In the

    United States Court of Appeals
                 For the Seventh Circuit
                     ____________________
No. 16-4007
ASHLEY GERSTNER,
                                                  Plaintiff-Appellant,

                                 v.

NANCY A. BERRYHILL,
                                                 Defendant-Appellee.
                     ____________________

         Appeal from the United States District Court for the
                   Eastern District of Wisconsin.
         No. 15-CV-1115 — Nancy Joseph, Magistrate Judge.
                     ____________________

    ARGUED OCTOBER 3, 2017 — DECIDED JANUARY 5, 2018
                ____________________

   Before KANNE, ROVNER, and SYKES, Circuit Judges.
    ROVNER, Circuit Judge. Ashley Gerstner challenges the
denial of her application for disability insurance benefits and
supplemental security income. An administrative law judge
found that she was severely impaired by anxiety, bipolar
disorder, panic disorder, depression, and fibromyalgia, and
that these impairments were not disabling. Gerstner con-
tends that the ALJ erred in assigning too little weight to her
treating psychiatrist’s opinions and in discrediting her com-
2                                                 No. 16-4007

plaints of fibromyalgia pain. We vacate the judgment and
remand.
                               I.
    Gerstner was 27 when she applied for disability benefits
and supplemental security income, alleging an onset date of
May 2011. Her mental impairments manifested during her
high school years. She was admitted to a psychiatric hospital
at 15 and has been treated with psychiatric and counseling
services. Because of anxiety, she was home schooled; she lat-
er transferred to a high school where she received her degree
by learning mostly on a computer by herself. She then
worked as an assistant manager at Blockbuster for five years
but quit abruptly after experiencing what she described as a
“mental breakdown.” She moved on to work at another re-
tail store in a managerial position that required fewer skills
than her previous job, but quit after six months because she
experienced another breakdown. Since May 2011, Gerstner
has remained unemployed.
    Between mid-2011 and mid-2012, Gerstner was treated
six times by Dr. Stephen Callaghan, M.D., a psychiatrist at
Psychiatric Treatment Services of Racine. In those visits,
Dr. Callaghan, who had treated Gerstner since 2006, diag-
nosed her with generalized anxiety disorder, depression,
and attention deficit disorder. He prescribed Xanax and oth-
er medications, and he frequently adjusted the dosages. But
Dr. Callaghan also noted that Gerstner appeared euthymic
(non-depressed) with normal affect.
   In connection with Gerstner’s application for benefits,
Dr. Callaghan completed a form assessment of Gerstner’s
mental health in mid-2012 and opined that she was extreme-
No. 16-4007                                                 3

ly limited socially and at work. He noted that since 2010
Gerstner’s mental health had deteriorated, and he estimated
that on average she could work only two to three hours per
day and likely would miss work seven days per month. In
response to a series of questions about “social adjustments,”
he checked boxes indicating that she had marked-to-extreme
limitations behaving in an emotionally stable manner, relat-
ing predictably in social situations, and demonstrating relia-
bility. He supported this assessment with findings that she
periodically felt suicidal and homicidal, had major problems
with social relations, and would withdraw from stressful
situations and not be able to function. On another part of the
form, in a section related to “occupational adjustments,” he
checked boxes reflecting that Gerstner had marked-to-
extreme limitations in her ability to deal with work stresses
and moderate-to-marked limits in maintaining attention—
findings that he based on her severe anxiety, depression, and
problems “handling any stress without shut[ting] down.”
Lastly, in response to a series of questions about “perfor-
mance adjustments,” he assessed Gerstner as markedly lim-
ited in her ability to understand and carry out detailed job
instructions because she would be overwhelmed by anxiety
and depression.
    After this assessment, Dr. Callaghan treated Gerstner
seven more times (all within a year), added diagnoses of bi-
polar disorder and panic disorder with agoraphobia, pre-
scribed medications to treat both, and noted that she had a
dysthymic mood (depressed) each visit. On one occasion
Dr. Callaghan noted that she had ideas of suicide and homi-
cide. But in subsequent exams, he noted that she no longer
had these thoughts and described her affect as normal.
4                                                No. 16-4007

     Dr. David Nichols, Ph.D., a psychologist who practiced
with Dr. Callaghan, met with Gerstner monthly (sometimes
more frequently) for hour-long visits. Dr. Nichols diagnosed
her in 2011 with major depressive and generalized anxiety
disorders, and in 2013 with bipolar disorder. Gerstner, after
filing her disability application, told Dr. Nichols that she
continued to look for a factory job.
    Gerstner was treated in 2013 by a nurse practitioner,
Nancy Maczka, who assessed her mental health on a form
identical to the one completed by Dr. Callaghan. She echoed
Dr. Callaghan’s findings that Gerstner had extreme limita-
tions with relating “predictably in social situations” and
“demonstrat[ing] reliability.” But unlike Dr. Callaghan, she
found Gerstner more limited in dealing with work stresses
and maintaining attention.
    In addition to her mental impairment, Gerstner says that
she was prevented from working by fibromyalgia. She first
complained of pain and weakness to Dr. Joseph Paukner,
M.D., in September 2011, and he referred her to a neurolo-
gist, Dr. Bhupendra Khatri, M.D., of the Center for Neuro-
logical Disorders in Milwaukee. Dr. Khatri examined Gerst-
ner in November 2011, concluded that she was “most likely”
suffering from fibromyalgia, and repeated this diagnosis at a
follow-up appointment in early January 2012 following an
MRI of Gerstner’s brain. (The MRI ruled out any neurologi-
cal change that might have accounted for her complaints of
worsening pain). Soon thereafter, Dr. Tracy Brenner, M.D., a
physician at the Milwaukee Rheumatology Center, found
that Gerstner had fourteen of eighteen positive tender
points, a finding that led the doctor to opine that Gerstner
had a “high suspicion for fibromyalgia.” Dr. Brenner de-
No. 16-4007                                                 5

ferred management of that condition to Dr. Paukner and Dr.
Callaghan. Dr. Paukner then diagnosed Gerstner with fi-
bromyalgia, for which he prescribed Lyrica, a pain reliever.
The next month, in response to Gerstner’s complaints of hav-
ing good and bad days, Dr. Paukner increased the dosage.
At an appointment later that year, Gerstner rated her pain
from fibromyalgia as a nine out of ten, and was prescribed
a stronger pain medication—methadone.
    In June 2012, the month before Dr. Callaghan completed
his assessment, a state-agency consultant, Dr. Craig Childs,
Ph.D., concluded from a review of Gerstner’s medical rec-
ords that she was only moderately limited in several tasks:
completing a normal workday and workweek, maintaining
concentration for extended periods, carrying out detailed
instructions, and interacting with the general public.
    Gerstner lost her health insurance in 2013. She ceased
treatment with Dr. Callaghan and went six months without
medication for her fibromyalgia pain.
    The Agency denied Gerstner’s application for disability-
insurance benefits and supplemental-security income, both
initially and on reconsideration.
   At a hearing before an ALJ in 2014, Gerstner described
how her health had deteriorated since 2010. She testified that
she had moved back into her parents’ house and experi-
enced trouble interacting with others, handling stress and
pressure, sleeping, and concentrating. She added that she
had difficulty making phone calls and leaving her house.
She said she usually took Xanax twice daily for her anxiety,
and she experienced shooting pain from her fibromyalgia
that was aggravated by stress, prolonged sitting and stand-
6                                                  No. 16-4007

ing, and exercise. Her flare-ups lasted from one to three
hours. She worried that working would exacerbate her pain
the next day. Since her fibromyalgia diagnosis, she had
gained one hundred pounds.
    The ALJ applied the standard 5-step analysis,
see 20 C.F.R. §§ 404.1520(a), 416.920(a), and concluded that
Gerstner was not disabled. The ALJ determined that she had
not engaged in substantial gainful activity since her alleged
onset date (step one); that her conditions—fibromyalgia, de-
pressive disorder, anxiety disorder, bipolar disorder, and
panic disorder—were severe impairments (step two); that
these impairments did not meet a listing for presumptive
disability, individually or in combination (step three); that
she had the residual functional capacity to perform light, un-
skilled work limited to “simple, routine and repetitive
tasks,” no interaction with the public, and “only occasional
interaction with coworkers” (step four); and that although
she could not perform her past work, she could work as a
night cleaner, price marker, or call router, as a vocational ex-
pert had concluded based on the ALJ’s assessment of Gerst-
ner’s residual functional capacity (step five).
    In deciding Gerstner’s residual functional capacity, the
ALJ discounted her account of disabling limitations and her
treating psychiatrist’s opinions. The ALJ found that her
statements about the “intensity, persistence, and pace” of her
symptoms of mental illnesses and fibromyalgia were “not
entirely credible.” He also gave “little weight” to the mid-
2012 opinions of Dr. Callaghan because the limitations he
marked were “extreme” compared to Dr. Callaghan’s own
“findings and observations,” which the ALJ said were “rela-
tively normal mental status examinations” with a “few ex-
No. 16-4007                                                   7

ceptions.” By contrast, the ALJ gave “great weight” to the
June 2012 opinion of the reviewing agency consultant, Dr.
Childs, who concluded that Gerstner had only moderate lim-
itations. The ALJ also purported to rely on “updated medical
evidence” to include “somewhat greater limitations” in
Gerstner’s residual functional capacity than what Dr. Childs
had opined.
   After the Appeals Council denied review, a magistrate
judge presiding by consent, see 28 U.S.C. § 636(c), affirmed
the decision of the Commissioner.
                                II.
   A. Treating Psychiatrist’s Opinions
    On appeal, Gerstner challenges the ALJ’s decision to give
“little weight” to Dr. Callaghan’s opinions in his mid-2012
assessment of her mental health. A treating physician's opin-
ion on the nature and severity of a medical condition is enti-
tled to controlling weight if it is well supported by medical
findings and is consistent with other evidence in the record.
20 C.F.R. § 404.1527(c)(1); Brown v. Colvin, 

, 252
(7th Cir. 2016). (This is the rule governing claims filed before
March 27, 2017, see 20 C.F.R. §§ 404.1520c(a) (2017).) The ALJ
declined to give Dr. Callaghan’s opinions controlling weight
because they were “extreme compared to his own findings
and observations. With a few exceptions, Dr. Callaghan not-
ed relatively normal mental status examinations.” Gerstner
argues that the ALJ cherry-picked Dr. Callaghan’s findings
about mood and affect and disregarded his diagnoses of de-
pression, anxiety, and other “abnormal findings.” She con-
tends also that he failed to consider the requisite factors for
8                                                 No. 16-4007

evaluating medical source opinions set forth in the applica-
ble regulation.
    We agree with Gerstner that the ALJ fixated on select
portions of Dr. Callaghan’s treatment notes and inadequate-
ly analyzed his opinions. First, with regard to Dr. Calla-
ghan’s reports from six exams before August 2012, the ALJ
focused on notes about mood and affect but ignored Dr. Cal-
laghan’s diagnoses of depression and anxiety disorder. “An
ALJ may not selectively discuss portions of a physician's re-
port that support a finding of non-disability while ignoring
other portions that suggest a disability.” Campbell v. Astrue,

, 301 (7th Cir. 2010). Here, the ALJ did not speci-
fy which of Dr. Callaghan’s findings were “normal,” but he
did say that Dr. Callaghan’s observations of Gerstner’s eu-
thymic mood and normal affect were inconsistent with her
“allegations” of “disabling affective disorders.” But
Dr. Callaghan made these observations in the same reports
before August 2012 in which he diagnosed her with depres-
sion and anxiety, and yet the ALJ ignored these diagnoses.
Moreover, the affect and mood notes that the ALJ empha-
sized simply described how Gerstner presented on the days of
her appointments. They were not general assessments.
    Second, the ALJ concluded from Dr. Callaghan’s notes af-
ter August 2012 that Gerstner’s mental health had improved,
disregarding other portions of those notes that undermine
his conclusion. After August 2012, Dr. Callaghan found dur-
ing each visit that Gerstner had a normal affect, no “im-
paired” thoughts, and no suicidal ideations, but also that she
reported a dysthymic mood and experienced anxiety, de-
pression, and problems sleeping and concentrating. The ALJ
considered only the signs of possible improvements in these
No. 16-4007                                                    9

notes and ignored the negative findings. But all findings in
psychiatric notes must be considered, even if they were
based on the patient’s own account of her mental symptoms,
see Price v. Colvin, 

, 839–40 (7th Cir. 2015) (citing
Adaire v. Colvin, 

, 688 (7th Cir. 2015)). Further-
more, every treatment note after August 2012 repeats Dr.
Callaghan’s diagnoses and treatments; he even changed the
medication for Gerstner’s bipolar disorder in January 2013
and increased the dosage three months later. These un-
changed diagnoses and the medication adjustments belie the
conclusion that Gerstner’s mental health had improved.
   Third, the ALJ ignored how Dr. Callaghan’s mid-2012
opinions of Gerstner’s limitations were supported by his re-
peated findings of depression and dysthymic mood. Dr. Cal-
laghan opined that Gerstner was “extremely limited” with
regard to such matters as emotional stability, reliability, and
predictability. This opinion conformed to his diagnosis of
depression since July 2011 and his observations since August
2012 that Gerstner had a dysthymic mood; dysthymia (de-
pression) is a condition that may “significantly interfere”
with work and relationships, Persistent depressive disorder
(dysthymia),                  MAYO                      CLINIC,
http://www.mayoclinic.org/diseases-conditions/persistent-
depressive-disorder/home/ovc-20166590 (visited Nov. 27,
2017).
    Fourth, the ALJ overlooked the extent to which
Dr. Callaghan’s opinions were consistent with the diagnoses
and opinions of other medical sources who treated Gerstner.
ALJs must consider psychologists’ and nurse practitioners’
opinions on the severity of a patient’s impairments.
See 42 U.S.C. § 423(d)(5)(B); 20 C.F.R. §§ 404.1502(a)(2),
10                                                          No. 16-4007

404.1527(a)(1),(b),(c); SSR 06-03P, 

(Aug. 9,
2006) (“[M]edical sources who are not ‘acceptable medical
sources,’ such as nurse practitioners  are important and
should be evaluated on key issues such as impairment sever-
ity and functional effects.”); see also 20 C.F.R. § 402.35(b)(1)
(“Social Security Rulings  are binding on all components
of the Social Security Administration.”). The ALJ failed to
mention that Dr. Nichols diagnosed Gerstner with the same
disorders found by Dr. Callaghan. Nor did the ALJ note that
Dr. Callaghan’s opinion was supported by Dr. Nichols’s rat-
ing of Gerstner’s Global Assessment of Functioning (GAF),
which measured her psychological, social, and occupational
abilities. See AMERICAN PSYCH. ASSOC., DIAGN. & STAT. MAN.
OF MENTAL DISORDERS 34 (4th Ed., Rev. 2000) (DSM-IV).
Gerstner’s GAF scores ranged from 40 to 55, indicating she
had serious difficulty (41 to 50) and moderate difficulty (51
to 60) in these abilities. Id.1 And a nurse practitioner,
Maczka, treated Gerstner and opined that her mental abili-
ties were limited to the same extent or more than that found
by Dr. Callaghan, but this supporting opinion, which the
ALJ wrongly discounted as coming from an unacceptable
medical source, went unmentioned in the ALJ’s analysis of
Dr. Callaghan’s opinions.




     1The DSM-V, the latest version of the Diagnostic and Statistical
Manual of Mental Disorders, has abandoned the GAF. AMERICAN PSYCH.
ASSOC., DIAGN. & STAT. MAN. OF MENTAL DISORDERS 16 (5th Ed., 2013)
(DSM-V). But the Social Security Administration still instructs ALJs to
treat GAF scores as medical-opinion evidence. See Craig v. Colvin, 659
F. App’x 381, 382 (9th Cir. 2016); Hughes v. Comm'r Soc. Sec., 643 F. App’x
116, 119 n.2 (3d Cir. 2016).
No. 16-4007                                                   11

     Fifth, even if there were sound reasons for refusing to
give Dr. Callaghan’s opinions controlling weight, the ALJ
still erred by assigning his opinions little weight without
considering relevant regulatory factors under 20 C.F.R.
§ 404.1527(c). ALJs must decide the weight of a treating phy-
sician’s non-controlling opinion by considering, to the extent
applicable, the treatment relationship’s length, nature, and
extent; the opinion’s consistency with other evidence; the
explanatory support for the opinion; and any specialty of the
treating physician. 20 C.F.R. § 404.1527(c); Moss v. Astrue,

, 561 (7th Cir. 2009); Bauer v. Astrue, 

,
608 (7th Cir. 2008). The ALJ here did not mention that
Dr. Callaghan is a psychiatrist, that he was the only psychia-
trist who treated Gerstner, that the treatments occurred reg-
ularly for six years (sometimes monthly), or that Dr. Calla-
ghan assessed her mental health thirteen times in the two
years between the onset date and the date when she lost
medical insurance. The ALJ also failed to consider the con-
sistency of Dr. Callaghan’s opinion with the opinions of oth-
er treating, examining, and reviewing medical sources. Alt-
hough the ALJ discussed the weight to afford these physi-
cians’ opinions, he did not specify how or to what extent he
considered these opinions when deciding to assign little
weight to Dr. Callaghan’s opinions.
   Because of these errors, substantial evidence does not
support the decision to afford little weight to Dr. Callaghan’s
opinions of Gerstner’s limitations from mental impairments,
and the case must be remanded for reconsideration of his
opinion, see Meuser v. Colvin, 

, 912 (7th Cir.
2016); Scott v. Astrue, 

, 740 (7th Cir. 2011).
12                                                 No. 16-4007

     B. Adverse Credibility Determination
    Gerstner also contends that the ALJ wrongly discounted
her testimony about the extent of her pain from fibromyal-
gia. The ALJ, using familiar boilerplate, said that the “claim-
ant’s statements concerning the intensity, persistence and
limiting effects of these symptoms are not entirely credible.”
Gerstner argues that the ALJ discredited her complaints of
intermittent fibromyalgia pain by overstating findings from
examining physicians’ diagnostic tests, in which she walked,
moved, sensed touch, and had no spasms or tenderness. She
argues further that the ALJ wrongly discredited her pain
complaints by misstating medical evidence from two exam-
ining physicians who recommended that she engage in un-
specified physical activity. She contends also that the ALJ
drew an unwarranted inference that her pain was not disa-
bling because she went six months without fibromyalgia
drugs, but the ALJ never considered the reason she offered
for that medication gap. Lastly, Gerstner argues that the ALJ
wrongly concluded that she could work based on a physi-
cian’s notation that she had searched for employment.
    We agree with Gerstner that the ALJ’s adverse credibility
determination must be overturned. This court will overturn
an ALJ’s adverse credibility finding if it is patently wrong.
See 

; Schaaf v. Astrue, 

, 875
(7th Cir. 2010). The ALJ here overstated test results and
treatment recommendations and drew unjustified inferences
from Gerstner’s medication gap and job search.
    First, the ALJ overstated findings from three diagnostic
tests to discredit Gerstner’s complaints of intermittent fi-
bromyalgia pain. The ALJ said that Gerstner’s pain com-
plaints were inconsistent with her abilities to sit, move, and
No. 16-4007                                                 13

walk for an unstated period in one exam, her normal sensa-
tion in extremities during another exam, and her lack of
spasms and tenderness in a third exam. But these findings
are consistent with Gerstner’s pain complaints. She never
testified that she had constant disabling pain, or that her
condition totally impaired the abilities tested in these exams.
Instead, she said that her pain was triggered by prolonged sit-
ting, standing, or activity and stress. The ALJ’s analysis re-
veals that he misunderstood the nature of her fibromyalgia
pain. The extent of fibromyalgia pain cannot be measured
with objective tests aside from a trigger-point assessment.
See 

; Fibromyalgia, Diagnosis,
MAYO        CLINIC,      http://www.mayoclinic.org/diseases-
conditions/fibromyalgia/diagnosis-treatment/diagnosis/dxc-
20317823 (visited Nov. 27, 2017). Trigger-point testing on
Gerstner pinpointed fibromyalgia as the source of her pain,
and her pain complaints were consistent with her prescrip-
tion for methadone, an opioid not intended for mild or acute
pain. Methadone Hydrochloride, PRSCBR’S. DIG. REFERENCE,
http://www.pdr.net/drug-summary/Methadone-
Hydrochloride-Intensol-Oral-Concentrate-methadone-
hydrochloride-3464 (visited Dec. 22, 2017).
    Second, the ALJ unjustifiably concluded that Gerstner’s
pain was not disabling because an examining physician rec-
ommended that she engage in “aerobic activity several times
a week” and a physician’s assistant who treated her was
concerned about her “inadequate physical activity.” These
general recommendations of physical activity do not contra-
dict Gerstner’s alleged limits from fibromyalgia. Gerstner
testified that she could not do “exercise,” but she did not ex-
plain what she considered exercise. Because the examining
physician and the physician’s assistant did not elaborate on
14                                                  No. 16-4007

the type, duration, or intensity of the physical activity they
would recommend, these medical sources may have had in
mind activity that was within Gerstner’s alleged limits—for
instance walking a few times a week for a few minutes each
time. Furthermore, the examining physician said her rec-
ommendation was only a “potential treatment,” probably
because she deferred treatment of Gerstner’s fibromyalgia to
other doctors who knew her limits. More importantly, the
record reflects that at the time of the hearing, Gerstner was
totally inactive, undermining any suggestion that she could
exercise beyond her alleged limits.
    Third, the ALJ questionably concluded that Gerstner ex-
aggerated her pain because she was able to function without
taking fibromyalgia drugs for six months when she was un-
insured. Gerstner and her attorney explained at the hearing
that the reason for the 6-month hiatus in her pain treatment
was her loss of health insurance. The ALJ, however, failed to
consider her explanation before inferring that this gap
somehow proved that her pain was not as severe as alleged.
See Garcia v. Colvin, 

, 761-62 (7th Cir. 2013); Craft
v. Astrue, 

, 679 (7th Cir. 2008).
    Lastly, the ALJ wrongly relied on psychiatric notes re-
flecting that Gerstner had searched for a factory job to con-
clude that she was able to work. The ALJ said this job search
“suggests that she did indeed retain the ability to perform
work-related activities,” and the Commissioner argued that
her job search undermines her complaints of disabling pain.
But the job search, on its own, is not evidence that she em-
bellished her pain, because a claimant who looks for work
after claiming a painful disability may have “a strong work
ethic or overly-optimistic outlook rather than an exaggerated
No. 16-4007                                                 15

condition.” See Ghiselli v. Colvin, 

, 778 (7th Cir.
2016); Hill v. Colvin, 

, 868 (7th Cir. 2015).
                               III.
   The ALJ failed to substantiate and explain his decision to
grant Dr. Callaghan’s opinion little weight, and the ALJ’s
adverse credibility determination regarding Gerstner’s com-
plaints of fibromyalgia pain is patently wrong. We therefore
vacate the judgment affirming the denial of benefits and re-
mand for further proceedings.