WILLIAMS, Circuit Judge.
Karen Murphy claims that as a result of a stroke she has impairments so severe that she has not been able to return to her job as a secretary or work in another capacity. She applied for Disability Insurance Benefits (“DIB”), but the Commissioner of the Social Security Administration (“SSA”) denied her application finding that she was not disabled. The Administrative Law Judge (“ALJ”) also agreed with the SSA.
On appeal, Murphy argues the ALJ’s credibility determination was flawed because it was not supported by substantial evidence, and we agree. The ALJ erred by not questioning Murphy further about her failure to fully comply with her home exercise program and the activities she participated in while on vacation. Murphy also asserts, and we agree, that the ALJ’s
residual functional capacity (“RFC”) assessment was flawed because it did not take into account Murphy’s potential inability to do light work. Finally, Murphy argues that the ALJ inappropriately used the medical-vocational grids (“grids”) to find her capable of working and we agree because the ALJ erroneously excluded from the RFC assessment information that should have been included. So we reverse and remand for further proceedings.
I. BACKGROUND
Karen Murphy was hospitalized after having a stroke on April 13, 2007. Before leaving the hospital, Murphy was examined by Dr. Joseph Mayer, who noted her past history of headaches and her diminished fluency in speech. Dr. Mayer noted that although Murphy could read a simple sentence, at times she substituted unintended words or phrases. He also noted that she had mild weakness on her right side, and a loss of sensation and proprio-ception (the ability to sense where her hand was in relation to her body without looking at it) in her right arm. Dr. Mayer recommended that Murphy see a physical therapist to help her rehabilitate.
Ten days after her stroke, Murphy started seeing a physical therapist who recommended a home exercise program to aid her rehabilitation, but instead of completing her physical therapy, she only attended two out of the four therapy sessions prescribed. The therapist recommended that Murphy be discharged from the program based on her attendance. It is unclear when Murphy returned to physical therapy, but on June 13, 2007, the same physical therapist wrote that Murphy had undergone seven weeks of physical therapy. Despite returning to the program, Murphy did not complete it and once again her therapist recommended that she be discharged from the program.
Dr. Mayer examined Murphy again less than two weeks after she was discharged from the hospital, at which time she complained that she felt light-headed, dizzy, and tired, and that she experienced occasional sharp pain in her right hand and spots in her left eye. At a follow-up visit on May 11, 2007, she stated that while her dizziness and light-headedness were gone, her headaches remained.
Two months after her stroke, Murphy met with Dr. Mayer who noted that her speech had improved, that her right foot was “essentially normal,” but that her pro-prioception in her right hand remained poor. At the end of July 2007, Murphy saw Melissa Schultz, Dr. Mayer’s physician assistant, and reported pain on the right side of her head. She told Dr. Mayer’s assistant that she had recently returned from vacation and did not notice the pain while she was away. She also reported continued numbness and discomfort in her right forearm. Upon examination, Schultz characterized the decreased sensation in Murphy’s right arm and hand as “mild” and “improving.” Two months later, in September 2007, Dr. Mayer noted that Murphy reported that her left-sided headaches were better, but she experienced periodic numbness along the right side of her face that sometimes developed into headache pain. Dr. Mayer also noted that Murphy’s speech had “significantly” improved. In April 2008, a year after her stroke, Murphy followed up with Dr. Mayer. He noted that Murphy still had difficulty speaking and “some significant loss of sensation.” He also noted that Murphy suffered from almost nightly headaches, but doubted that they were related to her stroke because she suffered from headaches before her stroke.
On September 29, 2008, Murphy applied for disability benefits, but her application was denied by the Commissioner of the
SSA. A hearing was held by an ALJ at which Murphy, her husband, and a vocational expert (“VE”) testified. Murphy testified about how her physical abilities— which included difficulty making a fist with her right hand, buttoning her shirts or jackets, picking up coins, writing or typing, and distinguishing hot from cold — impaired her ability to work. She also testified that her daily activities included sweeping, dusting, making the beds, and preparing simple meals. Murphy’s husband’s testimony largely mirrored her testimony. The ALJ asked Murphy and her husband about the vacation they went on in July 2007, but they both said they did not remember going on vacation.
Vocational Expert Pamela Tucker was asked to talk about the type of jobs that someone with certain hypothetical limitations could perform. In response to hypothetical questions posed by the ALJ, the VE testified that there were no sedentary jobs in the regional economy for a person who could neither work with the general public nor use her hands more than occasionally for fine manipulation with the dominant hand. The ALJ asked a second set of hypothetical questions that described a person who had the capacity to do light, unskilled work, but who could only occasionally perform fine hand manipulation. The VE testified that there were a significant number of jobs in the economy that this person could perform. In evaluating Murphy’s claim, the ALJ ruled that Murphy was not disabled because she had the RFC to perform the full range of light, unskilled work.
Murphy sought and obtained a review before the Appeals Council, which adopted the ALJ’s decision. The ALJ’s decision became the final decision of the Commissioner when the Appeals Council denied Murphy’s request for review. She appealed the Council’s decision to federal district court and the parties consented to a magistrate judge conducting the proceeding. Murphy sought review of the agency’s decision challenging the ALJ’s credibility determination, RFC determination, and application of the medical-vocational guidelines. The court affirmed the Commissioner’s decision that Murphy was not disabled and therefore was not entitled to disability insurance benefits. This appeal followed.
II. ANALYSIS
We review de novo the magistrate court’s decision, and reverse an ALJ’s determination only where it is not supported by substantial evidence, which means “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.”
Skinner v. Astrue,
478 F.3d 836, 841 (7th Cir.2007) (quoting
Richardson v. Perales,
402 U.S. 389, 401, 91 S.Ct. 1420, 28 L.Ed.2d 842 (1971)). “We will not, however, reweigh the evidence or substitute our judgment for that of the ALJ’s.”
Pepper v. Colvin,
712 F.3d 351, 362 (7th Cir.2013). In reaching its decision, the ALJ “must build a logical bridge from the evidence to his conclusion, but he need not provide a complete written evaluation of every piece of testimony and evidence.”
Schmidt v. Barnhart,
395 F.3d 737, 744 (7th Cir.2005) (internal citations and quotation marks omitted).
A. ALJ’s Credibility Determination Patently Wrong
Murphy argues that the ALJ improperly analyzed the credibility of her and her husband when the judge imper-missibly relied on oft-criticized boilerplate language and that the ALJ’s credibility determination was not supported by substantial evidence. ALJ credibility determinations are given deference because ALJs are in a special position to hear, see, and assess witnesses.
Shideler v. Astrue,
688 F.3d 306, 311 (7th Cir.2012). Therefore, we will only overturn the ALJ’s credibility determination if it is patently wrong, which means that the decision lacks any explanation or support.
Elder v. Astrue,
529 F.3d 408, 413-14 (7th Cir.2008). In drawing its conclusions, the ALJ must “explain her decision in such a way that allows us to determine whether she reached her decision in a rational manner, logically based on her specific findings and the evidence in the record.”
McKinzey v.
As
true,
641 F.3d 884, 890 (7th Cir.2011).
In reaching her credibility determination, the ALJ used boilerplate language that stated:
After careful consideration of the evidence, I find that the claimant’s medically determinable impairments could reasonably be expected to cause the alleged symptoms; however, the claimant’s statements concerning the intensity, persistence and limiting effects of these symptoms are not credible to the extent that they are inconsistent with the above residual functional capacity assessment.
We have often criticized the inclusion of such boilerplate language as “meaningless” because the language fails to connect the conclusory statement with objective evidence in the record or explain what the ALJ relied on when making her determination.
Pepper,
712 F.3d at 367. However, no matter how unhelpful the language is, simply because the ALJ “used boilerplate language does not automatically undermine or discredit the ALJ’s ultimate conclusion if [she] otherwise points to information that justifies [her] credibility determination.”
Id.
at 367-68. Put differently, the ALJ’s use of boilerplate language is reversible error if she did not give sufficient reasons, grounded in evidence in the record, to support her ultimate determination.
See id.
We must remand this case for further proceedings because the ALJ did not adequately explain its credibility determination and it was not supported by substantial evidence in the record. The ALJ deemed Murphy not credible because she did not attend all of her physical therapy sessions as instructed or fully comply with her home exercise program. The ALJ may deem an individual’s statements less credible if medical reports or records show that the individual is not following the treatment as prescribed.
See
SSR 96-7p, 1996 WL 374186, at *7;
Craft v. Astrue,
539 F.3d 668, 679 (7th Cir.2008). However, such evidence should not negatively affect an individual’s credibility if there are good reasons for the failure to complete the plan.
Craft,
539 F.3d at 679. Therefore, an ALJ may need to question the individual at the administrative proceeding to determine whether there are good reasons the individual did not seek medical treatment or fully comply with prescribed treatment.
See
SSR 96-7p;
Shauger v. Astrue,
675 F.3d 690, 696 (7th Cir.2012).
Here, the ALJ stated that Murphy’s documented failure to complete her prescribed medical treatment suggested that Murphy’s symptoms were not as limiting as she claimed and drew a negative inference as to Murphy’s credibility from her lack of follow through. But the ALJ did not ask Murphy why she did not attend all of her physical therapy sessions, or why she did not comply with her home exercise program. There may be a reasonable explanation behind Murphy’s actions, such as she may not have been able to afford the treatment, further treatment would have been ineffective, or the treatment created intolerable side effects.
See Shauger,
675 F.3d at 696. However, we cannot assess the validity of the ALJ’s credibility determination because the ALJ did not ask important questions to determine if Murphy’s actions were justifiable.
In addition, the ALJ found Murphy’s claim less than credible because Murphy took a vacation in July 2007, three months after her stroke, and again in July 2008. The ALJ stated that Murphy’s ability to go on vacation in July 2007, so soon after her stroke, suggested that Murphy’s symptoms were not as disabling as Murphy claimed. Once again, the ALJ’s assessment is problematic because the evidence does not support the inference the ALJ draws between Murphy’s symptoms and her ability to take a vacation. The ALJ’s assessment might have withstood scrutiny if, upon questioning Murphy and her husband, the ALJ found evidence that Murphy, for example, went on a Whitewater rafting vacation, walked with lions in Africa, or ran with the bulls in Spain. If Murphy was able to do these types of activities, legitimate questions would be raised as to the veracity of her claims. Even if the record showed that Murphy did something less strenuous, the ALJ’s determination might have withstood scrutiny, but the record does not indicate how going on vacation was inconsistent with Murphy’s claimed degree of physical limitation. One medical report from Schultz, Dr. Joseph Mayer’s physician assistant, at the end of July 2007 stated that Murphy reported head pain on her right side, but that Murphy had not noticed the symptom while she was on vacation. The report, however, does not suggest that because Murphy went on vacation that she could work.
A similar problem exists with the ALJ’s assessment of Murphy’s July 2008 vacation. The record shows that Murphy took a relaxing vacation with family members to Mexico in July 2008 where, according to Murphy, she mostly laid in the sun. For all we know, she could have been sunning herself on the beach while listening to smooth jazz. The record only indicates that Murphy regained the ability to talk, but it does not suggest that she did any strenuous activity while on vacation. In fact, she stated that other members of her family carried her bags. Given the limited information available on the record, such a vacation as described by Murphy would not be inconsistent with her symptoms to the point where her credibility would be diminished. Once again, we cannot assess the validity of the ALJ’s determination because the record is devoid of information that might support her assessment and the ALJ did not ask follow-up questions that might prove insightful. Therefore, we conclude that the ALJ’s credibility determination is not supported by substantial evidence.
B. ALJ’s RFC Determination Not Supported by Substantial Evidence
Murphy also argues that the ALJ erred in crafting its RFC assessment because, according to Murphy, the ALJ “cherrypicked” evidence that supported her decision while overlooking evidence that favored Murphy’s position. Murphy maintains that the purportedly overlooked evidence supports a finding that she has limitations that the ALJ failed to account for in evaluating her RFC.
The ALJ must determine an individual’s RFC, or “what an individual can still do despite his or her limitations,” SSR 96-8p, 1996 WL 374186, at *2, based upon medical evidence as well as “other evidence, such as testimony by the claimant or his friends and family,”
Craft,
539 F.3d at 676. In making a proper RFC determination, the ALJ must consider all of the relevant evidence in the record, “even [limitations] that are not severe, and may not dismiss a line of evidence contrary to the ruling.”
Villano v. Astrue,
556 F.3d 558, 563 (7th Cir.2009). However, a determination “need not contain a complete written evaluation of every piece of evi
dence,”
McKinzey,
641 F.3d at 891 (quoting
Schmidt,
395 F.3d at 744).
The ALJ in this case concluded that Murphy had the RFC to perform light work, which is defined as lifting no more than twenty pounds at a time with frequent lifting or carrying of objects weighing up to ten pounds. A job may also be considered light work if it requires “standing or walking, off and on, for a total of approximately six hours of an eight-hour workday” with intermittent sitting or “involves sitting most of the time with some pushing and pulling of arm or leg controls.” 20 C.F.R. § 404.1567(b);
Haynes v. Barnhart,
416 F.3d 621, 627 n. 1 (7th Cir.2005).
After examining the record, we find that the AL J’s RFC determination was not fully grounded in Murphy’s testimony or the medical evidence. Murphy said that in December 2007 she was only able to lift or carry ten pounds, walk one block, stand for fifteen minutes, and sit for only limited periods of time. There is no medical evidence in the record to contradict Murphy’s claim. No doctor conducted a functional assessment, which includes a function-by-function assessment of Murphy’s capability to perform light work.
Nor do Dr. Mayer’s treatment notes, which the ALJ primarily relied on, contradict Murphy’s testimony. In June 2007, two months after her stroke, Dr. Mayer noted that Murphy had a right-side facial droop, decreased mobility involving her entire right side, and difficulties with her speech and the ability to place words. In July 2007, Murphy complained of numbness and pain on the right side of her head and stated that although she did not experience the pain while she was on vacation, it still caused her discomfort. She also stated that she had noticed some improvement in her speech, but complained of numbness and discomfort in her right forearm. By September 2007, Murphy had noticed some improvement in her speech, but complained that it was still negatively impacted by the numbness she experienced on the right side of her face. She also complained of continuing difficulty placing words. In December 2007, Murphy once again saw Dr. Mayer and said that while she had “both good days and bad days,” the cold weather negatively affected her right side. Dr. Mayer noted that Murphy was using her right hand fairly well and that Murphy felt as though her condition was stable. He also noted that she had mild proprioception deficit in her right hand, but that she was doing better than in the past, had a normal gait, and was able to tandem walk (walking in a straight line where the toes of the back foot touch the heel of the front foot at each step) without support. In April 2008, Murphy’s speech was generally fluent, but she had occasional difficulty finding words. Murphy also had impaired sensation in her right hand and leg and a moderate decrease in proprioception on her right side, which was a little worse in her hand and foot. Finally, Murphy’s gait and tandem gait were normal. Based on Dr. Mayer’s notes, the ALJ concluded that Murphy was able to perform light work because her condition improved the year following her stroke.
The problem with the ALJ’s determination is that none of Dr. Mayer’s notes address the legal requirements one must be able to perform before the ALJ can determine that the individual is able to do light work. There was no commentary in Dr. Mayer’s notes that addressed Murphy’s ability to lift or carry any weight, let alone the weight that one must carry in order to be capable of doing light work. Similarly, there was no evidence that suggested that Murphy was able to walk or stand six hours of an eight-hour workday with intermittent sitting, which one must be able to do. Admittedly, Dr. Mayer
noted that Murphy’s gait and tandem gait had returned to normal, but that description is not informative as to whether Murphy could perform light work.
The government argues Dr. Mayer’s statements that characterize Murphy as medically improving and “neurologically stable” support the ALJ’s decision, but we disagree because those characterizations do not give us an accurate description of Murphy’s true neurological state. For example, Murphy could have been in terrible condition immediately after her stroke and still be characterized as “stable” by her doctor if her condition had not changed over a period of time. Moreover, one’s medical condition could improve drastically, but still be incapable of performing light work. The key is not whether one has improved (although that is important), but whether they have improved enough to meet the legal criteria of not being classified as disabled.
Although the ALJ stated the medical notes showed Murphy’s health consistently improved the twelve months after her stroke, there is no evidence to suggest that she had improved to the point where she could perform light work. Simply because one is characterized as “stable” or “improving” does not necessarily mean that she is capable of doing light work. The ALJ noted that Murphy still suffered from considerable limitations as a result of the stroke. For example, Murphy experienced difficulty knowing where her right hand was spatially located without looking at it in April 2008 and she still had issues related to her speech.
Based on the record, we conclude that the ALJ failed to build the accurate and logical bridge from the evidence to her conclusion so that, we as a reviewing court, could assess the validity of her ultimate findings and afford Murphy meaningful judicial review.
Young v. Barnhart,
362 F.3d 995, 1002 (7th Cir.2004). The ALJ’s RFC determination is inadequate because it is not supported by substantial evidence, such as a doctor’s functional assessment, or Dr. Mayer’s notes and Murphy was not discredited to the point where the ALJ could not rely on her testimony. Based on these facts, the RFC assessment does not take into account Murphy’s asserted inability to lift no more than twenty pounds at a time, carry objects weighing up to ten pounds, or stand or walk for six hours of an eight-hour workday.
C. ALJ’s Application of Vocational Grids Flawed
At the final step of the disability analysis, the ALJ concluded that Rule 202.21 governed her decision and that based on Murphy’s age, education, work experience, and ability to perform a full range of light work activity, Murphy was not disabled. According to Rule 202.21 of the medical-vocational guidelines, or “grids,” an individual with Murphy’s characteristics and a residual functional capacity for light work is “not disabled.”
See
20 C.F.R. pt. 404, subpt. P, app. 2, tbl. No. 2, Rule 202.21.
There are two broad types of limitations that may affect a claimant’s ability to work: exertional and non-exertional. Exertional imitations, such as sitting, standing, walking, and lifting affect only an individual’s ability to meet the strength demands of jobs.
Haynes v. Barnhart,
416 F.3d 621, 628 (7th Cir.2005). Non-exertional limitations, on the other hand, relate to a person’s ability to climb, balance, stoop, kneel, or crouch.
Id.
When a person has only exertional limitations, or when her non-exertional limitations are insignificant, the grids are dispositive on the issue of disability, and an ALJ may rely on the grids to determine whether a person is disabled.
McKinzey,
641 F.3d at 892. But if the claimant has exertional and non-
exertional limitations, an ALJ cannot rely solely on the grids.
Haynes,
416 F.3d at 628-29.
Murphy first argues that she had non-exertional limitations, such as forgetfulness, dizziness, depression, anxiety, and aphasia that precluded application of the grids and that the ALJ erred by using the grids exclusively. An RFC determination must account for all impairments, even those that are not severe in isolation.
Terry v. Astrue,
580 F.3d 471, 477 (7th Cir.2009). If the ALJ had properly assessed Murphy’s credibility and found that she did not have nonexertional limitations, then applying the vocational guidelines would not amount to reversible error.
See McKinzey,
641 F.3d at 892-93. However, as we previously discussed, the ALJ found Murphy less than credible. Since this decision was erroneous and the ALJ excluded Murphy’s non-exertional limitations based on this erroneous finding, Murphy’s RFC assessment did not include information that should have been included. Therefore, the ALJ could not solely rely on the vocational guideline grids, but rather needed to consult a vocational expert.
Though the ALJ asked a VE a number of hypothetical questions that pertained to Murphy’s ability to work, she ultimately did not rely on the VE’s opinion. However, even if the ALJ had relied on the VE’s opinion, we would have to remand the case because the ALJ excluded important information regarding Murphy’s ability to work. We have stated repeatedly that ALJs must provide VEs with a complete picture of a claimant’s residual functional capacity.
Jelinek v. Astrue,
662 F.3d 805, 813 (7th Cir.2011). The hypothetical question posed to a VE need not include every physical limitation of a claimant, provided that the VE had the opportunity to learn of the claimant’s limitations through, for example, an independent review of the medical records or through other questioning at the hearing.
See Young,
362 F.3d at 1003. However, if the hypothetical posed to the VE does not include all of the claimant’s limitations, there must be some amount of evidence in the record indicating that the VE knew the extent of the claimant’s limitations.
Id.
We require the VE to know about a claimant’s limitations so that the VE does not refer to work that the claimant is not capable of undertaking.
Steele v. Barnhart,
290 F.3d 936, 942 (7th Cir.2002).
Murphy also argues that the hypothetical questions the ALJ asked of the VE excluded information the VE needed to make an accurate assessment, and we agree. The ALJ is only required to incorporate into her hypothetical those impairments and limitations that she accepts as credible.
See Schmidt v. Astrue,
496 F.3d 833, 846 (7th Cir.2007). Here, the ALJ’s flawed credibility determination affected the ALJ’s RFC findings about the extent of Murphy’s limitations, which led the ALJ to ask more restrictive hypotheticals of the ALJ based upon only some, as opposed to all, of Murphy’s complaints. As noted above, Murphy alleged she was only able to lift or carry ten pounds, walk one block, stand for fifteen minutes, and sit for only limited periods of time, but the ALJ did not find her claims credible because they were inconsistent with the ALJ’s assessment of Murphy’s RFC. Therefore, the ALJ excluded those facts from the hypothetical questions she asked the VE. But as we stated above, the ALJ’s RFC was not supported by substantial evidence, nor was Murphy so discredited to the point that the ALJ could find her claims to be not credible. Since Murphy’s claims were not inconsistent with the RFC, the ALJ should have included them in the hypothetical questions she posed to the VE. The lack of such evidence did not focus the VE’s attention on jobs Murphy could perform. Moreover, there is nothing in the
record that reflects that the VE independently knew of all the limitations related to Murphy’s inabilities. Because the ALJ did not include in her hypotheticals information that should have been included, this case warrants remand.
III. CONCLUSION
The judgment is REVERSED and the case is REMANDED with instructions to return the case to the Social Security Administration for further proceedings consistent with this opinion.