodiazepines on cognition. J Clin Psychiatry 66 Suppl 29–13.
- 49. Zhang X, Bu R, Sha W, Wang X, Liu J, et al. (2013) Serum prolactin and smoking status in chronic antipsychotic-treated male patients with schizophrenia. Psychiatry Res 209: 239–241. doi: 10.1016/j.psychres.2013.04.026
- 50. Segarra R, Zabala A, Eguiluz JI, Ojeda N, Elizagarate E, et al. (2011) Cognitive performance and smoking in first-episode psychosis: The self-medication hypothesis. Eur Arch Psychiatry Clin Neurosci 261: 241–250. doi: 10.1007/s00406-010-0146-6