Introduction
There are several reasons to suspect that the risk of developing diabetes might be different between deficit (i.e., those with primary, enduring negative symptoms) and nondeficit schizophrenia. Increased rates of diabetes and insulin resistance in people with schizophrenia predate antipsychotics (1, 2). This indicates some innate biological predisposition to impaired glucose utilization and, since deficit schizophrenia may have a unique pathophysiology (3–5), people with deficit schizophrenia may show a different risk for diabetes when separated out from schizophrenia as a whole. Negative symptoms such as decreased activity, goals, and an absence of important relationships may also impair the ability of someone with deficit schizophrenia to implement the diet and exercise changes recommended for prediabetes (fasting blood sugar between 100–125 mg/dL) (6), leading to higher rates of progression to diabetes in this group.
Kirkpatrick et al. (7, 8) compared newly diagnosed, antipsychotic-naive deficit and nondeficit groups to healthy controls using an oral glucose tolerance test. There were significant differences among all three groups (controls 85.0±21.6 mg/dL, nondeficit 123.7±42.2 mg/dL, deficit 100.2±23.1 mg/dL). A second study by these authors replicated the finding that newly diagnosed deficit participants had lower blood sugar after glucose challenge compared to nondeficit participants (9). This difference between deficit and nondeficit participants supports the notion of differing biological risk.
We conducted a chart review to determine whether there is a differential risk of diabetes between deficit and nondeficit groups with chronic schizophrenia. Differential prevalence rates of diabetes would extend the work of Kirkpatrick and colleagues and further support differential biological mechanisms behind deficit and nondeficit schizophrenia. Differential prevalence rates between the groups might also support the utility of the deficit/nondeficit categorization in order to target scarce clinical resources toward one or the other group to produce the greatest impact on metabolic health in schizophrenia.
Methods
Data were abstracted from all current outpatient charts and from charts closed within the year from the Maryland Psychiatric Research Center (MPRC) Outpatient Research Program clinic. The deficit group was supplemented with data from discharged patients since 1988 and deficit patients from another MPRC clinic (the Schizophrenia Related Disorders Program). Patients were diagnosed with schizophrenia/schizoaffective disorder via best estimate approach (structured clinical interviews, examination of records, caregiver informants). Patients were categorized as deficit or nondeficit using the Schedule for the Deficit Syndrome (SDS) (10). Categorization was made by the SDS developers or people directly trained by them.
For all charts, the most recently documented laboratory work and medical information was collected: demographics (including antipsychotic), diabetes status (presence or absence and type 1 or 2), fasting plasma glucose level, hemoglobin A1c (HgA1c) percent, total cholesterol, weight, height, and blood pressure. If there was no diagnosis of diabetes, but fasting blood sugar (>125 mg/dL) or HgA1c (>6.5%) results were diagnostic for diabetes, then the patient was recorded as having type 2 diabetes. Records were connected via unique identifiers to a database containing deficit/nondeficit status. Type 1 diabetes was excluded from analysis as the etiopathophysiology is markedly different from type 2 diabetes.
This study was reviewed and approved by the University of Maryland Institutional Review Board.
Statistical analysis included chi-square and t-test comparisons as appropriate, and logistic regression using age, body mass index (BMI), sex, and deficit diagnosis.
Results
One hundred and seventy-six charts were reviewed. Seven new patients were excluded for lack of SDS, leaving 72 deficit and 97 nondeficit patients. There were no group differences in age, sex, ethnicity, or antipsychotic (see Table 1). However, there was a numerically higher proportion of males in the deficit group. There were no group differences in prevalence of diabetes, fasting blood sugar, weight/BMI, cholesterol, or blood pressure. Including only type 2 diabetes patients, there were no differences in fasting blood sugar or HgA1c between deficit and nondeficit groups.
Table 1.
Demographics, Diabetes Diagnoses, and Metabolic Indices in Deficit and Nondeficit Groups
| Deficit | Nondeficit | Statistic (df) | P-Value | |||
|---|---|---|---|---|---|---|
| N | Measure | N | Measure | |||
| Total Sample | ||||||
| Age, years, mean (SD) | 72 | 45 (12.5) | 97 | 43 (12.7) | t(164)=0.57 | 0.57 |
| Male | 53 | 74% | 63 | 65% | χ2(1)=1.44 | 0.25 |
| Race | ||||||
| White | 38 | 53% | 56 | 58% | χ2(3)=5.59 | 0.13 |
| African American | 28 | 39% | 38 | 39% | ||
| Asian | 2 | 3% | 3 | 3% | ||
| Other | 4 | 6% | 0 | 0 | ||
| Antipsychotic Treatment* | ||||||
| First Generation | 14 | 22% | 14 | 16% | χ2(2)=5.34 | 0.07 |
| Second Generation† | 31 | 47% | 30 | 34% | ||
| Clozapine | 20 | 31% | 43 | 49% | ||
| Diabetes, prevalence (%) | 72 | 19 (26.4%) | 97 | 24 (24.7%) | χ2(1)=0.06 | 0.81 |
| Fasting Blood Sugar, mg/dL, mean (SD) | 68 | 107 (34.7) | 96 | 104 (34.4) | t(162)=0.49 | 0.63 |
| HgA1c %, mean (SD) | 37 | 6.0% (0.68) | 95 | 6.0% (0.76) | t(130)=0.31 | 0.76 |
| Weight, kg, mean (SD) | 65 | 90.7 (24.4) | 94 | 95.3 (21.9) | t(157)= −1.17 | 0.25 |
| BMI, mean (SD) | 56 | 30.4 (8.0) | 93 | 31.2 (7.8) | t(146)= −0.84 | 0.40 |
| Total Cholesterol, mg/dL, mean (SD) | 65 | 181 (40.9) | 95 | 178 (35.6) | t(158)=0.53 | 0.60 |
| Blood Pressure, Systolic/Diastolic, mmHg, mean (SD) | 65 | 126/79 (23.5/13.0) | 94 | 126/77 (15.2/10.8) | t(101)=0.26/t(157)=0.96 | 0.80/0.33 |
| Sub-sample with Diabetes | ||||||
| Fasting Blood Sugar, mg/dL, mean (SD) | 19 | 127 (57.4) | 24 | 130 (59.1) | t(41)= −0.20 | 0.84 |
| HgA1c %, mean (SD) | 15 | 6.34% (0.89) | 24 | 6.68% (1.12) | t(37)= −0.99 | 0.33 |
The proportion of deficit (χ2=0.93, p=1.0) and nondeficit patients (χ2=0.44, p=0.63) diagnosed with diabetes did not differ as a function of clozapine status. Deficit: clozapine with diabetes=30.0%; clozapine without diabetes=70.0%; non-clozapine with diabetes=28.9%; non-cloazpine without diabetes=71.1%. Nondeficit: clozapine with diabetes=23.3%; clozapine without diabetes=76.7%; non-clozapine with diabetes=29.5%; non-cloazpine without diabetes=70.5%.
Non-clozapine second-generation antipsychotics.
BMI=Body Mass Index; df=degrees freedom; HgA1c=hemoglobin A1c; mmHg=millimeters of mercury; SD=standard deviation.
Logistic regression was calculated to determine whether diabetes diagnosis was predicted by age, BMI, sex, and deficit diagnosis. The equation correctly classified 73% of cases, model χ2(4)=9.61, p<0.05. BMI was the only significant predictor of diabetes (B=−0.06; OR 0.95; 95% CI, 0.90–1.00, p<0.04); deficit diagnosis (B=−0.35; OR 0.71; 95% CI, 0.33–1.53), age (B=−0.03; OR 0.98; 95% CI, 0.95–1.01), and sex (B=0.23; OR 0.95; 95% CI, 0.55–2.87) were not predictors.
Discussion
There was no indication that prevalence rates of diabetes differed between chronic deficit and nondeficit schizophrenia patients, nor was there a group difference in glycemic control. This indicates either there are no differences in the biological mechanism behind the development of diabetes in the two groups, or that different biological mechanisms in the two groups still lead to the same overall prevalence of diabetes. It also indicates the symptom constellation of deficit schizophrenia is not more likely to lead to poorer glycemic control compared to nondeficit schizophrenia.
This study had a number of limitations. First, data quality was based on the accuracy of the medical record. Because only the most recent lab results were abstracted, cases of diabetes could have been missed (no formal chart diagnosis, recent labs were normal). However, diabetes rates observed in this study are consistent with those expected in a population with schizophrenia (11, 12), so it is likely few cases were missed. Another limitation is that we were unable to compare the groups on longitudinal risk for diabetes. Without knowing exactly when diabetes was diagnosed or, in many cases, the precise onset of schizophrenia, we do not know if there is a differential risk of diabetes in the groups over time (i.e., we measured prevalence, not incidence).
In summary, we found no difference in prevalence rates of type 2 diabetes between deficit and nondeficit schizophrenia. There were no differences in glycemic control between deficit and nondeficit schizophrenia groups, or between the subset of the groups with diabetes. This finding does not rule out differences in the biological mechanisms behind the occurrence of diabetes in the groups, but does argue against the utility of making the deficit/nondeficit categorization in a non-research setting in order to target resources aimed at enhancing metabolic health.
Footnotes
Conflicts of Interest
None of the authors report any conflicts of interest. Drs. Strauss and Fischer report past funding from the United States National Institutes of Health and the Department of Veterans Affairs. Dr. Fischer has also been funded by the Brain and Behavior Research Foundation (formerly NARSAD).
References
- 1.Holt RI, Bushe C, Citrome L. Diabetes and schizophrenia 2005: are we any closer to understanding the link? J Psychopharmacol. 2005;19(6 Suppl):56–65. doi: 10.1177/0269881105058379. [DOI] [PubMed] [Google Scholar]
- 2.Kohen D. Diabetes mellitus and schizophrenia: historical perspective. Br J Psychiatry Suppl. 2004;47:S64–66. doi: 10.1192/bjp.184.47.s64. [DOI] [PubMed] [Google Scholar]
- 3.Kirkpatrick B, Buchanan RW, Ross DE, Carpenter WT., Jr A separate disease within the syndrome of schizophrenia. Arch Gen Psychiatry. 2001;58(2):165–171. doi: 10.1001/archpsyc.58.2.165. [DOI] [PubMed] [Google Scholar]
- 4.Kirkpatrick B, Galderisi S. Deficit schizophrenia: an update. World Psychiatry. 2008;7(3):143–147. doi: 10.1002/j.2051-5545.2008.tb00181.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fischer BA, Kirkpatrick B, Carpenter WT. The neurobiology of negative symptoms and the deficit syndrome. In: Javitt DC, Kantrowitz JT, editors. The handbook of neurochemistry and molecular neurobiology: schizophrenia. 3. New York: Springer Science and Business Media LLC; 2009. pp. 507–523. [Google Scholar]
- 6.Ratner RE, Sathasivam A. Treatment recommendations for prediabetes. Med Clin North Am. 2011;95(2):385–395. doi: 10.1016/j.mcna.2010.11.007. [DOI] [PubMed] [Google Scholar]
- 7.Kirkpatrick B, Fernandez-Egea E, Garcia-Rizo C, Bernardo M. Differences in glucose tolerance between deficit and nondeficit schizophrenia. Schizophr Res. 2009;107(2–3):122–127. doi: 10.1016/j.schres.2008.09.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fernandez-Egea E, Bernardo M, Donner T, Conget I, Parellada E, Justicia A, et al. Metabolic profile of antipsychotic-naive individuals with non-affective psychosis. Br J Psychiatry. 2009;194(5):434–438. doi: 10.1192/bjp.bp.108.052605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Garcia-Rizo C, Fernandez-Egea E, Oliveira C, Justicia A, Bernardo M, Kirkpatrick B. Inflammatory markers in antipsychotic-naïve patients with nonaffective psychosis and deficit vs. nondeficit features. Psychiatry Res. 2012;198(2):212–215. doi: 10.1016/j.psychres.2011.08.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kirkpatrick B, Buchanan RW, McKenney PD, Alphs LD, Carpenter WT., Jr The Schedule for the Deficit syndrome: an instrument for research in schizophrenia. Psychiatry Res. 1989;30(2):119–123. doi: 10.1016/0165-1781(89)90153-4. [DOI] [PubMed] [Google Scholar]
- 11.Dixon L, Weiden P, Delahanty J, Goldberg R, Postrado L, Lucksted A, et al. Prevalence and correlates of diabetes in national schizophrenia samples. Schizophr Bull. 2000;26(4):903–912. doi: 10.1093/oxfordjournals.schbul.a033504. [DOI] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention. National diabetes fact sheet: National estimates and general information on diabetes and prediabetes in the United States. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2011. [Google Scholar]