NOT FOR PUBLICATION FILED

UNITED STATES COURT OF APPEALS MAY 1 2018

MOLLY C. DWYER, CLERK

U.S. COURT OF APPEALS

FOR THE NINTH CIRCUIT

MICHAEL TROY KELLY, No. 16-17173

Plaintiff-Appellant, D.C. No. 2:15-cv-00455-MHB

v.

MEMORANDUM*

NANCY A. BERRYHILL, Acting Commissioner Social Security,

Defendant-Appellee.

Appeal from the United States District Court for the District of Arizona Michelle H. Burns, Magistrate Judge, Presiding

Argued and Submitted March 16, 2018 San Francisco, California

Before: PAEZ and IKUTA, Circuit Judges, and ADELMAN,** District Judge.

Michael Troy Kelly appeals the district court’s judgment affirming the

Commissioner of Social Security’s denial of his application for Disability

Insurance Benefits under Title II of the Social Security Act, 42 U.S.C. § 1381 et

* This disposition is not appropriate for publication and is not precedent except as provided by Ninth Circuit Rule 36-3. ** The Honorable Lynn S. Adelman, United States District Judge for the Eastern District of Wisconsin, sitting by designation. seq. We review de novo the district court’s decision affirming the denial of

benefits, and may set aside the decision of the administrative law judge (ALJ)

where that decision is based on legal error or where the findings of fact are not

supported by substantial evidence in the record taken as a whole. Tackett v. Apfel,

180 F.3d 1094, 1097 (9th Cir. 1999). We have jurisdiction pursuant to 28 U.S.C. §

1291, and we reverse and remand for further proceedings.

1. We first hold that the ALJ erred in rejecting the three opinions provided by

Kelly’s treating physician, Dr. Dale Ratcliffe. Where, as here, the treating doctor's

opinions are contradicted by another doctor, the ALJ may not reject the opinions

without providing “specific and legitimate reasons” supported by substantial

evidence in the record; the same is required for rejecting the treating doctor’s

“ultimate conclusions” as to disability. Lester v. Chater, 81 F.3d 821, 830 (9th Cir.

1995).

a. The ALJ rejected Dr. Ratcliffe’s May 2012 and April 2013 opinions by

stating they were “not supported by the objective medical evidence,” elaborating

that 1) Kelly’s “pain responded well to medications and other treatments”; 2)

“objective medical imaging did not indicate disabling impairment”; and 3) “[n]o

significant neurologic deficits are documented by any treating or consulting

doctor.” As to the ALJ’s first reason, “[w]hen viewed in its entirety,” Orn v.

Astrue, 495 F.3d 625, 634 (9th Cir. 2007), the record thoroughly contradicts the

2 ALJ’s assertion that Kelly’s pain “responded well to medications and other

treatment.” Where an ALJ’s reasoning is belied by the record, it cannot constitute

a “specific and legitimate” reason for rejecting the controverted opinion of a

treating physician. Id. at 634–35. Indeed, the record demonstrates the quite the

opposite: that Kelly—despite being treated with high doses of opioids, muscle

relaxers, epidural steroid injections, lumbar intrathecal opioid injections, and a

thoracic spinal cord stimulator, among other treatments—experienced pain that

was “minimally controlled;” that high doses of opioid medications at best provided

mild, occasionally moderate, improvement; and that his treating physician

considered his pain “refractory to treatment.”1

The ALJ’s assertion that “objective medical imaging did not indicate

disabling impairment” was error in several ways. First, it inappropriately

substituted the ALJ’s layperson opinion for that of Kelly’s treating physician

1 Although the dissent cites to several treatment records it views as supporting the ALJ’s assertion that Kelly’s pain was under control, the dissent errs in several ways. First, the dissent cites to evidence not relied upon by the ALJ, a practice we are forbidden from doing in reviewing an agency’s decision. See, e.g., Burrell v. Colvin, 775 F.3d 1133, 1138 (9th Cir. 2014). Second, the dissent—much like the ALJ—engages in impermissible cherry-picking of the record, rather than properly viewing the record in its entirety. See Ghanim v. Colvin, 763 F.3d 1154, 1164 (9th Cir. 2014); Trevizo v. Berryhill, 871 F.3d 664, 679 (9th Cir. 2017). Finally, the dissent’s citation to Dr. Chettri’s 2013 treatment note is somewhat curious, given that the same treatment note discusses how Kelly was suffering from “constant” pain in his right hip that “radiate[d] to the right leg” with “associated pain at the lumbar area”—pain severe enough that it caused him to have trouble walking.

3 without any support from other medical opinions in the record. See Day v.

Weinberger, 522 F.2d 1154, 1156 (9th Cir. 1975) (recognizing that an ALJ is “not

qualified as a medical expert”); Trevizo v. Berryhill, 871 F.3d 664, 676–77 (9th

Cir. 2017); cf. Taylor v. Comm'r of Soc. Sec. Admin., 659 F.3d 1228, 1235 (9th Cir.

2011) (holding that an ALJ may not substitute his layperson personal observations

of claimant for the opinions of claimant’s treating physicians). The ALJ

additionally “improperly cherry-picked” several normal findings from the MRI—

including “no acute lumbar spinal fracture” and “no moderate or severe canal

stenosis”—while ignoring a number of abnormal findings. See Ghanim v. Colvin,

763 F.3d 1154, 1164 (9th Cir. 2014). This selective cherry-picking of the MRI

results was also error because Dr. Ratcliffe’s opinion as to Kelly’s limitations was

not based on Kelly suffering from either “acute lumbar spinal fracture” or

“moderate or severe canal stenosis.” Where an ALJ’s reasoning is “not

responsive” to the basis of a physician’s opinion, it fails the “specific and

legitimate” standard. Orn, 495 F.3d at 634-35. Moreover, Dr. Ratcliffe explicitly

opined that his opinions as to Kelly’s limitations were based on objective clinical

and diagnostic findings.2 Here, such objective findings include not only the MRI

2 The dissent’s reliance on Dr. Ratcliffe’s statement that Kelly’s pain was “out of proportion to what is seen objectively on MRI, CT” and his accompanying notes that Kelly’s pain may be magnified by other issues (such as depression or opioid- induced hyperalgesia) is not persuasive. Not only did the ALJ not mention these notes—thus making them impermissible grounds on which to affirm the ALJ—but

4 results, but also the many positive straight-leg tests, positive Braggard maneuvers,

diminished lumbar range of motion, and other clinical findings. By isolating a few

normal findings from the MRI results while ignoring the numerous abnormal MRI

findings as well as all of the other abnormal clinical findings in the record, the ALJ

erred. See Ghanim, 763 F.3d at 1164; Holohan v. Massanari, 246 F.3d 1195, 1207

(9th Cir. 2001).

Further, the ALJ failed to discuss the more recent lumbar MRI in the record,

from May 2012. This MRI—which predated all of Dr. Ratcliffe’s opinions, was

ordered by Dr. Ratcliffe, and was undoubtedly taken into account by Dr. Ratcliffe

in his opinions—also showed abnormal findings.3 While the ALJ need not

the dissent also ignores that pain caused by a mix of psychological and physical impairments (if that is indeed the case with Kelly) is no less disabling under our case law and the applicable regulations. See Lester v. Chater, 81 F.3d 821, 829 (9th Cir. 1995) (“Lester’s condition, which the medical advisor referred to as ‘chronic pain syndrome,’ has both a physical and psychological component. Pain merges into and becomes a part of the mental and psychological responses that produce the functional impairments. The components are not neatly separable.”) (citation omitted). In fact, a claimant can still be disabled even if their pain stems entirely from psychological causes. See 20 C.F.R. pt. 404, subpt. P, app. 1, pt. A2, listing 12.07 (recognizing somatic symptom disorders as a basis for disability); see also McCollough v. Astrue, 247 F. App'x 925, 926 (9th Cir. 2007) (citing Carradine v. Barnhart, 360 F.3d 751, 754 (7th Cir. 2004)). 3 Specifically, it showed “mixed broad-based posterior disc-osteophyte complex/broad-based right paracentral disc protrusion and mild to moderate bilateral facet disease at the L4-L5 level” with “secondary abutment of the nondisplaced right descending nerve root,” as well as mixed broad-based posterior disc-osteophyte complex/shallow disc displacement with mild to moderate bilateral facet disease at L3-4; shallow broad-based posterior disc-osteophyte complex and moderate bilateral facet disease at L5-S1; moderate to mild disc desiccation and

5 “discuss every piece of evidence,” Howard ex rel. Wolff v. Barnhart, 341 F.3d

1006, 1012 (9th Cir. 2003), the ALJ is not free to ignore relevant, competent

evidence—such as a recent lumbar MRI for a claimant who suffers from lumbar

degenerative disc disease and lumbar spondylosis—that would lend support to a

claim of disability. See Gallant v. Heckler, 753 F.2d 1450, 1455–56 (9th Cir.

1984).

The ALJ accurately stated that “[n]o significant neurologic deficits are

documented by any treating or consulting doctor.” Yet the ALJ’s reasoning is

nonetheless legally erroneous under our precedent because it is “not responsive” to

the basis for Dr. Ratcliffe’s opinions. Orn, 495 F.3d at 634-35. At no point in Dr.

Ratcliffe’s opinions does he mention that “neurologic deficits” play any role in

Kelly’s work-related limitations, nor do any of his dozens of treatment notes

indicate as much.

b. The ALJ rejected Dr. Ratcliffe’s July 2012 opinion by stating that it

“mostly lists claimant’s symptoms, but provides no findings regarding the

claimant’s work abilities. This statement is assigned no weight as it does not

provide any analysis of the claimant’s work-related limitations.” It appears that the

ALJ was erroneously referring to a psychological assessment conducted by a

disc height loss at L3-L4, L4-5, and L5-S1; multilevel mild anterior endplate spurring; and mild left facet joint capsulitis at L1-L2, L2-L3, L3-4.

6 different doctor. We further note that while the July 2012 opinion goes to the

‘ultimate issue” of disability, it can still only be rejected “with specific and

legitimate reasons supported by substantial evidence in the record.” Reddick v.

Chater, 157 F.3d 715, 725 (9th Cir. 1998). On remand, the ALJ must correctly

weigh this opinion, considering it in the context of the “record as a whole,”

Tackett, 180 F.3d at 1097, including Dr. Ratcliffe’s other opinions and dozens of

treatment notes, along with any other relevant medical evidence.

c. The ALJ also erred in her treatment of Dr. Ratcliffe’s opinions by failing

to consider the factors outlined in 20 C.F.R. § 404.1527(c)(2)-(6) in assessing a

treating physician’s opinion, which we recently described as “reversible legal

error.”4 Trevizo, 871 F.3d at 676. On remand, the ALJ must consider these

factors, including the length of the treating relationship, the frequency of

examination, the nature and extent of the treatment relationship, and the

specialization of Dr. Ratcliffe. Id. at 676.

4 The dissent correctly states that the ALJ is not required “to make an express statement that she considered all the factors outlined in 20 C.F.R. § 404.1527(c).” Contrary to the dissent’s assertion, however, the ALJ’s decision here gives no indication that the factors were properly considered, other than a cursory acknowledgment of Dr. Ratcliffe as Kelly’s treating physician. Indeed, the fact that the ALJ appears to have mistaken an examining psychologist’s treatment note for one of Dr. Ratcliffe’s medical opinions strongly indicates that the ALJ did not give Dr. Ratcliffe’s opinions the careful consideration owed to them under the regulations.

7 2. We next hold that the ALJ erred in finding Kelly “not entirely credible.”

Having met the first step of our two-step subjective symptom inquiry, the ALJ

could reject Kelly’s testimony about the severity of his symptoms “only by

offering specific, clear and convincing reasons for doing so.” Garrison v. Colvin,

759 F.3d 995, 1014-15 (9th Cir. 2014). The ALJ provided five reasons for finding

Kelly “not entirely credible”; we address each in turn.

a. The ALJ’s first reason—“the objective medical evidence does not support

the degree of severity alleged by the claimant”—is clearly contradicted by our

precedent. We have consistently held that an ALJ “may not discredit the

claimant's testimony as to the severity of symptoms merely because they are

unsupported by objective medical evidence.” Reddick, 157 F.3d at 722.

b. The ALJ’s second reason— Kelly’s lifestyle is “more active” than Kelly

alleged in the hearing—fares little better. We have “repeatedly warned that ALJs

must be especially cautious in concluding that daily activities are inconsistent with

testimony about pain, because impairments that would unquestionably preclude

work and all the pressures of a workplace environment will often be consistent

with doing more than merely resting in bed all day.” Garrison, 759 F.3d at 1016.

The ALJ provided the following examples: 1) he helps care for his children, by

preparing meals, helping them with homework, watching them until his wife gets

home from work, and getting them ready for school; 2) he is independent in self-

8 care; 3) he “drives and rides in a car”; 4) he goes shopping in stores; 5) he prepares

meals occasionally during the day; and 6) he watches television and spends time on

the computer.

We first note that none of the activities the ALJ identified above are

genuinely incompatible with Kelly’s inability to sustain full-time competitive work

or with his alleged limitations, and thus cannot be used to attack Kelly’s

credibility. See Reddick, 157 F.3d at 722. Further, the ALJ’s examples are either

unsupported by the record or strip the record of critical nuance. For example, the

ALJ heavily emphasizes Kelly’s childcare responsibilities, while omitting that his

children were teenagers. The ALJ’s decision gives the impression that Kelly was

engaged in arduous caretaking of young children, but it is apparent from the record

that Kelly’s children actually engaged in substantial caretaking of Kelly—not the

other way around. Similarly, despite the ALJ’s assertion that Kelly “drives and

rides in a car,” Kelly’s function report and testimony both reflect that he no longer

drives due to his impairments unless he has to, but that generally his son or wife

drive him.5

c. The ALJ’s third reason also fails to meet the clear and convincing

standard. The ALJ determined that because “there is no evidence of a significant

5 One of Kelly’s wife’s statements checks “yes” in response to a question asking whether he can drive, but then explains that “[h]e does not drive very often. I often take him to his doctor appointments.”

9 deterioration in [Kelly’s] medical condition” from the time Kelly stopped working

in order to start his own staffing company, “a reasonable inference” is that he could

still perform the job he left. The ALJ’s rationale is erroneous for two reasons: (1)

it explicitly contradicts the ALJ’s own conclusion that Kelly was unable to perform

his past relevant work; and (2) it is roundly contradicted by the record, which

demonstrates that Kelly’s condition drastically deteriorated between 2005 (when

he left the job in question), his alleged date of onset (April 2010), and the time of

the hearing (May 2013).

d. The ALJ’s fourth reason—that Kelly’s conditions had responded to

treatment and medications and that he did not report significant side effects from

medications—is contradicted by the record, as explained in our analysis of the

ALJ’s treatment of Dr. Ratcliffe’s opinion. Neither of the record citations the ALJ

relies upon are persuasive. One is a pre-onset treatment note stating that he gained

“fairly good benefit” for several months following lumbar epidural steroid

injections, but nonetheless noting that even during this period of improvement, he

was not able to decrease his narcotic dose at all, and that after two months, the pain

had resumed. Significantly, no post-onset evidence demonstrates that steroid

injections helped Kelly—indeed, Dr. Page observed in October 2010 that the two

epidural steroid injections Dr. Page had performed “really did not help.” The other

10 was an April 2010 visit where Kelly stated that Vicodin “takes the edge off,”

which hardly signifies that Kelly considered his pain well-managed.

e. Finally, we conclude that the ALJ’s fifth reason—Kelly’s inconsistent

drug screens and medical marijuana usage—likewise fails to meet the clear and

convincing standard. We first note that most of the ALJ’s record citations are

irrelevant or incompetent, either because they predate Kelly’s onset date or

because they mischaracterize the record. For example, the ALJ stated that Kelly

was “evasive regarding his substance abuse problems during a consultative

examination with Dr. Rabara,” but the only “evasiveness” Dr. Rabara noted was

regarding where Kelly obtained his medical marijuana—not generalized

evasiveness regarding substance abuse. Cf. Thomas v. Barnhart, 278 F.3d 947,

959 (9th Cir. 2002). Likewise, the ALJ cited a December 2009 ER treatment

record showing that Kelly tested positive for amphetamines, THC, opiates, and

benzodiazepines; although the amphetamines and THC were illicit, this test

predated Kelly’s alleged onset by approximately six months. And while we have

no binding precedent on whether medical marijuana use may be a legitimate

ground for an adverse credibility finding, we have previously expressed skepticism

at the idea. See Buchholz v. Barnhart, 56 F. App'x 773, 776 (9th Cir. 2003). Here,

we conclude that the ALJ’s reliance on Kelly’s medical marijuana use to discredit

him was improper and does not meet the clear and convincing standard.

11 3. We next conclude that the ALJ erred in rejecting the third-party function

reports submitted by Kelly’s wife. “To reject third-party reports of a claimant’s

impairments” or other lay witness evidence, “an ALJ need only ‘give reasons that

are germane to each witness.’” Revels v. Berryhill, 874 F.3d 648, 655 (9th Cir.

2017) (quoting Molina v. Astrue, 674 F.3d 1104, 1114 (9th Cir. 2012)). The ALJ

reasoned that Kelly’s wife statements were “not persuasive” because (1) they were

“not supported by the objective medical evidence” and (2) Kelly’s wife “[wa]s not

qualified to make a diagnosis regarding [Kelly’s] impairments.”

The ALJ’s first reason for rejecting the third-party function reports was

legally erroneous under our precedent. See Bruce v. Astrue, 557 F.3d 1113, 1116

(9th Cir. 2009) (“Nor under our law could the ALJ discredit her lay testimony as

not supported by medical evidence in the record.”). Moreover, despite the ALJ’s

assertion, the reports were supported by objective medical evidence. An ALJ does

not provide a “germane” reason for rejecting lay witness evidence where the ALJ

erroneously states that such evidence is inconsistent with other evidence in the

record. See Revels, 874 F.3d at 668.

The ALJ’s second reason for rejecting the third-party function reports—that

Kelly’s wife was not qualified to make a diagnosis—was erroneous because her

reports did not constitute a diagnosis. While lay witness testimony is incompetent

where it purports to make a diagnosis, “‘lay witness testimony as to a claimant’s

12 symptoms or how an impairment affects ability to work is competent evidence and

therefore cannot be disregarded without comment.’” Tobeler v. Colvin, 749 F.3d

830, 833–34 (9th Cir. 2014) (quoting Nguyen v. Chater, 100 F.3d 1462, 1467 (9th

Cir. 1996)). Kelly’s wife’s reports described at length how “how [Kelly’s]

impairment affects [his] ability to work.” Id. Such reports were “plainly

competent,” id. at 834, and thus could not be disregarded in the absence of a

“germane” reason—which the ALJ did not provide. See id.

4. We decline to exercise our discretion to remand for an award of benefits.

We instead remand for further proceedings. See Treichler v. Comm’r of Soc. Sec.

Admin., 775 F.3d 1090, 1104-05 (9th Cir. 2014).

REVERSED and REMANDED for further proceedings consistent with this

disposition.

13

FILED

Kelly v. Berryhill, No. 16-17173

MAY 01 2018

Ikuta, J., dissenting

MOLLY C. DWYER, CLERK

U.S. COURT OF APPEALS

The majority combs through the record to find evidence supporting its

conclusion that the ALJ’s reasons for rejecting Kelly’s disability claim are not

sufficient. This is the wrong approach to reviewing an agency decision. Rather,

we must uphold the ALJ’s reasoning so long as it is supported by a fair review of

the record, see Batson v. Comm’r of Soc. Sec. Admin., 359 F.3d 1190, 1193 (9th

Cir. 2004), even if we would have reached a different conclusion.1 Where

“evidence is susceptible to more than one rational interpretation,” we are obligated

to uphold the ALJ’s decision. Orn v. Astrue, 495 F.3d 625, 630 (9th Cir. 2007)

1 The majority’s assertion that the court may not consider the entire record to determine whether the ALJ’s findings are supported by substantial evidence, Maj. at 3, n.1, is not only contrary to the applicable statute, see 42 U.S.C. § 405(g), but contrary to our case law and the majority’s own approach. As we have explained, we must consider whether the ALJ’s findings are “supported by substantial evidence in the record as a whole,” Molina v. Astrue, 674 F.3d 1104, 1110 (9th Cir. 2012), and in doing so we “must consider the record as a whole, weighing both the evidence that supports and the evidence that detracts from the Secretary’s conclusion.” Winans v. Bowen, 853 F.2d 643, 644 (9th Cir. 1987); see also Smolen v. Chater, 80 F.3d 1273, 1279 (9th Cir. 1996) (“We must give the facts a full review and must independently determine whether the Commissioner’s findings are supported by substantial evidence.”). We “may not reweigh the evidence, substitute [our] own judgment for the Secretary’s, or give vent to feelings of compassion,” Winans, 853 F.2d at 644–45 (quoting Bowman v. Heckler, 706 F.2d 564, 566 (5th Cir. 1983)) (alteration in original), and may not identify “other alleged inconsistencies” in the record not identified by the ALJ, Burrell v. Colvin, 775 F.3d 1133, 1138 (9th Cir. 2014) (emphasis in original). But we are certainly not limited to reviewing only the portions of the record expressly identified by the ALJ, and indeed, the majority does not do so. (citation omitted).

There are only two issues of significance. First, the majority claims that the

ALJ erred in rejecting Dr. Ratcliffe’s opinions that Kelly suffers from moderately

severe pain that prevents him from work-related activity. But there is ample

evidence in the record showing that the ALJ’s reasons for reaching this conclusion

were specific and legitimate. Among other evidence noted by the ALJ, Dr.

Briggs’s 2012 report stated that Kelly’s pain was “relieved by medications,” and

Dr. Chettri’s 2013 report stated that Kelly’s back pain was “under control.”

Indeed, even Dr. Ratcliffe noted that Kelly’s pain was “out of proportion to what is

seen objectively on MRI, CT,” suggesting that the pain was magnified by other

issues, such as “opioid-induced hyperalgesia.”2 Moreover, the ALJ was well

within her authority to reject Dr. Ratcliffe’s two sentence statement (scrawled on a

prescription note pad) that Kelly “would benefit from disability in that a return to

work at this time is not possible,” given that the ultimate conclusion regarding

disability is solely the province of the ALJ. See 20 C.F.R. § 404.1527(d)(1) (“A

2 The majority rejects Dr. Ratcliffe’s express, written opinion that Kelly’s pain was “out of proportion to what is seen objectively on MRI, CT,” based on its speculation that Kelly’s pain was caused by a mix of psychological and physical impairments. Maj. at 4 n.2. In relying on a conjecture that is not addressed by the ALJ, and for which the majority identifies no support in the record, the majority substitutes its own judgment for the ALJ’s, and thus violates our limited role in reviewing the ALJ’s decision. Cf. Winans, 853 F.2d at 644–45.

2 statement by a medical source that you are ‘disabled’ or ‘unable to work’ does not

mean that we will determine that you are disabled.”); see also McLeod v. Astrue,

640 F.3d 881, 885 (9th Cir. 2011). The majority may disagree with the ALJ’s

conclusion, but a “rational interpretation” of the record supports her reasoning.

Bayliss v. Barnhart, 427 F.3d 1211, 1214 n.1 (9th Cir. 2005).3

The second basis for the majority’s opinion is its disagreement with the

ALJ’s adverse credibility determination. We are limited to determining whether

the ALJ gave “specific, clear and convincing reasons,” Garrison v. Colvin, 759

F.3d 995, 1015 (9th Cir. 2014) (citation omitted), for concluding that Kelly’s

testimony about the severity of his symptoms was “not entirely credible.” As

noted above, Dr. Briggs’s and Dr. Chettri’s reports supported the ALJ’s conclusion

that Kelly’s symptoms improved with treatment. Moreover, the record contained

multiple findings that Kelly had a normal gait and no motor defect, contrary to his

claims about his impairments. There is also ample evidence that Kelly’s mental

state was not as bad as he claimed, including findings of “normal mood and affect”

3 As the majority concedes, neither the regulations nor our case law require the ALJ to make an express statement that she considered all the factors outlined in 20 C.F.R. § 404.1527(c). Here the ALJ’s opinion indicates she considered the relevant factors; she mentions Kelly’s relationship with his physicians and the frequency of treatment, and provides an in-depth discussion of the supportability and consistency of the various medical opinions.

3 in June 2011, June 2012, and February 2013, and a report from Dr. Rabara that

Kelly’s claims of depression were “somewhat questionable” in light of recent

records that “consistently describe[d] his mood and affect as ‘normal.’”

The ALJ’s determination that Kelly “leads a more active lifestyle than what

he alleged to during the hearing” was also well supported by the record. For

instance, Kelly’s testimony at the hearing that he did not do “anything” around the

house, was contradicted by his written statement that he “raise[s] his kids, feed,

homework, school,” and that he takes care of his pets, including feeding and

watering them. His testimony that he does not drive was directly contradicted by

his written statement that he drove when he had to. His testimony that he needed

assistance in dressing and had to sit down in the shower was contradicted by his

statement to Dr. Rabara that he needed no assistance with personal hygiene,

grooming, or dressing, and his written statement that he had “no problem” with

personal care, including dressing and bathing. To the extent Kelly’s reports

“contradict claims of a totally debilitating impairment,” they support the ALJ’s

reasons for her adverse credibility determination. Molina v. Astrue, 674 F.3d 1104,

1113 (9th Cir. 2012).

Finally, the record is replete with evidence that Kelly engaged in evasive

drug-seeking behavior. He was discharged from Dr. Page’s pain treatment

4 program because he was taking extra hydrocodone prescribed by another doctor.

He reported overuse of medications and sought early refills of drugs while at the

Desert Pain and Rehabilitation clinic. A May 2011 drug screen showed that Kelly

failed to take some prescribed medication and instead took several drugs that had

not been prescribed, including illegal ones. Kelly told Dr. Rabara that he used

marijuana he obtained from medical dispensaries in Arizona, but became evasive

and changed his story when he learned dispensaries were not yet legal.

By noting the conflict between Kelly’s claims of disability compared to his

own written statements and objective medical evidence, along with evidence of

evasive drug seeking behavior, the ALJ gave “specific, clear and convincing

reasons,” Garrison, 759 F.3d at 1015, for rejecting Kelly’s testimony. Because

Kelly’s wife echoed Kelly’s testimony, the ALJ can rely on the same reasons to

reject her testimony as well. Valentine v. Comm’r Soc. Sec. Admin., 574 F.3d 685,

694 (9th Cir. 2009); see also Molina, 674 F.3d at 1114 (“[I]f the ALJ gives

germane reasons for rejecting testimony by one witness, the ALJ need only point to

those reasons when rejecting similar testimony by a different witness.”).

We must give proper deference to the ALJ and reverse her decision only if it

was not supported by substantial evidence or if she committed a legal error.

Batson, 359 F.3d at 1198. Because the record amply supports both her reasoning

5 and conclusion, I would affirm the ALJ’s decision. Therefore, I dissent.

6