I. Background
The district court has written three opinions in this matter that carefully and clearly recite the facts of this case. See Tsombanidis v. City of W. Haven, 129 F.Supp.2d 136 (D.Conn.2001) (Tsombanidis I); Tsombanidis v. City of W. Haven, 180 F.Supp.2d 262 (D.Conn.2001) (Tsombanidis II); Tsombanidis v. City of W. Haven, 208 F.Supp.2d 263 (D.Conn.2002) (Tsombanidis III). We presume familiari-
ty with Judge Goettel’s writings and only summarize those facts necessary to resolve the issues now before us.
Beverly Tsombanidis, owner of a residence located at 421 Platt Avenue, in West Haven, Connecticut, also known as “Oxford House-Jones Hill” (“OH-JH”); eight “John Does,” current or future residents of OH-JH; and Oxford House, Inc. (“OHI”) brought this action against the First Fire District for the City of West Haven (“Fire District”) and the City of West Haven (“City”) under the Fair Housing Act of 1968, as amended by the Fair Housing Amendments Act of 1988, 42 U.S.C. § 3601, et seq. (“FHAA”) and Title II of the Americans with Disabilities Act, 42 U.S.C. §§ 12131-12165 (“ADA”).1 OHI oversees more than 900 independent Oxford Houses operating both in the United States and abroad that provide homes for recovering alcoholics and drug addicts. Oxford Houses operate on the premise that people recovering from drug and alcohol addictions will remain sober if they live in a supportive environment. As noted by the district court, “[s]tatistics indicate that the average length of stay in an Oxford House is thirteen months [and a] founder of Oxford House claims that eighty percent of those who live in an Oxford House maintain long-term sobriety.” Tsombani-dis II, 180 F.Supp.2d at 273. Neither the City nor the Fire District question these assertions.
The day-to-day affairs of Oxford Houses are governed democratically by the residents of each house without the presence of a medical or therapeutic professional. OHI has found that residents are more likely to succeed if houses are (1) located *571in single-family residential neighborhoods away from readily available drugs and alcohol; (2) close to sites for regular Alcoholics Anonymous and Narcotics Anonymous meetings; (3) near commercial areas substantial enough to provide easy access to basic necessities; (4) near a range of employment opportunities accessible by public transportation; and (5) large enough for a minimum of six people to live, yet small enough that bedrooms are shared by residents. Tsombanidis II, 180 F.Supp.2d at 273.
In 1997, Tsombanidis purchased a two-story house in a residential area of detached single-family houses in West Haven, Connecticut. She bought the property to start OH-JH and, in July 1997, entered into a lease with four persons recovering from alcohol and drug addictions.2 Within days after the original residents moved into OH-JH, neighbors began to question Tsombanidis about the house. After learning of its purpose, neighbors expressed their concerns and it became apparent throughout the fall of 1997 that there was significant community opposition to OH-JH locating in the neighborhood. An anonymous caller to the City complained that OH-JH was operating as an illegal boarding house. Soon thereafter a group of local residents visited Mayor H. Richard Borer complaining about the recovery facility, and a petition signed by eighty-four people was presented to the City Council “protesting the use of the property located at 421 Platt Avenue in a residential neighborhood ... as a rooming house for people in rehabilitation ... in violation of numerous planning and zoning codes.” Tsombanidis II, 180 F.Supp.2d at 274-75.
West Haven enforces its Zoning Regulations, Property Maintenance Code and State Building Code primarily by responding to complaints. After the City received the first complaint, OH-JH was inspected; city officials concluded that Tsombanidis was operating “an Illegal Boarding House in a residential zone” in violation of the City’s zoning regulations. The City also informed Tsombanidis that she was in violation of the City’s Property Maintenance Code § 202.0, regarding one-family dwellings, as well as nine other sections of the Maintenance Code. She was ordered to make alterations to the property and to reduce the number of tenants to three within fourteen days in order to avoid penalties for operating an illegal boarding house. Tsombanidis made the repairs but refused to evict the tenants. On September 22, a City official issued a citation ordering Tsombanidis to pay a fine of $99.00 for every day she was in violation of the zoning and property regulations.
In response to these actions, OHI wrote to City officials explaining the concept behind Oxford Houses. OHI also informed the City that it believed the City’s enforcement efforts to evict the residents were in violation of the FHAA and ADA. Despite continuing its enforcement actions, the City never responded to OHI. Later that fall, the City turned OH-JH’s file over to its counsel and enforcement proceedings were put in abeyance until further notice.
In December 1997, Richard Spreyer, Inspector for the Fire District, inspected OH-JH. Since six unrelated individuals were living together in the house, Spreyer informed Tsombanidis that, as the landlord, she was required to install additional safety measures to ensure compliance with Chapter 20, the “lodging and rooming” *572portion of the Connecticut Fire Safety Code (“fire code”).3 In March 1998, Spreyer notified Tsombanidis she had 15 days to comply with the lodging and rooming requirements or face possible civil proceedings and criminal penalties including a fine and incarceration. OHI responded that application of the fire code to OH-JH violated the FHAA and the ADA.
Upon receipt of OHI’s letter, Spreyer forwarded the OH-JH file to Douglas Peabody, Deputy State Fire Marshal, requesting a determination of the occupancy classification. Peabody stated that OH-JH should be designated as a lodging and rooming house because six unrelated individuals rented the house. He advised Spreyer to consult with counsel for the city to determine whether the FHAA and ADA applied. City counsel referred Spreyer to Assistant State Attorney Mary Galvin who advised him that the statutes would have no application in this instance because the fire code was at issue rather than a zoning code. On June 15, 1998, Spreyer re-inspected OH-JH, and one day later sent Tsombanidis a final notice of fire safety hazards, stating that imprisonment of up to six months and/or criminal fines from $200 to $1,000 would be imposed in the event she did not comply. He later suspended any enforcement of the abatement during the pendency of this action.
Plaintiffs brought the present case against the Fire District and the City alleging that both governmental entities violated the FHAA and ADA by intentionally discriminating against plaintiffs, implementing policies that disparately impacted plaintiffs, and failing to make reasonable accommodations. Both defendants moved for summary judgment. The district court held that there was sufficient evidence to go forward on plaintiffs’ claim of intentional discrimination against the City but not against the Fire District. See Tsombanidis I, 129 F.Supp.2d at 152-55. The court held that both disparate impact claims could proceed to trial but held that the reasonable accommodation claims were not ripe because plaintiffs had not yet utilized the appropriate administrative proceedings to obtain an accommodation. See id. at 159-61.
In response to the court’s ruling that the reasonable accommodation claims were not ripe, Tsombanidis applied to the City of West Haven Zoning Board of Appeals for a special-use exception to continue to use the property as OH-JH. The Zoning Board held a public hearing and subse*573quently denied the application. Only two months before trial, Tsombanidis requested that the State Fire Marshal exempt her from the fire code. At the subsequent bench trial, John Blaschik, a new Deputy State Fire Marshal, testified that one of the then seven residents of OH-JH could be considered a member of a single family, and the other six could be considered outsiders, making OH-JH a single-family dwelling under the fire code. Spreyer promptly informed Tsombanidis that he would follow this interpretation and that she should disregard the previous notices.
After an eight day bench trial, the district court held that the fire code had a disparate impact on the John Doe plaintiffs. See Tsombanidis II, 180 F.Supp.2d at 296-98. The court also found that plaintiffs failed to prove a reasonable accommodation claim against the Fire District because the plaintiffs received the accommodation they requested. See id. at 298. The court awarded plaintiffs attorney’s fees but no compensatory damages because plaintiffs had not proven intentional discrimination. See id. at 299. The Fire District appeals the district court’s disparate impact holding and a mootness issue included in that claim. Plaintiffs cross-appeal the court’s reasonable accommodation ruling and its failure to award compensatory damages upon a finding of disparate impact. Plaintiffs have abandoned their intentional discrimination claim against the Fire District.
With respect to the City, the court held that: (1) the City intentionally discriminated against OH-JH; (2) the zoning and maintenance regulations disparately impacted the residents; and (3) the City failed to reasonably accommodate the residents’ handicap after plaintiffs had sought a variance through proper governmental procedures. See id. at 284-92. The court awarded plaintiffs compensatory damages and attorney’s fees. See id. at 296. The City appeals the intentional discrimination claim, the reasonable accommodation claim and the damages award. It does not contest the disparate impact claim.
II. Discussion
A. Statutory Framework
Both the FHAA and the ADA prohibit governmental entities from implementing or enforcing housing policies in a discriminatory manner against persons with disabilities. The FHAA makes it unlawful “[t]o discriminate in the sale or rental, or to otherwise make unavailable or deny, a dwelling to any buyer or renter because of a handicap.” 42 U.S.C. § 3604(f)(1). Similarly, the ADA states “no qualified individual with a disability shall, by reason of such disability, be excluded from participation in or be denied the benefits of the services, programs, or activities of a public entity, or be subject to discrimination by any such entity.” 42 U.S.C. § 12132. Both statutes require “that covered entities make reasonable accommodations in order to provide qualified individuals with an equal opportunity to receive benefits from or to participate in programs run by such entities.” Reg’l Econ. Cmty. Action Program, Inc. v. City of Middletown, 294 F.3d 35, 45 (2d Cir.), cert. denied, 537 U.S. 813, 123 S.Ct. 74, 154 L.Ed.2d 16 (2002). To establish discrimination under either the FHAA or the ADA, plaintiffs have three available theories: (1) intentional discrimination (disparate treatment); (2) disparate impact; and (3) failure to make a reasonable accommodation. See id. at 48.4
*574Both statutes apply to municipal zoning decisions. See id. at 45-46 (citing Forest City Daly Hous., Inc. v. Town of North Hempstead, 175 F.3d 144, 151 (2d Cir.1999); Innovative Health Sys., Inc. v. City of White Plains, 117 F.3d 37, 44 (2d Cir.1997)). Furthermore, the Fire District has not specifically contested the application of either statute to the fire code. Cf. Marbrunak, Inc. v. City of Stow, 974 F.2d 43 (6th Cir.1992) (applying FHAA to a zoning ordinance imposing special safety requirements on residences housing individuals with developmental disabilities).5 In addition, defendants have not contested that the John Doe plaintiffs, as recovering alcoholics and drug addicts, are considered “handicapped” or persons with disabilities and therefore protected by both the FHAA and the ADA. Finally, defendants concede plaintiffs Tsombanidis and OHI have standing in this case.6
B. Claims Against the Fire District
The Fire District argues that the case against it became moot when Blaschik testified he had changed the interpretation of the fire code, and Spreyer informed Tsombanidis he would follow that interpretation. Thus, OH-JH, which has capacity only for seven recovering individuals, will not be required to implement the fire safety measures necessary for residences treated as lodging and rooming houses. One day after the testimony, Inspector Spreyer informed Tsombanidis and the other plaintiffs that all enforcement efforts would cease as long as OH-JH did not exceed seven members.
“[Fjederal courts may not adjudicate matters that no longer present an actual dispute between parties.” Catanzano v. Wing, 277 F.3d 99, 107 (2d Cir.2001). “The voluntary cessation of allegedly illegal activities will usually render a case moot ‘if the defendant can demonstrate that (1) there is no reasonable expectation that the alleged violation will recur and (2) interim relief or events have completely and irrevocably eradicated the effects of the alleged violation.’ ” Granite State Outdoor Advert., Inc. v. Town of Orange, 303 F.3d 450, 451 (2d Cir.2002) (quoting Campbell v. Greisberger, 80 F.3d 703, 706 (2d Cir.1996)). The defendant’s burden is a heavy one to ensure the allegedly illegal activities do not temporarily cease only to resume after the claims have been dismissed. Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 189, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000). The Fire District has not met its heavy burden because the interpretation of the code might change again — for example, upon a change in the State Fire Marshal’s administration. Thus, the claims are not moot.
1. Disparate Impact Analysis
The district court found the Fire District in violation of the FHAA and ADA holding the fire code disparately impacted the John Doe plaintiffs. Disparate impact analysis focuses on facially neutral policies or practices that may have a discriminatory effect. “To establish a prima facie case under this theory, the plaintiff *575must show: ‘(1) the occurrence of certain outwardly neutral practices, and (2) a significantly adverse or disproportionate impact on persons of a particular type produced by the defendant’s facially neutral acts or practices.’ ” City of Middletown, 294 F.3d at 52-53 (quoting Gamble v. City of Escondido, 104 F.3d 300, 306 (9th Cir.1997)) (emphasis added). A plaintiff need not show the defendant’s action was based on any discriminatory intent. Huntington Branch, NAACP v. Town of Huntington, 844 F.2d 926, 934-36 (2d Cir.1988). When establishing that a challenged practice has a significantly adverse or disproportionate impact on a protected group, a plaintiff must prove the practice “actually or predictably results in ... discrimination.” Hack v. President & Fellows of Yale Coll., 237 F.3d 81, 90 (2d Cir.2000) (quoting Toum of Huntington, 844 F.2d at 934). A plaintiff has not met its burden if it merely raises an inference of discriminatory impact. See Gamble, 104 F.3d at 306. Furthermore, the plaintiff must show a causal connection between the facially neutral policy and the alleged discriminatory effect. See Hack, 237 F.3d at 90-91. If a plaintiff makes a prima facie showing, the burden shifts to the defendant to “prove that its actions furthered, in theory and in practice, a legitimate, bona fide governmental interest and that no alternative would serve that interest with less discriminatory effect.” Town of Huntington, 844 F.2d at 936.
In this case, plaintiffs challenge the facially neutral provisions of Connecticut’s Fire Safety Code relating to lodging and rooming houses. If the residence is a lodging or rooming house, plaintiffs would be required to make a number of structural changes to the building. The district court found these requirements to be “prohibitively expensive for OH-JH and that the continued enforcement of these provisions would result in the constructive eviction of the John Doe plaintiffs from this one-family dwelling and would limit the housing opportunities available to Oxford House residents.” Tsombanidis II, 180 F.Supp.2d at 297. The court further held that plaintiffs had presented “substantial evidence of their need to live in a group home setting in a residential neighborhood, in order to facilitate their continued recovery from alcoholism and drug addiction,” and that this need for group living is not shared by “non-handicapped persons” to the same degree. Id. After finding that the fire code had an adverse impact on plaintiffs, it held that although the Fire District could point to a legitimate interest — fire safety — the Fire District presented no evidence that the regulations were the least restrictive means to serve its legitimate interest. Id.
We disagree with the district court’s legal analysis and find as a matter of law that plaintiffs failed to establish a prima facie claim of disparate impact. The basis for a successful disparate impact claim involves a comparison between two groups — those affected and those unaffected by the facially neutral policy. This comparison must reveal that although neutral, the policy in question imposes a “significantly adverse or disproportionate impact” on a protected group of individuals. When examining disparate impact claims under the FHAA and ADA, we use Title VII as a starting point. See Hack, 237 F.3d at 90; see also Town of Huntington, 844 F.2d at 934. To establish a prima facie case for employment discrimination, plaintiffs are ordinarily required to include statistical evidence to show disparity in outcome between groups. See, e.g., Smith v. Xerox Corp., 196 F.3d 358, 365 (2d Cir.1999).
Statistical evidence is also normally used in cases involving fair housing disparate *576impact claims. For example, in Town of Huntington, the district court found a shortage of affordable rental housing for low and middle-income households and that the impact of this shortage was three times greater on blacks than on the overall population. Town of Huntington, 844 F.2d at 929. Furthermore, the town had restricted new low-income housing to an area consisting of 52% minority residents while the entire town was 98% white. Id. at 937-38. Thus, the Court concluded the town’s refusal to amend the ordinance restricting multifamily housing projects to largely minority urban renewal areas disparately impacted minorities. Id. at 938. Similarly, in Resident Advisory Board v. Rizzo, 564 F.2d 126 (3d Cir.1977), the court held that plaintiffs had proven discriminatory effect by showing that the defendants’ urban renewal efforts had left the area in question an “all-white community” whereas the area had previously been integrated to the extent of having 45% black families. Id. at 149. The court concluded that there was no “doubt that the impact of the governmental defendants’ termination of the project was felt primarily by blacks, who make up a substantial proportion of those who would be eligible to reside there.” Id.
By contrast, in Hack, this Court dismissed plaintiffs’ disparate impact claims against a university housing policy that required students to live in co-educational residence halls, which were objectionable to Orthodox Jews on religious grounds. The court noted that “[t]he students do not allege that [the university’s] policy has resulted in or predictably will result in under-representation of Orthodox Jews in [university] housing. Therefore, their claim fails.” Hack, 237 F.3d at 91.
Here, plaintiffs did not present any statistical information nor did they show the fire code actually or predictably created a shortage of housing for recovering alcoholics in the community. Although there may be cases where statistics are not necessary, there must be some analytical mechanism to determine disproportionate impact. The district court merely required plaintiffs to show they could not live in the house they desired because of the code and that plaintiffs needed this type of housing due to their handicap. The court’s analysis would appear to apply to any facially neutral housing policy that prevents a handicapped person from living in a particular house. Such a standard is not sufficient for disparate impact purposes. If a handicapped person requested and was denied a reasonable accommodation to the neutral policy, there would be a strong argument that the denial violated the FHAA and ADA, but this would only be true under a reasonable accommodation theory as opposed to a disparate impact theory. See Henrietta D. v. Bloomberg, 331 F.3d 261, 276-77 (2d Cir.2003) (holding that the ADA permits both theories of disparate impact and reasonable accommodation and that each is a separate and distinct claim). To prevail on a theory of disparate impact, however, there must be some evidence that a significant number of people suffering the handicap need group living and that the fire code restricts a substantial portion of similarly handicapped individuals from doing so. The only evidence the court had in this regard cuts against plaintiffs’ theory. There are twenty-six Oxford Houses in Connecticut and six, not including OH-JH, in the greater New Haven area; it does not appear that the state-wide fire code has adversely affected these homes. See Tsombanidis II, 180 F.Supp.2d at 272.
Whether using statistics or some other analytical method, plaintiffs must also utilize the appropriate comparison groups. They must first identify members of a protected group that are affected by the *577neutral policy and then identify similarly situated persons who are unaffected by the policy. It is unclear from its decision what groups the district court compared. To the extent that the district court compared the handicapped plaintiffs to “a similarly sized family where the individuals were related by blood, marriage or adoption,” see Tsombanidis I, 129 F.Supp.2d at 157, that comparison was improper. It fails to include similar-sized groups that are not related by blood — seven college students wanting to live together, for example — but are still affected by the policy. See Gamble, 104 F.3d at 304. The district court also erred by merely comparing handicapped and non-handicapped persons. Rather, in this case, the proper comparison is between (1) recovering alcoholics and recovering drug abusers (“recover-ings”) and (2) people who are neither recovering alcoholics nor recovering drug abusers (“non-recoverings”). Such a comparison identifies the handicap and allows for a causal analysis between the claim of discrimination based on the handicap in question and the facially neutral policy.
In this case, plaintiffs might have been able to meet their burden by providing statistical evidence (1) that x% of all of the recoverings in West Haven need (or have good reason) to live in the “group settings” prohibited by the facially neutral fire regulations at issue, (2) that y% of all of the non-recoverings in West Haven need (or have good reason) to live in such group settings prohibited by the fire regulations, and, crucially, (3) that x is significantly greater than y. There is nothing in the district court’s decision nor do plaintiffs point to any evidence in the record alluding to such a statistical comparison.
A more qualitative comparison might also have supported the disparate impact theory. In such a comparison, plaintiffs would have to show that the average recovering in West Haven has a greater need — qualitatively—for group living than does the average non-recovering resident of West Haven. This would likely require some quantification of what each group “needs” from a living arrangement standpoint. A court could then conclude that, despite whether the quantitative test is met, there is a qualitatively disproportionate impact on recoverings in West Haven. “If a significant correlation exists between being disabled and living in group houses, a disparate impact on group housing could conceivably establish a prima facie disparate impact claim.” Gamble, 104 F.3d at 307 n. 2. No evidence was presented in this case that establishes a significant correlation between being disabled and living in group housing.
Plaintiffs seem to have taken the qualitative track, but again, they have not shown any proof that there are other re-coverings in West Haven who need group living of seven or more or any proof about non-recoverings’ needs. The district court found that plaintiffs had established that the Oxford House program was a highly successful rehabilitation method, especially when recoverings were attending Alcoholics Anonymous or Narcotics Anonymous meetings. Tsombanidis II, 180 F.Supp.2d at 274. The court also found that one of the criteria that made this program “much more likely to succeed” was its desire to use houses “large enough for a minimum of six people to live, yet small enough that bedrooms are shared by residents.” Id. at 273. This was insufficient to establish a “‘comparison class’ of ‘similarly situated individuals given preferential treatment.’ ” Henrietta D., 331 F.3d at 273 (quoting Olmstead v. L.C. ex rel. Zimring, 527 U.S. 581, 119 S.Ct. 2176, 144 L.Ed.2d 540 (1999) (plurality opinion)). Oxford House’s own experts noted that not all recoverings need group living and that other factors, including attending Alcoholics Anonymous meet*578ings and abstinence, also play a substantial factor in recovering. Since no quantitative or qualitative comparison was proven, plaintiffs did not establish a disparate impact claim.
2. Reasonable Accommodation Analysis
Plaintiffs also assert that the Fire District’s refusal to treat OH-JH as a one-family dwelling qualifies as' a failure to reasonably accommodate the John Doe plaintiffs’ handicap as required by the FHAA and the ADA. Plaintiffs contest both the district court’s original ripeness decision as well as the holding that the reasonable accommodation provisions were not violated. We affirm both rulings.
Under the FHAA and the ADA,7 a governmental entity engages in a discriminatory practice if it refuses to make a “reasonable accommodation” to “rules, policies, practices or services when such accommodation may be necessary to afford [a handicapped person] equal opportunity to use and enjoy a dwelling.” 42 U.S.C. § 3604(f)(3)(B); see also 42 U.S.C. § 12131(2) (“The term ‘qualified individual with a disability’ means an individual with a disability who, with or without reasonable modifications to rules, policies or practices ... meets the essential eligibility requirements for the receipt of services or the participation in programs or activities provided by a public entity.”). Thus, these statutes “require that changes be made to such traditional rules or practices if necessary to permit a person with handicaps an equal opportunity to use and enjoy a dwelling.” Shapiro v. Cadman Towers, Inc., 51 F.3d 328, 333 (2d Cir.1995) (quoting H.R.Rep. No. 711, reprinted in 1988 U.S.C.C.A.N. 2173, 2186 (footnotes omitted)). “Plaintiffs must show that, but for the accommodation, they likely will be denied an equal opportunity to enjoy the housing of their choice.” Smith & Lee Assocs., Inc. v. City of Taylor, 102 F.3d 781, 795 (6th Cir.1996). A defendant must incur reasonable costs and take modest, affirmative steps to accommodate the handicapped as long as the accommodations sought do not pose an undue hardship or a substantial burden. Shapiro, 51 F.3d at 334-35; see also Olmstead v. L.C. ex rel. Zimring, 527 U.S. 581, 605 n. 16, 119 S.Ct. 2176, 144 L.Ed.2d 540 (1999); Salute v. Stratford Greens Garden Apartments, 136 F.3d 293, 300 (2d Cir.1998) (requiring landlord to accept Section 8 housing tenants modified an integral aspect of the landlord’s rental policy).
Plaintiffs argue that neither statute requires the exhaustion of administrative remedies and, therefore, the district court erred in requiring plaintiffs to seek a variance from the State Fire Marshal before bringing this claim. Under Connecticut General Statute § 29-296, variations or exemptions from the fire code may be granted by the State Fire Marshal; similar authority is not vested in the local Fire District. Appellees never sought an accommodation. In fact, OHI’s March 24, 1998 letter notified Inspector Spreyer that Oxford House was not seeking an accommodation in this regard and that its position was that the code was facially invalid under the Federal Fair Housing Act as it was being applied to Oxford House-Jones Hill.
To prevail on a reasonable accommodation claim, plaintiffs must first provide the governmental entity an opportunity to accommodate them through the entity’s established procedures used to adjust the neutral policy in question. Oxford House-C v. City of St. Louis, 77 F.3d 249, *579253 (8th Cir.1996); see also United States v. Vill. of Palatine, 37 F.3d 1230, 1234 (7th Cir.1994) (holding that an administrative procedure must be used unless plaintiff can show such an action would be futile). Furthermore, requiring OH-JH to utilize facially neutral procedures to request an accommodation from the fire code is not by itself a failure to reasonably accommodate plaintiffs’ handicaps. A governmental entity must know what a plaintiff seeks prior to incurring liability for failing to affirmatively grant a reasonable accommodation. It may be that once the governmental entity denies such an accommodation, neither the FHAA nor the AHA require a plaintiff to exhaust the state or local administrative procedures. See Bryant Woods Inn, Inc. v. Howard County, 124 F.3d 597, 601 (4th Cir.1997) (holding that plaintiffs were not required to exhaust county administrative procedures after it received a final decision on its application for a variance to zoning restrictions); see also 42 U.S.C. § 3613(a)(2) (permitting private enforcement of the FHA “whether or not a complaint [to the Secretary] has been filed”). But a plaintiff must first use the procedures available to notify the governmental entity that it seeks an exception or variance from the facially neutral laws when pursuing a reasonable accommodation claim.8 Here, OH-JH specifically stated in its original letter it was not seeking an accommodation.
We also affirm the district court’s decision that the accommodation plaintiffs ultimately sought was provided two months after it was requested. Plaintiffs did not seek an exception to the fire code until August 2001. During the trial, the Deputy State Fire Marshal testified that under his interpretation of the fire code, seven individuals could live together and still be considered a single-family residence. The next day and before the close of the trial, local Inspector Spreyer informed plaintiffs that all abatement procedures against OH-JH would end. Thus, the accommodation plaintiffs sought — being classified a single-family residence — was granted.
C. Claims Against the City
As stated above, the district court found the City in violation of the FHAA and ADA on all three available theories. Since the City did not contest the disparate impact holding, we do not review the merits of that claim and only address the intentional discrimination and reasonable accommodation claims along with the damages award.
1. Intentional Discrimination
The City argues that the district court’s finding of discriminatory intent was clearly erroneous. See Harris Trust & Sav. Bank v. John Hancock Mut. Life Ins. Co., 302 F.3d 18, 26 (2d Cir.2002); Joseph v. New York City Bd. of Educ., 171 F.3d 87, 93 (2d Cir.1999). To establish intentional discrimination, plaintiffs must prove that a motivating factor behind the City’s refusal to classify OH-JH as a single family household was the residents’ status as recovering drug addicts and alco*580holics. See Vill. of Arlington Heights v. Metro. Housing Dev. Corp., 429 U.S. 252, 265-66, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977). Factors to be considered in evaluating a claim of intentional discrimination include: “(1) the discriminatory impact of the governmental decision; (2) the decision’s historical background; (3) the specific sequence of events leading up to the challenged decision; (4) departures from the normal procedural sequences; and (5) departures from normal substantive criteria.” Tsombanidis I, 129 F.Supp.2d at 152 (citing Vill. of Arlington Heights, 429 U.S. at 266-68, 97 S.Ct. 555). These factors are not exclusive or mandatory but merely a framework within which a court conducts its analysis.
The district court’s finding of intentional discrimination was not clearly erroneous. The court used the appropriate factors and the evidence presented supports its findings. Among other things, the district court noted the history of hostility of neighborhood residents to OH-JH and their pressure on the Mayor and other city officials. Evidence supports the court’s finding that this hostility motivated the City in initiating and continuing its enforcement efforts. See Innovative Health Sys., Inc. v. City of White Plains, 117 F.3d 37, 49 (2d Cir.1997) overruled on other grounds by Zervos v. Verizon New York, Inc., 252 F.3d 163, 171 n. 7 (2d Cir.2001). There was also evidence the City rarely took enforcement action against boarding houses in residential neighborhoods. Furthermore, the City failed to acknowledge multiple letters sent by OHI thoroughly explaining OH-JH’s policies and procedures and its argument that the residents had a right to be treated as a single-family residence. The court also cited the reaction of Michael McCur-ry, one of two Property Maintenance Code Officials for the City. McCurry expressed his personal dissatisfaction with OH-JH and ordered Tsombanidis to evict the residents without any authority in the City Code. Finally, there was record support for the court’s finding of bias in the denial of OH-JH’s request for a special use exception by the Zoning Board of Appeals. We therefore affirm the district court’s conclusion that the city intentionally discriminated.
2. Reasonable Accommodation
We also affirm the district court’s finding that plaintiffs requested a reasonable accommodation and the City failed to grant it. The City is not required to grant an exception for a group of people to live as a single family, but it cannot deny the variance request based solely on plaintiffs’ handicap where the requested accommodation is reasonable. The district court found that these plaintiffs operated much like a family. Additionally, there is evidence that these particular plaintiffs needed to live in group homes located in single-family areas. See Tsombanidis II, 180 F.Supp.2d at 293. The City concedes that, from a municipal services standpoint, it would bear minimal financial cost from the proposed accommodation. While legitimate concerns of residential zoning laws include the integrity of the City’s housing scheme and problems associated with large numbers of unrelated transient persons living together, such as traffic congestion and noise, see Vill. of Belle Terre v. Boraas, 416 U.S. 1, 9, 94 S.Ct. 1536, 39 L.Ed.2d 797 (1974); Oxford House-C, 77 F.3d at 252, the City points to no evidence that those concerns were present here. The district court’s finding was therefore not clearly erroneous.
3. Damages
We affirm the district court’s award of compensatory damages and at*581torney’s fees in its entirety and pause briefly to discuss its inclusion of a damages award for an attorney’s involvement in the Zoning Board appeal. “The standard of review of an award of attorney’s fees is highly deferential to the district court. Because attorney’s fees are dependent on the unique facts of each case, the resolution of this issue is committed to the discretion of the district court.” Baker v. Health Mgmt. Sys,, Inc., 264 F.3d 144, 149 (2d Cir.2001) (quoting Mautner v. Hirsch, 32 F.3d 37, 39 (2d Cir.1994)). The district court’s award of attorney’s fees, as well as its entire damages award, was not excessive but rather carefully calculated and reasonable.
The district court awarded plaintiffs’ attorney’s fees for work on the Zoning Board appeal. Under 42 U.S.C. § 3613(c)(2) a court may award reasonable attorney’s fees to the prevailing party in a private enforcement action.9 We believe the district court correctly analogized this case to Pennsylvania v. Delaware Valley Citizens’ Council, 478 U.S. 546, 106 S.Ct. 3088, 92 L.Ed.2d 439 (1986), which interpreted similar statutory language employed in the Clean Air Act. Both the FHAA, 42 U.S.C. § 3613(c)(2), and section 304 of the Clean Air Act, 42 U.S.C. § 7604(d), allow a prevailing party to obtain attorney’s fees for private enforcement “actions.” Both statutes use only the term “action” instead of “action or proceeding.” In Delaware Valley, however, the Court held that the Clean Air Act should be interpreted in a manner similar to § 1988. 478 U.S. at 559-61, 106 S.Ct. 3088. Section 1988 permits attorney’s fees “for time spent on administrative proceedings to enforce the civil rights claim prior to the litigation.” North Carolina Dep’t of Transp. v. Crest Street Cmty. Council, Inc., 479 U.S. 6, 15, 107 S.Ct. 336, 93 L.Ed.2d 188 (1986). To obtain the fees, the administrative proceeding must be “useful and of a type ordinarily necessary to secure the final result obtained from the litigation.” Delaware Valley, 478 U.S. at 561, 106 S.Ct. 3088 (internal quotations omitted). The Court employed the same reasoning for the Clean Air Act, because like § 1988, the Clean Air Act was “enacted to ensure that private citizens have a meaningful opportunity to vindicate their rights.” Id. at 559. The same can be said for private citizen suits brought pursuant to the FHAA. Thus, an administrative proceeding could be included in the calculation of reasonable attorney’s fees if it is “useful and of a type ordinarily necessary to secure the final result obtained from the litigation.” Id. at 561.
As we have noted earlier, see supra at 578-79, plaintiffs were required to use proper local procedures to request a reasonable accommodation from a governmental entity before bringing an action under the FHAA or ADA in this regard. Plaintiffs used the appropriate channels to seek a variance in the zoning regulations and to cure their ripeness problem. Thus, we agree with the district court that the proceeding before the Zoning Board was the type ordinarily necessary to secure the final result in an FHAA enforcement action brought under a reasonable accommodation theory. Thus, we affirm its inclusion of this award.
III. Conclusion
For the above reasons, the district court’s order of January 8, 2002 is hereby *582AfpiRmed in part and Reversed and Remanded in part.
. In the amended complaint, plaintiffs also alleged both defendants violated the Equal Protection Clause, and thus brought a claim under 42 U.S.C. § 1983. In its summary judgment decision, however, the district court dismissed this claim against both defendants and plaintiffs have not appealed that decision. See Tsombanidis I, 129 F.Supp.2d at 161-62.
. More residents moved into OH-JH soon after the original lease was signed. Although the number of residents has fluctuated since 1997, normally OH-JH has been operating at its capacity, which is seven residents. See Tsombanidis 11, 180 F.Supp.2d at 271.
. In December 1997, the Connecticut Fire Safety Code defined "lodging or rooming houses” as
buildings that provide sleeping accommodations for a total of 16 or fewer persons on either a transient or permanent basis, with or without meals, but without separate cooking facilities for individual occupants except as provided in Chapter 21.
Chapter 21 defined one and two-family dwellings as
buildings containing not more than two dwelling units in which each living unit is occupied by members or a single family with no more than five outsiders, if any, accommodated in rented rooms.
In April 2000, the Code was amended to define "lodging or rooming houses” as
buildings or portions thereof that do not qualify as a one or two-family dwelling that provide sleeping accommodations for a total of 16 but not fewer than seven persons on either a transient or permanent basis, with or without meals, but without separate cooking facilities for individual occupants except as provided in Chapter 21.
Chapter 21 was amended to define "one and two-family dwellings” as
buildings containing not more than two dwelling units in which each living unit is occupied by members of a single family with no more than six outsiders, if any accommodated in rented rooms.
Tsombanidis II, 180 F.Supp.2d at 279-80.
. Due to the similarities between the statutes, we interpret them in tandem. See City of Middletown, 294 F.3d at 45-46. Although there may be differences in the FHAA and *574ADA, we see no material differences presented in this case and the parties have not identified any.
. The Fire District does mention that most other cases also include zoning provisions, but it has not argued that the federal statutes do not apply to safety codes.
. As a justiciability issue, it is insufficient that defendants do not contest plaintiffs’ standing, but it is clear that both Tsombanidis, as owner of the home, and OHI, as the parent organization, will incur an injury and have standing in this case. See City of Middletown, 294 F.3d at 46 n. 2.
. Again, in this case the two statutes are treated similarly.
. Such a holding is not in conflict with our earlier ruling in Huntington Branch, NAACP v. Town of Huntington, 689 F.2d 391 (2d Cir.1982) (Huntington I). In that case, we held that plaintiffs were not required to exhaust local remedies by filing a formal application for rezoning. Id. at 393 n. 3. There, however, plaintiffs claimed the local entity’s policies were discriminatory on a disparate impact theory and did not assert failure to reasonably accommodate. Huntington II, 844 F.2d at 928. This is not an exhaustion requirement but merely a requirement that plaintiffs first use the proper procedure to seek an exception or variance. If denied this request, they do not need to exhaust the administrative appeal process.
. We reject the City's argument that we should analyze the availability of attorney’s fees for work on the Zoning Board appeal under 42 U.S.C. § 3612(p). 42 U.S.C. § 3612 applies only to enforcement actions brought by the Secretary of Housing and Urban Development, not to private actions, as in the instant case.