United States Court of Appeals
                           FOR THE EIGHTH CIRCUIT
                                   ___________

                                   No. 06-3640
                                   ___________

Lisa Cox,                            *
                                     *
             Appellant,              *
                                     * Appeal from the United States
      v.                             * District Court for the
                                     * Eastern District of Arkansas.
Michael J. Astrue, Commissioner      *
of Social Security,                  *
                                     *
             Appellee.               *
                                ___________

                             Submitted: April 13, 2007
                                Filed: July 26, 2007
                                 ___________

Before WOLLMAN, BEAM, and COLLOTON, Circuit Judges.
                          ___________

WOLLMAN, Circuit Judge.

      Lisa Cox appeals the district court’s1 order upholding the Social Security
Commissioner’s denial of her application for disability insurance benefits. Cox argues
that as a result of procedural errors and ambiguous medical evidence, the
administrative law judge’s (ALJ) determination that she was not disabled was not
premised on substantial evidence. We affirm.


      1
       The Honorable Henry L. Jones, Jr., United States Magistrate Judge for the
Eastern District of Arkansas, to whom the case was referred for final disposition by
consent of the parties pursuant to 28 U.S.C. § 636(c).
                                          I.

       Cox contends that she has been qualified for disability benefits since December
11, 2002, because anxiety, mental retardation, and a respiratory impairment prevent
her from working. At the time of the ALJ’s decision, Cox was thirty-seven years old.
According to her Social Security Administration disability form and testimony, she
reported having received a tenth grade education and having attended special
education classes. She attempted to receive a GED, but was unsuccessful. She also
asserted that she had worked full time, on and off, as a certified nurse’s aide (CNA)
from 1994 to 1996, although she acknowledges that she never received special job
training or attended a trade or vocational school, and had reported earnings averaging
approximately $2000 a year during that period.

       Cox has had a chronically tumultuous home life. She testified that she was
molested by her father from the age of nine to sixteen, and as a result has difficulty
concentrating, handling stress, and dealing with people. She has three children, and
she still lives with and takes care of her sixteen-year-old daughter, who suffers from
severe mental impairments. She informed a psychiatrist that she had been married
three times to abusive men, that she receives no child support from her children’s
father, and that she has had to care for various other family members.

       In April 2002, Angela McKinness, an advanced practice nurse, diagnosed Cox
with insomnia and generalized anxiety disorder. Nurse McKinness prescribed
medication to help Cox with these issues. On May 8, 2003, Dr. Mary Ellen Ziolko
performed a consultative psychological evaluation. Dr. Ziolko described Cox’s affect
and mood as depressed and anxious. An administration of the Wechsler Adult
Intelligence Scale indicated that Cox had full scale, verbal, and performance IQ scores
in the mid- to upper-sixties. These results were considered valid. Dr. Ziolko’s
summary report and diagnosis, however, made facially conflicting statements
concerning Cox’s status. Although she reported that Cox’s intellectual functioning

                                         -2-
falls in the “mild” retardation range,2 she also indicated in her evaluation of adaptive
functioning that there “did not appear to be significant limitations in two or more areas
of adaptive behavior. Adaptive behavior appeared more consistent with ‘borderline’
intellectual functioning than mental retardation.”

       Cox was subsequently treated by Dr. Mohammed Al-Taher for her depression,
anxiety, and insomnia. Dr. Al-Taher periodically adjusted Cox’s medication in
response to her needs. At various points, Dr. Al-Taher noted that the treatment
appeared to be yielding positive results, but Cox’s tumultuous family life and
manipulative daughter often resulted in the return of depressive episodes. The record
indicates that Dr. Al-Taher diagnosed Cox with mild depression and dependent
personality traits. Cox testified that she suffers from anxiety attacks two to three times
a week, does not have her anxiety and depression completely under control even with
medication, and would cry if criticized in a work environment. She stated that she is
routinely subject to crying spells and constantly thinks of her experience as a victim
of molestation. Nevertheless, she acknowledged that she was not plagued by most of
these problems when she worked as a CNA at a nursing home and her children were
younger. She left that job in order to take care of her children.

      After reviewing the entirety of the record, the ALJ found that although Cox’s
IQ scores were within the range of mild mental retardation, because of both her ability
to perform a wide variety of daily activities and Dr. Ziolko’s conclusion that her
adaptive functioning was more consistent with borderline intellectual functioning, Cox
did not have an impairment listed in, or medically equal to, those set forth in the
Federal Regulations.3 Furthermore, the ALJ found that her subjective complaints


      2
       The mild retardation conclusion appears in the diagnosis section of the report.
      3
        Specifically, the ALJ noted that Cox’s IQ scores are inconsistent with her
ability to do the following: work as a nursing assistant for over two years, care for
children, shop for groceries and clothes, pay bills, count change, cook, and drive.

                                           -3-
were not borne out by the record and were not fully credible. After the ALJ
determined Cox’s residual functional capacity (RFC), he posed hypotheticals to the
vocational expert (VE) consistent with Cox’s RFC. The VE indicated that an
individual with Cox’s RFC who can perform functionally light work could work as
a bench assembler or small products assembler. Accordingly, the ALJ concluded that
Cox lacked a cognizable disability as defined in the Social Security Act.

       On appeal, Cox contends that the ALJ erred by (1) not seeking clarification
from Dr. Ziolko, whose report contradicted itself by indicating that Cox had mild
retardation while simultaneously concluding that she had borderline intellectual
functioning inconsistent with mild retardation; (2) failing to recontact Dr. Al-Taher
and Nurse McKinness to determine how they believed her depression and anxiety
affect her ability to work; and (3) asking the vocational expert hypothetical questions
that did not include all of the relevant details of Cox’s residual functional capacity,
thereby rendering the answers unreliable.

                                           II.

        “It is not the role of this court to reweigh the evidence presented to the ALJ or
to try the issue in this case de novo.” Loving v. Dep’t of Health & Human Servs., 

, 969 (8th Cir. 1994). Instead, we review the ALJ’s decision to determine
whether it is supported by substantial evidence on the record as a whole. 

is less than a preponderance, but enough that a reasonable mind
might accept it as adequate to support a decision.” Cox v. Apfel, 

,
1206-07 (8th Cir. 1998). Our review extends beyond examining the record to find
substantial evidence in support of the ALJ’s decision; we also consider evidence in
the record that fairly detracts from that decision. 

If, after conducting this
review, we find that “‘it is possible to draw two inconsistent positions from the
evidence and one of those positions represents the [Secretary’s] findings, we must
affirm the decision’ of the Secretary.” Siemers v. Shalala, 

, 301 (8th Cir.

                                           -4-
1995) (alteration in original) (quoting Robinson v. Sullivan, 

, 838 (8th
Cir. 1992)).

       The ALJ considered Cox’s impairment by conducting the familiar five-step
evaluation set forth in 20 C.F.R. § 404.1520(a)-(g) (2004). Under the regulations, the
ALJ determines: (1) whether the claimant is currently engaged in substantial gainful
activity; (2) whether the claimant’s impairments are so severe that they significantly
limit the claimant’s physical or mental ability to perform basic work activities; (3)
whether the claimant has impairments that meet or equal a presumptively disabling
impairment specified in the regulations; (4) whether the claimant’s RFC is sufficient
for her to perform her past work; and finally, if the claimant cannot perform her past
work, the burden shifts to the Commissioner to prove that (5) there are other jobs in
the national economy that the claimant can perform given the claimant’s RFC, age,
education and work experience. See Cox v. 

. Cox’s claims
of error relate to steps three and five.

                               A. Mental Retardation

       The ALJ found that substantial evidence supported the conclusion that Cox’s
mental impairments did not meet or equal the listed requirements for mental
retardation. We agree. For Cox’s purposes, to qualify as presumptively disabled due
to mental retardation, substantial evidence must support the presence of a “valid
verbal, performance, or full scale IQ of 60 through 70” and “a physical or other mental
impairment imposing an additional and significant work-related limitation of
function,” whose onset had occurred by age twenty-two. 20 C.F.R. Pt. 404, Subpt. P,
App. 1, § 12.05(c); Maresh v. Barnhart, 

, 899 (8th Cir. 2006) (describing
the § 12.05(c) requirements).

      Generally, social security hearings are non-adversarial. Snead v. Barnhart, 

, 838 (8th Cir. 2004). An ALJ bears a responsibility to “develop the record

                                         -5-
fairly and fully, independent of the claimant’s burden to press his case.” 

then, that because Dr. Ziolko’s medical report at least appeared to diagnose
her with mental retardation, the ALJ should not have relied on contradictory language
indicating that her adaptive function was more in line with borderline functioning than
mental retardation to support its determination that Cox was not mentally retarded.
Instead, the ALJ should have determined that Cox was mentally retarded or else
consulted Dr. Ziolko for further clarification of the report’s discrepancy. See 

(“Because [the] evidence might have altered the outcome of the
disability determination, the ALJ’s failure to elicit it prejudiced [the claimant] in his
pursuit of benefits.”).

       It is clear that Dr. Ziolko did not intend to ultimately diagnose Cox with mental
retardation, however. The totality of the clinical record supports the ALJ’s conclusion
and was not ignored. Dr. Ziolko’s report recounts Cox’s ability to effectively
communicate, her generally successful social relationships as exemplified by her
relationships with her children, the many ways in which Cox has exhibited self-
sufficient behavior, her lack of physical problems, and the lack of any limitations in
her concentration, persistence, or pace. Each of these categories is a relevant adaptive
functioning skill area. Dr. Ziolko concluded from these findings that Cox’s adaptive
behavior is more consistent with “borderline intellectual functioning” than mental
retardation, and noted that there did not appear to be limitations in two or more areas
of adaptive behavior – a medical prerequisite for the mental retardation classification.4
Her discussion and conclusion, in fact, directly parallel the analytical considerations
considered essential for the determination of mental retardation as enumerated in the
American Psychiatric Association, Diagnostic and Statistical Manual of Mental
Disorders 41 (4th ed. text revision 2000) (hereinafter “DSM-IV”), which states that

      4
        We note that the medical standard for mental retardation is not identical to the
legal standard. Maresh v. Barnhart, 

, 899 (8th Cir. 2006). Here,
however, the medical definition is relevant for purposes of interpreting what was
stated in a medical document.

                                          -6-
“[t]he essential feature of Mental Retardation is significantly subaverage general
intellectual functioning . . . accompanied by significant limitations in adaptive
functioning in at least two [enumerated] skill areas . . . .”

       In light of this direct, precise, and extensive discussion in Dr. Ziolko’s report,
which supports and culminates in an effective diagnosis of borderline intellectual
functioning, we conclude that her remark indicating a contrary diagnosis of mild
mental retardation was the result of inadvertence or imprecision. To hold otherwise
would require the improbable conclusion that Dr. Ziolko had intended to offer a
cursory diagnosis in direct contradiction to the careful findings and conclusions she
thoroughly recounted and characterized on prior pages, and in direct contradiction to
the defining diagnostic characteristics of mental retardation as identified by the DSM-
IV that her analysis explicitly addressed. The ALJ’s treatment of her report here, then,
does not involve his substituting judicial conjecture for ambiguous medical opinion,
but instead involves recognizing what amounts to an obvious medical opinion
inopportunely attended by superficially distracting or misleading language. In these
circumstances, there is no need for further clarification. See 20 C.F.R. § 416.912(e)
(2006) (requiring the recontacting of a treating physician only if evidence from that
physician is inadequate to determine disability).

       Taking into account the medical report and other evidence on the record
concerning Cox’s impairments, then, substantial evidence supports the ALJ’s
determination that Cox’s impairments neither meet nor equal those limitations
required for mental retardation. We reach this conclusion with full awareness of the
very real difficulties Cox appears to experience. The difficulties are not sufficient to
merit reversal, however, because “we will not reverse the decision even if substantial
evidence also supports a different outcome.” Stormo v. Barnhart, 

, 805
(8th Cir. 2004) (citing Fredrickson v. Barnhart, 

, 976 (8th Cir. 2004)).
On balance, we consider the ALJ’s determination to be supported by substantial
evidence – particularly in light of the activities Cox acknowledged she could perform,

                                          -7-
not least of which was her assertion that she previously worked as a CNA, a
semiskilled job, without experiencing difficulties for more than two years. Her
counsel contends that given Cox’s limited education, lack of specialized training, and
marginal income from the CNA work, the job title she provides is “probably” little
more than an attempt to put a positive spin on what amounts to a nurse’s helper
position. This is no more than speculation, however. We have not been made aware
of any evidence on record controverting her assertion that she worked as a CNA.

                B. The Residual Functional Capacity Determination

      Cox next contends that the ALJ erroneously established Cox’s RFC in the
absence of any medical opinion by Nurse McKinness and Dr. Al-Taher directly
addressing how her depression and anxiety affect her ability to work. As a result, Cox
argues that the ALJ’s RFC determination amounted to no more than a “layman’s
guess” at the work-related restrictions imposed by Cox’s anxiety and depression.
Cox’s argument lacks merit because the medical evidence provided a sufficient basis
on which her RFC was determined. The ALJ determined that Cox had the residual
functional capacity to perform unskilled or semiskilled work and that Cox should have
only superficial incidental contact with the public and co-workers, experience few
changes in work setting, and perform only simple, routine, repetitive tasks involving
no more than limited decision making.

       Because a claimant’s RFC is a medical question, an ALJ’s assessment of it must
be supported by some medical evidence of the claimant’s ability to function in the
workplace. Lauer v. Apfel, 

, 704 (8th Cir. 2001) (citing Nevland v.
Apfel, 

, 858 (8th Cir. 2000)). Accordingly, the regulations provide that
treating physicians or psychologists will be recontacted by the Commissioner when
the medical evidence received from them is inadequate to determine a claimant’s
disability. 20 C.F.R. § 416.912(e). Nevertheless, in evaluating a claimant’s RFC, an
ALJ is not limited to considering medical evidence exclusively. Lauer, 245 F.3d at

                                         -8-
704; Dykes v. Apfel, 

, 866 (8th Cir. 2002) (per curiam) (“To the extent
[claimant] is arguing that residual functional capacity may be proved only by medical
evidence, we disagree.”). Even though the RFC assessment draws from medical
sources for support, it is ultimately an administrative determination reserved to the
Commissioner. 20 C.F.R. §§ 416.927(e)(2), 416.946 (2006).

       The ALJ’s RFC determination with respect to the work-related effects of Cox’s
depression and anxiety was sufficiently supported by the medical evidence. The ALJ
considered Dr. Al-Taher’s examination in which Dr. Al-Taher assessed only a mild
depressive disorder. Dr. Al-Taher indicated that Cox’s depression was largely due to
her chaotic lifestyle and family-life. The record also indicates that Dr. Al-Taher
assigned Cox a Global Assessment of Functioning (GAF) score of sixty-five, which
indicates that even though she suffers from some mild symptoms such as depressed
mood and mild insomnia, or experiences some difficulty in social or occupational
functioning, she generally functions reasonably well and is capable of having
meaningful interpersonal relationships.5 DMS-IV at 34. Furthermore, although Cox
reported intermittent depressive symptoms (usually spurred by a family crisis), Dr. Al-
Taher stated that they “partially improved fairly quickly.” Despite Cox’s ability to
show some partial recovery from depressive episodes, Dr. Al-Taher acknowledged
that some socially avoidant symptoms remain. Finally, the ALJ found to be
significant Dr. Al-Taher’s observation that during her treatment, Cox was capable of
superficial social contact as evidenced by her ability to go Christmas shopping at a
shopping mall for a five-hour period without difficulty despite her alleged inability
to be around others and her proclivity to experience panic attacks. In light of these
facts, observations, and medical conclusions which bear directly on the extent of


      5
        For added context, we note that a GAF score in the fifties may be associated
with a moderate impairment in occupational functioning, and a GAF score in the
forties may be associated with a serious impairment in occupational functioning.
American Psychiatric Association, Diagnostic and Statistical Manual of Mental
Disorders 34 (4th ed. text revision 2000).

                                         -9-
Cox’s ability to function in a work environment, Dr. Al-Taher’s records persuade us
that the ALJ’s RFC assessment is supported by substantial medical evidence.6

                            C. The Hypothetical Questions

        Finally, Cox contends that in step five of its analysis, the ALJ relied on the
vocational expert’s response to a hypothetical that did not include all of the limitations
contained in the RFC. Testimony from a vocational expert is substantial evidence
only when the testimony is based on a correctly phrased hypothetical question that
captures the concrete consequences of a claimant’s deficiencies. Porch v. Chater, 

, 572 (8th Cir. 1997) (citing Pickney v. Chater, 

, 297 (8th Cir.
1996)). Although the ALJ asked the vocational expert three hypotheticals, only the
first and third have bearing on our discussion of Cox’s claim. The relevant language
of each is as follows:

      Assume . . . that this person has very limited reading and writing ability
      at the elementary level, and very basic math ability. Assume further . . .
      that this individual has a capacity for medium work, but with the
      following additional limitations, the individual is limited to simple,


      6
        In support of her position, Cox cites a number of cases in which we reversed
and remanded with instructions to make a new determination after further
development of the record. The cases she cites, however, rely on inapposite fact
patterns or are otherwise distinguishable. For example, in Bowman v. Barnhart, 

(8th Cir. 2002), we remanded because the doctor’s “somewhat cursory”
entries did not adequately assess how the claimant’s impairments limited her work-
related activities. 

Cox appears to believe that Bowman supports her
contention that the absence of any explicit reference to “work” in close proximity to
the description of her various medically evaluated limitations makes it impossible for
the ALJ to ascertain her work-related limitations from that evaluation. It does not.
Such explicit language is unnecessary here because Dr. Al-Taher’s evaluations
describe Cox’s functional limitations with sufficient generalized clarity to allow for
an understanding of how those limitations function in a work environment.

                                          -10-
      repetitive, routine tasks, with low stress to be defined as limited decision
      making required, and limited changes of the work setting, and that the
      individual can only have incidental contact with the public and co-
      workers. Could such an individual perform the relevant work of Ms.
      Cox?

      ....

      Assume, if you would, we have the same requirement for unskilled work
      that I discussed with the limited decision-making and only incidental
      contact with the public, but restricted to a light functional level. Would
      there be any employment for such an individual?

Tr. 280-81.

       Cox alleges that the third hypothetical fails to incorporate all of the material in
the first. Cox parses the language of the third hypothetical too finely, however, and
does not appreciate the full relevance of its surrounding context. Cox does not
contend that the first hypothetical fails to capture the concrete consequences of her
deficiencies. Furthermore, our plain reading of the third hypothetical comports with
the district court’s understanding of it – that is, one hearing it would assume from its
language that it incorporates the same limitations described in the first hypothetical
except for a change in the permissible exertion level from medium to light.
Accordingly, because the vocational expert, in response to a hypothetical that captures
the consequences of Cox’s deficiencies, described readily available occupations in
which she could engage, the Commissioner successfully demonstrated Cox’s ability
to perform work in the economy. See Reed v. Sullivan, 

, 815-16 (8th
Cir. 1993) (describing the burden).

      The judgment is affirmed.
                      ______________________________



                                          -11-