FILED NOT FOR PUBLICATION APR 9 2021 UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS
FOR THE NINTH CIRCUIT
MARY M. KHAN, No. 19-17473
Plaintiff-Appellant, D.C. No. 1:19-cv-00005-ACK-WRP v.
ANDREW M. SAUL, Commissioner of MEMORANDUM* Social Security,
Defendant-Appellee.
Appeal from the United States District Court for the District of Hawaii Alan C. Kay, District Judge, Presiding
Submitted February 4, 2021** Honolulu, Hawaii
Before: CLIFTON, R. NELSON, and COLLINS, Circuit Judges. Dissent by Judge COLLINS
Plaintiff Mary M. Khan appeals from the district court’s order affirming the
Commissioner of Social Security’s final decision denying Khan’s application for
* This disposition is not appropriate for publication and is not precedent except as provided by Ninth Circuit Rule 36-3. ** The panel unanimously concludes this case is suitable for decision without oral argument. See Fed. R. App. P. 34(a)(2). benefits under Titles II and XVI of the Social Security Act. Reviewing the district
court’s order de novo, we will reverse only if the Administrative Law Judge’s
(“ALJ”) decision “was not supported by substantial evidence in the record as a
whole or if the ALJ applied the wrong legal standard.” Molina v. Astrue, 674 F.3d
1104, 1110 (9th Cir. 2012) (citations omitted). Where “the evidence is susceptible
to more than one rational interpretation,” this court “must uphold the ALJ’s
findings if they are supported by inferences reasonably drawn from the record.”
Id. at 1111 (citation omitted).
Khan challenges the weighing of three physicians’ opinions. First, Khan
objects to the weighing of treating physician Dr. Myers’ opinions. Due to
contradicting opinions in the record, the ALJ could only discount Dr. Myers’
opinion “by giving specific and legitimate reasons” supported by substantial
evidence. See Ford v. Saul, 950 F.3d 1141, 1154 (9th Cir. 2020) (citation
omitted).1 The ALJ set out a “detailed and thorough summary of the facts and
conflicting clinical evidence, stating [his] interpretation thereof, and making
findings.” Tommasetti v. Astrue, 533 F.3d 1035, 1041 (9th Cir. 2008) (quoting
1 Khan argues our more exacting standard should apply to Dr. Myers’ August 2016 opinion. But this opinion was offered as a rebuttal to Dr. Jonas’s opinion and was; therefore, inherently controverted. Further, even if the lower standard applied, given the conclusory nature of this opinion the ALJ did not err in discounting it. See Ford, 950 F.3d at 1154–55. 2 Magallanes v. Bowen, 881 F.2d 747, 751 (9th Cir. 1989)). Khan also concedes the
ALJ gave specific reasons but argues they were not legitimate. The ALJ properly
discounted Dr. Myers’ opinions as they were conclusory or cursory. See Ford, 950
F.3d at 1154–55 (“The ALJ need not accept the opinion of any physician,
including a treating physician, if that opinion is brief, conclusory, and inadequately
supported by clinical findings.” (quoting Thomas v. Barnhart, 278 F.3d 947, 957
(9th Cir. 2002)); see also 20 C.F.R. § 404.1527(c)(3).
Second, Khan argues that the ALJ improperly discounted the opinion of Dr.
Winger by substituting the ALJ’s medical judgment for that of the doctor.
However, the ALJ merely noted an inconsistency between (1) the test results Dr.
Wingert outlines in his report and his own conclusions which he stated were based,
in part, on the claimant’s self-reports; and (2) his conclusion and claimant’s
conservative treatment regimen. An ALJ may reject an opinion if it is based “to a
large extent” on the claimant’s self-reports. Tommasetti, 533 F.3d at 1041 (citation
omitted). An ALJ may also consider conservative treatment when evaluating a
medical expert’s conclusions. See Rollins v. Massanari, 261 F.3d 853, 856 (9th
Cir. 2001). As such, the ALJ provided specific and legitimate reasons supported
by substantial evidence for discounting Dr. Wingert’s conclusions. See Ford, 950
F.3d at 1154–55.
3 Third, Khan argues that the forms Dr. Kiyota completed for the State of
Hawaii Department of Human Services should not have been discounted by the
ALJ. The ALJ reasonably assigned less weight to Dr. Kiyota’s check-box forms
because of their lack of explanation and because they contradicted her longer-form
evaluation. Id. at 1155; see also Batson v. Comm’r of Soc. Sec., 359 F.3d 1190,
1195 (9th Cir. 2004) (an ALJ must resolve conflicts between medical opinions and
can look to the level of explanation for the various opinions to resolve conflicts).
Khan also argues that the ALJ improperly rejected her symptom testimony
regarding her inattention and need for hour-long breaks throughout the day. The
ALJ specified that while he found Khan’s symptoms to justify the limitations in the
residual functional capacity, they did not justify any further limitations or a finding
of disability. He gave four reasons for rejecting the extent of symptoms to which
Khan testified. Together they were “specific, clear and convincing reasons”
supported by substantial evidence to discount “the claimant’s testimony about the
severity of her symptoms.” Tommasetti, 533 F.3d at 1039 (quoting Smolen v.
Chater, 80 F.3d 1273, 1281 (9th Cir. 1996)). First, the ALJ noted Khan’s
conservative treatment. This reason alone can be a sufficient reason supported by
substantial evidence to discount claimant’s symptom testimony. See id.; see also
Parra v. Astrue, 481 F.3d 742, 750–51 (9th Cir. 2007) (“conservative treatment” is
4 sufficient to discount claimant’s testimony regarding severity of impairment);
Meanel v. Apfel, 172 F.3d 1111, 1114 (9th Cir. 1999) (holding subjective pain
complaints properly discredited where claimant received “minimal” and
“conservative” treatment). Second, the ALJ noted Khan’s focused demeanor at the
hearing as undermining the symptom testimony. Third, the ALJ considered
Khan’s own statements regarding her activities of daily life which included caring
for minor children and maintaining a household. See Rollins, 261 F.3d at 857
(holding taking care of children and maintaining a household to be a clear and
convincing reason to discredit symptom testimony). Fourth, the ALJ looked to the
objective medical evidence. Molina, 674 F.3d at 1113 (comparing subjective
symptom results to findings in clinical evaluations in upholding the ALJ’s adverse
credibility determination).
The ALJ’s decision was supported by substantial evidence.
AFFIRMED.
5 FILED Khan v. Saul, No. 19-17473 APR 9 2021
COLLINS, Circuit Judge, dissenting: MOLLY C. DWYER, CLERK U.S. COURT OF APPEALS
This is a textbook case of cherry-picking, in which the Administrative Law
Judge (“ALJ”) unevenly applied a sharply critical lens to the overwhelming
evidence supporting disability claimant Mary Khan, while ignoring that those same
criticisms also applied to the small subset of evidence on which the ALJ selectively
relied. Specifically, the administrative record contains at least 13 reports from the
relevant time period from either Khan’s treating psychologist or from one of two
examining psychologists. Of these 13 reports, 12 of them concluded that Khan’s
severe mental and emotional difficulties would make it difficult for her to work on
a full-time basis of eight hours per day, five days per week. The ALJ discounted
these 12 reports largely on the ground that they were supposedly lacking in
sufficient clinical detail. Instead, the ALJ chose to give “significant” weight to the
single report by an examining psychologist that reached a different conclusion
from the other 12, as well as to a conclusory opinion of a non-examining
psychiatrist. The ALJ did so even though these two opinions were less detailed
than some of those he discounted as lacking sufficient detail. The Appeals Council
then upheld the ALJ’s decision and uncritically adopted his relevant findings.
Because this kind of markedly uneven evaluation of the record violates applicable
administrative-law standards, I would vacate the ALJ’s denial of benefits and remand. I therefore respectfully dissent.
1. “Generally, a treating physician’s opinion carries more weight than an
examining physician’s, and an examining physician’s opinion carries more weight
than a reviewing physician’s.” 1 Holohan v. Massanari, 246 F.3d 1195, 1202 (9th
Cir. 2001); 20 C.F.R. § 404.1527(c)(1)–(2). More weight is also given “to
opinions that are explained than to those that are not.” Holohan, 246 F.3d at 1202;
20 C.F.R. § 404.1527(c)(3). “[I]f a treating physician’s medical opinion is
supported by medically acceptable diagnostic techniques and is not inconsistent
with other substantial evidence in the record, the treating physician’s opinion is
given controlling weight.” Holohan, 246 F.3d at 1202; 20 C.F.R.
§ 404.1527(c)(2). But “‘if the treating doctor’s opinion is contradicted by another
doctor’”—even a non-examining doctor—“the ALJ may discount the treating
physician’s opinion by giving ‘specific and legitimate reasons’ that are supported
by substantial evidence in the record.” Ford v. Saul, 950 F.3d 1141, 1154 (9th Cir.
2020) (citation omitted). The ALJ likewise must give specific and legitimate
reasons for discounting the opinion of an examining physician or psychologist. Id.
1 These principles apply equally to psychologists. See 20 C.F.R. § 404.1527(a)(1) (defining “[m]edical opinions” as “statements from acceptable medical sources that reflect judgments about the nature and severity of your impairment(s), including your symptoms, diagnosis and prognosis, what you can still do despite impairment(s), and your physical or mental restrictions”); id. § 404.1502(a)(2) (including “[l]icensed psychologist” among “[a]cceptable medical source[s]”).
2 at 1155. The ALJ failed to provide such reasons for rejecting the conclusions of
Khan’s treating psychologist, as well as those of two examining psychologists.
Instead, the ALJ selectively applied his criticisms in a way that produced an
internally inconsistent and slanted analysis of the medical evidence.
a. Between May 2013 and August 2016, Khan’s treating psychologist, Dr.
Mary Myers, prepared four reports indicating that Khan’s severe mental and
emotional problems prevented her from working full time, all four of which the
ALJ afforded less than full weight. The ALJ stated that he only afforded them
“some weight” because Dr. Myers assertedly did “not substantiate many of her
conclusions” and did “not provide detailed clinical findings to support her
conclusion[s].” This is a reasonable criticism to apply to two of Dr. Myers’s
reports, because her May 2013 letter included few details, and her June 2016 report
was primarily a check-the-box form. But Dr. Myers’s two other reports were
among the most detailed in the record, and they explained how and why Khan’s
mental conditions impeded her ability to handle a full-time work schedule.
Specifically, Dr. Myers’s August 2014 report noted that Khan’s complaints
and symptoms of “[a]nxiety, fear, worry, hypervigilance, recurrent intrusive
recollections of past traumas/deaths, [and] grief” resulted from her history of
“stress of being a single [mother] to 8 kids”; the deaths of her first husband, a
child, two sisters, and both parents; domestic violence by her second husband; and
3 a custody battle to reunite with her children after involvement with Child
Protective Services. Dr. Myers stated that, because Khan “continues to struggle
[with] past traumas” and “is most likely at or near her limit for coping [and]
stressors of any kind,” Khan “may not be able to” understand and remember
simple work instructions, maintain regular job attendance, and persist at simple and
repetitive work tasks under ordinary supervision “consistently on [a] daily basis for
employment.” Dr. Myers therefore concluded that Khan was “probably not”
capable of coping with a low-demand, entry-level job “on a full time consistent
basis in an employment situation.” The ALJ, however, disregarded these
statements and instead focused on the irrelevant fact that the August 2014 report
also mentioned several positive traits displayed by Khan, such as the fact that she
has a “pleasant” disposition and takes care of her children. See Ghanim v. Colvin,
763 F.3d 1154, 1161–62 (9th Cir. 2014) (positive observations “must be ‘read in
context of the overall diagnostic picture’ the provider draws” (citation omitted)).
In her August 2016 report two years later, Dr. Myers identified specific
limitations in Khan’s mental functioning, including “difficulty sustaining her
attention to complete tasks”; a “very limited” threshold for stress, which
“compromis[es] her ability to learn new tasks, remember what she has learned, and
make decisions”; a “lack of self-confidence”; “depressive symptoms”; and
“dissociative episodes in which she is not present in the sense of being aware of
4 what’s going on in her surroundings.” These limitations led Dr. Myers to the
conclusion that, as she found in August 2014, Khan could not consistently
understand and remember instructions, maintain regular attendance, or persist in
simple and repetitive tasks—i.e., Dr. Myers did “not expect that Mary could
maintain an 8 hour work day for 5 days a week.”
In view of these reports, the ALJ’s assertion that Dr. Myers did not
substantiate her conclusions and did not provide “detailed clinical findings” is not
supported by substantial evidence. Cf. Ford, 950 F.3d at 1154–55 (“‘The ALJ
need not accept the opinion of any physician, including a treating physician, if that
opinion is brief, conclusory, and inadequately supported by clinical findings.’”
(citation omitted)).
b. But the ALJ’s error goes beyond that. Even if the ALJ’s criticism had
some validity, the ALJ cannot legitimately discount one report for its supposedly
conclusory nature while crediting a contradictory report that suffers from the same
deficiencies. See, e.g., Lester v. Chater, 81 F.3d 821, 832 (9th Cir. 1995) (ALJ
may not reject examining psychologist’s opinion on the ground that it was based
on “limited observation” of the claimant when the competing opinion accepted by
the ALJ was based on no examination of the claimant). Strikingly, the ALJ here
gave “significant weight” to examining psychologist Dr. Heide Kiyota’s August
14, 2014 report—the only report by a treating or examining practitioner that found
5 Khan capable of full-time work, and the only one of Dr. Kiyota’s eight reports that
came to that conclusion—even though its level of detail was not materially
different from Dr. Myers’s August 2014 and August 2016 reports.
In her August 2014 report, Dr. Kiyota described Khan as anxious and
depressed but friendly, exhibiting “good hygiene,” able to relate to others, and
capable of doing chores and caring for her children. She also indicated that Khan’s
cognitive status was within normal limits and rated her global assessment of
function (“GAF”) at 55 out of 100. In affording the report “significant weight,”
the ALJ reasoned that the report’s “conclusions are supported by [Dr. Kiyota’s]
own findings, which are based on her observation of [Khan] on several occasions”
and that the report was “detailed and well-reasoned.” This conclusion is
indefensible. The descriptions in Dr. Kiyota’s August 2014 are less detailed than
Dr. Myers’s August 2014 report, which was filled out on an identical form. 2
Moreover, Dr. Kiyota had observed Khan on far fewer occasions than Dr. Myers,
who by then had observed her once or twice per month for nearly two years. In
other words, the reasons Dr. Kiyota’s report was given “significant weight” applied
with even greater force to Dr. Myers’s August 2014 report, but the ALJ failed to
grant the latter the same weight. This kind of obviously inconsistent and uneven
2 The only notable detail that Dr. Kiyota’s report added was that it gave Khan a GAF rating, whereas Dr. Myers’s report did not.
6 evaluation of the evidence does not constitute a “specific and legitimate reason”
for rejecting the treating psychologist’s opinion. Ford, 950 F.3d at 1154.3
The same uneven application of criticisms is reflected in the ALJ’s decision
to give “significant” weight to the testimony of Dr. Alfred Jonas, a non-examining
physician, who opined in July 2016 that Khan had no functional limitations. This
testimony, however, was based on Dr. Jonas’s review of incomplete records, from
which at least seven of the most recent reports by Khan’s treating and examining
psychologists were missing. The missing records included Dr. Myers’s June 2016
report, which opined that Khan had “marked” limitations on her ability to accept
instructions and respond appropriately to criticism, deal with normal work stress,
and carry out detailed instructions. And, of course, Dr. Myers’s August 2016
report—which she prepared in response to Dr. Jonas’s testimony—did not yet
exist. Dr. Jonas was also missing six of Dr. Kiyota’s reports, all of which opined
that Khan was incapable of working full time, and the last two of which—the
September 2015 and March 2016 reports—concluded that Khan had regressed to
the point of being unable to work at all. After Dr. Jonas was given these additional
records, he confirmed in September 2016—in a report that was even more
3 Further, in light of the three prior and four subsequent opinions by Dr. Kiyota concluding that Khan could not work full time, “it [was] error for [the] ALJ to pick out [the single] isolated instance[] of improvement over a period of . . . years and to treat [it] as a basis for concluding [Khan] is capable of working.” See Garrison v. Colvin, 759 F.3d 995, 1017 (9th Cir. 2014).
7 conclusory than some of the contradictory reports the ALJ had discounted for their
supposedly insufficient detail—that his opinion had not changed. Thus, just as
with Dr. Kiyota’s August 2014 report, the ALJ did not require from Dr. Jonas the
same level of supporting explanation and detail that the ALJ demanded from any
report that disagreed with the ALJ’s conclusion. An ALJ cannot justify his or her
preferred conclusion by selectively discounting only the “evidence in the record
that suggests an opposite result.” See Gallant v. Heckler, 753 F.2d 1450, 1456 (9th
Cir. 1984).
c. The ALJ’s impermissible cherry-picking is further illustrated by his effort
to use parts of Dr. Myers’s August 2014 report to criticize her later August 2016
report. The ALJ stated that, in contrast to the earlier report, Dr. Myers’s August
2016 report gave fewer “detailed findings to support her conclusions.” As
examples of the sort of “detailed findings” he was looking for, the ALJ pointed to
the August 2014 report’s statements that Khan had certain positive traits: she
“performs all [activities of daily living], provides for needs of her 8 children as a
single parent, [and] manages her own funds”; has an excellent ability to relate,
engages with others, and is pleasant; displays an “appropriate range of emotions
given topics discussed”; and has a well oriented cognitive status and an average
estimated IQ functioning. The ALJ’s suggestion that the August 2014 report was
more detailed is unfathomable, because the August 2016 report undeniably
8 contained more detail substantiating Dr. Myers’s conclusions than did the earlier
report. Indeed, Dr. Myers created the August 2016 report in response to a request
to expound upon the conclusions contained in the August 2014 report. In fact, the
only discernable difference between the details contained in these two reports—
both of which ultimately concluded that Khan could not sustain full-time
employment—is that the August 2014 report also describes areas of ability rather
than just areas of disability. This point further confirms the inescapable conclusion
that, under the ALJ’s ruling, a report had an adequate level of “detail” if it
conformed to the ALJ’s preferred conclusion that Khan was capable of full-time
work, and it had insufficient detail if it did not.
In addition to failing to assess the evidence in a consistent and even-handed
manner, the ALJ also failed to recognize that the cited descriptions of Khan’s
positive attributes are non-responsive to the question of whether Khan can work
full time. For example, the ALJ points to Dr. Myers’s description of Khan’s ability
to get along with others, provide for her children, and perform simple repetitive
tasks, and the ALJ concluded that these points are inconsistent with the conclusion
that Khan could not handle full-time work. But the question was never whether
Khan was able to work for short periods of time; it was whether she could
consistently work on a full-time basis, and almost every report of a treating or
examining psychologist concluded that she could not. See Smolen v. Chater, 80
9 F.3d 1273, 1284 n.7 (9th Cir. 1996) (“The Social Security Act does not require that
claimants be utterly incapacitated to be eligible for benefits, and many home
activities may not be easily transferable to a work environment where it might be
impossible to rest periodically or take medication.”). In other words, the “ALJ
improperly cherry-picked some of [the practitioners’] characterizations of [Khan’s]
rapport and demeanor instead of considering these factors in the context of [their]
diagnoses and observations of impairment.” Ghanim, 763 F.3d at 1164; see also
Orn v. Astrue, 495 F.3d 625, 635 (9th Cir. 2007) (a reason for rejecting an opinion
is not “legitimate” if it is “not responsive”); Holohan, 246 F.3d at 1205 (criticizing
an ALJ in part for being “selective in his reliance” on a physician’s notes, and
stating that a physician’s “statements must be read in context of the overall
diagnostic picture he [or she] draws.”).
2. Similar problems affected the ALJ’s selective crediting of the sole report
of Dr. John Wingert, who examined Khan in January 2015. To the extent that Dr.
Wingert opined that Khan could “maintain adequate concentration and attention in
a structured setting” and that she “demonstrated cognitive ability to make simple
work related decisions,” the ALJ concluded that Dr. Wingert’s opinions were
substantiated by his observations during the examination. But the ALJ then
discredited Dr. Wingert’s opinion that Khan’s “depressive symptoms might limit
her ability to complete a normal workday,” which the ALJ found was “mainly
10 derived from [Khan’s] subjective reports,” as opposed to her performance on Dr.
Wingert’s intelligence, cognitive functioning, and personality examinations. The
ALJ’s analysis overlooks the fact that, in addition to Khan’s self-reports, Dr.
Wingert also relied on his direct observations of Khan during their examination.
For example, Dr. Wingert stated that “[f]rom the onset of the session, [Khan]
appeared to be close to tears at times during the interview, and there was a close-
to-the-surface sense of irritability evident”; that Khan “maintained a serious and
intense composure and one does get the impression of an individual experiencing a
high level of adjustment difficulties”; that, based on the “Wechsler Adult
Intelligence Scale – IV,” Khan’s “[w]eaknesses were evident on the subtests
making up the Processing Speed Index, which scored within the Borderline range,
and this involves the speed of mental and graphal motor processing”; and that “one
does get the impression of an individual who is functioning at a somewhat reduced
level of psychological efficiency.” These comments all derived from observation,
not self-report. Moreover, we have repeatedly emphasized that psychological
evaluations, by their nature, “will always depend in part on the patient’s self-
report.” Buck v. Berryhill, 869 F.3d 1040, 1049 (9th Cir. 2017). But the most
troubling aspect is that the ALJ found Dr. Wingert’s descriptions of Khan’s
positive attributes to be backed up by his clinical observations, but then ignored Dr.
Wingert’s clinical observations when they emphasized the limitations associated
11 with Khan’s severe mental difficulties. Once again, the conclusion is inescapable
that the ALJ inconsistently evaluated the medical opinions in this case based on
whether he agreed with their ultimate conclusions.4
3. The ALJ’s errors in evaluating the medical evidence necessarily vitiate
the ALJ’s evaluation of Khan’s testimony describing her symptoms and
limitations, because a proper assessment of that evidence would reasonably be
expected to affect the overall evaluation of that testimony. For example, whether
Khan’s treatment was “conservative” might be viewed differently depending upon
the assessment of the underlying conditions. I would require that Khan’s
testimony be re-evaluated on remand after a proper weighing of the medical
evidence. See Hiler v. Astrue, 687 F.3d 1208, 1212 (9th Cir. 2012). I would note,
however, that the ALJ’s reliance upon his lay impression of Khan’s psychological
condition during her testimony is also questionable. Cf. Montijo v. Secretary of
Health & Human Servs., 729 F.2d 599, 602 (9th Cir. 1984) (“The administrative
4 Little, if any, weight can be given to the ALJ’s discussion of the reports of two non-examining “State agency psychological consultants.” Although the ALJ acknowledged that these psychologists “were unable to consider the evidence added after their review,” the ALJ nonetheless proceeded to conclude that “their assessment remains supported in some respects.” Given that the ALJ did not specify which aspects of their assessment were supported and which ones were not, their evaluations provide no basis for rejecting the assessments of the treating and examining psychologists. The ALJ merely mentioned their statement that Khan could perform particular tasks, but that limited point did not speak to the grounds for the other psychologists’ assessment that Khan’s psychological problems would prevent her from working full time.
12 law judge’s observation of the claimant at the hearing, at least in cases involving
alleged psychological disability, does not provide a sufficient reason to reject the
otherwise uncontroverted medical evidence.”).
* * *
I recognize that the findings of ALJs are subject to a highly deferential
standard of review. But that review is applied against a backdrop of
administrative-law principles that require the ALJ to provide an explanation that is
internally consistent and that fairly evaluates the competing items of evidence in an
even-handed way. See Garrison, 759 F.3d at 1017 n.23 (an ALJ is “not permitted
to ‘cherry-pick’ from . . . mixed results to support a denial of benefits” (citation
omitted)); Gallant, 753 F.2d at 1456 (an ALJ “cannot reach a conclusion first, and
then attempt to justify it by ignoring competent evidence in the record that suggests
an opposite result”). The ALJ failed to do that here. The substantial deference we
are to afford does not require us to rubberstamp the sort of highly selective and
uneven consideration of the opinion evidence reflected in the decision in this case.
I respectfully dissent.