Jeff Kasarsky suffers from lower back pain,. chronic depression, and a mild depressive disorder known as dysthymia. He applied for Disability Insurance Benefits (DIB) under Title II of the Social Security Act, 42 U.S.C. §§ 216(i), 223(d), and for Supplemental Security Income (SSI) under Title XVI of the Act, 42 U.S.C. §§ 1603, 1614(a)(30). After preliminary agency proceedings, an administrative law judge (ALJ) found that he was not disabled and thus not entitled to benefits under either program. The Appeals Council and the district court both affirmed the ALJ’s decision. We respectfully disagree for the reasons explained below, and remand this case to the agency for further proceedings.

I

Born in 1959, Kasarsky is now a forty-two-year-old man with an eleventh grade education. He has failed in four attempts to earn his GED and has spent most of his life working in various unskilled general labor positions. In the 1980s, he worked as a general laborer at two large factories. He also worked as a guard for a security company. Matters took a turn for the worse for him on May 4, 1992, when he was injured in an automobile accident. About six hours after the accident, he began experiencing severe pain in his lower back. This did not prevent him right away from working; after the accident, he held a series of positions such as commercial driver for an airline, general laborer for a temporary service, and a car porter for a rental car company. His most recent employment was in May 1996, when he performed general assembly work at a packaging company. He quit after only three days at work,. claiming that his back pain was too intense. He has not worked, since that time.

The relevant medical evidence in the record relates to both Kasarsky’s physical pain and his mental condition. On the physical side, Kasarsky began seeing a chiropractor, Dr. Michael-Fletcher, in February 1993. Dr. Fletcher found that, although Kasarsky continued to suffer from chronic back pain, he was not going to improve any further. Dr. Fletcher cleared Kasarsky to return to work. A year later, Kasarsky saw Dr. David Coleman, who diagnosed him with “thoracic myofascial pain syndrome” and opined that his condition was “becoming quite chronic.” Dr. Coleman referred Kasarsky to physical therapy and also prescribed Nortriptyline, an antidepressant, and Naprosyn, a mild painkiller. When Kasarsky returned to Dr. Coleman’s office a week later, he reported that his back had improved significantly. Dr. Coleman recommended that he continue with the same regimen of medications.

In July 1994, a few months after his visits to Dr. Coleman, Kasarsky began seeing Dr. Catherine Dremel, once again for the back. Kasarsky’s reports to her were[*542] somewhat positive. He told her that his back pain was decreasing and that he was sleeping better at night. Once the pain lessened, he had stopped taking the antidepressant and the painkillers. Dr. Dre-mel advised him to use the painkillers as necessary. Matters worsened, however, by September 1994; Kasarsky returned to Dr. Dremel then and reported a significant increase in the back pain. She prescribed more painkillers.

In October 1994, Dr. Dremel examined Kasarsky and ordered a bone scan, as he did not seem to be responding to physical therapy. 'The bone scan showed a “normal uptake of the spine and upper pelvis.” Later, she also ordered an MRI and a surgical consultation with neurosurgeon Dr. John Hutchinson. Dr. Hutchinson examined the MRI and advised against surgery. Kasarsky continued to take the pain medication and to visit Dr. Dremel monthly. She ultimately recommended that he consider further vocational training to prepare him for a job that required less bending, lifting, and twisting than his former jobs had required.

On the psychological side, in addition to the antidepressants Dr. Coleman had prescribed, Kasarsky began seeing a psychologist, Dr. Daniel Neunaber, in April 1995, for treatment of depression and anxiety resulting from the chronic back pain. He saw Dr. Neunaber four times, and Dr. Neunaber concluded that he suffered from a major depressive disorder as a result of the back problems. Later, Kasarsky had an operation to remove a herniated disk, which was performed by Dr. Kamljit Paul, another neurosurgeon. Dr. Paul also recommended follow-up treatments in the form of epidural steroid injections to control the pain.

Another psychologist entered the picture in the person of Dr. Steven Kaplan, a rehabilitation psychologist, who met Ka-sarsky in April 1996. Dr. Kaplan determined that Kasarsky functions at the fourth-grade reading and spelling level and a third-grade level in basic computations. Dr. Kaplan was of the view that Kasarsky probably would be unable to complete a job application or read a newspaper. Even though Kasarsky had completed the 11th grade, Dr. Kaplan noted that most of his time had been spent in special education classes while he was in school. Dr. Kaplan classified Kasarsky as a “slow learner” who would have trouble making change as a cashier or operating simple machinery. He recommended that Kasar-sky’s mental tasks be limited to a single repetitive action that would not require much instruction.

Kasarsky filed his application for benefits in the middle of this time period, on April 7, 1995. At his hearing before the ALJ in March 1997, he testified that he was still experiencing constant pain; he reported that he took pain medication daily, but that it was only effective' about twenty percent of the time. He also testified that he could sit only for approximately thirty minutes at a time and that he could stand only for about twenty minutes. Finally, he said that he had had suicidal thoughts in the past, but that he was not currently taking any medication for depression.

II

The ALJ performed the standard five-step analysis to determine whether Kasar-sky was disabled. See 20 C.F.R. § 404.1520. As is often the case, the controversy here centers on the ALJ’s conclusions at step five. The ALJ was satisfied that Kasarsky had not engaged in substantial gainful employment since November 1, 1994 (step 1), that the medical evidence established severe back pain, borderline intelligence, and chronic pain syndrome[*543] (step 2), that Kasarsky did not have an impairment that met or equaled anything on the official list in the regulations (step 3), and that Kasarsky could not return to any of his past employment (step 4).

At step 5, the burden shifts to the Commissioner to show that the claimant can perform some other kind of substantial gainful employment that exists in adequate numbers in the national economy. In order to evaluate this point, the ALJ here considered the testimony of a vocational expert, Robert Verkins. As is normally done, the ALJ posed several hypothetical questions to Verkins that reflected different combinations of the physical and mental constraints Kasarsky has. Eventually, he asked whether a person with the capacity to perform sedentary work, who was also limited by borderline intelligence, could find work. This question did not, however, focus Verkins’s attention on the additional limitations of deficiencies in ability to concentrate, persistence, and pace. Verkins testified that there were approximately 150,000 jobs in the national economy that were sedentary and accessible to a person with borderline intelligence, including product inspector, machine feeder, and hand packager. Based on this evidence, the ALJ concluded that Kasarsky was not disabled and rejected his claims.

The Appeals Council denied Kasarsky’s two requests for review and thus the ALJ’s decision became that of the Commissioner. Kasarsky then appealed to the district court, where both parties consented to review by a magistrate judge. See 28 U.S.C. § 636(c). The magistrate judge concluded that substantial evidence supported the ALJ’s decision and upheld the Commissioner’s judgment. This appeal followed.

Ill

In reviewing decisions of the ALJ, this court examines the entire record. We do not disturb the ALJ’s determination if it is supported by “substantial evidence.” See Clifford v. Apfel, 227 F.3d 863, 869 (7th Cir.2000). Nor do we “reweigh the evidence, resolve conflicts, decide questions of credibility, or substitute our own judgment for that of the Commissioner.” Id. In coming to his decision, however, the ALJ must confront evidence that does not support his conclusion and explain why it was rejected. See Herron v. Shalala, 19 F.3d 329, 333 (7th Cir.1994). Furthermore, to the extent the ALJ relies on testimony from a vocational expert, the question posed to the expert must incorporate all relevant limitations from which the claimant suffers. See id. at 337. Otherwise, the vocational testimony will not reveal whether there are jobs in the national economy that a person like the claimant could perform, and if so, how many.

In this case, the ALJ completed a Psychiatric Review Technique Form (PRTF), which at the time was required by Social Security regulations. See 20 C.F.R. § 404.1520(a). That form, which was appended to the ALJ’s decision and is thus part of the record, indicates that Kasarsky had symptoms of “affective disorders” and of “mental retardation and autism.” With respect to the affective disorders, the form shows that Kasarsky suffered from “other dysthymia not severe”; in the mental retardation category, the ALJ indicated that Kasarsky had a learning disability. Finally, in a section calling for the rating of impairment severity, the ALJ found that Kasarsky had no restrictions on his daily living activities and no difficulty maintaining his level of social functioning. Importantly, however, in a third category entitled “Deficiencies of. Concentration, Persistence or. Pace Resulting in Failure to Complete Tasks in a Timely Manner (in work settings or. elsewhere),” the ALJ checked the box marked “frequent.”

[*544] Kasarsky criticizes the ALJ’s decision for a comment in which he said that “[the PRTF] form is not part of this decision and should not be construed as’ such.” The point, however, is not whether the PRTF form was literally “part of the decision.” It is instead whether the limitations noted in the form, which the ALJ went on to acknowledge immediately after the quoted statement, were incorporated in the hypothetical he posed to the vocational expert. It appears that the question posed to Verkins reflected only the ALJ’s finding about Kasarsky’s residual functional capacity, which read as follows:

I find that the claimant has the residual functional capacity to perform the physical exertional and nonexertional requirements of light work, not requiring more than occasional bending, squatting, and kneeling, any scaffold or ladder climbing, or any frequent overhead reaching. Because of borderline intelligence, the claimant is serious[ly] limited, but not precluded from understanding, remembering, and carrying out detailed instructions.

We see nothing in this description, however, that takes into account the ALJ’s own earlier observation (both in his opinion and in the PRTF) that Kasarsky suffered from frequent deficiencies of concentration, persistence, or pace. It is possible, of course, that there is an explanation for this omission. Perhaps the ALJ thought that even with frequent deficiencies of this type, Kasarsky could still carry out detailed instructions in a way that would satisfy a potential employer. But we have no way of knowing that, and it is equally possible that Verkins might have found that there were no jobs for someone with (a) limited exertional abilities, (b) borderline intelligence, and (c) frequent deficiencies of concentration, persistence or pace. Employers are entitled to demand that their employees stick with the job, once they have been trained to do it; the length of time it takes someone with borderline intelligence to learn a job is not the same as the ability of that person to perform consistently once trained. The ALJ’s failure to incorporate the latter kind of limitation, fully supported by this record, in the hypothetical he posed to the vocational expert requires us to remand this case for further proceedings.

We add a few words about Kasarsky’s other arguments for the sake of completeness. Kasarsky claims that the ALJ did not consider the impact of his low intelligence level and non-severe dysthymia, but we disagree. Two of the three hypotheticals specifically included limitations based on Kasarsky’s own intelligence level, and the ALJ’s order contains an adequate discussion of his claims of depression. Next, Kasarsky argues that the implication of the vocational expert’s testimony that Kasarsky could not perform his former work because of his back pain was that Kasarsky was also unable to perform any other'sedentary work. We disagree. In fact, Verkins expressly distinguished between the former job at the packaging plant and “other” sedentary jobs that Kasarsky could still perform. Finally, we will not second-guess the ALJ’s decision that Kasarsky’s depression and dysthymia were not significant impairments. There was evidence in the record that tended to indicate that Kasarsky had been able to work despite these problems. Although Dr. Neunaber had diagnosed Kasarsky with a major depressive disorder in 1995, Kasarsky was no longer receiving treatment for depression by the time of the hearing, nor had any other doctor commented on any lingering effects he might have been experiencing. At the hearing, Kasarsky testified that he was no longer taking anti-depression medication. In short, the ALJ’s decision on this point was supported[*545] by substantial evidence and we see nothing that would justify setting it aside.

IV

The judgment of the district court is ReveRsed, and this case is Remanded to that court so that it in turn can remand the matter to the Commissioner for further proceedings consistent with this opinion.